The Journal of Family Practice

For well over a century, the periodic health exam has been associated with the delivery of preventive services¹—a model widely accepted by physicians and patients alike. Approximately 8% of ambulatory care visits are for check-ups, and more than 20% of US residents schedule a health exam annually.²

Periodic health exams, however, do not result in optimal preventive care. Evidence suggests that important preventive services, such as dietary counseling, occur at only 8% of such visits; that just 10 seconds, on average, is devoted to smoking cessation; and that 80% of the preventive services patients receive are delivered outside of scheduled health exams.² Because physicians and patients alike understandably prioritize acute problems, any discussion of health maintenance issues during a periodic check-up is likely to occur despite the visit’s agenda, not as part of it.^3-7

Primary care physicians and practices are increasingly being held accountable for their performance on preventive measures. With that in mind, being familiar with evidence-based guidelines relating to the delivery of preventive services in ambulatory care settings, such as those developed by the US Preventive Services Task Force (USPSTF),⁸ is crucial. Identifying elements of the chronic care model that can be effectively applied to preventive care—and recognizing that the reactive acute care model is ineffective for both chronic illness and preventive services—is essential, as well.

Preventive care that's evidence-based

Good practice guidelines should have the following features, according to the Institute of Medicine:⁹
• Validity. Application would lead to the desired health and cost outcomes.
• Reliability/reproducibility. Others using the same data/interpretation would reach the same conclusion.
• Clinical applicability. Patient populations are appropriately defined.
• Clinical flexibility. Known or generally expected exceptions are identified.
• Clarity. Guideline uses unambiguous language with precise definitions.
• Multidisciplinary process. Developed with the participation of all key stakeholders.
• Scheduled review. Periodically reviewed and revised to incorporate new evidence/changing consensus.
• Documentation. Procedures used in development are well documented.

It’s estimated that a primary care physician with an average-sized patient panel would need an additional 7.4 hours per day to achieve 100% compliance with all of the USPSTF recommendations.

The USPSTF guidelines are consistent with these attributes.⁸

Some brief interventions fail

Guidelines are useful as practice standards for establishing preventive care goals, but their existence alone does not ensure the delivery of high-quality preventive services in an office setting. One key factor is time. It is estimated that a primary care physician with an average-sized patient panel would require an additional 7.4 hours per day to achieve 100% compliance with all of the USPSTF recommendations.¹⁰

Counseling, in particular, is an intervention that may be more effective in theory than in practice. Evidence suggests that even when primary care providers are trained in preventive counseling (and many are not), many brief interventions are not effective at creating sustained behavior change or health improvement.¹¹

Others are effective

That’s not to say that there’s little that can be done. Use of standing orders for influenza vaccination is one example of an effective and easily implemented preventive measure that requires little or no additional physician time. Yet some doctors are resistant, fearing loss of control or lawsuits. In fact, the National Vaccine Injury Compensation Program specifically provides protection against vaccinerelated malpractice claims.¹² A standing order for office staff to arrange a screening mammogram is another effective intervention; it has been shown to improve screening rates by as much as 30%.¹³

Lessons from chronic illness management

Chronic illness care and preventive care have much in common.¹⁴ Both acknowledge the need for proactive screening and counseling to bring about behavior change. In addition, both require ongoing care and follow-up, as well as depend on links to community resources. And finally, both are resource-intensive—too resource-intensive, some say, to be delivered in a cost-effective manner.

The Chronic Care Model has been proposed as a framework for improving preventive services (TABLE).^15,16 Evidence suggests adopting some key components of chronic care can lead to significant gains in the delivery of preventive care, with the largest improvement seen when multiple components are used simultaneously.^17-20

Self-management support. A growing body of evidence shows that self-management activities are associated with improved outcomes.²¹ Instituting a peer support group for pregnant women at a federally qualified health center, for example, led to a 7% absolute risk reduction in preterm births.²²

Not all efforts to encourage self-management are effective, however, and evaluation is crucial to determine what works. In one large-scale trial, the use of patient reminder cards to facilitate a reduction in health risk behaviors such as tobacco use, risky drinking, unhealthy dietary patterns, and physical inactivity led to fewer health risk assessments being performed, fewer individual counseling encounters, and no change in these behaviors.²³

Decision support. Clinical decision support systems, which generate patient-specific evidence-based assessments and/or recommendations that are actionable as part of the workflow at the point of care, have been found to improve care.^24,25 An example of this is a prompt that reminds the physician to discuss chemoprevention with a patient at high risk for breast cancer.

Delivery system design. The patient-centered medical home (PCMH) is a well-known example of a redesign of health care delivery.²⁶ Conversion to this model is associated with a small positive effect on preventive interventions.²⁷ However, the persistence of a fee-for-service payment system—which does not include physician reimbursement for some of the added services incorporated into the PCMH—limits the implementation of the PCMH model.²⁸

Many practices are improving health care delivery by using nonphysicians for various tasks related to preventive care. Care managers, for example, typically have smaller caseloads and focus on reducing unnecessary treatment for patients with high-risk conditions, such as congestive heart failure, while patient navigators generally have less clinical expertise but more knowledge of community services.

In one study, practice-initiated phone conversations with nonphysicians increased colorectal cancer screening by up to 40%.²⁹ And in one pilot program, the use of ancillary providers led to an increase in colorectal cancer screening by as much as 123%.³⁰ The human touch seems to be key to the success of these interventions. Passive reminders, such as videos being shown on waiting room televisions, have not proven to be effective.³¹ 

Clinical information systems. Early on, the power of electronic health records (EHRs) to improve practices’ delivery of preventive services was recognized. As early as 1995, the use of a reminder system to highlight such services during acute care visits was linked to improvements in counseling about smoking cessation and higher rates of cervical cancer screening, among other preventive measures.^32,33 Overall, the use of EHRs alone has been shown to improve rates of preventive services by as much as 66%, with most practices reporting improvements of at least 20%.³⁴

Use of standing orders for influenza vaccination is one example of an effective and easily implemented preventive measure that requires little or no additional physician time.

Today, EHRs that are Meaningful Use Stage 2-compliant have the tools needed to improve care. Requirements include the ability to generate patient registries of all those with a given disease and to identify patients on the registry who have not received needed care.³⁵ To improve preventive care, registries should focus on the mitigation of risk factors, such as identifying—and contacting—patients with diabetes who are in need of, or overdue for, an annual eye exam.

Trials using the registry function, in combination with automated messaging to deliver targeted information to various patient groups (identified by the demographic information available from the EHR), are ongoing.

Community resources. Many clinicians have informal referral relationships with community organizations, such as the YMCA. Physician practices that establish links to community resources have the potential to have a large effect on unhealthy behaviors. (See “Putting theory into practice: 2 cases”.) However, a systematic review found that, while evidence to support such connections is mainly positive, research is limited and further evaluation is needed.³⁶

The Practical Playbook (practicalplaybook.org)³⁷ developed by the deBeaumont Foundation, Duke Department of Community and Family Medicine, and the Centers for Disease Control and Prevention, offers concrete examples of how physician practices are linking with community resources to improve the health of the population. For example, Duke University’s “Just for Us” program provides in-home chronic illness care to 350 high-risk elderly individuals. The LATCH program connects thousands of Latino immigrants to health care services and culturally and linguistically appropriate health education classes.³⁷

Growing emphasis on quality

Systemic changes in the US health care system are occurring rapidly, with an emphasis on quality and improved outcomes. Many physicians are now required to submit data to external agencies for payment, and much of the data is grounded in preventive standards. Medicare’s Physicians Quality Reporting System requires that all Medicare providers provide data on preventive and chronic illness care. Rates of vaccination, obesity screening, and tobacco use screening are examples of preventive services that will be reported publicly on the Centers for Medicare & Medicaid Services’ Physician Compare Web site.³⁸

Physicians who work in accountable care organizations are required to meet quality standards on the delivery of certain preventive services, including breast cancer screening, colorectal cancer screening, influenza and pneumonia immunization, body mass index screening and follow-up, tobacco use screening and cessation intervention, screening for high blood pressure and followup, and screening for clinical depression and follow-up.³⁹ As patients discover that the Affordable Care Act mandates that preventive services be covered with no cost sharing, they are likely to become more receptive to physician attempts to provide them.^40,41

CORRESPONDENCE
Gerald Liu, MD, 1504 Springhill Avenue, Suite 3414, Mobile, AL 36604-3207; gliu@health.southalabama.edu

For well over a century, the periodic health exam has been associated with the delivery of preventive services¹—a model widely accepted by physicians and patients alike. Approximately 8% of ambulatory care visits are for check-ups, and more than 20% of US residents schedule a health exam annually.²

Periodic health exams, however, do not result in optimal preventive care. Evidence suggests that important preventive services, such as dietary counseling, occur at only 8% of such visits; that just 10 seconds, on average, is devoted to smoking cessation; and that 80% of the preventive services patients receive are delivered outside of scheduled health exams.² Because physicians and patients alike understandably prioritize acute problems, any discussion of health maintenance issues during a periodic check-up is likely to occur despite the visit’s agenda, not as part of it.^3-7

Primary care physicians and practices are increasingly being held accountable for their performance on preventive measures. With that in mind, being familiar with evidence-based guidelines relating to the delivery of preventive services in ambulatory care settings, such as those developed by the US Preventive Services Task Force (USPSTF),⁸ is crucial. Identifying elements of the chronic care model that can be effectively applied to preventive care—and recognizing that the reactive acute care model is ineffective for both chronic illness and preventive services—is essential, as well.

Preventive care that's evidence-based

Good practice guidelines should have the following features, according to the Institute of Medicine:⁹
• Validity. Application would lead to the desired health and cost outcomes.
• Reliability/reproducibility. Others using the same data/interpretation would reach the same conclusion.
• Clinical applicability. Patient populations are appropriately defined.
• Clinical flexibility. Known or generally expected exceptions are identified.
• Clarity. Guideline uses unambiguous language with precise definitions.
• Multidisciplinary process. Developed with the participation of all key stakeholders.
• Scheduled review. Periodically reviewed and revised to incorporate new evidence/changing consensus.
• Documentation. Procedures used in development are well documented.

It’s estimated that a primary care physician with an average-sized patient panel would need an additional 7.4 hours per day to achieve 100% compliance with all of the USPSTF recommendations.

The USPSTF guidelines are consistent with these attributes.⁸

Some brief interventions fail

Guidelines are useful as practice standards for establishing preventive care goals, but their existence alone does not ensure the delivery of high-quality preventive services in an office setting. One key factor is time. It is estimated that a primary care physician with an average-sized patient panel would require an additional 7.4 hours per day to achieve 100% compliance with all of the USPSTF recommendations.¹⁰

Counseling, in particular, is an intervention that may be more effective in theory than in practice. Evidence suggests that even when primary care providers are trained in preventive counseling (and many are not), many brief interventions are not effective at creating sustained behavior change or health improvement.¹¹

Others are effective

That’s not to say that there’s little that can be done. Use of standing orders for influenza vaccination is one example of an effective and easily implemented preventive measure that requires little or no additional physician time. Yet some doctors are resistant, fearing loss of control or lawsuits. In fact, the National Vaccine Injury Compensation Program specifically provides protection against vaccinerelated malpractice claims.¹² A standing order for office staff to arrange a screening mammogram is another effective intervention; it has been shown to improve screening rates by as much as 30%.¹³

Lessons from chronic illness management

Chronic illness care and preventive care have much in common.¹⁴ Both acknowledge the need for proactive screening and counseling to bring about behavior change. In addition, both require ongoing care and follow-up, as well as depend on links to community resources. And finally, both are resource-intensive—too resource-intensive, some say, to be delivered in a cost-effective manner.

