Applied Evidence

We don’t do enough to protect our patients from the pain of herpes zoster (HZ). Consider:

• Each year in the United States, about 1 million new cases of herpes zoster (HZ) occur.¹ The incidence is estimated at 3 to 4 per 1,000 in the general population,² but climbs to more than 10 per 1000 among people 60 years of age and older.^3,4
• Overall, between 13% and 26% of patients with HZ develop postherpetic neuralgia (PHN), defined as pain that continues for more than 1 month after the rash has healed. Among patients who are 70 or older, however, the likelihood that HZ will progress to PHN is approximately 50%.⁵ (In a study of 7595 patients being treated for HZ or PHN by general practitioners or dermatologists in France, 45% reported pain that was severe or very severe, and 42% reported permanent pain.⁶ )
• Between 10% and 25% of HZ patients develop ocular complications, which have the potential to result in vision loss, facial scarring, or prolonged or permanent pain.⁷ Encephalitis, myelitis, and peripheral nerve palsies are potential complications, as well.

Yet HZ and its complications are largely preventable.

A live attenuated vaccine (Zostavax) received US Food and Drug Administration approval in 2006.⁸ But many patients have not yet heard of it, and many physicians fail to recommend it. (See “Herpes zoster vaccine: Why aren’t more people receiving it?”.)

As a family physician, you can play a key role in reducing the burden of shingles by rapidly identifying and treating HZ, minimizing the risk of prolonged pain, and, notably, by talking to older patients about the benefits of vaccination.

Start antiviral therapy without delay

Several meta-analyses and many (though not all) randomized controlled trials (RCTs) of HZ treatment have demonstrated that prompt antiviral therapy—with oral acyclovir (ACV), valacyclovir (VCV), or famciclovir (FCV)—reduces the duration of acute pain and the likelihood that PHN will develop.^9,10 Without antiviral therapy, up to 45% of patients over the age of 60 experience pain that persists for 6 months to a year. Even with therapy, studies have found that about 20% of patients older than 50 years continued to have pain for 6 months after their rash appeared.¹⁰ Risk factors for PHN include age (>50 years), sex (female), a disseminated rash, a severe pain presentation, and polymerase chain reaction-detectable varicella zoster virus viremia.¹¹

Which agent? What the research reveals

For most people with HZ, any of the 3 antiviral agents can be used, based on physician and patient preference. (See TABLE for treatment guidelines.) Here’s a look at the evidence for each.

Oral ACV has long been the mainstay of treatment for HZ, but its poor bioavailability and the need for 5 daily doses has led to the development of newer antiviral agents.¹² When initiated within 48 to 72 hours of the onset of the rash, ACV has demonstrated clinical benefit. (The value of starting ACV therapy beyond the 72-hour mark has not been established, though treatment should be considered if new lesions are still appearing.)

In 1 meta-analysis of 4 placebo-controlled trials, ACV accelerated resolution of acute pain, with the greatest effect in those older than 50 years.¹³ In a second meta-analysis, treatment with ACV reduced the incidence of PHN at 3 months by 46% (number needed to treat [NNT]=3.2-8).¹²

VCV is well absorbed in the gastrointestinal tract, providing 3- to 5-fold greater bioavailability compared with ACV.¹³ VCV’s efficacy was demonstrated in an RCT in which the researchers conducted an intent-to-treat analysis: Compared with ACV therapy for 7 to 14 days, VCV significantly accelerated the resolution of acute pain, reduced the duration of PHN, and decreased the proportion of patients with pain persisting for more than 6 months (19% vs 26%). The incidence of adverse events was similar in both groups.¹⁴

FCV has broad activity against varicella-zoster virus.¹⁵ In an RCT that evaluated oral FCV in 419 immunocompetent adults (mean age 50 years) with uncomplicated HZ, FCV was well tolerated and accelerated lesion healing compared with placebo. Among those who developed PHN, the pain resolved twice as fast for patients in the FCV group compared with the controls, and the median duration of PHN was reduced by 2 months.¹⁵

TABLE
Antiviral therapy dosing guidelines*

Analgesics and other drugs: What to consider

While antiviral therapy helps to relieve the pain of HZ, several trials have shown that none of the available agents completely alleviates it or routinely prevents the development of PHN. As a result, adjunctive therapy, including pain medication, is often required. But prescribing analgesics to frail elderly patients and those who have comorbidities and take multiple medications is not without risk.

The ability of a tricyclic antidepressant to alleviate pain and prevent PHN when therapy is initiated within 48 hours of the eruption of lesions was tested in a double-blind trial in which 72 patients 60 years of age or older were randomized to amitriptyline 25 mg daily for 90 days or placebo.¹⁶ Antiviral agents were administered according to the preference of the primary physician. At 6 weeks, the pain prevalence—the primary outcome measure—was reduced by about 50% in the amitriptyline group.¹⁶ There is no other evidence to support the use of tricyclics in the acute phase of HZ, however, and concerns about orthostatic hypotension and anticholinergic side effects limit their use, particularly in older patients.

Corticosteroids are sometimes used, too, often in conjunction with antiviral therapy, but there are problems with this approach, as well. One RCT comparing an ACV-prednisone combination with ACV alone in HZ patients over the age of 50 found that patients who received both drugs had faster resolution of acute pain and earlier discontinuation of analgesics.¹⁷ But several serious adverse effects of prednisone were reported in patients in the combination therapy group, despite the fact that individuals with contraindications to corticosteroids were excluded from the study. Overall, there is little evidence to suggest that steroids can be safely used to reduce the incidence or severity of PHN. There is no specific recommendation regarding analgesic therapy for PHN, but physicians often adopt a stepwise approach.

Recommend the shingles vaccine

In view of the toll that shingles often takes, vaccination is the best way to prevent HZ and its complications. In a randomized, double-blind, placebo-controlled study involving 38,546 adults who were 60 years of age or older, researchers demonstrated that cell-mediated immunity to the varicella-zoster virus was boosted by the live attenuated HZ vaccine.¹⁸ The enhanced immunity was associated with a 51% reduction in the incidence of HZ (NNT=58 to prevent 1 case over 3 years), a 66% reduction in the incidence of PHN (NNT=364 to prevent 1 case of PHN over 3 years), and a 61% reduction in disease burden. The vaccine was well tolerated, and injection site reactions were generally mild.¹⁸ Accurate cost-effectiveness analyses of immunization are not available because the duration of vaccine protection is unknown.¹⁹

FIGURE
A unilateral vesicular rash

The Advisory Committee on Immunization Practices (ACIP) recommends routine vaccination to prevent both HZ and PHN in healthy adults who are 60 years of age and older (individuals with primary or acquired immunodeficiencies or patients on immunosuppressive therapies should not be vaccinated), and suggests that practitioners offer the HZ vaccine to appropriate patients at their first visit.⁷ It is not necessary, however, to ask about the patient’s history of varicella or to conduct serologic testing to determine varicella immunity before administering the HZ vaccine.⁷ A history of shingles is not a contraindication, so advise patients who develop HZ to come in for vaccination soon after the rash and pain resolve.

CORRESPONDENCE
Pierre-Olivier Lang, MD, MPH, PhD, University Hospitals of Geneva, Department of Rehabilitation and Geriatrics, Chemin du Pont-Bochet, 3, CH-1226, Thonex-Geneva, Switzerland; [email protected]

We don’t do enough to protect our patients from the pain of herpes zoster (HZ). Consider:

• Each year in the United States, about 1 million new cases of herpes zoster (HZ) occur.¹ The incidence is estimated at 3 to 4 per 1,000 in the general population,² but climbs to more than 10 per 1000 among people 60 years of age and older.^3,4
• Overall, between 13% and 26% of patients with HZ develop postherpetic neuralgia (PHN), defined as pain that continues for more than 1 month after the rash has healed. Among patients who are 70 or older, however, the likelihood that HZ will progress to PHN is approximately 50%.⁵ (In a study of 7595 patients being treated for HZ or PHN by general practitioners or dermatologists in France, 45% reported pain that was severe or very severe, and 42% reported permanent pain.⁶ )
• Between 10% and 25% of HZ patients develop ocular complications, which have the potential to result in vision loss, facial scarring, or prolonged or permanent pain.⁷ Encephalitis, myelitis, and peripheral nerve palsies are potential complications, as well.

Yet HZ and its complications are largely preventable.

A live attenuated vaccine (Zostavax) received US Food and Drug Administration approval in 2006.⁸ But many patients have not yet heard of it, and many physicians fail to recommend it. (See “Herpes zoster vaccine: Why aren’t more people receiving it?”.)

As a family physician, you can play a key role in reducing the burden of shingles by rapidly identifying and treating HZ, minimizing the risk of prolonged pain, and, notably, by talking to older patients about the benefits of vaccination.

Start antiviral therapy without delay

Several meta-analyses and many (though not all) randomized controlled trials (RCTs) of HZ treatment have demonstrated that prompt antiviral therapy—with oral acyclovir (ACV), valacyclovir (VCV), or famciclovir (FCV)—reduces the duration of acute pain and the likelihood that PHN will develop.^9,10 Without antiviral therapy, up to 45% of patients over the age of 60 experience pain that persists for 6 months to a year. Even with therapy, studies have found that about 20% of patients older than 50 years continued to have pain for 6 months after their rash appeared.¹⁰ Risk factors for PHN include age (>50 years), sex (female), a disseminated rash, a severe pain presentation, and polymerase chain reaction-detectable varicella zoster virus viremia.¹¹

Which agent? What the research reveals

For most people with HZ, any of the 3 antiviral agents can be used, based on physician and patient preference. (See TABLE for treatment guidelines.) Here’s a look at the evidence for each.

Oral ACV has long been the mainstay of treatment for HZ, but its poor bioavailability and the need for 5 daily doses has led to the development of newer antiviral agents.¹² When initiated within 48 to 72 hours of the onset of the rash, ACV has demonstrated clinical benefit. (The value of starting ACV therapy beyond the 72-hour mark has not been established, though treatment should be considered if new lesions are still appearing.)

In 1 meta-analysis of 4 placebo-controlled trials, ACV accelerated resolution of acute pain, with the greatest effect in those older than 50 years.¹³ In a second meta-analysis, treatment with ACV reduced the incidence of PHN at 3 months by 46% (number needed to treat [NNT]=3.2-8).¹²

VCV is well absorbed in the gastrointestinal tract, providing 3- to 5-fold greater bioavailability compared with ACV.¹³ VCV’s efficacy was demonstrated in an RCT in which the researchers conducted an intent-to-treat analysis: Compared with ACV therapy for 7 to 14 days, VCV significantly accelerated the resolution of acute pain, reduced the duration of PHN, and decreased the proportion of patients with pain persisting for more than 6 months (19% vs 26%). The incidence of adverse events was similar in both groups.¹⁴

FCV has broad activity against varicella-zoster virus.¹⁵ In an RCT that evaluated oral FCV in 419 immunocompetent adults (mean age 50 years) with uncomplicated HZ, FCV was well tolerated and accelerated lesion healing compared with placebo. Among those who developed PHN, the pain resolved twice as fast for patients in the FCV group compared with the controls, and the median duration of PHN was reduced by 2 months.¹⁵

TABLE
Antiviral therapy dosing guidelines*

Analgesics and other drugs: What to consider

While antiviral therapy helps to relieve the pain of HZ, several trials have shown that none of the available agents completely alleviates it or routinely prevents the development of PHN. As a result, adjunctive therapy, including pain medication, is often required. But prescribing analgesics to frail elderly patients and those who have comorbidities and take multiple medications is not without risk.

The ability of a tricyclic antidepressant to alleviate pain and prevent PHN when therapy is initiated within 48 hours of the eruption of lesions was tested in a double-blind trial in which 72 patients 60 years of age or older were randomized to amitriptyline 25 mg daily for 90 days or placebo.¹⁶ Antiviral agents were administered according to the preference of the primary physician. At 6 weeks, the pain prevalence—the primary outcome measure—was reduced by about 50% in the amitriptyline group.¹⁶ There is no other evidence to support the use of tricyclics in the acute phase of HZ, however, and concerns about orthostatic hypotension and anticholinergic side effects limit their use, particularly in older patients.

Corticosteroids are sometimes used, too, often in conjunction with antiviral therapy, but there are problems with this approach, as well. One RCT comparing an ACV-prednisone combination with ACV alone in HZ patients over the age of 50 found that patients who received both drugs had faster resolution of acute pain and earlier discontinuation of analgesics.¹⁷ But several serious adverse effects of prednisone were reported in patients in the combination therapy group, despite the fact that individuals with contraindications to corticosteroids were excluded from the study. Overall, there is little evidence to suggest that steroids can be safely used to reduce the incidence or severity of PHN. There is no specific recommendation regarding analgesic therapy for PHN, but physicians often adopt a stepwise approach.

Recommend the shingles vaccine

In view of the toll that shingles often takes, vaccination is the best way to prevent HZ and its complications. In a randomized, double-blind, placebo-controlled study involving 38,546 adults who were 60 years of age or older, researchers demonstrated that cell-mediated immunity to the varicella-zoster virus was boosted by the live attenuated HZ vaccine.¹⁸ The enhanced immunity was associated with a 51% reduction in the incidence of HZ (NNT=58 to prevent 1 case over 3 years), a 66% reduction in the incidence of PHN (NNT=364 to prevent 1 case of PHN over 3 years), and a 61% reduction in disease burden. The vaccine was well tolerated, and injection site reactions were generally mild.¹⁸ Accurate cost-effectiveness analyses of immunization are not available because the duration of vaccine protection is unknown.¹⁹

FIGURE
A unilateral vesicular rash

The Advisory Committee on Immunization Practices (ACIP) recommends routine vaccination to prevent both HZ and PHN in healthy adults who are 60 years of age and older (individuals with primary or acquired immunodeficiencies or patients on immunosuppressive therapies should not be vaccinated), and suggests that practitioners offer the HZ vaccine to appropriate patients at their first visit.⁷ It is not necessary, however, to ask about the patient’s history of varicella or to conduct serologic testing to determine varicella immunity before administering the HZ vaccine.⁷ A history of shingles is not a contraindication, so advise patients who develop HZ to come in for vaccination soon after the rash and pain resolve.

CORRESPONDENCE
Pierre-Olivier Lang, MD, MPH, PhD, University Hospitals of Geneva, Department of Rehabilitation and Geriatrics, Chemin du Pont-Bochet, 3, CH-1226, Thonex-Geneva, Switzerland; [email protected]

We don’t do enough to protect our patients from the pain of herpes zoster (HZ). Consider:

• Each year in the United States, about 1 million new cases of herpes zoster (HZ) occur.¹ The incidence is estimated at 3 to 4 per 1,000 in the general population,² but climbs to more than 10 per 1000 among people 60 years of age and older.^3,4
• Overall, between 13% and 26% of patients with HZ develop postherpetic neuralgia (PHN), defined as pain that continues for more than 1 month after the rash has healed. Among patients who are 70 or older, however, the likelihood that HZ will progress to PHN is approximately 50%.⁵ (In a study of 7595 patients being treated for HZ or PHN by general practitioners or dermatologists in France, 45% reported pain that was severe or very severe, and 42% reported permanent pain.⁶ )
• Between 10% and 25% of HZ patients develop ocular complications, which have the potential to result in vision loss, facial scarring, or prolonged or permanent pain.⁷ Encephalitis, myelitis, and peripheral nerve palsies are potential complications, as well.

Yet HZ and its complications are largely preventable.

A live attenuated vaccine (Zostavax) received US Food and Drug Administration approval in 2006.⁸ But many patients have not yet heard of it, and many physicians fail to recommend it. (See “Herpes zoster vaccine: Why aren’t more people receiving it?”.)

As a family physician, you can play a key role in reducing the burden of shingles by rapidly identifying and treating HZ, minimizing the risk of prolonged pain, and, notably, by talking to older patients about the benefits of vaccination.

Start antiviral therapy without delay

Several meta-analyses and many (though not all) randomized controlled trials (RCTs) of HZ treatment have demonstrated that prompt antiviral therapy—with oral acyclovir (ACV), valacyclovir (VCV), or famciclovir (FCV)—reduces the duration of acute pain and the likelihood that PHN will develop.^9,10 Without antiviral therapy, up to 45% of patients over the age of 60 experience pain that persists for 6 months to a year. Even with therapy, studies have found that about 20% of patients older than 50 years continued to have pain for 6 months after their rash appeared.¹⁰ Risk factors for PHN include age (>50 years), sex (female), a disseminated rash, a severe pain presentation, and polymerase chain reaction-detectable varicella zoster virus viremia.¹¹

Which agent? What the research reveals

For most people with HZ, any of the 3 antiviral agents can be used, based on physician and patient preference. (See TABLE for treatment guidelines.) Here’s a look at the evidence for each.

Oral ACV has long been the mainstay of treatment for HZ, but its poor bioavailability and the need for 5 daily doses has led to the development of newer antiviral agents.¹² When initiated within 48 to 72 hours of the onset of the rash, ACV has demonstrated clinical benefit. (The value of starting ACV therapy beyond the 72-hour mark has not been established, though treatment should be considered if new lesions are still appearing.)

In 1 meta-analysis of 4 placebo-controlled trials, ACV accelerated resolution of acute pain, with the greatest effect in those older than 50 years.¹³ In a second meta-analysis, treatment with ACV reduced the incidence of PHN at 3 months by 46% (number needed to treat [NNT]=3.2-8).¹²

VCV is well absorbed in the gastrointestinal tract, providing 3- to 5-fold greater bioavailability compared with ACV.¹³ VCV’s efficacy was demonstrated in an RCT in which the researchers conducted an intent-to-treat analysis: Compared with ACV therapy for 7 to 14 days, VCV significantly accelerated the resolution of acute pain, reduced the duration of PHN, and decreased the proportion of patients with pain persisting for more than 6 months (19% vs 26%). The incidence of adverse events was similar in both groups.¹⁴

FCV has broad activity against varicella-zoster virus.¹⁵ In an RCT that evaluated oral FCV in 419 immunocompetent adults (mean age 50 years) with uncomplicated HZ, FCV was well tolerated and accelerated lesion healing compared with placebo. Among those who developed PHN, the pain resolved twice as fast for patients in the FCV group compared with the controls, and the median duration of PHN was reduced by 2 months.¹⁵

TABLE
Antiviral therapy dosing guidelines*

Analgesics and other drugs: What to consider

While antiviral therapy helps to relieve the pain of HZ, several trials have shown that none of the available agents completely alleviates it or routinely prevents the development of PHN. As a result, adjunctive therapy, including pain medication, is often required. But prescribing analgesics to frail elderly patients and those who have comorbidities and take multiple medications is not without risk.

The ability of a tricyclic antidepressant to alleviate pain and prevent PHN when therapy is initiated within 48 hours of the eruption of lesions was tested in a double-blind trial in which 72 patients 60 years of age or older were randomized to amitriptyline 25 mg daily for 90 days or placebo.¹⁶ Antiviral agents were administered according to the preference of the primary physician. At 6 weeks, the pain prevalence—the primary outcome measure—was reduced by about 50% in the amitriptyline group.¹⁶ There is no other evidence to support the use of tricyclics in the acute phase of HZ, however, and concerns about orthostatic hypotension and anticholinergic side effects limit their use, particularly in older patients.

Corticosteroids are sometimes used, too, often in conjunction with antiviral therapy, but there are problems with this approach, as well. One RCT comparing an ACV-prednisone combination with ACV alone in HZ patients over the age of 50 found that patients who received both drugs had faster resolution of acute pain and earlier discontinuation of analgesics.¹⁷ But several serious adverse effects of prednisone were reported in patients in the combination therapy group, despite the fact that individuals with contraindications to corticosteroids were excluded from the study. Overall, there is little evidence to suggest that steroids can be safely used to reduce the incidence or severity of PHN. There is no specific recommendation regarding analgesic therapy for PHN, but physicians often adopt a stepwise approach.

Recommend the shingles vaccine

In view of the toll that shingles often takes, vaccination is the best way to prevent HZ and its complications. In a randomized, double-blind, placebo-controlled study involving 38,546 adults who were 60 years of age or older, researchers demonstrated that cell-mediated immunity to the varicella-zoster virus was boosted by the live attenuated HZ vaccine.¹⁸ The enhanced immunity was associated with a 51% reduction in the incidence of HZ (NNT=58 to prevent 1 case over 3 years), a 66% reduction in the incidence of PHN (NNT=364 to prevent 1 case of PHN over 3 years), and a 61% reduction in disease burden. The vaccine was well tolerated, and injection site reactions were generally mild.¹⁸ Accurate cost-effectiveness analyses of immunization are not available because the duration of vaccine protection is unknown.¹⁹

FIGURE
A unilateral vesicular rash

The Advisory Committee on Immunization Practices (ACIP) recommends routine vaccination to prevent both HZ and PHN in healthy adults who are 60 years of age and older (individuals with primary or acquired immunodeficiencies or patients on immunosuppressive therapies should not be vaccinated), and suggests that practitioners offer the HZ vaccine to appropriate patients at their first visit.⁷ It is not necessary, however, to ask about the patient’s history of varicella or to conduct serologic testing to determine varicella immunity before administering the HZ vaccine.⁷ A history of shingles is not a contraindication, so advise patients who develop HZ to come in for vaccination soon after the rash and pain resolve.

CORRESPONDENCE
Pierre-Olivier Lang, MD, MPH, PhD, University Hospitals of Geneva, Department of Rehabilitation and Geriatrics, Chemin du Pont-Bochet, 3, CH-1226, Thonex-Geneva, Switzerland; [email protected]

CASE 1: A patient with mildly elevated ALT and AST

John M., a 63-year-old truck driver with a family history of diabetes and arterial hypertension, is complaining of persistent fatigue—again. He has type 2 diabetes and takes metformin and repaglinide, but his blood pressure is normal. Lab tests reveal a recurrent mild elevation of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels of unknown origin; Mr. M. has no history of virus or hepatotoxic drugs, and reports only modest alcohol intake. He is obese, however, with a BMI of 34.5 and a waist circumference of 41 inches.

CASE 2: A patient with abdominal swelling

Anna B., a slim 68-year-old, comes in with 2 acute conditions: About a week ago, she noticed abdominal swelling and low back pain that began suddenly, after she carried a heavy load. She’s taken nonsteroidal anti-inflammatory drugs for 6 days, but the pain has not improved. The patient’s only significant clinical history is a hysterectomy, with oophorectomy, at age 38, and a recurrent elevation of serum transaminase levels that has never been investigated. Examination reveals an important kyphosis, and finger pressure on the vertebral spine or a position change exacerbates the pain. Her abdomen is swollen and tense, with a tympanic sound on the upper abdomen and a dull sound on the lower portion.

