Arthroplasty/Joint Replacement

The Internet is becoming a primary source for obtaining medical information. This growing trend may have serious implications for the medical field. As patients increasingly regard the Internet as an essential tool for obtaining health-related information, questions have been raised regarding the quality of medical information available on the Internet.¹ Studies have shown that health-related sites often present inaccurate, inconsistent, and outdated information that may have a negative impact on health care decisions made by patients.²

According to the US Census Bureau, 71.7% of American households report having access to the Internet.³ Of those who have access to Internet, approximately 72% have sought health information online over the last year.⁴ Among people older than age 65 years living in the United States, there has been a growing trend toward using the Internet, from 14% in 2000 to almost 60% in 2013, according to the Pew Research Internet Project.⁵ Most medical websites are viewed for information on diseases and treatment options.⁶ Since most patients want to be informed about treatment options, as well as risks and benefits for each treatment, access to credible information is essential for proper decision-making.⁷

To assess the quality of information on the Internet, we used DISCERN, a standardized questionnaire to aid consumers in judging Internet content.⁸ The DISCERN instrument, available at www.discern.org.uk, was designed by an expert group in the United Kingdom. First, an expert panel developed and tested the instrument, and then health care providers and self-help group members tested it further.^8,9 The questionnaire had been found to have good interrater reliability, regardless of use by health professionals or consumers.^8-10

More than 53,000 shoulder arthroplasties are performed in the United States annually, and the number is growing, with the main goal of pain relief from glenohumeral degenerative joint disease.^11,12 The Internet has become a quasi–second opinion for patients trying to participate in their care. Given the prevalence of shoulder-related surgeries, it is critical to analyze and become familiar with the quality of information that patients read online in order to direct them to nonbiased, all-inclusive websites. In this study, we provide a summary assessment and comparison of the quality of online information pertaining to shoulder replacement, using medical (total shoulder replacement) and nontechnical (shoulder replacement) search terms.

Methods

Websites were identified using 3 search engines (Google, Yahoo, and Bing) and 2 search terms, shoulder replacement (SR) and total shoulder arthroplasty (TSA), on January 17, 2014. These 3 search engines were used because 77% of health care–related information online searches begin through a search engine (Google, Bing, Yahoo); only 13% begin at a health care–specialized website.⁴ These search terms were used after consulting with orthopedic residents and attending physicians in a focus group regarding the terminology used with patients. The first 30 websites in each search engine were identified consecutively and evaluated for category and quality of information using the DISCERN instrument.

A total of 180 websites (90 per search term) were reviewed. Each website was evaluated independently by 3 medical students. In the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram, we recorded how websites were identified, screened, and included (Figure 1).13 Websites that were duplicated within each search term and those that were inaccessible were used to determine the total number of noncommercial versus commercial websites, but were excluded from the final analysis. The first part of the analysis involved determining the type of website (eg, commercial vs noncommercial) based upon the html endings. All .com endings were classified as commercial websites; noncommercial included .gov, .org, .edu, and .net endings. Next, each website was categorized based on the target audience. Websites were grouped into health professional–oriented information, patient-oriented, advertisement, or “other.” These classifications were based on those described in previous works.^14,15 The “other” category included images, YouTube videos, another search engine, and open forums, which were also excluded from the final analysis because they were not easily evaluable with the DISCERN instrument. Websites were considered health professional–oriented if they included journal articles, scholarly articles, and/or rehabilitation protocols. Patient-directed websites clearly stated the information was directed to patients or provided a general overview. Advertisement included sites that displayed ads or products for sale. Websites were evaluated for quality using the DISCERN instrument (Figure 2).

DISCERN has 3 subdivision scores: the reliable score (composed of the first 8 questions), the treatment options (the next 7 questions), and 1 final question that addresses the overall quality of the website and is rated independently of the first 15 questions. DISCERN uses 2 scales, a binary scale anchored on both extremes with the number 1 equaling complete absence of the criteria being measured, and the number 5 at the upper extreme, representing completeness of the quality being assessed. In between 1 and 5 is a partial ordinal scale measuring from 2 to 4, which indicates the information is present to some extent but not complete. The ordinal scale allows ranking of the criteria being assessed. Summarizing values from each of the 2 scales poses some concern: the scale is not a true binary scale because of the ordinal scale of the middle numbers (2-4), and as such, is not amenable to being an interval scale to calculate arithmetic means. To summarize the values from the 2 scales, we calculated the harmonic mean, the arithmetic mean, the geometric mean, and the median. The means were empirically compared with the median, and we used the harmonic mean to summarize scale values because it was the best approximation of the medians.

Results

A total of 90 websites were assessed with the search term total shoulder arthroplasty and another 90 with shoulder replacement. When 37 duplicate websites for TSA and 52 for SR were eliminated, 53 (59%) and 38 (42%) unique websites were evaluated for each search term, respectively (Figure 1). (These unique websites are included in the Appendix.) Between the 2 search terms, 20 websites were duplicated. Figure 3 shows the distribution of websites by category. Total shoulder arthroplasty provided the highest percentage of health professional–oriented information; SR had the greatest percentage of patient-oriented information. Both TSA and SR had nearly the same number of advertisements and websites labeled “other.” The percentage of noncommercial websites from each search engine is represented in Figure 4. For SR, Google had 40% (12/30) noncommercial websites compared with Yahoo at 53% (16/30) and Bing at 46% (14/30). Total shoulder arthroplasty had 43% (13/30) noncommercial websites on Google, 27% (8/30) on Yahoo, and 40% (12/30) on Bing. In total, SR had more noncommercial websites, 47% (42/90), compared with 37% (33/90) for TSA.

The mean of all 3 raters for reliablity (DISCERN questions 1-8) and treatment options (DISCERN questions 9-15) is represented in the Table. For both search terms, we found that websites identified as health professional–oriented had the highest reliable mean scores, followed by patient-oriented, and advertisement at the lowest (SR: P = .054; TSA: P = .134). For SR, treatment mean scores demonstrated similar results with health professional–oriented websites receiving the highest, followed by patient-oriented and advertisement (P = .005). However, the treatment mean scores for TSA differed with patient-oriented websites receiving higher scores than health professional–oriented websites, but this was not statistically significant (P= .407). Regarding search terms, there were no significant differences between mean reliable and treatment scores across all categories.

The average overall DISCERN score for TSA websites was 2.5 (range, 1-5), compared with 2.3 (range, 1-5) for SR websites. The overall reliable score (DISCERN questions 1-8) for TSA websites was 2.6 and 2.5 for SR websites (P < .001). For TSA websites, 38% (20/53) were classified as good, having an overall DISCERN score ≥3, versus 26% (10/38) of SR websites. The overall DISCERN score for health professional–oriented websites was 2.7, patient-oriented websites received a score of 2.6, and advertisements had the lowest score at 2.4.

Discussion

Both patients and health professionals obtain information on health care subjects through the Internet, which has become the primary resource for patients.^15,16 However, there are no strict regulations of the content being written. This creates a challenge for the typical user to find credible and evidence-based information, which is important because misleading information could cause undue anxiety, among other effects.^17,18 The aims of this study were to determine the quality of Internet information for shoulder replacement surgeries using the medical terminology total shoulder arthroplasty (TSA) and the nontechnical term shoulder replacement (SR), and to compare the results.

After analyzing the types of websites returned for both total shoulder arthroplasty and shoulder replacement (Figure 4), it was interesting to find that using nonmedical terminology as the search term provided more noncommercial websites compared with total shoulder arthroplasty. Furthermore, Yahoo provided the highest yield of noncommercial websites at 16, with Bing at 14, when using SR as the search term. We believe the increase in noncommercial websites returned for SR was greater than for TSA because SR yielded more patient-oriented websites, which usually had html endings of .edu and .org, as shown in Figure 3 (48% of SR websites offered patient-oriented information).

Although there were more noncommercial websites for SR, the majority of the DISCERN values between the 2 search terms did not differ significantly. This is a direct result of the number of sites (20) that were duplicated across both search terms. However as seen in the Table, TSA had similar reliable mean scores for advertisements and patient-oriented websites but a slightly higher reliable score for health professional–oriented websites. We correlated this with the increased number of health professional–oriented websites returned when using TSA as the search term (Figure 3). The health professional–oriented websites explained their aims and cited their sources more consistently than did patient-oriented sites and advertisements, resulting in higher reliable scores. Although patient-oriented websites frequently lacked citations, they provided information about multiple treatment options, which were more relevant to consumers. This resulted in nearly equivalent reliable scores. Treatment means for advertisements in both SR and TSA were similar. However, treatment means for professional-oriented websites in TSA were lower than those for SR because health professional–oriented websites often were only moderately relevant to consumers, with their focus usually on 1 treatment option or on rehabilitation protocols. Although the DISCERN scores were similar between the search terms, total shoulder arthroplasty provided more websites (20) classified as good—overall DISCERN score, ≥3—than SR did (10). Advertisement websites had similar overall DISCERN scores, which we anticipated because most of the advertisements were duplicated across the search terms.

Using the 2 search terms, academic websites and commercial websites, such as WebMD, consistently received higher reliable and overall DISCERN scores. Advertisement websites, which need to deliver a clear message, frequently scored high on explicitly stating their aims and relevance to consumers, but focused on their products without discussing the benefits of other treatment options. This is significant because Internet search engines, such as Google, offer sponsor links for which organizations pay to appear at the top of the search results. This creates the potential for consumers to receive biased information because most individuals only visit the top 10 websites generated by a search engine.¹⁹

We concluded that the quality of online information relating to SR and TSA was highly variable and frequently of moderate-to-poor quality, with most overall DISCERN scores <3. The quality of information found online for this study using the DISCERN instrument is consistent with those studies using DISCERN to evaluate other medical conditions (eg, bunions, chronic pain, general anesthesia, and anterior cruciate ligament reconstruction).^2,9,15,19 These studies also concluded that online information varies tremendously in quality and completeness.

This study has several limitations. Websites were searched at a single time point and, because Internet resources are frequently updated, the results of this study could vary. Furthermore, although Google, Yahoo, and Bing are 3 of the most popular search engines, these are not the only resources patients use when searching the Internet for health-related information. Other search engines, such as Pubmed.gov and MSN.com, could provide additional websites for Internet users. Lastly, although DISCERN is validated to address the quality of information available online, it does not evaluate the accuracy of the information.⁸ Our use of DISCERN involves 2 scales, a binary yes/no (ratings, 1 and 5) and an ordinal scale (ratings, 2-4). As such, a single mean summary statistic cannot be calculated.

Conclusion

The information available on the Internet pertaining to TSA and SR is highly variable and provides mostly moderate-to-poor quality information based on the DISCERN instrument. Many websites failed to describe the benefits and the risks of different treatment options, including nonoperative management. Health care professionals should be aware that patients often refer to the Internet as a primary resource for obtaining medical information. It is important to direct patients to websites that provide accurate information, because patients who educate themselves about their conditions and actively participate in decision-making may have improved health outcomes.20-22 Overall, academic websites and commercial websites, such as WebMD and OrthoInfo, generally had higher DISCERN scores when using either search term. Of major concern is the potential for misleading advertisements or incorrect information that can negatively affect health outcomes. This study found that using nonmedical terminology (SR) provided more noncommercial and patient-oriented websites, especially through Yahoo. This study highlights the need for more comprehensive online information pertaining to shoulder replacement that can better serve as a resource for Internet users.

The Internet is becoming a primary source for obtaining medical information. This growing trend may have serious implications for the medical field. As patients increasingly regard the Internet as an essential tool for obtaining health-related information, questions have been raised regarding the quality of medical information available on the Internet.¹ Studies have shown that health-related sites often present inaccurate, inconsistent, and outdated information that may have a negative impact on health care decisions made by patients.²

According to the US Census Bureau, 71.7% of American households report having access to the Internet.³ Of those who have access to Internet, approximately 72% have sought health information online over the last year.⁴ Among people older than age 65 years living in the United States, there has been a growing trend toward using the Internet, from 14% in 2000 to almost 60% in 2013, according to the Pew Research Internet Project.⁵ Most medical websites are viewed for information on diseases and treatment options.⁶ Since most patients want to be informed about treatment options, as well as risks and benefits for each treatment, access to credible information is essential for proper decision-making.⁷

To assess the quality of information on the Internet, we used DISCERN, a standardized questionnaire to aid consumers in judging Internet content.⁸ The DISCERN instrument, available at www.discern.org.uk, was designed by an expert group in the United Kingdom. First, an expert panel developed and tested the instrument, and then health care providers and self-help group members tested it further.^8,9 The questionnaire had been found to have good interrater reliability, regardless of use by health professionals or consumers.^8-10

More than 53,000 shoulder arthroplasties are performed in the United States annually, and the number is growing, with the main goal of pain relief from glenohumeral degenerative joint disease.^11,12 The Internet has become a quasi–second opinion for patients trying to participate in their care. Given the prevalence of shoulder-related surgeries, it is critical to analyze and become familiar with the quality of information that patients read online in order to direct them to nonbiased, all-inclusive websites. In this study, we provide a summary assessment and comparison of the quality of online information pertaining to shoulder replacement, using medical (total shoulder replacement) and nontechnical (shoulder replacement) search terms.

Methods

Websites were identified using 3 search engines (Google, Yahoo, and Bing) and 2 search terms, shoulder replacement (SR) and total shoulder arthroplasty (TSA), on January 17, 2014. These 3 search engines were used because 77% of health care–related information online searches begin through a search engine (Google, Bing, Yahoo); only 13% begin at a health care–specialized website.⁴ These search terms were used after consulting with orthopedic residents and attending physicians in a focus group regarding the terminology used with patients. The first 30 websites in each search engine were identified consecutively and evaluated for category and quality of information using the DISCERN instrument.

A total of 180 websites (90 per search term) were reviewed. Each website was evaluated independently by 3 medical students. In the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram, we recorded how websites were identified, screened, and included (Figure 1).13 Websites that were duplicated within each search term and those that were inaccessible were used to determine the total number of noncommercial versus commercial websites, but were excluded from the final analysis. The first part of the analysis involved determining the type of website (eg, commercial vs noncommercial) based upon the html endings. All .com endings were classified as commercial websites; noncommercial included .gov, .org, .edu, and .net endings. Next, each website was categorized based on the target audience. Websites were grouped into health professional–oriented information, patient-oriented, advertisement, or “other.” These classifications were based on those described in previous works.^14,15 The “other” category included images, YouTube videos, another search engine, and open forums, which were also excluded from the final analysis because they were not easily evaluable with the DISCERN instrument. Websites were considered health professional–oriented if they included journal articles, scholarly articles, and/or rehabilitation protocols. Patient-directed websites clearly stated the information was directed to patients or provided a general overview. Advertisement included sites that displayed ads or products for sale. Websites were evaluated for quality using the DISCERN instrument (Figure 2).

DISCERN has 3 subdivision scores: the reliable score (composed of the first 8 questions), the treatment options (the next 7 questions), and 1 final question that addresses the overall quality of the website and is rated independently of the first 15 questions. DISCERN uses 2 scales, a binary scale anchored on both extremes with the number 1 equaling complete absence of the criteria being measured, and the number 5 at the upper extreme, representing completeness of the quality being assessed. In between 1 and 5 is a partial ordinal scale measuring from 2 to 4, which indicates the information is present to some extent but not complete. The ordinal scale allows ranking of the criteria being assessed. Summarizing values from each of the 2 scales poses some concern: the scale is not a true binary scale because of the ordinal scale of the middle numbers (2-4), and as such, is not amenable to being an interval scale to calculate arithmetic means. To summarize the values from the 2 scales, we calculated the harmonic mean, the arithmetic mean, the geometric mean, and the median. The means were empirically compared with the median, and we used the harmonic mean to summarize scale values because it was the best approximation of the medians.

Results

A total of 90 websites were assessed with the search term total shoulder arthroplasty and another 90 with shoulder replacement. When 37 duplicate websites for TSA and 52 for SR were eliminated, 53 (59%) and 38 (42%) unique websites were evaluated for each search term, respectively (Figure 1). (These unique websites are included in the Appendix.) Between the 2 search terms, 20 websites were duplicated. Figure 3 shows the distribution of websites by category. Total shoulder arthroplasty provided the highest percentage of health professional–oriented information; SR had the greatest percentage of patient-oriented information. Both TSA and SR had nearly the same number of advertisements and websites labeled “other.” The percentage of noncommercial websites from each search engine is represented in Figure 4. For SR, Google had 40% (12/30) noncommercial websites compared with Yahoo at 53% (16/30) and Bing at 46% (14/30). Total shoulder arthroplasty had 43% (13/30) noncommercial websites on Google, 27% (8/30) on Yahoo, and 40% (12/30) on Bing. In total, SR had more noncommercial websites, 47% (42/90), compared with 37% (33/90) for TSA.

The mean of all 3 raters for reliablity (DISCERN questions 1-8) and treatment options (DISCERN questions 9-15) is represented in the Table. For both search terms, we found that websites identified as health professional–oriented had the highest reliable mean scores, followed by patient-oriented, and advertisement at the lowest (SR: P = .054; TSA: P = .134). For SR, treatment mean scores demonstrated similar results with health professional–oriented websites receiving the highest, followed by patient-oriented and advertisement (P = .005). However, the treatment mean scores for TSA differed with patient-oriented websites receiving higher scores than health professional–oriented websites, but this was not statistically significant (P= .407). Regarding search terms, there were no significant differences between mean reliable and treatment scores across all categories.

The average overall DISCERN score for TSA websites was 2.5 (range, 1-5), compared with 2.3 (range, 1-5) for SR websites. The overall reliable score (DISCERN questions 1-8) for TSA websites was 2.6 and 2.5 for SR websites (P < .001). For TSA websites, 38% (20/53) were classified as good, having an overall DISCERN score ≥3, versus 26% (10/38) of SR websites. The overall DISCERN score for health professional–oriented websites was 2.7, patient-oriented websites received a score of 2.6, and advertisements had the lowest score at 2.4.

Discussion

Both patients and health professionals obtain information on health care subjects through the Internet, which has become the primary resource for patients.^15,16 However, there are no strict regulations of the content being written. This creates a challenge for the typical user to find credible and evidence-based information, which is important because misleading information could cause undue anxiety, among other effects.^17,18 The aims of this study were to determine the quality of Internet information for shoulder replacement surgeries using the medical terminology total shoulder arthroplasty (TSA) and the nontechnical term shoulder replacement (SR), and to compare the results.

After analyzing the types of websites returned for both total shoulder arthroplasty and shoulder replacement (Figure 4), it was interesting to find that using nonmedical terminology as the search term provided more noncommercial websites compared with total shoulder arthroplasty. Furthermore, Yahoo provided the highest yield of noncommercial websites at 16, with Bing at 14, when using SR as the search term. We believe the increase in noncommercial websites returned for SR was greater than for TSA because SR yielded more patient-oriented websites, which usually had html endings of .edu and .org, as shown in Figure 3 (48% of SR websites offered patient-oriented information).

Although there were more noncommercial websites for SR, the majority of the DISCERN values between the 2 search terms did not differ significantly. This is a direct result of the number of sites (20) that were duplicated across both search terms. However as seen in the Table, TSA had similar reliable mean scores for advertisements and patient-oriented websites but a slightly higher reliable score for health professional–oriented websites. We correlated this with the increased number of health professional–oriented websites returned when using TSA as the search term (Figure 3). The health professional–oriented websites explained their aims and cited their sources more consistently than did patient-oriented sites and advertisements, resulting in higher reliable scores. Although patient-oriented websites frequently lacked citations, they provided information about multiple treatment options, which were more relevant to consumers. This resulted in nearly equivalent reliable scores. Treatment means for advertisements in both SR and TSA were similar. However, treatment means for professional-oriented websites in TSA were lower than those for SR because health professional–oriented websites often were only moderately relevant to consumers, with their focus usually on 1 treatment option or on rehabilitation protocols. Although the DISCERN scores were similar between the search terms, total shoulder arthroplasty provided more websites (20) classified as good—overall DISCERN score, ≥3—than SR did (10). Advertisement websites had similar overall DISCERN scores, which we anticipated because most of the advertisements were duplicated across the search terms.

Using the 2 search terms, academic websites and commercial websites, such as WebMD, consistently received higher reliable and overall DISCERN scores. Advertisement websites, which need to deliver a clear message, frequently scored high on explicitly stating their aims and relevance to consumers, but focused on their products without discussing the benefits of other treatment options. This is significant because Internet search engines, such as Google, offer sponsor links for which organizations pay to appear at the top of the search results. This creates the potential for consumers to receive biased information because most individuals only visit the top 10 websites generated by a search engine.¹⁹

We concluded that the quality of online information relating to SR and TSA was highly variable and frequently of moderate-to-poor quality, with most overall DISCERN scores <3. The quality of information found online for this study using the DISCERN instrument is consistent with those studies using DISCERN to evaluate other medical conditions (eg, bunions, chronic pain, general anesthesia, and anterior cruciate ligament reconstruction).^2,9,15,19 These studies also concluded that online information varies tremendously in quality and completeness.

This study has several limitations. Websites were searched at a single time point and, because Internet resources are frequently updated, the results of this study could vary. Furthermore, although Google, Yahoo, and Bing are 3 of the most popular search engines, these are not the only resources patients use when searching the Internet for health-related information. Other search engines, such as Pubmed.gov and MSN.com, could provide additional websites for Internet users. Lastly, although DISCERN is validated to address the quality of information available online, it does not evaluate the accuracy of the information.⁸ Our use of DISCERN involves 2 scales, a binary yes/no (ratings, 1 and 5) and an ordinal scale (ratings, 2-4). As such, a single mean summary statistic cannot be calculated.

Conclusion

The information available on the Internet pertaining to TSA and SR is highly variable and provides mostly moderate-to-poor quality information based on the DISCERN instrument. Many websites failed to describe the benefits and the risks of different treatment options, including nonoperative management. Health care professionals should be aware that patients often refer to the Internet as a primary resource for obtaining medical information. It is important to direct patients to websites that provide accurate information, because patients who educate themselves about their conditions and actively participate in decision-making may have improved health outcomes.20-22 Overall, academic websites and commercial websites, such as WebMD and OrthoInfo, generally had higher DISCERN scores when using either search term. Of major concern is the potential for misleading advertisements or incorrect information that can negatively affect health outcomes. This study found that using nonmedical terminology (SR) provided more noncommercial and patient-oriented websites, especially through Yahoo. This study highlights the need for more comprehensive online information pertaining to shoulder replacement that can better serve as a resource for Internet users.

The Internet is becoming a primary source for obtaining medical information. This growing trend may have serious implications for the medical field. As patients increasingly regard the Internet as an essential tool for obtaining health-related information, questions have been raised regarding the quality of medical information available on the Internet.¹ Studies have shown that health-related sites often present inaccurate, inconsistent, and outdated information that may have a negative impact on health care decisions made by patients.²

According to the US Census Bureau, 71.7% of American households report having access to the Internet.³ Of those who have access to Internet, approximately 72% have sought health information online over the last year.⁴ Among people older than age 65 years living in the United States, there has been a growing trend toward using the Internet, from 14% in 2000 to almost 60% in 2013, according to the Pew Research Internet Project.⁵ Most medical websites are viewed for information on diseases and treatment options.⁶ Since most patients want to be informed about treatment options, as well as risks and benefits for each treatment, access to credible information is essential for proper decision-making.⁷

To assess the quality of information on the Internet, we used DISCERN, a standardized questionnaire to aid consumers in judging Internet content.⁸ The DISCERN instrument, available at www.discern.org.uk, was designed by an expert group in the United Kingdom. First, an expert panel developed and tested the instrument, and then health care providers and self-help group members tested it further.^8,9 The questionnaire had been found to have good interrater reliability, regardless of use by health professionals or consumers.^8-10

More than 53,000 shoulder arthroplasties are performed in the United States annually, and the number is growing, with the main goal of pain relief from glenohumeral degenerative joint disease.^11,12 The Internet has become a quasi–second opinion for patients trying to participate in their care. Given the prevalence of shoulder-related surgeries, it is critical to analyze and become familiar with the quality of information that patients read online in order to direct them to nonbiased, all-inclusive websites. In this study, we provide a summary assessment and comparison of the quality of online information pertaining to shoulder replacement, using medical (total shoulder replacement) and nontechnical (shoulder replacement) search terms.

Methods

Websites were identified using 3 search engines (Google, Yahoo, and Bing) and 2 search terms, shoulder replacement (SR) and total shoulder arthroplasty (TSA), on January 17, 2014. These 3 search engines were used because 77% of health care–related information online searches begin through a search engine (Google, Bing, Yahoo); only 13% begin at a health care–specialized website.⁴ These search terms were used after consulting with orthopedic residents and attending physicians in a focus group regarding the terminology used with patients. The first 30 websites in each search engine were identified consecutively and evaluated for category and quality of information using the DISCERN instrument.

A total of 180 websites (90 per search term) were reviewed. Each website was evaluated independently by 3 medical students. In the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram, we recorded how websites were identified, screened, and included (Figure 1).13 Websites that were duplicated within each search term and those that were inaccessible were used to determine the total number of noncommercial versus commercial websites, but were excluded from the final analysis. The first part of the analysis involved determining the type of website (eg, commercial vs noncommercial) based upon the html endings. All .com endings were classified as commercial websites; noncommercial included .gov, .org, .edu, and .net endings. Next, each website was categorized based on the target audience. Websites were grouped into health professional–oriented information, patient-oriented, advertisement, or “other.” These classifications were based on those described in previous works.^14,15 The “other” category included images, YouTube videos, another search engine, and open forums, which were also excluded from the final analysis because they were not easily evaluable with the DISCERN instrument. Websites were considered health professional–oriented if they included journal articles, scholarly articles, and/or rehabilitation protocols. Patient-directed websites clearly stated the information was directed to patients or provided a general overview. Advertisement included sites that displayed ads or products for sale. Websites were evaluated for quality using the DISCERN instrument (Figure 2).

DISCERN has 3 subdivision scores: the reliable score (composed of the first 8 questions), the treatment options (the next 7 questions), and 1 final question that addresses the overall quality of the website and is rated independently of the first 15 questions. DISCERN uses 2 scales, a binary scale anchored on both extremes with the number 1 equaling complete absence of the criteria being measured, and the number 5 at the upper extreme, representing completeness of the quality being assessed. In between 1 and 5 is a partial ordinal scale measuring from 2 to 4, which indicates the information is present to some extent but not complete. The ordinal scale allows ranking of the criteria being assessed. Summarizing values from each of the 2 scales poses some concern: the scale is not a true binary scale because of the ordinal scale of the middle numbers (2-4), and as such, is not amenable to being an interval scale to calculate arithmetic means. To summarize the values from the 2 scales, we calculated the harmonic mean, the arithmetic mean, the geometric mean, and the median. The means were empirically compared with the median, and we used the harmonic mean to summarize scale values because it was the best approximation of the medians.

