Orthopedics

It’s spring. Have you started your Meaningful Use reporting yet? More important, have you begun reporting at all?

“Say the words Meaningful Use to most orthopedists, and they usually roll their eyes or shake their heads,” says Cheyenne Brinson, MBA, CPA, a KarenZupko & Associates consultant who has been advising surgical practices on Meaningful Use since the program’s inception. Although many orthopedists are successfully using certified electronic health records (EHRs) to e-prescribe and enter radiology and laboratory orders, Brinson says many other requirements are misunderstood and perceived as overly complex. In many cases, practices are doing more work than they need to in order to attest.

“It’s actually not that complicated to meet Meaningful Use requirements,” she says. “The trick is to zero in on what’s relevant only for surgeons. This isn’t crystal clear in the CMS [Centers for Medicare & Medicaid Services] documents, and it’s not the forte of most EHR vendors or trainers either.” In fact, in Brinson’s experience, most EHR trainers present Meaningful Use to every practice as if it were primary care. Yet, the requirements for surgeons are different for primary care and are, frankly, less involved.

That’s good news. Because if you didn’t attest for Meaningful Use in 2014, the first year that reporting was required, you’re automatically getting dinged 2% on your Medicare payments in 2015. So, it’s time to get organized and get moving to avoid further penalties.

Avoid These Four Common Faux Pas

Brinson says the Clinical Quality Measures (CQMs) are hands down the most misunderstood component of Meaningful Use. “When I explain Meaningful Use to surgeons, I can’t jump up and down and wave my hands in the air enough to call attention to this,” she quips.

At issue: There are 64 CQMs, but very few are applicable to surgeons. Yet, many surgeons think they have to perform them for Meaningful Use. Not so, says Brinson. “Surgeons have to report a CQM only if it’s clinically relevant. If none of the CQMs are clinically relevant in your practice, it’s okay to report a zero value if you have not actually performed it.”

Here’s how this plays out. In Stage 2, physicians must report 9 CQMs across 3 domains; Population/Public Health, Patient Safety, and Efficient Use of Healthcare Resources are examples of domains that are most applicable to orthopedists. “If you choose Low Back Pain: Use of Imaging Studies as one of these, it’s possible an orthopedist would have a numeric value to report,” Brinson says. “But if you also choose Use of High-Risk Medications in the Elderly, an orthopedist will probably report a zero value. And that’s totally acceptable. You will not be penalized for reporting zero.”

Another common misconception is around the Vital Signs and Smoking Status measures. “We have worked with surgical practices that think Meaningful Use is requiring them to collect vital signs and smoking status at every visit, even though they may not be clinically relevant,” says Brinson. Again, not true.

“Height and/or weight and blood pressure, as well as smoking status measures, need to be reported only once per patient during the reporting period,” Brinson clarifies. “So from a practical standpoint, most orthopedic practices can collect this data from new patients and then again as clinically necessary,” adding there are even exclusions for physicians who attest that either height and weight and/or blood pressure has no relevance to their scope of practice at all.

Brinson also sees practices do more work than they need to when it comes to Patient Care Reminders. She recently worked with a surgery practice that sent reminders for colonoscopies. “Not exactly clinically relevant,” she says, “and an unnecessary step for staff.” That’s because physicians aren’t required to send reminders that aren’t relevant to their specialty.

The Federal Register states, “An eligible provider (EP) should use clinically relevant information stored within the [EHR] to identify patients who should receive reminders…. The EP is best positioned to decide which information is clinically relevant for this purpose.”

“In orthopedics, clinically relevant reminders could be for an outside referral, a follow-up on an MRI or other test, or a reminder to schedule a postoperative appointment,” Brinson explains. “Work with your EHR vendor to create the reminders that are most appropriate for your patient base.”

The final faux pas that Brinson finds: “Meaningful Use requires you to report data for all patients, not just Medicare patients. That seems to be a point of confusion for many.”

Three Cheers for the Patient Portal Requirement

Stage 2 saw the addition of the Patient Portal Requirement, and Brinson suggests that the benefits of this tool go far beyond Meaningful Use. “Patient portals are essential to a modern practice,” she says. “Patients use them to complete a health history prior to their appointment, pay their bill, schedule follow-up appointments, and more.” Further, the patient portal facilitates another Meaningful Use Stage 2 requirement: secure electronic messaging with patients. For both Meaningful Use and risk management, moving away from e-mail and texting and toward secure/encrypted messaging is a must. The patient portal has this feature already built in, and all messages are stored securely and archived—which meets the HIPAA (Health Insurance Portability and Accountability Act) Omnibus requirements, too.

So if you’ve implemented a patient portal, that’s good for your practice and your patients on many levels. But there is a caveat about meeting the Meaningful Use requirement. “For this requirement, 5% of the unique patients seen during the reporting period must ‘view, download, or transmit to a third party their health information,’” Brinson explains. “So the onus is on your practice to ‘sell’ the benefits of the patient portal and get patients to use it so you can achieve the 5% threshold.”

Clinical Decision Support and Summaries

The requirements of Clinical Decision Support Interventions and Clinical Summaries may seem daunting, but, if you think beyond Meaningful Use for a moment, both facilitate better care.

Take Clinical Decision Support Interventions. What would be helpful for you to know about a patient before surgery? What information would enable you to deliver better care?

“One surgeon told me that a family history of malignant hyperthermia would mean the difference between performing the case in the operating room versus the ambulatory surgery center,” Brinson says. “This is a good example of an intervention that a surgeon would work with their EHR vendor to set up.”

The objective states that each intervention is to be an evidence-based decision-support intervention based on each one and at least one combination of the following data: problem list, medication list, medication allergy list, demographics, laboratory tests and values/results, and vital signs. “Stage 1 requires physicians to implement 1 Clinical Decision Support Intervention, and Stage 2 requires 5,” reminds Brinson.

And here’s all you need to know about Clinical Summaries. Although there are 20 specific required elements of a clinical summary, physicians themselves need to provide details only for clinical instructions and the care plan, including goals and instructions. Ancillary staff can populate the other elements.

Brinson points out that surgeons are not expected to provide a copy of the patient’s note, or to complete the note, before the patient checks out. The requirement under Stage 2 is that the clinical summary is provided to the patient within 1 business day. “From a practical standpoint, practices can print the clinical summary for patients at checkout. A well-done clinical summary is a practice efficiency tool as much as a clinical document. It can reduce phone calls from patients asking, ‘Now what did the doctor tell me to do?’”

Often Overlooked

There are requirements that, Brinson says, surgeons often gloss over: Protect Electronic Health Information and Text-Searchable Progress Notes.

“Stage 2 requires physicians to conduct a privacy risk analysis to protect electronic health information,” she explains. “Most EHR vendors don’t offer this as part of their product, so it’s frequently overlooked.” Such an analysis typically requires an outside vendor, but there are free, do-it-yourself tools available, such as the Privacy and Security Toolkit for Small Provider Organizations,^* from the Healthcare Information and Management Systems Society (HIMSS).

The analysis should follow HIPAA guidelines, and the most intensive part of this requirement is to conduct or review a privacy risk analysis of the clinical technology. “You’ve also got to address data encryption and security in the EHR, and ensure HIPAA policies and procedures are in place,” Brinson states.

Text-Searchable Progress Notes are also a new requirement in Stage 2. All progress notes must be text searchable—practices can no longer include progress notes as scanned attachments. “That means no more PDFs,” Brinson says. “Surgeons can still dictate, but the dictation must be entered into the EHR in such a way that it’s searchable. In Stage 2, 30% of unique patients must have a minimum of 1 text-searchable electronic progress note created, edited, and signed in the EHR.”

Conclusion

Meaningful Use does not have to be cumbersome. Focus on what surgical practices need to know, and attestation won’t be as complicated as you think.

It’s spring. Have you started your Meaningful Use reporting yet? More important, have you begun reporting at all?

“Say the words Meaningful Use to most orthopedists, and they usually roll their eyes or shake their heads,” says Cheyenne Brinson, MBA, CPA, a KarenZupko & Associates consultant who has been advising surgical practices on Meaningful Use since the program’s inception. Although many orthopedists are successfully using certified electronic health records (EHRs) to e-prescribe and enter radiology and laboratory orders, Brinson says many other requirements are misunderstood and perceived as overly complex. In many cases, practices are doing more work than they need to in order to attest.

“It’s actually not that complicated to meet Meaningful Use requirements,” she says. “The trick is to zero in on what’s relevant only for surgeons. This isn’t crystal clear in the CMS [Centers for Medicare & Medicaid Services] documents, and it’s not the forte of most EHR vendors or trainers either.” In fact, in Brinson’s experience, most EHR trainers present Meaningful Use to every practice as if it were primary care. Yet, the requirements for surgeons are different for primary care and are, frankly, less involved.

That’s good news. Because if you didn’t attest for Meaningful Use in 2014, the first year that reporting was required, you’re automatically getting dinged 2% on your Medicare payments in 2015. So, it’s time to get organized and get moving to avoid further penalties.

Avoid These Four Common Faux Pas

Brinson says the Clinical Quality Measures (CQMs) are hands down the most misunderstood component of Meaningful Use. “When I explain Meaningful Use to surgeons, I can’t jump up and down and wave my hands in the air enough to call attention to this,” she quips.

At issue: There are 64 CQMs, but very few are applicable to surgeons. Yet, many surgeons think they have to perform them for Meaningful Use. Not so, says Brinson. “Surgeons have to report a CQM only if it’s clinically relevant. If none of the CQMs are clinically relevant in your practice, it’s okay to report a zero value if you have not actually performed it.”

Here’s how this plays out. In Stage 2, physicians must report 9 CQMs across 3 domains; Population/Public Health, Patient Safety, and Efficient Use of Healthcare Resources are examples of domains that are most applicable to orthopedists. “If you choose Low Back Pain: Use of Imaging Studies as one of these, it’s possible an orthopedist would have a numeric value to report,” Brinson says. “But if you also choose Use of High-Risk Medications in the Elderly, an orthopedist will probably report a zero value. And that’s totally acceptable. You will not be penalized for reporting zero.”

Another common misconception is around the Vital Signs and Smoking Status measures. “We have worked with surgical practices that think Meaningful Use is requiring them to collect vital signs and smoking status at every visit, even though they may not be clinically relevant,” says Brinson. Again, not true.

“Height and/or weight and blood pressure, as well as smoking status measures, need to be reported only once per patient during the reporting period,” Brinson clarifies. “So from a practical standpoint, most orthopedic practices can collect this data from new patients and then again as clinically necessary,” adding there are even exclusions for physicians who attest that either height and weight and/or blood pressure has no relevance to their scope of practice at all.

Brinson also sees practices do more work than they need to when it comes to Patient Care Reminders. She recently worked with a surgery practice that sent reminders for colonoscopies. “Not exactly clinically relevant,” she says, “and an unnecessary step for staff.” That’s because physicians aren’t required to send reminders that aren’t relevant to their specialty.

The Federal Register states, “An eligible provider (EP) should use clinically relevant information stored within the [EHR] to identify patients who should receive reminders…. The EP is best positioned to decide which information is clinically relevant for this purpose.”

“In orthopedics, clinically relevant reminders could be for an outside referral, a follow-up on an MRI or other test, or a reminder to schedule a postoperative appointment,” Brinson explains. “Work with your EHR vendor to create the reminders that are most appropriate for your patient base.”

The final faux pas that Brinson finds: “Meaningful Use requires you to report data for all patients, not just Medicare patients. That seems to be a point of confusion for many.”

Three Cheers for the Patient Portal Requirement

Stage 2 saw the addition of the Patient Portal Requirement, and Brinson suggests that the benefits of this tool go far beyond Meaningful Use. “Patient portals are essential to a modern practice,” she says. “Patients use them to complete a health history prior to their appointment, pay their bill, schedule follow-up appointments, and more.” Further, the patient portal facilitates another Meaningful Use Stage 2 requirement: secure electronic messaging with patients. For both Meaningful Use and risk management, moving away from e-mail and texting and toward secure/encrypted messaging is a must. The patient portal has this feature already built in, and all messages are stored securely and archived—which meets the HIPAA (Health Insurance Portability and Accountability Act) Omnibus requirements, too.

So if you’ve implemented a patient portal, that’s good for your practice and your patients on many levels. But there is a caveat about meeting the Meaningful Use requirement. “For this requirement, 5% of the unique patients seen during the reporting period must ‘view, download, or transmit to a third party their health information,’” Brinson explains. “So the onus is on your practice to ‘sell’ the benefits of the patient portal and get patients to use it so you can achieve the 5% threshold.”

Clinical Decision Support and Summaries

The requirements of Clinical Decision Support Interventions and Clinical Summaries may seem daunting, but, if you think beyond Meaningful Use for a moment, both facilitate better care.

Take Clinical Decision Support Interventions. What would be helpful for you to know about a patient before surgery? What information would enable you to deliver better care?

“One surgeon told me that a family history of malignant hyperthermia would mean the difference between performing the case in the operating room versus the ambulatory surgery center,” Brinson says. “This is a good example of an intervention that a surgeon would work with their EHR vendor to set up.”

The objective states that each intervention is to be an evidence-based decision-support intervention based on each one and at least one combination of the following data: problem list, medication list, medication allergy list, demographics, laboratory tests and values/results, and vital signs. “Stage 1 requires physicians to implement 1 Clinical Decision Support Intervention, and Stage 2 requires 5,” reminds Brinson.

And here’s all you need to know about Clinical Summaries. Although there are 20 specific required elements of a clinical summary, physicians themselves need to provide details only for clinical instructions and the care plan, including goals and instructions. Ancillary staff can populate the other elements.

Brinson points out that surgeons are not expected to provide a copy of the patient’s note, or to complete the note, before the patient checks out. The requirement under Stage 2 is that the clinical summary is provided to the patient within 1 business day. “From a practical standpoint, practices can print the clinical summary for patients at checkout. A well-done clinical summary is a practice efficiency tool as much as a clinical document. It can reduce phone calls from patients asking, ‘Now what did the doctor tell me to do?’”

Often Overlooked

There are requirements that, Brinson says, surgeons often gloss over: Protect Electronic Health Information and Text-Searchable Progress Notes.

“Stage 2 requires physicians to conduct a privacy risk analysis to protect electronic health information,” she explains. “Most EHR vendors don’t offer this as part of their product, so it’s frequently overlooked.” Such an analysis typically requires an outside vendor, but there are free, do-it-yourself tools available, such as the Privacy and Security Toolkit for Small Provider Organizations,^* from the Healthcare Information and Management Systems Society (HIMSS).

The analysis should follow HIPAA guidelines, and the most intensive part of this requirement is to conduct or review a privacy risk analysis of the clinical technology. “You’ve also got to address data encryption and security in the EHR, and ensure HIPAA policies and procedures are in place,” Brinson states.

Text-Searchable Progress Notes are also a new requirement in Stage 2. All progress notes must be text searchable—practices can no longer include progress notes as scanned attachments. “That means no more PDFs,” Brinson says. “Surgeons can still dictate, but the dictation must be entered into the EHR in such a way that it’s searchable. In Stage 2, 30% of unique patients must have a minimum of 1 text-searchable electronic progress note created, edited, and signed in the EHR.”

Conclusion

Meaningful Use does not have to be cumbersome. Focus on what surgical practices need to know, and attestation won’t be as complicated as you think.

It’s spring. Have you started your Meaningful Use reporting yet? More important, have you begun reporting at all?

“Say the words Meaningful Use to most orthopedists, and they usually roll their eyes or shake their heads,” says Cheyenne Brinson, MBA, CPA, a KarenZupko & Associates consultant who has been advising surgical practices on Meaningful Use since the program’s inception. Although many orthopedists are successfully using certified electronic health records (EHRs) to e-prescribe and enter radiology and laboratory orders, Brinson says many other requirements are misunderstood and perceived as overly complex. In many cases, practices are doing more work than they need to in order to attest.

“It’s actually not that complicated to meet Meaningful Use requirements,” she says. “The trick is to zero in on what’s relevant only for surgeons. This isn’t crystal clear in the CMS [Centers for Medicare & Medicaid Services] documents, and it’s not the forte of most EHR vendors or trainers either.” In fact, in Brinson’s experience, most EHR trainers present Meaningful Use to every practice as if it were primary care. Yet, the requirements for surgeons are different for primary care and are, frankly, less involved.

That’s good news. Because if you didn’t attest for Meaningful Use in 2014, the first year that reporting was required, you’re automatically getting dinged 2% on your Medicare payments in 2015. So, it’s time to get organized and get moving to avoid further penalties.

Avoid These Four Common Faux Pas

Brinson says the Clinical Quality Measures (CQMs) are hands down the most misunderstood component of Meaningful Use. “When I explain Meaningful Use to surgeons, I can’t jump up and down and wave my hands in the air enough to call attention to this,” she quips.

At issue: There are 64 CQMs, but very few are applicable to surgeons. Yet, many surgeons think they have to perform them for Meaningful Use. Not so, says Brinson. “Surgeons have to report a CQM only if it’s clinically relevant. If none of the CQMs are clinically relevant in your practice, it’s okay to report a zero value if you have not actually performed it.”

Here’s how this plays out. In Stage 2, physicians must report 9 CQMs across 3 domains; Population/Public Health, Patient Safety, and Efficient Use of Healthcare Resources are examples of domains that are most applicable to orthopedists. “If you choose Low Back Pain: Use of Imaging Studies as one of these, it’s possible an orthopedist would have a numeric value to report,” Brinson says. “But if you also choose Use of High-Risk Medications in the Elderly, an orthopedist will probably report a zero value. And that’s totally acceptable. You will not be penalized for reporting zero.”

Another common misconception is around the Vital Signs and Smoking Status measures. “We have worked with surgical practices that think Meaningful Use is requiring them to collect vital signs and smoking status at every visit, even though they may not be clinically relevant,” says Brinson. Again, not true.

“Height and/or weight and blood pressure, as well as smoking status measures, need to be reported only once per patient during the reporting period,” Brinson clarifies. “So from a practical standpoint, most orthopedic practices can collect this data from new patients and then again as clinically necessary,” adding there are even exclusions for physicians who attest that either height and weight and/or blood pressure has no relevance to their scope of practice at all.