The Chronic Care Model has been proposed as a framework for improving preventive services (TABLE).^15,16 Evidence suggests adopting some key components of chronic care can lead to significant gains in the delivery of preventive care, with the largest improvement seen when multiple components are used simultaneously.^17-20

Self-management support. A growing body of evidence shows that self-management activities are associated with improved outcomes.²¹ Instituting a peer support group for pregnant women at a federally qualified health center, for example, led to a 7% absolute risk reduction in preterm births.²²

Not all efforts to encourage self-management are effective, however, and evaluation is crucial to determine what works. In one large-scale trial, the use of patient reminder cards to facilitate a reduction in health risk behaviors such as tobacco use, risky drinking, unhealthy dietary patterns, and physical inactivity led to fewer health risk assessments being performed, fewer individual counseling encounters, and no change in these behaviors.²³

Decision support. Clinical decision support systems, which generate patient-specific evidence-based assessments and/or recommendations that are actionable as part of the workflow at the point of care, have been found to improve care.^24,25 An example of this is a prompt that reminds the physician to discuss chemoprevention with a patient at high risk for breast cancer.

Delivery system design. The patient-centered medical home (PCMH) is a well-known example of a redesign of health care delivery.²⁶ Conversion to this model is associated with a small positive effect on preventive interventions.²⁷ However, the persistence of a fee-for-service payment system—which does not include physician reimbursement for some of the added services incorporated into the PCMH—limits the implementation of the PCMH model.²⁸

Many practices are improving health care delivery by using nonphysicians for various tasks related to preventive care. Care managers, for example, typically have smaller caseloads and focus on reducing unnecessary treatment for patients with high-risk conditions, such as congestive heart failure, while patient navigators generally have less clinical expertise but more knowledge of community services.

In one study, practice-initiated phone conversations with nonphysicians increased colorectal cancer screening by up to 40%.²⁹ And in one pilot program, the use of ancillary providers led to an increase in colorectal cancer screening by as much as 123%.³⁰ The human touch seems to be key to the success of these interventions. Passive reminders, such as videos being shown on waiting room televisions, have not proven to be effective.³¹ 

Clinical information systems. Early on, the power of electronic health records (EHRs) to improve practices’ delivery of preventive services was recognized. As early as 1995, the use of a reminder system to highlight such services during acute care visits was linked to improvements in counseling about smoking cessation and higher rates of cervical cancer screening, among other preventive measures.^32,33 Overall, the use of EHRs alone has been shown to improve rates of preventive services by as much as 66%, with most practices reporting improvements of at least 20%.³⁴

Use of standing orders for influenza vaccination is one example of an effective and easily implemented preventive measure that requires little or no additional physician time.

Today, EHRs that are Meaningful Use Stage 2-compliant have the tools needed to improve care. Requirements include the ability to generate patient registries of all those with a given disease and to identify patients on the registry who have not received needed care.³⁵ To improve preventive care, registries should focus on the mitigation of risk factors, such as identifying—and contacting—patients with diabetes who are in need of, or overdue for, an annual eye exam.

Trials using the registry function, in combination with automated messaging to deliver targeted information to various patient groups (identified by the demographic information available from the EHR), are ongoing.

Community resources. Many clinicians have informal referral relationships with community organizations, such as the YMCA. Physician practices that establish links to community resources have the potential to have a large effect on unhealthy behaviors. (See “Putting theory into practice: 2 cases”.) However, a systematic review found that, while evidence to support such connections is mainly positive, research is limited and further evaluation is needed.³⁶

The Practical Playbook (practicalplaybook.org)³⁷ developed by the deBeaumont Foundation, Duke Department of Community and Family Medicine, and the Centers for Disease Control and Prevention, offers concrete examples of how physician practices are linking with community resources to improve the health of the population. For example, Duke University’s “Just for Us” program provides in-home chronic illness care to 350 high-risk elderly individuals. The LATCH program connects thousands of Latino immigrants to health care services and culturally and linguistically appropriate health education classes.³⁷

Growing emphasis on quality

Systemic changes in the US health care system are occurring rapidly, with an emphasis on quality and improved outcomes. Many physicians are now required to submit data to external agencies for payment, and much of the data is grounded in preventive standards. Medicare’s Physicians Quality Reporting System requires that all Medicare providers provide data on preventive and chronic illness care. Rates of vaccination, obesity screening, and tobacco use screening are examples of preventive services that will be reported publicly on the Centers for Medicare & Medicaid Services’ Physician Compare Web site.³⁸

Physicians who work in accountable care organizations are required to meet quality standards on the delivery of certain preventive services, including breast cancer screening, colorectal cancer screening, influenza and pneumonia immunization, body mass index screening and follow-up, tobacco use screening and cessation intervention, screening for high blood pressure and followup, and screening for clinical depression and follow-up.³⁹ As patients discover that the Affordable Care Act mandates that preventive services be covered with no cost sharing, they are likely to become more receptive to physician attempts to provide them.^40,41

CORRESPONDENCE
Gerald Liu, MD, 1504 Springhill Avenue, Suite 3414, Mobile, AL 36604-3207; gliu@health.southalabama.edu

For well over a century, the periodic health exam has been associated with the delivery of preventive services¹—a model widely accepted by physicians and patients alike. Approximately 8% of ambulatory care visits are for check-ups, and more than 20% of US residents schedule a health exam annually.²

Periodic health exams, however, do not result in optimal preventive care. Evidence suggests that important preventive services, such as dietary counseling, occur at only 8% of such visits; that just 10 seconds, on average, is devoted to smoking cessation; and that 80% of the preventive services patients receive are delivered outside of scheduled health exams.² Because physicians and patients alike understandably prioritize acute problems, any discussion of health maintenance issues during a periodic check-up is likely to occur despite the visit’s agenda, not as part of it.^3-7

Primary care physicians and practices are increasingly being held accountable for their performance on preventive measures. With that in mind, being familiar with evidence-based guidelines relating to the delivery of preventive services in ambulatory care settings, such as those developed by the US Preventive Services Task Force (USPSTF),⁸ is crucial. Identifying elements of the chronic care model that can be effectively applied to preventive care—and recognizing that the reactive acute care model is ineffective for both chronic illness and preventive services—is essential, as well.

Preventive care that's evidence-based

Good practice guidelines should have the following features, according to the Institute of Medicine:⁹
• Validity. Application would lead to the desired health and cost outcomes.
• Reliability/reproducibility. Others using the same data/interpretation would reach the same conclusion.
• Clinical applicability. Patient populations are appropriately defined.
• Clinical flexibility. Known or generally expected exceptions are identified.
• Clarity. Guideline uses unambiguous language with precise definitions.
• Multidisciplinary process. Developed with the participation of all key stakeholders.
• Scheduled review. Periodically reviewed and revised to incorporate new evidence/changing consensus.
• Documentation. Procedures used in development are well documented.

It’s estimated that a primary care physician with an average-sized patient panel would need an additional 7.4 hours per day to achieve 100% compliance with all of the USPSTF recommendations.

The USPSTF guidelines are consistent with these attributes.⁸

Some brief interventions fail

Guidelines are useful as practice standards for establishing preventive care goals, but their existence alone does not ensure the delivery of high-quality preventive services in an office setting. One key factor is time. It is estimated that a primary care physician with an average-sized patient panel would require an additional 7.4 hours per day to achieve 100% compliance with all of the USPSTF recommendations.¹⁰

Counseling, in particular, is an intervention that may be more effective in theory than in practice. Evidence suggests that even when primary care providers are trained in preventive counseling (and many are not), many brief interventions are not effective at creating sustained behavior change or health improvement.¹¹

Others are effective

That’s not to say that there’s little that can be done. Use of standing orders for influenza vaccination is one example of an effective and easily implemented preventive measure that requires little or no additional physician time. Yet some doctors are resistant, fearing loss of control or lawsuits. In fact, the National Vaccine Injury Compensation Program specifically provides protection against vaccinerelated malpractice claims.¹² A standing order for office staff to arrange a screening mammogram is another effective intervention; it has been shown to improve screening rates by as much as 30%.¹³

Lessons from chronic illness management

Chronic illness care and preventive care have much in common.¹⁴ Both acknowledge the need for proactive screening and counseling to bring about behavior change. In addition, both require ongoing care and follow-up, as well as depend on links to community resources. And finally, both are resource-intensive—too resource-intensive, some say, to be delivered in a cost-effective manner.

The Chronic Care Model has been proposed as a framework for improving preventive services (TABLE).^15,16 Evidence suggests adopting some key components of chronic care can lead to significant gains in the delivery of preventive care, with the largest improvement seen when multiple components are used simultaneously.^17-20

Self-management support. A growing body of evidence shows that self-management activities are associated with improved outcomes.²¹ Instituting a peer support group for pregnant women at a federally qualified health center, for example, led to a 7% absolute risk reduction in preterm births.²²

Not all efforts to encourage self-management are effective, however, and evaluation is crucial to determine what works. In one large-scale trial, the use of patient reminder cards to facilitate a reduction in health risk behaviors such as tobacco use, risky drinking, unhealthy dietary patterns, and physical inactivity led to fewer health risk assessments being performed, fewer individual counseling encounters, and no change in these behaviors.²³

Decision support. Clinical decision support systems, which generate patient-specific evidence-based assessments and/or recommendations that are actionable as part of the workflow at the point of care, have been found to improve care.^24,25 An example of this is a prompt that reminds the physician to discuss chemoprevention with a patient at high risk for breast cancer.

Delivery system design. The patient-centered medical home (PCMH) is a well-known example of a redesign of health care delivery.²⁶ Conversion to this model is associated with a small positive effect on preventive interventions.²⁷ However, the persistence of a fee-for-service payment system—which does not include physician reimbursement for some of the added services incorporated into the PCMH—limits the implementation of the PCMH model.²⁸

Many practices are improving health care delivery by using nonphysicians for various tasks related to preventive care. Care managers, for example, typically have smaller caseloads and focus on reducing unnecessary treatment for patients with high-risk conditions, such as congestive heart failure, while patient navigators generally have less clinical expertise but more knowledge of community services.

In one study, practice-initiated phone conversations with nonphysicians increased colorectal cancer screening by up to 40%.²⁹ And in one pilot program, the use of ancillary providers led to an increase in colorectal cancer screening by as much as 123%.³⁰ The human touch seems to be key to the success of these interventions. Passive reminders, such as videos being shown on waiting room televisions, have not proven to be effective.³¹ 

Clinical information systems. Early on, the power of electronic health records (EHRs) to improve practices’ delivery of preventive services was recognized. As early as 1995, the use of a reminder system to highlight such services during acute care visits was linked to improvements in counseling about smoking cessation and higher rates of cervical cancer screening, among other preventive measures.^32,33 Overall, the use of EHRs alone has been shown to improve rates of preventive services by as much as 66%, with most practices reporting improvements of at least 20%.³⁴

Use of standing orders for influenza vaccination is one example of an effective and easily implemented preventive measure that requires little or no additional physician time.

Today, EHRs that are Meaningful Use Stage 2-compliant have the tools needed to improve care. Requirements include the ability to generate patient registries of all those with a given disease and to identify patients on the registry who have not received needed care.³⁵ To improve preventive care, registries should focus on the mitigation of risk factors, such as identifying—and contacting—patients with diabetes who are in need of, or overdue for, an annual eye exam.