Results

A total of 90 websites were assessed with the search term total shoulder arthroplasty and another 90 with shoulder replacement. When 37 duplicate websites for TSA and 52 for SR were eliminated, 53 (59%) and 38 (42%) unique websites were evaluated for each search term, respectively (Figure 1). (These unique websites are included in the Appendix.) Between the 2 search terms, 20 websites were duplicated. Figure 3 shows the distribution of websites by category. Total shoulder arthroplasty provided the highest percentage of health professional–oriented information; SR had the greatest percentage of patient-oriented information. Both TSA and SR had nearly the same number of advertisements and websites labeled “other.” The percentage of noncommercial websites from each search engine is represented in Figure 4. For SR, Google had 40% (12/30) noncommercial websites compared with Yahoo at 53% (16/30) and Bing at 46% (14/30). Total shoulder arthroplasty had 43% (13/30) noncommercial websites on Google, 27% (8/30) on Yahoo, and 40% (12/30) on Bing. In total, SR had more noncommercial websites, 47% (42/90), compared with 37% (33/90) for TSA.

The mean of all 3 raters for reliablity (DISCERN questions 1-8) and treatment options (DISCERN questions 9-15) is represented in the Table. For both search terms, we found that websites identified as health professional–oriented had the highest reliable mean scores, followed by patient-oriented, and advertisement at the lowest (SR: P = .054; TSA: P = .134). For SR, treatment mean scores demonstrated similar results with health professional–oriented websites receiving the highest, followed by patient-oriented and advertisement (P = .005). However, the treatment mean scores for TSA differed with patient-oriented websites receiving higher scores than health professional–oriented websites, but this was not statistically significant (P= .407). Regarding search terms, there were no significant differences between mean reliable and treatment scores across all categories.

The average overall DISCERN score for TSA websites was 2.5 (range, 1-5), compared with 2.3 (range, 1-5) for SR websites. The overall reliable score (DISCERN questions 1-8) for TSA websites was 2.6 and 2.5 for SR websites (P < .001). For TSA websites, 38% (20/53) were classified as good, having an overall DISCERN score ≥3, versus 26% (10/38) of SR websites. The overall DISCERN score for health professional–oriented websites was 2.7, patient-oriented websites received a score of 2.6, and advertisements had the lowest score at 2.4.

Discussion

Both patients and health professionals obtain information on health care subjects through the Internet, which has become the primary resource for patients.^15,16 However, there are no strict regulations of the content being written. This creates a challenge for the typical user to find credible and evidence-based information, which is important because misleading information could cause undue anxiety, among other effects.^17,18 The aims of this study were to determine the quality of Internet information for shoulder replacement surgeries using the medical terminology total shoulder arthroplasty (TSA) and the nontechnical term shoulder replacement (SR), and to compare the results.

After analyzing the types of websites returned for both total shoulder arthroplasty and shoulder replacement (Figure 4), it was interesting to find that using nonmedical terminology as the search term provided more noncommercial websites compared with total shoulder arthroplasty. Furthermore, Yahoo provided the highest yield of noncommercial websites at 16, with Bing at 14, when using SR as the search term. We believe the increase in noncommercial websites returned for SR was greater than for TSA because SR yielded more patient-oriented websites, which usually had html endings of .edu and .org, as shown in Figure 3 (48% of SR websites offered patient-oriented information).

Although there were more noncommercial websites for SR, the majority of the DISCERN values between the 2 search terms did not differ significantly. This is a direct result of the number of sites (20) that were duplicated across both search terms. However as seen in the Table, TSA had similar reliable mean scores for advertisements and patient-oriented websites but a slightly higher reliable score for health professional–oriented websites. We correlated this with the increased number of health professional–oriented websites returned when using TSA as the search term (Figure 3). The health professional–oriented websites explained their aims and cited their sources more consistently than did patient-oriented sites and advertisements, resulting in higher reliable scores. Although patient-oriented websites frequently lacked citations, they provided information about multiple treatment options, which were more relevant to consumers. This resulted in nearly equivalent reliable scores. Treatment means for advertisements in both SR and TSA were similar. However, treatment means for professional-oriented websites in TSA were lower than those for SR because health professional–oriented websites often were only moderately relevant to consumers, with their focus usually on 1 treatment option or on rehabilitation protocols. Although the DISCERN scores were similar between the search terms, total shoulder arthroplasty provided more websites (20) classified as good—overall DISCERN score, ≥3—than SR did (10). Advertisement websites had similar overall DISCERN scores, which we anticipated because most of the advertisements were duplicated across the search terms.

Using the 2 search terms, academic websites and commercial websites, such as WebMD, consistently received higher reliable and overall DISCERN scores. Advertisement websites, which need to deliver a clear message, frequently scored high on explicitly stating their aims and relevance to consumers, but focused on their products without discussing the benefits of other treatment options. This is significant because Internet search engines, such as Google, offer sponsor links for which organizations pay to appear at the top of the search results. This creates the potential for consumers to receive biased information because most individuals only visit the top 10 websites generated by a search engine.¹⁹

We concluded that the quality of online information relating to SR and TSA was highly variable and frequently of moderate-to-poor quality, with most overall DISCERN scores <3. The quality of information found online for this study using the DISCERN instrument is consistent with those studies using DISCERN to evaluate other medical conditions (eg, bunions, chronic pain, general anesthesia, and anterior cruciate ligament reconstruction).^2,9,15,19 These studies also concluded that online information varies tremendously in quality and completeness.

This study has several limitations. Websites were searched at a single time point and, because Internet resources are frequently updated, the results of this study could vary. Furthermore, although Google, Yahoo, and Bing are 3 of the most popular search engines, these are not the only resources patients use when searching the Internet for health-related information. Other search engines, such as Pubmed.gov and MSN.com, could provide additional websites for Internet users. Lastly, although DISCERN is validated to address the quality of information available online, it does not evaluate the accuracy of the information.⁸ Our use of DISCERN involves 2 scales, a binary yes/no (ratings, 1 and 5) and an ordinal scale (ratings, 2-4). As such, a single mean summary statistic cannot be calculated.

Conclusion

The information available on the Internet pertaining to TSA and SR is highly variable and provides mostly moderate-to-poor quality information based on the DISCERN instrument. Many websites failed to describe the benefits and the risks of different treatment options, including nonoperative management. Health care professionals should be aware that patients often refer to the Internet as a primary resource for obtaining medical information. It is important to direct patients to websites that provide accurate information, because patients who educate themselves about their conditions and actively participate in decision-making may have improved health outcomes.20-22 Overall, academic websites and commercial websites, such as WebMD and OrthoInfo, generally had higher DISCERN scores when using either search term. Of major concern is the potential for misleading advertisements or incorrect information that can negatively affect health outcomes. This study found that using nonmedical terminology (SR) provided more noncommercial and patient-oriented websites, especially through Yahoo. This study highlights the need for more comprehensive online information pertaining to shoulder replacement that can better serve as a resource for Internet users.

Degenerative arthritis is a widespread chronic condition with an incidence of almost 43 million and annual health care costs of $60 billion in the United States alone.1 Although many cases can be managed symptomatically with medical therapy and intra-articular injections,2 many patients experience disease progression resulting in decreased ambulatory ability and work productivity. For these patients, elective hip and knee arthroplasties can drastically improve quality of life and functionality.3,4 Over the past decade, there has been a marked increase in the number of primary and revision total hip and knee arthroplasties performed in the United States. By 2030, the demand for primary total hip arthroplasties will grow an estimated 174%, to 572,000 procedures. Likewise, the demand for primary total knee arthroplasties is projected to grow by 673%, to 3.48 million procedures.5 However, though better surgical techniques and technology have led to improved functional outcomes, there is still substantial risk for complications in the perioperative period, especially in the geriatric population, in which substantial comorbidities are common.6-9

Acute kidney injury (AKI) is a common public health problem in hospitalized patients and in patients undergoing procedures. More than one-third of all AKI cases occur in surgical settings.^10,11 Over the past decade, both community-acquired and in-hospital AKIs rapidly increased in incidence in all major clinical settings.^12-14 Patients with AKI have high rates of adverse outcomes during hospitalization and discharge.^11,15 Sequelae of AKIs include worsening chronic kidney disease (CKD) and progression to end-stage renal disease, necessitating either long-term dialysis or transplantation.¹² This in turn leads to exacerbated disability, diminished quality of life, and disproportionate burden on health care resources.

Much of our knowledge about postoperative AKI has been derived from cardiovascular, thoracic, and abdominal surgery settings. However, there is a paucity of data on epidemiology and trends for either AKI or associated outcomes in patients undergoing major orthopedic surgery. The few studies to date either were single-center or had inadequate sample sizes for appropriately powered analysis of the risk factors and outcomes related to AKI.¹⁶

In the study reported here, we analyzed a large cohort of patients from a nationwide multicenter database to determine the incidence of and risk factors for AKI. We also examined the mortality and adverse discharges associated with AKI after major joint surgery. Lastly, we assessed temporal trends in both incidence and outcomes of AKI, including the death risk attributable to AKI.

Methods

Database

We extracted our study cohort from the Nationwide Inpatient Sample (NIS) and the National Inpatient Sample of Healthcare Cost and Utilization Project (HCUP) compiled by the Agency for Healthcare Research and Quality.¹⁷ NIS, the largest inpatient care database in the United States, stores data from almost 8 million stays in about 1000 hospitals across the country each year. Its participating hospital pool consists of about 20% of US community hospitals, resulting in a sampling frame comprising about 90% of all hospital discharges in the United States. This allows for calculation of precise, weighted nationwide estimates. Data elements within NIS are drawn from hospital discharge abstracts that indicate all procedures performed. NIS also stores information on patient characteristics, length of stay (LOS), discharge disposition, postoperative morbidity, and observed in-hospital mortality. However, it stores no information on long-term follow-up or complications after discharge.

Data Analysis

For the period 2002–2012, we queried the NIS database for hip and knee arthroplasties with primary diagnosis codes for osteoarthritis and secondary codes for AKI. We excluded patients under age 18 years and patients with diagnosis codes for hip and knee fracture/necrosis, inflammatory/infectious arthritis, or bone neoplasms (Table 1). We then extracted baseline characteristics of the study population. Patient-level characteristics included age, sex, race, quartile classification of median household income according to postal (ZIP) code, and primary payer (Medicare/Medicaid, private insurance, self-pay, no charge). Hospital-level characteristics included hospital location (urban, rural), hospital bed size (small, medium, large), region (Northeast, Midwest/North Central, South, West), and teaching status. We defined illness severity and likelihood of death using Deyo’s modification of the Charlson Comorbidity Index (CCI), which draws on principal and secondary ICD-9-CM (International Classification of Diseases, Ninth Revision-Clinical Modification) diagnosis codes, procedure codes, and patient demographics to estimate a patient’s mortality risk. This method reliably predicts mortality and readmission in the orthopedic population.^18,19 We assessed the effect of AKI on 4 outcomes, including in-hospital mortality, discharge disposition, LOS, and cost of stay. Discharge disposition was grouped by either (a) home or short-term facility or (b) adverse discharge. Home or short-term facility covered routine, short-term hospital, against medical advice, home intravenous provider, another rehabilitation facility, another institution for outpatient services, institution for outpatient services, discharged alive, and destination unknown; adverse discharge covered skilled nursing facility, intermediate care, hospice home, hospice medical facility, long-term care hospital, and certified nursing facility. This dichotomization of discharge disposition is often used in studies of NIS data.²⁰

Statistical Analyses

We compared the baseline characteristics of hospitalized patients with and without AKI. To test for significance, we used the χ² test for categorical variables, the Student t test for normally distributed continuous variables, the Wilcoxon rank sum test for non-normally distributed continuous variables, and the Cochran-Armitage test for trends in AKI incidence. We used survey logistic regression models to calculate adjusted odds ratios (ORs) with 95% confidence intervals (95% CIs) in order to estimate the predictors of AKI and the impact of AKI on hospital outcomes. We constructed final models after adjusting for confounders, testing for potential interactions, and ensuring no multicolinearity between covariates. Last, we computed the risk proportion of death attributable to AKI, indicating the proportion of deaths that could potentially be avoided if AKI and its complications were abrogated.²¹

We performed all statistical analyses with SAS Version 9.3 (SAS Institute) using designated weight values to produce weighted national estimates. The threshold for statistical significance was set at P < .01 (with ORs and 95% CIs that excluded 1).

Results

AKI Incidence, Risk Factors, and Trends

We identified 7,235,251 patients who underwent elective hip or knee arthroplasty for osteoarthritis between 2002 and 2012—an estimate consistent with data from the Centers for Disease Control and Prevention.²² Of that total, 94,367 (1.3%) had AKI. The proportion of discharges diagnosed with AKI increased rapidly over the decade, from 0.5% in 2002 to 1.8% to 1.9% in the period 2010–2012. This upward trend was highly significant (P_trend < .001) (Figure 1). Patients with AKI (vs patients without AKI) were more likely to be older (mean age, 70 vs 66 years; P < .001), male (50.8% vs 38.4%; P < .001), and black (10.07% vs 5.15%; P<. 001). They were also found to have a significantly higher comorbidity score (mean CCI, 2.8 vs 1.5; P < .001) and higher proportions of comorbidities, including hypertension, CKD, atrial fibrillation, diabetes mellitus (DM), congestive heart failure, chronic liver disease, and hepatitis C virus infection. In addition, AKI was associated with perioperative myocardial infarction (MI), sepsis, cardiac catheterization, and blood transfusion. Regarding socioeconomic characteristics, patients with AKI were more likely to have Medicare/Medicaid insurance (72.26% vs 58.06%; P < .001) and to belong to the extremes of income categories (Table 2).

Using multivariable logistic regression, we found that increased age (1.11 increase in adjusted OR for every year older; 95% CI, 1.09-1.14; P < .001), male sex (adjusted OR, 1.65; 95% CI, 1.60-1.71; P < .001), and black race (adjusted OR, 1.57; 95% CI, 1.45-1.69; P < .001) were significantly associated with postoperative AKI. Regarding comorbidities, baseline CKD (adjusted OR, 8.64; 95% CI, 8.14-9.18; P < .001) and congestive heart failure (adjusted OR, 2.74; 95% CI, 2.57-2.92; P< .0001) were most significantly associated with AKI. Perioperative events, including sepsis (adjusted OR, 35.64; 95% CI, 30.28-41.96; P < .0001), MI (adjusted OR, 6.14; 95% CI, 5.17-7.28; P < .0001), and blood transfusion (adjusted OR, 2.28; 95% CI, 2.15-2.42; P < .0001), were also strongly associated with postoperative AKI. Last, compared with urban hospitals and small hospital bed size, rural hospitals (adjusted OR, 0.70; 95% CI, 0.60-0.81; P< .001) and large bed size (adjusted OR, 0.82; 95% CI, 0.70-0.93; P = .003) were associated with lower probability of developing AKI (Table 3).

Figure 2 elucidates the frequency of AKI based on a combination of key preoperative comorbid conditions and postoperative complications—demonstrating that the proportion of AKI cases associated with other postoperative complications is significantly higher in the CKD and concomitant DM/CKD patient populations. Patients hospitalized with CKD exhibited higher rates of AKI in cases involving blood transfusion (20.9% vs 1.8%; P < .001), acute MI (48.9% vs 13.8%; P < .001), and sepsis (74.7% vs 36.3%;P< .001) relative to patients without CKD. Similarly, patients with concomitant DM/CKD exhibited higher rates of AKI in cases involving blood transfusion (23% vs 1.9%; P< .001), acute MI (51.1% vs 12.1%; P< .001), and sepsis (75% vs 38.2%; P < .001) relative to patients without either condition. However, patients hospitalized with DM alone exhibited only marginally higher rates of AKI in cases involving blood transfusion (4.7% vs 2%; P < .01) and acute MI (19.2% vs 16.7%; P< .01) and a lower rate in cases involving sepsis (38.2% vs 41.7%; P < .01) relative to patients without DM. These data suggest that CKD is the most significant clinically relevant risk factor for AKI and that CKD may synergize with DM to raise the risk for AKI.

Outcomes

We then analyzed the impact of AKI on hospital outcomes, including in-hospital mortality, discharge disposition, LOS, and cost of care. Mortality was significantly higher in patients with AKI than in patients without it (2.08% vs 0.06%; P < .001). Even after adjusting for confounders (eg, demographics, comorbidity burden, perioperative sepsis, hospital-level characteristics), AKI was still associated with strikingly higher odds of in-hospital death (adjusted OR, 11.32; 95% CI, 9.34-13.74; P < .001). However, analysis of temporal trends indicated that the odds for adjusted mortality associated with AKI decreased from 18.09 to 9.45 (P_trend = .01) over the period 2002–2012 (Figure 3). This decrease in odds of death was countered by an increase in incidence of AKI, resulting in a stable attributable risk proportion (97.9% in 2002 to 97.3% in 2012; P_trend = .90) (Table 4). Regarding discharge disposition, patients with AKI were much less likely to be discharged home (41.35% vs 62.59%; P < .001) and more likely to be discharged to long-term care (56.37% vs 37.03%; P< .001). After adjustment for confounders, AKI was associated with significantly increased odds of adverse discharge (adjusted OR, 2.24; 95% CI, 2.12-2.36; P< .001). Analysis of temporal trends revealed no appreciable decrease in the adjusted odds of adverse discharge between 2002 (adjusted OR, 1.87; 95% CI, 1.37-2.55; P < .001) and 2012 (adjusted OR, 1.93; 95% CI, 1.76-2.11; P < .001) (Figure 4, Table 5). Last, both mean LOS (5 days vs 3 days; P < .001) and mean cost of hospitalization (US $22,269 vs $15,757; P < .001) were significantly higher in patients with AKI.

Discussion

In this study, we found that the incidence of AKI among hospitalized patients increased 4-fold between 2002 and 2012. Moreover, we identified numerous patient-specific, hospital-specific, perioperative risk factors for AKI. Most important, we found that AKI was associated with a strikingly higher risk of in-hospital death, and surviving patients were more likely to experience adverse discharge. Although the adjusted mortality rate associated with AKI decreased over that decade, the attributable risk proportion remained stable.

Few studies have addressed this significant public health concern. In one recent study in Australia, Kimmel and colleagues¹⁶ identified risk factors for AKI but lacked data on AKI outcomes. In a study of complications and mortality occurring after orthopedic surgery, Belmont and colleagues²² categorized complications as either local or systemic but did not examine renal complications. Only 2 other major studies have been conducted on renal outcomes associated with major joint surgery, and both were limited to patients with acute hip fractures. The first included acute fracture surgery patients and omitted elective joint surgery patients, and it evaluated admission renal function but not postoperative AKI.²² The second study had a sample size of only 170 patients.²³ Thus, the literature leaves us with a crucial knowledge gap in renal outcomes and their postoperative impact in elective arthroplasties.

The present study filled this information gap by examining the incidence, risk factors, outcomes, and temporal trends of AKI after elective hip and knee arthroplasties. The increasing incidence of AKI in this surgical setting is similar to that of AKI in other surgical settings (cardiac and noncardiac).²¹ Although our analysis was limited by lack of perioperative management data, patients undergoing elective joint arthroplasty can experience kidney dysfunction for several reasons, including volume depletion, postoperative sepsis, and influence of medications, such as nonsteroidal anti-inflammatory drugs (NSAIDs), especially in older patients with more comorbidities and a higher burden of CKD. Each of these factors can cause renal dysfunction in patients having orthopedic procedures.²⁴ Moreover, NSAID use among elective joint arthroplasty patients is likely higher because of an emphasis on multimodal analgesia, as recent randomized controlled trials have demonstrated the efficacy of NSAID use in controlling pain without increasing bleeding.^25-27 Our results also demonstrated that the absolute incidence of AKI after orthopedic surgery is relatively low. One possible explanation for this phenomenon is that the definitions used were based on ICD-9-CM codes that underestimate the true incidence of AKI.

Consistent with other studies, we found that certain key preoperative comorbid conditions and postoperative events were associated with higher AKI risk. We stratified the rate of AKI associated with each postoperative event (sepsis, acute MI, cardiac catheterization, need for transfusion) by DM/CKD comorbidity. CKD was associated with significantly higher AKI risk across all postoperative complications. This information may provide clinicians with bedside information that can be used to determine which patients may be at higher or lower risk for AKI.

Our analysis of patient outcomes revealed that, though AKI was relatively uncommon, it increased the risk for death during hospitalization more than 10-fold between 2002 and 2012. Although the adjusted OR of in-hospital mortality decreased over the decade studied, the concurrent increase in AKI incidence caused the attributable risk of death associated with AKI to essentially remain the same. This observation is consistent with recent reports from cardiac surgery settings.²¹ These data together suggest that ameliorating occurrences of AKI would decrease mortality and increase quality of care for patients undergoing elective joint surgeries.

We also examined the effect of AKI on resource use by studying LOS, costs, and risk for adverse discharge. Much as in other surgical settings, AKI increased both LOS and overall hospitalization costs. More important, AKI was associated with increased adverse discharge (discharge to long-term care or nursing homes). Although exact reasons are unclear, we can speculate that postoperative renal dysfunction precludes early rehabilitation, impeding desired functional outcome and disposition.^28,29 Given the projected increases in primary and revision hip and knee arthroplasties,⁵ these data predict that the impact of AKI on health outcomes will increase alarmingly in coming years.

There are limitations to our study. First, it was based on administrative data and lacked patient-level and laboratory data. As reported, the sensitivity of AKI codes remains moderate,³⁰ so the true burden may be higher than indicated here. As the definition of AKI was based on administrative coding, we also could not estimate severity, though previous studies have found that administrative codes typically capture a more severe form of disease.³¹ Another limitation is that, because the data were deidentified, we could not delineate the risk for recurrent AKI in repeated surgical procedures, though this cohort unlikely was large enough to qualitatively affect our results. The third limitation is that, though we used CCI to adjust for the comorbidity burden, we were unable to account for other unmeasured confounders associated with increased AKI incidence, such as specific medication use. In addition, given the lack of patient-level data, we could not analyze the specific factors responsible for AKI in the perioperative period. Nevertheless, the strengths of a nationally representative sample, such as large sample size and generalizability, outweigh these limitations.

Conclusion

AKI is potentially an important quality indicator of elective joint surgery, and reducing its incidence is therefore essential for quality improvement. Given that hip and knee arthroplasties are projected to increase exponentially, as is the burden of comorbid conditions in this population, postoperative AKI will continue to have an incremental impact on health and health care resources. Thus, a carefully planned approach of interdisciplinary perioperative care is warranted to reduce both the risk and the consequences of this devastating condition.

Degenerative arthritis is a widespread chronic condition with an incidence of almost 43 million and annual health care costs of $60 billion in the United States alone.1 Although many cases can be managed symptomatically with medical therapy and intra-articular injections,2 many patients experience disease progression resulting in decreased ambulatory ability and work productivity. For these patients, elective hip and knee arthroplasties can drastically improve quality of life and functionality.3,4 Over the past decade, there has been a marked increase in the number of primary and revision total hip and knee arthroplasties performed in the United States. By 2030, the demand for primary total hip arthroplasties will grow an estimated 174%, to 572,000 procedures. Likewise, the demand for primary total knee arthroplasties is projected to grow by 673%, to 3.48 million procedures.5 However, though better surgical techniques and technology have led to improved functional outcomes, there is still substantial risk for complications in the perioperative period, especially in the geriatric population, in which substantial comorbidities are common.6-9

Acute kidney injury (AKI) is a common public health problem in hospitalized patients and in patients undergoing procedures. More than one-third of all AKI cases occur in surgical settings.^10,11 Over the past decade, both community-acquired and in-hospital AKIs rapidly increased in incidence in all major clinical settings.^12-14 Patients with AKI have high rates of adverse outcomes during hospitalization and discharge.^11,15 Sequelae of AKIs include worsening chronic kidney disease (CKD) and progression to end-stage renal disease, necessitating either long-term dialysis or transplantation.¹² This in turn leads to exacerbated disability, diminished quality of life, and disproportionate burden on health care resources.

Much of our knowledge about postoperative AKI has been derived from cardiovascular, thoracic, and abdominal surgery settings. However, there is a paucity of data on epidemiology and trends for either AKI or associated outcomes in patients undergoing major orthopedic surgery. The few studies to date either were single-center or had inadequate sample sizes for appropriately powered analysis of the risk factors and outcomes related to AKI.¹⁶

In the study reported here, we analyzed a large cohort of patients from a nationwide multicenter database to determine the incidence of and risk factors for AKI. We also examined the mortality and adverse discharges associated with AKI after major joint surgery. Lastly, we assessed temporal trends in both incidence and outcomes of AKI, including the death risk attributable to AKI.

Methods

Database

We extracted our study cohort from the Nationwide Inpatient Sample (NIS) and the National Inpatient Sample of Healthcare Cost and Utilization Project (HCUP) compiled by the Agency for Healthcare Research and Quality.¹⁷ NIS, the largest inpatient care database in the United States, stores data from almost 8 million stays in about 1000 hospitals across the country each year. Its participating hospital pool consists of about 20% of US community hospitals, resulting in a sampling frame comprising about 90% of all hospital discharges in the United States. This allows for calculation of precise, weighted nationwide estimates. Data elements within NIS are drawn from hospital discharge abstracts that indicate all procedures performed. NIS also stores information on patient characteristics, length of stay (LOS), discharge disposition, postoperative morbidity, and observed in-hospital mortality. However, it stores no information on long-term follow-up or complications after discharge.

Data Analysis

For the period 2002–2012, we queried the NIS database for hip and knee arthroplasties with primary diagnosis codes for osteoarthritis and secondary codes for AKI. We excluded patients under age 18 years and patients with diagnosis codes for hip and knee fracture/necrosis, inflammatory/infectious arthritis, or bone neoplasms (Table 1). We then extracted baseline characteristics of the study population. Patient-level characteristics included age, sex, race, quartile classification of median household income according to postal (ZIP) code, and primary payer (Medicare/Medicaid, private insurance, self-pay, no charge). Hospital-level characteristics included hospital location (urban, rural), hospital bed size (small, medium, large), region (Northeast, Midwest/North Central, South, West), and teaching status. We defined illness severity and likelihood of death using Deyo’s modification of the Charlson Comorbidity Index (CCI), which draws on principal and secondary ICD-9-CM (International Classification of Diseases, Ninth Revision-Clinical Modification) diagnosis codes, procedure codes, and patient demographics to estimate a patient’s mortality risk. This method reliably predicts mortality and readmission in the orthopedic population.^18,19 We assessed the effect of AKI on 4 outcomes, including in-hospital mortality, discharge disposition, LOS, and cost of stay. Discharge disposition was grouped by either (a) home or short-term facility or (b) adverse discharge. Home or short-term facility covered routine, short-term hospital, against medical advice, home intravenous provider, another rehabilitation facility, another institution for outpatient services, institution for outpatient services, discharged alive, and destination unknown; adverse discharge covered skilled nursing facility, intermediate care, hospice home, hospice medical facility, long-term care hospital, and certified nursing facility. This dichotomization of discharge disposition is often used in studies of NIS data.²⁰

Statistical Analyses

We compared the baseline characteristics of hospitalized patients with and without AKI. To test for significance, we used the χ² test for categorical variables, the Student t test for normally distributed continuous variables, the Wilcoxon rank sum test for non-normally distributed continuous variables, and the Cochran-Armitage test for trends in AKI incidence. We used survey logistic regression models to calculate adjusted odds ratios (ORs) with 95% confidence intervals (95% CIs) in order to estimate the predictors of AKI and the impact of AKI on hospital outcomes. We constructed final models after adjusting for confounders, testing for potential interactions, and ensuring no multicolinearity between covariates. Last, we computed the risk proportion of death attributable to AKI, indicating the proportion of deaths that could potentially be avoided if AKI and its complications were abrogated.²¹

We performed all statistical analyses with SAS Version 9.3 (SAS Institute) using designated weight values to produce weighted national estimates. The threshold for statistical significance was set at P < .01 (with ORs and 95% CIs that excluded 1).