Brinson also sees practices do more work than they need to when it comes to Patient Care Reminders. She recently worked with a surgery practice that sent reminders for colonoscopies. “Not exactly clinically relevant,” she says, “and an unnecessary step for staff.” That’s because physicians aren’t required to send reminders that aren’t relevant to their specialty.

The Federal Register states, “An eligible provider (EP) should use clinically relevant information stored within the [EHR] to identify patients who should receive reminders…. The EP is best positioned to decide which information is clinically relevant for this purpose.”

“In orthopedics, clinically relevant reminders could be for an outside referral, a follow-up on an MRI or other test, or a reminder to schedule a postoperative appointment,” Brinson explains. “Work with your EHR vendor to create the reminders that are most appropriate for your patient base.”

The final faux pas that Brinson finds: “Meaningful Use requires you to report data for all patients, not just Medicare patients. That seems to be a point of confusion for many.”

Three Cheers for the Patient Portal Requirement

Stage 2 saw the addition of the Patient Portal Requirement, and Brinson suggests that the benefits of this tool go far beyond Meaningful Use. “Patient portals are essential to a modern practice,” she says. “Patients use them to complete a health history prior to their appointment, pay their bill, schedule follow-up appointments, and more.” Further, the patient portal facilitates another Meaningful Use Stage 2 requirement: secure electronic messaging with patients. For both Meaningful Use and risk management, moving away from e-mail and texting and toward secure/encrypted messaging is a must. The patient portal has this feature already built in, and all messages are stored securely and archived—which meets the HIPAA (Health Insurance Portability and Accountability Act) Omnibus requirements, too.

So if you’ve implemented a patient portal, that’s good for your practice and your patients on many levels. But there is a caveat about meeting the Meaningful Use requirement. “For this requirement, 5% of the unique patients seen during the reporting period must ‘view, download, or transmit to a third party their health information,’” Brinson explains. “So the onus is on your practice to ‘sell’ the benefits of the patient portal and get patients to use it so you can achieve the 5% threshold.”

Clinical Decision Support and Summaries

The requirements of Clinical Decision Support Interventions and Clinical Summaries may seem daunting, but, if you think beyond Meaningful Use for a moment, both facilitate better care.

Take Clinical Decision Support Interventions. What would be helpful for you to know about a patient before surgery? What information would enable you to deliver better care?

“One surgeon told me that a family history of malignant hyperthermia would mean the difference between performing the case in the operating room versus the ambulatory surgery center,” Brinson says. “This is a good example of an intervention that a surgeon would work with their EHR vendor to set up.”

The objective states that each intervention is to be an evidence-based decision-support intervention based on each one and at least one combination of the following data: problem list, medication list, medication allergy list, demographics, laboratory tests and values/results, and vital signs. “Stage 1 requires physicians to implement 1 Clinical Decision Support Intervention, and Stage 2 requires 5,” reminds Brinson.

And here’s all you need to know about Clinical Summaries. Although there are 20 specific required elements of a clinical summary, physicians themselves need to provide details only for clinical instructions and the care plan, including goals and instructions. Ancillary staff can populate the other elements.

Brinson points out that surgeons are not expected to provide a copy of the patient’s note, or to complete the note, before the patient checks out. The requirement under Stage 2 is that the clinical summary is provided to the patient within 1 business day. “From a practical standpoint, practices can print the clinical summary for patients at checkout. A well-done clinical summary is a practice efficiency tool as much as a clinical document. It can reduce phone calls from patients asking, ‘Now what did the doctor tell me to do?’”

Often Overlooked

There are requirements that, Brinson says, surgeons often gloss over: Protect Electronic Health Information and Text-Searchable Progress Notes.

“Stage 2 requires physicians to conduct a privacy risk analysis to protect electronic health information,” she explains. “Most EHR vendors don’t offer this as part of their product, so it’s frequently overlooked.” Such an analysis typically requires an outside vendor, but there are free, do-it-yourself tools available, such as the Privacy and Security Toolkit for Small Provider Organizations,^* from the Healthcare Information and Management Systems Society (HIMSS).

The analysis should follow HIPAA guidelines, and the most intensive part of this requirement is to conduct or review a privacy risk analysis of the clinical technology. “You’ve also got to address data encryption and security in the EHR, and ensure HIPAA policies and procedures are in place,” Brinson states.

Text-Searchable Progress Notes are also a new requirement in Stage 2. All progress notes must be text searchable—practices can no longer include progress notes as scanned attachments. “That means no more PDFs,” Brinson says. “Surgeons can still dictate, but the dictation must be entered into the EHR in such a way that it’s searchable. In Stage 2, 30% of unique patients must have a minimum of 1 text-searchable electronic progress note created, edited, and signed in the EHR.”

Conclusion

Meaningful Use does not have to be cumbersome. Focus on what surgical practices need to know, and attestation won’t be as complicated as you think.

Osteoid osteoma (OO) is one of the most common benign tumors of bone, representing roughly 10% of all benign bone-forming tumors and 5% of all primary bone tumors.¹ The majority of cases occur in individuals under age 20 years and more frequently in males (2:1).² These lesions tend to be cortically based and most often located about the hip and in the diaphysis of long bones. They typically are characterized radiographically by a nidus less than 2 cm in diameter surrounded by dense, reactive bone of variable thickness.

The classic presentation of OO is localized, dull, aching pain that is worse at night and that is relieved with use of salicylates or other nonsteroidal anti-inflammatory drugs (NSAIDs).³ The diagnosis is made by patient history and plain radiographs, often supported by computed tomography (CT) or magnetic resonance imaging for appropriate identification of the tumor nidus. Despite effective pain relief with NSAIDs as well as evidence suggesting that the natural history of these tumors is self-limited, most patients forgo medical management in favor of elective surgical treatment.^4,5

Initially, treatment for OO focused on either symptom management or en bloc surgical resection of the tumor nidus. Several different minimally invasive therapies have since been developed, and good results reported.^6-8 More recently, use of percutaneous radiofrequency ablation (RFA) has increased, as this method has demonstrated high efficacy and minimal morbidity.^9-11 RFA for OO traditionally has been performed by radiologists under CT guidance in the radiology suite, but advances in intraoperative imaging techniques now allow orthopedic oncologists to perform image-guided RFA in the operating room.

To our knowledge, there have been no reports documenting use of intraoperative CT for localization of OO and use of RFA in the treatment of this lesion. In this article, we report the results of a series of 28 patients with OO treated with intraoperative CT-guided RFA by a single surgeon. We also provide a brief description of this novel technique.

Materials and Methods

The protocol used was approved by our institutional review board. All patients and/or their legal guardians provided informed consent to participate in the study and were informed at the time consent was obtained that case-related data would be submitted for publication.

Patients

Between September 2004 and December 2008, 28 patients (19 males, 9 females) with OO underwent intraoperative percutaneous image-guided RFA at a university hospital. Mean age was 19.5 years, median age was 16 years (range, 7-54 years). Patients were referred for RFA if they had clinical and radiographic features of OO (Figures 1, 2) and wanted to forgo continued medical management. As we selected only patients with lesions that we thought were amenable to percutaneous RFA—lesions involving the long and short bones of the upper or lower extremity and selected flat bones—en bloc surgical resection was not offered to these patients. Lesions were located in the upper extremity (n = 1), lower extremity (n = 24), and pelvis (n = 3) (Figure 3). Twenty-seven procedures were performed for initial tumor treatment and 1 for recurrence after previous open excision. Two additional procedures were later performed on separate patients with recurrent symptoms after the index procedure. All procedures were performed by the senior author (DML).

Procedure

With each patient, all options were discussed, including continued medical management versus surgical treatment, and informed consent was obtained. All procedures were performed with the patient under general anesthesia in the operating room. RFA for an upper extremity lesion was performed with the patient in the supine position with the ipsilateral extremity draped over a hand table. The 2 procedures for lesions in the talus or calcaneus were performed with the patient in the supine position using a standard table with the bottom of the table flexed down 90° to allow the nonaffected leg to hang over the end of the table. The affected extremity in each case was then positioned in a well-padded leg holder to allow the foot and ankle to be draped free for 360° imaging.

All other procedures for lower extremity diaphyseal or pelvic lesions were performed with a fracture table. After successful induction of general anesthesia, the patient was positioned supine on the table with the contralateral lower extremity abducted and externally rotated in a well-leg holder. The ipsilateral leg was held in the traction apparatus without traction applied and was prepared and draped accordingly (Figure 4). With use of the Siemens Siremobil ISO-C3D fluoroscopic C-arm (Siemens Medical Solutions, Malvern, Pennsylvania), a radiograph was taken of the affected area to identify the lesion. Local anesthetic was infiltrated into the surgical site down to the periosteum. A stab incision was made, and, with fluoroscopic guidance, a 0.062-mm Kirschner wire (K-wire) was placed into the lesion. Location within the tumor nidus was confirmed with biplanar fluoroscopic imaging. A Bonopty cannula (AprioMed, Uppsala, Sweden) was then passed over the K-wire. After the wire was removed, a 5-mm radiofrequency probe (Radionics, Burlington, Massachusetts) was placed through the cannula, and positioning within the nidus was confirmed with 3-dimensional (3-D) CT reconstructions in the sagittal, coronal, and axial planes (Figure 5). A radiofrequency generator (Radionics) was used to heat the lesion at 93°C for 7 minutes. The probe and trocar were then removed. Steri-strips and a sterile dressing were used to cover the wound, and the patient was taken to the recovery area after extubation. All patients were discharged home the day of the procedure.

Follow-Up

We phoned all the patients to ask about symptom recurrence, outside treatment, and satisfaction with RFA and to obtain informed consent to participate in our study. Only 1 of the 28 patients could not be reached and was lost to follow-up. Mean follow-up at time of study completion was 31.1 months (range, 5.2-55.8 months).

The 27 patients were asked a series of questions about their treatment: Have you had any recurrence of symptoms following treatment for your OO? Have you received treatment elsewhere? Were you satisfied with your treatment? Would you have the procedure again if you had a recurrence of symptoms?

Primary success was defined as complete pain relief after initial RFA with no evidence of recurrence at time of final follow-up, and secondary success was defined as presence of recurrent symptoms after initial RFA with complete pain relief after a second procedure with no evidence of recurrence.

Results

All RFAs were technically successful with adequate localization of the tumor nidus and subsequent probe placement within the lesion. There were no intraoperative or postoperative complications. All 28 patients were discharged home the day of procedure. Twenty-six patients (92.8%) experienced complete pain relief after primary RFA, had no evidence of recurrence at final follow-up, and denied symptom recurrence at time of study completion.

The other 2 patients reported symptom recurrence after the index treatment (1 proximal femur lesion, 1 distal femur lesion). One of these patients did well initially but had a recurrence about 2 months after the primary RFA; a second RFA provided complete resolution of pain with no evidence of recurrence at time of study completion. In the other patient’s case, intermittent pain persisted for 2 weeks after the primary RFA, and evidence of recurrence was documented 3 months after surgery; a second RFA was performed shortly thereafter, but the patient was subsequently lost to follow-up.

At time of study completion, all 27 patients who had been contacted by phone denied seeking additional treatment elsewhere and stated they would have the procedure again if their symptoms ever recurred.

Discussion

Osteoid osteoma is one of the most common benign tumors of bone. Over the past 2 decades, percutaneous RFA, in comparison with open excision, has emerged as a safe and effective treatment option with minimal patient morbidity.^9-11 RFA traditionally has been performed by radiologists under CT guidance in the radiology suite. However, now orthopedic surgeons can obtain advanced intraoperative imaging beyond standard fluoroscopy. The Siemens Siremobil ISO-C3D fluoroscopic C-arm is an innovative intraoperative imaging device that functions as a standard fluoroscope but also generates 3-D reconstructions of surgical anatomy. The isocentric design and integrated motor unit allow the C-arm to move through a 190º arc while centering its beam directly on the area of interest. This data set is transferred to a computer workstation, where it is reformatted so that CT-quality images are generated in axial, sagittal, and coronal planes. This acquisition process takes only minutes, and the multiplanar images produced may be simultaneously displayed and manipulated on the screen in real time.

One concern about this technology is the amount of radiation exposure for patients, surgeons, and operating room staff. The device measures only radiation time, and the amount of exposure during that time depends on the volume and density of the radiated body. We did not calculate the amount of exposure for this study. Mean exposure time was between 20 and 40 seconds, reflecting the number of attempts required to localize the lesion and the surgeon’s experience with the technique. Although the potential for increased exposure is a valid concern, previous studies using this technology have demonstrated that a similar average exposure time is equivalent to that of standard CT, and that use of the device, over conventional techniques, potentially can lead to decreased overall radiation exposure.^12,13

This series demonstrated that OO can be safely and effectively treated with intraoperative percutaneous RFA by an orthopedic oncologist. Our success rate is very similar to rates reported in the radiology literature. Studies are needed to confirm the efficacy of this novel technique in comparison with what has been reported in that literature. Given these promising preliminary results, and the relative ease of use and minimal learning curve associated with this technology, all orthopedic oncologists should be able to offer this treatment for OO. Furthermore, this technique allows orthopedic oncologists to provide appropriate definitive treatment and care directly, rather than by referring patients to radiologists.

In the treatment of OO, we reserve RFA for lesions involving the long and short bones of the upper and lower extremities, as well as selected flat bones, such as those in the pelvis. Although percutaneous RFA of spinal lesions has been reported in the literature, we think these represent a relative contraindication for this technique; image resolution, in our opinion, is not high enough to justify risking injury to the nerves in the spinal canal, lateral recesses, and neural foramina. In addition, given the radiation exposure, we recommend caution when using this technique for a pelvic or proximal femoral lesion in a woman of childbearing age.

Osteoid osteoma (OO) is one of the most common benign tumors of bone, representing roughly 10% of all benign bone-forming tumors and 5% of all primary bone tumors.¹ The majority of cases occur in individuals under age 20 years and more frequently in males (2:1).² These lesions tend to be cortically based and most often located about the hip and in the diaphysis of long bones. They typically are characterized radiographically by a nidus less than 2 cm in diameter surrounded by dense, reactive bone of variable thickness.

The classic presentation of OO is localized, dull, aching pain that is worse at night and that is relieved with use of salicylates or other nonsteroidal anti-inflammatory drugs (NSAIDs).³ The diagnosis is made by patient history and plain radiographs, often supported by computed tomography (CT) or magnetic resonance imaging for appropriate identification of the tumor nidus. Despite effective pain relief with NSAIDs as well as evidence suggesting that the natural history of these tumors is self-limited, most patients forgo medical management in favor of elective surgical treatment.^4,5

Initially, treatment for OO focused on either symptom management or en bloc surgical resection of the tumor nidus. Several different minimally invasive therapies have since been developed, and good results reported.^6-8 More recently, use of percutaneous radiofrequency ablation (RFA) has increased, as this method has demonstrated high efficacy and minimal morbidity.^9-11 RFA for OO traditionally has been performed by radiologists under CT guidance in the radiology suite, but advances in intraoperative imaging techniques now allow orthopedic oncologists to perform image-guided RFA in the operating room.

To our knowledge, there have been no reports documenting use of intraoperative CT for localization of OO and use of RFA in the treatment of this lesion. In this article, we report the results of a series of 28 patients with OO treated with intraoperative CT-guided RFA by a single surgeon. We also provide a brief description of this novel technique.

Materials and Methods

The protocol used was approved by our institutional review board. All patients and/or their legal guardians provided informed consent to participate in the study and were informed at the time consent was obtained that case-related data would be submitted for publication.

Patients

Between September 2004 and December 2008, 28 patients (19 males, 9 females) with OO underwent intraoperative percutaneous image-guided RFA at a university hospital. Mean age was 19.5 years, median age was 16 years (range, 7-54 years). Patients were referred for RFA if they had clinical and radiographic features of OO (Figures 1, 2) and wanted to forgo continued medical management. As we selected only patients with lesions that we thought were amenable to percutaneous RFA—lesions involving the long and short bones of the upper or lower extremity and selected flat bones—en bloc surgical resection was not offered to these patients. Lesions were located in the upper extremity (n = 1), lower extremity (n = 24), and pelvis (n = 3) (Figure 3). Twenty-seven procedures were performed for initial tumor treatment and 1 for recurrence after previous open excision. Two additional procedures were later performed on separate patients with recurrent symptoms after the index procedure. All procedures were performed by the senior author (DML).

Procedure

With each patient, all options were discussed, including continued medical management versus surgical treatment, and informed consent was obtained. All procedures were performed with the patient under general anesthesia in the operating room. RFA for an upper extremity lesion was performed with the patient in the supine position with the ipsilateral extremity draped over a hand table. The 2 procedures for lesions in the talus or calcaneus were performed with the patient in the supine position using a standard table with the bottom of the table flexed down 90° to allow the nonaffected leg to hang over the end of the table. The affected extremity in each case was then positioned in a well-padded leg holder to allow the foot and ankle to be draped free for 360° imaging.

All other procedures for lower extremity diaphyseal or pelvic lesions were performed with a fracture table. After successful induction of general anesthesia, the patient was positioned supine on the table with the contralateral lower extremity abducted and externally rotated in a well-leg holder. The ipsilateral leg was held in the traction apparatus without traction applied and was prepared and draped accordingly (Figure 4). With use of the Siemens Siremobil ISO-C3D fluoroscopic C-arm (Siemens Medical Solutions, Malvern, Pennsylvania), a radiograph was taken of the affected area to identify the lesion. Local anesthetic was infiltrated into the surgical site down to the periosteum. A stab incision was made, and, with fluoroscopic guidance, a 0.062-mm Kirschner wire (K-wire) was placed into the lesion. Location within the tumor nidus was confirmed with biplanar fluoroscopic imaging. A Bonopty cannula (AprioMed, Uppsala, Sweden) was then passed over the K-wire. After the wire was removed, a 5-mm radiofrequency probe (Radionics, Burlington, Massachusetts) was placed through the cannula, and positioning within the nidus was confirmed with 3-dimensional (3-D) CT reconstructions in the sagittal, coronal, and axial planes (Figure 5). A radiofrequency generator (Radionics) was used to heat the lesion at 93°C for 7 minutes. The probe and trocar were then removed. Steri-strips and a sterile dressing were used to cover the wound, and the patient was taken to the recovery area after extubation. All patients were discharged home the day of the procedure.