Trials using the registry function, in combination with automated messaging to deliver targeted information to various patient groups (identified by the demographic information available from the EHR), are ongoing.

Community resources. Many clinicians have informal referral relationships with community organizations, such as the YMCA. Physician practices that establish links to community resources have the potential to have a large effect on unhealthy behaviors. (See “Putting theory into practice: 2 cases”.) However, a systematic review found that, while evidence to support such connections is mainly positive, research is limited and further evaluation is needed.³⁶

The Practical Playbook (practicalplaybook.org)³⁷ developed by the deBeaumont Foundation, Duke Department of Community and Family Medicine, and the Centers for Disease Control and Prevention, offers concrete examples of how physician practices are linking with community resources to improve the health of the population. For example, Duke University’s “Just for Us” program provides in-home chronic illness care to 350 high-risk elderly individuals. The LATCH program connects thousands of Latino immigrants to health care services and culturally and linguistically appropriate health education classes.³⁷

Growing emphasis on quality

Systemic changes in the US health care system are occurring rapidly, with an emphasis on quality and improved outcomes. Many physicians are now required to submit data to external agencies for payment, and much of the data is grounded in preventive standards. Medicare’s Physicians Quality Reporting System requires that all Medicare providers provide data on preventive and chronic illness care. Rates of vaccination, obesity screening, and tobacco use screening are examples of preventive services that will be reported publicly on the Centers for Medicare & Medicaid Services’ Physician Compare Web site.³⁸

Physicians who work in accountable care organizations are required to meet quality standards on the delivery of certain preventive services, including breast cancer screening, colorectal cancer screening, influenza and pneumonia immunization, body mass index screening and follow-up, tobacco use screening and cessation intervention, screening for high blood pressure and followup, and screening for clinical depression and follow-up.³⁹ As patients discover that the Affordable Care Act mandates that preventive services be covered with no cost sharing, they are likely to become more receptive to physician attempts to provide them.^40,41

CORRESPONDENCE
Gerald Liu, MD, 1504 Springhill Avenue, Suite 3414, Mobile, AL 36604-3207; gliu@health.southalabama.edu

CASE › Joe M, age 59, seeks care at the local emergency department (ED) for shortness of breath. He also complains that his abdomen has been getting “bigger and bigger.” The ED physician recognizes that he is suffering from cirrhosis with secondary ascites and admits him. A paracentesis is performed and 7 L of fluid are removed. The patient is started on furosemide 40 mg/d and the health care team educates him about the relationship between his alcohol consumption and his enlarging abdomen. At discharge, he is told to follow up with his primary care physician.

Two weeks later, the patient arrives at your clinic for followup. What is the next step in managing this patient?

Cirrhosis—the end stage of chronic liver disease characterized by inflammation and fibrosis—is a relatively common and often fatal diagnosis. In the United States, an estimated 633,000 adults have cirrhosis,¹ and each year approximately 32,000 people die from the condition.² The most common causes of cirrhosis are heavy alcohol use, chronic hepatitis B or C infection, nonalcoholic fatty liver disease, and nonalcoholic steatohepatitis.³ Cirrhosis typically involves degeneration and necrosis of hepatocytes, which are replaced by fibrotic tissues and regenerative nodules, leading to loss of liver function.⁴

Patients with cirrhosis can be treated as outpatients—that is, until they decompensate. Obviously, treatment specific to the underlying causes of cirrhosis, such as interferon for a patient with hepatitis and abstinence for a patient with alcohol-related liver disease, should be the first concern. However, this article focuses on the family physician’s role in identifying and treating several of the most common complications of cirrhosis, including ascites, variceal bleeding, hepatic encephalopathy, and hepatorenal syndrome. We will also cover which patients should be referred for evaluation for liver transplantation. (For a guide to providing patient education for individuals with cirrhosis, see “Dx cirrhosis: What to teach your patient”.^3,5-10)

Sodium restriction, diuretics are first steps for ascites

The goals of ascites treatment are to prevent or relieve dyspnea and abdominal pain and to prevent life-threatening complications, such as spontaneous bacterial peritonitis (SBP) and hepatorenal syndrome.¹¹ Patient education is key regarding weight gain; that’s why it’s important to instruct patients to contact you if they gain more than 2 lbs/d for 3 consecutive days or more than 10 lbs.¹²

Approximately 10% of patients with ascites respond well to sodium restriction alone (1500-2000 mg/d).¹¹ In addition to sodium restriction, patients with grade 2 ascites (moderate ascites with proportionate abdominal distension) should receive a low-dose diuretic, such as spironolactone (initial dose, 50-100 mg/d; increase up to 200-300 mg/d)¹³ or amiloride (5-10 mg/d).⁵

Approximately 10% of patients with ascites respond well to sodium restriction alone.

Painful gynecomastia and hyperkalemia are the most common adverse effects of spironolactone.¹³ Amiloride has fewer adverse effects than spironolactone, but is less effective.¹¹ Low-dose furosemide (20-40 mg/d) may be added, although weight loss should be monitored to watch for excessive diuresis, which can lead to renal failure, hyponatremia, or encephalopathy.^5,13 Also monitor electrolytes to watch for hypokalemia or hyponatremia.¹³

Recommended weight loss to prevent renal failure is 300 to 500 g/d (.66-1.1 lbs/d) for patients without peripheral edema, and 800 to 1000 g/d (1.7-2.2 lbs/d) for patients with peripheral edema.^5,13

Patients with grade 3 (tense) or refractory ascites should have large-volume paracentesis (LVP) plus an albumin infusion.⁵ LVP (removal of >5 L of fluid) is more effective, faster, and has less risk of adverse effects than increasing the dosage of the patient’s diuretic.^5,13 LVP can be done in an outpatient setting and is considered safe—even for patients with a prolonged prothrombin time.^13,14 Rare complications of LVP include significant bleeding at the puncture site, infection, and intestinal perforation.⁵

Diuretics should be prescribed after LVP to prevent ascites recurrence.⁵ Plasma expanders can prevent hepatorenal syndrome, ascites recurrence, and dilutional hyponatremia.^5,11 Albumin is the most efficacious of these agents;^5,14 it is administered intravenously at a dose of 8 to 10 g/L of fluid removed.^13,15

Take steps to prevent variceal bleeding

Soon after a patient is diagnosed with cirrhosis, he or she should undergo esophagogastroduodenoscopy to screen for the presence and size of varices.¹⁶ Although they can’t prevent esophagogastric varices, nonselective beta-blockers (NSBBs) are the gold standard for preventing first variceal hemorrhage in patients with small varices with red wale signs on the varices and/or Child-Pugh Class B or C cirrhosis (TABLE¹⁷), and in all patients with medium or large varices.¹⁸ Propranolol is usually started at 20 mg BID, or nadolol is started at 20 to 40 mg/d.¹⁶ The NSBB dose is adjusted to the maximum tolerated dose, which occurs when the patient's heart rate is reduced to 55 to 60 beats/min.

NSBBs are associated with poor survival in patients with refractory ascites and thus are contraindicated in these patients.¹⁹ NSBBs also should not be taken by patients with SBP because use of these medications is associated with worse outcomes compared to those not receiving NSBBs.²⁰

Endoscopic variceal ligation is an alternative to NSBBs for the primary prophylaxis of variceal hemorrhage in patients with medium to large varices.¹⁸ In particular, ligation should be considered for patients with high-risk varices in whom beta-blockers are contraindicated or must be discontinued because of adverse effects.²¹

Avoid nitrates in patients with varices because these agents do not prevent first variceal hemorrhage and have been associated with higher mortality rates in patients older than 50.¹⁶ There is no significant additional benefit or mortality reduction associated with adding a nitrate to an NSBB.²² Transjugular intrahepatic portosystemic shunt (TIPS) or surgically created shunts are reserved for patients for whom medical therapy fails.¹⁸

Mental status changes suggest hepatic encephalopathy

Hepatic encephalopathy is a reversible impairment of neuropsychiatric function that is associated with impaired hepatic function. Because a patient with encephalopathy presents with an altered mental status, he or she may need to be admitted to the hospital for evaluation, diagnosis, and treatment.

The goals of hepatic encephalopathy treatment are to identify and correct precipitating causes and lower serum ammonia concentrations to improve mental status.¹⁵ Nutritional support should be provided without protein restriction unless the patient is severely proteinintolerant.²³ The recommended initial therapy is lactulose 30 to 45 mL 2 to 4 times per day, to decrease absorption of ammonia in the gut. The dose should be titrated until patients have 2 to 3 soft stools daily.²⁴

For patients who can’t tolerate lactulose or whose mental status doesn’t improve within 48 hours, rifaximin 400 mg orally 3 times daily or 550 mg 2 times daily is recommended.²⁵ Neomycin 500 mg orally 3 times a day or 1 g twice daily is a second-line agent reserved for patients who are unable to take rifaximin; however, its efficacy is not well established, and neomycin has been associated with ototoxicity and nephrotoxicity.²⁴

Watch for signs of kidney failure

Hepatorenal syndrome is renal failure induced by severe hepatic injury and characterized by azotemia and decreased renal blood flow and glomerular filtration rate.¹⁵ It is a diagnosis of exclusion. Hepatorenal syndrome is typically caused by arterial vasodilation in the splanchnic circulation in patients with portal hypertension.^15,26,27 Type 1 hepatorenal syndrome is characterized by at least a 2-fold increase in serum creatinine to a level of >2.5 mg/dL over more than 2 weeks. Patients typically have urine output <400 to 500 mL/d. Type 2 hepatorenal syndrome is characterized by less severe renal impairment; it is associated with ascites that does not improve with diuretics.²⁸

Endoscopic variceal ligation is an alternative to nonselective beta-blockers for preventing variceal hemorrhage in patients with medium to large varices.

Patients with hepatorenal syndrome should not use any nephrotoxic agents, such as nonsteroidal anti-inflammatory drugs. Inpatient treatment is usually required and may include norepinephrine with albumin, terlipressin with midodrine, or octreotide and albumin. Patients who fail to respond to medical therapy may benefit from TIPS as a bridge until they can undergo liver transplantation.²⁹

When to consider liver transplantation

The appropriateness and timing of liver transplantation should be determined on a case-by-case basis. For some patients with cirrhosis, transplantation may be the definitive treatment. For example, in some patients with hepatocellular carcinoma (HCC), liver transplantation is an option because transplantation can cure the tumor and underlying cirrhosis. However, while transplantation is a suitable option for early HCC in patients with cirrhosis, it has been shown to have limited efficacy in patients with advanced disease who are not selected using specific criteria.³⁰

Referral for evaluation for transplantation should be considered once a patient with cirrhosis experiences a major complication (eg, ascites, variceal hemorrhage, or hepatic encephalopathy).³¹ Another criterion for timing and allocation of liver transplantation is based on the statistical model for end-stage liver disease (MELD), which is used to predict 3-month survival in patients with cirrhosis based on the relationships between serum bilirubin, serum creatinine, and international normalized ratio values.¹⁵ Liver transplantation should be considered for patients with a MELD score ≥15.^15,31 Such patients should be promptly referred to a liver transplantation specialist to allow sufficient time for the appropriate psychosocial assessments and medical evaluations, and for patients and their families to receive appropriate education on things like the risks and benefits of transplantation.¹⁵

CASE › After evaluating Mr. M, you prescribe spironolactone 100 mg/d and furosemide 40 mg twice a day to address ascites, and propranolol—which you titrate to 80 mg twice a day—to prevent variceal hemorrhage. Mr. M is maintained on these medications and returns with his daughter, as he has been doing every 2 to 3 months. He is excited that he breathes easily as long as he avoids salt and takes his medications. He continues to see his hepatologist regularly, and his last paracentesis was 4 months ago. He has not used any alcohol since he was taught about the relationship between alcohol and his breathing.