If John M. and Anna B. were your patients, would you suspect that they both have advanced liver disease? What diagnostic tests would you order, and how would you manage their care?

Cirrhosis has always been associated with high rates of morbidity and mortality. It is the 12th most common cause of death in the United States;¹ in some parts of the world, its ranking as a cause of death is considerably higher.^2,3 In recent years, however, cirrhosis has become the focus of greater attention both here and abroad, for 2 reasons: The first is the increasing prevalence of viral hepatitis and steatohepatitis, both of which are prominent causes of cirrhosis. The second is the improvement we’ve made in treatment: Not only can we slow the progression of cirrhosis, but in some cases, we can even restore hepatic function.⁴

The key to successful management of cirrhosis lies in spotting subtle signs and symptoms well in advance of the serious complications that can arise down the road. Here’s what to look for.

Early warnings you can’t afford to overlook

While the clinical presentation of a patient with liver cirrhosis is often asymptomatic, serum transminases—included in many standard laboratory tests as part of a routine examination—often provide the first sign of a problem.

Mildly elevated ALT in an asymptomatic patient may be transient and benign, or an indication of chronic liver disease.⁵ In fact, signs suggestive of significant liver disease have been reported in more than 20% of patients with ALT elevation.² But because abnormal ALT values are common and frequently resolve, many primary care physicians pay little attention to this potentially important finding—and miss a key opportunity for early identification and treatment.⁶

Look at other lab values, risk factors, as well

Additional lab values that suggest the possibility of cirrhosis include an elevated AST/ALT ratio, a low platelet count (<150,000/L), elevated alkaline phosphatase, elevated bilirubin (>1.1 mg/dL), low serum albumin (<2.5 g/dL), and decreased prothrombin time (<100%). Potential causes include viral hepatitis, heavy alcohol use, hepatotoxic drugs, steatosis, and steatohepatitis.

The next step for a patient with any of these abnormal values is a thorough medication review and medical history. Identify all prescription and nonprescription drugs the patient is taking, as well as any herbal products and supplements, in search of hepatotoxic agents. Amiodarone and valproic acid, among other drugs, may cause steatosis, and some herbal products—particularly kava kava extract, used to treat anxiety and insomnia—have been linked to hepatitis and even liver failure.⁷ Question the patient about alcohol consumption and a history of conditions associated with liver disease, such as diabetes, hyperlipidemia, and thyroid disorders, as well.

At a minimum, schedule follow-up testing of asymptomatic patients with abnormal laboratory findings in no more than 6 months. Persistent ALT elevation in such patients is most commonly caused by major viruses, alcohol abuse, nonalcoholic fatty liver disease (NAFLD), or nonalcoholic steatohepatitis (NASH).⁸ Nonalcoholic fatty liver is especially likely in patients with clinical and demographic risk factors—those who, like John M., suffer from obesity or diabetes, or both.

Ultrasound yields further information

Further screening should be limited to patients who continue to have abnormal test results for 6 months or more or have multiple risk factors. While biopsy is still considered the gold standard for diagnosing and staging chronic liver disease, it should be considered, according to the American Gastroenterological Association, only if ultrasound and other tests have not been helpful in reaching a diagnosis.⁹

Often, though, ultrasonography aids in diagnosis. In the case of John M., ultrasound revealed an enlarged liver with diffuse echostructural dyshomogeneity and signs of severe steatosis and mild splenomegaly, but no increase in portal vein diameter and no ascites. For asymptomatic patients with cirrhosis or an earlier stage of liver disease, ultrasound at 6-month intervals, combined with blood alpha-fetoprotein measurement, can be used to track disease progression and screen for hepatocellular carcinoma.¹⁰

Newer, noninvasive methods aid in diagnosis

Noninvasive means of evaluating the presence and extent of liver fibrosis and differentiating cirrhosis from noncirrhosis, developed in recent years, have been found to have positive predictive values greater than 85% to 90%.¹¹ Transient elastography (FibroScan, London, England), which assesses liver stiffness, is 1 such method. Although it is often used successfully, however, morbid obesity, small intercostal spaces, and ascites limit the diagnostic capability of this medical device.¹²

Fibrosis can also be detected with the use of 1 or more algorithms—each testing blood samples for a different combination of serum surrogate markers for liver disease. Some widely used algorithms include the APRI (AST-to-platelets ratio index), the Fibrotest (aptoglobin, alpha-2 macroglobulin, apolipoprotein A1, gamma-glutamyl transpeptidase, and bilirubin), the Hepascore (bilirubin, gamma-glutamyl transpeptidase, haluronic acid, alpha-2 macroglobulin, age, sex), and the BARD (BMI, AST/ALT ratio, diabetes).

Hepatologists often use the results of ultrasonography, followed by transient elastography in conjunction with findings from 1 or more of these algorithms, to determine which patients are candidates for liver biopsy.^11,12

Staging is crucial, with or without biopsy

The decision to perform a liver biopsy should be based on a number of factors, including the patient’s age, lifestyle, liver chemistry abnormalities, desire for prognostic information, and associated comorbidities.⁹ Despite the value of biopsy, it is a costly procedure with potentially serious side effects and risks—and not always accepted by patients. In a recent survey of 1177 primary care physicians in France, as many as 59% of patients with chronic hepatitis C refused to undergo liver biopsy; what’s more, 22% of the doctors surveyed shared the patients’ hesitancy.¹³ Whether patients refuse biopsy or it is deemed unnecessary because ultrasound and other noninvasive tests result in a probable diagnosis, staging is necessary, both to guide therapy and to arrive at a prognosis.

Liver enzyme levels reveal little about organ integrity and are not useful for staging. But other parameters (specifically, bilirubin, albumin, and prothrombin time), combined with the presence (or absence) and severity of physical findings such as encephalopathy and ascites, are included in the Child-Pugh classification system ( TABLE 1 ),¹⁴ a widely used system that roughly indicates disease severity.¹⁵

The Model for End-stage Liver Disease (MELD)—and PELD, the pediatric model—use bilirubin, creatinine, and international normalized ratio values to classify disease severity. MELD and PELD scores are considered more accurate than the Child-Pugh score in determining short-term mortality,¹⁶ and are used by the United Network of Organ Sharing (UNOS) for liver allocation. You’ll find a calculator at http://optn.transplant.hrsa.gov/resources/MeldPeldCalculator.asp?index=97.¹⁷

Despite the progress in diagnostic techniques, the life expectancy and quality of life for patients with advanced cirrhosis remains poor. Patients routinely experience fatigue, pruritus, ascites, encephalopathy, and bleeding; dyspepsia and malnutrition are common, as well. Cirrhosis also carries the risk of life-threatening complications, partly due to comorbidities—most notably, osteoporosis, malabsorption, and rheumatic diseases. Liver transplantation has the potential to change the life expectancy of these patients, but because of the extensive waiting lists, candidates for transplant often die before a liver becomes available.

But for many patients who are in stable condition—those with compensated cirrhosis, that is—the prognosis is far more hopeful: In addition to providing standard medical care, including immunization, if necessary, and nutritional counseling, targeted therapy is crucial to slow, or stop, disease progression.

TABLE 1
Child-Pugh: Classifying cirrhosis, predicting survival*

Treatment for cirrhosis depends on the cause

Although primary care physicians can often provide most, or all, of the care for those in stable condition, a specialist may be helpful in determining further testing to identify the underlying cause of the cirrhosis, which is essential to determining the most appropriate treatment. What’s more, research has shown that patients with cirrhosis whose care is managed by a primary care physician and a hepatologist have better outcomes than those who are treated by a primary care doctor alone.¹⁸

What to test for?

Tests to determine the cause of cirrhosis are listed in TABLE 2 . For an individual patient, diagnostic tests would be based on the suspected cause. A patient with a family history of hereditary hemochromatosis would be tested for elevated serum ferritin levels and hepatic iron content on liver biopsy sample; the transferrin saturation index would also be obtained, and the patient might be tested for specific gene mutations. A patient who drinks heavily would be tested for elevated gamma-glutamyl transpeptidase and mean corpuscular volume. For someone with obesity, diabetes, and an enlarged liver, standard lab tests, including high-density lipoprotein (HDL) cholesterol, glucose, and triglycerides, may be sufficient.

Keep in mind, however, that cirrhosis may have more than 1 contributing factor—obesity or chronic alcohol use and a virus, for example; alcohol abuse and metabolic fatty liver; or virus and hemochromatosis. Thus, it may require more than 1 type of treatment.

Alcohol abuse is the cause of 25% of cases of liver cirrhosis, and a contributor to another 25% to 50%.¹⁹ The key treatment here—and an ideal role for a family physician—is to refer the patient to a detoxification and treatment program and provide ongoing monitoring and support. Antiviral treatment may be helpful for a recovering alcoholic who also tests positive for hepatitis B or C virus, but because of potential problems with compliance, some physicians delay therapy until the patient has had at least 6 months of continuous abstinence. Although this is not an absolute criterion, the same period of abstinence may be required before a patient becomes eligible for a liver transplant.

NAFLD/NASH is typically diagnosed on the basis of lab values and physical presentation. For a stable patient, the primary treatment includes lifestyle change—a low-calorie, low-carbohydrate diet and an exercise regimen—and a possible switch to insulin for better glycemic control.

For patients who are not candidates for such targeted treatments, either because their disease is too advanced or they’re unable to tolerate the recommended therapy, numerous pharmaceutical preparations claiming antioxidant or anti-inflammatory properties are available. But only 1—an herbal extract known as silymarin, derived from the milk thistle plant and taken with vitamin E—has been found to have some protective effects.²⁰

TABLE 2
Liver cirrhosis: Common causes, diagnostic tests, and treatments^4,34-38

Address systemic problems along with targeted treatment

Malnutrition is a serious problem for many patients with cirrhosis. Causes range from poor oral intake or malabsorption to ongoing alcohol use, chronic nausea, or early satiety because of compression from ascites. Dental problems that prevent the patient from chewing properly may be a contributing factor, as well.

Regardless of the cause, malnutrition is associated with muscle wasting, hypoalbuminemia, decreased resistance to infections, and variceal bleeding, and addressing it is a key part of treatment. Assess the nutritional status of every patient with cirrhosis, and stress the importance of multivitamin supplementation.²¹ If dental care is needed, take steps to see that the patient receives it.

Nutritional support, however, should be reserved for severely malnourished patients awaiting transplantation.²²

Osteoporosis. Reduced bone formation—the result of vitamin D deficiency, hypoparathyroidism, and hypogonadism—is a well-known complication of end-stage cirrhosis. However, osteopenia may occur in an earlier stage of disease, especially in patients with cholestatic disease and those receiving antiviral therapy. Prescribe bisphosphonates, together with calcium and vitamin D₃, to improve bone mineral density.²³

Diabetes. The relationship between diabetes and cirrhosis is particularly complex, because diabetes can be both a causal factor and a consequence of cirrhosis. Diabetes is common in patients with NASH, and prevalent among those with hepatitis C and hemochromatosis. Multivariate analyses have found that diabetes has an independent negative effect on the progression of liver disease.²⁴

Diet remains the first-line treatment for hyperglycemia, with metformin as the drug of choice if diet alone is unsuccessful. Sulfonylureas can be used, but require caution to avoid hypoglycemia. Glitazones are a newer alternative, but their value in patients with liver cirrhosis has not been studied. However, the use of any oral antidiabetic agent requires extra caution in patients with cirrhosis, and should be avoided in those with advanced liver disease. Although insulin requires intense self-monitoring of serum glucose levels, it is preferable to oral agents for this patient population.²⁵

Managing complications of cirrhosis

Hospital, home, or long-term care? Whether patients with advanced cirrhosis can be maintained at home or require hospitalization or long-term care is best decided in consultation with patient, family, and other members of the health care team. One helpful tool is the Karnofsky Performance Scale Index (http://www.pennmedicine.org/homecare/hcp/elig_worksheets/Karnofsky-Performance-Status.pdf), which scores patients from 0 to 100 based on their functional impairment.²⁶ (Patients with decompensated liver cirrhosis and limited self-sufficiency typically score <50, indicating that they require home health care, hospice, or institutional care.) Whatever the outcome, the patient may need to be reevaluated as the disease progresses and complications occur.

Ascites, the most common complication of cirrhosis,²⁷ is a primary reason for hospitalization, but may be managed on an outpatient basis, depending on the patient presentation. Determining factors include the presence or absence of portal hypertension, impaired albumin synthesis, decreased plasma oncotic pressure, and sodium retention. Diagnosis is based on physical exam and ultrasonography.

Initial treatment for ascites includes salt restriction^28,29 and avoidance of NSAIDs, which promote renal sodium retention, followed by spironolactone (100–400 mg/d). Add furosemide (40-160 mg/d) if the fluid retention does not begin to resolve after 3 to 5 days of treatment. If the condition persists despite maximum tolerable doses of diuretics, large-volume paracentesis to remove transudative fluid (albumin <1 g/dL; serum/ascites albumin gradient >1.1) may be needed. A patient with recurrent or refractory ascites should see a specialist for further evaluation and the possibility of a transjugular intrahepatic portosystemic shunt (TIPS).

Abdominal pain and an ascitic granulocyte count >250/mm³ suggest spontaneous bacterial peritonitis (SBP)—a severe complication of ascites that can result in renal and liver failure. In addition to pain, patients may present with tense ascites and fever, followed by encephalopathy, shock, and increased serum creatinine levels. Hospitalization is required for SBP; therapy includes high-dose albumin and intravenous antibiotics, typically cephalosporin. Long-term prophylaxis with norfloxacin to prevent the recurrence of SBP is indicated.³⁰

If your patient has ascites and is being cared for at home, talk to the patient and his or her family about the importance of a daily weight check. Tell them to contact you if the patient gains more than 4 to 8 lbs within a few days. Frequent electrolyte checks are needed, as well. An albumin infusion is required when serum levels are particularly low, or after large-volume paracentesis.³¹ Patients with SBP or refractory ascites generally have more advanced disease and a poor prognosis.

Portal hypertension/esophageal varices. The main aim of treating portal hypertension is to prevent esophageal variceal bleeding. The appearance of varices should be checked by endoscopy every 2 to 3 years, or yearly for patients at high risk of bleeding. Patients with varices can be managed with nonselective beta-blockers at doses that are sufficient to elicit a 25% reduction in resting heart rate. Those at high risk for bleeding and patients who have already had esophageal bleeding may require endoscopic band ligation.³² TIPS is an alternative for those whose previous treatments have failed.³³

Hepatic encephalopathy. This potentially reversible decrease in neuropsychiatric function mainly affects patients with portal hypertension. Caused by reduced hepatic clearance of gut-deriving neurotoxins, hepatic encephalopathy is associated with a range of signs and symptoms—from subtle personality changes to coma, with flapping tremor as a frequent initial finding. Acid-base and electrolyte disturbances, constipation, infections, gastrointestinal bleeding, and sedatives can precipitate encephalopathy. Hepatic encephalopathy is a diagnosis of exclusion, however, requiring the exclusion of all other etiologies of altered mental status.

Treatment consists of identifying and correcting the precipitating factors, and includes electrolyte correction, colon cleansing, and acidification with lactulose. Dietary protein restriction is no longer advocated, because it may facilitate malnutrition and complications. Oral rifaximine is useful and well tolerated for suppression of intestinal bacterial flora. Venous infusion of branched-chain amino acids or flumazenil may be effective in case of coma.

Fever and sepsis. Infection is a high-risk factor for mortality in patients with cirrhosis, as it can lead to renal and liver failure, variceal bleeding, and hepatic encephalopathy. However, individuals with cirrhosis often do not develop the typical signs and symptoms of infection; leukocytosis may be absent because of severe leukopenia, for instance, and patients may be afebrile.

Thus, the general appearance of systemic illness is an indication for antibiotics, with quinolones and cephalosporins as first-line agents. Infections most commonly involve the urinary tract (25%-55%) or the respiratory tract (20%), or are related to SBP (10%-30%).³³ Hospitalization is suggested in case of poor general health status or the appearance of organ dysfunction.

When medical therapy and other interventions fail to control complications, transplantation is the only alternative. Primary care physicians can play a role here, too, in referring potential candidates for liver transplants to specialists for further consideration.

CASE 1: Resolution

As we’ve already seen, John M.’s ultrasound revealed an enlarged liver. The results led to a probable diagnosis of an advanced form of NASH. Other lab tests indicated that he had poorly controlled diabetes, high triglyceride levels, and—for the first time—a low platelet count. His physician stressed the importance of following a low-calorie, low-carbohydrate diet and exercising regularly, prescribed insulin, and referred the patient to a hepatologist for further noninvasive evaluation of fibrosis and to determine whether liver biopsy was needed.

CASE 2: Resolution

Blood tests revealed that Anna B. had a low platelet count (64,000/mm³), elevated liver enzymes (AST 2× upper limit of normal [ULN], ALT 1.5× ULN, GGT 2.5× ULN), and high gamma-globulins (33.6%) with no monoclonal bands. Ultrasound revealed an enlarged liver with diffuse echostructural dyshomogeneity, portal vein dilatation, and moderate ascites. She also tested positive for HCV and had an HCV-RNA reading of 15×10⁶ IU/mL. No other cause of chronic liver disease emerged. Ms. B.’s physician told her that she had an osteoporotic vertebral fracture—a frequent comorbidity in patients with liver cirrhosis—and decompensated liver cirrhosis from an old HCV infection. He added that her abdomen was distended because of fluid retention. The physician recommended bed rest, prescribed paracetamol (1 g tid) and spironolactone (100 mg/d), and referred the patient to an orthopedist for treatment of the fracture and to a hepatologist to be evaluated for transplantation.

CORRESPONDENCE
Ignazio Grattagliano, MD, Department of Internal Medicine, University Medical School of Bari, P.zza G. Cesare, 11 – 70124, Bari, Italy; [email protected]

CASE 1: A patient with mildly elevated ALT and AST

John M., a 63-year-old truck driver with a family history of diabetes and arterial hypertension, is complaining of persistent fatigue—again. He has type 2 diabetes and takes metformin and repaglinide, but his blood pressure is normal. Lab tests reveal a recurrent mild elevation of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels of unknown origin; Mr. M. has no history of virus or hepatotoxic drugs, and reports only modest alcohol intake. He is obese, however, with a BMI of 34.5 and a waist circumference of 41 inches.

CASE 2: A patient with abdominal swelling

Anna B., a slim 68-year-old, comes in with 2 acute conditions: About a week ago, she noticed abdominal swelling and low back pain that began suddenly, after she carried a heavy load. She’s taken nonsteroidal anti-inflammatory drugs for 6 days, but the pain has not improved. The patient’s only significant clinical history is a hysterectomy, with oophorectomy, at age 38, and a recurrent elevation of serum transaminase levels that has never been investigated. Examination reveals an important kyphosis, and finger pressure on the vertebral spine or a position change exacerbates the pain. Her abdomen is swollen and tense, with a tympanic sound on the upper abdomen and a dull sound on the lower portion.

If John M. and Anna B. were your patients, would you suspect that they both have advanced liver disease? What diagnostic tests would you order, and how would you manage their care?

Cirrhosis has always been associated with high rates of morbidity and mortality. It is the 12th most common cause of death in the United States;¹ in some parts of the world, its ranking as a cause of death is considerably higher.^2,3 In recent years, however, cirrhosis has become the focus of greater attention both here and abroad, for 2 reasons: The first is the increasing prevalence of viral hepatitis and steatohepatitis, both of which are prominent causes of cirrhosis. The second is the improvement we’ve made in treatment: Not only can we slow the progression of cirrhosis, but in some cases, we can even restore hepatic function.⁴

The key to successful management of cirrhosis lies in spotting subtle signs and symptoms well in advance of the serious complications that can arise down the road. Here’s what to look for.

Early warnings you can’t afford to overlook

While the clinical presentation of a patient with liver cirrhosis is often asymptomatic, serum transminases—included in many standard laboratory tests as part of a routine examination—often provide the first sign of a problem.

Mildly elevated ALT in an asymptomatic patient may be transient and benign, or an indication of chronic liver disease.⁵ In fact, signs suggestive of significant liver disease have been reported in more than 20% of patients with ALT elevation.² But because abnormal ALT values are common and frequently resolve, many primary care physicians pay little attention to this potentially important finding—and miss a key opportunity for early identification and treatment.⁶

Look at other lab values, risk factors, as well

Additional lab values that suggest the possibility of cirrhosis include an elevated AST/ALT ratio, a low platelet count (<150,000/L), elevated alkaline phosphatase, elevated bilirubin (>1.1 mg/dL), low serum albumin (<2.5 g/dL), and decreased prothrombin time (<100%). Potential causes include viral hepatitis, heavy alcohol use, hepatotoxic drugs, steatosis, and steatohepatitis.

The next step for a patient with any of these abnormal values is a thorough medication review and medical history. Identify all prescription and nonprescription drugs the patient is taking, as well as any herbal products and supplements, in search of hepatotoxic agents. Amiodarone and valproic acid, among other drugs, may cause steatosis, and some herbal products—particularly kava kava extract, used to treat anxiety and insomnia—have been linked to hepatitis and even liver failure.⁷ Question the patient about alcohol consumption and a history of conditions associated with liver disease, such as diabetes, hyperlipidemia, and thyroid disorders, as well.

At a minimum, schedule follow-up testing of asymptomatic patients with abnormal laboratory findings in no more than 6 months. Persistent ALT elevation in such patients is most commonly caused by major viruses, alcohol abuse, nonalcoholic fatty liver disease (NAFLD), or nonalcoholic steatohepatitis (NASH).⁸ Nonalcoholic fatty liver is especially likely in patients with clinical and demographic risk factors—those who, like John M., suffer from obesity or diabetes, or both.