Follow-Up

We phoned all the patients to ask about symptom recurrence, outside treatment, and satisfaction with RFA and to obtain informed consent to participate in our study. Only 1 of the 28 patients could not be reached and was lost to follow-up. Mean follow-up at time of study completion was 31.1 months (range, 5.2-55.8 months).

The 27 patients were asked a series of questions about their treatment: Have you had any recurrence of symptoms following treatment for your OO? Have you received treatment elsewhere? Were you satisfied with your treatment? Would you have the procedure again if you had a recurrence of symptoms?

Primary success was defined as complete pain relief after initial RFA with no evidence of recurrence at time of final follow-up, and secondary success was defined as presence of recurrent symptoms after initial RFA with complete pain relief after a second procedure with no evidence of recurrence.

Results

All RFAs were technically successful with adequate localization of the tumor nidus and subsequent probe placement within the lesion. There were no intraoperative or postoperative complications. All 28 patients were discharged home the day of procedure. Twenty-six patients (92.8%) experienced complete pain relief after primary RFA, had no evidence of recurrence at final follow-up, and denied symptom recurrence at time of study completion.

The other 2 patients reported symptom recurrence after the index treatment (1 proximal femur lesion, 1 distal femur lesion). One of these patients did well initially but had a recurrence about 2 months after the primary RFA; a second RFA provided complete resolution of pain with no evidence of recurrence at time of study completion. In the other patient’s case, intermittent pain persisted for 2 weeks after the primary RFA, and evidence of recurrence was documented 3 months after surgery; a second RFA was performed shortly thereafter, but the patient was subsequently lost to follow-up.

At time of study completion, all 27 patients who had been contacted by phone denied seeking additional treatment elsewhere and stated they would have the procedure again if their symptoms ever recurred.

Discussion

Osteoid osteoma is one of the most common benign tumors of bone. Over the past 2 decades, percutaneous RFA, in comparison with open excision, has emerged as a safe and effective treatment option with minimal patient morbidity.^9-11 RFA traditionally has been performed by radiologists under CT guidance in the radiology suite. However, now orthopedic surgeons can obtain advanced intraoperative imaging beyond standard fluoroscopy. The Siemens Siremobil ISO-C3D fluoroscopic C-arm is an innovative intraoperative imaging device that functions as a standard fluoroscope but also generates 3-D reconstructions of surgical anatomy. The isocentric design and integrated motor unit allow the C-arm to move through a 190º arc while centering its beam directly on the area of interest. This data set is transferred to a computer workstation, where it is reformatted so that CT-quality images are generated in axial, sagittal, and coronal planes. This acquisition process takes only minutes, and the multiplanar images produced may be simultaneously displayed and manipulated on the screen in real time.

One concern about this technology is the amount of radiation exposure for patients, surgeons, and operating room staff. The device measures only radiation time, and the amount of exposure during that time depends on the volume and density of the radiated body. We did not calculate the amount of exposure for this study. Mean exposure time was between 20 and 40 seconds, reflecting the number of attempts required to localize the lesion and the surgeon’s experience with the technique. Although the potential for increased exposure is a valid concern, previous studies using this technology have demonstrated that a similar average exposure time is equivalent to that of standard CT, and that use of the device, over conventional techniques, potentially can lead to decreased overall radiation exposure.^12,13

This series demonstrated that OO can be safely and effectively treated with intraoperative percutaneous RFA by an orthopedic oncologist. Our success rate is very similar to rates reported in the radiology literature. Studies are needed to confirm the efficacy of this novel technique in comparison with what has been reported in that literature. Given these promising preliminary results, and the relative ease of use and minimal learning curve associated with this technology, all orthopedic oncologists should be able to offer this treatment for OO. Furthermore, this technique allows orthopedic oncologists to provide appropriate definitive treatment and care directly, rather than by referring patients to radiologists.

In the treatment of OO, we reserve RFA for lesions involving the long and short bones of the upper and lower extremities, as well as selected flat bones, such as those in the pelvis. Although percutaneous RFA of spinal lesions has been reported in the literature, we think these represent a relative contraindication for this technique; image resolution, in our opinion, is not high enough to justify risking injury to the nerves in the spinal canal, lateral recesses, and neural foramina. In addition, given the radiation exposure, we recommend caution when using this technique for a pelvic or proximal femoral lesion in a woman of childbearing age.

Osteoid osteoma (OO) is one of the most common benign tumors of bone, representing roughly 10% of all benign bone-forming tumors and 5% of all primary bone tumors.¹ The majority of cases occur in individuals under age 20 years and more frequently in males (2:1).² These lesions tend to be cortically based and most often located about the hip and in the diaphysis of long bones. They typically are characterized radiographically by a nidus less than 2 cm in diameter surrounded by dense, reactive bone of variable thickness.

The classic presentation of OO is localized, dull, aching pain that is worse at night and that is relieved with use of salicylates or other nonsteroidal anti-inflammatory drugs (NSAIDs).³ The diagnosis is made by patient history and plain radiographs, often supported by computed tomography (CT) or magnetic resonance imaging for appropriate identification of the tumor nidus. Despite effective pain relief with NSAIDs as well as evidence suggesting that the natural history of these tumors is self-limited, most patients forgo medical management in favor of elective surgical treatment.^4,5

Initially, treatment for OO focused on either symptom management or en bloc surgical resection of the tumor nidus. Several different minimally invasive therapies have since been developed, and good results reported.^6-8 More recently, use of percutaneous radiofrequency ablation (RFA) has increased, as this method has demonstrated high efficacy and minimal morbidity.^9-11 RFA for OO traditionally has been performed by radiologists under CT guidance in the radiology suite, but advances in intraoperative imaging techniques now allow orthopedic oncologists to perform image-guided RFA in the operating room.

To our knowledge, there have been no reports documenting use of intraoperative CT for localization of OO and use of RFA in the treatment of this lesion. In this article, we report the results of a series of 28 patients with OO treated with intraoperative CT-guided RFA by a single surgeon. We also provide a brief description of this novel technique.

Materials and Methods

The protocol used was approved by our institutional review board. All patients and/or their legal guardians provided informed consent to participate in the study and were informed at the time consent was obtained that case-related data would be submitted for publication.

Patients

Between September 2004 and December 2008, 28 patients (19 males, 9 females) with OO underwent intraoperative percutaneous image-guided RFA at a university hospital. Mean age was 19.5 years, median age was 16 years (range, 7-54 years). Patients were referred for RFA if they had clinical and radiographic features of OO (Figures 1, 2) and wanted to forgo continued medical management. As we selected only patients with lesions that we thought were amenable to percutaneous RFA—lesions involving the long and short bones of the upper or lower extremity and selected flat bones—en bloc surgical resection was not offered to these patients. Lesions were located in the upper extremity (n = 1), lower extremity (n = 24), and pelvis (n = 3) (Figure 3). Twenty-seven procedures were performed for initial tumor treatment and 1 for recurrence after previous open excision. Two additional procedures were later performed on separate patients with recurrent symptoms after the index procedure. All procedures were performed by the senior author (DML).

Procedure

With each patient, all options were discussed, including continued medical management versus surgical treatment, and informed consent was obtained. All procedures were performed with the patient under general anesthesia in the operating room. RFA for an upper extremity lesion was performed with the patient in the supine position with the ipsilateral extremity draped over a hand table. The 2 procedures for lesions in the talus or calcaneus were performed with the patient in the supine position using a standard table with the bottom of the table flexed down 90° to allow the nonaffected leg to hang over the end of the table. The affected extremity in each case was then positioned in a well-padded leg holder to allow the foot and ankle to be draped free for 360° imaging.

All other procedures for lower extremity diaphyseal or pelvic lesions were performed with a fracture table. After successful induction of general anesthesia, the patient was positioned supine on the table with the contralateral lower extremity abducted and externally rotated in a well-leg holder. The ipsilateral leg was held in the traction apparatus without traction applied and was prepared and draped accordingly (Figure 4). With use of the Siemens Siremobil ISO-C3D fluoroscopic C-arm (Siemens Medical Solutions, Malvern, Pennsylvania), a radiograph was taken of the affected area to identify the lesion. Local anesthetic was infiltrated into the surgical site down to the periosteum. A stab incision was made, and, with fluoroscopic guidance, a 0.062-mm Kirschner wire (K-wire) was placed into the lesion. Location within the tumor nidus was confirmed with biplanar fluoroscopic imaging. A Bonopty cannula (AprioMed, Uppsala, Sweden) was then passed over the K-wire. After the wire was removed, a 5-mm radiofrequency probe (Radionics, Burlington, Massachusetts) was placed through the cannula, and positioning within the nidus was confirmed with 3-dimensional (3-D) CT reconstructions in the sagittal, coronal, and axial planes (Figure 5). A radiofrequency generator (Radionics) was used to heat the lesion at 93°C for 7 minutes. The probe and trocar were then removed. Steri-strips and a sterile dressing were used to cover the wound, and the patient was taken to the recovery area after extubation. All patients were discharged home the day of the procedure.

Follow-Up

We phoned all the patients to ask about symptom recurrence, outside treatment, and satisfaction with RFA and to obtain informed consent to participate in our study. Only 1 of the 28 patients could not be reached and was lost to follow-up. Mean follow-up at time of study completion was 31.1 months (range, 5.2-55.8 months).

The 27 patients were asked a series of questions about their treatment: Have you had any recurrence of symptoms following treatment for your OO? Have you received treatment elsewhere? Were you satisfied with your treatment? Would you have the procedure again if you had a recurrence of symptoms?

Primary success was defined as complete pain relief after initial RFA with no evidence of recurrence at time of final follow-up, and secondary success was defined as presence of recurrent symptoms after initial RFA with complete pain relief after a second procedure with no evidence of recurrence.

Results

All RFAs were technically successful with adequate localization of the tumor nidus and subsequent probe placement within the lesion. There were no intraoperative or postoperative complications. All 28 patients were discharged home the day of procedure. Twenty-six patients (92.8%) experienced complete pain relief after primary RFA, had no evidence of recurrence at final follow-up, and denied symptom recurrence at time of study completion.

The other 2 patients reported symptom recurrence after the index treatment (1 proximal femur lesion, 1 distal femur lesion). One of these patients did well initially but had a recurrence about 2 months after the primary RFA; a second RFA provided complete resolution of pain with no evidence of recurrence at time of study completion. In the other patient’s case, intermittent pain persisted for 2 weeks after the primary RFA, and evidence of recurrence was documented 3 months after surgery; a second RFA was performed shortly thereafter, but the patient was subsequently lost to follow-up.

At time of study completion, all 27 patients who had been contacted by phone denied seeking additional treatment elsewhere and stated they would have the procedure again if their symptoms ever recurred.

Discussion

Osteoid osteoma is one of the most common benign tumors of bone. Over the past 2 decades, percutaneous RFA, in comparison with open excision, has emerged as a safe and effective treatment option with minimal patient morbidity.^9-11 RFA traditionally has been performed by radiologists under CT guidance in the radiology suite. However, now orthopedic surgeons can obtain advanced intraoperative imaging beyond standard fluoroscopy. The Siemens Siremobil ISO-C3D fluoroscopic C-arm is an innovative intraoperative imaging device that functions as a standard fluoroscope but also generates 3-D reconstructions of surgical anatomy. The isocentric design and integrated motor unit allow the C-arm to move through a 190º arc while centering its beam directly on the area of interest. This data set is transferred to a computer workstation, where it is reformatted so that CT-quality images are generated in axial, sagittal, and coronal planes. This acquisition process takes only minutes, and the multiplanar images produced may be simultaneously displayed and manipulated on the screen in real time.

One concern about this technology is the amount of radiation exposure for patients, surgeons, and operating room staff. The device measures only radiation time, and the amount of exposure during that time depends on the volume and density of the radiated body. We did not calculate the amount of exposure for this study. Mean exposure time was between 20 and 40 seconds, reflecting the number of attempts required to localize the lesion and the surgeon’s experience with the technique. Although the potential for increased exposure is a valid concern, previous studies using this technology have demonstrated that a similar average exposure time is equivalent to that of standard CT, and that use of the device, over conventional techniques, potentially can lead to decreased overall radiation exposure.^12,13

This series demonstrated that OO can be safely and effectively treated with intraoperative percutaneous RFA by an orthopedic oncologist. Our success rate is very similar to rates reported in the radiology literature. Studies are needed to confirm the efficacy of this novel technique in comparison with what has been reported in that literature. Given these promising preliminary results, and the relative ease of use and minimal learning curve associated with this technology, all orthopedic oncologists should be able to offer this treatment for OO. Furthermore, this technique allows orthopedic oncologists to provide appropriate definitive treatment and care directly, rather than by referring patients to radiologists.

In the treatment of OO, we reserve RFA for lesions involving the long and short bones of the upper and lower extremities, as well as selected flat bones, such as those in the pelvis. Although percutaneous RFA of spinal lesions has been reported in the literature, we think these represent a relative contraindication for this technique; image resolution, in our opinion, is not high enough to justify risking injury to the nerves in the spinal canal, lateral recesses, and neural foramina. In addition, given the radiation exposure, we recommend caution when using this technique for a pelvic or proximal femoral lesion in a woman of childbearing age.

The proximal interphalangeal (PIP) joint plays a crucial role in hand function, accounting for an estimated 85% of the motion required to grasp an object.¹ The anatomy and biomechanics of the PIP joint, however, make it particularly prone to injury.^2,3 Dorsal PIP fracture-dislocations represent a subset of PIP injuries that often require surgical intervention.² The stability of these fracture-dislocations largely depends on the extent of articular involvement of the base of the middle phalanx. Fractures that involve less than 30% of the joint surface typically remain stable after reduction.^2,4,5 In cases in which involvement ranges from 30% to 50%, PIP joint stability is more tenuous, and more joint flexion is required to maintain concentric reduction. Fractures that involve more than 50% of the articular surface are unstable and require operative intervention.^2,5,6 Fractures that require more than 30° of flexion for reduction maintenance are generally considered unstable and may benefit from surgical intervention.²

The goals of treatment for this injury are to restore a stable, concentrically reduced joint and initiate early joint mobilization to prevent stiffness, pain, recurrent instability, and posttraumatic arthritis.^3,7 Numerous surgical interventions for unstable PIP fracture-dislocations have been proposed, including open reduction and internal fixation (ORIF),^8-10 extension-block pinning (EBP),^11-13 dynamic external fixation,^14-17 volar plate arthroplasty,^18,19 and hemi-hamate resurfacing arthroplasty.^20,21 Many of these techniques can be technically demanding and may require prolonged immobilization. EBP can be performed easily and efficiently and allows for early joint motion.

Extension-block pinning—placing a Kirschner wire (K-wire) into the head of the proximal phalanx at an angle that blocks PIP extension and prevents joint subluxation—was first described by Sugawa and colleagues¹² in 1979. In a study by Inoue and Tamura,¹¹ patients treated with EBP had a mean PIP range of motion (ROM) of 94° at a mean follow-up of 14 months. In a series of 3 case reports, Viegas²² noted an inverse relationship between extent of articular surface involvement and postoperative ROM in patients treated with EBP.

We conducted a study to expand on previous research on pain, function, and satisfaction outcomes in addition to ROM. We hypothesized that percutaneous EBP is an effective treatment for unstable dorsal PIP fracture-dislocations and has efficacy similar to that of more complex and technically demanding methods of treatment.

Materials and Methods

We retrospectively reviewed patient charts to identify candidates for this study. Inclusion criteria were unstable dorsal PIP fracture-dislocations treated with EBP and minimum 4-month follow-up. (Fracture-dislocations were deemed unstable if they involved at least 30% of the articular surface or required more than 30° of flexion for reduction maintenance.) Exclusion criteria were open injury, neurovascular or tendon injury, or any prior injury to the PIP joint.

Twelve patients (5 females, 7 males) treated over a 4-year period (2002–2006) met the inclusion criteria. Mean age was 30 years (range, 15-64 years). Each surgery was performed by Dr. Hagberg or Dr. Balk. Half the cases involved the dominant hand. Two small fingers, 4 ring fingers, 2 long fingers, and 4 index fingers were injured. The injuries were sustained in an all-terrain vehicle accident (n = 1), in falls (n = 2), while swimming (n = 1), or while playing softball (n = 3), football (n = 4), or soccer (n = 1). Mean time from injury to surgery was 7.5 days (range, 4-27 days). Extent of articular surface involvement of the base of the fractured middle phalanx was calculated using preoperatively obtained lateral radiographs.

Surgical intervention was performed in a reproducible fashion. All patients were treated with closed reduction of the PIP joint under fluoroscopic guidance. Before pinning, joint stability was assessed fluoroscopically both at rest and through an arc of motion. A single smooth 0.045-in K-wire was then inserted percutaneously into the distal and dorsal aspects of the proximal phalanx in retrograde fashion (Figure 1). During wire insertion, the distal interphalangeal joint was flexed to relax the intrinsic mechanism, and the central slip tendon was pierced just proximal to its insertion. We have not noted significant adhesion formation about the central slip with this technique, likely because of limited tendon excursion in this location. Stable joint reduction was confirmed with fluoroscopy. No attempt was made to reduce the intra-articular fracture at the base of the middle phalanx.

A therapy program was initiated 2 to 9 days after surgery. At the first postoperative visit, patients were allowed to perform active ROM (AROM) with the pin in place (Figure 1). K-wires were removed a mean of 25 days (range, 17-31 days) after surgery. A static dorsal block splint was then applied, and patients were encouraged to remove it several times per day for AROM between 20° and full flexion until 6 weeks after surgery. At that time, formal occupational therapy was commenced for another 6 weeks. If there was residual flexion contracture of the PIP joint, dynamic extension splinting was initiated after fracture consolidation.

