Robert Finn

SAN FRANCISCO – A growing body of evidence suggests that insulin sensitizers may avert cognitive decline in people with Alzheimer's disease, Suzanne Craft, Ph.D., said at the Third World Congress on Insulin Resistance Syndrome.

In one randomized, placebo-controlled clinical trial, rosiglitazone appeared to be effective in preserving memory and selective attention in a small group of patients with early Alzheimer's disease (AD) or mild cognitive impairment (MCI), said Dr. Craft of the University of Washington, Seattle.

Another randomized trial involved elderly patients with impaired glucose tolerance (IGT) or type 2 diabetes mellitus (T2DM) and contained three arms: placebo, pioglitazone, and nateglinide. Both pioglitazone and nateglinide improve glucose tolerance, but only pioglitazone improved insulin sensitivity and reduced insulin levels. In this trial, pioglitazone, but not nateglinide or placebo, improved performance in a memory test.

“If you pick up an endocrinology textbook from about 15 years ago, you may very well read that 'the brain is an insulin-insensitive organ,' “Dr. Craft said. “We're coming to understand that's very much not the case. There is a critical relationship between insulin resistance and key aspects of brain function.”

In fact, the brain contains many insulin receptors, which tend to be clustered in the hypothalamus, where insulin likely helps regulate eating behavior, and in cortical regions closely linked to cognition.

In earlier studies on healthy individuals Dr. Craft and her colleagues determined that experimentally induced hyperinsulinemia increases the production of the 42-peptide form of β-amyloid (Aβ), the precursor of the amyloid plaques that are the hallmarks of AD. The relationship between insulin levels and Aβ turned out to be age related, and was especially apparent in people over the age of 70.

The rosiglitazone trial compared 20 patients taking 4 mg of the drug daily for 6 months with 10 patients taking placebo. All of the patients had a confirmed diagnosis of AD or MCI and averaged 73 years of age. Patients taking rosiglitazone did have significantly lower fasting plasma insulin levels than control patients, but there were no differences between the groups on fasting glucose, lipids, or liver enzymes (Am. J. Geriatr. Psychiatry 2005;13:950–8).

At 4 and 6 months into the study, patients taking rosiglitazone performed significantly better than those taking placebo on a delayed memory task (the Buschke Selective Reminding Test). At the 6-month time point patients taking rosiglitazone performed significantly better on a measure of selective attention (the Stroop Color-Word Test).

The other trial, which has not yet been published, involved 71 patients over the age of 55 with IGT or T2DM. Patients were randomly assigned to receive placebo, pioglitazone (30 mg/day), or nateglinide (360 mg/day) for 4 months.

Compared with baseline, patients taking pioglitazone showed significant improvement in performance on a story recall test, while patients taking nateglinide or placebo showed no such improvement. Furthermore, among the patients taking pioglitazone, the improvement was directly proportional to the extent of their metabolic treatment response, as measured by a 2-hour oral glucose tolerance test.

A subset of these patients received PET scans at baseline and after 4 months of treatment. As expected, insulin-resistant patients showed hypometabolism in the left temporal lobe and the right parietal region, both areas that are known to be affected in the very early stages of AD.

In addition, while performing a task of word memory in the PET scanner, normal subjects showed a pattern of increased glucose metabolism in the right frontal lobe. Patients with IGT showed a significantly smaller increase in glucose metabolism in that area, and patients with T2DM showed a smaller increase still.

Pioglitazone and rosiglitazone were both tolerated well in the two trials.

Although these studies are small and preliminary, if confirmed in larger studies, they open the possibility that insulin sensitizers may provide an effective treatment for the cognitive decline seen in Alzheimer's disease.

SAN FRANCISCO – A growing body of evidence suggests that insulin sensitizers may avert cognitive decline in people with Alzheimer's disease, Suzanne Craft, Ph.D., said at the Third World Congress on Insulin Resistance Syndrome.

In one randomized, placebo-controlled clinical trial, rosiglitazone appeared to be effective in preserving memory and selective attention in a small group of patients with early Alzheimer's disease (AD) or mild cognitive impairment (MCI), said Dr. Craft of the University of Washington, Seattle.

Another randomized trial involved elderly patients with impaired glucose tolerance (IGT) or type 2 diabetes mellitus (T2DM) and contained three arms: placebo, pioglitazone, and nateglinide. Both pioglitazone and nateglinide improve glucose tolerance, but only pioglitazone improved insulin sensitivity and reduced insulin levels. In this trial, pioglitazone, but not nateglinide or placebo, improved performance in a memory test.