Ultrasound yields further information

Further screening should be limited to patients who continue to have abnormal test results for 6 months or more or have multiple risk factors. While biopsy is still considered the gold standard for diagnosing and staging chronic liver disease, it should be considered, according to the American Gastroenterological Association, only if ultrasound and other tests have not been helpful in reaching a diagnosis.⁹

Often, though, ultrasonography aids in diagnosis. In the case of John M., ultrasound revealed an enlarged liver with diffuse echostructural dyshomogeneity and signs of severe steatosis and mild splenomegaly, but no increase in portal vein diameter and no ascites. For asymptomatic patients with cirrhosis or an earlier stage of liver disease, ultrasound at 6-month intervals, combined with blood alpha-fetoprotein measurement, can be used to track disease progression and screen for hepatocellular carcinoma.¹⁰

Newer, noninvasive methods aid in diagnosis

Noninvasive means of evaluating the presence and extent of liver fibrosis and differentiating cirrhosis from noncirrhosis, developed in recent years, have been found to have positive predictive values greater than 85% to 90%.¹¹ Transient elastography (FibroScan, London, England), which assesses liver stiffness, is 1 such method. Although it is often used successfully, however, morbid obesity, small intercostal spaces, and ascites limit the diagnostic capability of this medical device.¹²

Fibrosis can also be detected with the use of 1 or more algorithms—each testing blood samples for a different combination of serum surrogate markers for liver disease. Some widely used algorithms include the APRI (AST-to-platelets ratio index), the Fibrotest (aptoglobin, alpha-2 macroglobulin, apolipoprotein A1, gamma-glutamyl transpeptidase, and bilirubin), the Hepascore (bilirubin, gamma-glutamyl transpeptidase, haluronic acid, alpha-2 macroglobulin, age, sex), and the BARD (BMI, AST/ALT ratio, diabetes).

Hepatologists often use the results of ultrasonography, followed by transient elastography in conjunction with findings from 1 or more of these algorithms, to determine which patients are candidates for liver biopsy.^11,12

Staging is crucial, with or without biopsy

The decision to perform a liver biopsy should be based on a number of factors, including the patient’s age, lifestyle, liver chemistry abnormalities, desire for prognostic information, and associated comorbidities.⁹ Despite the value of biopsy, it is a costly procedure with potentially serious side effects and risks—and not always accepted by patients. In a recent survey of 1177 primary care physicians in France, as many as 59% of patients with chronic hepatitis C refused to undergo liver biopsy; what’s more, 22% of the doctors surveyed shared the patients’ hesitancy.¹³ Whether patients refuse biopsy or it is deemed unnecessary because ultrasound and other noninvasive tests result in a probable diagnosis, staging is necessary, both to guide therapy and to arrive at a prognosis.

Liver enzyme levels reveal little about organ integrity and are not useful for staging. But other parameters (specifically, bilirubin, albumin, and prothrombin time), combined with the presence (or absence) and severity of physical findings such as encephalopathy and ascites, are included in the Child-Pugh classification system ( TABLE 1 ),¹⁴ a widely used system that roughly indicates disease severity.¹⁵

The Model for End-stage Liver Disease (MELD)—and PELD, the pediatric model—use bilirubin, creatinine, and international normalized ratio values to classify disease severity. MELD and PELD scores are considered more accurate than the Child-Pugh score in determining short-term mortality,¹⁶ and are used by the United Network of Organ Sharing (UNOS) for liver allocation. You’ll find a calculator at http://optn.transplant.hrsa.gov/resources/MeldPeldCalculator.asp?index=97.¹⁷

Despite the progress in diagnostic techniques, the life expectancy and quality of life for patients with advanced cirrhosis remains poor. Patients routinely experience fatigue, pruritus, ascites, encephalopathy, and bleeding; dyspepsia and malnutrition are common, as well. Cirrhosis also carries the risk of life-threatening complications, partly due to comorbidities—most notably, osteoporosis, malabsorption, and rheumatic diseases. Liver transplantation has the potential to change the life expectancy of these patients, but because of the extensive waiting lists, candidates for transplant often die before a liver becomes available.

But for many patients who are in stable condition—those with compensated cirrhosis, that is—the prognosis is far more hopeful: In addition to providing standard medical care, including immunization, if necessary, and nutritional counseling, targeted therapy is crucial to slow, or stop, disease progression.

TABLE 1
Child-Pugh: Classifying cirrhosis, predicting survival*

Treatment for cirrhosis depends on the cause

Although primary care physicians can often provide most, or all, of the care for those in stable condition, a specialist may be helpful in determining further testing to identify the underlying cause of the cirrhosis, which is essential to determining the most appropriate treatment. What’s more, research has shown that patients with cirrhosis whose care is managed by a primary care physician and a hepatologist have better outcomes than those who are treated by a primary care doctor alone.¹⁸

What to test for?

Tests to determine the cause of cirrhosis are listed in TABLE 2 . For an individual patient, diagnostic tests would be based on the suspected cause. A patient with a family history of hereditary hemochromatosis would be tested for elevated serum ferritin levels and hepatic iron content on liver biopsy sample; the transferrin saturation index would also be obtained, and the patient might be tested for specific gene mutations. A patient who drinks heavily would be tested for elevated gamma-glutamyl transpeptidase and mean corpuscular volume. For someone with obesity, diabetes, and an enlarged liver, standard lab tests, including high-density lipoprotein (HDL) cholesterol, glucose, and triglycerides, may be sufficient.

Keep in mind, however, that cirrhosis may have more than 1 contributing factor—obesity or chronic alcohol use and a virus, for example; alcohol abuse and metabolic fatty liver; or virus and hemochromatosis. Thus, it may require more than 1 type of treatment.

Alcohol abuse is the cause of 25% of cases of liver cirrhosis, and a contributor to another 25% to 50%.¹⁹ The key treatment here—and an ideal role for a family physician—is to refer the patient to a detoxification and treatment program and provide ongoing monitoring and support. Antiviral treatment may be helpful for a recovering alcoholic who also tests positive for hepatitis B or C virus, but because of potential problems with compliance, some physicians delay therapy until the patient has had at least 6 months of continuous abstinence. Although this is not an absolute criterion, the same period of abstinence may be required before a patient becomes eligible for a liver transplant.

NAFLD/NASH is typically diagnosed on the basis of lab values and physical presentation. For a stable patient, the primary treatment includes lifestyle change—a low-calorie, low-carbohydrate diet and an exercise regimen—and a possible switch to insulin for better glycemic control.

For patients who are not candidates for such targeted treatments, either because their disease is too advanced or they’re unable to tolerate the recommended therapy, numerous pharmaceutical preparations claiming antioxidant or anti-inflammatory properties are available. But only 1—an herbal extract known as silymarin, derived from the milk thistle plant and taken with vitamin E—has been found to have some protective effects.²⁰

TABLE 2
Liver cirrhosis: Common causes, diagnostic tests, and treatments^4,34-38

Address systemic problems along with targeted treatment

Malnutrition is a serious problem for many patients with cirrhosis. Causes range from poor oral intake or malabsorption to ongoing alcohol use, chronic nausea, or early satiety because of compression from ascites. Dental problems that prevent the patient from chewing properly may be a contributing factor, as well.

Regardless of the cause, malnutrition is associated with muscle wasting, hypoalbuminemia, decreased resistance to infections, and variceal bleeding, and addressing it is a key part of treatment. Assess the nutritional status of every patient with cirrhosis, and stress the importance of multivitamin supplementation.²¹ If dental care is needed, take steps to see that the patient receives it.

Nutritional support, however, should be reserved for severely malnourished patients awaiting transplantation.²²

Osteoporosis. Reduced bone formation—the result of vitamin D deficiency, hypoparathyroidism, and hypogonadism—is a well-known complication of end-stage cirrhosis. However, osteopenia may occur in an earlier stage of disease, especially in patients with cholestatic disease and those receiving antiviral therapy. Prescribe bisphosphonates, together with calcium and vitamin D₃, to improve bone mineral density.²³

Diabetes. The relationship between diabetes and cirrhosis is particularly complex, because diabetes can be both a causal factor and a consequence of cirrhosis. Diabetes is common in patients with NASH, and prevalent among those with hepatitis C and hemochromatosis. Multivariate analyses have found that diabetes has an independent negative effect on the progression of liver disease.²⁴

Diet remains the first-line treatment for hyperglycemia, with metformin as the drug of choice if diet alone is unsuccessful. Sulfonylureas can be used, but require caution to avoid hypoglycemia. Glitazones are a newer alternative, but their value in patients with liver cirrhosis has not been studied. However, the use of any oral antidiabetic agent requires extra caution in patients with cirrhosis, and should be avoided in those with advanced liver disease. Although insulin requires intense self-monitoring of serum glucose levels, it is preferable to oral agents for this patient population.²⁵

Managing complications of cirrhosis

Hospital, home, or long-term care? Whether patients with advanced cirrhosis can be maintained at home or require hospitalization or long-term care is best decided in consultation with patient, family, and other members of the health care team. One helpful tool is the Karnofsky Performance Scale Index (http://www.pennmedicine.org/homecare/hcp/elig_worksheets/Karnofsky-Performance-Status.pdf), which scores patients from 0 to 100 based on their functional impairment.²⁶ (Patients with decompensated liver cirrhosis and limited self-sufficiency typically score <50, indicating that they require home health care, hospice, or institutional care.) Whatever the outcome, the patient may need to be reevaluated as the disease progresses and complications occur.

Ascites, the most common complication of cirrhosis,²⁷ is a primary reason for hospitalization, but may be managed on an outpatient basis, depending on the patient presentation. Determining factors include the presence or absence of portal hypertension, impaired albumin synthesis, decreased plasma oncotic pressure, and sodium retention. Diagnosis is based on physical exam and ultrasonography.

Initial treatment for ascites includes salt restriction^28,29 and avoidance of NSAIDs, which promote renal sodium retention, followed by spironolactone (100–400 mg/d). Add furosemide (40-160 mg/d) if the fluid retention does not begin to resolve after 3 to 5 days of treatment. If the condition persists despite maximum tolerable doses of diuretics, large-volume paracentesis to remove transudative fluid (albumin <1 g/dL; serum/ascites albumin gradient >1.1) may be needed. A patient with recurrent or refractory ascites should see a specialist for further evaluation and the possibility of a transjugular intrahepatic portosystemic shunt (TIPS).

Abdominal pain and an ascitic granulocyte count >250/mm³ suggest spontaneous bacterial peritonitis (SBP)—a severe complication of ascites that can result in renal and liver failure. In addition to pain, patients may present with tense ascites and fever, followed by encephalopathy, shock, and increased serum creatinine levels. Hospitalization is required for SBP; therapy includes high-dose albumin and intravenous antibiotics, typically cephalosporin. Long-term prophylaxis with norfloxacin to prevent the recurrence of SBP is indicated.³⁰

If your patient has ascites and is being cared for at home, talk to the patient and his or her family about the importance of a daily weight check. Tell them to contact you if the patient gains more than 4 to 8 lbs within a few days. Frequent electrolyte checks are needed, as well. An albumin infusion is required when serum levels are particularly low, or after large-volume paracentesis.³¹ Patients with SBP or refractory ascites generally have more advanced disease and a poor prognosis.

Portal hypertension/esophageal varices. The main aim of treating portal hypertension is to prevent esophageal variceal bleeding. The appearance of varices should be checked by endoscopy every 2 to 3 years, or yearly for patients at high risk of bleeding. Patients with varices can be managed with nonselective beta-blockers at doses that are sufficient to elicit a 25% reduction in resting heart rate. Those at high risk for bleeding and patients who have already had esophageal bleeding may require endoscopic band ligation.³² TIPS is an alternative for those whose previous treatments have failed.³³

Hepatic encephalopathy. This potentially reversible decrease in neuropsychiatric function mainly affects patients with portal hypertension. Caused by reduced hepatic clearance of gut-deriving neurotoxins, hepatic encephalopathy is associated with a range of signs and symptoms—from subtle personality changes to coma, with flapping tremor as a frequent initial finding. Acid-base and electrolyte disturbances, constipation, infections, gastrointestinal bleeding, and sedatives can precipitate encephalopathy. Hepatic encephalopathy is a diagnosis of exclusion, however, requiring the exclusion of all other etiologies of altered mental status.

Treatment consists of identifying and correcting the precipitating factors, and includes electrolyte correction, colon cleansing, and acidification with lactulose. Dietary protein restriction is no longer advocated, because it may facilitate malnutrition and complications. Oral rifaximine is useful and well tolerated for suppression of intestinal bacterial flora. Venous infusion of branched-chain amino acids or flumazenil may be effective in case of coma.

Fever and sepsis. Infection is a high-risk factor for mortality in patients with cirrhosis, as it can lead to renal and liver failure, variceal bleeding, and hepatic encephalopathy. However, individuals with cirrhosis often do not develop the typical signs and symptoms of infection; leukocytosis may be absent because of severe leukopenia, for instance, and patients may be afebrile.

Thus, the general appearance of systemic illness is an indication for antibiotics, with quinolones and cephalosporins as first-line agents. Infections most commonly involve the urinary tract (25%-55%) or the respiratory tract (20%), or are related to SBP (10%-30%).³³ Hospitalization is suggested in case of poor general health status or the appearance of organ dysfunction.

When medical therapy and other interventions fail to control complications, transplantation is the only alternative. Primary care physicians can play a role here, too, in referring potential candidates for liver transplants to specialists for further consideration.

CASE 1: Resolution

As we’ve already seen, John M.’s ultrasound revealed an enlarged liver. The results led to a probable diagnosis of an advanced form of NASH. Other lab tests indicated that he had poorly controlled diabetes, high triglyceride levels, and—for the first time—a low platelet count. His physician stressed the importance of following a low-calorie, low-carbohydrate diet and exercising regularly, prescribed insulin, and referred the patient to a hepatologist for further noninvasive evaluation of fibrosis and to determine whether liver biopsy was needed.

CASE 2: Resolution

Blood tests revealed that Anna B. had a low platelet count (64,000/mm³), elevated liver enzymes (AST 2× upper limit of normal [ULN], ALT 1.5× ULN, GGT 2.5× ULN), and high gamma-globulins (33.6%) with no monoclonal bands. Ultrasound revealed an enlarged liver with diffuse echostructural dyshomogeneity, portal vein dilatation, and moderate ascites. She also tested positive for HCV and had an HCV-RNA reading of 15×10⁶ IU/mL. No other cause of chronic liver disease emerged. Ms. B.’s physician told her that she had an osteoporotic vertebral fracture—a frequent comorbidity in patients with liver cirrhosis—and decompensated liver cirrhosis from an old HCV infection. He added that her abdomen was distended because of fluid retention. The physician recommended bed rest, prescribed paracetamol (1 g tid) and spironolactone (100 mg/d), and referred the patient to an orthopedist for treatment of the fracture and to a hepatologist to be evaluated for transplantation.

CORRESPONDENCE
Ignazio Grattagliano, MD, Department of Internal Medicine, University Medical School of Bari, P.zza G. Cesare, 11 – 70124, Bari, Italy; [email protected]

CASE 1: A patient with mildly elevated ALT and AST

John M., a 63-year-old truck driver with a family history of diabetes and arterial hypertension, is complaining of persistent fatigue—again. He has type 2 diabetes and takes metformin and repaglinide, but his blood pressure is normal. Lab tests reveal a recurrent mild elevation of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels of unknown origin; Mr. M. has no history of virus or hepatotoxic drugs, and reports only modest alcohol intake. He is obese, however, with a BMI of 34.5 and a waist circumference of 41 inches.

CASE 2: A patient with abdominal swelling

Anna B., a slim 68-year-old, comes in with 2 acute conditions: About a week ago, she noticed abdominal swelling and low back pain that began suddenly, after she carried a heavy load. She’s taken nonsteroidal anti-inflammatory drugs for 6 days, but the pain has not improved. The patient’s only significant clinical history is a hysterectomy, with oophorectomy, at age 38, and a recurrent elevation of serum transaminase levels that has never been investigated. Examination reveals an important kyphosis, and finger pressure on the vertebral spine or a position change exacerbates the pain. Her abdomen is swollen and tense, with a tympanic sound on the upper abdomen and a dull sound on the lower portion.

If John M. and Anna B. were your patients, would you suspect that they both have advanced liver disease? What diagnostic tests would you order, and how would you manage their care?

Cirrhosis has always been associated with high rates of morbidity and mortality. It is the 12th most common cause of death in the United States;¹ in some parts of the world, its ranking as a cause of death is considerably higher.^2,3 In recent years, however, cirrhosis has become the focus of greater attention both here and abroad, for 2 reasons: The first is the increasing prevalence of viral hepatitis and steatohepatitis, both of which are prominent causes of cirrhosis. The second is the improvement we’ve made in treatment: Not only can we slow the progression of cirrhosis, but in some cases, we can even restore hepatic function.⁴

The key to successful management of cirrhosis lies in spotting subtle signs and symptoms well in advance of the serious complications that can arise down the road. Here’s what to look for.

Early warnings you can’t afford to overlook

While the clinical presentation of a patient with liver cirrhosis is often asymptomatic, serum transminases—included in many standard laboratory tests as part of a routine examination—often provide the first sign of a problem.

Mildly elevated ALT in an asymptomatic patient may be transient and benign, or an indication of chronic liver disease.⁵ In fact, signs suggestive of significant liver disease have been reported in more than 20% of patients with ALT elevation.² But because abnormal ALT values are common and frequently resolve, many primary care physicians pay little attention to this potentially important finding—and miss a key opportunity for early identification and treatment.⁶

Look at other lab values, risk factors, as well

Additional lab values that suggest the possibility of cirrhosis include an elevated AST/ALT ratio, a low platelet count (<150,000/L), elevated alkaline phosphatase, elevated bilirubin (>1.1 mg/dL), low serum albumin (<2.5 g/dL), and decreased prothrombin time (<100%). Potential causes include viral hepatitis, heavy alcohol use, hepatotoxic drugs, steatosis, and steatohepatitis.

The next step for a patient with any of these abnormal values is a thorough medication review and medical history. Identify all prescription and nonprescription drugs the patient is taking, as well as any herbal products and supplements, in search of hepatotoxic agents. Amiodarone and valproic acid, among other drugs, may cause steatosis, and some herbal products—particularly kava kava extract, used to treat anxiety and insomnia—have been linked to hepatitis and even liver failure.⁷ Question the patient about alcohol consumption and a history of conditions associated with liver disease, such as diabetes, hyperlipidemia, and thyroid disorders, as well.

At a minimum, schedule follow-up testing of asymptomatic patients with abnormal laboratory findings in no more than 6 months. Persistent ALT elevation in such patients is most commonly caused by major viruses, alcohol abuse, nonalcoholic fatty liver disease (NAFLD), or nonalcoholic steatohepatitis (NASH).⁸ Nonalcoholic fatty liver is especially likely in patients with clinical and demographic risk factors—those who, like John M., suffer from obesity or diabetes, or both.

Ultrasound yields further information

Further screening should be limited to patients who continue to have abnormal test results for 6 months or more or have multiple risk factors. While biopsy is still considered the gold standard for diagnosing and staging chronic liver disease, it should be considered, according to the American Gastroenterological Association, only if ultrasound and other tests have not been helpful in reaching a diagnosis.⁹

Often, though, ultrasonography aids in diagnosis. In the case of John M., ultrasound revealed an enlarged liver with diffuse echostructural dyshomogeneity and signs of severe steatosis and mild splenomegaly, but no increase in portal vein diameter and no ascites. For asymptomatic patients with cirrhosis or an earlier stage of liver disease, ultrasound at 6-month intervals, combined with blood alpha-fetoprotein measurement, can be used to track disease progression and screen for hepatocellular carcinoma.¹⁰

Newer, noninvasive methods aid in diagnosis

Noninvasive means of evaluating the presence and extent of liver fibrosis and differentiating cirrhosis from noncirrhosis, developed in recent years, have been found to have positive predictive values greater than 85% to 90%.¹¹ Transient elastography (FibroScan, London, England), which assesses liver stiffness, is 1 such method. Although it is often used successfully, however, morbid obesity, small intercostal spaces, and ascites limit the diagnostic capability of this medical device.¹²

Fibrosis can also be detected with the use of 1 or more algorithms—each testing blood samples for a different combination of serum surrogate markers for liver disease. Some widely used algorithms include the APRI (AST-to-platelets ratio index), the Fibrotest (aptoglobin, alpha-2 macroglobulin, apolipoprotein A1, gamma-glutamyl transpeptidase, and bilirubin), the Hepascore (bilirubin, gamma-glutamyl transpeptidase, haluronic acid, alpha-2 macroglobulin, age, sex), and the BARD (BMI, AST/ALT ratio, diabetes).

Hepatologists often use the results of ultrasonography, followed by transient elastography in conjunction with findings from 1 or more of these algorithms, to determine which patients are candidates for liver biopsy.^11,12

Staging is crucial, with or without biopsy

The decision to perform a liver biopsy should be based on a number of factors, including the patient’s age, lifestyle, liver chemistry abnormalities, desire for prognostic information, and associated comorbidities.⁹ Despite the value of biopsy, it is a costly procedure with potentially serious side effects and risks—and not always accepted by patients. In a recent survey of 1177 primary care physicians in France, as many as 59% of patients with chronic hepatitis C refused to undergo liver biopsy; what’s more, 22% of the doctors surveyed shared the patients’ hesitancy.¹³ Whether patients refuse biopsy or it is deemed unnecessary because ultrasound and other noninvasive tests result in a probable diagnosis, staging is necessary, both to guide therapy and to arrive at a prognosis.

Liver enzyme levels reveal little about organ integrity and are not useful for staging. But other parameters (specifically, bilirubin, albumin, and prothrombin time), combined with the presence (or absence) and severity of physical findings such as encephalopathy and ascites, are included in the Child-Pugh classification system ( TABLE 1 ),¹⁴ a widely used system that roughly indicates disease severity.¹⁵

The Model for End-stage Liver Disease (MELD)—and PELD, the pediatric model—use bilirubin, creatinine, and international normalized ratio values to classify disease severity. MELD and PELD scores are considered more accurate than the Child-Pugh score in determining short-term mortality,¹⁶ and are used by the United Network of Organ Sharing (UNOS) for liver allocation. You’ll find a calculator at http://optn.transplant.hrsa.gov/resources/MeldPeldCalculator.asp?index=97.¹⁷

Despite the progress in diagnostic techniques, the life expectancy and quality of life for patients with advanced cirrhosis remains poor. Patients routinely experience fatigue, pruritus, ascites, encephalopathy, and bleeding; dyspepsia and malnutrition are common, as well. Cirrhosis also carries the risk of life-threatening complications, partly due to comorbidities—most notably, osteoporosis, malabsorption, and rheumatic diseases. Liver transplantation has the potential to change the life expectancy of these patients, but because of the extensive waiting lists, candidates for transplant often die before a liver becomes available.