Mean follow-up was 35.5 months (range, 4-94 months). Postoperative anteroposterior and lateral radiographs were used to evaluate maintenance of joint congruity, fracture union, remodeling, and evidence of degenerative changes. At final follow-up, grip strength of injured and contralateral hands was measured with a dynamometer (Jamar; Patterson Medical, Warrenville, Illinois). AROM and passive ROM (PROM) of the PIP joint was documented at follow-up visits. In addition, patients rated their pain on a 0-to-10 visual analog scale (VAS), with 0 representing no pain and 10 representing excruciating pain. Patients also completed a questionnaire assessing satisfaction with surgical outcome. Physical function and disability were assessed with the Quick Disabilities of the Arm, Shoulder, and Hand (QuickDASH) questionnaire. Any complications, including the need for further surgeries, were documented. Pearson correlation coefficients and Student t tests (with significance set at P < .05) were used to compare outcomes.

Results

Radiographic reduction of joint dislocation was achieved and maintained in 11 of the 12 patients at a mean follow-up of 35.5 months (range, 4-94 months). Extent of joint surface involvement, based on preoperative lateral radiographs, averaged 43% (range, 25%-75%). Although no direct articular reduction was performed, remodeling of the joint surface was consistently noted at follow-up (Figure 2). Mild radiographic degenerative changes were noted at final follow-up in 4 patients, and moderate changes were noted in 1 patient. Radiographic union was achieved in all cases, and no pin-tract infections were noted.

Mean AROM of the PIP joint at final follow-up was 84° (range, 50°-110°), with patients lacking a mean of 7° of full extension and achieving mean flexion of 91°. Mean PROM was 93° (range, 75°-110°). There was no correlation between extent of articular surface involvement and ROM. Furthermore, no correlation was found between time from injury to surgery and ROM. Patients regained full grip strength in the operative hand. At final follow-up, mean grip strength was 79.4 pounds in the operative hand and 79.6 pounds in the contralateral hand, demonstrating equal grip strengths bilaterally.

Patients overall had very low levels of pain; mean VAS score was 0.64 (range, 0-3). Mean QuickDASH score was 5.7 (range, 0-30), suggesting minimal functional impairment. One patient developed a malunion of the middle phalanx fracture resulting in a rotational deformity and required corrective osteotomy. This patient’s VAS score (3) and QuickDASH score (30) were significantly higher than those of the other patients in the study. No other complications were noted by final follow-up.

A higher level of patient satisfaction was found to be directly related to length of follow-up (P < .05). Satisfaction was inversely related to higher VAS score (P < .05) and higher QuickDASH score (P < .001). Pain at work correlated with lower satisfaction level (P < .05). There was no correlation between patient satisfaction and AROM or PROM.

Discussion

The results of this study demonstrate the efficacy of EBP in the treatment of dorsal PIP joint fracture-dislocations. EBP maintained joint dislocation reduction and allowed for early mobilization, which resulted in good ROM, minimal pain, and good functional outcomes. Of note, postoperative patient satisfaction correlated with pain but not with ROM. It is possible that EBP yielded sufficient functional ROM in all patients such that improvement beyond this threshold did not lead to further improvement in satisfaction. Hume and colleagues²³ found that mean PIP joint flexion of 60° is needed for activities of daily living. As mean PIP active flexion was 91° (range, 70°-105°) in the present study, it is possible that satisfaction did not correlate with ROM, as all 12 patients achieved active flexion of more than 60°. Despite the lack of correlation between ROM and satisfaction, early PIP joint mobilization is likely a key contributor to positive outcomes because of its significant role in cartilage healing.²⁴

Postoperative ROM in the present study is consistent with that in other reports of patients with PIP joint fracture-dislocations treated with EBP.^11,12,22 In a study by Inoue and Tamura,¹¹ 14 such patients had mean PIP ROM of 94° at a mean follow-up of 14 months. Viegas²² followed a series of 3 patients for a mean of 7 weeks. At final follow-up, their mean PIP arc of motion was 71°; they lacked 12° of full extension and achieved 83° of flexion. The larger PIP arc of motion (84°) found in the present study may be due to our significantly longer follow-up (35 months). Unlike us, Viegas²² noted an inverse relationship between extent of articular surface involvement and postoperative ROM. Our finding a lack of correlation may be a result of the significant amount of joint remodeling noted on follow-up radiographs.

Studies of transarticular pinning of PIP joints after dorsal PIP fracture-dislocations have reported outcomes similar to ours.^25,26 Newington and colleagues²⁵ evaluated 10 cases of transarticular pinning of the PIP joint and found mean arc of motion of 85° and equal grip strengths between injured and contralateral hands. In a series of 19 patients with PIP fracture-dislocations, Aladin and Davis²⁶ noted similar outcomes of transarticular K-wire fixation and ORIF. In both of their treatment groups, however, there was evidence of PIP joint incongruity and subluxation. Of note, PIP arc motion was lower in their study than in ours.

Recent studies have evaluated unstable PIP fracture-dislocations treated with both EBP and percutaneous reduction and pinning with a second K-wire.^13,27 At a mean follow-up of 18 months, Vitale and colleagues¹³ noted maintenance of concentric fracture reduction, good PIP ROM (mean range, 4°-93°), and low VAS and DASH scores (1.4 and 8, respectively). Waris and Alanen²⁷ noted mean PIP AROM of 83° and low VAS and DASH scores (1 and 4, respectively). The EBP technique used in the present study did not involve percutaneous fracture reduction but achieved equally good ROM and VAS and QuickDASH scores.

Clinical outcomes of EBP of PIP joint fracture-dislocations are also comparable to outcomes of more complex treatment methods.^{8-10,15-19,21,26,28-33} Dynamic distraction external fixation has led to equally good ROM (mean AROM, 80°-85°^15,16) and VAS scores, but with a higher incidence of pin-site infection.^14-17 ORIF of the intra-articular middle phalanx fracture has the advantage of obtaining a direct anatomical reduction, but clinical outcomes are similar to those in the present study (mean AROM, 70°; 78% pain-free⁹), and flexion contractures have been noted.^8-10 Furthermore, reduction of the fractured PIP joint articular surface has not been shown to be necessary for good outcomes.^16,34 This may be explained in part by PIP joint remodeling, which has been routinely observed on long-term follow-up by the senior authors of the present study. Hemi-hamate autografting and volar plate arthroplasty are other options that have had promising results in the treatment of acute and chronic unstable PIP fracture-dislocations.^18-21 However, the postoperative ROM (mean AROM, 61°-85°^18,21), VAS scores, and patient satisfaction (91% very satisfied²¹) of these operations are similar to those of EBP in the present study and may not justify the longer operative times and technical challenges associated with these techniques.

We believe that our study group’s 1 complication, a malunion that was treated with corrective osteotomy, resulted from lack of appreciation of the degree of injury. The teenaged female patient’s index finger PIP joint had a rotational malalignment that was not appreciated before or during surgery. After pinning and after ROM was restored, the index finger was observed crossing over the middle finger with digital flexion. The patient returned to the operating room for corrective osteotomy.

We recommend that surgeons assess alignment carefully, before and during surgery, when considering this technique. Although complications are rare, the technique is not for patients with rotational malalignment; ORIF may be more suitable in these cases. In addition, though EBP may be appropriate for pilon-type injuries, as it allows for early AROM, our procedure of choice for pilon fracture is dynamic external fixation, which in addition to allowing for AROM provides ligamentotaxis. In the event that a large volar articular fragment extends into the middle phalanx diaphysis, we typically proceed with ORIF through a volar shotgun approach. At our institution, injuries lasting more than 3 months are often treated with volar plate arthroplasty or hemi-hamate resurfacing. Finally, we believe that caution should be exercised when using this technique in patients with more than 50% articular involvement. In the present study, though we used this treatment in cases of up to 75% surface involvement, alternative techniques, such as hemi-hamate resurfacing arthroplasty, may provide a better volar bony buttress and limit the risk for recurrent instability. Despite its relative contraindications, our technique has been appropriate for more than 90% of the acute PIP fracture-dislocations we have seen.

This study expands on prior research by demonstrating good function, satisfaction, and pain outcomes of percutaneous EBP in the treatment of unstable dorsal PIP fracture-dislocations. In addition, this study demonstrated that the efficacy of EBP is similar to that of more complex and technically demanding methods of treatment. Our technique has the advantage of simplicity. It obviates the soft-tissue damage required for ORIF and more complex fixation techniques. Furthermore, use of this simple technique may save time and costs and lead to more reproducible outcomes.

One limitation of this study is its small sample size. It is possible that outcomes may have been different with a larger sample. Furthermore, we did not make a direct comparison with other treatment methods. To better determine the optimal treatment method for this fracture type, future studies should prospectively evaluate outcomes for multiple treatment modalities in a randomized fashion.

The proximal interphalangeal (PIP) joint plays a crucial role in hand function, accounting for an estimated 85% of the motion required to grasp an object.¹ The anatomy and biomechanics of the PIP joint, however, make it particularly prone to injury.^2,3 Dorsal PIP fracture-dislocations represent a subset of PIP injuries that often require surgical intervention.² The stability of these fracture-dislocations largely depends on the extent of articular involvement of the base of the middle phalanx. Fractures that involve less than 30% of the joint surface typically remain stable after reduction.^2,4,5 In cases in which involvement ranges from 30% to 50%, PIP joint stability is more tenuous, and more joint flexion is required to maintain concentric reduction. Fractures that involve more than 50% of the articular surface are unstable and require operative intervention.^2,5,6 Fractures that require more than 30° of flexion for reduction maintenance are generally considered unstable and may benefit from surgical intervention.²

The goals of treatment for this injury are to restore a stable, concentrically reduced joint and initiate early joint mobilization to prevent stiffness, pain, recurrent instability, and posttraumatic arthritis.^3,7 Numerous surgical interventions for unstable PIP fracture-dislocations have been proposed, including open reduction and internal fixation (ORIF),^8-10 extension-block pinning (EBP),^11-13 dynamic external fixation,^14-17 volar plate arthroplasty,^18,19 and hemi-hamate resurfacing arthroplasty.^20,21 Many of these techniques can be technically demanding and may require prolonged immobilization. EBP can be performed easily and efficiently and allows for early joint motion.

Extension-block pinning—placing a Kirschner wire (K-wire) into the head of the proximal phalanx at an angle that blocks PIP extension and prevents joint subluxation—was first described by Sugawa and colleagues¹² in 1979. In a study by Inoue and Tamura,¹¹ patients treated with EBP had a mean PIP range of motion (ROM) of 94° at a mean follow-up of 14 months. In a series of 3 case reports, Viegas²² noted an inverse relationship between extent of articular surface involvement and postoperative ROM in patients treated with EBP.

We conducted a study to expand on previous research on pain, function, and satisfaction outcomes in addition to ROM. We hypothesized that percutaneous EBP is an effective treatment for unstable dorsal PIP fracture-dislocations and has efficacy similar to that of more complex and technically demanding methods of treatment.

Materials and Methods

We retrospectively reviewed patient charts to identify candidates for this study. Inclusion criteria were unstable dorsal PIP fracture-dislocations treated with EBP and minimum 4-month follow-up. (Fracture-dislocations were deemed unstable if they involved at least 30% of the articular surface or required more than 30° of flexion for reduction maintenance.) Exclusion criteria were open injury, neurovascular or tendon injury, or any prior injury to the PIP joint.

Twelve patients (5 females, 7 males) treated over a 4-year period (2002–2006) met the inclusion criteria. Mean age was 30 years (range, 15-64 years). Each surgery was performed by Dr. Hagberg or Dr. Balk. Half the cases involved the dominant hand. Two small fingers, 4 ring fingers, 2 long fingers, and 4 index fingers were injured. The injuries were sustained in an all-terrain vehicle accident (n = 1), in falls (n = 2), while swimming (n = 1), or while playing softball (n = 3), football (n = 4), or soccer (n = 1). Mean time from injury to surgery was 7.5 days (range, 4-27 days). Extent of articular surface involvement of the base of the fractured middle phalanx was calculated using preoperatively obtained lateral radiographs.

Surgical intervention was performed in a reproducible fashion. All patients were treated with closed reduction of the PIP joint under fluoroscopic guidance. Before pinning, joint stability was assessed fluoroscopically both at rest and through an arc of motion. A single smooth 0.045-in K-wire was then inserted percutaneously into the distal and dorsal aspects of the proximal phalanx in retrograde fashion (Figure 1). During wire insertion, the distal interphalangeal joint was flexed to relax the intrinsic mechanism, and the central slip tendon was pierced just proximal to its insertion. We have not noted significant adhesion formation about the central slip with this technique, likely because of limited tendon excursion in this location. Stable joint reduction was confirmed with fluoroscopy. No attempt was made to reduce the intra-articular fracture at the base of the middle phalanx.

A therapy program was initiated 2 to 9 days after surgery. At the first postoperative visit, patients were allowed to perform active ROM (AROM) with the pin in place (Figure 1). K-wires were removed a mean of 25 days (range, 17-31 days) after surgery. A static dorsal block splint was then applied, and patients were encouraged to remove it several times per day for AROM between 20° and full flexion until 6 weeks after surgery. At that time, formal occupational therapy was commenced for another 6 weeks. If there was residual flexion contracture of the PIP joint, dynamic extension splinting was initiated after fracture consolidation.

Mean follow-up was 35.5 months (range, 4-94 months). Postoperative anteroposterior and lateral radiographs were used to evaluate maintenance of joint congruity, fracture union, remodeling, and evidence of degenerative changes. At final follow-up, grip strength of injured and contralateral hands was measured with a dynamometer (Jamar; Patterson Medical, Warrenville, Illinois). AROM and passive ROM (PROM) of the PIP joint was documented at follow-up visits. In addition, patients rated their pain on a 0-to-10 visual analog scale (VAS), with 0 representing no pain and 10 representing excruciating pain. Patients also completed a questionnaire assessing satisfaction with surgical outcome. Physical function and disability were assessed with the Quick Disabilities of the Arm, Shoulder, and Hand (QuickDASH) questionnaire. Any complications, including the need for further surgeries, were documented. Pearson correlation coefficients and Student t tests (with significance set at P < .05) were used to compare outcomes.

Results

Radiographic reduction of joint dislocation was achieved and maintained in 11 of the 12 patients at a mean follow-up of 35.5 months (range, 4-94 months). Extent of joint surface involvement, based on preoperative lateral radiographs, averaged 43% (range, 25%-75%). Although no direct articular reduction was performed, remodeling of the joint surface was consistently noted at follow-up (Figure 2). Mild radiographic degenerative changes were noted at final follow-up in 4 patients, and moderate changes were noted in 1 patient. Radiographic union was achieved in all cases, and no pin-tract infections were noted.

Mean AROM of the PIP joint at final follow-up was 84° (range, 50°-110°), with patients lacking a mean of 7° of full extension and achieving mean flexion of 91°. Mean PROM was 93° (range, 75°-110°). There was no correlation between extent of articular surface involvement and ROM. Furthermore, no correlation was found between time from injury to surgery and ROM. Patients regained full grip strength in the operative hand. At final follow-up, mean grip strength was 79.4 pounds in the operative hand and 79.6 pounds in the contralateral hand, demonstrating equal grip strengths bilaterally.

Patients overall had very low levels of pain; mean VAS score was 0.64 (range, 0-3). Mean QuickDASH score was 5.7 (range, 0-30), suggesting minimal functional impairment. One patient developed a malunion of the middle phalanx fracture resulting in a rotational deformity and required corrective osteotomy. This patient’s VAS score (3) and QuickDASH score (30) were significantly higher than those of the other patients in the study. No other complications were noted by final follow-up.

A higher level of patient satisfaction was found to be directly related to length of follow-up (P < .05). Satisfaction was inversely related to higher VAS score (P < .05) and higher QuickDASH score (P < .001). Pain at work correlated with lower satisfaction level (P < .05). There was no correlation between patient satisfaction and AROM or PROM.

Discussion

The results of this study demonstrate the efficacy of EBP in the treatment of dorsal PIP joint fracture-dislocations. EBP maintained joint dislocation reduction and allowed for early mobilization, which resulted in good ROM, minimal pain, and good functional outcomes. Of note, postoperative patient satisfaction correlated with pain but not with ROM. It is possible that EBP yielded sufficient functional ROM in all patients such that improvement beyond this threshold did not lead to further improvement in satisfaction. Hume and colleagues²³ found that mean PIP joint flexion of 60° is needed for activities of daily living. As mean PIP active flexion was 91° (range, 70°-105°) in the present study, it is possible that satisfaction did not correlate with ROM, as all 12 patients achieved active flexion of more than 60°. Despite the lack of correlation between ROM and satisfaction, early PIP joint mobilization is likely a key contributor to positive outcomes because of its significant role in cartilage healing.²⁴

Postoperative ROM in the present study is consistent with that in other reports of patients with PIP joint fracture-dislocations treated with EBP.^11,12,22 In a study by Inoue and Tamura,¹¹ 14 such patients had mean PIP ROM of 94° at a mean follow-up of 14 months. Viegas²² followed a series of 3 patients for a mean of 7 weeks. At final follow-up, their mean PIP arc of motion was 71°; they lacked 12° of full extension and achieved 83° of flexion. The larger PIP arc of motion (84°) found in the present study may be due to our significantly longer follow-up (35 months). Unlike us, Viegas²² noted an inverse relationship between extent of articular surface involvement and postoperative ROM. Our finding a lack of correlation may be a result of the significant amount of joint remodeling noted on follow-up radiographs.

Studies of transarticular pinning of PIP joints after dorsal PIP fracture-dislocations have reported outcomes similar to ours.^25,26 Newington and colleagues²⁵ evaluated 10 cases of transarticular pinning of the PIP joint and found mean arc of motion of 85° and equal grip strengths between injured and contralateral hands. In a series of 19 patients with PIP fracture-dislocations, Aladin and Davis²⁶ noted similar outcomes of transarticular K-wire fixation and ORIF. In both of their treatment groups, however, there was evidence of PIP joint incongruity and subluxation. Of note, PIP arc motion was lower in their study than in ours.