CORRESPONDENCE
Suzanne Minor, MD, FAAF P, Assistant Professor of Family Medicine, Department of Humanities, Health, & Society, Florida International University, Herbert Wertheim College of Medicine, 11200 SW 8th Street, AHC II, 554A, Miami, FL 33199; seminor@fiu.edu

CASE › Joe M, age 59, seeks care at the local emergency department (ED) for shortness of breath. He also complains that his abdomen has been getting “bigger and bigger.” The ED physician recognizes that he is suffering from cirrhosis with secondary ascites and admits him. A paracentesis is performed and 7 L of fluid are removed. The patient is started on furosemide 40 mg/d and the health care team educates him about the relationship between his alcohol consumption and his enlarging abdomen. At discharge, he is told to follow up with his primary care physician.

Two weeks later, the patient arrives at your clinic for followup. What is the next step in managing this patient?

Cirrhosis—the end stage of chronic liver disease characterized by inflammation and fibrosis—is a relatively common and often fatal diagnosis. In the United States, an estimated 633,000 adults have cirrhosis,¹ and each year approximately 32,000 people die from the condition.² The most common causes of cirrhosis are heavy alcohol use, chronic hepatitis B or C infection, nonalcoholic fatty liver disease, and nonalcoholic steatohepatitis.³ Cirrhosis typically involves degeneration and necrosis of hepatocytes, which are replaced by fibrotic tissues and regenerative nodules, leading to loss of liver function.⁴

Patients with cirrhosis can be treated as outpatients—that is, until they decompensate. Obviously, treatment specific to the underlying causes of cirrhosis, such as interferon for a patient with hepatitis and abstinence for a patient with alcohol-related liver disease, should be the first concern. However, this article focuses on the family physician’s role in identifying and treating several of the most common complications of cirrhosis, including ascites, variceal bleeding, hepatic encephalopathy, and hepatorenal syndrome. We will also cover which patients should be referred for evaluation for liver transplantation. (For a guide to providing patient education for individuals with cirrhosis, see “Dx cirrhosis: What to teach your patient”.^3,5-10)

Sodium restriction, diuretics are first steps for ascites

The goals of ascites treatment are to prevent or relieve dyspnea and abdominal pain and to prevent life-threatening complications, such as spontaneous bacterial peritonitis (SBP) and hepatorenal syndrome.¹¹ Patient education is key regarding weight gain; that’s why it’s important to instruct patients to contact you if they gain more than 2 lbs/d for 3 consecutive days or more than 10 lbs.¹²

Approximately 10% of patients with ascites respond well to sodium restriction alone (1500-2000 mg/d).¹¹ In addition to sodium restriction, patients with grade 2 ascites (moderate ascites with proportionate abdominal distension) should receive a low-dose diuretic, such as spironolactone (initial dose, 50-100 mg/d; increase up to 200-300 mg/d)¹³ or amiloride (5-10 mg/d).⁵

Approximately 10% of patients with ascites respond well to sodium restriction alone.

Painful gynecomastia and hyperkalemia are the most common adverse effects of spironolactone.¹³ Amiloride has fewer adverse effects than spironolactone, but is less effective.¹¹ Low-dose furosemide (20-40 mg/d) may be added, although weight loss should be monitored to watch for excessive diuresis, which can lead to renal failure, hyponatremia, or encephalopathy.^5,13 Also monitor electrolytes to watch for hypokalemia or hyponatremia.¹³

Recommended weight loss to prevent renal failure is 300 to 500 g/d (.66-1.1 lbs/d) for patients without peripheral edema, and 800 to 1000 g/d (1.7-2.2 lbs/d) for patients with peripheral edema.^5,13

Patients with grade 3 (tense) or refractory ascites should have large-volume paracentesis (LVP) plus an albumin infusion.⁵ LVP (removal of >5 L of fluid) is more effective, faster, and has less risk of adverse effects than increasing the dosage of the patient’s diuretic.^5,13 LVP can be done in an outpatient setting and is considered safe—even for patients with a prolonged prothrombin time.^13,14 Rare complications of LVP include significant bleeding at the puncture site, infection, and intestinal perforation.⁵

Diuretics should be prescribed after LVP to prevent ascites recurrence.⁵ Plasma expanders can prevent hepatorenal syndrome, ascites recurrence, and dilutional hyponatremia.^5,11 Albumin is the most efficacious of these agents;^5,14 it is administered intravenously at a dose of 8 to 10 g/L of fluid removed.^13,15

Take steps to prevent variceal bleeding

Soon after a patient is diagnosed with cirrhosis, he or she should undergo esophagogastroduodenoscopy to screen for the presence and size of varices.¹⁶ Although they can’t prevent esophagogastric varices, nonselective beta-blockers (NSBBs) are the gold standard for preventing first variceal hemorrhage in patients with small varices with red wale signs on the varices and/or Child-Pugh Class B or C cirrhosis (TABLE¹⁷), and in all patients with medium or large varices.¹⁸ Propranolol is usually started at 20 mg BID, or nadolol is started at 20 to 40 mg/d.¹⁶ The NSBB dose is adjusted to the maximum tolerated dose, which occurs when the patient's heart rate is reduced to 55 to 60 beats/min.

NSBBs are associated with poor survival in patients with refractory ascites and thus are contraindicated in these patients.¹⁹ NSBBs also should not be taken by patients with SBP because use of these medications is associated with worse outcomes compared to those not receiving NSBBs.²⁰

Endoscopic variceal ligation is an alternative to NSBBs for the primary prophylaxis of variceal hemorrhage in patients with medium to large varices.¹⁸ In particular, ligation should be considered for patients with high-risk varices in whom beta-blockers are contraindicated or must be discontinued because of adverse effects.²¹

Avoid nitrates in patients with varices because these agents do not prevent first variceal hemorrhage and have been associated with higher mortality rates in patients older than 50.¹⁶ There is no significant additional benefit or mortality reduction associated with adding a nitrate to an NSBB.²² Transjugular intrahepatic portosystemic shunt (TIPS) or surgically created shunts are reserved for patients for whom medical therapy fails.¹⁸

Mental status changes suggest hepatic encephalopathy

Hepatic encephalopathy is a reversible impairment of neuropsychiatric function that is associated with impaired hepatic function. Because a patient with encephalopathy presents with an altered mental status, he or she may need to be admitted to the hospital for evaluation, diagnosis, and treatment.

The goals of hepatic encephalopathy treatment are to identify and correct precipitating causes and lower serum ammonia concentrations to improve mental status.¹⁵ Nutritional support should be provided without protein restriction unless the patient is severely proteinintolerant.²³ The recommended initial therapy is lactulose 30 to 45 mL 2 to 4 times per day, to decrease absorption of ammonia in the gut. The dose should be titrated until patients have 2 to 3 soft stools daily.²⁴

For patients who can’t tolerate lactulose or whose mental status doesn’t improve within 48 hours, rifaximin 400 mg orally 3 times daily or 550 mg 2 times daily is recommended.²⁵ Neomycin 500 mg orally 3 times a day or 1 g twice daily is a second-line agent reserved for patients who are unable to take rifaximin; however, its efficacy is not well established, and neomycin has been associated with ototoxicity and nephrotoxicity.²⁴

Watch for signs of kidney failure

Hepatorenal syndrome is renal failure induced by severe hepatic injury and characterized by azotemia and decreased renal blood flow and glomerular filtration rate.¹⁵ It is a diagnosis of exclusion. Hepatorenal syndrome is typically caused by arterial vasodilation in the splanchnic circulation in patients with portal hypertension.^15,26,27 Type 1 hepatorenal syndrome is characterized by at least a 2-fold increase in serum creatinine to a level of >2.5 mg/dL over more than 2 weeks. Patients typically have urine output <400 to 500 mL/d. Type 2 hepatorenal syndrome is characterized by less severe renal impairment; it is associated with ascites that does not improve with diuretics.²⁸

Endoscopic variceal ligation is an alternative to nonselective beta-blockers for preventing variceal hemorrhage in patients with medium to large varices.

Patients with hepatorenal syndrome should not use any nephrotoxic agents, such as nonsteroidal anti-inflammatory drugs. Inpatient treatment is usually required and may include norepinephrine with albumin, terlipressin with midodrine, or octreotide and albumin. Patients who fail to respond to medical therapy may benefit from TIPS as a bridge until they can undergo liver transplantation.²⁹

When to consider liver transplantation

The appropriateness and timing of liver transplantation should be determined on a case-by-case basis. For some patients with cirrhosis, transplantation may be the definitive treatment. For example, in some patients with hepatocellular carcinoma (HCC), liver transplantation is an option because transplantation can cure the tumor and underlying cirrhosis. However, while transplantation is a suitable option for early HCC in patients with cirrhosis, it has been shown to have limited efficacy in patients with advanced disease who are not selected using specific criteria.³⁰

Referral for evaluation for transplantation should be considered once a patient with cirrhosis experiences a major complication (eg, ascites, variceal hemorrhage, or hepatic encephalopathy).³¹ Another criterion for timing and allocation of liver transplantation is based on the statistical model for end-stage liver disease (MELD), which is used to predict 3-month survival in patients with cirrhosis based on the relationships between serum bilirubin, serum creatinine, and international normalized ratio values.¹⁵ Liver transplantation should be considered for patients with a MELD score ≥15.^15,31 Such patients should be promptly referred to a liver transplantation specialist to allow sufficient time for the appropriate psychosocial assessments and medical evaluations, and for patients and their families to receive appropriate education on things like the risks and benefits of transplantation.¹⁵

CASE › After evaluating Mr. M, you prescribe spironolactone 100 mg/d and furosemide 40 mg twice a day to address ascites, and propranolol—which you titrate to 80 mg twice a day—to prevent variceal hemorrhage. Mr. M is maintained on these medications and returns with his daughter, as he has been doing every 2 to 3 months. He is excited that he breathes easily as long as he avoids salt and takes his medications. He continues to see his hepatologist regularly, and his last paracentesis was 4 months ago. He has not used any alcohol since he was taught about the relationship between alcohol and his breathing.

CORRESPONDENCE
Suzanne Minor, MD, FAAF P, Assistant Professor of Family Medicine, Department of Humanities, Health, & Society, Florida International University, Herbert Wertheim College of Medicine, 11200 SW 8th Street, AHC II, 554A, Miami, FL 33199; seminor@fiu.edu

CASE › Joe M, age 59, seeks care at the local emergency department (ED) for shortness of breath. He also complains that his abdomen has been getting “bigger and bigger.” The ED physician recognizes that he is suffering from cirrhosis with secondary ascites and admits him. A paracentesis is performed and 7 L of fluid are removed. The patient is started on furosemide 40 mg/d and the health care team educates him about the relationship between his alcohol consumption and his enlarging abdomen. At discharge, he is told to follow up with his primary care physician.

Two weeks later, the patient arrives at your clinic for followup. What is the next step in managing this patient?