Results

AKI Incidence, Risk Factors, and Trends

We identified 7,235,251 patients who underwent elective hip or knee arthroplasty for osteoarthritis between 2002 and 2012—an estimate consistent with data from the Centers for Disease Control and Prevention.²² Of that total, 94,367 (1.3%) had AKI. The proportion of discharges diagnosed with AKI increased rapidly over the decade, from 0.5% in 2002 to 1.8% to 1.9% in the period 2010–2012. This upward trend was highly significant (P_trend < .001) (Figure 1). Patients with AKI (vs patients without AKI) were more likely to be older (mean age, 70 vs 66 years; P < .001), male (50.8% vs 38.4%; P < .001), and black (10.07% vs 5.15%; P<. 001). They were also found to have a significantly higher comorbidity score (mean CCI, 2.8 vs 1.5; P < .001) and higher proportions of comorbidities, including hypertension, CKD, atrial fibrillation, diabetes mellitus (DM), congestive heart failure, chronic liver disease, and hepatitis C virus infection. In addition, AKI was associated with perioperative myocardial infarction (MI), sepsis, cardiac catheterization, and blood transfusion. Regarding socioeconomic characteristics, patients with AKI were more likely to have Medicare/Medicaid insurance (72.26% vs 58.06%; P < .001) and to belong to the extremes of income categories (Table 2).

Using multivariable logistic regression, we found that increased age (1.11 increase in adjusted OR for every year older; 95% CI, 1.09-1.14; P < .001), male sex (adjusted OR, 1.65; 95% CI, 1.60-1.71; P < .001), and black race (adjusted OR, 1.57; 95% CI, 1.45-1.69; P < .001) were significantly associated with postoperative AKI. Regarding comorbidities, baseline CKD (adjusted OR, 8.64; 95% CI, 8.14-9.18; P < .001) and congestive heart failure (adjusted OR, 2.74; 95% CI, 2.57-2.92; P< .0001) were most significantly associated with AKI. Perioperative events, including sepsis (adjusted OR, 35.64; 95% CI, 30.28-41.96; P < .0001), MI (adjusted OR, 6.14; 95% CI, 5.17-7.28; P < .0001), and blood transfusion (adjusted OR, 2.28; 95% CI, 2.15-2.42; P < .0001), were also strongly associated with postoperative AKI. Last, compared with urban hospitals and small hospital bed size, rural hospitals (adjusted OR, 0.70; 95% CI, 0.60-0.81; P< .001) and large bed size (adjusted OR, 0.82; 95% CI, 0.70-0.93; P = .003) were associated with lower probability of developing AKI (Table 3).

Figure 2 elucidates the frequency of AKI based on a combination of key preoperative comorbid conditions and postoperative complications—demonstrating that the proportion of AKI cases associated with other postoperative complications is significantly higher in the CKD and concomitant DM/CKD patient populations. Patients hospitalized with CKD exhibited higher rates of AKI in cases involving blood transfusion (20.9% vs 1.8%; P < .001), acute MI (48.9% vs 13.8%; P < .001), and sepsis (74.7% vs 36.3%;P< .001) relative to patients without CKD. Similarly, patients with concomitant DM/CKD exhibited higher rates of AKI in cases involving blood transfusion (23% vs 1.9%; P< .001), acute MI (51.1% vs 12.1%; P< .001), and sepsis (75% vs 38.2%; P < .001) relative to patients without either condition. However, patients hospitalized with DM alone exhibited only marginally higher rates of AKI in cases involving blood transfusion (4.7% vs 2%; P < .01) and acute MI (19.2% vs 16.7%; P< .01) and a lower rate in cases involving sepsis (38.2% vs 41.7%; P < .01) relative to patients without DM. These data suggest that CKD is the most significant clinically relevant risk factor for AKI and that CKD may synergize with DM to raise the risk for AKI.

Outcomes

We then analyzed the impact of AKI on hospital outcomes, including in-hospital mortality, discharge disposition, LOS, and cost of care. Mortality was significantly higher in patients with AKI than in patients without it (2.08% vs 0.06%; P < .001). Even after adjusting for confounders (eg, demographics, comorbidity burden, perioperative sepsis, hospital-level characteristics), AKI was still associated with strikingly higher odds of in-hospital death (adjusted OR, 11.32; 95% CI, 9.34-13.74; P < .001). However, analysis of temporal trends indicated that the odds for adjusted mortality associated with AKI decreased from 18.09 to 9.45 (P_trend = .01) over the period 2002–2012 (Figure 3). This decrease in odds of death was countered by an increase in incidence of AKI, resulting in a stable attributable risk proportion (97.9% in 2002 to 97.3% in 2012; P_trend = .90) (Table 4). Regarding discharge disposition, patients with AKI were much less likely to be discharged home (41.35% vs 62.59%; P < .001) and more likely to be discharged to long-term care (56.37% vs 37.03%; P< .001). After adjustment for confounders, AKI was associated with significantly increased odds of adverse discharge (adjusted OR, 2.24; 95% CI, 2.12-2.36; P< .001). Analysis of temporal trends revealed no appreciable decrease in the adjusted odds of adverse discharge between 2002 (adjusted OR, 1.87; 95% CI, 1.37-2.55; P < .001) and 2012 (adjusted OR, 1.93; 95% CI, 1.76-2.11; P < .001) (Figure 4, Table 5). Last, both mean LOS (5 days vs 3 days; P < .001) and mean cost of hospitalization (US $22,269 vs $15,757; P < .001) were significantly higher in patients with AKI.

Discussion

In this study, we found that the incidence of AKI among hospitalized patients increased 4-fold between 2002 and 2012. Moreover, we identified numerous patient-specific, hospital-specific, perioperative risk factors for AKI. Most important, we found that AKI was associated with a strikingly higher risk of in-hospital death, and surviving patients were more likely to experience adverse discharge. Although the adjusted mortality rate associated with AKI decreased over that decade, the attributable risk proportion remained stable.

Few studies have addressed this significant public health concern. In one recent study in Australia, Kimmel and colleagues¹⁶ identified risk factors for AKI but lacked data on AKI outcomes. In a study of complications and mortality occurring after orthopedic surgery, Belmont and colleagues²² categorized complications as either local or systemic but did not examine renal complications. Only 2 other major studies have been conducted on renal outcomes associated with major joint surgery, and both were limited to patients with acute hip fractures. The first included acute fracture surgery patients and omitted elective joint surgery patients, and it evaluated admission renal function but not postoperative AKI.²² The second study had a sample size of only 170 patients.²³ Thus, the literature leaves us with a crucial knowledge gap in renal outcomes and their postoperative impact in elective arthroplasties.

The present study filled this information gap by examining the incidence, risk factors, outcomes, and temporal trends of AKI after elective hip and knee arthroplasties. The increasing incidence of AKI in this surgical setting is similar to that of AKI in other surgical settings (cardiac and noncardiac).²¹ Although our analysis was limited by lack of perioperative management data, patients undergoing elective joint arthroplasty can experience kidney dysfunction for several reasons, including volume depletion, postoperative sepsis, and influence of medications, such as nonsteroidal anti-inflammatory drugs (NSAIDs), especially in older patients with more comorbidities and a higher burden of CKD. Each of these factors can cause renal dysfunction in patients having orthopedic procedures.²⁴ Moreover, NSAID use among elective joint arthroplasty patients is likely higher because of an emphasis on multimodal analgesia, as recent randomized controlled trials have demonstrated the efficacy of NSAID use in controlling pain without increasing bleeding.^25-27 Our results also demonstrated that the absolute incidence of AKI after orthopedic surgery is relatively low. One possible explanation for this phenomenon is that the definitions used were based on ICD-9-CM codes that underestimate the true incidence of AKI.

Consistent with other studies, we found that certain key preoperative comorbid conditions and postoperative events were associated with higher AKI risk. We stratified the rate of AKI associated with each postoperative event (sepsis, acute MI, cardiac catheterization, need for transfusion) by DM/CKD comorbidity. CKD was associated with significantly higher AKI risk across all postoperative complications. This information may provide clinicians with bedside information that can be used to determine which patients may be at higher or lower risk for AKI.

Our analysis of patient outcomes revealed that, though AKI was relatively uncommon, it increased the risk for death during hospitalization more than 10-fold between 2002 and 2012. Although the adjusted OR of in-hospital mortality decreased over the decade studied, the concurrent increase in AKI incidence caused the attributable risk of death associated with AKI to essentially remain the same. This observation is consistent with recent reports from cardiac surgery settings.²¹ These data together suggest that ameliorating occurrences of AKI would decrease mortality and increase quality of care for patients undergoing elective joint surgeries.

We also examined the effect of AKI on resource use by studying LOS, costs, and risk for adverse discharge. Much as in other surgical settings, AKI increased both LOS and overall hospitalization costs. More important, AKI was associated with increased adverse discharge (discharge to long-term care or nursing homes). Although exact reasons are unclear, we can speculate that postoperative renal dysfunction precludes early rehabilitation, impeding desired functional outcome and disposition.^28,29 Given the projected increases in primary and revision hip and knee arthroplasties,⁵ these data predict that the impact of AKI on health outcomes will increase alarmingly in coming years.

There are limitations to our study. First, it was based on administrative data and lacked patient-level and laboratory data. As reported, the sensitivity of AKI codes remains moderate,³⁰ so the true burden may be higher than indicated here. As the definition of AKI was based on administrative coding, we also could not estimate severity, though previous studies have found that administrative codes typically capture a more severe form of disease.³¹ Another limitation is that, because the data were deidentified, we could not delineate the risk for recurrent AKI in repeated surgical procedures, though this cohort unlikely was large enough to qualitatively affect our results. The third limitation is that, though we used CCI to adjust for the comorbidity burden, we were unable to account for other unmeasured confounders associated with increased AKI incidence, such as specific medication use. In addition, given the lack of patient-level data, we could not analyze the specific factors responsible for AKI in the perioperative period. Nevertheless, the strengths of a nationally representative sample, such as large sample size and generalizability, outweigh these limitations.

Conclusion

AKI is potentially an important quality indicator of elective joint surgery, and reducing its incidence is therefore essential for quality improvement. Given that hip and knee arthroplasties are projected to increase exponentially, as is the burden of comorbid conditions in this population, postoperative AKI will continue to have an incremental impact on health and health care resources. Thus, a carefully planned approach of interdisciplinary perioperative care is warranted to reduce both the risk and the consequences of this devastating condition.

Degenerative arthritis is a widespread chronic condition with an incidence of almost 43 million and annual health care costs of $60 billion in the United States alone.1 Although many cases can be managed symptomatically with medical therapy and intra-articular injections,2 many patients experience disease progression resulting in decreased ambulatory ability and work productivity. For these patients, elective hip and knee arthroplasties can drastically improve quality of life and functionality.3,4 Over the past decade, there has been a marked increase in the number of primary and revision total hip and knee arthroplasties performed in the United States. By 2030, the demand for primary total hip arthroplasties will grow an estimated 174%, to 572,000 procedures. Likewise, the demand for primary total knee arthroplasties is projected to grow by 673%, to 3.48 million procedures.5 However, though better surgical techniques and technology have led to improved functional outcomes, there is still substantial risk for complications in the perioperative period, especially in the geriatric population, in which substantial comorbidities are common.6-9

Acute kidney injury (AKI) is a common public health problem in hospitalized patients and in patients undergoing procedures. More than one-third of all AKI cases occur in surgical settings.^10,11 Over the past decade, both community-acquired and in-hospital AKIs rapidly increased in incidence in all major clinical settings.^12-14 Patients with AKI have high rates of adverse outcomes during hospitalization and discharge.^11,15 Sequelae of AKIs include worsening chronic kidney disease (CKD) and progression to end-stage renal disease, necessitating either long-term dialysis or transplantation.¹² This in turn leads to exacerbated disability, diminished quality of life, and disproportionate burden on health care resources.

Much of our knowledge about postoperative AKI has been derived from cardiovascular, thoracic, and abdominal surgery settings. However, there is a paucity of data on epidemiology and trends for either AKI or associated outcomes in patients undergoing major orthopedic surgery. The few studies to date either were single-center or had inadequate sample sizes for appropriately powered analysis of the risk factors and outcomes related to AKI.¹⁶

In the study reported here, we analyzed a large cohort of patients from a nationwide multicenter database to determine the incidence of and risk factors for AKI. We also examined the mortality and adverse discharges associated with AKI after major joint surgery. Lastly, we assessed temporal trends in both incidence and outcomes of AKI, including the death risk attributable to AKI.

Methods

Database

We extracted our study cohort from the Nationwide Inpatient Sample (NIS) and the National Inpatient Sample of Healthcare Cost and Utilization Project (HCUP) compiled by the Agency for Healthcare Research and Quality.¹⁷ NIS, the largest inpatient care database in the United States, stores data from almost 8 million stays in about 1000 hospitals across the country each year. Its participating hospital pool consists of about 20% of US community hospitals, resulting in a sampling frame comprising about 90% of all hospital discharges in the United States. This allows for calculation of precise, weighted nationwide estimates. Data elements within NIS are drawn from hospital discharge abstracts that indicate all procedures performed. NIS also stores information on patient characteristics, length of stay (LOS), discharge disposition, postoperative morbidity, and observed in-hospital mortality. However, it stores no information on long-term follow-up or complications after discharge.

Data Analysis

For the period 2002–2012, we queried the NIS database for hip and knee arthroplasties with primary diagnosis codes for osteoarthritis and secondary codes for AKI. We excluded patients under age 18 years and patients with diagnosis codes for hip and knee fracture/necrosis, inflammatory/infectious arthritis, or bone neoplasms (Table 1). We then extracted baseline characteristics of the study population. Patient-level characteristics included age, sex, race, quartile classification of median household income according to postal (ZIP) code, and primary payer (Medicare/Medicaid, private insurance, self-pay, no charge). Hospital-level characteristics included hospital location (urban, rural), hospital bed size (small, medium, large), region (Northeast, Midwest/North Central, South, West), and teaching status. We defined illness severity and likelihood of death using Deyo’s modification of the Charlson Comorbidity Index (CCI), which draws on principal and secondary ICD-9-CM (International Classification of Diseases, Ninth Revision-Clinical Modification) diagnosis codes, procedure codes, and patient demographics to estimate a patient’s mortality risk. This method reliably predicts mortality and readmission in the orthopedic population.^18,19 We assessed the effect of AKI on 4 outcomes, including in-hospital mortality, discharge disposition, LOS, and cost of stay. Discharge disposition was grouped by either (a) home or short-term facility or (b) adverse discharge. Home or short-term facility covered routine, short-term hospital, against medical advice, home intravenous provider, another rehabilitation facility, another institution for outpatient services, institution for outpatient services, discharged alive, and destination unknown; adverse discharge covered skilled nursing facility, intermediate care, hospice home, hospice medical facility, long-term care hospital, and certified nursing facility. This dichotomization of discharge disposition is often used in studies of NIS data.²⁰

Statistical Analyses

We compared the baseline characteristics of hospitalized patients with and without AKI. To test for significance, we used the χ² test for categorical variables, the Student t test for normally distributed continuous variables, the Wilcoxon rank sum test for non-normally distributed continuous variables, and the Cochran-Armitage test for trends in AKI incidence. We used survey logistic regression models to calculate adjusted odds ratios (ORs) with 95% confidence intervals (95% CIs) in order to estimate the predictors of AKI and the impact of AKI on hospital outcomes. We constructed final models after adjusting for confounders, testing for potential interactions, and ensuring no multicolinearity between covariates. Last, we computed the risk proportion of death attributable to AKI, indicating the proportion of deaths that could potentially be avoided if AKI and its complications were abrogated.²¹

We performed all statistical analyses with SAS Version 9.3 (SAS Institute) using designated weight values to produce weighted national estimates. The threshold for statistical significance was set at P < .01 (with ORs and 95% CIs that excluded 1).

Results

AKI Incidence, Risk Factors, and Trends

We identified 7,235,251 patients who underwent elective hip or knee arthroplasty for osteoarthritis between 2002 and 2012—an estimate consistent with data from the Centers for Disease Control and Prevention.²² Of that total, 94,367 (1.3%) had AKI. The proportion of discharges diagnosed with AKI increased rapidly over the decade, from 0.5% in 2002 to 1.8% to 1.9% in the period 2010–2012. This upward trend was highly significant (P_trend < .001) (Figure 1). Patients with AKI (vs patients without AKI) were more likely to be older (mean age, 70 vs 66 years; P < .001), male (50.8% vs 38.4%; P < .001), and black (10.07% vs 5.15%; P<. 001). They were also found to have a significantly higher comorbidity score (mean CCI, 2.8 vs 1.5; P < .001) and higher proportions of comorbidities, including hypertension, CKD, atrial fibrillation, diabetes mellitus (DM), congestive heart failure, chronic liver disease, and hepatitis C virus infection. In addition, AKI was associated with perioperative myocardial infarction (MI), sepsis, cardiac catheterization, and blood transfusion. Regarding socioeconomic characteristics, patients with AKI were more likely to have Medicare/Medicaid insurance (72.26% vs 58.06%; P < .001) and to belong to the extremes of income categories (Table 2).

Using multivariable logistic regression, we found that increased age (1.11 increase in adjusted OR for every year older; 95% CI, 1.09-1.14; P < .001), male sex (adjusted OR, 1.65; 95% CI, 1.60-1.71; P < .001), and black race (adjusted OR, 1.57; 95% CI, 1.45-1.69; P < .001) were significantly associated with postoperative AKI. Regarding comorbidities, baseline CKD (adjusted OR, 8.64; 95% CI, 8.14-9.18; P < .001) and congestive heart failure (adjusted OR, 2.74; 95% CI, 2.57-2.92; P< .0001) were most significantly associated with AKI. Perioperative events, including sepsis (adjusted OR, 35.64; 95% CI, 30.28-41.96; P < .0001), MI (adjusted OR, 6.14; 95% CI, 5.17-7.28; P < .0001), and blood transfusion (adjusted OR, 2.28; 95% CI, 2.15-2.42; P < .0001), were also strongly associated with postoperative AKI. Last, compared with urban hospitals and small hospital bed size, rural hospitals (adjusted OR, 0.70; 95% CI, 0.60-0.81; P< .001) and large bed size (adjusted OR, 0.82; 95% CI, 0.70-0.93; P = .003) were associated with lower probability of developing AKI (Table 3).

Figure 2 elucidates the frequency of AKI based on a combination of key preoperative comorbid conditions and postoperative complications—demonstrating that the proportion of AKI cases associated with other postoperative complications is significantly higher in the CKD and concomitant DM/CKD patient populations. Patients hospitalized with CKD exhibited higher rates of AKI in cases involving blood transfusion (20.9% vs 1.8%; P < .001), acute MI (48.9% vs 13.8%; P < .001), and sepsis (74.7% vs 36.3%;P< .001) relative to patients without CKD. Similarly, patients with concomitant DM/CKD exhibited higher rates of AKI in cases involving blood transfusion (23% vs 1.9%; P< .001), acute MI (51.1% vs 12.1%; P< .001), and sepsis (75% vs 38.2%; P < .001) relative to patients without either condition. However, patients hospitalized with DM alone exhibited only marginally higher rates of AKI in cases involving blood transfusion (4.7% vs 2%; P < .01) and acute MI (19.2% vs 16.7%; P< .01) and a lower rate in cases involving sepsis (38.2% vs 41.7%; P < .01) relative to patients without DM. These data suggest that CKD is the most significant clinically relevant risk factor for AKI and that CKD may synergize with DM to raise the risk for AKI.

Outcomes

We then analyzed the impact of AKI on hospital outcomes, including in-hospital mortality, discharge disposition, LOS, and cost of care. Mortality was significantly higher in patients with AKI than in patients without it (2.08% vs 0.06%; P < .001). Even after adjusting for confounders (eg, demographics, comorbidity burden, perioperative sepsis, hospital-level characteristics), AKI was still associated with strikingly higher odds of in-hospital death (adjusted OR, 11.32; 95% CI, 9.34-13.74; P < .001). However, analysis of temporal trends indicated that the odds for adjusted mortality associated with AKI decreased from 18.09 to 9.45 (P_trend = .01) over the period 2002–2012 (Figure 3). This decrease in odds of death was countered by an increase in incidence of AKI, resulting in a stable attributable risk proportion (97.9% in 2002 to 97.3% in 2012; P_trend = .90) (Table 4). Regarding discharge disposition, patients with AKI were much less likely to be discharged home (41.35% vs 62.59%; P < .001) and more likely to be discharged to long-term care (56.37% vs 37.03%; P< .001). After adjustment for confounders, AKI was associated with significantly increased odds of adverse discharge (adjusted OR, 2.24; 95% CI, 2.12-2.36; P< .001). Analysis of temporal trends revealed no appreciable decrease in the adjusted odds of adverse discharge between 2002 (adjusted OR, 1.87; 95% CI, 1.37-2.55; P < .001) and 2012 (adjusted OR, 1.93; 95% CI, 1.76-2.11; P < .001) (Figure 4, Table 5). Last, both mean LOS (5 days vs 3 days; P < .001) and mean cost of hospitalization (US $22,269 vs $15,757; P < .001) were significantly higher in patients with AKI.

Discussion

In this study, we found that the incidence of AKI among hospitalized patients increased 4-fold between 2002 and 2012. Moreover, we identified numerous patient-specific, hospital-specific, perioperative risk factors for AKI. Most important, we found that AKI was associated with a strikingly higher risk of in-hospital death, and surviving patients were more likely to experience adverse discharge. Although the adjusted mortality rate associated with AKI decreased over that decade, the attributable risk proportion remained stable.

Few studies have addressed this significant public health concern. In one recent study in Australia, Kimmel and colleagues¹⁶ identified risk factors for AKI but lacked data on AKI outcomes. In a study of complications and mortality occurring after orthopedic surgery, Belmont and colleagues²² categorized complications as either local or systemic but did not examine renal complications. Only 2 other major studies have been conducted on renal outcomes associated with major joint surgery, and both were limited to patients with acute hip fractures. The first included acute fracture surgery patients and omitted elective joint surgery patients, and it evaluated admission renal function but not postoperative AKI.²² The second study had a sample size of only 170 patients.²³ Thus, the literature leaves us with a crucial knowledge gap in renal outcomes and their postoperative impact in elective arthroplasties.

The present study filled this information gap by examining the incidence, risk factors, outcomes, and temporal trends of AKI after elective hip and knee arthroplasties. The increasing incidence of AKI in this surgical setting is similar to that of AKI in other surgical settings (cardiac and noncardiac).²¹ Although our analysis was limited by lack of perioperative management data, patients undergoing elective joint arthroplasty can experience kidney dysfunction for several reasons, including volume depletion, postoperative sepsis, and influence of medications, such as nonsteroidal anti-inflammatory drugs (NSAIDs), especially in older patients with more comorbidities and a higher burden of CKD. Each of these factors can cause renal dysfunction in patients having orthopedic procedures.²⁴ Moreover, NSAID use among elective joint arthroplasty patients is likely higher because of an emphasis on multimodal analgesia, as recent randomized controlled trials have demonstrated the efficacy of NSAID use in controlling pain without increasing bleeding.^25-27 Our results also demonstrated that the absolute incidence of AKI after orthopedic surgery is relatively low. One possible explanation for this phenomenon is that the definitions used were based on ICD-9-CM codes that underestimate the true incidence of AKI.

Consistent with other studies, we found that certain key preoperative comorbid conditions and postoperative events were associated with higher AKI risk. We stratified the rate of AKI associated with each postoperative event (sepsis, acute MI, cardiac catheterization, need for transfusion) by DM/CKD comorbidity. CKD was associated with significantly higher AKI risk across all postoperative complications. This information may provide clinicians with bedside information that can be used to determine which patients may be at higher or lower risk for AKI.

Our analysis of patient outcomes revealed that, though AKI was relatively uncommon, it increased the risk for death during hospitalization more than 10-fold between 2002 and 2012. Although the adjusted OR of in-hospital mortality decreased over the decade studied, the concurrent increase in AKI incidence caused the attributable risk of death associated with AKI to essentially remain the same. This observation is consistent with recent reports from cardiac surgery settings.²¹ These data together suggest that ameliorating occurrences of AKI would decrease mortality and increase quality of care for patients undergoing elective joint surgeries.

We also examined the effect of AKI on resource use by studying LOS, costs, and risk for adverse discharge. Much as in other surgical settings, AKI increased both LOS and overall hospitalization costs. More important, AKI was associated with increased adverse discharge (discharge to long-term care or nursing homes). Although exact reasons are unclear, we can speculate that postoperative renal dysfunction precludes early rehabilitation, impeding desired functional outcome and disposition.^28,29 Given the projected increases in primary and revision hip and knee arthroplasties,⁵ these data predict that the impact of AKI on health outcomes will increase alarmingly in coming years.

There are limitations to our study. First, it was based on administrative data and lacked patient-level and laboratory data. As reported, the sensitivity of AKI codes remains moderate,³⁰ so the true burden may be higher than indicated here. As the definition of AKI was based on administrative coding, we also could not estimate severity, though previous studies have found that administrative codes typically capture a more severe form of disease.³¹ Another limitation is that, because the data were deidentified, we could not delineate the risk for recurrent AKI in repeated surgical procedures, though this cohort unlikely was large enough to qualitatively affect our results. The third limitation is that, though we used CCI to adjust for the comorbidity burden, we were unable to account for other unmeasured confounders associated with increased AKI incidence, such as specific medication use. In addition, given the lack of patient-level data, we could not analyze the specific factors responsible for AKI in the perioperative period. Nevertheless, the strengths of a nationally representative sample, such as large sample size and generalizability, outweigh these limitations.

Conclusion

AKI is potentially an important quality indicator of elective joint surgery, and reducing its incidence is therefore essential for quality improvement. Given that hip and knee arthroplasties are projected to increase exponentially, as is the burden of comorbid conditions in this population, postoperative AKI will continue to have an incremental impact on health and health care resources. Thus, a carefully planned approach of interdisciplinary perioperative care is warranted to reduce both the risk and the consequences of this devastating condition.

Medial-compartment partial knee arthroplasty (unicompartmental replacement) is an accepted surgical intervention for anteromedial osteoarthritis of the knee.1 The radiographic investigations required in the workup of these patients should include weight-bearing standing anteroposterior (AP), lateral, and sunrise (Merchant) views, as well as a valgus stress AP radiograph to assess the functionality of the lateral compartment. The method of properly obtaining the valgus stress film has been well described by the Oxford Group.2 Its recommended radiographic technique requires that a surgeon or a radiologic technologist perform the valgus stress maneuver, manually, while another technologist shoots the film. The 2 consequences of this technique are that it requires 2 individuals to obtain the film, and it subjects the individual who is applying the stress to some level of radiation exposure, which is undesirable. Because of this and the time inconvenience, many surgeons omit the valgus stress radiograph, which can lead to the adverse outcome of missing a lateral compartment that is functionally incompetent, resulting in the potential for early lateral compartment progression of osteoarthritis and the need for revision surgery, usually to a total knee arthroplasty.