But for many patients who are in stable condition—those with compensated cirrhosis, that is—the prognosis is far more hopeful: In addition to providing standard medical care, including immunization, if necessary, and nutritional counseling, targeted therapy is crucial to slow, or stop, disease progression.

TABLE 1
Child-Pugh: Classifying cirrhosis, predicting survival*

Treatment for cirrhosis depends on the cause

Although primary care physicians can often provide most, or all, of the care for those in stable condition, a specialist may be helpful in determining further testing to identify the underlying cause of the cirrhosis, which is essential to determining the most appropriate treatment. What’s more, research has shown that patients with cirrhosis whose care is managed by a primary care physician and a hepatologist have better outcomes than those who are treated by a primary care doctor alone.¹⁸

What to test for?

Tests to determine the cause of cirrhosis are listed in TABLE 2 . For an individual patient, diagnostic tests would be based on the suspected cause. A patient with a family history of hereditary hemochromatosis would be tested for elevated serum ferritin levels and hepatic iron content on liver biopsy sample; the transferrin saturation index would also be obtained, and the patient might be tested for specific gene mutations. A patient who drinks heavily would be tested for elevated gamma-glutamyl transpeptidase and mean corpuscular volume. For someone with obesity, diabetes, and an enlarged liver, standard lab tests, including high-density lipoprotein (HDL) cholesterol, glucose, and triglycerides, may be sufficient.

Keep in mind, however, that cirrhosis may have more than 1 contributing factor—obesity or chronic alcohol use and a virus, for example; alcohol abuse and metabolic fatty liver; or virus and hemochromatosis. Thus, it may require more than 1 type of treatment.

Alcohol abuse is the cause of 25% of cases of liver cirrhosis, and a contributor to another 25% to 50%.¹⁹ The key treatment here—and an ideal role for a family physician—is to refer the patient to a detoxification and treatment program and provide ongoing monitoring and support. Antiviral treatment may be helpful for a recovering alcoholic who also tests positive for hepatitis B or C virus, but because of potential problems with compliance, some physicians delay therapy until the patient has had at least 6 months of continuous abstinence. Although this is not an absolute criterion, the same period of abstinence may be required before a patient becomes eligible for a liver transplant.

NAFLD/NASH is typically diagnosed on the basis of lab values and physical presentation. For a stable patient, the primary treatment includes lifestyle change—a low-calorie, low-carbohydrate diet and an exercise regimen—and a possible switch to insulin for better glycemic control.

For patients who are not candidates for such targeted treatments, either because their disease is too advanced or they’re unable to tolerate the recommended therapy, numerous pharmaceutical preparations claiming antioxidant or anti-inflammatory properties are available. But only 1—an herbal extract known as silymarin, derived from the milk thistle plant and taken with vitamin E—has been found to have some protective effects.²⁰

TABLE 2
Liver cirrhosis: Common causes, diagnostic tests, and treatments^4,34-38

Address systemic problems along with targeted treatment

Malnutrition is a serious problem for many patients with cirrhosis. Causes range from poor oral intake or malabsorption to ongoing alcohol use, chronic nausea, or early satiety because of compression from ascites. Dental problems that prevent the patient from chewing properly may be a contributing factor, as well.

Regardless of the cause, malnutrition is associated with muscle wasting, hypoalbuminemia, decreased resistance to infections, and variceal bleeding, and addressing it is a key part of treatment. Assess the nutritional status of every patient with cirrhosis, and stress the importance of multivitamin supplementation.²¹ If dental care is needed, take steps to see that the patient receives it.

Nutritional support, however, should be reserved for severely malnourished patients awaiting transplantation.²²

Osteoporosis. Reduced bone formation—the result of vitamin D deficiency, hypoparathyroidism, and hypogonadism—is a well-known complication of end-stage cirrhosis. However, osteopenia may occur in an earlier stage of disease, especially in patients with cholestatic disease and those receiving antiviral therapy. Prescribe bisphosphonates, together with calcium and vitamin D₃, to improve bone mineral density.²³

Diabetes. The relationship between diabetes and cirrhosis is particularly complex, because diabetes can be both a causal factor and a consequence of cirrhosis. Diabetes is common in patients with NASH, and prevalent among those with hepatitis C and hemochromatosis. Multivariate analyses have found that diabetes has an independent negative effect on the progression of liver disease.²⁴

Diet remains the first-line treatment for hyperglycemia, with metformin as the drug of choice if diet alone is unsuccessful. Sulfonylureas can be used, but require caution to avoid hypoglycemia. Glitazones are a newer alternative, but their value in patients with liver cirrhosis has not been studied. However, the use of any oral antidiabetic agent requires extra caution in patients with cirrhosis, and should be avoided in those with advanced liver disease. Although insulin requires intense self-monitoring of serum glucose levels, it is preferable to oral agents for this patient population.²⁵

Managing complications of cirrhosis

Hospital, home, or long-term care? Whether patients with advanced cirrhosis can be maintained at home or require hospitalization or long-term care is best decided in consultation with patient, family, and other members of the health care team. One helpful tool is the Karnofsky Performance Scale Index (http://www.pennmedicine.org/homecare/hcp/elig_worksheets/Karnofsky-Performance-Status.pdf), which scores patients from 0 to 100 based on their functional impairment.²⁶ (Patients with decompensated liver cirrhosis and limited self-sufficiency typically score <50, indicating that they require home health care, hospice, or institutional care.) Whatever the outcome, the patient may need to be reevaluated as the disease progresses and complications occur.

Ascites, the most common complication of cirrhosis,²⁷ is a primary reason for hospitalization, but may be managed on an outpatient basis, depending on the patient presentation. Determining factors include the presence or absence of portal hypertension, impaired albumin synthesis, decreased plasma oncotic pressure, and sodium retention. Diagnosis is based on physical exam and ultrasonography.

Initial treatment for ascites includes salt restriction^28,29 and avoidance of NSAIDs, which promote renal sodium retention, followed by spironolactone (100–400 mg/d). Add furosemide (40-160 mg/d) if the fluid retention does not begin to resolve after 3 to 5 days of treatment. If the condition persists despite maximum tolerable doses of diuretics, large-volume paracentesis to remove transudative fluid (albumin <1 g/dL; serum/ascites albumin gradient >1.1) may be needed. A patient with recurrent or refractory ascites should see a specialist for further evaluation and the possibility of a transjugular intrahepatic portosystemic shunt (TIPS).

Abdominal pain and an ascitic granulocyte count >250/mm³ suggest spontaneous bacterial peritonitis (SBP)—a severe complication of ascites that can result in renal and liver failure. In addition to pain, patients may present with tense ascites and fever, followed by encephalopathy, shock, and increased serum creatinine levels. Hospitalization is required for SBP; therapy includes high-dose albumin and intravenous antibiotics, typically cephalosporin. Long-term prophylaxis with norfloxacin to prevent the recurrence of SBP is indicated.³⁰

If your patient has ascites and is being cared for at home, talk to the patient and his or her family about the importance of a daily weight check. Tell them to contact you if the patient gains more than 4 to 8 lbs within a few days. Frequent electrolyte checks are needed, as well. An albumin infusion is required when serum levels are particularly low, or after large-volume paracentesis.³¹ Patients with SBP or refractory ascites generally have more advanced disease and a poor prognosis.

Portal hypertension/esophageal varices. The main aim of treating portal hypertension is to prevent esophageal variceal bleeding. The appearance of varices should be checked by endoscopy every 2 to 3 years, or yearly for patients at high risk of bleeding. Patients with varices can be managed with nonselective beta-blockers at doses that are sufficient to elicit a 25% reduction in resting heart rate. Those at high risk for bleeding and patients who have already had esophageal bleeding may require endoscopic band ligation.³² TIPS is an alternative for those whose previous treatments have failed.³³

Hepatic encephalopathy. This potentially reversible decrease in neuropsychiatric function mainly affects patients with portal hypertension. Caused by reduced hepatic clearance of gut-deriving neurotoxins, hepatic encephalopathy is associated with a range of signs and symptoms—from subtle personality changes to coma, with flapping tremor as a frequent initial finding. Acid-base and electrolyte disturbances, constipation, infections, gastrointestinal bleeding, and sedatives can precipitate encephalopathy. Hepatic encephalopathy is a diagnosis of exclusion, however, requiring the exclusion of all other etiologies of altered mental status.

Treatment consists of identifying and correcting the precipitating factors, and includes electrolyte correction, colon cleansing, and acidification with lactulose. Dietary protein restriction is no longer advocated, because it may facilitate malnutrition and complications. Oral rifaximine is useful and well tolerated for suppression of intestinal bacterial flora. Venous infusion of branched-chain amino acids or flumazenil may be effective in case of coma.

Fever and sepsis. Infection is a high-risk factor for mortality in patients with cirrhosis, as it can lead to renal and liver failure, variceal bleeding, and hepatic encephalopathy. However, individuals with cirrhosis often do not develop the typical signs and symptoms of infection; leukocytosis may be absent because of severe leukopenia, for instance, and patients may be afebrile.

Thus, the general appearance of systemic illness is an indication for antibiotics, with quinolones and cephalosporins as first-line agents. Infections most commonly involve the urinary tract (25%-55%) or the respiratory tract (20%), or are related to SBP (10%-30%).³³ Hospitalization is suggested in case of poor general health status or the appearance of organ dysfunction.

When medical therapy and other interventions fail to control complications, transplantation is the only alternative. Primary care physicians can play a role here, too, in referring potential candidates for liver transplants to specialists for further consideration.

CASE 1: Resolution

As we’ve already seen, John M.’s ultrasound revealed an enlarged liver. The results led to a probable diagnosis of an advanced form of NASH. Other lab tests indicated that he had poorly controlled diabetes, high triglyceride levels, and—for the first time—a low platelet count. His physician stressed the importance of following a low-calorie, low-carbohydrate diet and exercising regularly, prescribed insulin, and referred the patient to a hepatologist for further noninvasive evaluation of fibrosis and to determine whether liver biopsy was needed.

CASE 2: Resolution

Blood tests revealed that Anna B. had a low platelet count (64,000/mm³), elevated liver enzymes (AST 2× upper limit of normal [ULN], ALT 1.5× ULN, GGT 2.5× ULN), and high gamma-globulins (33.6%) with no monoclonal bands. Ultrasound revealed an enlarged liver with diffuse echostructural dyshomogeneity, portal vein dilatation, and moderate ascites. She also tested positive for HCV and had an HCV-RNA reading of 15×10⁶ IU/mL. No other cause of chronic liver disease emerged. Ms. B.’s physician told her that she had an osteoporotic vertebral fracture—a frequent comorbidity in patients with liver cirrhosis—and decompensated liver cirrhosis from an old HCV infection. He added that her abdomen was distended because of fluid retention. The physician recommended bed rest, prescribed paracetamol (1 g tid) and spironolactone (100 mg/d), and referred the patient to an orthopedist for treatment of the fracture and to a hepatologist to be evaluated for transplantation.

CORRESPONDENCE
Ignazio Grattagliano, MD, Department of Internal Medicine, University Medical School of Bari, P.zza G. Cesare, 11 – 70124, Bari, Italy; [email protected]

Type 2 diabetes (T2D) is characterized by altered glucose homeostasis, including decreased insulin sensitivity of target tissues, a gradual decline in β-cell insulin production and secretion, and a progressive inability to suppress pancreatic α-cell glucagon secretion.¹ In the past, goals for therapy have focused primarily on insulin secretion, sensitization, and replacement. However, newer T2D medications utilize the incretin gut hormone pathway, a focus of scientific and clinical research for decades.² The so-called insulin effect, known today as the incretin effect³ —ie, greater insulin secretion in response to nutrient ingestion—was identified in 1964 when Elrick et al⁴ demonstrated that orally administered glucose produced a significant and sustained increase in plasma insulin, whereas intravenously administered glucose produced a smaller and transient insulin increase. This finding was paramount in bringing incretin-based therapies to clinical practice.

The 2 most well characterized incretin hormones are glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) ( Figure 1 ). Currently, therapeutic agents, acting as either an incretin mimetic (via GLP-1 analogs) or to inhibit the breakdown of GLP-1 (via dipeptidyl peptidase-4 [DPP-4] inhibitors) are available for treatment.^2,3 Various DPP-4 inhibitors are in development, and 2 are approved by the US Food and Drug Administration (FDA): sitagliptin and the recently approved DPP-4 inhibitor, saxagliptin, are indicated for use in a broad range of patients, including those who are drug naïve or who have inadequate glycemic control on another oral antidiabetic drug (OAD). Both agents are approved as monotherapy and as an add-on to current antihyperglycemic therapy (ie, metformin [MET], sulfonylurea [SU], thiazolidinedione [TZD]), and are also approved as initial combination therapy with MET.^5-7 Another DPP-4 inhibitor, alogliptin, failed to gain approval from the FDA, which indicated the need for additional data (Takeda Pharmaceutical Company Limited) ( Table 1 ).

Figure 1
DPP-4 Inhibitors: Mechanism of Glucose Control²⁸

Post–meal ingestion, GLP-1 and GIP are released from the small intestine and are rapidly degraded by the enzyme DPP-4. Inhibition of DPP-4 prevents the breakdown of GLP-1 and GIP and enhances glucose-stimulated insulin secretion (incretin action). GLP-1 and GIP act on the pancreatic β-cell to increase insulin release. GLP-1 also acts on the α-cell to suppress glucagon release and ultimately suppress hepatic glucose production. Together, the increased cellular glucose uptake and the decreased hepatic glucose output offer physiologic glucose control.

Table 1
DPP-4 Inhibitor Status and Availability

DPP-4 Inhibitors Offer Comprehensive Glycemic Management

Glycemic control efforts should be directed at the multiple pathophysiologic defects of T2D and should involve routine assessment of HbA_1c, monitoring of daily blood glucose values, and treatment with combination regimens that target glucose levels both before and after meals. As monotherapy, DPP-4 inhibitors demonstrate reductions of approximately 0.5% to 0.8% in HbA_1c, with concomitant reductions in postprandial glucose (PPG) and fasting plasma glucose (FPG) (eg, PPG reductions of –43 to –45 mg/dL at 120 minutes of an oral glucose tolerance test [OGTT] and FPG reductions of –9 to –15 mg/dL with saxagliptin monotherapy).^8,9 Durability of that effect has been demonstrated with up to 2 years of treatment.^10,11 DPP-4 inhibitors added to other OADs with complementary mechanisms of action have been proven to be particularly effective in lowering all 3 glycemic parameters in multiple studies ( Table 2 ), regardless of age ( ≤65 and >65 years), gender, race/ethnicity, or body mass index.^12-18 DPP-4 inhibitors are also effective when used concurrently (initial combination) with MET in drug-naïve patients, lowering HbA_1c by as much as 2.5% with concomitant decreases in FPG and PPG ( Figure 2 ) as demonstrated during a 3-hour OGTT.¹⁹

Table 2
DPP-4 Inhibitors in Combination with Other Oral Antidiabetic Drugs*

Figure 2
Changes in Glucose After 24 Weeks of Saxagliptin and Metformin Initial Combination Therapy¹⁹

TOLERABILITY

DPP-4 inhibitors offer strategic advantages with regard to tolerability, including few side effects or drug interactions with commonly used agents as well as simple oral dosing. These agents have demonstrated low risk for hypoglycemia and are weight neutral.^8,20,21 Tolerability profiles have been shown to be similar in patients aged ≤65 years and older patients (aged >65 years) with vildagliptin monotherapy.²² The incidence of discontinuation due to clinical adverse reactions has been shown to be similar to placebo.⁵ Sitagliptin requires a 2-step dose reduction; one for moderate renal impairment and a second for severe impairment and end-stage renal disease patients.⁵ Saxagliptin requires only a 1-step dose reduction if creatine clearance is ≤50 mL/min.⁷ DPP-4 inhibitors in combination with MET, an SU, or a TZD demonstrated favorable tolerability with an overall adverse event profile similar to monotherapy with MET, an SU, or a TZD.^{13,16-19,23-26} Modest increases in hypoglycemic events have been shown in studies with DPP-4 inhibitors in combination with an SU,^17,27 whereas others have shown no increase in hypoglycemic events.¹⁴ When used in combination with MET, gastrointestinal side effects are not increased above those seen with MET alone.¹³

Conclusion

DPP-4 inhibitors are a new class of agents that improve long-term, 24-hour control of HbA_1c, FPG (before meal) levels, and PPG (after meal) levels through decreased DPP-4–mediated degradation of incretin hormones. DPP-4 inhibitors provide a complementary mechanism of action to existing OADs and demonstrate significant efficacy when added to MET, an SU, or a TZD, with a well-tolerated profile, including a low risk for hypoglycemia and weight neutrality. DPP-4 inhibitors have been studied in a broad range of patients and have demonstrated similar efficacy, regardless of age, gender, or race/ethnicity. These agents offer an important addition to the treatment of patients with T2D by providing another mechanism to address the multiple pathophysiologic defects present in this disease.

Disclosure

Dr. Cobble has served on advisory boards for Abbott, AstraZeneca, Bristol-Myers Squibb, and Eli Lilly and Company; is the Chief Medical Officer for Atherotech Cardiodiagnostic Lipid Company; and has served as a lecturer for Pri-Med and the American Diabetes Association.

Acknowledgements

Funding for editorial support for this newsletter was provided by Bristol-Myers Squibb and AstraZeneca. Technical writing and editorial assistance for this newsletter was provided by Trina Ricci, PhD, of Innovex Medical Communications. This newsletter has been edited and peer reviewed by The Journal of Family Practice.

Type 2 diabetes (T2D) is characterized by altered glucose homeostasis, including decreased insulin sensitivity of target tissues, a gradual decline in β-cell insulin production and secretion, and a progressive inability to suppress pancreatic α-cell glucagon secretion.¹ In the past, goals for therapy have focused primarily on insulin secretion, sensitization, and replacement. However, newer T2D medications utilize the incretin gut hormone pathway, a focus of scientific and clinical research for decades.² The so-called insulin effect, known today as the incretin effect³ —ie, greater insulin secretion in response to nutrient ingestion—was identified in 1964 when Elrick et al⁴ demonstrated that orally administered glucose produced a significant and sustained increase in plasma insulin, whereas intravenously administered glucose produced a smaller and transient insulin increase. This finding was paramount in bringing incretin-based therapies to clinical practice.

The 2 most well characterized incretin hormones are glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) ( Figure 1 ). Currently, therapeutic agents, acting as either an incretin mimetic (via GLP-1 analogs) or to inhibit the breakdown of GLP-1 (via dipeptidyl peptidase-4 [DPP-4] inhibitors) are available for treatment.^2,3 Various DPP-4 inhibitors are in development, and 2 are approved by the US Food and Drug Administration (FDA): sitagliptin and the recently approved DPP-4 inhibitor, saxagliptin, are indicated for use in a broad range of patients, including those who are drug naïve or who have inadequate glycemic control on another oral antidiabetic drug (OAD). Both agents are approved as monotherapy and as an add-on to current antihyperglycemic therapy (ie, metformin [MET], sulfonylurea [SU], thiazolidinedione [TZD]), and are also approved as initial combination therapy with MET.^5-7 Another DPP-4 inhibitor, alogliptin, failed to gain approval from the FDA, which indicated the need for additional data (Takeda Pharmaceutical Company Limited) ( Table 1 ).

Figure 1
DPP-4 Inhibitors: Mechanism of Glucose Control²⁸

Post–meal ingestion, GLP-1 and GIP are released from the small intestine and are rapidly degraded by the enzyme DPP-4. Inhibition of DPP-4 prevents the breakdown of GLP-1 and GIP and enhances glucose-stimulated insulin secretion (incretin action). GLP-1 and GIP act on the pancreatic β-cell to increase insulin release. GLP-1 also acts on the α-cell to suppress glucagon release and ultimately suppress hepatic glucose production. Together, the increased cellular glucose uptake and the decreased hepatic glucose output offer physiologic glucose control.

Table 1
DPP-4 Inhibitor Status and Availability

DPP-4 Inhibitors Offer Comprehensive Glycemic Management

Glycemic control efforts should be directed at the multiple pathophysiologic defects of T2D and should involve routine assessment of HbA_1c, monitoring of daily blood glucose values, and treatment with combination regimens that target glucose levels both before and after meals. As monotherapy, DPP-4 inhibitors demonstrate reductions of approximately 0.5% to 0.8% in HbA_1c, with concomitant reductions in postprandial glucose (PPG) and fasting plasma glucose (FPG) (eg, PPG reductions of –43 to –45 mg/dL at 120 minutes of an oral glucose tolerance test [OGTT] and FPG reductions of –9 to –15 mg/dL with saxagliptin monotherapy).^8,9 Durability of that effect has been demonstrated with up to 2 years of treatment.^10,11 DPP-4 inhibitors added to other OADs with complementary mechanisms of action have been proven to be particularly effective in lowering all 3 glycemic parameters in multiple studies ( Table 2 ), regardless of age ( ≤65 and >65 years), gender, race/ethnicity, or body mass index.^12-18 DPP-4 inhibitors are also effective when used concurrently (initial combination) with MET in drug-naïve patients, lowering HbA_1c by as much as 2.5% with concomitant decreases in FPG and PPG ( Figure 2 ) as demonstrated during a 3-hour OGTT.¹⁹

Table 2
DPP-4 Inhibitors in Combination with Other Oral Antidiabetic Drugs*

Figure 2
Changes in Glucose After 24 Weeks of Saxagliptin and Metformin Initial Combination Therapy¹⁹

TOLERABILITY

DPP-4 inhibitors offer strategic advantages with regard to tolerability, including few side effects or drug interactions with commonly used agents as well as simple oral dosing. These agents have demonstrated low risk for hypoglycemia and are weight neutral.^8,20,21 Tolerability profiles have been shown to be similar in patients aged ≤65 years and older patients (aged >65 years) with vildagliptin monotherapy.²² The incidence of discontinuation due to clinical adverse reactions has been shown to be similar to placebo.⁵ Sitagliptin requires a 2-step dose reduction; one for moderate renal impairment and a second for severe impairment and end-stage renal disease patients.⁵ Saxagliptin requires only a 1-step dose reduction if creatine clearance is ≤50 mL/min.⁷ DPP-4 inhibitors in combination with MET, an SU, or a TZD demonstrated favorable tolerability with an overall adverse event profile similar to monotherapy with MET, an SU, or a TZD.^{13,16-19,23-26} Modest increases in hypoglycemic events have been shown in studies with DPP-4 inhibitors in combination with an SU,^17,27 whereas others have shown no increase in hypoglycemic events.¹⁴ When used in combination with MET, gastrointestinal side effects are not increased above those seen with MET alone.¹³

Conclusion

DPP-4 inhibitors are a new class of agents that improve long-term, 24-hour control of HbA_1c, FPG (before meal) levels, and PPG (after meal) levels through decreased DPP-4–mediated degradation of incretin hormones. DPP-4 inhibitors provide a complementary mechanism of action to existing OADs and demonstrate significant efficacy when added to MET, an SU, or a TZD, with a well-tolerated profile, including a low risk for hypoglycemia and weight neutrality. DPP-4 inhibitors have been studied in a broad range of patients and have demonstrated similar efficacy, regardless of age, gender, or race/ethnicity. These agents offer an important addition to the treatment of patients with T2D by providing another mechanism to address the multiple pathophysiologic defects present in this disease.