Cirrhosis—the end stage of chronic liver disease characterized by inflammation and fibrosis—is a relatively common and often fatal diagnosis. In the United States, an estimated 633,000 adults have cirrhosis,¹ and each year approximately 32,000 people die from the condition.² The most common causes of cirrhosis are heavy alcohol use, chronic hepatitis B or C infection, nonalcoholic fatty liver disease, and nonalcoholic steatohepatitis.³ Cirrhosis typically involves degeneration and necrosis of hepatocytes, which are replaced by fibrotic tissues and regenerative nodules, leading to loss of liver function.⁴

Patients with cirrhosis can be treated as outpatients—that is, until they decompensate. Obviously, treatment specific to the underlying causes of cirrhosis, such as interferon for a patient with hepatitis and abstinence for a patient with alcohol-related liver disease, should be the first concern. However, this article focuses on the family physician’s role in identifying and treating several of the most common complications of cirrhosis, including ascites, variceal bleeding, hepatic encephalopathy, and hepatorenal syndrome. We will also cover which patients should be referred for evaluation for liver transplantation. (For a guide to providing patient education for individuals with cirrhosis, see “Dx cirrhosis: What to teach your patient”.^3,5-10)

Sodium restriction, diuretics are first steps for ascites

The goals of ascites treatment are to prevent or relieve dyspnea and abdominal pain and to prevent life-threatening complications, such as spontaneous bacterial peritonitis (SBP) and hepatorenal syndrome.¹¹ Patient education is key regarding weight gain; that’s why it’s important to instruct patients to contact you if they gain more than 2 lbs/d for 3 consecutive days or more than 10 lbs.¹²

Approximately 10% of patients with ascites respond well to sodium restriction alone (1500-2000 mg/d).¹¹ In addition to sodium restriction, patients with grade 2 ascites (moderate ascites with proportionate abdominal distension) should receive a low-dose diuretic, such as spironolactone (initial dose, 50-100 mg/d; increase up to 200-300 mg/d)¹³ or amiloride (5-10 mg/d).⁵

Approximately 10% of patients with ascites respond well to sodium restriction alone.

Painful gynecomastia and hyperkalemia are the most common adverse effects of spironolactone.¹³ Amiloride has fewer adverse effects than spironolactone, but is less effective.¹¹ Low-dose furosemide (20-40 mg/d) may be added, although weight loss should be monitored to watch for excessive diuresis, which can lead to renal failure, hyponatremia, or encephalopathy.^5,13 Also monitor electrolytes to watch for hypokalemia or hyponatremia.¹³

Recommended weight loss to prevent renal failure is 300 to 500 g/d (.66-1.1 lbs/d) for patients without peripheral edema, and 800 to 1000 g/d (1.7-2.2 lbs/d) for patients with peripheral edema.^5,13

Patients with grade 3 (tense) or refractory ascites should have large-volume paracentesis (LVP) plus an albumin infusion.⁵ LVP (removal of >5 L of fluid) is more effective, faster, and has less risk of adverse effects than increasing the dosage of the patient’s diuretic.^5,13 LVP can be done in an outpatient setting and is considered safe—even for patients with a prolonged prothrombin time.^13,14 Rare complications of LVP include significant bleeding at the puncture site, infection, and intestinal perforation.⁵

Diuretics should be prescribed after LVP to prevent ascites recurrence.⁵ Plasma expanders can prevent hepatorenal syndrome, ascites recurrence, and dilutional hyponatremia.^5,11 Albumin is the most efficacious of these agents;^5,14 it is administered intravenously at a dose of 8 to 10 g/L of fluid removed.^13,15

Take steps to prevent variceal bleeding

Soon after a patient is diagnosed with cirrhosis, he or she should undergo esophagogastroduodenoscopy to screen for the presence and size of varices.¹⁶ Although they can’t prevent esophagogastric varices, nonselective beta-blockers (NSBBs) are the gold standard for preventing first variceal hemorrhage in patients with small varices with red wale signs on the varices and/or Child-Pugh Class B or C cirrhosis (TABLE¹⁷), and in all patients with medium or large varices.¹⁸ Propranolol is usually started at 20 mg BID, or nadolol is started at 20 to 40 mg/d.¹⁶ The NSBB dose is adjusted to the maximum tolerated dose, which occurs when the patient's heart rate is reduced to 55 to 60 beats/min.

NSBBs are associated with poor survival in patients with refractory ascites and thus are contraindicated in these patients.¹⁹ NSBBs also should not be taken by patients with SBP because use of these medications is associated with worse outcomes compared to those not receiving NSBBs.²⁰

Endoscopic variceal ligation is an alternative to NSBBs for the primary prophylaxis of variceal hemorrhage in patients with medium to large varices.¹⁸ In particular, ligation should be considered for patients with high-risk varices in whom beta-blockers are contraindicated or must be discontinued because of adverse effects.²¹

Avoid nitrates in patients with varices because these agents do not prevent first variceal hemorrhage and have been associated with higher mortality rates in patients older than 50.¹⁶ There is no significant additional benefit or mortality reduction associated with adding a nitrate to an NSBB.²² Transjugular intrahepatic portosystemic shunt (TIPS) or surgically created shunts are reserved for patients for whom medical therapy fails.¹⁸

Mental status changes suggest hepatic encephalopathy

Hepatic encephalopathy is a reversible impairment of neuropsychiatric function that is associated with impaired hepatic function. Because a patient with encephalopathy presents with an altered mental status, he or she may need to be admitted to the hospital for evaluation, diagnosis, and treatment.

The goals of hepatic encephalopathy treatment are to identify and correct precipitating causes and lower serum ammonia concentrations to improve mental status.¹⁵ Nutritional support should be provided without protein restriction unless the patient is severely proteinintolerant.²³ The recommended initial therapy is lactulose 30 to 45 mL 2 to 4 times per day, to decrease absorption of ammonia in the gut. The dose should be titrated until patients have 2 to 3 soft stools daily.²⁴

For patients who can’t tolerate lactulose or whose mental status doesn’t improve within 48 hours, rifaximin 400 mg orally 3 times daily or 550 mg 2 times daily is recommended.²⁵ Neomycin 500 mg orally 3 times a day or 1 g twice daily is a second-line agent reserved for patients who are unable to take rifaximin; however, its efficacy is not well established, and neomycin has been associated with ototoxicity and nephrotoxicity.²⁴

Watch for signs of kidney failure

Hepatorenal syndrome is renal failure induced by severe hepatic injury and characterized by azotemia and decreased renal blood flow and glomerular filtration rate.¹⁵ It is a diagnosis of exclusion. Hepatorenal syndrome is typically caused by arterial vasodilation in the splanchnic circulation in patients with portal hypertension.^15,26,27 Type 1 hepatorenal syndrome is characterized by at least a 2-fold increase in serum creatinine to a level of >2.5 mg/dL over more than 2 weeks. Patients typically have urine output <400 to 500 mL/d. Type 2 hepatorenal syndrome is characterized by less severe renal impairment; it is associated with ascites that does not improve with diuretics.²⁸

Endoscopic variceal ligation is an alternative to nonselective beta-blockers for preventing variceal hemorrhage in patients with medium to large varices.

Patients with hepatorenal syndrome should not use any nephrotoxic agents, such as nonsteroidal anti-inflammatory drugs. Inpatient treatment is usually required and may include norepinephrine with albumin, terlipressin with midodrine, or octreotide and albumin. Patients who fail to respond to medical therapy may benefit from TIPS as a bridge until they can undergo liver transplantation.²⁹

When to consider liver transplantation

The appropriateness and timing of liver transplantation should be determined on a case-by-case basis. For some patients with cirrhosis, transplantation may be the definitive treatment. For example, in some patients with hepatocellular carcinoma (HCC), liver transplantation is an option because transplantation can cure the tumor and underlying cirrhosis. However, while transplantation is a suitable option for early HCC in patients with cirrhosis, it has been shown to have limited efficacy in patients with advanced disease who are not selected using specific criteria.³⁰

Referral for evaluation for transplantation should be considered once a patient with cirrhosis experiences a major complication (eg, ascites, variceal hemorrhage, or hepatic encephalopathy).³¹ Another criterion for timing and allocation of liver transplantation is based on the statistical model for end-stage liver disease (MELD), which is used to predict 3-month survival in patients with cirrhosis based on the relationships between serum bilirubin, serum creatinine, and international normalized ratio values.¹⁵ Liver transplantation should be considered for patients with a MELD score ≥15.^15,31 Such patients should be promptly referred to a liver transplantation specialist to allow sufficient time for the appropriate psychosocial assessments and medical evaluations, and for patients and their families to receive appropriate education on things like the risks and benefits of transplantation.¹⁵

CASE › After evaluating Mr. M, you prescribe spironolactone 100 mg/d and furosemide 40 mg twice a day to address ascites, and propranolol—which you titrate to 80 mg twice a day—to prevent variceal hemorrhage. Mr. M is maintained on these medications and returns with his daughter, as he has been doing every 2 to 3 months. He is excited that he breathes easily as long as he avoids salt and takes his medications. He continues to see his hepatologist regularly, and his last paracentesis was 4 months ago. He has not used any alcohol since he was taught about the relationship between alcohol and his breathing.

CORRESPONDENCE
Suzanne Minor, MD, FAAF P, Assistant Professor of Family Medicine, Department of Humanities, Health, & Society, Florida International University, Herbert Wertheim College of Medicine, 11200 SW 8th Street, AHC II, 554A, Miami, FL 33199; seminor@fiu.edu

One of my favorite professional activities is teaching an evidence-based continuing medical education course each year at state Academy of Family Physicians meetings. In 12 intensive hours, 4 evidence-based medicine (EBM) experts guide family physicians, nurse practitioners, and physician assistants through nearly 400 abstracts that summarize recent studies that impact primary care practice.

In some cases, the new studies support current practice and standards of care, but for many topics, the new evidence suggests we ought to change our practice, either by stopping something we are currently doing or by starting to do something new. Who would have thought, for instance, that we should abandon the routine bimanual pelvic exam because the potential for harm is greater than the potential for benefit?

Frequently, however, we conclude a talk by describing the uncertainty surrounding particular issues and the need for more high-quality research. For example, there is scant evidence that vitamin D supplementation in healthy Americans leads to any positive outcomes compared to a decent diet and 15 minutes in the sun each day. Luckily, there are several large randomized trials currently underway that will evaluate vitamin D supplementation.

Who would have thought that we should abandon the routine bimanual pelvic exam? And yet, that is what the evidence tells us.

The strength of the scientific evidence to support screening tests and treatments is important to consider. A study examining changes in 11 American College of Cardiology/American Heart Association guidelines found that, out of 619 recommendations, 90% were unchanged in the updated version if supported by multiple randomized trials, and 74% were unchanged if supported by expert opinion.¹

In The Journal of Family Practice, we use the Strength of Recommendation Taxonomy (SORT) that was developed by family physician EBM experts² because it is an approach to grading evidence that takes into account “patient-oriented evidence that matters.” An A-level recommendation is based on consistent and good-quality patient-oriented evidence; a B-level recommendation is based on inconsistent or limited-quality patient-oriented evidence; and a C-level recommendation is based on consensus, usual practice, opinion, disease-oriented evidence, or case series.

We ask our authors to carefully select the level of evidence supporting their clinical recommendations. But your input—and the lively discussion that can often follow—is important, too. Just last month, we published a letter from 2 readers who challenged the evidence-based answer to a Clinical Inquiries question on breastfeeding.