In an attempt to mitigate these barriers to obtaining the necessary valgus stress radiograph, Dr. Mauerhan’s team developed a technique that could be done with the assistance of the patient and would require only 1 technologist to perform. Additionally, this project was a quality improvement initiative, because it lowered radiation exposure to all personnel involved in obtaining the correct films.

Materials and Methods

We initiated the project using weight-bearing strategies to impart the valgus stress view of the knee. After trying several different wedges and blocks, and varying patient instructions, we realized a different approach to this problem would be required to find an acceptable solution. We redirected our efforts to effectively performing the stress view with the patient in a supine position on the radiograph table. Ultimately, we decided that a much stiffer wedge and a denser object to squeeze would facilitate obtaining a proper film. Considering all available options, a youth size 4 soccer ball (diameter, 11 in) was introduced along with a slightly larger positioning wedge. The soccer ball was wrapped with 4-in Coban wrap (3M) to create a nonslip surface. This change in patient positioning, along with a standardized 7º to 10º cephalic radiographic tube angulation, helped to correct issues with tibial plateau visualization. Once these changes were enacted, we obtained fairly consistent positive results, and we instituted this patient-directed valgus stress view of the knee, along with a manual valgus stress view for comparison.

The protocol for obtaining the patient-directed valgus stress view of the knee is as follows: The patient lays supine with a dense 45º spine-positioning wedge (Burlington Medical Supplies) placed under both knees and the patient’s heels on the examining table. The radiographic tube is angled cephalad 7º to 10º centered on the inferior pole of the patella, using a 40-in source to image-receptor distance, collimated to part; the image receptor is placed under the affected knee, below the positioning wedge. The affected knee is rotated to the “true” AP position (the patella will be centered between the femoral condyles on the AP exposure), and the ball is placed between the patient’s legs just above the ankle joint. The technologist demonstrates to the patient how to squeeze the ball while maintaining contact of heels with the table. The technologist can exit the room and obtain the exposure, which is taken while the patient is squeezing the ball, as shown in Figures 1A and 1B. Examples of the standing AP, manual stress, and patient-directed valgus radiographs are shown in Figures 2A-2C. The entire technique is demonstrated in the Video.

Vidyard Video

 

 

Results

During the 9 months of this quality improvement project, 78 examinations were performed. Five studies did not show complete correction of the varus deformity. Of these, 3 showed complete correction on a manual valgus stress radiograph, and 2 did not, contraindicating the use of partial knee replacement. Three patients displayed collapse of the lateral compartment, indicating a nonfunctional lateral compartment, and, therefore, were also a contraindication to partial knee arthroplasty. The remaining 70 patients had identical radiographic results with both the manual and patient-directed valgus stress tests. There was no instance of examination failure or need to repeat as a result of difficulty of the examination for the patient. Repeat films because of positioning errors were very rare, usually early in the learning curve, and no more prevalent than when using the manual stress method. The technique was reproducible and easy to teach and adopt.

 

 

Discussion

In total, 73 patients (93.5%) with the patient-directed stress film showed the desired result, either correction of the medial compartment narrowing in conjunction with an intact lateral compartment or narrowing of the lateral compartment. Of the 5 patients (6.5%) whose patient-directed stress films did not show correction of the varus deformity, 3 patients displayed correction with a manually applied stress radiograph and 2 did not. Based on this observation, our recommendation would be for those patients who do not show adequate correction on the patient-directed stress radiograph to have a manual examination to establish the presence or absence of the desired correction.

Performing a valgus stress radiograph is an integral part of the investigation to determine if the patient is an appropriate candidate for partial knee arthroplasty.3 The historical, manually performed valgus stress radiograph requires 2 individuals, 1 to apply the stress with the patient on the table and 1 to shoot the exposure. For the individual or individuals applying this stress, there is an increased radiation exposure that would be undesirable over a long career. The authors developed a new technique using a commercially available spinal positioning wedge and 11-in youth soccer ball wrapped with Coban wrap, as described, which is economical and easy to obtain and use in the clinical setting. We believe this cost-effective method will offer surgeons who perform partial knee arthroplasty a novel method to obtain the important information gleaned from the valgus stress radiograph and to improve surgical outcomes through the preoperative assessment of the lateral compartment. Additionally, as a quality and safety improvement initiative, we believe this technique will reduce radiographic exposure for those performing these studies, and, because the examination can be carried out by a single technologist, it will significantly improve efficiency in the radiology suite.

Conclusion

We have developed a new method of obtaining the important valgus stress radiograph as part of the workup of patients with medial-compartment osteoarthritis of the knee. The technique can be performed with easily obtainable, commercially available products and is reliable 93.5% of the time. It also adds to the efficiency of the radiology suite and reduces radiographic exposure for technologists.

Medial-compartment partial knee arthroplasty (unicompartmental replacement) is an accepted surgical intervention for anteromedial osteoarthritis of the knee.1 The radiographic investigations required in the workup of these patients should include weight-bearing standing anteroposterior (AP), lateral, and sunrise (Merchant) views, as well as a valgus stress AP radiograph to assess the functionality of the lateral compartment. The method of properly obtaining the valgus stress film has been well described by the Oxford Group.2 Its recommended radiographic technique requires that a surgeon or a radiologic technologist perform the valgus stress maneuver, manually, while another technologist shoots the film. The 2 consequences of this technique are that it requires 2 individuals to obtain the film, and it subjects the individual who is applying the stress to some level of radiation exposure, which is undesirable. Because of this and the time inconvenience, many surgeons omit the valgus stress radiograph, which can lead to the adverse outcome of missing a lateral compartment that is functionally incompetent, resulting in the potential for early lateral compartment progression of osteoarthritis and the need for revision surgery, usually to a total knee arthroplasty.

In an attempt to mitigate these barriers to obtaining the necessary valgus stress radiograph, Dr. Mauerhan’s team developed a technique that could be done with the assistance of the patient and would require only 1 technologist to perform. Additionally, this project was a quality improvement initiative, because it lowered radiation exposure to all personnel involved in obtaining the correct films.

Materials and Methods

We initiated the project using weight-bearing strategies to impart the valgus stress view of the knee. After trying several different wedges and blocks, and varying patient instructions, we realized a different approach to this problem would be required to find an acceptable solution. We redirected our efforts to effectively performing the stress view with the patient in a supine position on the radiograph table. Ultimately, we decided that a much stiffer wedge and a denser object to squeeze would facilitate obtaining a proper film. Considering all available options, a youth size 4 soccer ball (diameter, 11 in) was introduced along with a slightly larger positioning wedge. The soccer ball was wrapped with 4-in Coban wrap (3M) to create a nonslip surface. This change in patient positioning, along with a standardized 7º to 10º cephalic radiographic tube angulation, helped to correct issues with tibial plateau visualization. Once these changes were enacted, we obtained fairly consistent positive results, and we instituted this patient-directed valgus stress view of the knee, along with a manual valgus stress view for comparison.

The protocol for obtaining the patient-directed valgus stress view of the knee is as follows: The patient lays supine with a dense 45º spine-positioning wedge (Burlington Medical Supplies) placed under both knees and the patient’s heels on the examining table. The radiographic tube is angled cephalad 7º to 10º centered on the inferior pole of the patella, using a 40-in source to image-receptor distance, collimated to part; the image receptor is placed under the affected knee, below the positioning wedge. The affected knee is rotated to the “true” AP position (the patella will be centered between the femoral condyles on the AP exposure), and the ball is placed between the patient’s legs just above the ankle joint. The technologist demonstrates to the patient how to squeeze the ball while maintaining contact of heels with the table. The technologist can exit the room and obtain the exposure, which is taken while the patient is squeezing the ball, as shown in Figures 1A and 1B. Examples of the standing AP, manual stress, and patient-directed valgus radiographs are shown in Figures 2A-2C. The entire technique is demonstrated in the Video.

Vidyard Video

 

 

Results

During the 9 months of this quality improvement project, 78 examinations were performed. Five studies did not show complete correction of the varus deformity. Of these, 3 showed complete correction on a manual valgus stress radiograph, and 2 did not, contraindicating the use of partial knee replacement. Three patients displayed collapse of the lateral compartment, indicating a nonfunctional lateral compartment, and, therefore, were also a contraindication to partial knee arthroplasty. The remaining 70 patients had identical radiographic results with both the manual and patient-directed valgus stress tests. There was no instance of examination failure or need to repeat as a result of difficulty of the examination for the patient. Repeat films because of positioning errors were very rare, usually early in the learning curve, and no more prevalent than when using the manual stress method. The technique was reproducible and easy to teach and adopt.

 

 

Discussion

In total, 73 patients (93.5%) with the patient-directed stress film showed the desired result, either correction of the medial compartment narrowing in conjunction with an intact lateral compartment or narrowing of the lateral compartment. Of the 5 patients (6.5%) whose patient-directed stress films did not show correction of the varus deformity, 3 patients displayed correction with a manually applied stress radiograph and 2 did not. Based on this observation, our recommendation would be for those patients who do not show adequate correction on the patient-directed stress radiograph to have a manual examination to establish the presence or absence of the desired correction.

Performing a valgus stress radiograph is an integral part of the investigation to determine if the patient is an appropriate candidate for partial knee arthroplasty.3 The historical, manually performed valgus stress radiograph requires 2 individuals, 1 to apply the stress with the patient on the table and 1 to shoot the exposure. For the individual or individuals applying this stress, there is an increased radiation exposure that would be undesirable over a long career. The authors developed a new technique using a commercially available spinal positioning wedge and 11-in youth soccer ball wrapped with Coban wrap, as described, which is economical and easy to obtain and use in the clinical setting. We believe this cost-effective method will offer surgeons who perform partial knee arthroplasty a novel method to obtain the important information gleaned from the valgus stress radiograph and to improve surgical outcomes through the preoperative assessment of the lateral compartment. Additionally, as a quality and safety improvement initiative, we believe this technique will reduce radiographic exposure for those performing these studies, and, because the examination can be carried out by a single technologist, it will significantly improve efficiency in the radiology suite.

Conclusion

We have developed a new method of obtaining the important valgus stress radiograph as part of the workup of patients with medial-compartment osteoarthritis of the knee. The technique can be performed with easily obtainable, commercially available products and is reliable 93.5% of the time. It also adds to the efficiency of the radiology suite and reduces radiographic exposure for technologists.

Medial-compartment partial knee arthroplasty (unicompartmental replacement) is an accepted surgical intervention for anteromedial osteoarthritis of the knee.1 The radiographic investigations required in the workup of these patients should include weight-bearing standing anteroposterior (AP), lateral, and sunrise (Merchant) views, as well as a valgus stress AP radiograph to assess the functionality of the lateral compartment. The method of properly obtaining the valgus stress film has been well described by the Oxford Group.2 Its recommended radiographic technique requires that a surgeon or a radiologic technologist perform the valgus stress maneuver, manually, while another technologist shoots the film. The 2 consequences of this technique are that it requires 2 individuals to obtain the film, and it subjects the individual who is applying the stress to some level of radiation exposure, which is undesirable. Because of this and the time inconvenience, many surgeons omit the valgus stress radiograph, which can lead to the adverse outcome of missing a lateral compartment that is functionally incompetent, resulting in the potential for early lateral compartment progression of osteoarthritis and the need for revision surgery, usually to a total knee arthroplasty.

In an attempt to mitigate these barriers to obtaining the necessary valgus stress radiograph, Dr. Mauerhan’s team developed a technique that could be done with the assistance of the patient and would require only 1 technologist to perform. Additionally, this project was a quality improvement initiative, because it lowered radiation exposure to all personnel involved in obtaining the correct films.

Materials and Methods

We initiated the project using weight-bearing strategies to impart the valgus stress view of the knee. After trying several different wedges and blocks, and varying patient instructions, we realized a different approach to this problem would be required to find an acceptable solution. We redirected our efforts to effectively performing the stress view with the patient in a supine position on the radiograph table. Ultimately, we decided that a much stiffer wedge and a denser object to squeeze would facilitate obtaining a proper film. Considering all available options, a youth size 4 soccer ball (diameter, 11 in) was introduced along with a slightly larger positioning wedge. The soccer ball was wrapped with 4-in Coban wrap (3M) to create a nonslip surface. This change in patient positioning, along with a standardized 7º to 10º cephalic radiographic tube angulation, helped to correct issues with tibial plateau visualization. Once these changes were enacted, we obtained fairly consistent positive results, and we instituted this patient-directed valgus stress view of the knee, along with a manual valgus stress view for comparison.

The protocol for obtaining the patient-directed valgus stress view of the knee is as follows: The patient lays supine with a dense 45º spine-positioning wedge (Burlington Medical Supplies) placed under both knees and the patient’s heels on the examining table. The radiographic tube is angled cephalad 7º to 10º centered on the inferior pole of the patella, using a 40-in source to image-receptor distance, collimated to part; the image receptor is placed under the affected knee, below the positioning wedge. The affected knee is rotated to the “true” AP position (the patella will be centered between the femoral condyles on the AP exposure), and the ball is placed between the patient’s legs just above the ankle joint. The technologist demonstrates to the patient how to squeeze the ball while maintaining contact of heels with the table. The technologist can exit the room and obtain the exposure, which is taken while the patient is squeezing the ball, as shown in Figures 1A and 1B. Examples of the standing AP, manual stress, and patient-directed valgus radiographs are shown in Figures 2A-2C. The entire technique is demonstrated in the Video.

Vidyard Video

 

 

Results

During the 9 months of this quality improvement project, 78 examinations were performed. Five studies did not show complete correction of the varus deformity. Of these, 3 showed complete correction on a manual valgus stress radiograph, and 2 did not, contraindicating the use of partial knee replacement. Three patients displayed collapse of the lateral compartment, indicating a nonfunctional lateral compartment, and, therefore, were also a contraindication to partial knee arthroplasty. The remaining 70 patients had identical radiographic results with both the manual and patient-directed valgus stress tests. There was no instance of examination failure or need to repeat as a result of difficulty of the examination for the patient. Repeat films because of positioning errors were very rare, usually early in the learning curve, and no more prevalent than when using the manual stress method. The technique was reproducible and easy to teach and adopt.

 

 

Discussion

In total, 73 patients (93.5%) with the patient-directed stress film showed the desired result, either correction of the medial compartment narrowing in conjunction with an intact lateral compartment or narrowing of the lateral compartment. Of the 5 patients (6.5%) whose patient-directed stress films did not show correction of the varus deformity, 3 patients displayed correction with a manually applied stress radiograph and 2 did not. Based on this observation, our recommendation would be for those patients who do not show adequate correction on the patient-directed stress radiograph to have a manual examination to establish the presence or absence of the desired correction.

Performing a valgus stress radiograph is an integral part of the investigation to determine if the patient is an appropriate candidate for partial knee arthroplasty.3 The historical, manually performed valgus stress radiograph requires 2 individuals, 1 to apply the stress with the patient on the table and 1 to shoot the exposure. For the individual or individuals applying this stress, there is an increased radiation exposure that would be undesirable over a long career. The authors developed a new technique using a commercially available spinal positioning wedge and 11-in youth soccer ball wrapped with Coban wrap, as described, which is economical and easy to obtain and use in the clinical setting. We believe this cost-effective method will offer surgeons who perform partial knee arthroplasty a novel method to obtain the important information gleaned from the valgus stress radiograph and to improve surgical outcomes through the preoperative assessment of the lateral compartment. Additionally, as a quality and safety improvement initiative, we believe this technique will reduce radiographic exposure for those performing these studies, and, because the examination can be carried out by a single technologist, it will significantly improve efficiency in the radiology suite.

Conclusion

We have developed a new method of obtaining the important valgus stress radiograph as part of the workup of patients with medial-compartment osteoarthritis of the knee. The technique can be performed with easily obtainable, commercially available products and is reliable 93.5% of the time. It also adds to the efficiency of the radiology suite and reduces radiographic exposure for technologists.

Distal radius fracture is a common injury treated by orthopedic surgeons. Fifty percent or more of distal radius fractures (DRFs) occur with concomitant ulnar styloid fractures (USFs)^1-3 (Figure). The base of the ulnar styloid is the insertion site for portions of the triangular fibrocartilaginous complex (TFCC), which is a primary stabilizer of the distal radioulnar joint (DRUJ).^4,5

Although the topic has received significant attention in the literature, there remains a lack of consensus on the prognostic and clinical significance of USF occurring with DRF. In a series reported by May and colleagues,⁶ all patients with DRUJ instability after DRF also had an USF. Some authors have reported USF as a poor prognostic indicator for DRF, as the occurrence of USF was taken as a proxy for DRUJ instability.^7,8 Conversely, other authors have reported that USF nonunion has no effect on the outcome of volar plating of DRF.^9-11 In a retrospective cohort study of 182 patients, Li and colleagues¹² found no clinically significant difference in outcome between presence or absence of USF with DRF. They also reported that the quality of the DRF reduction was the main determinant of clinical outcome in patients with USF.

We examined a large cohort of patients treated for DRF to identify any possible effect of an associated USF on clinical outcome. All patients provided written informed consent for study inclusion.

Materials and Methods

We retrospectively evaluated 315 cases of DRFs treated (184 operatively, 131 nonoperatively) by members of the Trauma and Hand divisions at our institution over a 7-year period. All cases had sufficient follow-up. In each group, patients with concomitant USF were identified.

At presentation, all displaced fractures underwent closed reduction and immobilization with a sugar-tong splint. Baseline demographic data, injury information, and baseline functional scores on the Disabilities of the Arm, Shoulder, and Hand (DASH) questionnaire and the 36-Item Short Form Health Survey (SF-36) were recorded. Complete histories were taken and physical examinations performed. Standard radiographs of the injured and contralateral wrists were obtained at time of initial injury.¹³

Surgery was indicated in patients with an open fracture and in patients with an inherently unstable fracture pattern, using the instability criteria of Cooney and colleagues.¹⁴ According to these criteria, unstable fractures have lost alignment after closed reduction or have more than 20° of dorsal angulation, more than 10 mm of longitudinal shortening, or more than 2 mm of articular displacement.¹⁴ Patients were treated with either a volar locked plate or bridging external fixation with supplemental Kirschner-wire fixation (usually 2 or 3 wires). Patients in both groups (operative, nonoperative) participated in a formal outpatient therapy program that emphasized active and passive range of motion (ROM) of the finger, wrist motion (if clinically appropriate), and forearm motion. Mean clinical follow-up was 12 months (range, 8-18 months). At each clinic visit, we used a handheld dynamometer to measure ROM, grip strength, and other parameters and compared them with the same parameters on the uninjured side, along with functional outcome.

Differences in demographic characteristics were evaluated with 2 tests—the χ² test for categorical variables (eg, USF incidence, sex, hand dominance, fracture pattern) and the Student t test for continuous variables. Mann-Whitney U tests were used to assess differences between groups in DASH and SF-36 scores at long-term follow-up, as well as differences in ROM and radiographic measurements. Statistical significance was set at P < .05.

Results

DRFs occurred in the dominant-side wrist more commonly (P < .05) in the nonoperative group than in the operative group, though there was no difference in hand dominance and presence or absence of USF. There was a significant correlation of intra-articular fractures in the operative group (70%) compared with the nonoperative group (34%), though no association was found between presence of USF and intra-articular fracture location.

The percentage of concomitant USF was higher (P< .0002) in patients treated operatively (64.1%) than in those treated nonoperatively (38.9%). Mean (SD) pain score was higher (P = .0001) for patients with USF, 1.80 (2.43), than for patients without USF, 0.80 (1.55). This relationship held in both the operative group, 1.95 (2.48) versus 1.04 (1.58) (P = .027), and the nonoperative group, 1.29 (2.09) versus 0.66 (1.53) (P = .048). Similarly, at long-term follow-up for the entire patient cohort, mean (SD) DASH score was negatively affected by presence of USF, 17.03 (18.94) versus 9.21 (14.06) (P = .001), as was mean (SD) SF-36 score, 77.16 (17.69) versus 82.68 (16.10) (P = .022). This relationship also held in the operative and nonoperative groups with respect to pain and DASH scores, though there were only trends in this direction with respect to SF-36 scores. At final follow-up, there was no significant correlation of pain, SF-36, or DASH scores with presence of an intra-articular fracture as compared with an extra-articular fracture.

Time to radiographic healing was not influenced by presence of USF compared with absence of USF (11 vs 10.06 weeks; P > .05). Similarly, healing was no different in intra-articular fractures compared with extra-articular fractures (11 vs 10 weeks; P > .05).

Wrist ROM at final follow-up was not affected by presence of USF; there was no significant difference in wrist flexion, extension, or forearm rotation. In addition, mean (SD) grip strength was unaffected (P = .132) by presence or absence of USF with DRF overall, 45.45% (31.92) of contralateral versus 52.88% (30.03). However, grip strength was negatively affected (P = .035) by presence of USF in the nonoperative group, 37.79% (20.58) versus 54.52% (31.89) (Table).

Discussion

In this study, we determined that presence of USF was a negative predictor for clinical outcomes after DRF. Given the higher incidence of USF in operatively treated DRFs, USF likely represents a higher-energy mechanism of injury. We think these inferior clinical results are attributable to other wrist pathologies that commonly occur with these injuries. These pathologies, identified in the past, include stylocarpal impaction, extensor carpi ulnaris tendinitis, and pain at USF site.^6,10,15 In addition, intracarpal ligamentous injuries, including damage to scapholunate and lunotriquetral ligaments, have been shown to occur in roughly 80% of patients who sustain DRFs, with TFCC injuries occurring at a rate of 60%.¹⁶

Patient outcome is multifactorial and depends on initial injury characteristics, reduction quality, associated injuries, and patient demographics and lifestyle factors. Li and colleagues¹² showed that the quality of the DRF reduction influenced outcomes in these injuries, as the ulnar styloid and its associated TFCC are in turn reduced more anatomically with a restored DRF reduction. This concept applies to injuries treated both operatively and nonoperatively. Similarly, Xarchas and colleagues¹⁷ identified malunion of the ulnar styloid as causing chronic wrist pain because of triquetral impingement, which was treated successfully with ulnar styloidectomy. The poor results at final follow-up in their study may reflect severity of the initial injury, as reported by Frykman.¹⁸

Additional factors may compromise clinical outcomes after such injuries. For example, the effect of USF fragment size on outcome has been suggested and debated. In a retrospective series, May and colleagues⁶ identified fractures involving the base of the ulnar styloid or fovea as potentially destabilizing the DRUJ and in turn leading to chronic instability. This mechanism should be considered a potential contributor to protracted clinical recovery. Other studies have shown that, irrespective of USF fragment size, presence of USF with DRF is not a reliable predictor of DRUJ instability.^2,10,19 In the present study, we simply identified presence or absence of USF, irrespective of either stability or fragment size. In cases in which there was an USF without instability, we fixed the DRF in isolation, without surgically addressing the USF. Our data demonstrated that, even in the absence of DRUJ instability, presence of USF was a negative prognostic indicator for patient outcome.

This study had several limitations. First, its design was retrospective. A prospective study would have been ideal for eliminating certain inherent bias. Second, USF represents a higher association with DRUJ instability.⁶ As there are no validated tests for this clinical entity, identification is somewhat subjective. We did not separate patients by presence or absence of DRUJ instability and thus were not able to directly correlate the connection between USF, DRUJ instability, and poor outcomes in association with DRF. In addition, management of an unstable DRUJ after operative fixation of DRF is controversial, with techniques ranging from splinting in supination to pinning the DRUJ. This inconsistency likely contributed to some error between groups of patients in this study. Last, we did not stratify patients by USF fragment size, as previously discussed, which may have affected outcomes within patient groups.

Our data add to the evidence showing that USF in association with DRF portends poorer clinical outcomes. Concomitant USF should alert the treating physician to a higher-energy mechanism of injury and raise the index of suspicion for other associated injuries in the carpus.

Distal radius fracture is a common injury treated by orthopedic surgeons. Fifty percent or more of distal radius fractures (DRFs) occur with concomitant ulnar styloid fractures (USFs)^1-3 (Figure). The base of the ulnar styloid is the insertion site for portions of the triangular fibrocartilaginous complex (TFCC), which is a primary stabilizer of the distal radioulnar joint (DRUJ).^4,5

Although the topic has received significant attention in the literature, there remains a lack of consensus on the prognostic and clinical significance of USF occurring with DRF. In a series reported by May and colleagues,⁶ all patients with DRUJ instability after DRF also had an USF. Some authors have reported USF as a poor prognostic indicator for DRF, as the occurrence of USF was taken as a proxy for DRUJ instability.^7,8 Conversely, other authors have reported that USF nonunion has no effect on the outcome of volar plating of DRF.^9-11 In a retrospective cohort study of 182 patients, Li and colleagues¹² found no clinically significant difference in outcome between presence or absence of USF with DRF. They also reported that the quality of the DRF reduction was the main determinant of clinical outcome in patients with USF.

We examined a large cohort of patients treated for DRF to identify any possible effect of an associated USF on clinical outcome. All patients provided written informed consent for study inclusion.

Materials and Methods

We retrospectively evaluated 315 cases of DRFs treated (184 operatively, 131 nonoperatively) by members of the Trauma and Hand divisions at our institution over a 7-year period. All cases had sufficient follow-up. In each group, patients with concomitant USF were identified.

At presentation, all displaced fractures underwent closed reduction and immobilization with a sugar-tong splint. Baseline demographic data, injury information, and baseline functional scores on the Disabilities of the Arm, Shoulder, and Hand (DASH) questionnaire and the 36-Item Short Form Health Survey (SF-36) were recorded. Complete histories were taken and physical examinations performed. Standard radiographs of the injured and contralateral wrists were obtained at time of initial injury.¹³

Surgery was indicated in patients with an open fracture and in patients with an inherently unstable fracture pattern, using the instability criteria of Cooney and colleagues.¹⁴ According to these criteria, unstable fractures have lost alignment after closed reduction or have more than 20° of dorsal angulation, more than 10 mm of longitudinal shortening, or more than 2 mm of articular displacement.¹⁴ Patients were treated with either a volar locked plate or bridging external fixation with supplemental Kirschner-wire fixation (usually 2 or 3 wires). Patients in both groups (operative, nonoperative) participated in a formal outpatient therapy program that emphasized active and passive range of motion (ROM) of the finger, wrist motion (if clinically appropriate), and forearm motion. Mean clinical follow-up was 12 months (range, 8-18 months). At each clinic visit, we used a handheld dynamometer to measure ROM, grip strength, and other parameters and compared them with the same parameters on the uninjured side, along with functional outcome.

Differences in demographic characteristics were evaluated with 2 tests—the χ² test for categorical variables (eg, USF incidence, sex, hand dominance, fracture pattern) and the Student t test for continuous variables. Mann-Whitney U tests were used to assess differences between groups in DASH and SF-36 scores at long-term follow-up, as well as differences in ROM and radiographic measurements. Statistical significance was set at P < .05.