Recent studies have evaluated unstable PIP fracture-dislocations treated with both EBP and percutaneous reduction and pinning with a second K-wire.^13,27 At a mean follow-up of 18 months, Vitale and colleagues¹³ noted maintenance of concentric fracture reduction, good PIP ROM (mean range, 4°-93°), and low VAS and DASH scores (1.4 and 8, respectively). Waris and Alanen²⁷ noted mean PIP AROM of 83° and low VAS and DASH scores (1 and 4, respectively). The EBP technique used in the present study did not involve percutaneous fracture reduction but achieved equally good ROM and VAS and QuickDASH scores.

Clinical outcomes of EBP of PIP joint fracture-dislocations are also comparable to outcomes of more complex treatment methods.^{8-10,15-19,21,26,28-33} Dynamic distraction external fixation has led to equally good ROM (mean AROM, 80°-85°^15,16) and VAS scores, but with a higher incidence of pin-site infection.^14-17 ORIF of the intra-articular middle phalanx fracture has the advantage of obtaining a direct anatomical reduction, but clinical outcomes are similar to those in the present study (mean AROM, 70°; 78% pain-free⁹), and flexion contractures have been noted.^8-10 Furthermore, reduction of the fractured PIP joint articular surface has not been shown to be necessary for good outcomes.^16,34 This may be explained in part by PIP joint remodeling, which has been routinely observed on long-term follow-up by the senior authors of the present study. Hemi-hamate autografting and volar plate arthroplasty are other options that have had promising results in the treatment of acute and chronic unstable PIP fracture-dislocations.^18-21 However, the postoperative ROM (mean AROM, 61°-85°^18,21), VAS scores, and patient satisfaction (91% very satisfied²¹) of these operations are similar to those of EBP in the present study and may not justify the longer operative times and technical challenges associated with these techniques.

We believe that our study group’s 1 complication, a malunion that was treated with corrective osteotomy, resulted from lack of appreciation of the degree of injury. The teenaged female patient’s index finger PIP joint had a rotational malalignment that was not appreciated before or during surgery. After pinning and after ROM was restored, the index finger was observed crossing over the middle finger with digital flexion. The patient returned to the operating room for corrective osteotomy.

We recommend that surgeons assess alignment carefully, before and during surgery, when considering this technique. Although complications are rare, the technique is not for patients with rotational malalignment; ORIF may be more suitable in these cases. In addition, though EBP may be appropriate for pilon-type injuries, as it allows for early AROM, our procedure of choice for pilon fracture is dynamic external fixation, which in addition to allowing for AROM provides ligamentotaxis. In the event that a large volar articular fragment extends into the middle phalanx diaphysis, we typically proceed with ORIF through a volar shotgun approach. At our institution, injuries lasting more than 3 months are often treated with volar plate arthroplasty or hemi-hamate resurfacing. Finally, we believe that caution should be exercised when using this technique in patients with more than 50% articular involvement. In the present study, though we used this treatment in cases of up to 75% surface involvement, alternative techniques, such as hemi-hamate resurfacing arthroplasty, may provide a better volar bony buttress and limit the risk for recurrent instability. Despite its relative contraindications, our technique has been appropriate for more than 90% of the acute PIP fracture-dislocations we have seen.

This study expands on prior research by demonstrating good function, satisfaction, and pain outcomes of percutaneous EBP in the treatment of unstable dorsal PIP fracture-dislocations. In addition, this study demonstrated that the efficacy of EBP is similar to that of more complex and technically demanding methods of treatment. Our technique has the advantage of simplicity. It obviates the soft-tissue damage required for ORIF and more complex fixation techniques. Furthermore, use of this simple technique may save time and costs and lead to more reproducible outcomes.

One limitation of this study is its small sample size. It is possible that outcomes may have been different with a larger sample. Furthermore, we did not make a direct comparison with other treatment methods. To better determine the optimal treatment method for this fracture type, future studies should prospectively evaluate outcomes for multiple treatment modalities in a randomized fashion.

The proximal interphalangeal (PIP) joint plays a crucial role in hand function, accounting for an estimated 85% of the motion required to grasp an object.¹ The anatomy and biomechanics of the PIP joint, however, make it particularly prone to injury.^2,3 Dorsal PIP fracture-dislocations represent a subset of PIP injuries that often require surgical intervention.² The stability of these fracture-dislocations largely depends on the extent of articular involvement of the base of the middle phalanx. Fractures that involve less than 30% of the joint surface typically remain stable after reduction.^2,4,5 In cases in which involvement ranges from 30% to 50%, PIP joint stability is more tenuous, and more joint flexion is required to maintain concentric reduction. Fractures that involve more than 50% of the articular surface are unstable and require operative intervention.^2,5,6 Fractures that require more than 30° of flexion for reduction maintenance are generally considered unstable and may benefit from surgical intervention.²

The goals of treatment for this injury are to restore a stable, concentrically reduced joint and initiate early joint mobilization to prevent stiffness, pain, recurrent instability, and posttraumatic arthritis.^3,7 Numerous surgical interventions for unstable PIP fracture-dislocations have been proposed, including open reduction and internal fixation (ORIF),^8-10 extension-block pinning (EBP),^11-13 dynamic external fixation,^14-17 volar plate arthroplasty,^18,19 and hemi-hamate resurfacing arthroplasty.^20,21 Many of these techniques can be technically demanding and may require prolonged immobilization. EBP can be performed easily and efficiently and allows for early joint motion.

Extension-block pinning—placing a Kirschner wire (K-wire) into the head of the proximal phalanx at an angle that blocks PIP extension and prevents joint subluxation—was first described by Sugawa and colleagues¹² in 1979. In a study by Inoue and Tamura,¹¹ patients treated with EBP had a mean PIP range of motion (ROM) of 94° at a mean follow-up of 14 months. In a series of 3 case reports, Viegas²² noted an inverse relationship between extent of articular surface involvement and postoperative ROM in patients treated with EBP.

We conducted a study to expand on previous research on pain, function, and satisfaction outcomes in addition to ROM. We hypothesized that percutaneous EBP is an effective treatment for unstable dorsal PIP fracture-dislocations and has efficacy similar to that of more complex and technically demanding methods of treatment.

Materials and Methods

We retrospectively reviewed patient charts to identify candidates for this study. Inclusion criteria were unstable dorsal PIP fracture-dislocations treated with EBP and minimum 4-month follow-up. (Fracture-dislocations were deemed unstable if they involved at least 30% of the articular surface or required more than 30° of flexion for reduction maintenance.) Exclusion criteria were open injury, neurovascular or tendon injury, or any prior injury to the PIP joint.

Twelve patients (5 females, 7 males) treated over a 4-year period (2002–2006) met the inclusion criteria. Mean age was 30 years (range, 15-64 years). Each surgery was performed by Dr. Hagberg or Dr. Balk. Half the cases involved the dominant hand. Two small fingers, 4 ring fingers, 2 long fingers, and 4 index fingers were injured. The injuries were sustained in an all-terrain vehicle accident (n = 1), in falls (n = 2), while swimming (n = 1), or while playing softball (n = 3), football (n = 4), or soccer (n = 1). Mean time from injury to surgery was 7.5 days (range, 4-27 days). Extent of articular surface involvement of the base of the fractured middle phalanx was calculated using preoperatively obtained lateral radiographs.

Surgical intervention was performed in a reproducible fashion. All patients were treated with closed reduction of the PIP joint under fluoroscopic guidance. Before pinning, joint stability was assessed fluoroscopically both at rest and through an arc of motion. A single smooth 0.045-in K-wire was then inserted percutaneously into the distal and dorsal aspects of the proximal phalanx in retrograde fashion (Figure 1). During wire insertion, the distal interphalangeal joint was flexed to relax the intrinsic mechanism, and the central slip tendon was pierced just proximal to its insertion. We have not noted significant adhesion formation about the central slip with this technique, likely because of limited tendon excursion in this location. Stable joint reduction was confirmed with fluoroscopy. No attempt was made to reduce the intra-articular fracture at the base of the middle phalanx.

A therapy program was initiated 2 to 9 days after surgery. At the first postoperative visit, patients were allowed to perform active ROM (AROM) with the pin in place (Figure 1). K-wires were removed a mean of 25 days (range, 17-31 days) after surgery. A static dorsal block splint was then applied, and patients were encouraged to remove it several times per day for AROM between 20° and full flexion until 6 weeks after surgery. At that time, formal occupational therapy was commenced for another 6 weeks. If there was residual flexion contracture of the PIP joint, dynamic extension splinting was initiated after fracture consolidation.

Mean follow-up was 35.5 months (range, 4-94 months). Postoperative anteroposterior and lateral radiographs were used to evaluate maintenance of joint congruity, fracture union, remodeling, and evidence of degenerative changes. At final follow-up, grip strength of injured and contralateral hands was measured with a dynamometer (Jamar; Patterson Medical, Warrenville, Illinois). AROM and passive ROM (PROM) of the PIP joint was documented at follow-up visits. In addition, patients rated their pain on a 0-to-10 visual analog scale (VAS), with 0 representing no pain and 10 representing excruciating pain. Patients also completed a questionnaire assessing satisfaction with surgical outcome. Physical function and disability were assessed with the Quick Disabilities of the Arm, Shoulder, and Hand (QuickDASH) questionnaire. Any complications, including the need for further surgeries, were documented. Pearson correlation coefficients and Student t tests (with significance set at P < .05) were used to compare outcomes.

Results

Radiographic reduction of joint dislocation was achieved and maintained in 11 of the 12 patients at a mean follow-up of 35.5 months (range, 4-94 months). Extent of joint surface involvement, based on preoperative lateral radiographs, averaged 43% (range, 25%-75%). Although no direct articular reduction was performed, remodeling of the joint surface was consistently noted at follow-up (Figure 2). Mild radiographic degenerative changes were noted at final follow-up in 4 patients, and moderate changes were noted in 1 patient. Radiographic union was achieved in all cases, and no pin-tract infections were noted.

Mean AROM of the PIP joint at final follow-up was 84° (range, 50°-110°), with patients lacking a mean of 7° of full extension and achieving mean flexion of 91°. Mean PROM was 93° (range, 75°-110°). There was no correlation between extent of articular surface involvement and ROM. Furthermore, no correlation was found between time from injury to surgery and ROM. Patients regained full grip strength in the operative hand. At final follow-up, mean grip strength was 79.4 pounds in the operative hand and 79.6 pounds in the contralateral hand, demonstrating equal grip strengths bilaterally.

Patients overall had very low levels of pain; mean VAS score was 0.64 (range, 0-3). Mean QuickDASH score was 5.7 (range, 0-30), suggesting minimal functional impairment. One patient developed a malunion of the middle phalanx fracture resulting in a rotational deformity and required corrective osteotomy. This patient’s VAS score (3) and QuickDASH score (30) were significantly higher than those of the other patients in the study. No other complications were noted by final follow-up.

A higher level of patient satisfaction was found to be directly related to length of follow-up (P < .05). Satisfaction was inversely related to higher VAS score (P < .05) and higher QuickDASH score (P < .001). Pain at work correlated with lower satisfaction level (P < .05). There was no correlation between patient satisfaction and AROM or PROM.

Discussion

The results of this study demonstrate the efficacy of EBP in the treatment of dorsal PIP joint fracture-dislocations. EBP maintained joint dislocation reduction and allowed for early mobilization, which resulted in good ROM, minimal pain, and good functional outcomes. Of note, postoperative patient satisfaction correlated with pain but not with ROM. It is possible that EBP yielded sufficient functional ROM in all patients such that improvement beyond this threshold did not lead to further improvement in satisfaction. Hume and colleagues²³ found that mean PIP joint flexion of 60° is needed for activities of daily living. As mean PIP active flexion was 91° (range, 70°-105°) in the present study, it is possible that satisfaction did not correlate with ROM, as all 12 patients achieved active flexion of more than 60°. Despite the lack of correlation between ROM and satisfaction, early PIP joint mobilization is likely a key contributor to positive outcomes because of its significant role in cartilage healing.²⁴

Postoperative ROM in the present study is consistent with that in other reports of patients with PIP joint fracture-dislocations treated with EBP.^11,12,22 In a study by Inoue and Tamura,¹¹ 14 such patients had mean PIP ROM of 94° at a mean follow-up of 14 months. Viegas²² followed a series of 3 patients for a mean of 7 weeks. At final follow-up, their mean PIP arc of motion was 71°; they lacked 12° of full extension and achieved 83° of flexion. The larger PIP arc of motion (84°) found in the present study may be due to our significantly longer follow-up (35 months). Unlike us, Viegas²² noted an inverse relationship between extent of articular surface involvement and postoperative ROM. Our finding a lack of correlation may be a result of the significant amount of joint remodeling noted on follow-up radiographs.

Studies of transarticular pinning of PIP joints after dorsal PIP fracture-dislocations have reported outcomes similar to ours.^25,26 Newington and colleagues²⁵ evaluated 10 cases of transarticular pinning of the PIP joint and found mean arc of motion of 85° and equal grip strengths between injured and contralateral hands. In a series of 19 patients with PIP fracture-dislocations, Aladin and Davis²⁶ noted similar outcomes of transarticular K-wire fixation and ORIF. In both of their treatment groups, however, there was evidence of PIP joint incongruity and subluxation. Of note, PIP arc motion was lower in their study than in ours.

Recent studies have evaluated unstable PIP fracture-dislocations treated with both EBP and percutaneous reduction and pinning with a second K-wire.^13,27 At a mean follow-up of 18 months, Vitale and colleagues¹³ noted maintenance of concentric fracture reduction, good PIP ROM (mean range, 4°-93°), and low VAS and DASH scores (1.4 and 8, respectively). Waris and Alanen²⁷ noted mean PIP AROM of 83° and low VAS and DASH scores (1 and 4, respectively). The EBP technique used in the present study did not involve percutaneous fracture reduction but achieved equally good ROM and VAS and QuickDASH scores.

Clinical outcomes of EBP of PIP joint fracture-dislocations are also comparable to outcomes of more complex treatment methods.^{8-10,15-19,21,26,28-33} Dynamic distraction external fixation has led to equally good ROM (mean AROM, 80°-85°^15,16) and VAS scores, but with a higher incidence of pin-site infection.^14-17 ORIF of the intra-articular middle phalanx fracture has the advantage of obtaining a direct anatomical reduction, but clinical outcomes are similar to those in the present study (mean AROM, 70°; 78% pain-free⁹), and flexion contractures have been noted.^8-10 Furthermore, reduction of the fractured PIP joint articular surface has not been shown to be necessary for good outcomes.^16,34 This may be explained in part by PIP joint remodeling, which has been routinely observed on long-term follow-up by the senior authors of the present study. Hemi-hamate autografting and volar plate arthroplasty are other options that have had promising results in the treatment of acute and chronic unstable PIP fracture-dislocations.^18-21 However, the postoperative ROM (mean AROM, 61°-85°^18,21), VAS scores, and patient satisfaction (91% very satisfied²¹) of these operations are similar to those of EBP in the present study and may not justify the longer operative times and technical challenges associated with these techniques.

We believe that our study group’s 1 complication, a malunion that was treated with corrective osteotomy, resulted from lack of appreciation of the degree of injury. The teenaged female patient’s index finger PIP joint had a rotational malalignment that was not appreciated before or during surgery. After pinning and after ROM was restored, the index finger was observed crossing over the middle finger with digital flexion. The patient returned to the operating room for corrective osteotomy.

We recommend that surgeons assess alignment carefully, before and during surgery, when considering this technique. Although complications are rare, the technique is not for patients with rotational malalignment; ORIF may be more suitable in these cases. In addition, though EBP may be appropriate for pilon-type injuries, as it allows for early AROM, our procedure of choice for pilon fracture is dynamic external fixation, which in addition to allowing for AROM provides ligamentotaxis. In the event that a large volar articular fragment extends into the middle phalanx diaphysis, we typically proceed with ORIF through a volar shotgun approach. At our institution, injuries lasting more than 3 months are often treated with volar plate arthroplasty or hemi-hamate resurfacing. Finally, we believe that caution should be exercised when using this technique in patients with more than 50% articular involvement. In the present study, though we used this treatment in cases of up to 75% surface involvement, alternative techniques, such as hemi-hamate resurfacing arthroplasty, may provide a better volar bony buttress and limit the risk for recurrent instability. Despite its relative contraindications, our technique has been appropriate for more than 90% of the acute PIP fracture-dislocations we have seen.

This study expands on prior research by demonstrating good function, satisfaction, and pain outcomes of percutaneous EBP in the treatment of unstable dorsal PIP fracture-dislocations. In addition, this study demonstrated that the efficacy of EBP is similar to that of more complex and technically demanding methods of treatment. Our technique has the advantage of simplicity. It obviates the soft-tissue damage required for ORIF and more complex fixation techniques. Furthermore, use of this simple technique may save time and costs and lead to more reproducible outcomes.

One limitation of this study is its small sample size. It is possible that outcomes may have been different with a larger sample. Furthermore, we did not make a direct comparison with other treatment methods. To better determine the optimal treatment method for this fracture type, future studies should prospectively evaluate outcomes for multiple treatment modalities in a randomized fashion.

Soft-tissue complications are a known problem in the treatment of pilon fractures of the distal end of the tibia. These fractures typically occur as the result of a high-energy mechanism, and axial load and shear forces often lead to a severe soft-tissue injury. In many cases, these injuries may require additional procedures to provide adequate soft-tissue coverage. These procedures can include use of either a rotational muscle flap or a free flap transfer. In some cases, however, these flaps are not possible secondary to vascular compromise.

In this article, we report the case of a pilon fracture combined with severe soft-tissue injury and vascular compromise of the leg. A cross-leg fasciocutaneous flap was performed as a salvage procedure for coverage of the soft-tissue defect. The patient provided written informed consent for print and electronic publication of this case report.

Case Report

A 23-year-old man sustained a left grade III open pilon fracture after a fall off a cherry picker. He was initially treated with irrigation and débridement of the open anteromedial wound, wound closure, application of external fixation, and open reduction and internal fixation (ORIF) of the concomitant comminuted fibular fracture. Operative fixation of the pilon was performed 3 weeks after injury, once skin and soft tissues were in acceptable condition (Figure 1). Skin closure was performed with 2-0 Vicryl sutures (Ethicon, Inc, Somerville, New Jersey) followed by 3-0 nylon skin sutures and No. 2 nylon retention sutures to reduce tension at the incision.