Disclosure

Dr. Cobble has served on advisory boards for Abbott, AstraZeneca, Bristol-Myers Squibb, and Eli Lilly and Company; is the Chief Medical Officer for Atherotech Cardiodiagnostic Lipid Company; and has served as a lecturer for Pri-Med and the American Diabetes Association.

Acknowledgements

Funding for editorial support for this newsletter was provided by Bristol-Myers Squibb and AstraZeneca. Technical writing and editorial assistance for this newsletter was provided by Trina Ricci, PhD, of Innovex Medical Communications. This newsletter has been edited and peer reviewed by The Journal of Family Practice.

Type 2 diabetes (T2D) is characterized by altered glucose homeostasis, including decreased insulin sensitivity of target tissues, a gradual decline in β-cell insulin production and secretion, and a progressive inability to suppress pancreatic α-cell glucagon secretion.¹ In the past, goals for therapy have focused primarily on insulin secretion, sensitization, and replacement. However, newer T2D medications utilize the incretin gut hormone pathway, a focus of scientific and clinical research for decades.² The so-called insulin effect, known today as the incretin effect³ —ie, greater insulin secretion in response to nutrient ingestion—was identified in 1964 when Elrick et al⁴ demonstrated that orally administered glucose produced a significant and sustained increase in plasma insulin, whereas intravenously administered glucose produced a smaller and transient insulin increase. This finding was paramount in bringing incretin-based therapies to clinical practice.

The 2 most well characterized incretin hormones are glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) ( Figure 1 ). Currently, therapeutic agents, acting as either an incretin mimetic (via GLP-1 analogs) or to inhibit the breakdown of GLP-1 (via dipeptidyl peptidase-4 [DPP-4] inhibitors) are available for treatment.^2,3 Various DPP-4 inhibitors are in development, and 2 are approved by the US Food and Drug Administration (FDA): sitagliptin and the recently approved DPP-4 inhibitor, saxagliptin, are indicated for use in a broad range of patients, including those who are drug naïve or who have inadequate glycemic control on another oral antidiabetic drug (OAD). Both agents are approved as monotherapy and as an add-on to current antihyperglycemic therapy (ie, metformin [MET], sulfonylurea [SU], thiazolidinedione [TZD]), and are also approved as initial combination therapy with MET.^5-7 Another DPP-4 inhibitor, alogliptin, failed to gain approval from the FDA, which indicated the need for additional data (Takeda Pharmaceutical Company Limited) ( Table 1 ).

Figure 1
DPP-4 Inhibitors: Mechanism of Glucose Control²⁸

Post–meal ingestion, GLP-1 and GIP are released from the small intestine and are rapidly degraded by the enzyme DPP-4. Inhibition of DPP-4 prevents the breakdown of GLP-1 and GIP and enhances glucose-stimulated insulin secretion (incretin action). GLP-1 and GIP act on the pancreatic β-cell to increase insulin release. GLP-1 also acts on the α-cell to suppress glucagon release and ultimately suppress hepatic glucose production. Together, the increased cellular glucose uptake and the decreased hepatic glucose output offer physiologic glucose control.

Table 1
DPP-4 Inhibitor Status and Availability

DPP-4 Inhibitors Offer Comprehensive Glycemic Management

Glycemic control efforts should be directed at the multiple pathophysiologic defects of T2D and should involve routine assessment of HbA_1c, monitoring of daily blood glucose values, and treatment with combination regimens that target glucose levels both before and after meals. As monotherapy, DPP-4 inhibitors demonstrate reductions of approximately 0.5% to 0.8% in HbA_1c, with concomitant reductions in postprandial glucose (PPG) and fasting plasma glucose (FPG) (eg, PPG reductions of –43 to –45 mg/dL at 120 minutes of an oral glucose tolerance test [OGTT] and FPG reductions of –9 to –15 mg/dL with saxagliptin monotherapy).^8,9 Durability of that effect has been demonstrated with up to 2 years of treatment.^10,11 DPP-4 inhibitors added to other OADs with complementary mechanisms of action have been proven to be particularly effective in lowering all 3 glycemic parameters in multiple studies ( Table 2 ), regardless of age ( ≤65 and >65 years), gender, race/ethnicity, or body mass index.^12-18 DPP-4 inhibitors are also effective when used concurrently (initial combination) with MET in drug-naïve patients, lowering HbA_1c by as much as 2.5% with concomitant decreases in FPG and PPG ( Figure 2 ) as demonstrated during a 3-hour OGTT.¹⁹

Table 2
DPP-4 Inhibitors in Combination with Other Oral Antidiabetic Drugs*

Figure 2
Changes in Glucose After 24 Weeks of Saxagliptin and Metformin Initial Combination Therapy¹⁹

TOLERABILITY

DPP-4 inhibitors offer strategic advantages with regard to tolerability, including few side effects or drug interactions with commonly used agents as well as simple oral dosing. These agents have demonstrated low risk for hypoglycemia and are weight neutral.^8,20,21 Tolerability profiles have been shown to be similar in patients aged ≤65 years and older patients (aged >65 years) with vildagliptin monotherapy.²² The incidence of discontinuation due to clinical adverse reactions has been shown to be similar to placebo.⁵ Sitagliptin requires a 2-step dose reduction; one for moderate renal impairment and a second for severe impairment and end-stage renal disease patients.⁵ Saxagliptin requires only a 1-step dose reduction if creatine clearance is ≤50 mL/min.⁷ DPP-4 inhibitors in combination with MET, an SU, or a TZD demonstrated favorable tolerability with an overall adverse event profile similar to monotherapy with MET, an SU, or a TZD.^{13,16-19,23-26} Modest increases in hypoglycemic events have been shown in studies with DPP-4 inhibitors in combination with an SU,^17,27 whereas others have shown no increase in hypoglycemic events.¹⁴ When used in combination with MET, gastrointestinal side effects are not increased above those seen with MET alone.¹³

Conclusion

DPP-4 inhibitors are a new class of agents that improve long-term, 24-hour control of HbA_1c, FPG (before meal) levels, and PPG (after meal) levels through decreased DPP-4–mediated degradation of incretin hormones. DPP-4 inhibitors provide a complementary mechanism of action to existing OADs and demonstrate significant efficacy when added to MET, an SU, or a TZD, with a well-tolerated profile, including a low risk for hypoglycemia and weight neutrality. DPP-4 inhibitors have been studied in a broad range of patients and have demonstrated similar efficacy, regardless of age, gender, or race/ethnicity. These agents offer an important addition to the treatment of patients with T2D by providing another mechanism to address the multiple pathophysiologic defects present in this disease.

Disclosure

Dr. Cobble has served on advisory boards for Abbott, AstraZeneca, Bristol-Myers Squibb, and Eli Lilly and Company; is the Chief Medical Officer for Atherotech Cardiodiagnostic Lipid Company; and has served as a lecturer for Pri-Med and the American Diabetes Association.

Acknowledgements

Funding for editorial support for this newsletter was provided by Bristol-Myers Squibb and AstraZeneca. Technical writing and editorial assistance for this newsletter was provided by Trina Ricci, PhD, of Innovex Medical Communications. This newsletter has been edited and peer reviewed by The Journal of Family Practice.

Position statement

The American College of Preventive Medicine (ACPM) maintains that primary care providers should screen all adults for depression and that all primary care providers should have systems in place, either within the primary care setting itself or through collaborations with mental health professionals, to ensure the accurate diagnosis and treatment of this condition. The earliest and best opportunities to identify depression are in the clinics of primary care providers. Thus, the ACPM supports the recommendations of the US Preventive Services Task Force (USPSTF), and further suggests that all primary care practices should have such systems of care in place.

Why a position statement? The rationale

Primary care physicians have already been urged by the USPSTF and other authorities to consider screening adults for depression an essential aspect of care.^1-5 So why is the ACPM issuing a new position statement on the subject? Because, the College believes, controversy over how to apply this mandate in the primary care setting is ongoing. Primary care providers—whether they practice family medicine, internal medicine, obstetrics/gynecology, or are in general practice—need to know what role they should play in screening adults for depression and ensuring adequate diagnosis and treatment.

The USPSTF recommendation

In May 2002, the USPSTF made a category B recommendation (high certainty of moderate net benefit) that adults should be screened for depression in “clinical practices that have systems in place to assure accurate diagnosis, effective treatment, and follow-up” of depression¹ (emphasis added). The less-than-clear aspect of this recommendation is italicized: Just what constitutes the “system” that primary care providers, the first and often the only point of contact adults have with the health care system, are told to have “in place”? And how can they go about providing such systems? That’s what the College has set out to elucidate.

The toll depression takes

Depression is a potentially life-threatening disorder that affects up to 6.7% of the population 18 years of age and older, or approximately 14.8 million Americans, in a given year.⁶ Many people younger than age 18 are also affected. The extensive STAR*D (Sequenced Treatment Alternatives to Relieve Depression) study of outpatients with depression found that nearly 40% of respondents had their first depressive episode before the age of 18.^7,8

The ripple effects

Depression is the leading cause of disability in the United States for individuals between the ages of 15 and 44 years.⁹ But the burden of this illness is not borne only by those diagnosed with the disorder; depression has a serious impact on the patient’s family, caregivers, colleagues, and society at large.

Medical costs. Depression contributes to a higher morbidity and mortality of other medical conditions. For example, people who have a myocardial infarction (MI) with comorbid depression have worse outcomes than those having an MI without depression. However, if the depressive episode is treated successfully, medical and surgical outcomes improve.¹⁰ Furthermore, studies documenting increased cardiovascular morbidity and mortality in patients with depressive symptoms or major depression suggest that depression is an independent risk factor in the pathophysiologic progression of cardiovascular disease, not just a secondary emotional response to the illness.¹¹

Economic costs. Depression is a leading contributor to lost productivity, not only from worker absenteeism, but also from what is known as “presenteeism”—being physically at work but not fully engaged. Thus, depression may be a leading cause of poor organizational performance. Further, employees diagnosed with depression have a higher incidence and longer duration of both work-related and nonwork-related disabilities.

Screening: Easier than you think

Screening instruments with acceptable sensitivity and specificity are available. These brief, paper-and-pencil instruments can be quickly completed by patients in your waiting room. Some of the most commonly used self-administered measures are the Beck Depression Inventory (BDI), the Center for Epidemiologic Studies Depression Scale, Revised (CES-DR), the Zung Self-Rating Depression Scale, and the Patient Health Questionnaire (PHQ-9).^12-16 These tools take approximately 5 to 10 minutes for patients to complete and do not interfere with clinical practice. You can choose the tests that are appropriate for screening your patient population, and you can also use the same instruments for ongoing monitoring of patients receiving treatment for depression.

Two questions. For an even briefer screen, ask your patients these 2 questions:

• Over the past month, have you felt down, depressed, or hopeless?
• Over the past month, have you felt little interest or pleasure in doing things?

Patients who answer Yes may need more in-depth screening and clinical assessment.

When these 2 questions were tested in a primary care setting with patients not receiving psychotropic drugs, they had a sensitivity of 97% and a specificity of 67%.¹⁷ Other research also shows that simple questions about depression perform as well as longer questionnaires, further proof that screening for depression need not add undue length to the clinical assessment in primary care.¹⁸

Which instrument is best? Selection of a screening measure, whether it be the 2 simple questions noted above or a longer, more comprehensive tool, is the first step in the process of detecting depression in primary care settings. In making your choice, consider characteristics of the population being screened, psychometric properties of the instrument, time required to complete the measure, time required to score the measure, ease of use, and cost. A review of available screening instruments suitable for use by primary care physicians has been published in American Family Physician.¹⁹ The review includes screening measures developed specifically for adolescents, such as the Reynolds Adolescent Depression Scale, and those developed for older adults, such as the Geriatric Depression Scale.

What comes next? Making the diagnosis

Screening tools provide only a preliminary assessment. Elevated scores must be confirmed with diagnostic interviewing. Without proper follow-up, false-positive scores can lead to harmful labeling, unnecessary additional testing, and inappropriate treatment.

The diagnostic interview. Primary care physicians may feel competent to perform the diagnostic interview themselves, or they may refer patients identified by screening to a mental health professional. The interview should determine whether a patient meets the diagnostic criteria for a depressive disorder—including major depressive disorder or dysthymic disorder—found in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR).²⁰ Bear in mind that criteria important for diagnosis, such as duration of symptoms, degree of impairment, and comorbid psychiatric or substance use disorders, are not revealed in the screening instruments.

Differential diagnosis. Depressive symptoms often overlap with medical conditions, such as hypothyroidism, and with other psychiatric illnesses, such as generalized anxiety disorder. That’s why the differential diagnosis is crucial.

Screen for bipolar disorder. Patients meeting criteria for a depressive disorder should be screened for bipolar disorder, because the 2 conditions are managed differently. Screening instruments for bipolar disorder have been less extensively studied than the depression screening instruments described earlier. The Mood Disorder Questionnaire is a brief, easy-to-use, self-report screening instrument for bipolar-spectrum disorders.²¹ This single-page measure screens for a lifetime history (“has there ever been a period of time”) of manic or hypomanic symptoms using 13 yes/no items, as well as 2 items assessing whether several symptoms were experienced during the same time period, and the level of functional impairment associated with such symptoms. As noted earlier, elevated scores on this and other screening instruments must be confirmed with diagnostic interviewing.

Use DSM criteria. Because of the varying clinical manifestations of depression, clinical judgment sometimes must supersede strict adherence to DSM-IV-TR criteria. Nevertheless, reliance on these well-established criteria is generally recommended as the best way to avoid over- or underdiagnosis, billing problems, and legal problems arising from an inaccurate diagnosis or inappropriate use of medications.

Treating depression

Depression is a highly treatable condition with generally good outcomes.²² A variety of antidepressant medications and psychotherapeutic modalities are available. Consensus-based guidelines have been developed to guide clinicians in the evaluation and treatment of depression.²³ Remission—not simply treatment response or an improvement in symptoms—should be the targeted endpoint. STAR*D data revealed that “better but not remitted” patients consistently have a worse prognosis and higher relapse rates than those achieving full remission.^8,24

Why me?

The answer: Primary care providers are the principal contacts for more than 50% of patients with mental illnesses. Approximately 35% of patients seen in primary care meet criteria for some form of depression and 10% suffer from major depression.^25,26 Because individuals with depression use health care more frequently, the prevalence of major depression is 2 to 3 times higher in primary care settings than in the general population.²⁷ Yet, a substantial proportion of primary care patients with major depression go undiagnosed, leading to a dangerous situation in which symptoms may worsen and suicidal ideation can develop.²⁸ That’s why you, as a primary care practitioner, have such an important role to play in assessing, diagnosing, and treating depression.

Making a difference. Interventions initiated in the primary care setting have been shown to be effective for the treatment of depression.²⁹ Findings of the STAR*D study confirm that primary care providers, when given the time, staffing, and reimbursement support, can provide high-quality, appropriate care for patients with depression, especially in uncomplicated cases.⁸

What you need are “systems in place”

The USPSTF recommended screening for depression in “clinical practices that have systems in place to assure accurate diagnosis, effective treatment, and follow-up.” Given the prevalence and gravity of the problem, the ACPM recommends that all primary care practices have such systems in place. These systems may be either:

• within your own practice, using clinical guidelines for the diagnosis and treatment of depression, or
• within an established system of referral to mental health professionals.

If you treat patients with depression within your own practice, keep in mind that using standardized treatments with established efficacy in psychiatric patients has been shown to be more effective than “usual care.”³⁰ If you choose to partner with mental health professionals, you can employ various levels of collaboration. Depending on the size of your practice, you may want to include a mental health professional as an integral part of your staff, or you may prefer to hire a part-time consultant. Another alternative is to establish an ongoing, collaborative relationship with a mental health provider in private practice. In some instances, working with a patient’s employee assistance program may be the best way to ensure that he or she receives treatment and follow-up. Primary care practices serving disadvantaged or impoverished communities may need to develop partnerships with public-sector community mental health centers.

Whatever setup works best for you, the goal is to make sure that your patients with depression have access to ongoing screening, diagnostic, and treatment services. That goal is worthy of your best efforts.

CORRESPONDENCE
Michele Surricchio, MPH, American College of Preventive Medicine, 1307 New York Ave., NW, Suite 200, Washington, DC 20005; [email protected]

Position statement

The American College of Preventive Medicine (ACPM) maintains that primary care providers should screen all adults for depression and that all primary care providers should have systems in place, either within the primary care setting itself or through collaborations with mental health professionals, to ensure the accurate diagnosis and treatment of this condition. The earliest and best opportunities to identify depression are in the clinics of primary care providers. Thus, the ACPM supports the recommendations of the US Preventive Services Task Force (USPSTF), and further suggests that all primary care practices should have such systems of care in place.

Why a position statement? The rationale

Primary care physicians have already been urged by the USPSTF and other authorities to consider screening adults for depression an essential aspect of care.^1-5 So why is the ACPM issuing a new position statement on the subject? Because, the College believes, controversy over how to apply this mandate in the primary care setting is ongoing. Primary care providers—whether they practice family medicine, internal medicine, obstetrics/gynecology, or are in general practice—need to know what role they should play in screening adults for depression and ensuring adequate diagnosis and treatment.

The USPSTF recommendation

In May 2002, the USPSTF made a category B recommendation (high certainty of moderate net benefit) that adults should be screened for depression in “clinical practices that have systems in place to assure accurate diagnosis, effective treatment, and follow-up” of depression¹ (emphasis added). The less-than-clear aspect of this recommendation is italicized: Just what constitutes the “system” that primary care providers, the first and often the only point of contact adults have with the health care system, are told to have “in place”? And how can they go about providing such systems? That’s what the College has set out to elucidate.

The toll depression takes

Depression is a potentially life-threatening disorder that affects up to 6.7% of the population 18 years of age and older, or approximately 14.8 million Americans, in a given year.⁶ Many people younger than age 18 are also affected. The extensive STAR*D (Sequenced Treatment Alternatives to Relieve Depression) study of outpatients with depression found that nearly 40% of respondents had their first depressive episode before the age of 18.^7,8

The ripple effects

Depression is the leading cause of disability in the United States for individuals between the ages of 15 and 44 years.⁹ But the burden of this illness is not borne only by those diagnosed with the disorder; depression has a serious impact on the patient’s family, caregivers, colleagues, and society at large.

Medical costs. Depression contributes to a higher morbidity and mortality of other medical conditions. For example, people who have a myocardial infarction (MI) with comorbid depression have worse outcomes than those having an MI without depression. However, if the depressive episode is treated successfully, medical and surgical outcomes improve.¹⁰ Furthermore, studies documenting increased cardiovascular morbidity and mortality in patients with depressive symptoms or major depression suggest that depression is an independent risk factor in the pathophysiologic progression of cardiovascular disease, not just a secondary emotional response to the illness.¹¹

Economic costs. Depression is a leading contributor to lost productivity, not only from worker absenteeism, but also from what is known as “presenteeism”—being physically at work but not fully engaged. Thus, depression may be a leading cause of poor organizational performance. Further, employees diagnosed with depression have a higher incidence and longer duration of both work-related and nonwork-related disabilities.

Screening: Easier than you think

Screening instruments with acceptable sensitivity and specificity are available. These brief, paper-and-pencil instruments can be quickly completed by patients in your waiting room. Some of the most commonly used self-administered measures are the Beck Depression Inventory (BDI), the Center for Epidemiologic Studies Depression Scale, Revised (CES-DR), the Zung Self-Rating Depression Scale, and the Patient Health Questionnaire (PHQ-9).^12-16 These tools take approximately 5 to 10 minutes for patients to complete and do not interfere with clinical practice. You can choose the tests that are appropriate for screening your patient population, and you can also use the same instruments for ongoing monitoring of patients receiving treatment for depression.

Two questions. For an even briefer screen, ask your patients these 2 questions:

• Over the past month, have you felt down, depressed, or hopeless?
• Over the past month, have you felt little interest or pleasure in doing things?

Patients who answer Yes may need more in-depth screening and clinical assessment.

When these 2 questions were tested in a primary care setting with patients not receiving psychotropic drugs, they had a sensitivity of 97% and a specificity of 67%.¹⁷ Other research also shows that simple questions about depression perform as well as longer questionnaires, further proof that screening for depression need not add undue length to the clinical assessment in primary care.¹⁸

Which instrument is best? Selection of a screening measure, whether it be the 2 simple questions noted above or a longer, more comprehensive tool, is the first step in the process of detecting depression in primary care settings. In making your choice, consider characteristics of the population being screened, psychometric properties of the instrument, time required to complete the measure, time required to score the measure, ease of use, and cost. A review of available screening instruments suitable for use by primary care physicians has been published in American Family Physician.¹⁹ The review includes screening measures developed specifically for adolescents, such as the Reynolds Adolescent Depression Scale, and those developed for older adults, such as the Geriatric Depression Scale.

What comes next? Making the diagnosis

Screening tools provide only a preliminary assessment. Elevated scores must be confirmed with diagnostic interviewing. Without proper follow-up, false-positive scores can lead to harmful labeling, unnecessary additional testing, and inappropriate treatment.

The diagnostic interview. Primary care physicians may feel competent to perform the diagnostic interview themselves, or they may refer patients identified by screening to a mental health professional. The interview should determine whether a patient meets the diagnostic criteria for a depressive disorder—including major depressive disorder or dysthymic disorder—found in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR).²⁰ Bear in mind that criteria important for diagnosis, such as duration of symptoms, degree of impairment, and comorbid psychiatric or substance use disorders, are not revealed in the screening instruments.