Results

DRFs occurred in the dominant-side wrist more commonly (P < .05) in the nonoperative group than in the operative group, though there was no difference in hand dominance and presence or absence of USF. There was a significant correlation of intra-articular fractures in the operative group (70%) compared with the nonoperative group (34%), though no association was found between presence of USF and intra-articular fracture location.

The percentage of concomitant USF was higher (P< .0002) in patients treated operatively (64.1%) than in those treated nonoperatively (38.9%). Mean (SD) pain score was higher (P = .0001) for patients with USF, 1.80 (2.43), than for patients without USF, 0.80 (1.55). This relationship held in both the operative group, 1.95 (2.48) versus 1.04 (1.58) (P = .027), and the nonoperative group, 1.29 (2.09) versus 0.66 (1.53) (P = .048). Similarly, at long-term follow-up for the entire patient cohort, mean (SD) DASH score was negatively affected by presence of USF, 17.03 (18.94) versus 9.21 (14.06) (P = .001), as was mean (SD) SF-36 score, 77.16 (17.69) versus 82.68 (16.10) (P = .022). This relationship also held in the operative and nonoperative groups with respect to pain and DASH scores, though there were only trends in this direction with respect to SF-36 scores. At final follow-up, there was no significant correlation of pain, SF-36, or DASH scores with presence of an intra-articular fracture as compared with an extra-articular fracture.

Time to radiographic healing was not influenced by presence of USF compared with absence of USF (11 vs 10.06 weeks; P > .05). Similarly, healing was no different in intra-articular fractures compared with extra-articular fractures (11 vs 10 weeks; P > .05).

Wrist ROM at final follow-up was not affected by presence of USF; there was no significant difference in wrist flexion, extension, or forearm rotation. In addition, mean (SD) grip strength was unaffected (P = .132) by presence or absence of USF with DRF overall, 45.45% (31.92) of contralateral versus 52.88% (30.03). However, grip strength was negatively affected (P = .035) by presence of USF in the nonoperative group, 37.79% (20.58) versus 54.52% (31.89) (Table).

Discussion

In this study, we determined that presence of USF was a negative predictor for clinical outcomes after DRF. Given the higher incidence of USF in operatively treated DRFs, USF likely represents a higher-energy mechanism of injury. We think these inferior clinical results are attributable to other wrist pathologies that commonly occur with these injuries. These pathologies, identified in the past, include stylocarpal impaction, extensor carpi ulnaris tendinitis, and pain at USF site.^6,10,15 In addition, intracarpal ligamentous injuries, including damage to scapholunate and lunotriquetral ligaments, have been shown to occur in roughly 80% of patients who sustain DRFs, with TFCC injuries occurring at a rate of 60%.¹⁶

Patient outcome is multifactorial and depends on initial injury characteristics, reduction quality, associated injuries, and patient demographics and lifestyle factors. Li and colleagues¹² showed that the quality of the DRF reduction influenced outcomes in these injuries, as the ulnar styloid and its associated TFCC are in turn reduced more anatomically with a restored DRF reduction. This concept applies to injuries treated both operatively and nonoperatively. Similarly, Xarchas and colleagues¹⁷ identified malunion of the ulnar styloid as causing chronic wrist pain because of triquetral impingement, which was treated successfully with ulnar styloidectomy. The poor results at final follow-up in their study may reflect severity of the initial injury, as reported by Frykman.¹⁸

Additional factors may compromise clinical outcomes after such injuries. For example, the effect of USF fragment size on outcome has been suggested and debated. In a retrospective series, May and colleagues⁶ identified fractures involving the base of the ulnar styloid or fovea as potentially destabilizing the DRUJ and in turn leading to chronic instability. This mechanism should be considered a potential contributor to protracted clinical recovery. Other studies have shown that, irrespective of USF fragment size, presence of USF with DRF is not a reliable predictor of DRUJ instability.^2,10,19 In the present study, we simply identified presence or absence of USF, irrespective of either stability or fragment size. In cases in which there was an USF without instability, we fixed the DRF in isolation, without surgically addressing the USF. Our data demonstrated that, even in the absence of DRUJ instability, presence of USF was a negative prognostic indicator for patient outcome.

This study had several limitations. First, its design was retrospective. A prospective study would have been ideal for eliminating certain inherent bias. Second, USF represents a higher association with DRUJ instability.⁶ As there are no validated tests for this clinical entity, identification is somewhat subjective. We did not separate patients by presence or absence of DRUJ instability and thus were not able to directly correlate the connection between USF, DRUJ instability, and poor outcomes in association with DRF. In addition, management of an unstable DRUJ after operative fixation of DRF is controversial, with techniques ranging from splinting in supination to pinning the DRUJ. This inconsistency likely contributed to some error between groups of patients in this study. Last, we did not stratify patients by USF fragment size, as previously discussed, which may have affected outcomes within patient groups.

Our data add to the evidence showing that USF in association with DRF portends poorer clinical outcomes. Concomitant USF should alert the treating physician to a higher-energy mechanism of injury and raise the index of suspicion for other associated injuries in the carpus.

Distal radius fracture is a common injury treated by orthopedic surgeons. Fifty percent or more of distal radius fractures (DRFs) occur with concomitant ulnar styloid fractures (USFs)^1-3 (Figure). The base of the ulnar styloid is the insertion site for portions of the triangular fibrocartilaginous complex (TFCC), which is a primary stabilizer of the distal radioulnar joint (DRUJ).^4,5

Although the topic has received significant attention in the literature, there remains a lack of consensus on the prognostic and clinical significance of USF occurring with DRF. In a series reported by May and colleagues,⁶ all patients with DRUJ instability after DRF also had an USF. Some authors have reported USF as a poor prognostic indicator for DRF, as the occurrence of USF was taken as a proxy for DRUJ instability.^7,8 Conversely, other authors have reported that USF nonunion has no effect on the outcome of volar plating of DRF.^9-11 In a retrospective cohort study of 182 patients, Li and colleagues¹² found no clinically significant difference in outcome between presence or absence of USF with DRF. They also reported that the quality of the DRF reduction was the main determinant of clinical outcome in patients with USF.

We examined a large cohort of patients treated for DRF to identify any possible effect of an associated USF on clinical outcome. All patients provided written informed consent for study inclusion.

Materials and Methods

We retrospectively evaluated 315 cases of DRFs treated (184 operatively, 131 nonoperatively) by members of the Trauma and Hand divisions at our institution over a 7-year period. All cases had sufficient follow-up. In each group, patients with concomitant USF were identified.

At presentation, all displaced fractures underwent closed reduction and immobilization with a sugar-tong splint. Baseline demographic data, injury information, and baseline functional scores on the Disabilities of the Arm, Shoulder, and Hand (DASH) questionnaire and the 36-Item Short Form Health Survey (SF-36) were recorded. Complete histories were taken and physical examinations performed. Standard radiographs of the injured and contralateral wrists were obtained at time of initial injury.¹³

Surgery was indicated in patients with an open fracture and in patients with an inherently unstable fracture pattern, using the instability criteria of Cooney and colleagues.¹⁴ According to these criteria, unstable fractures have lost alignment after closed reduction or have more than 20° of dorsal angulation, more than 10 mm of longitudinal shortening, or more than 2 mm of articular displacement.¹⁴ Patients were treated with either a volar locked plate or bridging external fixation with supplemental Kirschner-wire fixation (usually 2 or 3 wires). Patients in both groups (operative, nonoperative) participated in a formal outpatient therapy program that emphasized active and passive range of motion (ROM) of the finger, wrist motion (if clinically appropriate), and forearm motion. Mean clinical follow-up was 12 months (range, 8-18 months). At each clinic visit, we used a handheld dynamometer to measure ROM, grip strength, and other parameters and compared them with the same parameters on the uninjured side, along with functional outcome.

Differences in demographic characteristics were evaluated with 2 tests—the χ² test for categorical variables (eg, USF incidence, sex, hand dominance, fracture pattern) and the Student t test for continuous variables. Mann-Whitney U tests were used to assess differences between groups in DASH and SF-36 scores at long-term follow-up, as well as differences in ROM and radiographic measurements. Statistical significance was set at P < .05.

Results

DRFs occurred in the dominant-side wrist more commonly (P < .05) in the nonoperative group than in the operative group, though there was no difference in hand dominance and presence or absence of USF. There was a significant correlation of intra-articular fractures in the operative group (70%) compared with the nonoperative group (34%), though no association was found between presence of USF and intra-articular fracture location.

The percentage of concomitant USF was higher (P< .0002) in patients treated operatively (64.1%) than in those treated nonoperatively (38.9%). Mean (SD) pain score was higher (P = .0001) for patients with USF, 1.80 (2.43), than for patients without USF, 0.80 (1.55). This relationship held in both the operative group, 1.95 (2.48) versus 1.04 (1.58) (P = .027), and the nonoperative group, 1.29 (2.09) versus 0.66 (1.53) (P = .048). Similarly, at long-term follow-up for the entire patient cohort, mean (SD) DASH score was negatively affected by presence of USF, 17.03 (18.94) versus 9.21 (14.06) (P = .001), as was mean (SD) SF-36 score, 77.16 (17.69) versus 82.68 (16.10) (P = .022). This relationship also held in the operative and nonoperative groups with respect to pain and DASH scores, though there were only trends in this direction with respect to SF-36 scores. At final follow-up, there was no significant correlation of pain, SF-36, or DASH scores with presence of an intra-articular fracture as compared with an extra-articular fracture.

Time to radiographic healing was not influenced by presence of USF compared with absence of USF (11 vs 10.06 weeks; P > .05). Similarly, healing was no different in intra-articular fractures compared with extra-articular fractures (11 vs 10 weeks; P > .05).

Wrist ROM at final follow-up was not affected by presence of USF; there was no significant difference in wrist flexion, extension, or forearm rotation. In addition, mean (SD) grip strength was unaffected (P = .132) by presence or absence of USF with DRF overall, 45.45% (31.92) of contralateral versus 52.88% (30.03). However, grip strength was negatively affected (P = .035) by presence of USF in the nonoperative group, 37.79% (20.58) versus 54.52% (31.89) (Table).

Discussion

In this study, we determined that presence of USF was a negative predictor for clinical outcomes after DRF. Given the higher incidence of USF in operatively treated DRFs, USF likely represents a higher-energy mechanism of injury. We think these inferior clinical results are attributable to other wrist pathologies that commonly occur with these injuries. These pathologies, identified in the past, include stylocarpal impaction, extensor carpi ulnaris tendinitis, and pain at USF site.^6,10,15 In addition, intracarpal ligamentous injuries, including damage to scapholunate and lunotriquetral ligaments, have been shown to occur in roughly 80% of patients who sustain DRFs, with TFCC injuries occurring at a rate of 60%.¹⁶

Patient outcome is multifactorial and depends on initial injury characteristics, reduction quality, associated injuries, and patient demographics and lifestyle factors. Li and colleagues¹² showed that the quality of the DRF reduction influenced outcomes in these injuries, as the ulnar styloid and its associated TFCC are in turn reduced more anatomically with a restored DRF reduction. This concept applies to injuries treated both operatively and nonoperatively. Similarly, Xarchas and colleagues¹⁷ identified malunion of the ulnar styloid as causing chronic wrist pain because of triquetral impingement, which was treated successfully with ulnar styloidectomy. The poor results at final follow-up in their study may reflect severity of the initial injury, as reported by Frykman.¹⁸

Additional factors may compromise clinical outcomes after such injuries. For example, the effect of USF fragment size on outcome has been suggested and debated. In a retrospective series, May and colleagues⁶ identified fractures involving the base of the ulnar styloid or fovea as potentially destabilizing the DRUJ and in turn leading to chronic instability. This mechanism should be considered a potential contributor to protracted clinical recovery. Other studies have shown that, irrespective of USF fragment size, presence of USF with DRF is not a reliable predictor of DRUJ instability.^2,10,19 In the present study, we simply identified presence or absence of USF, irrespective of either stability or fragment size. In cases in which there was an USF without instability, we fixed the DRF in isolation, without surgically addressing the USF. Our data demonstrated that, even in the absence of DRUJ instability, presence of USF was a negative prognostic indicator for patient outcome.

This study had several limitations. First, its design was retrospective. A prospective study would have been ideal for eliminating certain inherent bias. Second, USF represents a higher association with DRUJ instability.⁶ As there are no validated tests for this clinical entity, identification is somewhat subjective. We did not separate patients by presence or absence of DRUJ instability and thus were not able to directly correlate the connection between USF, DRUJ instability, and poor outcomes in association with DRF. In addition, management of an unstable DRUJ after operative fixation of DRF is controversial, with techniques ranging from splinting in supination to pinning the DRUJ. This inconsistency likely contributed to some error between groups of patients in this study. Last, we did not stratify patients by USF fragment size, as previously discussed, which may have affected outcomes within patient groups.

Our data add to the evidence showing that USF in association with DRF portends poorer clinical outcomes. Concomitant USF should alert the treating physician to a higher-energy mechanism of injury and raise the index of suspicion for other associated injuries in the carpus.

Pigmented villonodular synovitis (PVNS) is a rare monoarticular disorder that affects the joints, bursae, or tendon sheaths of 1.8 per million patients.^1,2 PVNS is defined by exuberant proliferation of synovial villi and nodules. Although its etiology is unknown, PVNS behaves much as a neoplastic process does, with occasional chromosomal abnormalities, local tissue invasion, and the potential for malignant transformation.^3,4 Radiographs show cystic erosions or joint space narrowing, and magnetic resonance imaging shows characteristic low-signal intensity (on T1- and T2-weighted sequences) because of high hemosiderin content. Biopsy remains the gold standard for diagnosis and reveals hemosiderin-laden macrophages, vascularized villi, mononuclear cell infiltration, and sporadic mitotic figures.⁵ Diffuse PVNS appears as a thickened synovium with matted villi and synovial folds; localized PVNS presents as a pedunculated, firm yellow nodule.⁶

PVNS has a predilection for large joints, most commonly the knee (up to 80% of cases) and the hip.^1,2,7 Treatment strategies for knee PVNS have been well studied and, as an aggregate, show no superiority of arthroscopic or open techniques.⁸ The literature on hip PVNS is less abundant and more case-based, making it difficult to reach a consensus on effective treatment. Open synovectomy and arthroplasty have been the mainstays of treatment over the past 60 years, but the advent of hip arthroscopy has introduced a new treatment modality.^1,9 As arthroscopic management becomes more readily available, it is important to understand and compare the effectiveness of synovectomy and arthroplasty.

We systematically reviewed the treatment modalities for PVNS of the hip to determine how synovectomy and arthroplasty compare with respect to efficacy and revision rates.

Methods

Search Strategy

We systematically reviewed the literature according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines using the PRISMA checklist.¹⁰ Searches were completed in July 2014 using the PubMed Medline database and the Cochrane Central Register of Clinical Trials. Keyword selection was designed to capture all level I to V evidence English-language studies that reported clinical and/or radiographic outcomes. This was accomplished with a keyword search of all available titles and manuscript abstracts: (pigmented [Title/Abstract] AND villonodular [Title/Abstract] AND synovitis [Title/Abstract]) AND (hip [Title/Abstract]) AND (English [lang])). Abstracts from the 75 resulting studies were reviewed for exclusion criteria, which consisted of any cadaveric, biomechanical, histologic, and/or kinematic results, as well as a lack of any clinical and/or radiographic data (eg, review or technique articles). Studies were also excluded if they did not have clinical follow-up of at least 2 years. Studies not dedicated to hip PVNS specifically were not immediately excluded but were reviewed for outcomes data specific to the hip PVNS subpopulation. If a specific hip PVNS population could be distinguished from other patients, that study was included for review. If a study could not be deconstructed as such or was entirely devoted to one of our exclusion criteria, that study was excluded from our review. This initial search strategy yielded 16 studies.^1,6,7,11-28

Bibliographical review of these 16 studies yielded several more for review. To ensure that no patients were counted twice, each study’s authors, data collection period, and ethnic population were reviewed and compared with those of the other studies. If there was any overlap in authorship, period, and place, only the study with the most relevant or comprehensive data was included. After accounting for all inclusion and exclusion criteria, we selected a total of 21 studies with 82 patients (86 hips) for inclusion (Figure 1).

Data Extraction

Details of study design, sample size, and patient demographics, including age, sex, and duration of symptoms, were recorded. Use of diagnostic biopsy, joint space narrowing on radiographs, treatment method, and use of radiation therapy were also abstracted. Some studies described multiple treatment methods. If those methods could not be differentiated into distinct outcomes groups, the study would have been excluded for lack of specific clinical data. Studies with sufficient data were deconstructed such that the patients from each treatment group were isolated.

Fewer than 5 studies reported physical examination findings, validated survey scores, and/or radiographic results. Therefore, the primary outcomes reported and compared between treatment groups were disease recurrence, clinical worsening defined as progressive pain or loss of function, and revision surgery. Revision surgery was subdivided into repeat synovectomy and eventual arthroplasty, arthrodesis, or revision arthroplasty. Time to revision surgery was also documented. Each study’s methodologic quality and bias were evaluated with the Modified Coleman Methodology Score (MCMS), described by Cowan and colleagues.29 MCMS is a 15-item instrument that has been used to assess randomized and nonrandomized patient trials.^30,31 It has a scaled potential score ranging from 0 to 100, with scores from 85 through 100 indicating excellent, 70 through 84 good, 55 through 69 fair, and under 55 poor.

Statistical Analysis

We report our data as weighted means (SDs). A mean was calculated for each study reporting on a respective data point, and each mean was then weighted according to the sample size of that study. We multiplied each study’s individual mean by the number of patients enrolled in that study and divided the sum of all the studies’ weighted data points by the number of eligible patients in all relevant studies. The result is that the nonweighted means from studies with a smaller sample size did not carry as much weight as those from larger studies. We then compared 2 groups of patients: those who had only a synovectomy and those who had a combination of synovectomy and arthroplasty. The synovectomy-only group was also compared with a group that underwent total hip arthroplasty (THA) specifically (Figure 2). Groups were compared with Student t test (SPSS Version 18, IBM), and statistical significance was set at α = 0.05.

Results

Twenty-one studies (82 patients) were included in the final dataset (Table 1). Of these studies, 19 were retrospective case series (level IV evidence) in which the number of eligible hip PVNS patients ranged from 1 to 15. The other 2 studies were case reports (level V evidence). Mean (SD) MCMS was 25.0 (10.9).

Fifty-one patients (59.3%) were female. Mean (SD) age of all patients was 33.2 (12.6) years. Mean (SD) duration of symptoms was 4.2 (2.7) years. The right hip was affected in 59.5% of patients in whom laterality was documented. Sixty-eight patients (79.1%) had biopsy-proven PVNS; presence or absence of a biopsy was not documented for the other 18 patients.

Of the 82 patients in the study, 45 (54.9%) underwent synovectomy without arthroplasty. Staged radiation was used to augment the synovectomy in 2 of these 45 cases. One series in this group consisted of 15 cases of arthroscopic synovectomy.¹ The 37 patients (45.1%) in the other treatment group had arthroplasty at time of synovectomy. These patients underwent 22 THAs, 8 cup arthroplasties, 2 metal-on-metal hip resurfacings, and 1 hemiarthroplasty. The remaining 4 patients were treated nonoperatively (3) or with primary arthrodesis (1).

Comparisons between the synovectomy-only and synovectomy-with-arthroplasty groups are listed in Table 2. Synovectomy patients were younger on average than arthroplasty patients, but the difference was not statistically significant (P = .28). Only 6 studies distinguished between local and diffuse PVNS histology, and the diffuse type was detected in 87.0%, with insufficient data to detect a difference between the synovectomy and arthroplasty groups. In studies with documented radiographic findings, 75.0% of patients had evidence of joint space narrowing, which was significantly (P = .03) more common in the arthroplasty group (96.7% vs 31.3%).

Mean (SD) clinical follow-up was 8.4 (5.9) years for all patients. A larger percentage of synovectomy-only patients experienced recurrence and worsened symptoms, but neither trend achieved statistical significance. The rate of eventual THA or arthrodesis after synovectomy alone was almost identical (P = .17) to the rate of revision THA in the synovectomy-with-arthroplasty group (26.2% vs 24.3%). Time to revision surgery, however, was significantly (P = .02) longer in the arthroplasty group. Two additional patients in the synovectomy-with-arthroplasty group underwent repeat synovectomy alone, but no patients in the synovectomy-only group underwent repeat synovectomy without arthroplasty.

One nonoperatively managed patient experienced symptom progression over the course of 10 years. The other 2 patients were stable after 2- and 4-year follow-up. The arthrodesis patient did not experience recurrence or have a revision operation in the 5 years after the index procedure.

Discussion

PVNS is a proliferative disorder of synovial tissue with a high risk of recurrence.^15,32 Metastasis is extremely rare; there is only 1 case report of a fatality, which occurred within 42 months.¹² Chiari and colleagues¹⁵ suggested that the PVNS recurrence rate is highest in the large joints. Therefore, in hip PVNS, early surgical resection is needed to limit articular destruction and the potential for recurrence. The primary treatment modalities are synovectomy alone and synovectomy with arthroplasty, which includes THA, cup arthroplasty, hip resurfacing, and hemiarthroplasty. According to our systematic review, about one-fourth of all patients in both treatment groups ultimately underwent revision surgery. Mean time to revision was significantly longer for synovectomy-with-arthroplasty patients (almost 12 years) than for synovectomy-only patients (6.5 years). One potential explanation is that arthroplasty component fixation may take longer to loosen than an inadequately synovectomized joint takes to recur. The synovectomy-only group did have a higher recurrence rate, though the difference was not statistically significant.

Open synovectomy is the most widely described technique for addressing hip PVNS. The precise pathophysiology of PVNS remains largely unknown, but most authors agree that aggressive débridement is required to halt its locally invasive course. Scott²⁴ described the invasion of vascular foramina from synovium into bone and thought that radical synovectomy was essential to remove the stalks of these synovial villi. Furthermore, PVNS most commonly affects adults in the third through fifth decades of life,7 and many surgeons want to avoid prosthetic components (which may loosen over time) in this age group. Synovectomy, however, has persistently high recurrence rates, and, without removal of the femoral head and neck, it can be difficult to obtain adequate exposure for complete débridement. Although adjuvant external beam radiation has been used by some authors,^17,19,33 its utility is unproven, and other authors have cautioned against unnecessary irradiation of reproductive organs.^1,24,34

The high rates of bony involvement, joint destruction, and recurrence after synovectomy have prompted many surgeons to turn to arthroplasty. González Della Valle and colleagues18 theorized that joint space narrowing is more common in hip PVNS because of the poor distensibility of the hip capsule compared with that of the knee and other joints. In turn, bony lesions and arthritis present earlier in hip PVNS.¹⁴ Yoo and colleagues¹⁴ found a statistically significant increase in Harris Hip Scale (HHS) scores and a high rate of return to athletic activity after THA for PVNS. However, they also reported revisions for component loosening and osteolysis in 2 of 8 patients and periprosthetic osteolysis without loosening in another 2 patients. Vastel and colleagues¹⁶ similarly reported aseptic loosening of the acetabular component in half their patient cohort. No studies have determined which condition—PVNS recurrence or debris-related osteolysis—causes the accelerated loosening in this demographic.

Byrd and colleagues¹ recently described use of hip arthroscopy in the treatment of PVNS. In a cohort of 13 patients, they found statistically significant improvements in HHS scores, no postoperative complications, and only 1 revision (THA 6 years after surgery). Although there is a prevailing perception that nodular (vs diffuse) PVNS is more appropriately treated with arthroscopic excision, no studies have provided data on this effect, and Byrd and colleagues¹ in fact showed a trend of slightly better outcomes in diffuse cases than in nodular cases. The main challenges of hip arthroscopy are the steep learning curve and adequate exposure. Recent innovations include additional arthroscopic portals and enlarged T-capsulotomy, which may be contributing to decreased complication rates in hip arthroscopy in general.³⁵

The limitations of this systematic review were largely imposed by the studies analyzed. The primary limitation was the relative paucity of clinical and radiographic data on hip PVNS. To our knowledge, studies on the treatment of hip PVNS have reported evidence levels no higher than IV. In addition, the studies we reviewed often had only 1 or 2 patient cases satisfying our inclusion criteria. For this reason, we included case reports, which further lowered the level of evidence of studies used. There were no consistently reported physical examination, survey, or radiographic findings that could be used to compare studies. All studies with sufficient data on hip PVNS treatment outcomes were rated poorly with the Modified Coleman Methodology Scoring system.²⁹ Selection bias was minimized by the inclusive nature of studies with level I to V evidence, but this led to a study design bias in that most studies consisted of level IV evidence.

Conclusion

Although the hip PVNS literature is limited, our review provides insight into expected outcomes. No matter which surgery is to be performed, surgeons must counsel patients about the high revision rate. One in 4 patients ultimately undergoes a second surgery, which may be required within 6 or 7 years after synovectomy without arthroplasty. Further development and innovation in hip arthroscopy may transform the treatment of PVNS. We encourage other investigators to conduct prospective, comparative trials with higher evidence levels to assess the utility of arthroscopy and other treatment modalities.

Pigmented villonodular synovitis (PVNS) is a rare monoarticular disorder that affects the joints, bursae, or tendon sheaths of 1.8 per million patients.^1,2 PVNS is defined by exuberant proliferation of synovial villi and nodules. Although its etiology is unknown, PVNS behaves much as a neoplastic process does, with occasional chromosomal abnormalities, local tissue invasion, and the potential for malignant transformation.^3,4 Radiographs show cystic erosions or joint space narrowing, and magnetic resonance imaging shows characteristic low-signal intensity (on T1- and T2-weighted sequences) because of high hemosiderin content. Biopsy remains the gold standard for diagnosis and reveals hemosiderin-laden macrophages, vascularized villi, mononuclear cell infiltration, and sporadic mitotic figures.⁵ Diffuse PVNS appears as a thickened synovium with matted villi and synovial folds; localized PVNS presents as a pedunculated, firm yellow nodule.⁶

PVNS has a predilection for large joints, most commonly the knee (up to 80% of cases) and the hip.^1,2,7 Treatment strategies for knee PVNS have been well studied and, as an aggregate, show no superiority of arthroscopic or open techniques.⁸ The literature on hip PVNS is less abundant and more case-based, making it difficult to reach a consensus on effective treatment. Open synovectomy and arthroplasty have been the mainstays of treatment over the past 60 years, but the advent of hip arthroscopy has introduced a new treatment modality.^1,9 As arthroscopic management becomes more readily available, it is important to understand and compare the effectiveness of synovectomy and arthroplasty.

We systematically reviewed the treatment modalities for PVNS of the hip to determine how synovectomy and arthroplasty compare with respect to efficacy and revision rates.