On postoperative day 17, the patient was found to have skin necrosis with exposed hardware over the medial laceration that had resulted from the open fracture (Figure 2). The wound measured 7×6 cm. The plastic surgery team was consulted, and a soft-tissue flap was recommended. Preoperative computed tomography angiogram (Figure 3) revealed 1 vessel runoff in the leg, constituting the peroneal artery, and a conventional angiogram confirmed this finding (Figure 4). Despite these findings, the patient was taken to the operating room 4 weeks after initial injury to try to find a vessel compatible with anastomosis. Intraoperative wound exploration confirmed no patent blood supply for local soft-tissue flap coverage. Therefore, the wound was irrigated and débrided, and a vacuum-assisted closure (VAC) dressing was applied despite exposed hardware and bone. A decision was then made to attempt a cross-leg flap as a salvage procedure, and VAC dressing therapy was continued for several weeks to prepare the recipient site (Figure 5).

Seven weeks after injury, the patient was taken to the operating room by the orthopedic surgery and plastic surgery teams. After débridement, a fasciocutaneous flap was raised from the middle third of the contralateral leg (Figure 6) based on a posterior tibial artery perforator. The flap, which measured 7×7 cm (sufficient to cover the defect), was raised from lateral to medial from the posterior aspect of the leg with the pedicle located on the medial aspect of the right leg. Flap placement was facilitated by flexing the left knee to 80°. The flap was sutured into place with 4-0 Vicryl deep sutures followed by 4-0 nylon and superficial sutures in an interrupted fashion (Figure 7). Rigid external fixation was then applied to both extremities, bridging them together in optimal position (Figure 8). This construct included 2 short bars that would elevate the patient’s heels off the bed to reduce the chance of heel decubiti. Although including the feet in the external fixator construct may help prevent equinus contracture, we splinted the ankles in neutral position immediately after surgery so that we could begin early range-of-motion (ROM) exercises of the ankles to prevent stiffness. Ankle ROM exercises were started once the flap incorporated, 3 weeks after placement of the external fixator. Lacking medical insurance coverage, the patient could not be admitted to a rehabilitation facility or receive home care. He lived independently and had no help at home, so he had to remain hospitalized after placement of the external fixator. While hospitalized, the surgical site was treated with frequent dressing changes, including use of bacitracin and nonadherent dressing.

After flap coverage and 4 weeks of bed rest, a base clamping test confirmed the flap was incorporated into the recipient bed. The patient was then returned to the operating room for removal of the external fixator and skin grafting of the donor site. After surgery, he was started on physical therapy, including exercises for bilateral hip, knee, and ankle ROM and strengthening of the lower extremities. Four months after initial injury, the fracture was healed, based on bone consolidation, seen on radiographs, that is consistent with other pilon fractures treated at our institution. Six months after external fixator removal, the patient was able to ambulate independently with minimal discomfort (Figure 9). Passive and active ankle ROM was 20° of dorsiflexion and 25° of plantarflexion, compared with 25° of dorsiflexion and 45° of plantarflexion on the contralateral extremity. Subtalar motion had some stiffness with a 10° arc, compared with a 25° arc on the contralateral extremity. On simple manual testing, the patient had 5/5 motor strength with dorsiflexion, plantarflexion, inversion, and eversion. He returned to full duty as a landscaper about 1 year after initial injury and had no recurrence of wound complications or infection.

Discussion

Fractures of the distal tibia are commonly known as pilon or plafond fractures. They represent up to 10% of all tibial fractures. The injury consists of an intra-articular fracture of the tibiotalar joint with varying degrees of proximal extension into the tibial metaphysis. The etiology is an axial load on the tibia with or without a rotational force.¹ Treatment is challenging. The literature includes many reports of wound and soft-tissue complications after ORIF. In 1969, Rüedi and Allgöwer² published recommendations that have become the standard for treatment of pilon fractures. Twelve percent of the 84 fractures included in their study were associated with wound complications. In 2004, Sirkin and colleagues³ suggested that wound problems associated with ORIF of pilon fractures may be caused by attempts at immediate fixation through swollen soft tissue. They postulated that staging the procedure and waiting for decreased soft-tissue swelling may reduce the incidence of wound complications. In their series, only 2.9% of closed pilon fractures and only 9.1% of open fractures had any wound complications, and none of their patients required skin grafts, rotation flaps, or free tissue transfers.

However, soft-tissue complications still remain a significant threat in the treatment of pilon fracture, and cases that require additional procedures for soft-tissue coverage are common. In some cases, wound necrosis may lead to below-knee amputation.⁴ There are several coverage options, including local rotational flaps using the soleus muscle^5,6 as well as free flaps using the latissimus dorsi, gracilis, or rectus abdominis muscles.⁷ These options require a sufficient blood supply to the region.

Many high-energy pilon fractures may be associated with vascular injury, and therefore flap survival may be compromised. We have reported such a case in the present article. Our patient’s preoperative angiogram indicated he had 1-vessel runoff to the distal leg—a situation incompatible with free tissue transfer. It is not clear whether this finding is secondary to trauma to the leg or is caused by an anatomical anomaly. Nevertheless, the poor vascularity posed a challenge to providing soft-tissue coverage. Cross-finger⁸ and cross-foot⁹ flaps have been described in upper and lower extremity injuries. In 2006, Zhao and colleagues¹⁰ reported on 5 patients with tibia and/or hardware exposure after operative fixation of tibia fractures. These patients had poor local soft tissue around the wound and therefore underwent cross-leg flap for coverage. It is not clear where the soft-tissue defects were located and whether any studies were performed to assess the local blood flow.

From our patient’s case, we learned that multiple factors should be considered when assessing such high-energy injuries. First, respecting the soft tissues is of paramount importance. Our initial management on presentation consisted of irrigation and débridement of the wound, fixation of the fibula, and application of an external fixator to allow for soft-tissue healing before definitive fixation of the pilon. Although ultimately the patient required soft-tissue coverage, soft-tissue healing and viability are important in preventing unnecessary soft-tissue procedures, and therefore we would not have handled our initial treatment differently.

Patient selection is also important. The ideal candidate for a cross-leg flap is a young, healthy person who is compliant and has a strong support system to help with activities of daily living. Unfortunately, because of financial issues and lack of home support, our patient remained hospitalized during his treatment course. For a patient who has support, it is possible to be discharged either home or to a rehabilitation facility once flap viability has been confirmed after surgery.

Another consideration is type of immobilization. Immobilization options include casting, use of Kirschner wires (K-wires), and use of rigid external fixation. For cross-leg flaps, external fixation is superior to casting and K-wires, as it provides a more rigid construct and easier access to the flap for serial evaluation. Further, it is easier for the patient to maintain personal hygiene, and it can provide heel rises to avoid pressure ulcers.

Conclusion

To our knowledge, there have been no reports of using a cross-leg flap for wound complications in high-energy pilon fractures. As already mentioned, many of these fractures may be associated with severe soft-tissue injury and may need flap coverage. A cross-leg flap with external fixation of both legs provides a limb salvage option with satisfactory patient outcomes.

Soft-tissue complications are a known problem in the treatment of pilon fractures of the distal end of the tibia. These fractures typically occur as the result of a high-energy mechanism, and axial load and shear forces often lead to a severe soft-tissue injury. In many cases, these injuries may require additional procedures to provide adequate soft-tissue coverage. These procedures can include use of either a rotational muscle flap or a free flap transfer. In some cases, however, these flaps are not possible secondary to vascular compromise.

In this article, we report the case of a pilon fracture combined with severe soft-tissue injury and vascular compromise of the leg. A cross-leg fasciocutaneous flap was performed as a salvage procedure for coverage of the soft-tissue defect. The patient provided written informed consent for print and electronic publication of this case report.

Case Report

A 23-year-old man sustained a left grade III open pilon fracture after a fall off a cherry picker. He was initially treated with irrigation and débridement of the open anteromedial wound, wound closure, application of external fixation, and open reduction and internal fixation (ORIF) of the concomitant comminuted fibular fracture. Operative fixation of the pilon was performed 3 weeks after injury, once skin and soft tissues were in acceptable condition (Figure 1). Skin closure was performed with 2-0 Vicryl sutures (Ethicon, Inc, Somerville, New Jersey) followed by 3-0 nylon skin sutures and No. 2 nylon retention sutures to reduce tension at the incision.

On postoperative day 17, the patient was found to have skin necrosis with exposed hardware over the medial laceration that had resulted from the open fracture (Figure 2). The wound measured 7×6 cm. The plastic surgery team was consulted, and a soft-tissue flap was recommended. Preoperative computed tomography angiogram (Figure 3) revealed 1 vessel runoff in the leg, constituting the peroneal artery, and a conventional angiogram confirmed this finding (Figure 4). Despite these findings, the patient was taken to the operating room 4 weeks after initial injury to try to find a vessel compatible with anastomosis. Intraoperative wound exploration confirmed no patent blood supply for local soft-tissue flap coverage. Therefore, the wound was irrigated and débrided, and a vacuum-assisted closure (VAC) dressing was applied despite exposed hardware and bone. A decision was then made to attempt a cross-leg flap as a salvage procedure, and VAC dressing therapy was continued for several weeks to prepare the recipient site (Figure 5).

Seven weeks after injury, the patient was taken to the operating room by the orthopedic surgery and plastic surgery teams. After débridement, a fasciocutaneous flap was raised from the middle third of the contralateral leg (Figure 6) based on a posterior tibial artery perforator. The flap, which measured 7×7 cm (sufficient to cover the defect), was raised from lateral to medial from the posterior aspect of the leg with the pedicle located on the medial aspect of the right leg. Flap placement was facilitated by flexing the left knee to 80°. The flap was sutured into place with 4-0 Vicryl deep sutures followed by 4-0 nylon and superficial sutures in an interrupted fashion (Figure 7). Rigid external fixation was then applied to both extremities, bridging them together in optimal position (Figure 8). This construct included 2 short bars that would elevate the patient’s heels off the bed to reduce the chance of heel decubiti. Although including the feet in the external fixator construct may help prevent equinus contracture, we splinted the ankles in neutral position immediately after surgery so that we could begin early range-of-motion (ROM) exercises of the ankles to prevent stiffness. Ankle ROM exercises were started once the flap incorporated, 3 weeks after placement of the external fixator. Lacking medical insurance coverage, the patient could not be admitted to a rehabilitation facility or receive home care. He lived independently and had no help at home, so he had to remain hospitalized after placement of the external fixator. While hospitalized, the surgical site was treated with frequent dressing changes, including use of bacitracin and nonadherent dressing.

After flap coverage and 4 weeks of bed rest, a base clamping test confirmed the flap was incorporated into the recipient bed. The patient was then returned to the operating room for removal of the external fixator and skin grafting of the donor site. After surgery, he was started on physical therapy, including exercises for bilateral hip, knee, and ankle ROM and strengthening of the lower extremities. Four months after initial injury, the fracture was healed, based on bone consolidation, seen on radiographs, that is consistent with other pilon fractures treated at our institution. Six months after external fixator removal, the patient was able to ambulate independently with minimal discomfort (Figure 9). Passive and active ankle ROM was 20° of dorsiflexion and 25° of plantarflexion, compared with 25° of dorsiflexion and 45° of plantarflexion on the contralateral extremity. Subtalar motion had some stiffness with a 10° arc, compared with a 25° arc on the contralateral extremity. On simple manual testing, the patient had 5/5 motor strength with dorsiflexion, plantarflexion, inversion, and eversion. He returned to full duty as a landscaper about 1 year after initial injury and had no recurrence of wound complications or infection.

Discussion

Fractures of the distal tibia are commonly known as pilon or plafond fractures. They represent up to 10% of all tibial fractures. The injury consists of an intra-articular fracture of the tibiotalar joint with varying degrees of proximal extension into the tibial metaphysis. The etiology is an axial load on the tibia with or without a rotational force.¹ Treatment is challenging. The literature includes many reports of wound and soft-tissue complications after ORIF. In 1969, Rüedi and Allgöwer² published recommendations that have become the standard for treatment of pilon fractures. Twelve percent of the 84 fractures included in their study were associated with wound complications. In 2004, Sirkin and colleagues³ suggested that wound problems associated with ORIF of pilon fractures may be caused by attempts at immediate fixation through swollen soft tissue. They postulated that staging the procedure and waiting for decreased soft-tissue swelling may reduce the incidence of wound complications. In their series, only 2.9% of closed pilon fractures and only 9.1% of open fractures had any wound complications, and none of their patients required skin grafts, rotation flaps, or free tissue transfers.

However, soft-tissue complications still remain a significant threat in the treatment of pilon fracture, and cases that require additional procedures for soft-tissue coverage are common. In some cases, wound necrosis may lead to below-knee amputation.⁴ There are several coverage options, including local rotational flaps using the soleus muscle^5,6 as well as free flaps using the latissimus dorsi, gracilis, or rectus abdominis muscles.⁷ These options require a sufficient blood supply to the region.

Many high-energy pilon fractures may be associated with vascular injury, and therefore flap survival may be compromised. We have reported such a case in the present article. Our patient’s preoperative angiogram indicated he had 1-vessel runoff to the distal leg—a situation incompatible with free tissue transfer. It is not clear whether this finding is secondary to trauma to the leg or is caused by an anatomical anomaly. Nevertheless, the poor vascularity posed a challenge to providing soft-tissue coverage. Cross-finger⁸ and cross-foot⁹ flaps have been described in upper and lower extremity injuries. In 2006, Zhao and colleagues¹⁰ reported on 5 patients with tibia and/or hardware exposure after operative fixation of tibia fractures. These patients had poor local soft tissue around the wound and therefore underwent cross-leg flap for coverage. It is not clear where the soft-tissue defects were located and whether any studies were performed to assess the local blood flow.

From our patient’s case, we learned that multiple factors should be considered when assessing such high-energy injuries. First, respecting the soft tissues is of paramount importance. Our initial management on presentation consisted of irrigation and débridement of the wound, fixation of the fibula, and application of an external fixator to allow for soft-tissue healing before definitive fixation of the pilon. Although ultimately the patient required soft-tissue coverage, soft-tissue healing and viability are important in preventing unnecessary soft-tissue procedures, and therefore we would not have handled our initial treatment differently.

Patient selection is also important. The ideal candidate for a cross-leg flap is a young, healthy person who is compliant and has a strong support system to help with activities of daily living. Unfortunately, because of financial issues and lack of home support, our patient remained hospitalized during his treatment course. For a patient who has support, it is possible to be discharged either home or to a rehabilitation facility once flap viability has been confirmed after surgery.

Another consideration is type of immobilization. Immobilization options include casting, use of Kirschner wires (K-wires), and use of rigid external fixation. For cross-leg flaps, external fixation is superior to casting and K-wires, as it provides a more rigid construct and easier access to the flap for serial evaluation. Further, it is easier for the patient to maintain personal hygiene, and it can provide heel rises to avoid pressure ulcers.

Conclusion

To our knowledge, there have been no reports of using a cross-leg flap for wound complications in high-energy pilon fractures. As already mentioned, many of these fractures may be associated with severe soft-tissue injury and may need flap coverage. A cross-leg flap with external fixation of both legs provides a limb salvage option with satisfactory patient outcomes.

Soft-tissue complications are a known problem in the treatment of pilon fractures of the distal end of the tibia. These fractures typically occur as the result of a high-energy mechanism, and axial load and shear forces often lead to a severe soft-tissue injury. In many cases, these injuries may require additional procedures to provide adequate soft-tissue coverage. These procedures can include use of either a rotational muscle flap or a free flap transfer. In some cases, however, these flaps are not possible secondary to vascular compromise.

In this article, we report the case of a pilon fracture combined with severe soft-tissue injury and vascular compromise of the leg. A cross-leg fasciocutaneous flap was performed as a salvage procedure for coverage of the soft-tissue defect. The patient provided written informed consent for print and electronic publication of this case report.

Case Report

A 23-year-old man sustained a left grade III open pilon fracture after a fall off a cherry picker. He was initially treated with irrigation and débridement of the open anteromedial wound, wound closure, application of external fixation, and open reduction and internal fixation (ORIF) of the concomitant comminuted fibular fracture. Operative fixation of the pilon was performed 3 weeks after injury, once skin and soft tissues were in acceptable condition (Figure 1). Skin closure was performed with 2-0 Vicryl sutures (Ethicon, Inc, Somerville, New Jersey) followed by 3-0 nylon skin sutures and No. 2 nylon retention sutures to reduce tension at the incision.

On postoperative day 17, the patient was found to have skin necrosis with exposed hardware over the medial laceration that had resulted from the open fracture (Figure 2). The wound measured 7×6 cm. The plastic surgery team was consulted, and a soft-tissue flap was recommended. Preoperative computed tomography angiogram (Figure 3) revealed 1 vessel runoff in the leg, constituting the peroneal artery, and a conventional angiogram confirmed this finding (Figure 4). Despite these findings, the patient was taken to the operating room 4 weeks after initial injury to try to find a vessel compatible with anastomosis. Intraoperative wound exploration confirmed no patent blood supply for local soft-tissue flap coverage. Therefore, the wound was irrigated and débrided, and a vacuum-assisted closure (VAC) dressing was applied despite exposed hardware and bone. A decision was then made to attempt a cross-leg flap as a salvage procedure, and VAC dressing therapy was continued for several weeks to prepare the recipient site (Figure 5).

Seven weeks after injury, the patient was taken to the operating room by the orthopedic surgery and plastic surgery teams. After débridement, a fasciocutaneous flap was raised from the middle third of the contralateral leg (Figure 6) based on a posterior tibial artery perforator. The flap, which measured 7×7 cm (sufficient to cover the defect), was raised from lateral to medial from the posterior aspect of the leg with the pedicle located on the medial aspect of the right leg. Flap placement was facilitated by flexing the left knee to 80°. The flap was sutured into place with 4-0 Vicryl deep sutures followed by 4-0 nylon and superficial sutures in an interrupted fashion (Figure 7). Rigid external fixation was then applied to both extremities, bridging them together in optimal position (Figure 8). This construct included 2 short bars that would elevate the patient’s heels off the bed to reduce the chance of heel decubiti. Although including the feet in the external fixator construct may help prevent equinus contracture, we splinted the ankles in neutral position immediately after surgery so that we could begin early range-of-motion (ROM) exercises of the ankles to prevent stiffness. Ankle ROM exercises were started once the flap incorporated, 3 weeks after placement of the external fixator. Lacking medical insurance coverage, the patient could not be admitted to a rehabilitation facility or receive home care. He lived independently and had no help at home, so he had to remain hospitalized after placement of the external fixator. While hospitalized, the surgical site was treated with frequent dressing changes, including use of bacitracin and nonadherent dressing.