Differential diagnosis. Depressive symptoms often overlap with medical conditions, such as hypothyroidism, and with other psychiatric illnesses, such as generalized anxiety disorder. That’s why the differential diagnosis is crucial.

Screen for bipolar disorder. Patients meeting criteria for a depressive disorder should be screened for bipolar disorder, because the 2 conditions are managed differently. Screening instruments for bipolar disorder have been less extensively studied than the depression screening instruments described earlier. The Mood Disorder Questionnaire is a brief, easy-to-use, self-report screening instrument for bipolar-spectrum disorders.²¹ This single-page measure screens for a lifetime history (“has there ever been a period of time”) of manic or hypomanic symptoms using 13 yes/no items, as well as 2 items assessing whether several symptoms were experienced during the same time period, and the level of functional impairment associated with such symptoms. As noted earlier, elevated scores on this and other screening instruments must be confirmed with diagnostic interviewing.

Use DSM criteria. Because of the varying clinical manifestations of depression, clinical judgment sometimes must supersede strict adherence to DSM-IV-TR criteria. Nevertheless, reliance on these well-established criteria is generally recommended as the best way to avoid over- or underdiagnosis, billing problems, and legal problems arising from an inaccurate diagnosis or inappropriate use of medications.

Treating depression

Depression is a highly treatable condition with generally good outcomes.²² A variety of antidepressant medications and psychotherapeutic modalities are available. Consensus-based guidelines have been developed to guide clinicians in the evaluation and treatment of depression.²³ Remission—not simply treatment response or an improvement in symptoms—should be the targeted endpoint. STAR*D data revealed that “better but not remitted” patients consistently have a worse prognosis and higher relapse rates than those achieving full remission.^8,24

Why me?

The answer: Primary care providers are the principal contacts for more than 50% of patients with mental illnesses. Approximately 35% of patients seen in primary care meet criteria for some form of depression and 10% suffer from major depression.^25,26 Because individuals with depression use health care more frequently, the prevalence of major depression is 2 to 3 times higher in primary care settings than in the general population.²⁷ Yet, a substantial proportion of primary care patients with major depression go undiagnosed, leading to a dangerous situation in which symptoms may worsen and suicidal ideation can develop.²⁸ That’s why you, as a primary care practitioner, have such an important role to play in assessing, diagnosing, and treating depression.

Making a difference. Interventions initiated in the primary care setting have been shown to be effective for the treatment of depression.²⁹ Findings of the STAR*D study confirm that primary care providers, when given the time, staffing, and reimbursement support, can provide high-quality, appropriate care for patients with depression, especially in uncomplicated cases.⁸

What you need are “systems in place”

The USPSTF recommended screening for depression in “clinical practices that have systems in place to assure accurate diagnosis, effective treatment, and follow-up.” Given the prevalence and gravity of the problem, the ACPM recommends that all primary care practices have such systems in place. These systems may be either:

• within your own practice, using clinical guidelines for the diagnosis and treatment of depression, or
• within an established system of referral to mental health professionals.

If you treat patients with depression within your own practice, keep in mind that using standardized treatments with established efficacy in psychiatric patients has been shown to be more effective than “usual care.”³⁰ If you choose to partner with mental health professionals, you can employ various levels of collaboration. Depending on the size of your practice, you may want to include a mental health professional as an integral part of your staff, or you may prefer to hire a part-time consultant. Another alternative is to establish an ongoing, collaborative relationship with a mental health provider in private practice. In some instances, working with a patient’s employee assistance program may be the best way to ensure that he or she receives treatment and follow-up. Primary care practices serving disadvantaged or impoverished communities may need to develop partnerships with public-sector community mental health centers.

Whatever setup works best for you, the goal is to make sure that your patients with depression have access to ongoing screening, diagnostic, and treatment services. That goal is worthy of your best efforts.

CORRESPONDENCE
Michele Surricchio, MPH, American College of Preventive Medicine, 1307 New York Ave., NW, Suite 200, Washington, DC 20005; [email protected]

Position statement

The American College of Preventive Medicine (ACPM) maintains that primary care providers should screen all adults for depression and that all primary care providers should have systems in place, either within the primary care setting itself or through collaborations with mental health professionals, to ensure the accurate diagnosis and treatment of this condition. The earliest and best opportunities to identify depression are in the clinics of primary care providers. Thus, the ACPM supports the recommendations of the US Preventive Services Task Force (USPSTF), and further suggests that all primary care practices should have such systems of care in place.

Why a position statement? The rationale

Primary care physicians have already been urged by the USPSTF and other authorities to consider screening adults for depression an essential aspect of care.^1-5 So why is the ACPM issuing a new position statement on the subject? Because, the College believes, controversy over how to apply this mandate in the primary care setting is ongoing. Primary care providers—whether they practice family medicine, internal medicine, obstetrics/gynecology, or are in general practice—need to know what role they should play in screening adults for depression and ensuring adequate diagnosis and treatment.

The USPSTF recommendation

In May 2002, the USPSTF made a category B recommendation (high certainty of moderate net benefit) that adults should be screened for depression in “clinical practices that have systems in place to assure accurate diagnosis, effective treatment, and follow-up” of depression¹ (emphasis added). The less-than-clear aspect of this recommendation is italicized: Just what constitutes the “system” that primary care providers, the first and often the only point of contact adults have with the health care system, are told to have “in place”? And how can they go about providing such systems? That’s what the College has set out to elucidate.

The toll depression takes

Depression is a potentially life-threatening disorder that affects up to 6.7% of the population 18 years of age and older, or approximately 14.8 million Americans, in a given year.⁶ Many people younger than age 18 are also affected. The extensive STAR*D (Sequenced Treatment Alternatives to Relieve Depression) study of outpatients with depression found that nearly 40% of respondents had their first depressive episode before the age of 18.^7,8

The ripple effects

Depression is the leading cause of disability in the United States for individuals between the ages of 15 and 44 years.⁹ But the burden of this illness is not borne only by those diagnosed with the disorder; depression has a serious impact on the patient’s family, caregivers, colleagues, and society at large.

Medical costs. Depression contributes to a higher morbidity and mortality of other medical conditions. For example, people who have a myocardial infarction (MI) with comorbid depression have worse outcomes than those having an MI without depression. However, if the depressive episode is treated successfully, medical and surgical outcomes improve.¹⁰ Furthermore, studies documenting increased cardiovascular morbidity and mortality in patients with depressive symptoms or major depression suggest that depression is an independent risk factor in the pathophysiologic progression of cardiovascular disease, not just a secondary emotional response to the illness.¹¹

Economic costs. Depression is a leading contributor to lost productivity, not only from worker absenteeism, but also from what is known as “presenteeism”—being physically at work but not fully engaged. Thus, depression may be a leading cause of poor organizational performance. Further, employees diagnosed with depression have a higher incidence and longer duration of both work-related and nonwork-related disabilities.

Screening: Easier than you think

Screening instruments with acceptable sensitivity and specificity are available. These brief, paper-and-pencil instruments can be quickly completed by patients in your waiting room. Some of the most commonly used self-administered measures are the Beck Depression Inventory (BDI), the Center for Epidemiologic Studies Depression Scale, Revised (CES-DR), the Zung Self-Rating Depression Scale, and the Patient Health Questionnaire (PHQ-9).^12-16 These tools take approximately 5 to 10 minutes for patients to complete and do not interfere with clinical practice. You can choose the tests that are appropriate for screening your patient population, and you can also use the same instruments for ongoing monitoring of patients receiving treatment for depression.

Two questions. For an even briefer screen, ask your patients these 2 questions:

• Over the past month, have you felt down, depressed, or hopeless?
• Over the past month, have you felt little interest or pleasure in doing things?

Patients who answer Yes may need more in-depth screening and clinical assessment.

When these 2 questions were tested in a primary care setting with patients not receiving psychotropic drugs, they had a sensitivity of 97% and a specificity of 67%.¹⁷ Other research also shows that simple questions about depression perform as well as longer questionnaires, further proof that screening for depression need not add undue length to the clinical assessment in primary care.¹⁸

Which instrument is best? Selection of a screening measure, whether it be the 2 simple questions noted above or a longer, more comprehensive tool, is the first step in the process of detecting depression in primary care settings. In making your choice, consider characteristics of the population being screened, psychometric properties of the instrument, time required to complete the measure, time required to score the measure, ease of use, and cost. A review of available screening instruments suitable for use by primary care physicians has been published in American Family Physician.¹⁹ The review includes screening measures developed specifically for adolescents, such as the Reynolds Adolescent Depression Scale, and those developed for older adults, such as the Geriatric Depression Scale.

What comes next? Making the diagnosis

Screening tools provide only a preliminary assessment. Elevated scores must be confirmed with diagnostic interviewing. Without proper follow-up, false-positive scores can lead to harmful labeling, unnecessary additional testing, and inappropriate treatment.

The diagnostic interview. Primary care physicians may feel competent to perform the diagnostic interview themselves, or they may refer patients identified by screening to a mental health professional. The interview should determine whether a patient meets the diagnostic criteria for a depressive disorder—including major depressive disorder or dysthymic disorder—found in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR).²⁰ Bear in mind that criteria important for diagnosis, such as duration of symptoms, degree of impairment, and comorbid psychiatric or substance use disorders, are not revealed in the screening instruments.

Differential diagnosis. Depressive symptoms often overlap with medical conditions, such as hypothyroidism, and with other psychiatric illnesses, such as generalized anxiety disorder. That’s why the differential diagnosis is crucial.

Screen for bipolar disorder. Patients meeting criteria for a depressive disorder should be screened for bipolar disorder, because the 2 conditions are managed differently. Screening instruments for bipolar disorder have been less extensively studied than the depression screening instruments described earlier. The Mood Disorder Questionnaire is a brief, easy-to-use, self-report screening instrument for bipolar-spectrum disorders.²¹ This single-page measure screens for a lifetime history (“has there ever been a period of time”) of manic or hypomanic symptoms using 13 yes/no items, as well as 2 items assessing whether several symptoms were experienced during the same time period, and the level of functional impairment associated with such symptoms. As noted earlier, elevated scores on this and other screening instruments must be confirmed with diagnostic interviewing.

Use DSM criteria. Because of the varying clinical manifestations of depression, clinical judgment sometimes must supersede strict adherence to DSM-IV-TR criteria. Nevertheless, reliance on these well-established criteria is generally recommended as the best way to avoid over- or underdiagnosis, billing problems, and legal problems arising from an inaccurate diagnosis or inappropriate use of medications.

Treating depression

Depression is a highly treatable condition with generally good outcomes.²² A variety of antidepressant medications and psychotherapeutic modalities are available. Consensus-based guidelines have been developed to guide clinicians in the evaluation and treatment of depression.²³ Remission—not simply treatment response or an improvement in symptoms—should be the targeted endpoint. STAR*D data revealed that “better but not remitted” patients consistently have a worse prognosis and higher relapse rates than those achieving full remission.^8,24

Why me?

The answer: Primary care providers are the principal contacts for more than 50% of patients with mental illnesses. Approximately 35% of patients seen in primary care meet criteria for some form of depression and 10% suffer from major depression.^25,26 Because individuals with depression use health care more frequently, the prevalence of major depression is 2 to 3 times higher in primary care settings than in the general population.²⁷ Yet, a substantial proportion of primary care patients with major depression go undiagnosed, leading to a dangerous situation in which symptoms may worsen and suicidal ideation can develop.²⁸ That’s why you, as a primary care practitioner, have such an important role to play in assessing, diagnosing, and treating depression.

Making a difference. Interventions initiated in the primary care setting have been shown to be effective for the treatment of depression.²⁹ Findings of the STAR*D study confirm that primary care providers, when given the time, staffing, and reimbursement support, can provide high-quality, appropriate care for patients with depression, especially in uncomplicated cases.⁸

What you need are “systems in place”

The USPSTF recommended screening for depression in “clinical practices that have systems in place to assure accurate diagnosis, effective treatment, and follow-up.” Given the prevalence and gravity of the problem, the ACPM recommends that all primary care practices have such systems in place. These systems may be either:

• within your own practice, using clinical guidelines for the diagnosis and treatment of depression, or
• within an established system of referral to mental health professionals.

If you treat patients with depression within your own practice, keep in mind that using standardized treatments with established efficacy in psychiatric patients has been shown to be more effective than “usual care.”³⁰ If you choose to partner with mental health professionals, you can employ various levels of collaboration. Depending on the size of your practice, you may want to include a mental health professional as an integral part of your staff, or you may prefer to hire a part-time consultant. Another alternative is to establish an ongoing, collaborative relationship with a mental health provider in private practice. In some instances, working with a patient’s employee assistance program may be the best way to ensure that he or she receives treatment and follow-up. Primary care practices serving disadvantaged or impoverished communities may need to develop partnerships with public-sector community mental health centers.

Whatever setup works best for you, the goal is to make sure that your patients with depression have access to ongoing screening, diagnostic, and treatment services. That goal is worthy of your best efforts.

CORRESPONDENCE
Michele Surricchio, MPH, American College of Preventive Medicine, 1307 New York Ave., NW, Suite 200, Washington, DC 20005; [email protected]

CASE

Kevin M., a 35-year-old researcher who often works 12- to 14-hour days, is in your office again, following his second trip to the emergency room for bleeding from a protruding varicose vein proximal to his left ankle. He has complained of leg aching and ankle swelling in the past, usually after he’s been on his feet for hours.

A previous ultrasound showed no evidence of deep vein thrombosis (DVT), but did reveal reflux along the left great saphenous vein from the saphenofemoral junction to the ankle. On examination, Kevin’s lower extremity pulses are intact bilaterally, but there are multiple tortuous varicosities branching from the medial thigh to the left medial malleolus, with evidence of hemosiderin hyperpigmentation and dermatitis on the left lower leg. Previous lab studies have been normal, and Kevin has no other chronic conditions. Kevin’s main concern is to minimize “down time” from work, and he asks you about the newer, minimally invasive vein treatments he’s read about on the Internet.

What should you tell him?

Chronic venous insufficiency is a common and costly condition, affecting nearly one-third of the US population.¹ While many people mistakenly think of varicose veins primarily as a cosmetic issue, venous insufficiency often results in painful, even debilitating, signs and symptoms—from swelling and aching legs to skin changes that range from stasis dermatitis to open ulceration.

Venous stripping and ambulatory phlebectomy were long considered the primary means of addressing saphenous venous insufficiency,² and compression therapy with gradient stockings, short stretch bandages, and Unna’s dressings remains a therapeutic cornerstone for every stage of chronic venous disease. Compression has also been shown to dramatically decrease the risk of post-thrombotic syndrome following DVT³ and, when combined with anticoagulation, to prevent DVT after surgery.⁴

Outcomes of surgical treatment of large varicose veins arising from saphenous truncal vein reflux, however, have historically been poor.⁵ This is primarily because of neovascularization and residual refluxing veins. Average failure rates of 25% for saphenous vein stripping and 43% for saphenofemoral junction ligation have been reported.⁶ A subsequent study of 140 patients who underwent saphenous vein stripping found that 20% experienced neovascularization.⁷

In the last decade, a number of minimally invasive procedures have been developed and tested, giving physicians and their patients more choices for medical management of venous insufficiency. A closer look at 3 procedures, which include 2 types of thermal ablation, will help you direct patients like Kevin to the best possible treatment.

2 thermal ablation procedures use only local anesthesia

Collectively termed endovenous thermal ablation (EVTA), radiofrequency ablation (RFA) and endovenous laser treatment (ELT) are similar. Performed under duplex ultrasound visualization, both offer a number of benefits.

Outpatient treatment. EVTA procedures are typically done on an outpatient basis, with only local anesthesia—0.05% to 0.1% lidocaine, injected as a perivenous tumescent solution under ultrasound guidance.^8,9 The physician places and advances an endovenous catheter (for RFA) or a laser fiber (for ELT) into the vein (FIGURE 1). When the catheter or laser is in proper position, the RF generator or laser is activated. Heat, delivered through the tip of the catheter or fiber, ablates the lumen of the vein as the physician withdraws the catheter.

Mild side effects. Temporary discomfort is common with EVTA. Mild bruising, swelling, and nodularity of the treated veins are the most frequent side effects. More serious adverse effects, including skin burns, paresthesias, DVT, and pulmonary embolism, are rare.^10-12

High success rates. Successful treatment of both the great and small saphenous veins and long-term maintenance of vein closure are the norm, with rates ranging from 88% to 100% for both RFA^13,14 and ELT, regardless of vein size.^15-19 Thermal ablation has been shown to be superior to surgery in other ways as well, eliminating the need for general anesthesia and inpatient treatment and resulting in less postoperative pain, a shorter recovery period, an improved quality of life, and lower costs (TABLE).^20-23

TABLE
Endovenous thermal ablation vs standard surgery

FIGURE 1
Endovenous laser treatment

Not for every patient

Despite the usefulness of these minimally invasive procedures, RFA and ELT are not always possible. Veins that are tortuous, too small, or too close to the surface of the skin are not well suited for thermal ablation; in some cases, chemical ablation, discussed below, is the only option. However, some conditions—including pregnancy, active DVT, poor health status with limited mobility, and severe peripheral vascular disease—are contraindications for both thermal and chemical ablation.²⁴

Thermal won’t work? Consider chemical ablation

Endovenous chemoablation (ECA), like thermal ablation, is performed on an outpatient basis without sedation. Hypertonic saline, an osmotic agent, has been used for decades in treating small-caliber varicose veins, or “spider veins.” Detergenttype sclerosants, which we’ll discuss here, are now widely used to treat larger veins, including the great and small saphenous vein trunks and their affected tributaries.

Better suited for tortuous veins. Patients who have had prior stripping often have branches that are too tortuous to access with thermal ablation. In such cases, ECA is the only option. Those who have never undergone vein stripping often need a combination of thermal and chemical ablation, as well, with ECA reserved for saphenous veins that are inaccessible by RFA or laser.

Foamed sclerosants allow visualization. When mixed through a 3-way stopcock with an inert gas such as room air or carbon dioxide, the detergent-type sclerosants produce a micro-foamed preparation that is echogenic, and allows direct visualization with ultrasound (FIGURE 2). The foam displaces blood from within the vein and promotes contact with the venous endothelium, and intense vasospasm results. The sclerosant foam mixture, which makes it possible to watch the effect (and extent) of the foam as it disperses throughout the target vessel, has enhanced the treatment of larger viscosities.^25-26 Pretreatment venous ultrasound mapping—a process in which the patient stands upright while compression is applied to the saphenous veins and their tributaries to test the valves for normal flow—combined with realtime duplex imaging, minimizes the chance of arterial injection, extravasation, and tissue necrosis.²⁵

Serious adverse effects are rare. Hemosiderin hyperpigmentation and pain at the injection site are among the most common side effects of ultrasound-guided sclerotherapy. Serious adverse events, including anaphylaxis, DVT/pulmonary embolus, cutaneous necrosis, transient ischemic attack, and arterial injection, are rare.²⁷

To date, there is only 1 published prospective, randomized study that directly compares foam sclerotherapy with stripping/ligation surgery. The researchers found the results of both treatments to be comparable, but reported that surgery was more expensive than foam sclerotherapy and produced less improvement in quality of life.²⁸ A meta-analysis of treatment for 12,000 limbs concluded that foam sclerotherapy and EVTA were as effective as surgical stripping.²⁹ There are no prospective, randomized studies comparing foam sclerotherapy with thermal ablation.

FIGURE 2
Endovenous chemoablation

Advising Kevin M.: What’s his best bet?

You’ve talked to your patient in the past about vein stripping, but he’s been reluctant to undergo surgery. When you describe the minimally invasive approaches to chronic venous insufficiency with him, however, he asks to see a specialist—and to undergo ultrasound mapping.

You give him a prescription for graduated compression stockings for immediate use and a referral to a phlebologist. Soon after, you receive a report from the specialist, indicating that your patient had a combination of ELT of the refluxing left great saphenous vein and subsequent ultrasound-guided ECA of the saphenous tributaries.

The next time you see Kevin, he reports that his leg no longer aches—and you can see that the bulging, tortuous varicosities and swelling have improved significantly.

Acknowledgments

The authors thank John Hickner, MD, MSc, for his input during the manuscript process.

CORRESPONDENCE
Jeffery P. Schoonover, MD, FAAFP, Vein Clinics of America, 9000 Keystone Crossing, Suite 960, Indianapolis, IN 46240; [email protected]

CASE

Kevin M., a 35-year-old researcher who often works 12- to 14-hour days, is in your office again, following his second trip to the emergency room for bleeding from a protruding varicose vein proximal to his left ankle. He has complained of leg aching and ankle swelling in the past, usually after he’s been on his feet for hours.

A previous ultrasound showed no evidence of deep vein thrombosis (DVT), but did reveal reflux along the left great saphenous vein from the saphenofemoral junction to the ankle. On examination, Kevin’s lower extremity pulses are intact bilaterally, but there are multiple tortuous varicosities branching from the medial thigh to the left medial malleolus, with evidence of hemosiderin hyperpigmentation and dermatitis on the left lower leg. Previous lab studies have been normal, and Kevin has no other chronic conditions. Kevin’s main concern is to minimize “down time” from work, and he asks you about the newer, minimally invasive vein treatments he’s read about on the Internet.

What should you tell him?

Chronic venous insufficiency is a common and costly condition, affecting nearly one-third of the US population.¹ While many people mistakenly think of varicose veins primarily as a cosmetic issue, venous insufficiency often results in painful, even debilitating, signs and symptoms—from swelling and aching legs to skin changes that range from stasis dermatitis to open ulceration.

Venous stripping and ambulatory phlebectomy were long considered the primary means of addressing saphenous venous insufficiency,² and compression therapy with gradient stockings, short stretch bandages, and Unna’s dressings remains a therapeutic cornerstone for every stage of chronic venous disease. Compression has also been shown to dramatically decrease the risk of post-thrombotic syndrome following DVT³ and, when combined with anticoagulation, to prevent DVT after surgery.⁴

Outcomes of surgical treatment of large varicose veins arising from saphenous truncal vein reflux, however, have historically been poor.⁵ This is primarily because of neovascularization and residual refluxing veins. Average failure rates of 25% for saphenous vein stripping and 43% for saphenofemoral junction ligation have been reported.⁶ A subsequent study of 140 patients who underwent saphenous vein stripping found that 20% experienced neovascularization.⁷

In the last decade, a number of minimally invasive procedures have been developed and tested, giving physicians and their patients more choices for medical management of venous insufficiency. A closer look at 3 procedures, which include 2 types of thermal ablation, will help you direct patients like Kevin to the best possible treatment.