Methods

Search Strategy

We systematically reviewed the literature according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines using the PRISMA checklist.¹⁰ Searches were completed in July 2014 using the PubMed Medline database and the Cochrane Central Register of Clinical Trials. Keyword selection was designed to capture all level I to V evidence English-language studies that reported clinical and/or radiographic outcomes. This was accomplished with a keyword search of all available titles and manuscript abstracts: (pigmented [Title/Abstract] AND villonodular [Title/Abstract] AND synovitis [Title/Abstract]) AND (hip [Title/Abstract]) AND (English [lang])). Abstracts from the 75 resulting studies were reviewed for exclusion criteria, which consisted of any cadaveric, biomechanical, histologic, and/or kinematic results, as well as a lack of any clinical and/or radiographic data (eg, review or technique articles). Studies were also excluded if they did not have clinical follow-up of at least 2 years. Studies not dedicated to hip PVNS specifically were not immediately excluded but were reviewed for outcomes data specific to the hip PVNS subpopulation. If a specific hip PVNS population could be distinguished from other patients, that study was included for review. If a study could not be deconstructed as such or was entirely devoted to one of our exclusion criteria, that study was excluded from our review. This initial search strategy yielded 16 studies.^1,6,7,11-28

Bibliographical review of these 16 studies yielded several more for review. To ensure that no patients were counted twice, each study’s authors, data collection period, and ethnic population were reviewed and compared with those of the other studies. If there was any overlap in authorship, period, and place, only the study with the most relevant or comprehensive data was included. After accounting for all inclusion and exclusion criteria, we selected a total of 21 studies with 82 patients (86 hips) for inclusion (Figure 1).

Data Extraction

Details of study design, sample size, and patient demographics, including age, sex, and duration of symptoms, were recorded. Use of diagnostic biopsy, joint space narrowing on radiographs, treatment method, and use of radiation therapy were also abstracted. Some studies described multiple treatment methods. If those methods could not be differentiated into distinct outcomes groups, the study would have been excluded for lack of specific clinical data. Studies with sufficient data were deconstructed such that the patients from each treatment group were isolated.

Fewer than 5 studies reported physical examination findings, validated survey scores, and/or radiographic results. Therefore, the primary outcomes reported and compared between treatment groups were disease recurrence, clinical worsening defined as progressive pain or loss of function, and revision surgery. Revision surgery was subdivided into repeat synovectomy and eventual arthroplasty, arthrodesis, or revision arthroplasty. Time to revision surgery was also documented. Each study’s methodologic quality and bias were evaluated with the Modified Coleman Methodology Score (MCMS), described by Cowan and colleagues.29 MCMS is a 15-item instrument that has been used to assess randomized and nonrandomized patient trials.^30,31 It has a scaled potential score ranging from 0 to 100, with scores from 85 through 100 indicating excellent, 70 through 84 good, 55 through 69 fair, and under 55 poor.

Statistical Analysis

We report our data as weighted means (SDs). A mean was calculated for each study reporting on a respective data point, and each mean was then weighted according to the sample size of that study. We multiplied each study’s individual mean by the number of patients enrolled in that study and divided the sum of all the studies’ weighted data points by the number of eligible patients in all relevant studies. The result is that the nonweighted means from studies with a smaller sample size did not carry as much weight as those from larger studies. We then compared 2 groups of patients: those who had only a synovectomy and those who had a combination of synovectomy and arthroplasty. The synovectomy-only group was also compared with a group that underwent total hip arthroplasty (THA) specifically (Figure 2). Groups were compared with Student t test (SPSS Version 18, IBM), and statistical significance was set at α = 0.05.

Results

Twenty-one studies (82 patients) were included in the final dataset (Table 1). Of these studies, 19 were retrospective case series (level IV evidence) in which the number of eligible hip PVNS patients ranged from 1 to 15. The other 2 studies were case reports (level V evidence). Mean (SD) MCMS was 25.0 (10.9).

Fifty-one patients (59.3%) were female. Mean (SD) age of all patients was 33.2 (12.6) years. Mean (SD) duration of symptoms was 4.2 (2.7) years. The right hip was affected in 59.5% of patients in whom laterality was documented. Sixty-eight patients (79.1%) had biopsy-proven PVNS; presence or absence of a biopsy was not documented for the other 18 patients.

Of the 82 patients in the study, 45 (54.9%) underwent synovectomy without arthroplasty. Staged radiation was used to augment the synovectomy in 2 of these 45 cases. One series in this group consisted of 15 cases of arthroscopic synovectomy.¹ The 37 patients (45.1%) in the other treatment group had arthroplasty at time of synovectomy. These patients underwent 22 THAs, 8 cup arthroplasties, 2 metal-on-metal hip resurfacings, and 1 hemiarthroplasty. The remaining 4 patients were treated nonoperatively (3) or with primary arthrodesis (1).

Comparisons between the synovectomy-only and synovectomy-with-arthroplasty groups are listed in Table 2. Synovectomy patients were younger on average than arthroplasty patients, but the difference was not statistically significant (P = .28). Only 6 studies distinguished between local and diffuse PVNS histology, and the diffuse type was detected in 87.0%, with insufficient data to detect a difference between the synovectomy and arthroplasty groups. In studies with documented radiographic findings, 75.0% of patients had evidence of joint space narrowing, which was significantly (P = .03) more common in the arthroplasty group (96.7% vs 31.3%).

Mean (SD) clinical follow-up was 8.4 (5.9) years for all patients. A larger percentage of synovectomy-only patients experienced recurrence and worsened symptoms, but neither trend achieved statistical significance. The rate of eventual THA or arthrodesis after synovectomy alone was almost identical (P = .17) to the rate of revision THA in the synovectomy-with-arthroplasty group (26.2% vs 24.3%). Time to revision surgery, however, was significantly (P = .02) longer in the arthroplasty group. Two additional patients in the synovectomy-with-arthroplasty group underwent repeat synovectomy alone, but no patients in the synovectomy-only group underwent repeat synovectomy without arthroplasty.

One nonoperatively managed patient experienced symptom progression over the course of 10 years. The other 2 patients were stable after 2- and 4-year follow-up. The arthrodesis patient did not experience recurrence or have a revision operation in the 5 years after the index procedure.

Discussion

PVNS is a proliferative disorder of synovial tissue with a high risk of recurrence.^15,32 Metastasis is extremely rare; there is only 1 case report of a fatality, which occurred within 42 months.¹² Chiari and colleagues¹⁵ suggested that the PVNS recurrence rate is highest in the large joints. Therefore, in hip PVNS, early surgical resection is needed to limit articular destruction and the potential for recurrence. The primary treatment modalities are synovectomy alone and synovectomy with arthroplasty, which includes THA, cup arthroplasty, hip resurfacing, and hemiarthroplasty. According to our systematic review, about one-fourth of all patients in both treatment groups ultimately underwent revision surgery. Mean time to revision was significantly longer for synovectomy-with-arthroplasty patients (almost 12 years) than for synovectomy-only patients (6.5 years). One potential explanation is that arthroplasty component fixation may take longer to loosen than an inadequately synovectomized joint takes to recur. The synovectomy-only group did have a higher recurrence rate, though the difference was not statistically significant.

Open synovectomy is the most widely described technique for addressing hip PVNS. The precise pathophysiology of PVNS remains largely unknown, but most authors agree that aggressive débridement is required to halt its locally invasive course. Scott²⁴ described the invasion of vascular foramina from synovium into bone and thought that radical synovectomy was essential to remove the stalks of these synovial villi. Furthermore, PVNS most commonly affects adults in the third through fifth decades of life,7 and many surgeons want to avoid prosthetic components (which may loosen over time) in this age group. Synovectomy, however, has persistently high recurrence rates, and, without removal of the femoral head and neck, it can be difficult to obtain adequate exposure for complete débridement. Although adjuvant external beam radiation has been used by some authors,^17,19,33 its utility is unproven, and other authors have cautioned against unnecessary irradiation of reproductive organs.^1,24,34

The high rates of bony involvement, joint destruction, and recurrence after synovectomy have prompted many surgeons to turn to arthroplasty. González Della Valle and colleagues18 theorized that joint space narrowing is more common in hip PVNS because of the poor distensibility of the hip capsule compared with that of the knee and other joints. In turn, bony lesions and arthritis present earlier in hip PVNS.¹⁴ Yoo and colleagues¹⁴ found a statistically significant increase in Harris Hip Scale (HHS) scores and a high rate of return to athletic activity after THA for PVNS. However, they also reported revisions for component loosening and osteolysis in 2 of 8 patients and periprosthetic osteolysis without loosening in another 2 patients. Vastel and colleagues¹⁶ similarly reported aseptic loosening of the acetabular component in half their patient cohort. No studies have determined which condition—PVNS recurrence or debris-related osteolysis—causes the accelerated loosening in this demographic.

Byrd and colleagues¹ recently described use of hip arthroscopy in the treatment of PVNS. In a cohort of 13 patients, they found statistically significant improvements in HHS scores, no postoperative complications, and only 1 revision (THA 6 years after surgery). Although there is a prevailing perception that nodular (vs diffuse) PVNS is more appropriately treated with arthroscopic excision, no studies have provided data on this effect, and Byrd and colleagues¹ in fact showed a trend of slightly better outcomes in diffuse cases than in nodular cases. The main challenges of hip arthroscopy are the steep learning curve and adequate exposure. Recent innovations include additional arthroscopic portals and enlarged T-capsulotomy, which may be contributing to decreased complication rates in hip arthroscopy in general.³⁵

The limitations of this systematic review were largely imposed by the studies analyzed. The primary limitation was the relative paucity of clinical and radiographic data on hip PVNS. To our knowledge, studies on the treatment of hip PVNS have reported evidence levels no higher than IV. In addition, the studies we reviewed often had only 1 or 2 patient cases satisfying our inclusion criteria. For this reason, we included case reports, which further lowered the level of evidence of studies used. There were no consistently reported physical examination, survey, or radiographic findings that could be used to compare studies. All studies with sufficient data on hip PVNS treatment outcomes were rated poorly with the Modified Coleman Methodology Scoring system.²⁹ Selection bias was minimized by the inclusive nature of studies with level I to V evidence, but this led to a study design bias in that most studies consisted of level IV evidence.

Conclusion

Although the hip PVNS literature is limited, our review provides insight into expected outcomes. No matter which surgery is to be performed, surgeons must counsel patients about the high revision rate. One in 4 patients ultimately undergoes a second surgery, which may be required within 6 or 7 years after synovectomy without arthroplasty. Further development and innovation in hip arthroscopy may transform the treatment of PVNS. We encourage other investigators to conduct prospective, comparative trials with higher evidence levels to assess the utility of arthroscopy and other treatment modalities.

Pigmented villonodular synovitis (PVNS) is a rare monoarticular disorder that affects the joints, bursae, or tendon sheaths of 1.8 per million patients.^1,2 PVNS is defined by exuberant proliferation of synovial villi and nodules. Although its etiology is unknown, PVNS behaves much as a neoplastic process does, with occasional chromosomal abnormalities, local tissue invasion, and the potential for malignant transformation.^3,4 Radiographs show cystic erosions or joint space narrowing, and magnetic resonance imaging shows characteristic low-signal intensity (on T1- and T2-weighted sequences) because of high hemosiderin content. Biopsy remains the gold standard for diagnosis and reveals hemosiderin-laden macrophages, vascularized villi, mononuclear cell infiltration, and sporadic mitotic figures.⁵ Diffuse PVNS appears as a thickened synovium with matted villi and synovial folds; localized PVNS presents as a pedunculated, firm yellow nodule.⁶

PVNS has a predilection for large joints, most commonly the knee (up to 80% of cases) and the hip.^1,2,7 Treatment strategies for knee PVNS have been well studied and, as an aggregate, show no superiority of arthroscopic or open techniques.⁸ The literature on hip PVNS is less abundant and more case-based, making it difficult to reach a consensus on effective treatment. Open synovectomy and arthroplasty have been the mainstays of treatment over the past 60 years, but the advent of hip arthroscopy has introduced a new treatment modality.^1,9 As arthroscopic management becomes more readily available, it is important to understand and compare the effectiveness of synovectomy and arthroplasty.

We systematically reviewed the treatment modalities for PVNS of the hip to determine how synovectomy and arthroplasty compare with respect to efficacy and revision rates.

Methods

Search Strategy

We systematically reviewed the literature according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines using the PRISMA checklist.¹⁰ Searches were completed in July 2014 using the PubMed Medline database and the Cochrane Central Register of Clinical Trials. Keyword selection was designed to capture all level I to V evidence English-language studies that reported clinical and/or radiographic outcomes. This was accomplished with a keyword search of all available titles and manuscript abstracts: (pigmented [Title/Abstract] AND villonodular [Title/Abstract] AND synovitis [Title/Abstract]) AND (hip [Title/Abstract]) AND (English [lang])). Abstracts from the 75 resulting studies were reviewed for exclusion criteria, which consisted of any cadaveric, biomechanical, histologic, and/or kinematic results, as well as a lack of any clinical and/or radiographic data (eg, review or technique articles). Studies were also excluded if they did not have clinical follow-up of at least 2 years. Studies not dedicated to hip PVNS specifically were not immediately excluded but were reviewed for outcomes data specific to the hip PVNS subpopulation. If a specific hip PVNS population could be distinguished from other patients, that study was included for review. If a study could not be deconstructed as such or was entirely devoted to one of our exclusion criteria, that study was excluded from our review. This initial search strategy yielded 16 studies.^1,6,7,11-28

Bibliographical review of these 16 studies yielded several more for review. To ensure that no patients were counted twice, each study’s authors, data collection period, and ethnic population were reviewed and compared with those of the other studies. If there was any overlap in authorship, period, and place, only the study with the most relevant or comprehensive data was included. After accounting for all inclusion and exclusion criteria, we selected a total of 21 studies with 82 patients (86 hips) for inclusion (Figure 1).

Data Extraction

Details of study design, sample size, and patient demographics, including age, sex, and duration of symptoms, were recorded. Use of diagnostic biopsy, joint space narrowing on radiographs, treatment method, and use of radiation therapy were also abstracted. Some studies described multiple treatment methods. If those methods could not be differentiated into distinct outcomes groups, the study would have been excluded for lack of specific clinical data. Studies with sufficient data were deconstructed such that the patients from each treatment group were isolated.

Fewer than 5 studies reported physical examination findings, validated survey scores, and/or radiographic results. Therefore, the primary outcomes reported and compared between treatment groups were disease recurrence, clinical worsening defined as progressive pain or loss of function, and revision surgery. Revision surgery was subdivided into repeat synovectomy and eventual arthroplasty, arthrodesis, or revision arthroplasty. Time to revision surgery was also documented. Each study’s methodologic quality and bias were evaluated with the Modified Coleman Methodology Score (MCMS), described by Cowan and colleagues.29 MCMS is a 15-item instrument that has been used to assess randomized and nonrandomized patient trials.^30,31 It has a scaled potential score ranging from 0 to 100, with scores from 85 through 100 indicating excellent, 70 through 84 good, 55 through 69 fair, and under 55 poor.

Statistical Analysis

We report our data as weighted means (SDs). A mean was calculated for each study reporting on a respective data point, and each mean was then weighted according to the sample size of that study. We multiplied each study’s individual mean by the number of patients enrolled in that study and divided the sum of all the studies’ weighted data points by the number of eligible patients in all relevant studies. The result is that the nonweighted means from studies with a smaller sample size did not carry as much weight as those from larger studies. We then compared 2 groups of patients: those who had only a synovectomy and those who had a combination of synovectomy and arthroplasty. The synovectomy-only group was also compared with a group that underwent total hip arthroplasty (THA) specifically (Figure 2). Groups were compared with Student t test (SPSS Version 18, IBM), and statistical significance was set at α = 0.05.

Results

Twenty-one studies (82 patients) were included in the final dataset (Table 1). Of these studies, 19 were retrospective case series (level IV evidence) in which the number of eligible hip PVNS patients ranged from 1 to 15. The other 2 studies were case reports (level V evidence). Mean (SD) MCMS was 25.0 (10.9).

Fifty-one patients (59.3%) were female. Mean (SD) age of all patients was 33.2 (12.6) years. Mean (SD) duration of symptoms was 4.2 (2.7) years. The right hip was affected in 59.5% of patients in whom laterality was documented. Sixty-eight patients (79.1%) had biopsy-proven PVNS; presence or absence of a biopsy was not documented for the other 18 patients.

Of the 82 patients in the study, 45 (54.9%) underwent synovectomy without arthroplasty. Staged radiation was used to augment the synovectomy in 2 of these 45 cases. One series in this group consisted of 15 cases of arthroscopic synovectomy.¹ The 37 patients (45.1%) in the other treatment group had arthroplasty at time of synovectomy. These patients underwent 22 THAs, 8 cup arthroplasties, 2 metal-on-metal hip resurfacings, and 1 hemiarthroplasty. The remaining 4 patients were treated nonoperatively (3) or with primary arthrodesis (1).

Comparisons between the synovectomy-only and synovectomy-with-arthroplasty groups are listed in Table 2. Synovectomy patients were younger on average than arthroplasty patients, but the difference was not statistically significant (P = .28). Only 6 studies distinguished between local and diffuse PVNS histology, and the diffuse type was detected in 87.0%, with insufficient data to detect a difference between the synovectomy and arthroplasty groups. In studies with documented radiographic findings, 75.0% of patients had evidence of joint space narrowing, which was significantly (P = .03) more common in the arthroplasty group (96.7% vs 31.3%).

Mean (SD) clinical follow-up was 8.4 (5.9) years for all patients. A larger percentage of synovectomy-only patients experienced recurrence and worsened symptoms, but neither trend achieved statistical significance. The rate of eventual THA or arthrodesis after synovectomy alone was almost identical (P = .17) to the rate of revision THA in the synovectomy-with-arthroplasty group (26.2% vs 24.3%). Time to revision surgery, however, was significantly (P = .02) longer in the arthroplasty group. Two additional patients in the synovectomy-with-arthroplasty group underwent repeat synovectomy alone, but no patients in the synovectomy-only group underwent repeat synovectomy without arthroplasty.

One nonoperatively managed patient experienced symptom progression over the course of 10 years. The other 2 patients were stable after 2- and 4-year follow-up. The arthrodesis patient did not experience recurrence or have a revision operation in the 5 years after the index procedure.

Discussion

PVNS is a proliferative disorder of synovial tissue with a high risk of recurrence.^15,32 Metastasis is extremely rare; there is only 1 case report of a fatality, which occurred within 42 months.¹² Chiari and colleagues¹⁵ suggested that the PVNS recurrence rate is highest in the large joints. Therefore, in hip PVNS, early surgical resection is needed to limit articular destruction and the potential for recurrence. The primary treatment modalities are synovectomy alone and synovectomy with arthroplasty, which includes THA, cup arthroplasty, hip resurfacing, and hemiarthroplasty. According to our systematic review, about one-fourth of all patients in both treatment groups ultimately underwent revision surgery. Mean time to revision was significantly longer for synovectomy-with-arthroplasty patients (almost 12 years) than for synovectomy-only patients (6.5 years). One potential explanation is that arthroplasty component fixation may take longer to loosen than an inadequately synovectomized joint takes to recur. The synovectomy-only group did have a higher recurrence rate, though the difference was not statistically significant.

Open synovectomy is the most widely described technique for addressing hip PVNS. The precise pathophysiology of PVNS remains largely unknown, but most authors agree that aggressive débridement is required to halt its locally invasive course. Scott²⁴ described the invasion of vascular foramina from synovium into bone and thought that radical synovectomy was essential to remove the stalks of these synovial villi. Furthermore, PVNS most commonly affects adults in the third through fifth decades of life,7 and many surgeons want to avoid prosthetic components (which may loosen over time) in this age group. Synovectomy, however, has persistently high recurrence rates, and, without removal of the femoral head and neck, it can be difficult to obtain adequate exposure for complete débridement. Although adjuvant external beam radiation has been used by some authors,^17,19,33 its utility is unproven, and other authors have cautioned against unnecessary irradiation of reproductive organs.^1,24,34

The high rates of bony involvement, joint destruction, and recurrence after synovectomy have prompted many surgeons to turn to arthroplasty. González Della Valle and colleagues18 theorized that joint space narrowing is more common in hip PVNS because of the poor distensibility of the hip capsule compared with that of the knee and other joints. In turn, bony lesions and arthritis present earlier in hip PVNS.¹⁴ Yoo and colleagues¹⁴ found a statistically significant increase in Harris Hip Scale (HHS) scores and a high rate of return to athletic activity after THA for PVNS. However, they also reported revisions for component loosening and osteolysis in 2 of 8 patients and periprosthetic osteolysis without loosening in another 2 patients. Vastel and colleagues¹⁶ similarly reported aseptic loosening of the acetabular component in half their patient cohort. No studies have determined which condition—PVNS recurrence or debris-related osteolysis—causes the accelerated loosening in this demographic.

Byrd and colleagues¹ recently described use of hip arthroscopy in the treatment of PVNS. In a cohort of 13 patients, they found statistically significant improvements in HHS scores, no postoperative complications, and only 1 revision (THA 6 years after surgery). Although there is a prevailing perception that nodular (vs diffuse) PVNS is more appropriately treated with arthroscopic excision, no studies have provided data on this effect, and Byrd and colleagues¹ in fact showed a trend of slightly better outcomes in diffuse cases than in nodular cases. The main challenges of hip arthroscopy are the steep learning curve and adequate exposure. Recent innovations include additional arthroscopic portals and enlarged T-capsulotomy, which may be contributing to decreased complication rates in hip arthroscopy in general.³⁵

The limitations of this systematic review were largely imposed by the studies analyzed. The primary limitation was the relative paucity of clinical and radiographic data on hip PVNS. To our knowledge, studies on the treatment of hip PVNS have reported evidence levels no higher than IV. In addition, the studies we reviewed often had only 1 or 2 patient cases satisfying our inclusion criteria. For this reason, we included case reports, which further lowered the level of evidence of studies used. There were no consistently reported physical examination, survey, or radiographic findings that could be used to compare studies. All studies with sufficient data on hip PVNS treatment outcomes were rated poorly with the Modified Coleman Methodology Scoring system.²⁹ Selection bias was minimized by the inclusive nature of studies with level I to V evidence, but this led to a study design bias in that most studies consisted of level IV evidence.

Conclusion

Although the hip PVNS literature is limited, our review provides insight into expected outcomes. No matter which surgery is to be performed, surgeons must counsel patients about the high revision rate. One in 4 patients ultimately undergoes a second surgery, which may be required within 6 or 7 years after synovectomy without arthroplasty. Further development and innovation in hip arthroscopy may transform the treatment of PVNS. We encourage other investigators to conduct prospective, comparative trials with higher evidence levels to assess the utility of arthroscopy and other treatment modalities.

Chronic hemophilic arthropathy, a well-known complication of hemophilia, develops as a long-term consequence of recurrent joint bleeds resulting in synovial hypertrophy (chronic proliferative synovitis) and joint cartilage destruction. Hemophilic arthropathy mostly affects the knees, ankles, and elbows and causes chronic joint pain and functional impairment in relatively young patients who have not received adequate primary prophylactic replacement therapy with factor concentrates from early childhood.^1-3

In the late stages of hemophilic arthropathy of the knee, total knee arthroplasty (TKA) provides dramatic joint pain relief, improves knee functional status, and reduces rebleeding into the joint.^4-8 TKA performed on a patient with hemophilia was first reported in the mid-1970s.^9,10 In these cases, the surgical procedure itself is often complicated by severe fibrosis developing in the joint soft tissues, flexion joint contracture, and poor quality of the joint bone structures. Even though TKA significantly reduces joint pain in patients with chronic hemophilic arthropathy, some authors have achieved only modest functional outcomes and experienced a high rate of complications (infection, prosthetic loosening).^11-13 Data on TKA outcomes are still scarce, and most studies have enrolled a limited number of patients.

We retrospectively evaluated the outcomes of 88 primary TKAs performed on patients with severe hemophilia at a single institution. Clinical outcomes and complications were assessed with a special focus on prosthetic survival and infection.

Patients and Methods

Ninety-one primary TKAs were performed in 77 patients with severe hemophilia A and B (factor VIII [FVIII] and factor IX plasma concentration, <1% each) between January 1, 1999, and December 31, 2011, and the medical records of all these patients were thoroughly reviewed in 2013. The cases of 3 patients who died shortly after surgery were excluded from analysis. Thus, 88 TKAs and 74 patients (74 males) were finally available for evaluation. Fourteen patients underwent bilateral TKAs but none concurrently. The patients provided written informed consent for print and electronic publication of their outcomes.

We recorded demographic data, type and severity of hemophilia, human immunodeficiency virus (HIV) status, hepatitis C virus (HCV) status, and Knee Society Scale (KSS) scores.¹⁴ KSS scores include Knee score (pain, range of motion [ROM], stability) and Function score (walking, stairs), both of which range from 0 (normal knee) to 100 (most affected knee). Prosthetic infection was classified (Segawa and colleagues¹⁵) as early or late, depending on timing of symptom onset (4 weeks after replacement surgery was the threshold used).

Patients received an intravenous bolus infusion of the deficient factor concentrate followed by continuous infusion to reach a plasma factor level of 100% just before surgery and during the first 7 postoperative days and 50% over the next 7 days (Table 1). Patients with a circulating inhibitor (3 overall) received bypassing agents FEIBA (FVIII inhibitor bypassing agent) or rFVIIa (recombinant factor VII activated) (Table 2). Patients were not given any antifibrinolytic treatment or thromboprophylaxis.

Surgery was performed in a standard surgical room. Patients were placed on the operating table in decubitus supinus position. A parapatellar medial incision was made on a bloodless surgical field (achieved with tourniquet ischemia). The prosthesis model used was always the cemented (gentamicin bone cement) NexGen (Zimmer). Patellar resurfacing was done in all cases (Figures 1A–1D). All TKAs were performed by Dr. Rodríguez-Merchán. Intravenous antibiotic prophylaxis was administered at anesthetic induction and during the first 48 hours after surgery (3 further doses). Active exercises were started on postoperative day 1. Joint load aided with 2 crutches was allowed starting on postoperative day 2.

Mean patient age was 38.2 years (range, 24-73 years). Of the 74 patients, 55 had a diagnosis of severe hemophilia A, and 19 had a diagnosis of severe hemophilia B. During the follow-up period, 23 patients died (mean time, 6.4 years; range, 4-9 years). Causes of death were acquired immune deficiency syndrome (AIDS), liver cirrhosis, and intracranial bleeding. Mean follow-up for the full series of patients was 8 years (range, 1-13 years).

Descriptive statistical analysis was performed with SPSS Windows Version 18.0. Prosthetic failure was regarded as implant removal for any reason. Student t test was used to compare continuous variables, and either χ² test or Fisher exact test was used to compare categorical variables. P < .05 (2-sided) was considered significant.

Results

Prosthetic survival rates with implant removal for any reason regarded as final endpoint was 92%. Causes of failure were prosthetic infection (6 cases, 6.8%) and loosening (2 cases, 2.2%). Of the 6 prosthetic infections, 5 were regarded as late and 1 as early. Late infections were successfully sorted by performing 2-stage revision TKA with the Constrained Condylar Knee (Zimmer). Acute infections were managed by open joint débridement and polyethylene exchange. Both cases of aseptic loosening of the TKA were successfully managed with 1-stage revision TKA using the same implant model (Figures 2A–2D).

Mean KSS Knee score improved from 79 before surgery to 36 after surgery, and mean KSS Function score improved from 63 to 33. KSS Pain score, which is included in the Knee score, 0 (no pain) to 50 (most severe pain), improved from 47 to 8. Patients receiving inhibitors and patients who were HIV- or HCV-positive did not have poorer outcomes relative to those of patients not receiving inhibitors and patients who were HIV- or HCV-negative. Patients with liver cirrhosis had a lower prosthetic survival rate and lower Knee scores.