Such ongoing dialogue is useful and enlightening. And we encourage you to write us if you disagree with any of the SORT ratings published in the journal. Let’s keep talking about what the evidence says.

One of my favorite professional activities is teaching an evidence-based continuing medical education course each year at state Academy of Family Physicians meetings. In 12 intensive hours, 4 evidence-based medicine (EBM) experts guide family physicians, nurse practitioners, and physician assistants through nearly 400 abstracts that summarize recent studies that impact primary care practice.

In some cases, the new studies support current practice and standards of care, but for many topics, the new evidence suggests we ought to change our practice, either by stopping something we are currently doing or by starting to do something new. Who would have thought, for instance, that we should abandon the routine bimanual pelvic exam because the potential for harm is greater than the potential for benefit?

Frequently, however, we conclude a talk by describing the uncertainty surrounding particular issues and the need for more high-quality research. For example, there is scant evidence that vitamin D supplementation in healthy Americans leads to any positive outcomes compared to a decent diet and 15 minutes in the sun each day. Luckily, there are several large randomized trials currently underway that will evaluate vitamin D supplementation.

Who would have thought that we should abandon the routine bimanual pelvic exam? And yet, that is what the evidence tells us.

The strength of the scientific evidence to support screening tests and treatments is important to consider. A study examining changes in 11 American College of Cardiology/American Heart Association guidelines found that, out of 619 recommendations, 90% were unchanged in the updated version if supported by multiple randomized trials, and 74% were unchanged if supported by expert opinion.¹

In The Journal of Family Practice, we use the Strength of Recommendation Taxonomy (SORT) that was developed by family physician EBM experts² because it is an approach to grading evidence that takes into account “patient-oriented evidence that matters.” An A-level recommendation is based on consistent and good-quality patient-oriented evidence; a B-level recommendation is based on inconsistent or limited-quality patient-oriented evidence; and a C-level recommendation is based on consensus, usual practice, opinion, disease-oriented evidence, or case series.

We ask our authors to carefully select the level of evidence supporting their clinical recommendations. But your input—and the lively discussion that can often follow—is important, too. Just last month, we published a letter from 2 readers who challenged the evidence-based answer to a Clinical Inquiries question on breastfeeding.

Such ongoing dialogue is useful and enlightening. And we encourage you to write us if you disagree with any of the SORT ratings published in the journal. Let’s keep talking about what the evidence says.

One of my favorite professional activities is teaching an evidence-based continuing medical education course each year at state Academy of Family Physicians meetings. In 12 intensive hours, 4 evidence-based medicine (EBM) experts guide family physicians, nurse practitioners, and physician assistants through nearly 400 abstracts that summarize recent studies that impact primary care practice.

In some cases, the new studies support current practice and standards of care, but for many topics, the new evidence suggests we ought to change our practice, either by stopping something we are currently doing or by starting to do something new. Who would have thought, for instance, that we should abandon the routine bimanual pelvic exam because the potential for harm is greater than the potential for benefit?

Frequently, however, we conclude a talk by describing the uncertainty surrounding particular issues and the need for more high-quality research. For example, there is scant evidence that vitamin D supplementation in healthy Americans leads to any positive outcomes compared to a decent diet and 15 minutes in the sun each day. Luckily, there are several large randomized trials currently underway that will evaluate vitamin D supplementation.

Who would have thought that we should abandon the routine bimanual pelvic exam? And yet, that is what the evidence tells us.

The strength of the scientific evidence to support screening tests and treatments is important to consider. A study examining changes in 11 American College of Cardiology/American Heart Association guidelines found that, out of 619 recommendations, 90% were unchanged in the updated version if supported by multiple randomized trials, and 74% were unchanged if supported by expert opinion.¹

In The Journal of Family Practice, we use the Strength of Recommendation Taxonomy (SORT) that was developed by family physician EBM experts² because it is an approach to grading evidence that takes into account “patient-oriented evidence that matters.” An A-level recommendation is based on consistent and good-quality patient-oriented evidence; a B-level recommendation is based on inconsistent or limited-quality patient-oriented evidence; and a C-level recommendation is based on consensus, usual practice, opinion, disease-oriented evidence, or case series.

We ask our authors to carefully select the level of evidence supporting their clinical recommendations. But your input—and the lively discussion that can often follow—is important, too. Just last month, we published a letter from 2 readers who challenged the evidence-based answer to a Clinical Inquiries question on breastfeeding.

Such ongoing dialogue is useful and enlightening. And we encourage you to write us if you disagree with any of the SORT ratings published in the journal. Let’s keep talking about what the evidence says.

THE CASE

A 27-year-old woman in the 21st week of her first pregnancy came to our clinic complaining of a constant burning pain that spread around her left chest wall to her back. She graded the pain as a 10 on a 0 to 10 visual analog scale. The pain, which began 3 months earlier, became worse when she took a deep breath, ate, or walked, but was alleviated by applying warm compresses. Our patient hadn’t slept well since the pain began. Her medical history was noteworthy for chickenpox at age 5.

During the physical examination, palpating her left upper abdominal quadrant and left lower chest wall elicited tenderness. We noted allodynia and hyperesthesia in these regions, and the left T5 dermatome revealed extreme sensitivity.

THE DIAGNOSIS

We decided to test for antibodies to the varicella-zoster virus (VZV) based on the location of the pain along a dermatome. A serum anti-VZV immunoglobulin G (IgG) level was high at 1.9. Since our patient hadn’t been vaccinated against VZV, her high IgG level may have been the result of reactivation of the virus. Based on this test result and our patient’s history and physical exam findings (ie, neuropathic pain along a dermatome without a typical herpes zoster rash), we diagnosed zoster sine herpete (ZSH).

DISCUSSION

One million new cases of herpes zoster (shingles) are diagnosed in the United States each year, with a rate of 3 to 4 cases per 1000 people.¹ One in 3 patients develops postherpetic neuralgia, depending on age and immunocompetence.¹

In ZSH, the neuropathic pain of herpes zoster occurs without the typical zoster rash.² Since the rash is absent, the diagnosis is often missed. The incidence of ZSH is unknown.

Although many pregnant women suffer from thoracic and/or abdominal neuropathic pain, there are no reports in the literature that describe ZSH in pregnant women.³

Suspect zoster sine herpete in a patient who has neuropathic radicular pain but no rash.

The appropriate diagnostic tests for ZSH are polymerase chain reaction for VZV DNA and anti-VZV IgG.^2,4-7 A definitive diagnosis can be reached by identifying herpes zoster DNA in cerebrospinal fluid (CSF) and organism-specific immunoglobulins. However, a high titer of serum IgG antibodies or a positive IgM antibodies test typically provides a high degree of certainty for the diagnosis.⁸ For our patient, we decided not to test her CSF because we felt that her clinical course and positive IgG test were sufficient to establish the diagnosis.

The differential diagnosis of radicular pain during pregnancy includes cutaneous nerve entrapment. The expanding uterus could increase pressure on cutaneous nerves in the abdominal wall and cause pain. Although nerve entrapment would be expected to cause impingement and sometimes hypoesthesia, ZSH usually causes allodynia and hyperesthesia, as was the case in our patient.³

Pregnancy affects choice of treatment

Treatments for ZSH include acyclovir and local anesthesia.⁸ A single injection of lidocaine (8 cc) may completely eliminate the ZSH pain by affecting the nerve action potential.⁹ Corticosteroids are used to suppress inflammation and decrease erythema, swelling, warmth at the site, and local tenderness.

Our patient. We decided to treat our patient with only a nerve block because the potential adverse effects of acyclovir in the second trimester of pregnancy are unclear.¹⁰ She received 1 cc of betamethasone acetate (3 mg) and betamethasone sodium phosphate (3 mg) and 8 cc of 2% lidocaine. The patient reported immediate pain relief, which lasted until delivery.

THE TAKEAWAY

ZSH is characterized by neuropathic pain along a dermatome that’s associated with herpes zoster and is not accompanied by the characteristic rash. Many pregnant women suffer from thoracic and abdominal wall neuropathic pain. Neuropathic radicular pain in the absence of a rash should raise suspicion of ZSH. Considering this syndrome at an early stage can avert unnecessary testing and reduce the patient’s pain.

THE CASE

A 27-year-old woman in the 21st week of her first pregnancy came to our clinic complaining of a constant burning pain that spread around her left chest wall to her back. She graded the pain as a 10 on a 0 to 10 visual analog scale. The pain, which began 3 months earlier, became worse when she took a deep breath, ate, or walked, but was alleviated by applying warm compresses. Our patient hadn’t slept well since the pain began. Her medical history was noteworthy for chickenpox at age 5.

During the physical examination, palpating her left upper abdominal quadrant and left lower chest wall elicited tenderness. We noted allodynia and hyperesthesia in these regions, and the left T5 dermatome revealed extreme sensitivity.

THE DIAGNOSIS

We decided to test for antibodies to the varicella-zoster virus (VZV) based on the location of the pain along a dermatome. A serum anti-VZV immunoglobulin G (IgG) level was high at 1.9. Since our patient hadn’t been vaccinated against VZV, her high IgG level may have been the result of reactivation of the virus. Based on this test result and our patient’s history and physical exam findings (ie, neuropathic pain along a dermatome without a typical herpes zoster rash), we diagnosed zoster sine herpete (ZSH).

DISCUSSION

One million new cases of herpes zoster (shingles) are diagnosed in the United States each year, with a rate of 3 to 4 cases per 1000 people.¹ One in 3 patients develops postherpetic neuralgia, depending on age and immunocompetence.¹

In ZSH, the neuropathic pain of herpes zoster occurs without the typical zoster rash.² Since the rash is absent, the diagnosis is often missed. The incidence of ZSH is unknown.

Although many pregnant women suffer from thoracic and/or abdominal neuropathic pain, there are no reports in the literature that describe ZSH in pregnant women.³

Suspect zoster sine herpete in a patient who has neuropathic radicular pain but no rash.

The appropriate diagnostic tests for ZSH are polymerase chain reaction for VZV DNA and anti-VZV IgG.^2,4-7 A definitive diagnosis can be reached by identifying herpes zoster DNA in cerebrospinal fluid (CSF) and organism-specific immunoglobulins. However, a high titer of serum IgG antibodies or a positive IgM antibodies test typically provides a high degree of certainty for the diagnosis.⁸ For our patient, we decided not to test her CSF because we felt that her clinical course and positive IgG test were sufficient to establish the diagnosis.

The differential diagnosis of radicular pain during pregnancy includes cutaneous nerve entrapment. The expanding uterus could increase pressure on cutaneous nerves in the abdominal wall and cause pain. Although nerve entrapment would be expected to cause impingement and sometimes hypoesthesia, ZSH usually causes allodynia and hyperesthesia, as was the case in our patient.³

Pregnancy affects choice of treatment

Treatments for ZSH include acyclovir and local anesthesia.⁸ A single injection of lidocaine (8 cc) may completely eliminate the ZSH pain by affecting the nerve action potential.⁹ Corticosteroids are used to suppress inflammation and decrease erythema, swelling, warmth at the site, and local tenderness.

Our patient. We decided to treat our patient with only a nerve block because the potential adverse effects of acyclovir in the second trimester of pregnancy are unclear.¹⁰ She received 1 cc of betamethasone acetate (3 mg) and betamethasone sodium phosphate (3 mg) and 8 cc of 2% lidocaine. The patient reported immediate pain relief, which lasted until delivery.