2 thermal ablation procedures use only local anesthesia

Collectively termed endovenous thermal ablation (EVTA), radiofrequency ablation (RFA) and endovenous laser treatment (ELT) are similar. Performed under duplex ultrasound visualization, both offer a number of benefits.

Outpatient treatment. EVTA procedures are typically done on an outpatient basis, with only local anesthesia—0.05% to 0.1% lidocaine, injected as a perivenous tumescent solution under ultrasound guidance.^8,9 The physician places and advances an endovenous catheter (for RFA) or a laser fiber (for ELT) into the vein (FIGURE 1). When the catheter or laser is in proper position, the RF generator or laser is activated. Heat, delivered through the tip of the catheter or fiber, ablates the lumen of the vein as the physician withdraws the catheter.

Mild side effects. Temporary discomfort is common with EVTA. Mild bruising, swelling, and nodularity of the treated veins are the most frequent side effects. More serious adverse effects, including skin burns, paresthesias, DVT, and pulmonary embolism, are rare.^10-12

High success rates. Successful treatment of both the great and small saphenous veins and long-term maintenance of vein closure are the norm, with rates ranging from 88% to 100% for both RFA^13,14 and ELT, regardless of vein size.^15-19 Thermal ablation has been shown to be superior to surgery in other ways as well, eliminating the need for general anesthesia and inpatient treatment and resulting in less postoperative pain, a shorter recovery period, an improved quality of life, and lower costs (TABLE).^20-23

TABLE
Endovenous thermal ablation vs standard surgery

FIGURE 1
Endovenous laser treatment

Not for every patient

Despite the usefulness of these minimally invasive procedures, RFA and ELT are not always possible. Veins that are tortuous, too small, or too close to the surface of the skin are not well suited for thermal ablation; in some cases, chemical ablation, discussed below, is the only option. However, some conditions—including pregnancy, active DVT, poor health status with limited mobility, and severe peripheral vascular disease—are contraindications for both thermal and chemical ablation.²⁴

Thermal won’t work? Consider chemical ablation

Endovenous chemoablation (ECA), like thermal ablation, is performed on an outpatient basis without sedation. Hypertonic saline, an osmotic agent, has been used for decades in treating small-caliber varicose veins, or “spider veins.” Detergenttype sclerosants, which we’ll discuss here, are now widely used to treat larger veins, including the great and small saphenous vein trunks and their affected tributaries.

Better suited for tortuous veins. Patients who have had prior stripping often have branches that are too tortuous to access with thermal ablation. In such cases, ECA is the only option. Those who have never undergone vein stripping often need a combination of thermal and chemical ablation, as well, with ECA reserved for saphenous veins that are inaccessible by RFA or laser.

Foamed sclerosants allow visualization. When mixed through a 3-way stopcock with an inert gas such as room air or carbon dioxide, the detergent-type sclerosants produce a micro-foamed preparation that is echogenic, and allows direct visualization with ultrasound (FIGURE 2). The foam displaces blood from within the vein and promotes contact with the venous endothelium, and intense vasospasm results. The sclerosant foam mixture, which makes it possible to watch the effect (and extent) of the foam as it disperses throughout the target vessel, has enhanced the treatment of larger viscosities.^25-26 Pretreatment venous ultrasound mapping—a process in which the patient stands upright while compression is applied to the saphenous veins and their tributaries to test the valves for normal flow—combined with realtime duplex imaging, minimizes the chance of arterial injection, extravasation, and tissue necrosis.²⁵

Serious adverse effects are rare. Hemosiderin hyperpigmentation and pain at the injection site are among the most common side effects of ultrasound-guided sclerotherapy. Serious adverse events, including anaphylaxis, DVT/pulmonary embolus, cutaneous necrosis, transient ischemic attack, and arterial injection, are rare.²⁷

To date, there is only 1 published prospective, randomized study that directly compares foam sclerotherapy with stripping/ligation surgery. The researchers found the results of both treatments to be comparable, but reported that surgery was more expensive than foam sclerotherapy and produced less improvement in quality of life.²⁸ A meta-analysis of treatment for 12,000 limbs concluded that foam sclerotherapy and EVTA were as effective as surgical stripping.²⁹ There are no prospective, randomized studies comparing foam sclerotherapy with thermal ablation.

FIGURE 2
Endovenous chemoablation

Advising Kevin M.: What’s his best bet?

You’ve talked to your patient in the past about vein stripping, but he’s been reluctant to undergo surgery. When you describe the minimally invasive approaches to chronic venous insufficiency with him, however, he asks to see a specialist—and to undergo ultrasound mapping.

You give him a prescription for graduated compression stockings for immediate use and a referral to a phlebologist. Soon after, you receive a report from the specialist, indicating that your patient had a combination of ELT of the refluxing left great saphenous vein and subsequent ultrasound-guided ECA of the saphenous tributaries.

The next time you see Kevin, he reports that his leg no longer aches—and you can see that the bulging, tortuous varicosities and swelling have improved significantly.

Acknowledgments

The authors thank John Hickner, MD, MSc, for his input during the manuscript process.

CORRESPONDENCE
Jeffery P. Schoonover, MD, FAAFP, Vein Clinics of America, 9000 Keystone Crossing, Suite 960, Indianapolis, IN 46240; [email protected]

CASE

Kevin M., a 35-year-old researcher who often works 12- to 14-hour days, is in your office again, following his second trip to the emergency room for bleeding from a protruding varicose vein proximal to his left ankle. He has complained of leg aching and ankle swelling in the past, usually after he’s been on his feet for hours.

A previous ultrasound showed no evidence of deep vein thrombosis (DVT), but did reveal reflux along the left great saphenous vein from the saphenofemoral junction to the ankle. On examination, Kevin’s lower extremity pulses are intact bilaterally, but there are multiple tortuous varicosities branching from the medial thigh to the left medial malleolus, with evidence of hemosiderin hyperpigmentation and dermatitis on the left lower leg. Previous lab studies have been normal, and Kevin has no other chronic conditions. Kevin’s main concern is to minimize “down time” from work, and he asks you about the newer, minimally invasive vein treatments he’s read about on the Internet.

What should you tell him?

Chronic venous insufficiency is a common and costly condition, affecting nearly one-third of the US population.¹ While many people mistakenly think of varicose veins primarily as a cosmetic issue, venous insufficiency often results in painful, even debilitating, signs and symptoms—from swelling and aching legs to skin changes that range from stasis dermatitis to open ulceration.

Venous stripping and ambulatory phlebectomy were long considered the primary means of addressing saphenous venous insufficiency,² and compression therapy with gradient stockings, short stretch bandages, and Unna’s dressings remains a therapeutic cornerstone for every stage of chronic venous disease. Compression has also been shown to dramatically decrease the risk of post-thrombotic syndrome following DVT³ and, when combined with anticoagulation, to prevent DVT after surgery.⁴

Outcomes of surgical treatment of large varicose veins arising from saphenous truncal vein reflux, however, have historically been poor.⁵ This is primarily because of neovascularization and residual refluxing veins. Average failure rates of 25% for saphenous vein stripping and 43% for saphenofemoral junction ligation have been reported.⁶ A subsequent study of 140 patients who underwent saphenous vein stripping found that 20% experienced neovascularization.⁷

In the last decade, a number of minimally invasive procedures have been developed and tested, giving physicians and their patients more choices for medical management of venous insufficiency. A closer look at 3 procedures, which include 2 types of thermal ablation, will help you direct patients like Kevin to the best possible treatment.

2 thermal ablation procedures use only local anesthesia

Collectively termed endovenous thermal ablation (EVTA), radiofrequency ablation (RFA) and endovenous laser treatment (ELT) are similar. Performed under duplex ultrasound visualization, both offer a number of benefits.

Outpatient treatment. EVTA procedures are typically done on an outpatient basis, with only local anesthesia—0.05% to 0.1% lidocaine, injected as a perivenous tumescent solution under ultrasound guidance.^8,9 The physician places and advances an endovenous catheter (for RFA) or a laser fiber (for ELT) into the vein (FIGURE 1). When the catheter or laser is in proper position, the RF generator or laser is activated. Heat, delivered through the tip of the catheter or fiber, ablates the lumen of the vein as the physician withdraws the catheter.

Mild side effects. Temporary discomfort is common with EVTA. Mild bruising, swelling, and nodularity of the treated veins are the most frequent side effects. More serious adverse effects, including skin burns, paresthesias, DVT, and pulmonary embolism, are rare.^10-12

High success rates. Successful treatment of both the great and small saphenous veins and long-term maintenance of vein closure are the norm, with rates ranging from 88% to 100% for both RFA^13,14 and ELT, regardless of vein size.^15-19 Thermal ablation has been shown to be superior to surgery in other ways as well, eliminating the need for general anesthesia and inpatient treatment and resulting in less postoperative pain, a shorter recovery period, an improved quality of life, and lower costs (TABLE).^20-23

TABLE
Endovenous thermal ablation vs standard surgery

FIGURE 1
Endovenous laser treatment

Not for every patient

Despite the usefulness of these minimally invasive procedures, RFA and ELT are not always possible. Veins that are tortuous, too small, or too close to the surface of the skin are not well suited for thermal ablation; in some cases, chemical ablation, discussed below, is the only option. However, some conditions—including pregnancy, active DVT, poor health status with limited mobility, and severe peripheral vascular disease—are contraindications for both thermal and chemical ablation.²⁴

Thermal won’t work? Consider chemical ablation

Endovenous chemoablation (ECA), like thermal ablation, is performed on an outpatient basis without sedation. Hypertonic saline, an osmotic agent, has been used for decades in treating small-caliber varicose veins, or “spider veins.” Detergenttype sclerosants, which we’ll discuss here, are now widely used to treat larger veins, including the great and small saphenous vein trunks and their affected tributaries.

Better suited for tortuous veins. Patients who have had prior stripping often have branches that are too tortuous to access with thermal ablation. In such cases, ECA is the only option. Those who have never undergone vein stripping often need a combination of thermal and chemical ablation, as well, with ECA reserved for saphenous veins that are inaccessible by RFA or laser.

Foamed sclerosants allow visualization. When mixed through a 3-way stopcock with an inert gas such as room air or carbon dioxide, the detergent-type sclerosants produce a micro-foamed preparation that is echogenic, and allows direct visualization with ultrasound (FIGURE 2). The foam displaces blood from within the vein and promotes contact with the venous endothelium, and intense vasospasm results. The sclerosant foam mixture, which makes it possible to watch the effect (and extent) of the foam as it disperses throughout the target vessel, has enhanced the treatment of larger viscosities.^25-26 Pretreatment venous ultrasound mapping—a process in which the patient stands upright while compression is applied to the saphenous veins and their tributaries to test the valves for normal flow—combined with realtime duplex imaging, minimizes the chance of arterial injection, extravasation, and tissue necrosis.²⁵

Serious adverse effects are rare. Hemosiderin hyperpigmentation and pain at the injection site are among the most common side effects of ultrasound-guided sclerotherapy. Serious adverse events, including anaphylaxis, DVT/pulmonary embolus, cutaneous necrosis, transient ischemic attack, and arterial injection, are rare.²⁷

To date, there is only 1 published prospective, randomized study that directly compares foam sclerotherapy with stripping/ligation surgery. The researchers found the results of both treatments to be comparable, but reported that surgery was more expensive than foam sclerotherapy and produced less improvement in quality of life.²⁸ A meta-analysis of treatment for 12,000 limbs concluded that foam sclerotherapy and EVTA were as effective as surgical stripping.²⁹ There are no prospective, randomized studies comparing foam sclerotherapy with thermal ablation.

FIGURE 2
Endovenous chemoablation

Advising Kevin M.: What’s his best bet?

You’ve talked to your patient in the past about vein stripping, but he’s been reluctant to undergo surgery. When you describe the minimally invasive approaches to chronic venous insufficiency with him, however, he asks to see a specialist—and to undergo ultrasound mapping.

You give him a prescription for graduated compression stockings for immediate use and a referral to a phlebologist. Soon after, you receive a report from the specialist, indicating that your patient had a combination of ELT of the refluxing left great saphenous vein and subsequent ultrasound-guided ECA of the saphenous tributaries.

The next time you see Kevin, he reports that his leg no longer aches—and you can see that the bulging, tortuous varicosities and swelling have improved significantly.

Acknowledgments

The authors thank John Hickner, MD, MSc, for his input during the manuscript process.

CORRESPONDENCE
Jeffery P. Schoonover, MD, FAAFP, Vein Clinics of America, 9000 Keystone Crossing, Suite 960, Indianapolis, IN 46240; [email protected]

Esther V. is an 81-year-old Caucasian woman who has been in your practice for a number of years. Her history is significant for a hip fracture 5 years ago and a subsequent diagnosis of osteoporosis. But she’s still able to live on her own, and takes only 1 medication—alendronate 70 mg once a week. Today Mrs. V.’s blood pressure is 150/80 mm Hg, and a chart review indicates that it has been in that range for the past 2 years. In view of her history and her age, would you proceed with treatment, and if so, how would you address her elevated systolic pressure?

Treating high blood pressure in elderly patients like Esther V. is a complicated proposition: While it’s important to mitigate hypertension’s negative effects, physicians need to be mindful that for this patient population, antihypertensive therapy itself involves elevated risks and unique concerns. These include:

• drug-drug interactions, exacerbated by the multiple medications that many older patients take;
• the side effect profile of antihypertensive agents, including orthostatic hypotension and the possibility of exacerbating an already heightened risk of falls; and
• uncertainty about when to initiate therapy, which drugs to choose if a second antihypertensive agent is needed, and when (or whether) a patient is too old to benefit from treatment.

Noncompliance is another potential complication in the elderly, who may have difficulty following a drug regimen or avoid prescription medications because they’re worried about the adverse effects—or the cost—of the drugs.

Thus, it’s not surprising that hypertension so often remains under- or untreated in older patients, a problem highlighted by a 2003 review in which only 27% of elderly patients were found to have reached their blood pressure goal.¹

Despite the difficulty of adequately addressing hypertension in this patient population, the benefits do outweigh the risks. The following practical guide will help you overcome common barriers to treatment, avoid dangerous drug combinations, and customize your patient’s care to maximum benefit.

Treat hypertension, regardless of age

The Seventh Report of the Joint National Committee on Prevention, Evaluation, and Treatment of Hypertension (JNC 7) recommends a goal of <140/90 mm Hg for most individuals, and <130/80 mm Hg for those with diabetes mellitus or chronic kidney disease.² But as patients age, presentations like that of Mrs. V.—elevated systolic pressure and normal diastolic pressure—are increasingly common. Isolated systolic hypertension is thought to be a consequence of the aging process, which results in a reduction in elasticity and compliance of the large arteries, degradation of arterial elastin, and atherosclerosis-associated accumulation of arterial calcium and collagen.³

Certainly, we know that hypertension of any kind is a major risk factor for a number of conditions with significant morbidity and mortality, including coronary artery disease,⁴ stroke,⁵ and heart failure.⁶ In the elderly, treatment of systolic hypertension, even in the absence of diastolic hypertension, has been proven to reduce cardiovascular and renal disease and death.^7-9

Extending lives: What the evidence shows

Treating the very old. Hypertension in the Very Elderly (HYVET), the most recent trial, randomized more than 3800 hypertensive patients over the age of 80 in Europe, China, Australia, and Tunisia to receive either a diuretic or placebo. After 2 years, the treatment group had significant reductions in fatal and nonfatal stroke (number needed to treat [NNT]=19), all-cause mortality (NNT=8), cardiovascular death (NNT=15), and heart failure (NNT=67).¹⁰

A subset of this trial (HYVET-COG) reviewed the effects of antihypertensive therapy on the development of dementia in the very old. The researchers did not find a statistically significant reduction in the incidence of dementia in the treatment group. But when the HYVET-COG data were combined in a meta-analysis with data from 3 other antihypertensive trials in the elderly, treatment for hypertension was associated with a 13% relative risk reduction for dementia.¹¹ Despite a major limitation of the HYVET trial—participants were typically healthier than the general population within their age group—the findings highlight the benefits of treating hypertension even in the very old.

Treating systolic hypertension. A Cochrane review conducted more than 10 years ago to assess the effectiveness of treating systolic hypertension in otherwise healthy older patients found that treatment reduced cardiovascular morbidity and mortality by 53 events per 1000 patient-years.¹² Several prospective, double-blind, randomized, placebo-controlled studies found that treatment of systolic hypertension in healthy elderly patients reduced the incidence of total mortality (NNT=59), fatal and nonfatal cardiovascular events (NNT=26), and fatal and nonfatal stroke (NNT=79).¹³

Getting past common barriers to treatment

Despite these findings, many elderly patients with hypertension do not receive adequate treatment, often because of clinicians’ concerns about uncertain parameters and treatment risks. We’ve identified some common barriers and suggested ways to get beyond them.

BARRIER: Unclear parameters, concerns about aggressive Tx

There is little doubt of the value of treating stage 2 systolic hypertension (≥160 mm Hg) in the elderly. Not so for stage 1 systolic hypertension (140-159 mm Hg): A 2004 systematic review of studies evaluating the treatment of elevated systolic pressure in this patient population concluded that evidence for treating stage 1 systolic hypertension was not as strong.¹⁴

Observational studies suggesting a link between aggressive blood pressure-lowering and increased mortality in elderly patients have also been a cause for concern among some physicians.^15,16 The INDANA meta-analysis, conducted in 1999, reviewed data from patients 80 years of age and older enrolled in antihypertensive trials and found a nonsignificant increase in death rates among patients who received treatment.¹⁷

How to respond? Carefully weigh the potential benefits of antihypertensive therapy in view of the overall health, functional status, and risks of side effects for a particular patient. The findings of the INDANA meta-analysis, while not statistically significant, serve as a reminder of the need to individualize treatment.

In the absence of a definitive treatment approach, it is also important to discuss the options with the patient. The authors of the systematic review that evaluated the benefits of treatment for stage 1 systolic hypertension recommended that antihypertensive therapy be based, in part, on patient preference and tolerance of therapy.¹⁴

BARRIER: Risks associated with multiple medications

The concurrent use of multiple medications, which is very common in the elderly, can interfere with treatment of hypertension in 2 ways: Some drugs have the potential to cause (or exacerbate) elevated blood pressure. Others may counter the effects of antihypertensive therapy.

Among the drugs that contribute to hypertension are the sympathomimetic agents found in many over-the-counter (OTC) cough and cold preparations in which the mechanism of action is direct vasoconstriction.¹⁸ Tricyclic antidepressants, which are commonly used to treat pain syndromes in the elderly, have also been shown to raise both systolic and diastolic pressure.¹⁹

On the other hand, nonsteroidal anti-inflammatory drugs (NSAIDs), often used by elderly patients to treat osteoarthritis, may counter the effects of many antihypertensive agents. Both OTC and prescription NSAIDs inhibit cyclo-oxygenase-2 (COX-2) in the kidneys, resulting in a reduction in sodium excretion and an increase in plasma volume.²⁰

How to respond? Take a thorough medication history, including OTC and prescription drugs and supplements, before initiating treatment. Caution patients to avoid NSAIDs, OTC cough and cold preparations, and excessive caffeine, all of which may elevate their blood pressure. Remind patients that any new medication prescribed by another physician may affect their blood pressure, and advise them to maintain a current list of all medications to be reviewed at each visit.

BARRIER: Concerns about side effects

The elderly are more susceptible to orthostatic hypotension than younger patients because of age-related decreases in the baroreflex.²¹ Estimates suggest that anywhere from 5% to 50% of individuals older than 65 years are affected.²¹ Orthostatic hypotension, a side effect of many antihypertensive agents, is associated with higher rates of dizziness, syncope, and falls—a particular concern in elderly patients with a history of osteoporosis, pathological fracture, or gait dysfunction (See “Screening for hearing loss, risk of falls”).

How to respond? Before initiating antihypertensive therapy, take your patient’s blood pressure in both a sitting and standing position. Educate the patient and his or her family about the presentation of orthostatic hypotension, and advise the patient to stop taking the new medication and contact you if he or she develops the associated signs and symptoms. Schedule frequent follow-up appointments to closely monitor the effects of antihypertensive therapy, as well.

Stress the importance of lifestyle interventions

JNC 7 recommends several nonpharmacological treatments for all hypertensive patients, regardless of age. These include:

• healthy diet (such as Dietary Approaches to Stop Hypertension [DASH])
• sodium restriction (daily intake of <100 mmol/d)
• limited alcohol intake
• weight reduction (a decrease of 5-20 mm Hg for every 10 kg weight loss has been reported)
• regular physical activity (reported decrease of 4-9 mm Hg).²

No long-term trials have assessed the effects of lifestyle modification on morbidity and mortality. A subset of patients in the original DASH trial with stage 1 systolic hypertension did benefit from the diet, but the subgroup was small and the median age was 54.7 years.²² No large-scale studies have addressed the effectiveness of the diet in older patients with systolic hypertension.

Customizing therapy: What to consider

Evidence supports the use of various classes of antihypertensive medications in the treatment of high blood pressure, including thiazide diuretics, beta-blockers, angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), and calcium channel blockers (CCBs).^10,23-27 JNC 7 recommends starting with 1 medication for patients with stage 1 hypertension (140-159/90-99 mm Hg) and 2 medications for patients with stage 2 hypertension (≥160/100 mm Hg).² As previously noted, however, clear-cut evidence in support of treating elderly patients with stage 1 hypertension is lacking.

When you do initiate treatment for an elderly patient, begin with a low dose and titrate slowly, carefully monitoring for side effects. If 2 antihypertensive agents are needed, start 1 agent at a time in patients who are at risk for significant side effects. The type of antihypertensive agent you prescribe should be based on your patient’s health status, comorbidities, and treatment for other conditions, as well as on the known effects of the particular class of drugs (TABLE).