Discussion

The prosthetic survival rate found in this study compares well with other reported rates for patients with hemophilia and other bleeding disorders. However, evidence regarding long-term prosthesis survival in TKAs performed for patients with hemophilia is limited. Table 3 summarizes the main reported series of patients with hemophilia with 10-year prosthetic survival rates, number of TKAs performed, and mean follow-up period; in all these series, implant removal for any reason was regarded as the final endpoint.^5-8,16,17 Mean follow-up in our study was 8 years. Clinical outcomes of TKA in patients with severe hemophilia and related disorders are expected to be inferior to those achieved in patients without a bleeding condition. The overall 10-year prosthetic survival rate for cemented TKA implants, as reported by the Norwegian Arthroplasty Register, was on the order of 93%.¹⁸ Mean age of our patients at time of surgery was only 38.2 years. TKAs performed in younger patients without a bleeding disorder have been associated with shorter implant survival times relative to those of elderly patients.¹⁹ Thus, Diduch and colleagues²⁰ reported a prosthetic survival rate of 87% at 18 years in 108 TKAs performed on patients under age 55 years. Lonner and colleagues²¹ reported a better implant survival rate (90% at 8 years) in a series of patients under age 40 years (32 TKAs). In a study by Duffy and colleagues,²² the implant survival rate was 85% at 15 years in patients under age 55 years (74 TKAs). The results from our retrospective case assessment are quite similar to the overall prosthetic survival rates reported for TKAs performed on patients without hemophilia.

Rates of periprosthetic infection after primary TKA in patients with hemophilia and other bleeding conditions are much higher (up to 11%), with a mean infection rate of 6.2% (range, 1% to 11%), consistent with the rate found in our series of patients (6.8%)^{7,16,17,23,24} (Table 4). This rate is much higher than that reported after primary TKA in patients without hemophilia but is similar to some rates reported for patients with hemophilia. In our experience, most periprosthetic infections (5/6) were sorted as late.

Late infection is a major concern after TKA in patients with hemophilia, and various factors have been hypothesized as contributing to the high prevalence. An important factor is the high rate of HIV-positive patients among patients with hemophilia—which acts as a strong predisposing factor because of the often low CD4 counts and associated immune deficiency,²⁵ but different reports have provided conflicting results in this respect.^5,6,12 We found no relationship between HIV status and risk for periprosthetic infection, but conclusions are limited by the low number of HIV-positive patients in our series (14/74, 18.9%). Our patients’ late periprosthetic infections were diagnosed several years after TKA, suggesting hematogenous spread of infection. Most of these patients either were on regular prophylactic factor infusions or were being treated on demand, which might entail a risk for contamination of infusions by skin bacteria from the puncture site. Therefore, having an aseptic technique for administering coagulation factor concentrates is of paramount importance for patients with hemophilia and a knee implant.

Another important complication of TKA surgery is aseptic loosening of the prosthesis. Aseptic loosening occurred in 2.2% of our patients, but higher rates have been reported elsewhere.^11,26 Rates of this complication increase over follow-up, and some authors have linked this complication to TKA polyethylene wear.²⁷ Development of a reactive and destructive bone–cement interface and microhemorrhages into such interface might be implicated in the higher rate of loosening observed among patients with hemophilia.²⁸

In the present study, preoperative and postoperative functional outcomes differed significantly. A modest postoperative total ROM of 69º to 79º has been reported by several authors.^5,6 Postoperative ROM may vary—may be slightly increased, remain unchanged, or may even be reduced.^4,23,26 Even though little improvement in total ROM is achieved after TKA, many authors have reported reduced flexion contracture and hence an easier gait. However, along with functional improvement, dramatic pain relief after TKA is perhaps the most remarkable aspect, and it has a strong effect on patient satisfaction after surgery.^5,7,8,18,23

Our study had 2 main limitations. First, it was a retrospective case series evaluation with the usual issues of potential inaccuracy of medical records and information bias. Second, the study did not include a control group.

Conclusion

The primary TKAs performed in our patients with hemophilia have had a good prosthetic survival rate. Even though such a result is slightly inferior to results in patients without hemophilia, our prosthetic survival rate is not significantly different from the rates reported in other, younger patient subsets. Late periprosthetic infections are a major concern, and taking precautions to avoid hematogenous spread of infections during factor concentrate infusions is strongly encouraged.

Chronic hemophilic arthropathy, a well-known complication of hemophilia, develops as a long-term consequence of recurrent joint bleeds resulting in synovial hypertrophy (chronic proliferative synovitis) and joint cartilage destruction. Hemophilic arthropathy mostly affects the knees, ankles, and elbows and causes chronic joint pain and functional impairment in relatively young patients who have not received adequate primary prophylactic replacement therapy with factor concentrates from early childhood.^1-3

In the late stages of hemophilic arthropathy of the knee, total knee arthroplasty (TKA) provides dramatic joint pain relief, improves knee functional status, and reduces rebleeding into the joint.^4-8 TKA performed on a patient with hemophilia was first reported in the mid-1970s.^9,10 In these cases, the surgical procedure itself is often complicated by severe fibrosis developing in the joint soft tissues, flexion joint contracture, and poor quality of the joint bone structures. Even though TKA significantly reduces joint pain in patients with chronic hemophilic arthropathy, some authors have achieved only modest functional outcomes and experienced a high rate of complications (infection, prosthetic loosening).^11-13 Data on TKA outcomes are still scarce, and most studies have enrolled a limited number of patients.

We retrospectively evaluated the outcomes of 88 primary TKAs performed on patients with severe hemophilia at a single institution. Clinical outcomes and complications were assessed with a special focus on prosthetic survival and infection.

Patients and Methods

Ninety-one primary TKAs were performed in 77 patients with severe hemophilia A and B (factor VIII [FVIII] and factor IX plasma concentration, <1% each) between January 1, 1999, and December 31, 2011, and the medical records of all these patients were thoroughly reviewed in 2013. The cases of 3 patients who died shortly after surgery were excluded from analysis. Thus, 88 TKAs and 74 patients (74 males) were finally available for evaluation. Fourteen patients underwent bilateral TKAs but none concurrently. The patients provided written informed consent for print and electronic publication of their outcomes.

We recorded demographic data, type and severity of hemophilia, human immunodeficiency virus (HIV) status, hepatitis C virus (HCV) status, and Knee Society Scale (KSS) scores.¹⁴ KSS scores include Knee score (pain, range of motion [ROM], stability) and Function score (walking, stairs), both of which range from 0 (normal knee) to 100 (most affected knee). Prosthetic infection was classified (Segawa and colleagues¹⁵) as early or late, depending on timing of symptom onset (4 weeks after replacement surgery was the threshold used).

Patients received an intravenous bolus infusion of the deficient factor concentrate followed by continuous infusion to reach a plasma factor level of 100% just before surgery and during the first 7 postoperative days and 50% over the next 7 days (Table 1). Patients with a circulating inhibitor (3 overall) received bypassing agents FEIBA (FVIII inhibitor bypassing agent) or rFVIIa (recombinant factor VII activated) (Table 2). Patients were not given any antifibrinolytic treatment or thromboprophylaxis.

Surgery was performed in a standard surgical room. Patients were placed on the operating table in decubitus supinus position. A parapatellar medial incision was made on a bloodless surgical field (achieved with tourniquet ischemia). The prosthesis model used was always the cemented (gentamicin bone cement) NexGen (Zimmer). Patellar resurfacing was done in all cases (Figures 1A–1D). All TKAs were performed by Dr. Rodríguez-Merchán. Intravenous antibiotic prophylaxis was administered at anesthetic induction and during the first 48 hours after surgery (3 further doses). Active exercises were started on postoperative day 1. Joint load aided with 2 crutches was allowed starting on postoperative day 2.

Mean patient age was 38.2 years (range, 24-73 years). Of the 74 patients, 55 had a diagnosis of severe hemophilia A, and 19 had a diagnosis of severe hemophilia B. During the follow-up period, 23 patients died (mean time, 6.4 years; range, 4-9 years). Causes of death were acquired immune deficiency syndrome (AIDS), liver cirrhosis, and intracranial bleeding. Mean follow-up for the full series of patients was 8 years (range, 1-13 years).

Descriptive statistical analysis was performed with SPSS Windows Version 18.0. Prosthetic failure was regarded as implant removal for any reason. Student t test was used to compare continuous variables, and either χ² test or Fisher exact test was used to compare categorical variables. P < .05 (2-sided) was considered significant.

Results

Prosthetic survival rates with implant removal for any reason regarded as final endpoint was 92%. Causes of failure were prosthetic infection (6 cases, 6.8%) and loosening (2 cases, 2.2%). Of the 6 prosthetic infections, 5 were regarded as late and 1 as early. Late infections were successfully sorted by performing 2-stage revision TKA with the Constrained Condylar Knee (Zimmer). Acute infections were managed by open joint débridement and polyethylene exchange. Both cases of aseptic loosening of the TKA were successfully managed with 1-stage revision TKA using the same implant model (Figures 2A–2D).

Mean KSS Knee score improved from 79 before surgery to 36 after surgery, and mean KSS Function score improved from 63 to 33. KSS Pain score, which is included in the Knee score, 0 (no pain) to 50 (most severe pain), improved from 47 to 8. Patients receiving inhibitors and patients who were HIV- or HCV-positive did not have poorer outcomes relative to those of patients not receiving inhibitors and patients who were HIV- or HCV-negative. Patients with liver cirrhosis had a lower prosthetic survival rate and lower Knee scores.

Discussion

The prosthetic survival rate found in this study compares well with other reported rates for patients with hemophilia and other bleeding disorders. However, evidence regarding long-term prosthesis survival in TKAs performed for patients with hemophilia is limited. Table 3 summarizes the main reported series of patients with hemophilia with 10-year prosthetic survival rates, number of TKAs performed, and mean follow-up period; in all these series, implant removal for any reason was regarded as the final endpoint.^5-8,16,17 Mean follow-up in our study was 8 years. Clinical outcomes of TKA in patients with severe hemophilia and related disorders are expected to be inferior to those achieved in patients without a bleeding condition. The overall 10-year prosthetic survival rate for cemented TKA implants, as reported by the Norwegian Arthroplasty Register, was on the order of 93%.¹⁸ Mean age of our patients at time of surgery was only 38.2 years. TKAs performed in younger patients without a bleeding disorder have been associated with shorter implant survival times relative to those of elderly patients.¹⁹ Thus, Diduch and colleagues²⁰ reported a prosthetic survival rate of 87% at 18 years in 108 TKAs performed on patients under age 55 years. Lonner and colleagues²¹ reported a better implant survival rate (90% at 8 years) in a series of patients under age 40 years (32 TKAs). In a study by Duffy and colleagues,²² the implant survival rate was 85% at 15 years in patients under age 55 years (74 TKAs). The results from our retrospective case assessment are quite similar to the overall prosthetic survival rates reported for TKAs performed on patients without hemophilia.

Rates of periprosthetic infection after primary TKA in patients with hemophilia and other bleeding conditions are much higher (up to 11%), with a mean infection rate of 6.2% (range, 1% to 11%), consistent with the rate found in our series of patients (6.8%)^{7,16,17,23,24} (Table 4). This rate is much higher than that reported after primary TKA in patients without hemophilia but is similar to some rates reported for patients with hemophilia. In our experience, most periprosthetic infections (5/6) were sorted as late.

Late infection is a major concern after TKA in patients with hemophilia, and various factors have been hypothesized as contributing to the high prevalence. An important factor is the high rate of HIV-positive patients among patients with hemophilia—which acts as a strong predisposing factor because of the often low CD4 counts and associated immune deficiency,²⁵ but different reports have provided conflicting results in this respect.^5,6,12 We found no relationship between HIV status and risk for periprosthetic infection, but conclusions are limited by the low number of HIV-positive patients in our series (14/74, 18.9%). Our patients’ late periprosthetic infections were diagnosed several years after TKA, suggesting hematogenous spread of infection. Most of these patients either were on regular prophylactic factor infusions or were being treated on demand, which might entail a risk for contamination of infusions by skin bacteria from the puncture site. Therefore, having an aseptic technique for administering coagulation factor concentrates is of paramount importance for patients with hemophilia and a knee implant.

Another important complication of TKA surgery is aseptic loosening of the prosthesis. Aseptic loosening occurred in 2.2% of our patients, but higher rates have been reported elsewhere.^11,26 Rates of this complication increase over follow-up, and some authors have linked this complication to TKA polyethylene wear.²⁷ Development of a reactive and destructive bone–cement interface and microhemorrhages into such interface might be implicated in the higher rate of loosening observed among patients with hemophilia.²⁸

In the present study, preoperative and postoperative functional outcomes differed significantly. A modest postoperative total ROM of 69º to 79º has been reported by several authors.^5,6 Postoperative ROM may vary—may be slightly increased, remain unchanged, or may even be reduced.^4,23,26 Even though little improvement in total ROM is achieved after TKA, many authors have reported reduced flexion contracture and hence an easier gait. However, along with functional improvement, dramatic pain relief after TKA is perhaps the most remarkable aspect, and it has a strong effect on patient satisfaction after surgery.^5,7,8,18,23

Our study had 2 main limitations. First, it was a retrospective case series evaluation with the usual issues of potential inaccuracy of medical records and information bias. Second, the study did not include a control group.

Conclusion

The primary TKAs performed in our patients with hemophilia have had a good prosthetic survival rate. Even though such a result is slightly inferior to results in patients without hemophilia, our prosthetic survival rate is not significantly different from the rates reported in other, younger patient subsets. Late periprosthetic infections are a major concern, and taking precautions to avoid hematogenous spread of infections during factor concentrate infusions is strongly encouraged.

Chronic hemophilic arthropathy, a well-known complication of hemophilia, develops as a long-term consequence of recurrent joint bleeds resulting in synovial hypertrophy (chronic proliferative synovitis) and joint cartilage destruction. Hemophilic arthropathy mostly affects the knees, ankles, and elbows and causes chronic joint pain and functional impairment in relatively young patients who have not received adequate primary prophylactic replacement therapy with factor concentrates from early childhood.^1-3

In the late stages of hemophilic arthropathy of the knee, total knee arthroplasty (TKA) provides dramatic joint pain relief, improves knee functional status, and reduces rebleeding into the joint.^4-8 TKA performed on a patient with hemophilia was first reported in the mid-1970s.^9,10 In these cases, the surgical procedure itself is often complicated by severe fibrosis developing in the joint soft tissues, flexion joint contracture, and poor quality of the joint bone structures. Even though TKA significantly reduces joint pain in patients with chronic hemophilic arthropathy, some authors have achieved only modest functional outcomes and experienced a high rate of complications (infection, prosthetic loosening).^11-13 Data on TKA outcomes are still scarce, and most studies have enrolled a limited number of patients.

We retrospectively evaluated the outcomes of 88 primary TKAs performed on patients with severe hemophilia at a single institution. Clinical outcomes and complications were assessed with a special focus on prosthetic survival and infection.

Patients and Methods

Ninety-one primary TKAs were performed in 77 patients with severe hemophilia A and B (factor VIII [FVIII] and factor IX plasma concentration, <1% each) between January 1, 1999, and December 31, 2011, and the medical records of all these patients were thoroughly reviewed in 2013. The cases of 3 patients who died shortly after surgery were excluded from analysis. Thus, 88 TKAs and 74 patients (74 males) were finally available for evaluation. Fourteen patients underwent bilateral TKAs but none concurrently. The patients provided written informed consent for print and electronic publication of their outcomes.

We recorded demographic data, type and severity of hemophilia, human immunodeficiency virus (HIV) status, hepatitis C virus (HCV) status, and Knee Society Scale (KSS) scores.¹⁴ KSS scores include Knee score (pain, range of motion [ROM], stability) and Function score (walking, stairs), both of which range from 0 (normal knee) to 100 (most affected knee). Prosthetic infection was classified (Segawa and colleagues¹⁵) as early or late, depending on timing of symptom onset (4 weeks after replacement surgery was the threshold used).

Patients received an intravenous bolus infusion of the deficient factor concentrate followed by continuous infusion to reach a plasma factor level of 100% just before surgery and during the first 7 postoperative days and 50% over the next 7 days (Table 1). Patients with a circulating inhibitor (3 overall) received bypassing agents FEIBA (FVIII inhibitor bypassing agent) or rFVIIa (recombinant factor VII activated) (Table 2). Patients were not given any antifibrinolytic treatment or thromboprophylaxis.

Surgery was performed in a standard surgical room. Patients were placed on the operating table in decubitus supinus position. A parapatellar medial incision was made on a bloodless surgical field (achieved with tourniquet ischemia). The prosthesis model used was always the cemented (gentamicin bone cement) NexGen (Zimmer). Patellar resurfacing was done in all cases (Figures 1A–1D). All TKAs were performed by Dr. Rodríguez-Merchán. Intravenous antibiotic prophylaxis was administered at anesthetic induction and during the first 48 hours after surgery (3 further doses). Active exercises were started on postoperative day 1. Joint load aided with 2 crutches was allowed starting on postoperative day 2.

Mean patient age was 38.2 years (range, 24-73 years). Of the 74 patients, 55 had a diagnosis of severe hemophilia A, and 19 had a diagnosis of severe hemophilia B. During the follow-up period, 23 patients died (mean time, 6.4 years; range, 4-9 years). Causes of death were acquired immune deficiency syndrome (AIDS), liver cirrhosis, and intracranial bleeding. Mean follow-up for the full series of patients was 8 years (range, 1-13 years).

Descriptive statistical analysis was performed with SPSS Windows Version 18.0. Prosthetic failure was regarded as implant removal for any reason. Student t test was used to compare continuous variables, and either χ² test or Fisher exact test was used to compare categorical variables. P < .05 (2-sided) was considered significant.

Results

Prosthetic survival rates with implant removal for any reason regarded as final endpoint was 92%. Causes of failure were prosthetic infection (6 cases, 6.8%) and loosening (2 cases, 2.2%). Of the 6 prosthetic infections, 5 were regarded as late and 1 as early. Late infections were successfully sorted by performing 2-stage revision TKA with the Constrained Condylar Knee (Zimmer). Acute infections were managed by open joint débridement and polyethylene exchange. Both cases of aseptic loosening of the TKA were successfully managed with 1-stage revision TKA using the same implant model (Figures 2A–2D).

Mean KSS Knee score improved from 79 before surgery to 36 after surgery, and mean KSS Function score improved from 63 to 33. KSS Pain score, which is included in the Knee score, 0 (no pain) to 50 (most severe pain), improved from 47 to 8. Patients receiving inhibitors and patients who were HIV- or HCV-positive did not have poorer outcomes relative to those of patients not receiving inhibitors and patients who were HIV- or HCV-negative. Patients with liver cirrhosis had a lower prosthetic survival rate and lower Knee scores.

Discussion

The prosthetic survival rate found in this study compares well with other reported rates for patients with hemophilia and other bleeding disorders. However, evidence regarding long-term prosthesis survival in TKAs performed for patients with hemophilia is limited. Table 3 summarizes the main reported series of patients with hemophilia with 10-year prosthetic survival rates, number of TKAs performed, and mean follow-up period; in all these series, implant removal for any reason was regarded as the final endpoint.^5-8,16,17 Mean follow-up in our study was 8 years. Clinical outcomes of TKA in patients with severe hemophilia and related disorders are expected to be inferior to those achieved in patients without a bleeding condition. The overall 10-year prosthetic survival rate for cemented TKA implants, as reported by the Norwegian Arthroplasty Register, was on the order of 93%.¹⁸ Mean age of our patients at time of surgery was only 38.2 years. TKAs performed in younger patients without a bleeding disorder have been associated with shorter implant survival times relative to those of elderly patients.¹⁹ Thus, Diduch and colleagues²⁰ reported a prosthetic survival rate of 87% at 18 years in 108 TKAs performed on patients under age 55 years. Lonner and colleagues²¹ reported a better implant survival rate (90% at 8 years) in a series of patients under age 40 years (32 TKAs). In a study by Duffy and colleagues,²² the implant survival rate was 85% at 15 years in patients under age 55 years (74 TKAs). The results from our retrospective case assessment are quite similar to the overall prosthetic survival rates reported for TKAs performed on patients without hemophilia.

Rates of periprosthetic infection after primary TKA in patients with hemophilia and other bleeding conditions are much higher (up to 11%), with a mean infection rate of 6.2% (range, 1% to 11%), consistent with the rate found in our series of patients (6.8%)^{7,16,17,23,24} (Table 4). This rate is much higher than that reported after primary TKA in patients without hemophilia but is similar to some rates reported for patients with hemophilia. In our experience, most periprosthetic infections (5/6) were sorted as late.

Late infection is a major concern after TKA in patients with hemophilia, and various factors have been hypothesized as contributing to the high prevalence. An important factor is the high rate of HIV-positive patients among patients with hemophilia—which acts as a strong predisposing factor because of the often low CD4 counts and associated immune deficiency,²⁵ but different reports have provided conflicting results in this respect.^5,6,12 We found no relationship between HIV status and risk for periprosthetic infection, but conclusions are limited by the low number of HIV-positive patients in our series (14/74, 18.9%). Our patients’ late periprosthetic infections were diagnosed several years after TKA, suggesting hematogenous spread of infection. Most of these patients either were on regular prophylactic factor infusions or were being treated on demand, which might entail a risk for contamination of infusions by skin bacteria from the puncture site. Therefore, having an aseptic technique for administering coagulation factor concentrates is of paramount importance for patients with hemophilia and a knee implant.

Another important complication of TKA surgery is aseptic loosening of the prosthesis. Aseptic loosening occurred in 2.2% of our patients, but higher rates have been reported elsewhere.^11,26 Rates of this complication increase over follow-up, and some authors have linked this complication to TKA polyethylene wear.²⁷ Development of a reactive and destructive bone–cement interface and microhemorrhages into such interface might be implicated in the higher rate of loosening observed among patients with hemophilia.²⁸

In the present study, preoperative and postoperative functional outcomes differed significantly. A modest postoperative total ROM of 69º to 79º has been reported by several authors.^5,6 Postoperative ROM may vary—may be slightly increased, remain unchanged, or may even be reduced.^4,23,26 Even though little improvement in total ROM is achieved after TKA, many authors have reported reduced flexion contracture and hence an easier gait. However, along with functional improvement, dramatic pain relief after TKA is perhaps the most remarkable aspect, and it has a strong effect on patient satisfaction after surgery.^5,7,8,18,23

Our study had 2 main limitations. First, it was a retrospective case series evaluation with the usual issues of potential inaccuracy of medical records and information bias. Second, the study did not include a control group.

Conclusion

The primary TKAs performed in our patients with hemophilia have had a good prosthetic survival rate. Even though such a result is slightly inferior to results in patients without hemophilia, our prosthetic survival rate is not significantly different from the rates reported in other, younger patient subsets. Late periprosthetic infections are a major concern, and taking precautions to avoid hematogenous spread of infections during factor concentrate infusions is strongly encouraged.

Many organizations and work groups have issued recommendations regarding which patients should undergo bone densitometry. In 2004, the US Surgeon General recommended bone mineral density (BMD) evaluation for all women over age 65 years and for women and men with fragility fractures.¹ The Centers for Medicare & Medicaid Services recommended BMD assessment for estrogen-deficient patients, for patients with vertebral abnormalities or hyperparathyroidism, and for patients receiving either steroid therapy or osteoporosis medications approved by the US Food and Drug Administration.² The US Preventive Services Task Force and the National Osteoporosis Foundation each recommended screening for all women age 65 years or older and for postmenopausal women (age, 60-64 years) at high risk.^3,4 The International Society for Clinical Densitometry (ISCD) recommended screening for all women age 65 years or older, all men age 70 years or older, and high-risk women under age 65 years.⁵

These current recommendations for BMD evaluation focus on women over age 65 years. More recent studies of postmenopausal women with distal radius fractures (DRFs) have found that both younger women (age, 45-65 years) and older women (age, ≥65 years) can have lower BMD and increased risk for hip and spine fracture.^6,7 The authors of those studies recommended that all postmenopausal women with DRFs be evaluated for low BMD and that fracture prevention treatment be initiated. Earnshaw and colleagues⁸ and Oyen and colleagues⁹ found that men and women (age, ≥50 years) with DRFs had low BMD and elevated 10-year fracture rates. They concluded that BMD should be evaluated and treated in all DRF patients age 50 years or older. Other studies have shown low BMD in the contralateral distal radius of patients of all ages who presented with Colles fractures.^10,11 These 2 studies did not measure spine or hip BMD.

The literature on BMD of younger women with DRFs is limited, relying solely on data collected for the contralateral distal radius.^10,11 The ISCD recommended measuring both hip and spine BMD in premenopausal women. They also stated that z scores, not t scores, should be used for premenopausal women.⁵ The causes of low BMD in women over age 55 years are primarily nutritional deficiency and normal aging.¹ In younger females, low BMD results from secondary causes, such as diet, medications, medical conditions, and endocrine disorders. When the secondary cause of low BMD can be identified and treated, osteoporosis can be stopped and even reversed in younger patients.^12-14 Low BMD is more amenable to treatment in younger patients than in postmenopausal women. Younger patients with low BMD carry a higher lifetime fracture risk because they have more years of life with low BMD; therefore, early identification and treatment have a more significant impact on fracture prevention in these patients.

In the present study, we determined the prevalence of osteoporosis and osteopenia in younger women (age, 35-50 years) with DRFs and compared BMD measurements from younger women (age, 35-50 years) and older women (age, >50 years) with DRFs. The main goal was to determine which patients should be referred for bone densitometry and subsequent treatment.

Patients and Methods

This study received institutional review board approval. During a 5-year period (January 2005–August 2010), we prospectively collected dual-energy x-ray absorptiometry (DXA) scans for 128 women (age, >35 years) who presented with DRFs to our level I trauma center. Age ranged from 35 to 86 years. Data on mechanism of injury, treatment, and body mass index (BMI) were collected. The 128 patients were divided into a younger group (47 women; age range, 35-50 years; mean age, 44 years) and an older group (81 women; age, ≥51 years; mean age, 61 years). Mean BMI was 29.3 in the younger group and 28.8 in the older group (P = .88) (Table).

BMD was measured with a General Electric Lunar Prodigy Advance scanner that was tested annually for accuracy and precision. BMD of hips and lumbar spines was measured with a 76-Kv x-ray source. All DXA scans were analyzed by the same physician. BMD was omitted in cases of patients with a history of lumbar spine or hip fracture.

Two-sample Student t test was used to compare the 2 groups’ data. When multiple groups were being compared, analysis of variance was used. Spearman rank-order test was used to calculate a correlation coefficient for evaluation of the relationships between age and BMD.

Results

Mean lumbar spine (L1–L4) BMD was 1.12 in the younger group and 1.063 in the older group (P = .02); t scores were –0.63 and –1.132, respectively (P = .02); and mean z scores were –0.69 and –0.61, respectively (P = .81). Mean femoral neck BMD was 0.91 in the younger group and 0.80 in the older group (P < .05); t scores were –0.87 and –1.65, respectively (P < .01), and mean femoral neck z scores were –0.69 and –0.67, respectively (P = .92).