THE TAKEAWAY

ZSH is characterized by neuropathic pain along a dermatome that’s associated with herpes zoster and is not accompanied by the characteristic rash. Many pregnant women suffer from thoracic and abdominal wall neuropathic pain. Neuropathic radicular pain in the absence of a rash should raise suspicion of ZSH. Considering this syndrome at an early stage can avert unnecessary testing and reduce the patient’s pain.

THE CASE

A 27-year-old woman in the 21st week of her first pregnancy came to our clinic complaining of a constant burning pain that spread around her left chest wall to her back. She graded the pain as a 10 on a 0 to 10 visual analog scale. The pain, which began 3 months earlier, became worse when she took a deep breath, ate, or walked, but was alleviated by applying warm compresses. Our patient hadn’t slept well since the pain began. Her medical history was noteworthy for chickenpox at age 5.

During the physical examination, palpating her left upper abdominal quadrant and left lower chest wall elicited tenderness. We noted allodynia and hyperesthesia in these regions, and the left T5 dermatome revealed extreme sensitivity.

THE DIAGNOSIS

We decided to test for antibodies to the varicella-zoster virus (VZV) based on the location of the pain along a dermatome. A serum anti-VZV immunoglobulin G (IgG) level was high at 1.9. Since our patient hadn’t been vaccinated against VZV, her high IgG level may have been the result of reactivation of the virus. Based on this test result and our patient’s history and physical exam findings (ie, neuropathic pain along a dermatome without a typical herpes zoster rash), we diagnosed zoster sine herpete (ZSH).

DISCUSSION

One million new cases of herpes zoster (shingles) are diagnosed in the United States each year, with a rate of 3 to 4 cases per 1000 people.¹ One in 3 patients develops postherpetic neuralgia, depending on age and immunocompetence.¹

In ZSH, the neuropathic pain of herpes zoster occurs without the typical zoster rash.² Since the rash is absent, the diagnosis is often missed. The incidence of ZSH is unknown.

Although many pregnant women suffer from thoracic and/or abdominal neuropathic pain, there are no reports in the literature that describe ZSH in pregnant women.³

Suspect zoster sine herpete in a patient who has neuropathic radicular pain but no rash.

The appropriate diagnostic tests for ZSH are polymerase chain reaction for VZV DNA and anti-VZV IgG.^2,4-7 A definitive diagnosis can be reached by identifying herpes zoster DNA in cerebrospinal fluid (CSF) and organism-specific immunoglobulins. However, a high titer of serum IgG antibodies or a positive IgM antibodies test typically provides a high degree of certainty for the diagnosis.⁸ For our patient, we decided not to test her CSF because we felt that her clinical course and positive IgG test were sufficient to establish the diagnosis.

The differential diagnosis of radicular pain during pregnancy includes cutaneous nerve entrapment. The expanding uterus could increase pressure on cutaneous nerves in the abdominal wall and cause pain. Although nerve entrapment would be expected to cause impingement and sometimes hypoesthesia, ZSH usually causes allodynia and hyperesthesia, as was the case in our patient.³

Pregnancy affects choice of treatment

Treatments for ZSH include acyclovir and local anesthesia.⁸ A single injection of lidocaine (8 cc) may completely eliminate the ZSH pain by affecting the nerve action potential.⁹ Corticosteroids are used to suppress inflammation and decrease erythema, swelling, warmth at the site, and local tenderness.

Our patient. We decided to treat our patient with only a nerve block because the potential adverse effects of acyclovir in the second trimester of pregnancy are unclear.¹⁰ She received 1 cc of betamethasone acetate (3 mg) and betamethasone sodium phosphate (3 mg) and 8 cc of 2% lidocaine. The patient reported immediate pain relief, which lasted until delivery.

THE TAKEAWAY

ZSH is characterized by neuropathic pain along a dermatome that’s associated with herpes zoster and is not accompanied by the characteristic rash. Many pregnant women suffer from thoracic and abdominal wall neuropathic pain. Neuropathic radicular pain in the absence of a rash should raise suspicion of ZSH. Considering this syndrome at an early stage can avert unnecessary testing and reduce the patient’s pain.

THE CASE

A 55-year-old woman presented to the emergency department (ED) with a bifrontal headache that she’d had for one day. She also had blurred vision and was vomiting shortly before coming to the hospital. The patient had no history of hypertension, migraine headaches, seizure disorder, autoimmune disorders, or cerebrovascular disease.

Her vital signs, including a blood pressure of 114/63 mm Hg, were normal, but a physical examination revealed subjective vision loss. She was only able to see objects moving on a horizontal plane. Her finger-to-nose exam, pupillary reflexes, and extra-ocular movements were normal, but peripheral vision was limited on her left side. No other neurologic deficits were noted.

The patient was admitted to the hospital and most of her laboratory work-up was normal, including a basic metabolic panel, complete blood count, coagulation studies, brain natriuretic peptide test, and cardiac enzymes. Her white blood cell count was 19,700/mcL, but no source of infection was found. A computed tomography (CT) scan of her head without contrast showed low-density, patchy areas in the subcortical regions of the parietal and occipital lobes bilaterally (FIGURE 1, arrows), with relative sparing of the cortex.

THE DIAGNOSIS

Based on our patient’s presentation and radiologic findings, we made a diagnosis of posterior reversible encephalopathy syndrome (PRES). However, because we could not rule out an ischemic cerebrovascular event at the time of presentation, we started the patient on aspirin and clopidogrel 75 mg to prevent possible future ischemic events. The next day, we ordered magnetic resonance imaging (MRI) of the head and neck, which documented the edema and confirmed the diagnosis of PRES (FIGURE 2).

DISCUSSION

PRES is a neurotoxic state associated with a unique pattern of brain vasogenic edema seen on CT or MRI. The edema is often widespread but is predominantly found in the parietal and occipital regions.¹ PRES is seen in patients with a variety of conditions, including hypertension and bone marrow or organ transplantation, as well as in those receiving immunosuppressive or cytotoxic medications.¹ Patients with PRES typically present with headaches and seizures.² Visual abnormalities (most commonly cortical blindness), occur in 15% to 20% of patients with PRES.^2-4

Hinchey et al³ first described reversible posterior leukoencephalopathy syndrome (which later became known as PRES) in 1996. Most of the 15 patients included in this original report had a history of hypertension or immunosuppression. These cases were associated with cerebral edema in portions of the posterior cerebral white matter. It is thought that hypertension alters the blood-brain barrier and causes the acute changes that occur in PRES.³

Besides hypertension and immunosuppression, the risk factors most commonly associated with PRES include preeclampsia/eclampsia; sepsis, particularly due to grampositive organisms; Wegener’s granulomatosis, scleroderma, and polyarteritis nodosa; cancer chemotherapy; bone marrow or stem cell transplantation; and renal disease.^1,4-6

Although a clear cause of PRES has not yet been established, researchers have proposed 2 theories. The first postulates that a sudden increase in systemic blood pressure causes vasoconstriction, which leads to ischemia and edema.^1-4,7,8 However, several studies have also described cases of PRES in patients with mild elevations in blood pressure,^1,5-7 and mild edema has been observed even in normotensive patients^1,5 (as was the case with our patient).

The second theory links PRES to the loss of brain autoregulation, a function that maintains steady blood flow when blood pressure fluctuates.⁶ A loss of this regulatory mechanism causes endothelial dysfunction, capillary leakage, and disruption in the blood-brain barrier.^1,2,4,6-8 These changes then lead to cerebral vasodilatation and edema.² Immunotherapy has also been associated with increased endothelial dysfunction.²

The evidence on the link between the severity of PRES and clinical outcomes is conflicting.

One study that followed 113 PRES patients over 6 years did not find an association between the severity of clinical presentation and the extent of vasogenic edema found on imaging studies.⁵ Of these 113 patients, 69 had PRES primarily due to hypertension, and 21 were receiving cytotoxic medications.⁵ In contrast, a larger retrospective study that followed patients with PRES for 12 years found that severe cases, which included patients with severe cerebral edema and altered mental status, had poor outcomes.⁴ Small studies have reported that 14% of patients with PRES develop cerebral hemorrhage.⁸

When to suspect this condition. PRES should be part of the differential diagnosis for any patient who presents with headache and vision loss. It is important to distinguish PRES from an acute cerebrovascular accident (CVA) because the 2 conditions are managed differently.² In addition, PRES lesions can be misdiagnosed as tumors, especially in a patient with a history of malignant disease in whom the condition appears after chemotherapy.⁹

Treatment targets the underlying causes

Treatment options for PRES are limited. Hypertension in a patient with PRES requires prompt intervention to avoid progression of the disease.² The use of intravenous (IV) calcium-channel blockers or IV beta-blockers for these patients is common.^2,8

Patients with seizures should be treated with anticonvulsant medication, but longterm antiepileptic treatment usually is not required.² Patients who take immunosuppressant or cytotoxic drugs should stop them indefinitely upon presenting with PRES.²

PRES lesions can be misdiagnosed as tumors—especially in patients with a history of malignant disease who have undergone chemotherapy.

For a pregnant woman with preeclampsia/eclampsia, delivery of the placenta, which is considered to be the cause of PRES in these cases, is curative.¹ However, women can develop PRES several weeks after delivery.¹

In most cases, the symptoms associated with PRES will resolve once treatment is initiated, and neurologic recovery can be expected within 2 weeks.²

Our patient regained her sight the following morning and was discharged home 2 days after admission. Her blood pressure remained normal. She returned to the hospital unresponsive the day after she had been discharged. Family members stated that she had taken 15 packets of an aspirin/caffeine combination to control a new headache.

Her blood pressure was elevated at 159/79 mm Hg. A CT of the brain showed a hemorrhagic stroke within the left occipital lobe and posterior parietal lobe with a midline shift of 8 mm. We don’t know if the aspirin use contributed to the hemorrhagic event or if it was a sequela of PRES.

The patient died 4 days later.

THE TAKEAWAY

PRES is a neurotoxic condition that causes headache, seizures, and vision loss. Most patients will present with elevated blood pressure and imaging studies will reveal a specific pattern of vasogenic edema that is predominately found in the parietal and occipital regions.

Treating the hypertension may result in a more favorable recovery. Normotensive patients are harder to treat because there is no specific therapy for PRES. Follow-up imaging may help to assess the resolution of the syndrome.

THE CASE

A 55-year-old woman presented to the emergency department (ED) with a bifrontal headache that she’d had for one day. She also had blurred vision and was vomiting shortly before coming to the hospital. The patient had no history of hypertension, migraine headaches, seizure disorder, autoimmune disorders, or cerebrovascular disease.

Her vital signs, including a blood pressure of 114/63 mm Hg, were normal, but a physical examination revealed subjective vision loss. She was only able to see objects moving on a horizontal plane. Her finger-to-nose exam, pupillary reflexes, and extra-ocular movements were normal, but peripheral vision was limited on her left side. No other neurologic deficits were noted.

The patient was admitted to the hospital and most of her laboratory work-up was normal, including a basic metabolic panel, complete blood count, coagulation studies, brain natriuretic peptide test, and cardiac enzymes. Her white blood cell count was 19,700/mcL, but no source of infection was found. A computed tomography (CT) scan of her head without contrast showed low-density, patchy areas in the subcortical regions of the parietal and occipital lobes bilaterally (FIGURE 1, arrows), with relative sparing of the cortex.