After flap coverage and 4 weeks of bed rest, a base clamping test confirmed the flap was incorporated into the recipient bed. The patient was then returned to the operating room for removal of the external fixator and skin grafting of the donor site. After surgery, he was started on physical therapy, including exercises for bilateral hip, knee, and ankle ROM and strengthening of the lower extremities. Four months after initial injury, the fracture was healed, based on bone consolidation, seen on radiographs, that is consistent with other pilon fractures treated at our institution. Six months after external fixator removal, the patient was able to ambulate independently with minimal discomfort (Figure 9). Passive and active ankle ROM was 20° of dorsiflexion and 25° of plantarflexion, compared with 25° of dorsiflexion and 45° of plantarflexion on the contralateral extremity. Subtalar motion had some stiffness with a 10° arc, compared with a 25° arc on the contralateral extremity. On simple manual testing, the patient had 5/5 motor strength with dorsiflexion, plantarflexion, inversion, and eversion. He returned to full duty as a landscaper about 1 year after initial injury and had no recurrence of wound complications or infection.

Discussion

Fractures of the distal tibia are commonly known as pilon or plafond fractures. They represent up to 10% of all tibial fractures. The injury consists of an intra-articular fracture of the tibiotalar joint with varying degrees of proximal extension into the tibial metaphysis. The etiology is an axial load on the tibia with or without a rotational force.¹ Treatment is challenging. The literature includes many reports of wound and soft-tissue complications after ORIF. In 1969, Rüedi and Allgöwer² published recommendations that have become the standard for treatment of pilon fractures. Twelve percent of the 84 fractures included in their study were associated with wound complications. In 2004, Sirkin and colleagues³ suggested that wound problems associated with ORIF of pilon fractures may be caused by attempts at immediate fixation through swollen soft tissue. They postulated that staging the procedure and waiting for decreased soft-tissue swelling may reduce the incidence of wound complications. In their series, only 2.9% of closed pilon fractures and only 9.1% of open fractures had any wound complications, and none of their patients required skin grafts, rotation flaps, or free tissue transfers.

However, soft-tissue complications still remain a significant threat in the treatment of pilon fracture, and cases that require additional procedures for soft-tissue coverage are common. In some cases, wound necrosis may lead to below-knee amputation.⁴ There are several coverage options, including local rotational flaps using the soleus muscle^5,6 as well as free flaps using the latissimus dorsi, gracilis, or rectus abdominis muscles.⁷ These options require a sufficient blood supply to the region.

Many high-energy pilon fractures may be associated with vascular injury, and therefore flap survival may be compromised. We have reported such a case in the present article. Our patient’s preoperative angiogram indicated he had 1-vessel runoff to the distal leg—a situation incompatible with free tissue transfer. It is not clear whether this finding is secondary to trauma to the leg or is caused by an anatomical anomaly. Nevertheless, the poor vascularity posed a challenge to providing soft-tissue coverage. Cross-finger⁸ and cross-foot⁹ flaps have been described in upper and lower extremity injuries. In 2006, Zhao and colleagues¹⁰ reported on 5 patients with tibia and/or hardware exposure after operative fixation of tibia fractures. These patients had poor local soft tissue around the wound and therefore underwent cross-leg flap for coverage. It is not clear where the soft-tissue defects were located and whether any studies were performed to assess the local blood flow.

From our patient’s case, we learned that multiple factors should be considered when assessing such high-energy injuries. First, respecting the soft tissues is of paramount importance. Our initial management on presentation consisted of irrigation and débridement of the wound, fixation of the fibula, and application of an external fixator to allow for soft-tissue healing before definitive fixation of the pilon. Although ultimately the patient required soft-tissue coverage, soft-tissue healing and viability are important in preventing unnecessary soft-tissue procedures, and therefore we would not have handled our initial treatment differently.

Patient selection is also important. The ideal candidate for a cross-leg flap is a young, healthy person who is compliant and has a strong support system to help with activities of daily living. Unfortunately, because of financial issues and lack of home support, our patient remained hospitalized during his treatment course. For a patient who has support, it is possible to be discharged either home or to a rehabilitation facility once flap viability has been confirmed after surgery.

Another consideration is type of immobilization. Immobilization options include casting, use of Kirschner wires (K-wires), and use of rigid external fixation. For cross-leg flaps, external fixation is superior to casting and K-wires, as it provides a more rigid construct and easier access to the flap for serial evaluation. Further, it is easier for the patient to maintain personal hygiene, and it can provide heel rises to avoid pressure ulcers.

Conclusion

To our knowledge, there have been no reports of using a cross-leg flap for wound complications in high-energy pilon fractures. As already mentioned, many of these fractures may be associated with severe soft-tissue injury and may need flap coverage. A cross-leg flap with external fixation of both legs provides a limb salvage option with satisfactory patient outcomes.

Open treatment of ankle fractures is one of the most common procedures performed by orthopedic surgeons.¹ Among the younger patient population, ankle fractures represent a significant proportion of orthopedic injuries.² The reported incidence of illicit drug and alcohol use in the urban trauma population ranges from 36% to 86%,² and medical and anesthetic complications associated with illicit drug use have been well documented in surgical patients.² However, patients with a recent history of drug abuse may be subject to a separate but related set of complications of open treatment of ankle fractures.

The perioperative complications associated with open treatment of ankle fractures in patients with diabetes mellitus have been well described.^3-6 Similarly, previous studies have suggested that peripheral vascular disease, complicated diabetes, and smoking are risk factors for poor outcomes in patients who require open reduction and internal fixation (ORIF) in lower extremity trauma.^7-9 However, there are few data on the complications specifically associated with illicit drug use and orthopedic surgery. Properly identifying these high-risk groups and being cognizant of commonly associated complications are likely important in ensuring proper perioperative care and may alter follow-up protocols in these patients.

We conducted a study to identify the complications associated with open treatment of ankle fractures in patients who tested positive for illicit drugs on urine drug screen (UDS). We hypothesized that patients who had a history of positive UDS and underwent ORIF of an ankle fracture would have a higher incidence of major and minor complications.

Materials and Methods

After obtaining institutional review board approval, we retrospectively reviewed the cases of 142 patients who underwent open treatment of an ankle fracture between 2006 and 2010. Data sources included patient demographic information, radiographs, preoperative UDS, attending surgeons’ clinical office notes, and clinical laboratory data. Our institution’s standard protocol for ankle fractures was followed for all patients in the study. All patients were evaluated by an orthopedic physician, in either the emergency department or the office, during application of a well-padded Jones splint before surgery. Oral narcotic pain medication was routinely prescribed. All patients were seen, within 10 days of injury, for surgery planning. A board-certified orthopedic surgeon surgically stabilized the ankle fractures. The postoperative treatment regimen, per protocol, included non-weight-bearing in a padded Jones splint dressing; oral narcotic pain medication; physical therapy; and routine scheduled follow-up. In open fracture cases, patients were taken urgently to the operating room for irrigation and débridement with stabilization. Which treatment would be initially used—external fixation or ORIF—was determined on a case-by-case basis.

The sample consisted of adults (age, >18 years) who had undergone definitive ORIF of a lateral malleolar, bimalleolar, or trimalleolar ankle fracture during the study period. Polytrauma patients, patients with external fixation as definitive treatment, and patients with nonoperative treatment were excluded. Before surgical management, all patients were tested for recent illicit drug use by UDS (standard protocol at our institution). UDS, measured for cocaine, marijuana, PCP (phencyclidine), opiates, and barbiturates, was obtained in the office setting or emergency department or on day of surgery. The patients were divided into 2 groups, positive and negative UDS. Patients with documented receipt of narcotic pain medication before UDS were excluded.

The outcomes identified as dependent variables included nonunion, malunion, superficial or deep infection, amputation, delay in treatment, days to healing, repeat surgery, long-term bracing, and loss to follow-up. A nonunion was defined as lasting longer than 9 months and not showing radiographic signs of progression toward healing for 3 consecutive months. These complications were identified with use of attending surgeon clinical progress notes, laboratory values, radiographic parameters, and inpatient readmissions/surgeries associated with these outcomes. Nonunion, malunion, superficial or deep infection, and amputation were then grouped as major complications and analyzed as pooled major complications.

The Fisher exact test was used to analyze categorical variables with respect to UDS. The Wilcoxon rank sum test was used to determine statistical significance for continuous variables. Univariate logistic regression examined both continuous and categorical variables to evaluate predictors for a selected outcome. Statistical significance was set a priori at P ≤ .05, with significant factors indicating an increase (or decrease) in the outcome variable being tested.

Results

We retrospectively reviewed the cases of 142 patients. Table 1 lists the number of cases by fracture type. Bimalleolar fractures were most common, accounting for 99 (69.8%) of the 142 cases. Isolated lateral malleolar fractures accounted for 16 cases (11.2%), and trimalleolar fractures accounted for 27 cases (19%).

Twenty-five (18%) of the 142 patients tested positive for illicit drugs. Mean age was 45.2 years for positive UDS patients and 41.5 years for negative UDS patients. Open fracture cases represented 4.3% of negative UDS patients and 16% of positive UDS patients. Fifty-two percent of positive UDS patients and 32% of negative UDS patients were also tobacco users. These data were statistically significant (P = .003) There were no significant differences in age, sex, incidence of diabetes, incidence of open fracture, or time to surgery between the groups (Table 2).

Incidence of nonunion was higher in positive UDS patients (n = 5; P = .01), as was incidence of deep infection (n = 4; P = .05) (Table 3).

Mean time to radiographic healing was 50.7 days in negative UDS patients and 82.8 days in positive UDS patients (P > .99). Incidence of nonunion was 3.5% in negative UDS patients and 20% in positive UDS patients (P = .01). There were no malunions in negative UDS patients and 2 malunions in positive UDS patients. Incidence of deep infections was 2.5% in negative UDS patients and 16% in positive UDS patients (P = .04). No significant differences were found in incidence of malunions, superficial infections, amputations, need for repeat surgery, continued bracing, or loss to follow-up.

Major complications were defined as superficial or deep infections, amputations, malunions, and nonunions. The rate of major complication was significantly (P = .03) higher in positive UDS patients (24.24%) than in negative UDS patients (7.69%) (Table 4).

Discussion

In the present study, we retrospectively reviewed the cases of patients treated with ORIF for varying types of ankle fractures. Important major and minor complications were analyzed. The overall incidence of major complications in negative UDS patients was only 7.69%, consistent with previously reported results in patients with ankle fractures.^6,10 However, a statistically significant (P = .03) increased incidence of major complications—an alarmingly high rate of almost 1 in 4—was found in positive UDS patients. Our results also demonstrated a significantly higher rate of nonunion and deep infection in positive UDS patients. Calculated odds ratios were 7.37 and 4.27 for nonunion and deep infection, respectively—arguably 2 of the most devastating postoperative complications in positive UDS patients.

Previous studies have found that open fractures, age, and medical comorbidities are significant predictors of short-term complications, such as wound healing, infection, persistent pain, and delayed union.^3-6 Levy and colleagues¹¹ examined the incidence of orthopedic trauma in positive UDS patients. These patients had orthopedic injuries that were more severe and required longer hospitalization. However, the study did not address patients with ankle fractures or the incidence of major complications. Diabetes and peripheral vascular disease are significant risk factors for many surgical procedures in orthopedic surgery.^3,7-9,12,13 Tight glycemic control and optimization of medical comorbidities decrease postoperative complications.^12,13 SooHoo and colleagues⁶ found that history of diabetes and history of peripheral vascular disease were significant predictors of short-term complications of mortality, infection, reoperation, and amputation. The rate of infection in the complicated diabetes group was statistically higher as well. The effect of illicit drug use was not analyzed in that study. We think the findings of the present study highlight the importance of screening for high-risk populations (eg, patients with diabetes, patients with peripheral vascular disease, drug abusers) before orthopedic surgery, especially during definitive treatment of ankle fracture.

Recently, Nåsell and colleagues¹⁰ found that a well-implemented smoking cessation program was associated with a statistically significant reduction in complications 6 and 12 weeks after surgery. The target treatment groups were patients who underwent major lower extremity and upper extremity orthopedic surgery. The most common surgery performed in the study was ORIF of ankle fractures. The authors concluded that a smoking cessation intervention program during the first 6 weeks after acute fracture surgery decreases the risk for postoperative complications. However, no recommendations were made for treating patients with other addictions, such as alcohol and illicit drug addictions.

To our knowledge, our study is the first to critically examine postoperative complications in ankle fracture patients with a history of illicit drug abuse as determined by preoperative positive UDS. These data suggest the importance of critically evaluating this patient population. The rates of deep infection, nonunion, and pooled major complications were all notable. Furthermore, compared with negative UDS patients, positive UDS patients were more than 7 times likely to develop a nonunion and more than 4 times likely to develop a deep infection. The reasons are likely multifactorial but may involve factors such as injury severity, poor nutrition, suboptimal living conditions, difficulty complying with weight-bearing restrictions, and, possibly, poor compliance with wound-care recommendations. Determining the influence of each factor was beyond the scope of this study. However, further investigation is warranted.

The difference in incidence of smoking between the 2 groups was statistically significant. As smoking has been well documented as contributing to poor wound and bone healing,^14-16 it is likely to have been a contributory factor. However, nicotine levels are not routinely part of UDS, and people who quit smoking typically take 7 to 10 days to demonstrate a measurable drop in cotinine levels. On the other hand, screening for drugs takes only a few minutes and can provide useful information during the preoperative period. It was suggested that positive UDS patients were significantly likely to be tobacco users as well.

The 2 groups were not significantly different with respect to mean follow-up time or loss to follow-up. Although mean follow-up was longer in negative UDS patients, the standard deviation was large in both groups. Given the positive UDS patients’ higher incidence of deep infection and nonunion, both of which typically prolong the course of treatment, the results were likely deceptive. Patients with a history of illicit drug use have confounding variables (eg, psychiatric disorders, financial strife) that make treatment compliance and follow-up difficult.¹⁷

Some of the weaknesses of this study are inherent to its retrospective design and limited sample size. Furthermore, patient satisfaction scores and ankle-specific outcome measures, such as AOFAS (American Orthopaedic Foot and Ankle Society) scores, were not considered. Prospective collection of data that include patient satisfaction scores and ankle-specific outcome measures would be optimal. Our current recommendation is to obtain preoperative UDS and illicit drug use history for all trauma patients. In addition, operating surgeons should exercise caution when caring for patients who test positive for illicit drugs.

Conclusion

We evaluated the incidence of complications experienced by positive UDS patients undergoing surgical treatment of ankle fractures. It is well documented that illicit drug users who receive general anesthesia have complications. However, little is known about the untoward effects of illicit drugs on postoperative complications. Furthermore, the efficacy of drug cessation programs in minimizing these complications has not been fully explored.

In conclusion, similar to patients with diabetes, patients with a history of recent illicit drug use, as evidenced by preoperative positive UDS, are at increased risk for complications during treatment for ankle fracture. These data suggest that practicing orthopedists should be more vigilant when caring for ankle fracture patients with preoperative positive UDS.

Open treatment of ankle fractures is one of the most common procedures performed by orthopedic surgeons.¹ Among the younger patient population, ankle fractures represent a significant proportion of orthopedic injuries.² The reported incidence of illicit drug and alcohol use in the urban trauma population ranges from 36% to 86%,² and medical and anesthetic complications associated with illicit drug use have been well documented in surgical patients.² However, patients with a recent history of drug abuse may be subject to a separate but related set of complications of open treatment of ankle fractures.

The perioperative complications associated with open treatment of ankle fractures in patients with diabetes mellitus have been well described.^3-6 Similarly, previous studies have suggested that peripheral vascular disease, complicated diabetes, and smoking are risk factors for poor outcomes in patients who require open reduction and internal fixation (ORIF) in lower extremity trauma.^7-9 However, there are few data on the complications specifically associated with illicit drug use and orthopedic surgery. Properly identifying these high-risk groups and being cognizant of commonly associated complications are likely important in ensuring proper perioperative care and may alter follow-up protocols in these patients.

We conducted a study to identify the complications associated with open treatment of ankle fractures in patients who tested positive for illicit drugs on urine drug screen (UDS). We hypothesized that patients who had a history of positive UDS and underwent ORIF of an ankle fracture would have a higher incidence of major and minor complications.

Materials and Methods

After obtaining institutional review board approval, we retrospectively reviewed the cases of 142 patients who underwent open treatment of an ankle fracture between 2006 and 2010. Data sources included patient demographic information, radiographs, preoperative UDS, attending surgeons’ clinical office notes, and clinical laboratory data. Our institution’s standard protocol for ankle fractures was followed for all patients in the study. All patients were evaluated by an orthopedic physician, in either the emergency department or the office, during application of a well-padded Jones splint before surgery. Oral narcotic pain medication was routinely prescribed. All patients were seen, within 10 days of injury, for surgery planning. A board-certified orthopedic surgeon surgically stabilized the ankle fractures. The postoperative treatment regimen, per protocol, included non-weight-bearing in a padded Jones splint dressing; oral narcotic pain medication; physical therapy; and routine scheduled follow-up. In open fracture cases, patients were taken urgently to the operating room for irrigation and débridement with stabilization. Which treatment would be initially used—external fixation or ORIF—was determined on a case-by-case basis.