To further analyze BMD of specific age groups, we divided patients by decade: 35-39, 40-49, 50-59, 60-69, 70-79, 80-89 years. Among all 6 decades, there were no statistically significant differences between hip z scores (P = .83) (Figure 1). Spearman rank-order correlation test showed a moderate inverse correlation between age and femoral neck BMD (R = –0.42) and t score (R = –0.43). There was a weak correlation between increasing age and decreasing spine BMD, t score, and z score (Rs = –0.27, –0.31, 0.03). There was no correlation between age and femoral neck z score (R = –0.04).

According to the WHO classification system, 11 (23%) of the 47 women in the younger group were osteopenic, and 8 (17%) were osteoporotic, based on spine BMD. Hip BMD values indicated that 20 patients (43%) were osteopenic, and 3 (6%) were osteoporotic. One patient in the younger group had a hip z score of less than –2, and 14 patients (39%) had a hip z score between –2 and –1. Six patients (18%) had a spine z score of less than –2, and 6 patients (18%) had a spine z score between –2 and –1. Of the 81 older patients, 22 (27%) were osteopenic, and 21 (26%) were osteoporotic, according to spine measurements. The femoral neck data indicated that 39 (48%) of the older patients were osteopenic, and 22 (27%) were osteoporotic.

In both groups, mechanisms of injury were identified. Of the 47 younger patients, 26 fell from standing, 7 fell from a height of more than 6 feet, and 14 were injured in motor vehicle collisions (MVCs). Of the 81 older patients, 2 sustained a direct blow, 64 fell from standing, 4 fell from a height of more than 6 feet, and 11 were injured in MVCs. The differences in z scores based on mechanism of injury were not statistically significant (P = .22) (Figure 2).

Discussion

Several studies have shown that older women with DRFs have low BMD in the spine and femoral neck.^8,9 These studies focused on older women who sustained low-energy fractures caused by a fall from a standing height. Studies of younger women with DRFs focused on BMD of the contralateral distal radius, not the spine or femoral neck.^10,11 Those study groups also had low BMD. Findings from a multitude of studies have established that patients who are older than 50 years when they sustain distal radius fragility fractures should be referred for bone densitometry studies, and there is increasing evidence that younger patients with fragility fractures should undergo this evaluation as well.

The present study was designed to expand the range of patients and mechanisms of injury. Women in this study were 35 years or older. In addition to collecting data from patients injured in a fall from standing, we examined the medical records of women injured in MVCs, in falls from heights of more than 6 feet, and from direct trauma to the wrist. We measured the BMD of the spine and femoral neck and of the contralateral distal radius.

For this discussion, several key points should be made about BMD evaluation in younger versus older women. Most organizations caution against using spine BMD in older women. The ISCD, however, recommended measuring both hip and spine BMD; whereas BMD can be falsely elevated by spine osteoarthritis in older patients, spine BMD measurements are accurate in younger patients not affected by osteoarthritis. The ISCD also stipulated that z scores should be used in examining BMD in younger patients. The z score is a value of how many standard deviations BMD differs from a matched population of the same age, sex, ethnicity, and weight. The t score, which is useful in evaluating older patients, compares a patient’s BMD with that of an average 30-year-old.¹²

According to the WHO classification system (intended for older women), osteopenia is indicated by a t score between –1.0 and –2.5, and osteoporosis is indicated by a t score of less than –2.5. In the present study, about 43% of the younger patients (age, 35-50 years) with DRFs were osteopenic, and 6% of these patients were osteoporotic. In concert with previous studies,⁹ 48% of our older women (age, >50 years) with DRFs were osteopenic, and 27% were osteoporotic. The difference in mean spinal z scores between the younger and older groups was not statistically significant (P = .81).

As mentioned, when examining BMD of younger patients, it is imperative to use spine z scores. About 18% of our younger patients had a z score of less than –2, and 18% had a z score between –2 and –1. In our comparison of patients from 5 different age decades (range, 35-79 years), there was no statistically significant difference in z scores (P = .83). In addition, there was no correlation between increasing age and decreasing z score (R = –0.04).

Secondary causes of osteoporosis have been documented in 30% of premenopausal women and 55% of men with vertebral fractures.^13-15 Primary osteoporosis results from the normal aging process; secondary osteoporosis results from reversible causes, including medications, gastrointestinal disorders, renal disease, endocrine disorders, and sedentary lifestyle.^15,16 When a secondary cause of osteoporosis is identified, treatment can be initiated to increase BMD. As younger patients can reverse bone loss and even increase BMD, it is important to identify reversible causes of osteopenia and osteoporosis in this age group. It is well documented that both younger and older patients with DRFs are at increased risk for subsequent fractures.⁶ Preventing further bone loss at a younger age may drastically decrease lifetime fracture risk.^12,17

Most previous studies of BMD in women were limited to patients with DRFs caused by a low-energy mechanism or by a fall from standing. Current recommendations for BMD testing focus on postmenopausal women who have sustained a fragility or low-energy DRF. When an osteoporotic or osteopenic patient’s distal radius is subjected to a high-energy force, a fracture is likely. Therefore, we expanded our study to include high-energy mechanisms of injury. Our analysis of BMD in patients with DRFs sustained in MVCs indicated that 12% of this group were osteoporotic, and 44% were osteopenic. Forty-three percent of our younger patients with a DRF fractured in a MVC were osteopenic, and 6% were osteoporotic. Among 4 mechanisms of injury for DRFs, there was no statistically significant difference in z scores (P = .22) (Figure 2). This provides evidence that a significant portion of patients with DRFs from both high- and low-energy mechanisms are osteoporotic or osteopenic. Patients with DRFs sustained in MVCs or in falls from heights of more than 6 feet should be referred for BMD evaluation.         

Conclusion

A significant proportion of younger patients with DRFs are osteopenic or osteoporotic (43% and 6%, respectively), and their z scores are comparable to those of older patients with DRFs. There was no statistically significant difference in BMD z scores between younger and older patients and no difference in mechanisms of injury. This is evidence that younger patients with DRFs caused by a high- or low-energy mechanism of injury should undergo both DXA scan and BMD evaluation. If osteoporosis or osteopenia can be diagnosed at an earlier age, and if these patients can be properly treated, subsequent fractures could be prevented. The present study provides evidence supporting a simplification of the current recommendations for BMD evaluation: All women with DRFs should undergo bone densitometry.

Many organizations and work groups have issued recommendations regarding which patients should undergo bone densitometry. In 2004, the US Surgeon General recommended bone mineral density (BMD) evaluation for all women over age 65 years and for women and men with fragility fractures.¹ The Centers for Medicare & Medicaid Services recommended BMD assessment for estrogen-deficient patients, for patients with vertebral abnormalities or hyperparathyroidism, and for patients receiving either steroid therapy or osteoporosis medications approved by the US Food and Drug Administration.² The US Preventive Services Task Force and the National Osteoporosis Foundation each recommended screening for all women age 65 years or older and for postmenopausal women (age, 60-64 years) at high risk.^3,4 The International Society for Clinical Densitometry (ISCD) recommended screening for all women age 65 years or older, all men age 70 years or older, and high-risk women under age 65 years.⁵

These current recommendations for BMD evaluation focus on women over age 65 years. More recent studies of postmenopausal women with distal radius fractures (DRFs) have found that both younger women (age, 45-65 years) and older women (age, ≥65 years) can have lower BMD and increased risk for hip and spine fracture.^6,7 The authors of those studies recommended that all postmenopausal women with DRFs be evaluated for low BMD and that fracture prevention treatment be initiated. Earnshaw and colleagues⁸ and Oyen and colleagues⁹ found that men and women (age, ≥50 years) with DRFs had low BMD and elevated 10-year fracture rates. They concluded that BMD should be evaluated and treated in all DRF patients age 50 years or older. Other studies have shown low BMD in the contralateral distal radius of patients of all ages who presented with Colles fractures.^10,11 These 2 studies did not measure spine or hip BMD.

The literature on BMD of younger women with DRFs is limited, relying solely on data collected for the contralateral distal radius.^10,11 The ISCD recommended measuring both hip and spine BMD in premenopausal women. They also stated that z scores, not t scores, should be used for premenopausal women.⁵ The causes of low BMD in women over age 55 years are primarily nutritional deficiency and normal aging.¹ In younger females, low BMD results from secondary causes, such as diet, medications, medical conditions, and endocrine disorders. When the secondary cause of low BMD can be identified and treated, osteoporosis can be stopped and even reversed in younger patients.^12-14 Low BMD is more amenable to treatment in younger patients than in postmenopausal women. Younger patients with low BMD carry a higher lifetime fracture risk because they have more years of life with low BMD; therefore, early identification and treatment have a more significant impact on fracture prevention in these patients.

In the present study, we determined the prevalence of osteoporosis and osteopenia in younger women (age, 35-50 years) with DRFs and compared BMD measurements from younger women (age, 35-50 years) and older women (age, >50 years) with DRFs. The main goal was to determine which patients should be referred for bone densitometry and subsequent treatment.

Patients and Methods

This study received institutional review board approval. During a 5-year period (January 2005–August 2010), we prospectively collected dual-energy x-ray absorptiometry (DXA) scans for 128 women (age, >35 years) who presented with DRFs to our level I trauma center. Age ranged from 35 to 86 years. Data on mechanism of injury, treatment, and body mass index (BMI) were collected. The 128 patients were divided into a younger group (47 women; age range, 35-50 years; mean age, 44 years) and an older group (81 women; age, ≥51 years; mean age, 61 years). Mean BMI was 29.3 in the younger group and 28.8 in the older group (P = .88) (Table).

BMD was measured with a General Electric Lunar Prodigy Advance scanner that was tested annually for accuracy and precision. BMD of hips and lumbar spines was measured with a 76-Kv x-ray source. All DXA scans were analyzed by the same physician. BMD was omitted in cases of patients with a history of lumbar spine or hip fracture.

Two-sample Student t test was used to compare the 2 groups’ data. When multiple groups were being compared, analysis of variance was used. Spearman rank-order test was used to calculate a correlation coefficient for evaluation of the relationships between age and BMD.

Results

Mean lumbar spine (L1–L4) BMD was 1.12 in the younger group and 1.063 in the older group (P = .02); t scores were –0.63 and –1.132, respectively (P = .02); and mean z scores were –0.69 and –0.61, respectively (P = .81). Mean femoral neck BMD was 0.91 in the younger group and 0.80 in the older group (P < .05); t scores were –0.87 and –1.65, respectively (P < .01), and mean femoral neck z scores were –0.69 and –0.67, respectively (P = .92).

To further analyze BMD of specific age groups, we divided patients by decade: 35-39, 40-49, 50-59, 60-69, 70-79, 80-89 years. Among all 6 decades, there were no statistically significant differences between hip z scores (P = .83) (Figure 1). Spearman rank-order correlation test showed a moderate inverse correlation between age and femoral neck BMD (R = –0.42) and t score (R = –0.43). There was a weak correlation between increasing age and decreasing spine BMD, t score, and z score (Rs = –0.27, –0.31, 0.03). There was no correlation between age and femoral neck z score (R = –0.04).

According to the WHO classification system, 11 (23%) of the 47 women in the younger group were osteopenic, and 8 (17%) were osteoporotic, based on spine BMD. Hip BMD values indicated that 20 patients (43%) were osteopenic, and 3 (6%) were osteoporotic. One patient in the younger group had a hip z score of less than –2, and 14 patients (39%) had a hip z score between –2 and –1. Six patients (18%) had a spine z score of less than –2, and 6 patients (18%) had a spine z score between –2 and –1. Of the 81 older patients, 22 (27%) were osteopenic, and 21 (26%) were osteoporotic, according to spine measurements. The femoral neck data indicated that 39 (48%) of the older patients were osteopenic, and 22 (27%) were osteoporotic.

In both groups, mechanisms of injury were identified. Of the 47 younger patients, 26 fell from standing, 7 fell from a height of more than 6 feet, and 14 were injured in motor vehicle collisions (MVCs). Of the 81 older patients, 2 sustained a direct blow, 64 fell from standing, 4 fell from a height of more than 6 feet, and 11 were injured in MVCs. The differences in z scores based on mechanism of injury were not statistically significant (P = .22) (Figure 2).

Discussion

Several studies have shown that older women with DRFs have low BMD in the spine and femoral neck.^8,9 These studies focused on older women who sustained low-energy fractures caused by a fall from a standing height. Studies of younger women with DRFs focused on BMD of the contralateral distal radius, not the spine or femoral neck.^10,11 Those study groups also had low BMD. Findings from a multitude of studies have established that patients who are older than 50 years when they sustain distal radius fragility fractures should be referred for bone densitometry studies, and there is increasing evidence that younger patients with fragility fractures should undergo this evaluation as well.

The present study was designed to expand the range of patients and mechanisms of injury. Women in this study were 35 years or older. In addition to collecting data from patients injured in a fall from standing, we examined the medical records of women injured in MVCs, in falls from heights of more than 6 feet, and from direct trauma to the wrist. We measured the BMD of the spine and femoral neck and of the contralateral distal radius.

For this discussion, several key points should be made about BMD evaluation in younger versus older women. Most organizations caution against using spine BMD in older women. The ISCD, however, recommended measuring both hip and spine BMD; whereas BMD can be falsely elevated by spine osteoarthritis in older patients, spine BMD measurements are accurate in younger patients not affected by osteoarthritis. The ISCD also stipulated that z scores should be used in examining BMD in younger patients. The z score is a value of how many standard deviations BMD differs from a matched population of the same age, sex, ethnicity, and weight. The t score, which is useful in evaluating older patients, compares a patient’s BMD with that of an average 30-year-old.¹²

According to the WHO classification system (intended for older women), osteopenia is indicated by a t score between –1.0 and –2.5, and osteoporosis is indicated by a t score of less than –2.5. In the present study, about 43% of the younger patients (age, 35-50 years) with DRFs were osteopenic, and 6% of these patients were osteoporotic. In concert with previous studies,⁹ 48% of our older women (age, >50 years) with DRFs were osteopenic, and 27% were osteoporotic. The difference in mean spinal z scores between the younger and older groups was not statistically significant (P = .81).

As mentioned, when examining BMD of younger patients, it is imperative to use spine z scores. About 18% of our younger patients had a z score of less than –2, and 18% had a z score between –2 and –1. In our comparison of patients from 5 different age decades (range, 35-79 years), there was no statistically significant difference in z scores (P = .83). In addition, there was no correlation between increasing age and decreasing z score (R = –0.04).

Secondary causes of osteoporosis have been documented in 30% of premenopausal women and 55% of men with vertebral fractures.^13-15 Primary osteoporosis results from the normal aging process; secondary osteoporosis results from reversible causes, including medications, gastrointestinal disorders, renal disease, endocrine disorders, and sedentary lifestyle.^15,16 When a secondary cause of osteoporosis is identified, treatment can be initiated to increase BMD. As younger patients can reverse bone loss and even increase BMD, it is important to identify reversible causes of osteopenia and osteoporosis in this age group. It is well documented that both younger and older patients with DRFs are at increased risk for subsequent fractures.⁶ Preventing further bone loss at a younger age may drastically decrease lifetime fracture risk.^12,17

Most previous studies of BMD in women were limited to patients with DRFs caused by a low-energy mechanism or by a fall from standing. Current recommendations for BMD testing focus on postmenopausal women who have sustained a fragility or low-energy DRF. When an osteoporotic or osteopenic patient’s distal radius is subjected to a high-energy force, a fracture is likely. Therefore, we expanded our study to include high-energy mechanisms of injury. Our analysis of BMD in patients with DRFs sustained in MVCs indicated that 12% of this group were osteoporotic, and 44% were osteopenic. Forty-three percent of our younger patients with a DRF fractured in a MVC were osteopenic, and 6% were osteoporotic. Among 4 mechanisms of injury for DRFs, there was no statistically significant difference in z scores (P = .22) (Figure 2). This provides evidence that a significant portion of patients with DRFs from both high- and low-energy mechanisms are osteoporotic or osteopenic. Patients with DRFs sustained in MVCs or in falls from heights of more than 6 feet should be referred for BMD evaluation.         

Conclusion

A significant proportion of younger patients with DRFs are osteopenic or osteoporotic (43% and 6%, respectively), and their z scores are comparable to those of older patients with DRFs. There was no statistically significant difference in BMD z scores between younger and older patients and no difference in mechanisms of injury. This is evidence that younger patients with DRFs caused by a high- or low-energy mechanism of injury should undergo both DXA scan and BMD evaluation. If osteoporosis or osteopenia can be diagnosed at an earlier age, and if these patients can be properly treated, subsequent fractures could be prevented. The present study provides evidence supporting a simplification of the current recommendations for BMD evaluation: All women with DRFs should undergo bone densitometry.

Many organizations and work groups have issued recommendations regarding which patients should undergo bone densitometry. In 2004, the US Surgeon General recommended bone mineral density (BMD) evaluation for all women over age 65 years and for women and men with fragility fractures.¹ The Centers for Medicare & Medicaid Services recommended BMD assessment for estrogen-deficient patients, for patients with vertebral abnormalities or hyperparathyroidism, and for patients receiving either steroid therapy or osteoporosis medications approved by the US Food and Drug Administration.² The US Preventive Services Task Force and the National Osteoporosis Foundation each recommended screening for all women age 65 years or older and for postmenopausal women (age, 60-64 years) at high risk.^3,4 The International Society for Clinical Densitometry (ISCD) recommended screening for all women age 65 years or older, all men age 70 years or older, and high-risk women under age 65 years.⁵

These current recommendations for BMD evaluation focus on women over age 65 years. More recent studies of postmenopausal women with distal radius fractures (DRFs) have found that both younger women (age, 45-65 years) and older women (age, ≥65 years) can have lower BMD and increased risk for hip and spine fracture.^6,7 The authors of those studies recommended that all postmenopausal women with DRFs be evaluated for low BMD and that fracture prevention treatment be initiated. Earnshaw and colleagues⁸ and Oyen and colleagues⁹ found that men and women (age, ≥50 years) with DRFs had low BMD and elevated 10-year fracture rates. They concluded that BMD should be evaluated and treated in all DRF patients age 50 years or older. Other studies have shown low BMD in the contralateral distal radius of patients of all ages who presented with Colles fractures.^10,11 These 2 studies did not measure spine or hip BMD.

The literature on BMD of younger women with DRFs is limited, relying solely on data collected for the contralateral distal radius.^10,11 The ISCD recommended measuring both hip and spine BMD in premenopausal women. They also stated that z scores, not t scores, should be used for premenopausal women.⁵ The causes of low BMD in women over age 55 years are primarily nutritional deficiency and normal aging.¹ In younger females, low BMD results from secondary causes, such as diet, medications, medical conditions, and endocrine disorders. When the secondary cause of low BMD can be identified and treated, osteoporosis can be stopped and even reversed in younger patients.^12-14 Low BMD is more amenable to treatment in younger patients than in postmenopausal women. Younger patients with low BMD carry a higher lifetime fracture risk because they have more years of life with low BMD; therefore, early identification and treatment have a more significant impact on fracture prevention in these patients.

In the present study, we determined the prevalence of osteoporosis and osteopenia in younger women (age, 35-50 years) with DRFs and compared BMD measurements from younger women (age, 35-50 years) and older women (age, >50 years) with DRFs. The main goal was to determine which patients should be referred for bone densitometry and subsequent treatment.

Patients and Methods

This study received institutional review board approval. During a 5-year period (January 2005–August 2010), we prospectively collected dual-energy x-ray absorptiometry (DXA) scans for 128 women (age, >35 years) who presented with DRFs to our level I trauma center. Age ranged from 35 to 86 years. Data on mechanism of injury, treatment, and body mass index (BMI) were collected. The 128 patients were divided into a younger group (47 women; age range, 35-50 years; mean age, 44 years) and an older group (81 women; age, ≥51 years; mean age, 61 years). Mean BMI was 29.3 in the younger group and 28.8 in the older group (P = .88) (Table).

BMD was measured with a General Electric Lunar Prodigy Advance scanner that was tested annually for accuracy and precision. BMD of hips and lumbar spines was measured with a 76-Kv x-ray source. All DXA scans were analyzed by the same physician. BMD was omitted in cases of patients with a history of lumbar spine or hip fracture.

Two-sample Student t test was used to compare the 2 groups’ data. When multiple groups were being compared, analysis of variance was used. Spearman rank-order test was used to calculate a correlation coefficient for evaluation of the relationships between age and BMD.

Results

Mean lumbar spine (L1–L4) BMD was 1.12 in the younger group and 1.063 in the older group (P = .02); t scores were –0.63 and –1.132, respectively (P = .02); and mean z scores were –0.69 and –0.61, respectively (P = .81). Mean femoral neck BMD was 0.91 in the younger group and 0.80 in the older group (P < .05); t scores were –0.87 and –1.65, respectively (P < .01), and mean femoral neck z scores were –0.69 and –0.67, respectively (P = .92).

To further analyze BMD of specific age groups, we divided patients by decade: 35-39, 40-49, 50-59, 60-69, 70-79, 80-89 years. Among all 6 decades, there were no statistically significant differences between hip z scores (P = .83) (Figure 1). Spearman rank-order correlation test showed a moderate inverse correlation between age and femoral neck BMD (R = –0.42) and t score (R = –0.43). There was a weak correlation between increasing age and decreasing spine BMD, t score, and z score (Rs = –0.27, –0.31, 0.03). There was no correlation between age and femoral neck z score (R = –0.04).

According to the WHO classification system, 11 (23%) of the 47 women in the younger group were osteopenic, and 8 (17%) were osteoporotic, based on spine BMD. Hip BMD values indicated that 20 patients (43%) were osteopenic, and 3 (6%) were osteoporotic. One patient in the younger group had a hip z score of less than –2, and 14 patients (39%) had a hip z score between –2 and –1. Six patients (18%) had a spine z score of less than –2, and 6 patients (18%) had a spine z score between –2 and –1. Of the 81 older patients, 22 (27%) were osteopenic, and 21 (26%) were osteoporotic, according to spine measurements. The femoral neck data indicated that 39 (48%) of the older patients were osteopenic, and 22 (27%) were osteoporotic.

In both groups, mechanisms of injury were identified. Of the 47 younger patients, 26 fell from standing, 7 fell from a height of more than 6 feet, and 14 were injured in motor vehicle collisions (MVCs). Of the 81 older patients, 2 sustained a direct blow, 64 fell from standing, 4 fell from a height of more than 6 feet, and 11 were injured in MVCs. The differences in z scores based on mechanism of injury were not statistically significant (P = .22) (Figure 2).

Discussion

Several studies have shown that older women with DRFs have low BMD in the spine and femoral neck.^8,9 These studies focused on older women who sustained low-energy fractures caused by a fall from a standing height. Studies of younger women with DRFs focused on BMD of the contralateral distal radius, not the spine or femoral neck.^10,11 Those study groups also had low BMD. Findings from a multitude of studies have established that patients who are older than 50 years when they sustain distal radius fragility fractures should be referred for bone densitometry studies, and there is increasing evidence that younger patients with fragility fractures should undergo this evaluation as well.

The present study was designed to expand the range of patients and mechanisms of injury. Women in this study were 35 years or older. In addition to collecting data from patients injured in a fall from standing, we examined the medical records of women injured in MVCs, in falls from heights of more than 6 feet, and from direct trauma to the wrist. We measured the BMD of the spine and femoral neck and of the contralateral distal radius.

For this discussion, several key points should be made about BMD evaluation in younger versus older women. Most organizations caution against using spine BMD in older women. The ISCD, however, recommended measuring both hip and spine BMD; whereas BMD can be falsely elevated by spine osteoarthritis in older patients, spine BMD measurements are accurate in younger patients not affected by osteoarthritis. The ISCD also stipulated that z scores should be used in examining BMD in younger patients. The z score is a value of how many standard deviations BMD differs from a matched population of the same age, sex, ethnicity, and weight. The t score, which is useful in evaluating older patients, compares a patient’s BMD with that of an average 30-year-old.¹²

According to the WHO classification system (intended for older women), osteopenia is indicated by a t score between –1.0 and –2.5, and osteoporosis is indicated by a t score of less than –2.5. In the present study, about 43% of the younger patients (age, 35-50 years) with DRFs were osteopenic, and 6% of these patients were osteoporotic. In concert with previous studies,⁹ 48% of our older women (age, >50 years) with DRFs were osteopenic, and 27% were osteoporotic. The difference in mean spinal z scores between the younger and older groups was not statistically significant (P = .81).

As mentioned, when examining BMD of younger patients, it is imperative to use spine z scores. About 18% of our younger patients had a z score of less than –2, and 18% had a z score between –2 and –1. In our comparison of patients from 5 different age decades (range, 35-79 years), there was no statistically significant difference in z scores (P = .83). In addition, there was no correlation between increasing age and decreasing z score (R = –0.04).

Secondary causes of osteoporosis have been documented in 30% of premenopausal women and 55% of men with vertebral fractures.^13-15 Primary osteoporosis results from the normal aging process; secondary osteoporosis results from reversible causes, including medications, gastrointestinal disorders, renal disease, endocrine disorders, and sedentary lifestyle.^15,16 When a secondary cause of osteoporosis is identified, treatment can be initiated to increase BMD. As younger patients can reverse bone loss and even increase BMD, it is important to identify reversible causes of osteopenia and osteoporosis in this age group. It is well documented that both younger and older patients with DRFs are at increased risk for subsequent fractures.⁶ Preventing further bone loss at a younger age may drastically decrease lifetime fracture risk.^12,17

Most previous studies of BMD in women were limited to patients with DRFs caused by a low-energy mechanism or by a fall from standing. Current recommendations for BMD testing focus on postmenopausal women who have sustained a fragility or low-energy DRF. When an osteoporotic or osteopenic patient’s distal radius is subjected to a high-energy force, a fracture is likely. Therefore, we expanded our study to include high-energy mechanisms of injury. Our analysis of BMD in patients with DRFs sustained in MVCs indicated that 12% of this group were osteoporotic, and 44% were osteopenic. Forty-three percent of our younger patients with a DRF fractured in a MVC were osteopenic, and 6% were osteoporotic. Among 4 mechanisms of injury for DRFs, there was no statistically significant difference in z scores (P = .22) (Figure 2). This provides evidence that a significant portion of patients with DRFs from both high- and low-energy mechanisms are osteoporotic or osteopenic. Patients with DRFs sustained in MVCs or in falls from heights of more than 6 feet should be referred for BMD evaluation.         

Conclusion

A significant proportion of younger patients with DRFs are osteopenic or osteoporotic (43% and 6%, respectively), and their z scores are comparable to those of older patients with DRFs. There was no statistically significant difference in BMD z scores between younger and older patients and no difference in mechanisms of injury. This is evidence that younger patients with DRFs caused by a high- or low-energy mechanism of injury should undergo both DXA scan and BMD evaluation. If osteoporosis or osteopenia can be diagnosed at an earlier age, and if these patients can be properly treated, subsequent fractures could be prevented. The present study provides evidence supporting a simplification of the current recommendations for BMD evaluation: All women with DRFs should undergo bone densitometry.