THE DIAGNOSIS

Based on our patient’s presentation and radiologic findings, we made a diagnosis of posterior reversible encephalopathy syndrome (PRES). However, because we could not rule out an ischemic cerebrovascular event at the time of presentation, we started the patient on aspirin and clopidogrel 75 mg to prevent possible future ischemic events. The next day, we ordered magnetic resonance imaging (MRI) of the head and neck, which documented the edema and confirmed the diagnosis of PRES (FIGURE 2).

DISCUSSION

PRES is a neurotoxic state associated with a unique pattern of brain vasogenic edema seen on CT or MRI. The edema is often widespread but is predominantly found in the parietal and occipital regions.¹ PRES is seen in patients with a variety of conditions, including hypertension and bone marrow or organ transplantation, as well as in those receiving immunosuppressive or cytotoxic medications.¹ Patients with PRES typically present with headaches and seizures.² Visual abnormalities (most commonly cortical blindness), occur in 15% to 20% of patients with PRES.^2-4

Hinchey et al³ first described reversible posterior leukoencephalopathy syndrome (which later became known as PRES) in 1996. Most of the 15 patients included in this original report had a history of hypertension or immunosuppression. These cases were associated with cerebral edema in portions of the posterior cerebral white matter. It is thought that hypertension alters the blood-brain barrier and causes the acute changes that occur in PRES.³

Besides hypertension and immunosuppression, the risk factors most commonly associated with PRES include preeclampsia/eclampsia; sepsis, particularly due to grampositive organisms; Wegener’s granulomatosis, scleroderma, and polyarteritis nodosa; cancer chemotherapy; bone marrow or stem cell transplantation; and renal disease.^1,4-6

Although a clear cause of PRES has not yet been established, researchers have proposed 2 theories. The first postulates that a sudden increase in systemic blood pressure causes vasoconstriction, which leads to ischemia and edema.^1-4,7,8 However, several studies have also described cases of PRES in patients with mild elevations in blood pressure,^1,5-7 and mild edema has been observed even in normotensive patients^1,5 (as was the case with our patient).

The second theory links PRES to the loss of brain autoregulation, a function that maintains steady blood flow when blood pressure fluctuates.⁶ A loss of this regulatory mechanism causes endothelial dysfunction, capillary leakage, and disruption in the blood-brain barrier.^1,2,4,6-8 These changes then lead to cerebral vasodilatation and edema.² Immunotherapy has also been associated with increased endothelial dysfunction.²

The evidence on the link between the severity of PRES and clinical outcomes is conflicting.

One study that followed 113 PRES patients over 6 years did not find an association between the severity of clinical presentation and the extent of vasogenic edema found on imaging studies.⁵ Of these 113 patients, 69 had PRES primarily due to hypertension, and 21 were receiving cytotoxic medications.⁵ In contrast, a larger retrospective study that followed patients with PRES for 12 years found that severe cases, which included patients with severe cerebral edema and altered mental status, had poor outcomes.⁴ Small studies have reported that 14% of patients with PRES develop cerebral hemorrhage.⁸

When to suspect this condition. PRES should be part of the differential diagnosis for any patient who presents with headache and vision loss. It is important to distinguish PRES from an acute cerebrovascular accident (CVA) because the 2 conditions are managed differently.² In addition, PRES lesions can be misdiagnosed as tumors, especially in a patient with a history of malignant disease in whom the condition appears after chemotherapy.⁹

Treatment targets the underlying causes

Treatment options for PRES are limited. Hypertension in a patient with PRES requires prompt intervention to avoid progression of the disease.² The use of intravenous (IV) calcium-channel blockers or IV beta-blockers for these patients is common.^2,8

Patients with seizures should be treated with anticonvulsant medication, but longterm antiepileptic treatment usually is not required.² Patients who take immunosuppressant or cytotoxic drugs should stop them indefinitely upon presenting with PRES.²

PRES lesions can be misdiagnosed as tumors—especially in patients with a history of malignant disease who have undergone chemotherapy.

For a pregnant woman with preeclampsia/eclampsia, delivery of the placenta, which is considered to be the cause of PRES in these cases, is curative.¹ However, women can develop PRES several weeks after delivery.¹

In most cases, the symptoms associated with PRES will resolve once treatment is initiated, and neurologic recovery can be expected within 2 weeks.²

Our patient regained her sight the following morning and was discharged home 2 days after admission. Her blood pressure remained normal. She returned to the hospital unresponsive the day after she had been discharged. Family members stated that she had taken 15 packets of an aspirin/caffeine combination to control a new headache.

Her blood pressure was elevated at 159/79 mm Hg. A CT of the brain showed a hemorrhagic stroke within the left occipital lobe and posterior parietal lobe with a midline shift of 8 mm. We don’t know if the aspirin use contributed to the hemorrhagic event or if it was a sequela of PRES.

The patient died 4 days later.

THE TAKEAWAY

PRES is a neurotoxic condition that causes headache, seizures, and vision loss. Most patients will present with elevated blood pressure and imaging studies will reveal a specific pattern of vasogenic edema that is predominately found in the parietal and occipital regions.

Treating the hypertension may result in a more favorable recovery. Normotensive patients are harder to treat because there is no specific therapy for PRES. Follow-up imaging may help to assess the resolution of the syndrome.

THE CASE

A 55-year-old woman presented to the emergency department (ED) with a bifrontal headache that she’d had for one day. She also had blurred vision and was vomiting shortly before coming to the hospital. The patient had no history of hypertension, migraine headaches, seizure disorder, autoimmune disorders, or cerebrovascular disease.

Her vital signs, including a blood pressure of 114/63 mm Hg, were normal, but a physical examination revealed subjective vision loss. She was only able to see objects moving on a horizontal plane. Her finger-to-nose exam, pupillary reflexes, and extra-ocular movements were normal, but peripheral vision was limited on her left side. No other neurologic deficits were noted.

The patient was admitted to the hospital and most of her laboratory work-up was normal, including a basic metabolic panel, complete blood count, coagulation studies, brain natriuretic peptide test, and cardiac enzymes. Her white blood cell count was 19,700/mcL, but no source of infection was found. A computed tomography (CT) scan of her head without contrast showed low-density, patchy areas in the subcortical regions of the parietal and occipital lobes bilaterally (FIGURE 1, arrows), with relative sparing of the cortex.

THE DIAGNOSIS

Based on our patient’s presentation and radiologic findings, we made a diagnosis of posterior reversible encephalopathy syndrome (PRES). However, because we could not rule out an ischemic cerebrovascular event at the time of presentation, we started the patient on aspirin and clopidogrel 75 mg to prevent possible future ischemic events. The next day, we ordered magnetic resonance imaging (MRI) of the head and neck, which documented the edema and confirmed the diagnosis of PRES (FIGURE 2).

DISCUSSION

PRES is a neurotoxic state associated with a unique pattern of brain vasogenic edema seen on CT or MRI. The edema is often widespread but is predominantly found in the parietal and occipital regions.¹ PRES is seen in patients with a variety of conditions, including hypertension and bone marrow or organ transplantation, as well as in those receiving immunosuppressive or cytotoxic medications.¹ Patients with PRES typically present with headaches and seizures.² Visual abnormalities (most commonly cortical blindness), occur in 15% to 20% of patients with PRES.^2-4

Hinchey et al³ first described reversible posterior leukoencephalopathy syndrome (which later became known as PRES) in 1996. Most of the 15 patients included in this original report had a history of hypertension or immunosuppression. These cases were associated with cerebral edema in portions of the posterior cerebral white matter. It is thought that hypertension alters the blood-brain barrier and causes the acute changes that occur in PRES.³

Besides hypertension and immunosuppression, the risk factors most commonly associated with PRES include preeclampsia/eclampsia; sepsis, particularly due to grampositive organisms; Wegener’s granulomatosis, scleroderma, and polyarteritis nodosa; cancer chemotherapy; bone marrow or stem cell transplantation; and renal disease.^1,4-6

Although a clear cause of PRES has not yet been established, researchers have proposed 2 theories. The first postulates that a sudden increase in systemic blood pressure causes vasoconstriction, which leads to ischemia and edema.^1-4,7,8 However, several studies have also described cases of PRES in patients with mild elevations in blood pressure,^1,5-7 and mild edema has been observed even in normotensive patients^1,5 (as was the case with our patient).

The second theory links PRES to the loss of brain autoregulation, a function that maintains steady blood flow when blood pressure fluctuates.⁶ A loss of this regulatory mechanism causes endothelial dysfunction, capillary leakage, and disruption in the blood-brain barrier.^1,2,4,6-8 These changes then lead to cerebral vasodilatation and edema.² Immunotherapy has also been associated with increased endothelial dysfunction.²

The evidence on the link between the severity of PRES and clinical outcomes is conflicting.

One study that followed 113 PRES patients over 6 years did not find an association between the severity of clinical presentation and the extent of vasogenic edema found on imaging studies.⁵ Of these 113 patients, 69 had PRES primarily due to hypertension, and 21 were receiving cytotoxic medications.⁵ In contrast, a larger retrospective study that followed patients with PRES for 12 years found that severe cases, which included patients with severe cerebral edema and altered mental status, had poor outcomes.⁴ Small studies have reported that 14% of patients with PRES develop cerebral hemorrhage.⁸

When to suspect this condition. PRES should be part of the differential diagnosis for any patient who presents with headache and vision loss. It is important to distinguish PRES from an acute cerebrovascular accident (CVA) because the 2 conditions are managed differently.² In addition, PRES lesions can be misdiagnosed as tumors, especially in a patient with a history of malignant disease in whom the condition appears after chemotherapy.⁹

Treatment targets the underlying causes

Treatment options for PRES are limited. Hypertension in a patient with PRES requires prompt intervention to avoid progression of the disease.² The use of intravenous (IV) calcium-channel blockers or IV beta-blockers for these patients is common.^2,8

Patients with seizures should be treated with anticonvulsant medication, but longterm antiepileptic treatment usually is not required.² Patients who take immunosuppressant or cytotoxic drugs should stop them indefinitely upon presenting with PRES.²

PRES lesions can be misdiagnosed as tumors—especially in patients with a history of malignant disease who have undergone chemotherapy.

For a pregnant woman with preeclampsia/eclampsia, delivery of the placenta, which is considered to be the cause of PRES in these cases, is curative.¹ However, women can develop PRES several weeks after delivery.¹

In most cases, the symptoms associated with PRES will resolve once treatment is initiated, and neurologic recovery can be expected within 2 weeks.²

Our patient regained her sight the following morning and was discharged home 2 days after admission. Her blood pressure remained normal. She returned to the hospital unresponsive the day after she had been discharged. Family members stated that she had taken 15 packets of an aspirin/caffeine combination to control a new headache.

Her blood pressure was elevated at 159/79 mm Hg. A CT of the brain showed a hemorrhagic stroke within the left occipital lobe and posterior parietal lobe with a midline shift of 8 mm. We don’t know if the aspirin use contributed to the hemorrhagic event or if it was a sequela of PRES.

The patient died 4 days later.

THE TAKEAWAY

PRES is a neurotoxic condition that causes headache, seizures, and vision loss. Most patients will present with elevated blood pressure and imaging studies will reveal a specific pattern of vasogenic edema that is predominately found in the parietal and occipital regions.

Treating the hypertension may result in a more favorable recovery. Normotensive patients are harder to treat because there is no specific therapy for PRES. Follow-up imaging may help to assess the resolution of the syndrome.