Thiazide diuretics. JNC 7 recommends thiazide diuretics as first-line treatment based on cost and effectiveness. Because thiazide diuretics decrease the urinary excretion of calcium, JNC 7 recommends their use in patients with osteoporosis.²

But potential side effects, such as dehydration, orthostatic hypotension, and hypokalemia, are more common in elderly patients, and may limit the use of these drugs in clinical practice. NSAIDs may reduce the antihypertensive effects of thiazide diuretics through a decrease in sodium excretion and an increase in plasma volume, and should be used with caution—if at all—in this patient population.²⁰

Hydrochlorothiazide, a commonly used thiazide diuretic, should not be used in doses >50 mg/d because higher doses increase the risk of side effects without increasing efficacy.²⁸ In the Systolic Hypertension in the Elderly Program (SHEP) study, the positive effects of another thiazide diuretic—chlorthalidone—were lost in elderly patients with serum potassium <3.5 mmol/L, so serum electrolytes should be routinely assessed during therapy.^2,7 Thiazide diuretics lose their effectiveness in patients with a glomerular filtration rate of <30 mL/min/1.73 m² and should not be used in this group.²⁹

Beta-blockers. Beta-blockers are indicated in patients with a previous history of myocardial infarction, as they have been shown to decrease the rate of new coronary events.³⁰ Other comorbidities that may be mitigated by beta-blocker administration include angina pectoris, atrial fibrillation with a rapid ventricular rate, compensated heart failure, preoperative hypertension, and essential tremor.²

Side effects of beta-blockers include sedation, depression, sexual dysfunction, bradycardia, conduction abnormalities, and exacerbation of severe reactive airway disease.² Central nervous system effects of beta-blockers tend to be lowest in hydrophilic agents, such as atenolol, and highest in lipophilic agents, such as propranolol, which readily cross the blood-brain barrier.³¹

A 2007 Cochrane review looked at the effectiveness of beta-blockers as first-line agents in uncomplicated hypertension and concluded that the evidence does not support their use. The reviewers noted, however, that the majority of the trials used atenolol and cautioned that the results might not apply to other beta-blockers—or to subgroups, such as the elderly.³²

ACE inhibitors and ARBs. Both ACE inhibitors and ARBs inhibit the renin-angiotensin-aldosterone pathway, and both classes are indicated in patients with diabetes, chronic kidney disease, or heart failure.² Cough is a side effect of ACE inhibitors that affects 5% to 35% of patients,³³ but angioedema—estimated to occur in 0.1% to 0.7% of patients taking ACE inhibitors—is the most serious side effect.³⁴

An elevation in serum potassium and serum creatinine may occur after starting an ACE inhibitor in patients with heart failure, dehydration, or significant renal insufficiency, so assess serum creatinine and potassium 1 week after initiating therapy. An increase in serum creatinine >30% may require discontinuation or dose reduction.^2,35

Elderly patients who take NSAIDs while on ACE inhibitors are at particular risk for acute renal failure; reviews of iatrogenic acute renal failure in the elderly have shown that most cases are related to the concurrent use of these medications.^36,37 This is yet another reason to avoid giving NSAIDs to elderly patients with hypertension, particularly if they’re taking diuretics or ACE inhibitors.

Beware of ACE inhibitor-ARB combinations. The American College of Cardiology and the American Heart Association recommend substituting an ARB if side effects from an ACE inhibitor occur in patients with heart failure, but extreme caution is required if the patient experienced ACE inhibitor-associated angioedema.³⁸ TRANSCEND, a randomized controlled trial of almost 6000 patients, specifically studied the use of ARBs in patients unable to tolerate ACE inhibitors and concluded that they can be safely given to patients who had side effects from ACE inhibitors.³⁹

Since ACE inhibitors and ARBs affect the renin-angiotensin-aldosterone pathway at different points, there has been interest in the effectiveness of combining these agents. The CHARM-Added trial found that the combination of these agents reduced both cardiovascular events and mortality in patients with heart failure.⁴⁰

Combining ACE inhibitors and ARBs in patients without heart failure was evaluated in the ONTARGET trial. While the combination treatment group had a greater reduction in blood pressure, there was no significant cardiovascular benefit over patients who were on ACE inhibitors alone. But there was an increase in hypotension, syncope, and renal dysfunction among those in the combination treatment group, all of which are significant concerns in the elderly. The authors concluded that the ACE inhibitor-ARB combination should not be used in patients who do not have heart failure.⁴¹

Calcium channel blockers. CCBs block the entrance of calcium into vascular cells, producing dilation in the coronary arteries and peripheral vasculature, and are effective in the treatment of hypertension in elderly patients.⁴² Nondihydropyridine CCBs (diltiazem and verapamil) are useful in patients with atrial fibrillation and supraventricular tachycardia because of their negative chronotropic effects.² And African American patients respond better to CCBs—with a greater reduction in blood pressure and cardiovascular complications—than to ACE inhibitors.^2,25

Dihydropyridine CCBs (amlodipine and long-acting felodipine and nifedipine) are safe in patients with heart failure or chronic stable angina,⁴³ but short-acting nifedipine has been found to increase the risk of mortality.⁴⁴

Dihydropyridine CCBs have been reported to cause peripheral edema in 7% to 8% of patients taking long-acting nifedipine and up to 16% of elderly patient taking amlodipine.⁴⁵ Verapamil has the highest incidence of constipation, and should be avoided in elderly patients.²

TABLE
Which drugs, for which patients?²

Steer clear of these drugs

Significant side effects, including hypotension, sedation, depression, and dry mouth, limit the use of other antihypertensives, such as peripheral alpha blockers, centrally acting agents, and vasodilators, in elderly patients.² In the ALLHAT trial, the alpha-blocker arm was stopped prematurely because of an increased rate of cardiovascular events and heart failure compared with the thiazide diuretic arm.²⁵ That finding led to the recommendation that alpha-blockers not be used as first-line agents for treating hypertension.

Centrally acting alpha-agonists (clonidine, methyldopa, and reserpine) have a high incidence of sedation, dry mouth, and depression, and elderly patients are more likely to experience orthostatic hypotension and rebound hypertension if these agents are discontinued abruptly or doses are missed.² Vasodilators, including minoxidil and hydralazine, cause sodium and fluid retention and reflex tachycardia.² Because of their unfavorable side effects and lack of outcomes data, alpha-blockers, centrally acting alpha-agonists, and vasodilators should not be used routinely for the treatment of hypertension in elderly patients.

Considerations in selecting a second agent

Patients who fail to achieve adequate blood pressure control despite lifestyle modifications and a single agent, and those who are initially diagnosed with stage 2 hypertension, will require 2 or more medications. An elderly patient who has not responded to a single agent should be asked about any other medications (or substances) that might be interfering with the antihypertensive agent, especially NSAIDs, tricyclic antidepressants, sympathomimetics, and caffeine. Find out, too, whether the patient is following the prescribed regimen.

When you prescribe 2 antihypertensives, choose agents with complementary mechanisms of action—a diuretic and an ACE inhibitor, for example, or an ACE inhibitor and a CCB. In addition to avoiding an ACE/ARB combination in patients who do not have heart failure, avoid concurrent use of beta-blockers and ACE inhibitors⁴⁶ and a dihydropyridine CCB/thiazide diuretic combination⁴⁷ in all elderly patients.¹⁴

How does the evidence apply to your patient?

As we saw earlier, Esther V. has isolated systolic hypertension, which increases her risk of myocardial infarction, stroke, and heart failure. Even though she is 81 years old, studies such as HYVET support the treatment of her hypertension. Certainly, you’ll encourage her to limit her salt intake, follow a healthy diet, and exercise daily, as tolerated.

But because Mrs. V. has stage 1 hypertension, her preference and her ability to tolerate therapy will play a key role in the decision to initiate treatment. If her blood pressure does not respond to lifestyle modifications and she is agreeable to medical therapy, you would recommend 1 antihypertensive agent.

Which drug class is best? Given her history of osteoporosis, a low-dose thiazide diuretic might be a reasonable choice. But because she has a history of hip fracture from a fall, it would be prudent to avoid agents associated with volume depletion —and to opt for either an ACE inhibitor or a CCB as first-line therapy instead. It is extremely important to obtain a list of any OTC medications Mrs. V. is currently taking and to advise her to avoid the use of NSAIDs.

You would also want to obtain the results of Mrs. V.’s most recent bone mineral density test, and talk to her about the importance of taking calcium with vitamin D. If there are concerns about her gait, you would consider a formal gait evaluation, as well. Finally, you would review potential side effects, including orthostatic hypotension, and tell Mrs. V. to call if they occur, and schedule an appointment to monitor her serum potassium and renal function in a few weeks.

Correspondence Robert C. Langan, MD, St. Luke’s Family Medicine Residency Program, 2830 Easton Avenue, Bethlehem, PA 18017; [email protected]

Esther V. is an 81-year-old Caucasian woman who has been in your practice for a number of years. Her history is significant for a hip fracture 5 years ago and a subsequent diagnosis of osteoporosis. But she’s still able to live on her own, and takes only 1 medication—alendronate 70 mg once a week. Today Mrs. V.’s blood pressure is 150/80 mm Hg, and a chart review indicates that it has been in that range for the past 2 years. In view of her history and her age, would you proceed with treatment, and if so, how would you address her elevated systolic pressure?

Treating high blood pressure in elderly patients like Esther V. is a complicated proposition: While it’s important to mitigate hypertension’s negative effects, physicians need to be mindful that for this patient population, antihypertensive therapy itself involves elevated risks and unique concerns. These include:

• drug-drug interactions, exacerbated by the multiple medications that many older patients take;
• the side effect profile of antihypertensive agents, including orthostatic hypotension and the possibility of exacerbating an already heightened risk of falls; and
• uncertainty about when to initiate therapy, which drugs to choose if a second antihypertensive agent is needed, and when (or whether) a patient is too old to benefit from treatment.

Noncompliance is another potential complication in the elderly, who may have difficulty following a drug regimen or avoid prescription medications because they’re worried about the adverse effects—or the cost—of the drugs.

Thus, it’s not surprising that hypertension so often remains under- or untreated in older patients, a problem highlighted by a 2003 review in which only 27% of elderly patients were found to have reached their blood pressure goal.¹

Despite the difficulty of adequately addressing hypertension in this patient population, the benefits do outweigh the risks. The following practical guide will help you overcome common barriers to treatment, avoid dangerous drug combinations, and customize your patient’s care to maximum benefit.

Treat hypertension, regardless of age

The Seventh Report of the Joint National Committee on Prevention, Evaluation, and Treatment of Hypertension (JNC 7) recommends a goal of <140/90 mm Hg for most individuals, and <130/80 mm Hg for those with diabetes mellitus or chronic kidney disease.² But as patients age, presentations like that of Mrs. V.—elevated systolic pressure and normal diastolic pressure—are increasingly common. Isolated systolic hypertension is thought to be a consequence of the aging process, which results in a reduction in elasticity and compliance of the large arteries, degradation of arterial elastin, and atherosclerosis-associated accumulation of arterial calcium and collagen.³

Certainly, we know that hypertension of any kind is a major risk factor for a number of conditions with significant morbidity and mortality, including coronary artery disease,⁴ stroke,⁵ and heart failure.⁶ In the elderly, treatment of systolic hypertension, even in the absence of diastolic hypertension, has been proven to reduce cardiovascular and renal disease and death.^7-9

Extending lives: What the evidence shows

Treating the very old. Hypertension in the Very Elderly (HYVET), the most recent trial, randomized more than 3800 hypertensive patients over the age of 80 in Europe, China, Australia, and Tunisia to receive either a diuretic or placebo. After 2 years, the treatment group had significant reductions in fatal and nonfatal stroke (number needed to treat [NNT]=19), all-cause mortality (NNT=8), cardiovascular death (NNT=15), and heart failure (NNT=67).¹⁰

A subset of this trial (HYVET-COG) reviewed the effects of antihypertensive therapy on the development of dementia in the very old. The researchers did not find a statistically significant reduction in the incidence of dementia in the treatment group. But when the HYVET-COG data were combined in a meta-analysis with data from 3 other antihypertensive trials in the elderly, treatment for hypertension was associated with a 13% relative risk reduction for dementia.¹¹ Despite a major limitation of the HYVET trial—participants were typically healthier than the general population within their age group—the findings highlight the benefits of treating hypertension even in the very old.

Treating systolic hypertension. A Cochrane review conducted more than 10 years ago to assess the effectiveness of treating systolic hypertension in otherwise healthy older patients found that treatment reduced cardiovascular morbidity and mortality by 53 events per 1000 patient-years.¹² Several prospective, double-blind, randomized, placebo-controlled studies found that treatment of systolic hypertension in healthy elderly patients reduced the incidence of total mortality (NNT=59), fatal and nonfatal cardiovascular events (NNT=26), and fatal and nonfatal stroke (NNT=79).¹³

Getting past common barriers to treatment

Despite these findings, many elderly patients with hypertension do not receive adequate treatment, often because of clinicians’ concerns about uncertain parameters and treatment risks. We’ve identified some common barriers and suggested ways to get beyond them.

BARRIER: Unclear parameters, concerns about aggressive Tx

There is little doubt of the value of treating stage 2 systolic hypertension (≥160 mm Hg) in the elderly. Not so for stage 1 systolic hypertension (140-159 mm Hg): A 2004 systematic review of studies evaluating the treatment of elevated systolic pressure in this patient population concluded that evidence for treating stage 1 systolic hypertension was not as strong.¹⁴

Observational studies suggesting a link between aggressive blood pressure-lowering and increased mortality in elderly patients have also been a cause for concern among some physicians.^15,16 The INDANA meta-analysis, conducted in 1999, reviewed data from patients 80 years of age and older enrolled in antihypertensive trials and found a nonsignificant increase in death rates among patients who received treatment.¹⁷

How to respond? Carefully weigh the potential benefits of antihypertensive therapy in view of the overall health, functional status, and risks of side effects for a particular patient. The findings of the INDANA meta-analysis, while not statistically significant, serve as a reminder of the need to individualize treatment.

In the absence of a definitive treatment approach, it is also important to discuss the options with the patient. The authors of the systematic review that evaluated the benefits of treatment for stage 1 systolic hypertension recommended that antihypertensive therapy be based, in part, on patient preference and tolerance of therapy.¹⁴

BARRIER: Risks associated with multiple medications

The concurrent use of multiple medications, which is very common in the elderly, can interfere with treatment of hypertension in 2 ways: Some drugs have the potential to cause (or exacerbate) elevated blood pressure. Others may counter the effects of antihypertensive therapy.

Among the drugs that contribute to hypertension are the sympathomimetic agents found in many over-the-counter (OTC) cough and cold preparations in which the mechanism of action is direct vasoconstriction.¹⁸ Tricyclic antidepressants, which are commonly used to treat pain syndromes in the elderly, have also been shown to raise both systolic and diastolic pressure.¹⁹

On the other hand, nonsteroidal anti-inflammatory drugs (NSAIDs), often used by elderly patients to treat osteoarthritis, may counter the effects of many antihypertensive agents. Both OTC and prescription NSAIDs inhibit cyclo-oxygenase-2 (COX-2) in the kidneys, resulting in a reduction in sodium excretion and an increase in plasma volume.²⁰

How to respond? Take a thorough medication history, including OTC and prescription drugs and supplements, before initiating treatment. Caution patients to avoid NSAIDs, OTC cough and cold preparations, and excessive caffeine, all of which may elevate their blood pressure. Remind patients that any new medication prescribed by another physician may affect their blood pressure, and advise them to maintain a current list of all medications to be reviewed at each visit.

BARRIER: Concerns about side effects

The elderly are more susceptible to orthostatic hypotension than younger patients because of age-related decreases in the baroreflex.²¹ Estimates suggest that anywhere from 5% to 50% of individuals older than 65 years are affected.²¹ Orthostatic hypotension, a side effect of many antihypertensive agents, is associated with higher rates of dizziness, syncope, and falls—a particular concern in elderly patients with a history of osteoporosis, pathological fracture, or gait dysfunction (See “Screening for hearing loss, risk of falls”).

How to respond? Before initiating antihypertensive therapy, take your patient’s blood pressure in both a sitting and standing position. Educate the patient and his or her family about the presentation of orthostatic hypotension, and advise the patient to stop taking the new medication and contact you if he or she develops the associated signs and symptoms. Schedule frequent follow-up appointments to closely monitor the effects of antihypertensive therapy, as well.

Stress the importance of lifestyle interventions

JNC 7 recommends several nonpharmacological treatments for all hypertensive patients, regardless of age. These include:

• healthy diet (such as Dietary Approaches to Stop Hypertension [DASH])
• sodium restriction (daily intake of <100 mmol/d)
• limited alcohol intake
• weight reduction (a decrease of 5-20 mm Hg for every 10 kg weight loss has been reported)
• regular physical activity (reported decrease of 4-9 mm Hg).²

No long-term trials have assessed the effects of lifestyle modification on morbidity and mortality. A subset of patients in the original DASH trial with stage 1 systolic hypertension did benefit from the diet, but the subgroup was small and the median age was 54.7 years.²² No large-scale studies have addressed the effectiveness of the diet in older patients with systolic hypertension.

Customizing therapy: What to consider

Evidence supports the use of various classes of antihypertensive medications in the treatment of high blood pressure, including thiazide diuretics, beta-blockers, angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), and calcium channel blockers (CCBs).^10,23-27 JNC 7 recommends starting with 1 medication for patients with stage 1 hypertension (140-159/90-99 mm Hg) and 2 medications for patients with stage 2 hypertension (≥160/100 mm Hg).² As previously noted, however, clear-cut evidence in support of treating elderly patients with stage 1 hypertension is lacking.

When you do initiate treatment for an elderly patient, begin with a low dose and titrate slowly, carefully monitoring for side effects. If 2 antihypertensive agents are needed, start 1 agent at a time in patients who are at risk for significant side effects. The type of antihypertensive agent you prescribe should be based on your patient’s health status, comorbidities, and treatment for other conditions, as well as on the known effects of the particular class of drugs (TABLE).

Thiazide diuretics. JNC 7 recommends thiazide diuretics as first-line treatment based on cost and effectiveness. Because thiazide diuretics decrease the urinary excretion of calcium, JNC 7 recommends their use in patients with osteoporosis.²

But potential side effects, such as dehydration, orthostatic hypotension, and hypokalemia, are more common in elderly patients, and may limit the use of these drugs in clinical practice. NSAIDs may reduce the antihypertensive effects of thiazide diuretics through a decrease in sodium excretion and an increase in plasma volume, and should be used with caution—if at all—in this patient population.²⁰

Hydrochlorothiazide, a commonly used thiazide diuretic, should not be used in doses >50 mg/d because higher doses increase the risk of side effects without increasing efficacy.²⁸ In the Systolic Hypertension in the Elderly Program (SHEP) study, the positive effects of another thiazide diuretic—chlorthalidone—were lost in elderly patients with serum potassium <3.5 mmol/L, so serum electrolytes should be routinely assessed during therapy.^2,7 Thiazide diuretics lose their effectiveness in patients with a glomerular filtration rate of <30 mL/min/1.73 m² and should not be used in this group.²⁹

Beta-blockers. Beta-blockers are indicated in patients with a previous history of myocardial infarction, as they have been shown to decrease the rate of new coronary events.³⁰ Other comorbidities that may be mitigated by beta-blocker administration include angina pectoris, atrial fibrillation with a rapid ventricular rate, compensated heart failure, preoperative hypertension, and essential tremor.²

Side effects of beta-blockers include sedation, depression, sexual dysfunction, bradycardia, conduction abnormalities, and exacerbation of severe reactive airway disease.² Central nervous system effects of beta-blockers tend to be lowest in hydrophilic agents, such as atenolol, and highest in lipophilic agents, such as propranolol, which readily cross the blood-brain barrier.³¹

A 2007 Cochrane review looked at the effectiveness of beta-blockers as first-line agents in uncomplicated hypertension and concluded that the evidence does not support their use. The reviewers noted, however, that the majority of the trials used atenolol and cautioned that the results might not apply to other beta-blockers—or to subgroups, such as the elderly.³²

ACE inhibitors and ARBs. Both ACE inhibitors and ARBs inhibit the renin-angiotensin-aldosterone pathway, and both classes are indicated in patients with diabetes, chronic kidney disease, or heart failure.² Cough is a side effect of ACE inhibitors that affects 5% to 35% of patients,³³ but angioedema—estimated to occur in 0.1% to 0.7% of patients taking ACE inhibitors—is the most serious side effect.³⁴

An elevation in serum potassium and serum creatinine may occur after starting an ACE inhibitor in patients with heart failure, dehydration, or significant renal insufficiency, so assess serum creatinine and potassium 1 week after initiating therapy. An increase in serum creatinine >30% may require discontinuation or dose reduction.^2,35

Elderly patients who take NSAIDs while on ACE inhibitors are at particular risk for acute renal failure; reviews of iatrogenic acute renal failure in the elderly have shown that most cases are related to the concurrent use of these medications.^36,37 This is yet another reason to avoid giving NSAIDs to elderly patients with hypertension, particularly if they’re taking diuretics or ACE inhibitors.

Beware of ACE inhibitor-ARB combinations. The American College of Cardiology and the American Heart Association recommend substituting an ARB if side effects from an ACE inhibitor occur in patients with heart failure, but extreme caution is required if the patient experienced ACE inhibitor-associated angioedema.³⁸ TRANSCEND, a randomized controlled trial of almost 6000 patients, specifically studied the use of ARBs in patients unable to tolerate ACE inhibitors and concluded that they can be safely given to patients who had side effects from ACE inhibitors.³⁹

Since ACE inhibitors and ARBs affect the renin-angiotensin-aldosterone pathway at different points, there has been interest in the effectiveness of combining these agents. The CHARM-Added trial found that the combination of these agents reduced both cardiovascular events and mortality in patients with heart failure.⁴⁰

Combining ACE inhibitors and ARBs in patients without heart failure was evaluated in the ONTARGET trial. While the combination treatment group had a greater reduction in blood pressure, there was no significant cardiovascular benefit over patients who were on ACE inhibitors alone. But there was an increase in hypotension, syncope, and renal dysfunction among those in the combination treatment group, all of which are significant concerns in the elderly. The authors concluded that the ACE inhibitor-ARB combination should not be used in patients who do not have heart failure.⁴¹

Calcium channel blockers. CCBs block the entrance of calcium into vascular cells, producing dilation in the coronary arteries and peripheral vasculature, and are effective in the treatment of hypertension in elderly patients.⁴² Nondihydropyridine CCBs (diltiazem and verapamil) are useful in patients with atrial fibrillation and supraventricular tachycardia because of their negative chronotropic effects.² And African American patients respond better to CCBs—with a greater reduction in blood pressure and cardiovascular complications—than to ACE inhibitors.^2,25

Dihydropyridine CCBs (amlodipine and long-acting felodipine and nifedipine) are safe in patients with heart failure or chronic stable angina,⁴³ but short-acting nifedipine has been found to increase the risk of mortality.⁴⁴