The sample consisted of adults (age, >18 years) who had undergone definitive ORIF of a lateral malleolar, bimalleolar, or trimalleolar ankle fracture during the study period. Polytrauma patients, patients with external fixation as definitive treatment, and patients with nonoperative treatment were excluded. Before surgical management, all patients were tested for recent illicit drug use by UDS (standard protocol at our institution). UDS, measured for cocaine, marijuana, PCP (phencyclidine), opiates, and barbiturates, was obtained in the office setting or emergency department or on day of surgery. The patients were divided into 2 groups, positive and negative UDS. Patients with documented receipt of narcotic pain medication before UDS were excluded.

The outcomes identified as dependent variables included nonunion, malunion, superficial or deep infection, amputation, delay in treatment, days to healing, repeat surgery, long-term bracing, and loss to follow-up. A nonunion was defined as lasting longer than 9 months and not showing radiographic signs of progression toward healing for 3 consecutive months. These complications were identified with use of attending surgeon clinical progress notes, laboratory values, radiographic parameters, and inpatient readmissions/surgeries associated with these outcomes. Nonunion, malunion, superficial or deep infection, and amputation were then grouped as major complications and analyzed as pooled major complications.

The Fisher exact test was used to analyze categorical variables with respect to UDS. The Wilcoxon rank sum test was used to determine statistical significance for continuous variables. Univariate logistic regression examined both continuous and categorical variables to evaluate predictors for a selected outcome. Statistical significance was set a priori at P ≤ .05, with significant factors indicating an increase (or decrease) in the outcome variable being tested.

Results

We retrospectively reviewed the cases of 142 patients. Table 1 lists the number of cases by fracture type. Bimalleolar fractures were most common, accounting for 99 (69.8%) of the 142 cases. Isolated lateral malleolar fractures accounted for 16 cases (11.2%), and trimalleolar fractures accounted for 27 cases (19%).

Twenty-five (18%) of the 142 patients tested positive for illicit drugs. Mean age was 45.2 years for positive UDS patients and 41.5 years for negative UDS patients. Open fracture cases represented 4.3% of negative UDS patients and 16% of positive UDS patients. Fifty-two percent of positive UDS patients and 32% of negative UDS patients were also tobacco users. These data were statistically significant (P = .003) There were no significant differences in age, sex, incidence of diabetes, incidence of open fracture, or time to surgery between the groups (Table 2).

Incidence of nonunion was higher in positive UDS patients (n = 5; P = .01), as was incidence of deep infection (n = 4; P = .05) (Table 3).

Mean time to radiographic healing was 50.7 days in negative UDS patients and 82.8 days in positive UDS patients (P > .99). Incidence of nonunion was 3.5% in negative UDS patients and 20% in positive UDS patients (P = .01). There were no malunions in negative UDS patients and 2 malunions in positive UDS patients. Incidence of deep infections was 2.5% in negative UDS patients and 16% in positive UDS patients (P = .04). No significant differences were found in incidence of malunions, superficial infections, amputations, need for repeat surgery, continued bracing, or loss to follow-up.

Major complications were defined as superficial or deep infections, amputations, malunions, and nonunions. The rate of major complication was significantly (P = .03) higher in positive UDS patients (24.24%) than in negative UDS patients (7.69%) (Table 4).

Discussion

In the present study, we retrospectively reviewed the cases of patients treated with ORIF for varying types of ankle fractures. Important major and minor complications were analyzed. The overall incidence of major complications in negative UDS patients was only 7.69%, consistent with previously reported results in patients with ankle fractures.^6,10 However, a statistically significant (P = .03) increased incidence of major complications—an alarmingly high rate of almost 1 in 4—was found in positive UDS patients. Our results also demonstrated a significantly higher rate of nonunion and deep infection in positive UDS patients. Calculated odds ratios were 7.37 and 4.27 for nonunion and deep infection, respectively—arguably 2 of the most devastating postoperative complications in positive UDS patients.

Previous studies have found that open fractures, age, and medical comorbidities are significant predictors of short-term complications, such as wound healing, infection, persistent pain, and delayed union.^3-6 Levy and colleagues¹¹ examined the incidence of orthopedic trauma in positive UDS patients. These patients had orthopedic injuries that were more severe and required longer hospitalization. However, the study did not address patients with ankle fractures or the incidence of major complications. Diabetes and peripheral vascular disease are significant risk factors for many surgical procedures in orthopedic surgery.^3,7-9,12,13 Tight glycemic control and optimization of medical comorbidities decrease postoperative complications.^12,13 SooHoo and colleagues⁶ found that history of diabetes and history of peripheral vascular disease were significant predictors of short-term complications of mortality, infection, reoperation, and amputation. The rate of infection in the complicated diabetes group was statistically higher as well. The effect of illicit drug use was not analyzed in that study. We think the findings of the present study highlight the importance of screening for high-risk populations (eg, patients with diabetes, patients with peripheral vascular disease, drug abusers) before orthopedic surgery, especially during definitive treatment of ankle fracture.

Recently, Nåsell and colleagues¹⁰ found that a well-implemented smoking cessation program was associated with a statistically significant reduction in complications 6 and 12 weeks after surgery. The target treatment groups were patients who underwent major lower extremity and upper extremity orthopedic surgery. The most common surgery performed in the study was ORIF of ankle fractures. The authors concluded that a smoking cessation intervention program during the first 6 weeks after acute fracture surgery decreases the risk for postoperative complications. However, no recommendations were made for treating patients with other addictions, such as alcohol and illicit drug addictions.

To our knowledge, our study is the first to critically examine postoperative complications in ankle fracture patients with a history of illicit drug abuse as determined by preoperative positive UDS. These data suggest the importance of critically evaluating this patient population. The rates of deep infection, nonunion, and pooled major complications were all notable. Furthermore, compared with negative UDS patients, positive UDS patients were more than 7 times likely to develop a nonunion and more than 4 times likely to develop a deep infection. The reasons are likely multifactorial but may involve factors such as injury severity, poor nutrition, suboptimal living conditions, difficulty complying with weight-bearing restrictions, and, possibly, poor compliance with wound-care recommendations. Determining the influence of each factor was beyond the scope of this study. However, further investigation is warranted.

The difference in incidence of smoking between the 2 groups was statistically significant. As smoking has been well documented as contributing to poor wound and bone healing,^14-16 it is likely to have been a contributory factor. However, nicotine levels are not routinely part of UDS, and people who quit smoking typically take 7 to 10 days to demonstrate a measurable drop in cotinine levels. On the other hand, screening for drugs takes only a few minutes and can provide useful information during the preoperative period. It was suggested that positive UDS patients were significantly likely to be tobacco users as well.

The 2 groups were not significantly different with respect to mean follow-up time or loss to follow-up. Although mean follow-up was longer in negative UDS patients, the standard deviation was large in both groups. Given the positive UDS patients’ higher incidence of deep infection and nonunion, both of which typically prolong the course of treatment, the results were likely deceptive. Patients with a history of illicit drug use have confounding variables (eg, psychiatric disorders, financial strife) that make treatment compliance and follow-up difficult.¹⁷

Some of the weaknesses of this study are inherent to its retrospective design and limited sample size. Furthermore, patient satisfaction scores and ankle-specific outcome measures, such as AOFAS (American Orthopaedic Foot and Ankle Society) scores, were not considered. Prospective collection of data that include patient satisfaction scores and ankle-specific outcome measures would be optimal. Our current recommendation is to obtain preoperative UDS and illicit drug use history for all trauma patients. In addition, operating surgeons should exercise caution when caring for patients who test positive for illicit drugs.

Conclusion

We evaluated the incidence of complications experienced by positive UDS patients undergoing surgical treatment of ankle fractures. It is well documented that illicit drug users who receive general anesthesia have complications. However, little is known about the untoward effects of illicit drugs on postoperative complications. Furthermore, the efficacy of drug cessation programs in minimizing these complications has not been fully explored.

In conclusion, similar to patients with diabetes, patients with a history of recent illicit drug use, as evidenced by preoperative positive UDS, are at increased risk for complications during treatment for ankle fracture. These data suggest that practicing orthopedists should be more vigilant when caring for ankle fracture patients with preoperative positive UDS.

Open treatment of ankle fractures is one of the most common procedures performed by orthopedic surgeons.¹ Among the younger patient population, ankle fractures represent a significant proportion of orthopedic injuries.² The reported incidence of illicit drug and alcohol use in the urban trauma population ranges from 36% to 86%,² and medical and anesthetic complications associated with illicit drug use have been well documented in surgical patients.² However, patients with a recent history of drug abuse may be subject to a separate but related set of complications of open treatment of ankle fractures.

The perioperative complications associated with open treatment of ankle fractures in patients with diabetes mellitus have been well described.^3-6 Similarly, previous studies have suggested that peripheral vascular disease, complicated diabetes, and smoking are risk factors for poor outcomes in patients who require open reduction and internal fixation (ORIF) in lower extremity trauma.^7-9 However, there are few data on the complications specifically associated with illicit drug use and orthopedic surgery. Properly identifying these high-risk groups and being cognizant of commonly associated complications are likely important in ensuring proper perioperative care and may alter follow-up protocols in these patients.

We conducted a study to identify the complications associated with open treatment of ankle fractures in patients who tested positive for illicit drugs on urine drug screen (UDS). We hypothesized that patients who had a history of positive UDS and underwent ORIF of an ankle fracture would have a higher incidence of major and minor complications.

Materials and Methods

After obtaining institutional review board approval, we retrospectively reviewed the cases of 142 patients who underwent open treatment of an ankle fracture between 2006 and 2010. Data sources included patient demographic information, radiographs, preoperative UDS, attending surgeons’ clinical office notes, and clinical laboratory data. Our institution’s standard protocol for ankle fractures was followed for all patients in the study. All patients were evaluated by an orthopedic physician, in either the emergency department or the office, during application of a well-padded Jones splint before surgery. Oral narcotic pain medication was routinely prescribed. All patients were seen, within 10 days of injury, for surgery planning. A board-certified orthopedic surgeon surgically stabilized the ankle fractures. The postoperative treatment regimen, per protocol, included non-weight-bearing in a padded Jones splint dressing; oral narcotic pain medication; physical therapy; and routine scheduled follow-up. In open fracture cases, patients were taken urgently to the operating room for irrigation and débridement with stabilization. Which treatment would be initially used—external fixation or ORIF—was determined on a case-by-case basis.

The sample consisted of adults (age, >18 years) who had undergone definitive ORIF of a lateral malleolar, bimalleolar, or trimalleolar ankle fracture during the study period. Polytrauma patients, patients with external fixation as definitive treatment, and patients with nonoperative treatment were excluded. Before surgical management, all patients were tested for recent illicit drug use by UDS (standard protocol at our institution). UDS, measured for cocaine, marijuana, PCP (phencyclidine), opiates, and barbiturates, was obtained in the office setting or emergency department or on day of surgery. The patients were divided into 2 groups, positive and negative UDS. Patients with documented receipt of narcotic pain medication before UDS were excluded.

The outcomes identified as dependent variables included nonunion, malunion, superficial or deep infection, amputation, delay in treatment, days to healing, repeat surgery, long-term bracing, and loss to follow-up. A nonunion was defined as lasting longer than 9 months and not showing radiographic signs of progression toward healing for 3 consecutive months. These complications were identified with use of attending surgeon clinical progress notes, laboratory values, radiographic parameters, and inpatient readmissions/surgeries associated with these outcomes. Nonunion, malunion, superficial or deep infection, and amputation were then grouped as major complications and analyzed as pooled major complications.

The Fisher exact test was used to analyze categorical variables with respect to UDS. The Wilcoxon rank sum test was used to determine statistical significance for continuous variables. Univariate logistic regression examined both continuous and categorical variables to evaluate predictors for a selected outcome. Statistical significance was set a priori at P ≤ .05, with significant factors indicating an increase (or decrease) in the outcome variable being tested.

Results

We retrospectively reviewed the cases of 142 patients. Table 1 lists the number of cases by fracture type. Bimalleolar fractures were most common, accounting for 99 (69.8%) of the 142 cases. Isolated lateral malleolar fractures accounted for 16 cases (11.2%), and trimalleolar fractures accounted for 27 cases (19%).

Twenty-five (18%) of the 142 patients tested positive for illicit drugs. Mean age was 45.2 years for positive UDS patients and 41.5 years for negative UDS patients. Open fracture cases represented 4.3% of negative UDS patients and 16% of positive UDS patients. Fifty-two percent of positive UDS patients and 32% of negative UDS patients were also tobacco users. These data were statistically significant (P = .003) There were no significant differences in age, sex, incidence of diabetes, incidence of open fracture, or time to surgery between the groups (Table 2).

Incidence of nonunion was higher in positive UDS patients (n = 5; P = .01), as was incidence of deep infection (n = 4; P = .05) (Table 3).

Mean time to radiographic healing was 50.7 days in negative UDS patients and 82.8 days in positive UDS patients (P > .99). Incidence of nonunion was 3.5% in negative UDS patients and 20% in positive UDS patients (P = .01). There were no malunions in negative UDS patients and 2 malunions in positive UDS patients. Incidence of deep infections was 2.5% in negative UDS patients and 16% in positive UDS patients (P = .04). No significant differences were found in incidence of malunions, superficial infections, amputations, need for repeat surgery, continued bracing, or loss to follow-up.

Major complications were defined as superficial or deep infections, amputations, malunions, and nonunions. The rate of major complication was significantly (P = .03) higher in positive UDS patients (24.24%) than in negative UDS patients (7.69%) (Table 4).

Discussion

In the present study, we retrospectively reviewed the cases of patients treated with ORIF for varying types of ankle fractures. Important major and minor complications were analyzed. The overall incidence of major complications in negative UDS patients was only 7.69%, consistent with previously reported results in patients with ankle fractures.^6,10 However, a statistically significant (P = .03) increased incidence of major complications—an alarmingly high rate of almost 1 in 4—was found in positive UDS patients. Our results also demonstrated a significantly higher rate of nonunion and deep infection in positive UDS patients. Calculated odds ratios were 7.37 and 4.27 for nonunion and deep infection, respectively—arguably 2 of the most devastating postoperative complications in positive UDS patients.

Previous studies have found that open fractures, age, and medical comorbidities are significant predictors of short-term complications, such as wound healing, infection, persistent pain, and delayed union.^3-6 Levy and colleagues¹¹ examined the incidence of orthopedic trauma in positive UDS patients. These patients had orthopedic injuries that were more severe and required longer hospitalization. However, the study did not address patients with ankle fractures or the incidence of major complications. Diabetes and peripheral vascular disease are significant risk factors for many surgical procedures in orthopedic surgery.^3,7-9,12,13 Tight glycemic control and optimization of medical comorbidities decrease postoperative complications.^12,13 SooHoo and colleagues⁶ found that history of diabetes and history of peripheral vascular disease were significant predictors of short-term complications of mortality, infection, reoperation, and amputation. The rate of infection in the complicated diabetes group was statistically higher as well. The effect of illicit drug use was not analyzed in that study. We think the findings of the present study highlight the importance of screening for high-risk populations (eg, patients with diabetes, patients with peripheral vascular disease, drug abusers) before orthopedic surgery, especially during definitive treatment of ankle fracture.

Recently, Nåsell and colleagues¹⁰ found that a well-implemented smoking cessation program was associated with a statistically significant reduction in complications 6 and 12 weeks after surgery. The target treatment groups were patients who underwent major lower extremity and upper extremity orthopedic surgery. The most common surgery performed in the study was ORIF of ankle fractures. The authors concluded that a smoking cessation intervention program during the first 6 weeks after acute fracture surgery decreases the risk for postoperative complications. However, no recommendations were made for treating patients with other addictions, such as alcohol and illicit drug addictions.

To our knowledge, our study is the first to critically examine postoperative complications in ankle fracture patients with a history of illicit drug abuse as determined by preoperative positive UDS. These data suggest the importance of critically evaluating this patient population. The rates of deep infection, nonunion, and pooled major complications were all notable. Furthermore, compared with negative UDS patients, positive UDS patients were more than 7 times likely to develop a nonunion and more than 4 times likely to develop a deep infection. The reasons are likely multifactorial but may involve factors such as injury severity, poor nutrition, suboptimal living conditions, difficulty complying with weight-bearing restrictions, and, possibly, poor compliance with wound-care recommendations. Determining the influence of each factor was beyond the scope of this study. However, further investigation is warranted.

The difference in incidence of smoking between the 2 groups was statistically significant. As smoking has been well documented as contributing to poor wound and bone healing,^14-16 it is likely to have been a contributory factor. However, nicotine levels are not routinely part of UDS, and people who quit smoking typically take 7 to 10 days to demonstrate a measurable drop in cotinine levels. On the other hand, screening for drugs takes only a few minutes and can provide useful information during the preoperative period. It was suggested that positive UDS patients were significantly likely to be tobacco users as well.

The 2 groups were not significantly different with respect to mean follow-up time or loss to follow-up. Although mean follow-up was longer in negative UDS patients, the standard deviation was large in both groups. Given the positive UDS patients’ higher incidence of deep infection and nonunion, both of which typically prolong the course of treatment, the results were likely deceptive. Patients with a history of illicit drug use have confounding variables (eg, psychiatric disorders, financial strife) that make treatment compliance and follow-up difficult.¹⁷

Some of the weaknesses of this study are inherent to its retrospective design and limited sample size. Furthermore, patient satisfaction scores and ankle-specific outcome measures, such as AOFAS (American Orthopaedic Foot and Ankle Society) scores, were not considered. Prospective collection of data that include patient satisfaction scores and ankle-specific outcome measures would be optimal. Our current recommendation is to obtain preoperative UDS and illicit drug use history for all trauma patients. In addition, operating surgeons should exercise caution when caring for patients who test positive for illicit drugs.

Conclusion

We evaluated the incidence of complications experienced by positive UDS patients undergoing surgical treatment of ankle fractures. It is well documented that illicit drug users who receive general anesthesia have complications. However, little is known about the untoward effects of illicit drugs on postoperative complications. Furthermore, the efficacy of drug cessation programs in minimizing these complications has not been fully explored.

In conclusion, similar to patients with diabetes, patients with a history of recent illicit drug use, as evidenced by preoperative positive UDS, are at increased risk for complications during treatment for ankle fracture. These data suggest that practicing orthopedists should be more vigilant when caring for ankle fracture patients with preoperative positive UDS.