“If you pick up an endocrinology textbook from about 15 years ago, you may very well read that 'the brain is an insulin-insensitive organ,' “Dr. Craft said. “We're coming to understand that's very much not the case. There is a critical relationship between insulin resistance and key aspects of brain function.”

In fact, the brain contains many insulin receptors, which tend to be clustered in the hypothalamus, where insulin likely helps regulate eating behavior, and in cortical regions closely linked to cognition.

In earlier studies on healthy individuals Dr. Craft and her colleagues determined that experimentally induced hyperinsulinemia increases the production of the 42-peptide form of β-amyloid (Aβ), the precursor of the amyloid plaques that are the hallmarks of AD. The relationship between insulin levels and Aβ turned out to be age related, and was especially apparent in people over the age of 70.

The rosiglitazone trial compared 20 patients taking 4 mg of the drug daily for 6 months with 10 patients taking placebo. All of the patients had a confirmed diagnosis of AD or MCI and averaged 73 years of age. Patients taking rosiglitazone did have significantly lower fasting plasma insulin levels than control patients, but there were no differences between the groups on fasting glucose, lipids, or liver enzymes (Am. J. Geriatr. Psychiatry 2005;13:950–8).

At 4 and 6 months into the study, patients taking rosiglitazone performed significantly better than those taking placebo on a delayed memory task (the Buschke Selective Reminding Test). At the 6-month time point patients taking rosiglitazone performed significantly better on a measure of selective attention (the Stroop Color-Word Test).

The other trial, which has not yet been published, involved 71 patients over the age of 55 with IGT or T2DM. Patients were randomly assigned to receive placebo, pioglitazone (30 mg/day), or nateglinide (360 mg/day) for 4 months.

Compared with baseline, patients taking pioglitazone showed significant improvement in performance on a story recall test, while patients taking nateglinide or placebo showed no such improvement. Furthermore, among the patients taking pioglitazone, the improvement was directly proportional to the extent of their metabolic treatment response, as measured by a 2-hour oral glucose tolerance test.

A subset of these patients received PET scans at baseline and after 4 months of treatment. As expected, insulin-resistant patients showed hypometabolism in the left temporal lobe and the right parietal region, both areas that are known to be affected in the very early stages of AD.

In addition, while performing a task of word memory in the PET scanner, normal subjects showed a pattern of increased glucose metabolism in the right frontal lobe. Patients with IGT showed a significantly smaller increase in glucose metabolism in that area, and patients with T2DM showed a smaller increase still.

Pioglitazone and rosiglitazone were both tolerated well in the two trials.

Although these studies are small and preliminary, if confirmed in larger studies, they open the possibility that insulin sensitizers may provide an effective treatment for the cognitive decline seen in Alzheimer's disease.

SAN FRANCISCO – A growing body of evidence suggests that insulin sensitizers may avert cognitive decline in people with Alzheimer's disease, Suzanne Craft, Ph.D., said at the Third World Congress on Insulin Resistance Syndrome.

In one randomized, placebo-controlled clinical trial, rosiglitazone appeared to be effective in preserving memory and selective attention in a small group of patients with early Alzheimer's disease (AD) or mild cognitive impairment (MCI), said Dr. Craft of the University of Washington, Seattle.

Another randomized trial involved elderly patients with impaired glucose tolerance (IGT) or type 2 diabetes mellitus (T2DM) and contained three arms: placebo, pioglitazone, and nateglinide. Both pioglitazone and nateglinide improve glucose tolerance, but only pioglitazone improved insulin sensitivity and reduced insulin levels. In this trial, pioglitazone, but not nateglinide or placebo, improved performance in a memory test.

“If you pick up an endocrinology textbook from about 15 years ago, you may very well read that 'the brain is an insulin-insensitive organ,' “Dr. Craft said. “We're coming to understand that's very much not the case. There is a critical relationship between insulin resistance and key aspects of brain function.”

In fact, the brain contains many insulin receptors, which tend to be clustered in the hypothalamus, where insulin likely helps regulate eating behavior, and in cortical regions closely linked to cognition.

In earlier studies on healthy individuals Dr. Craft and her colleagues determined that experimentally induced hyperinsulinemia increases the production of the 42-peptide form of β-amyloid (Aβ), the precursor of the amyloid plaques that are the hallmarks of AD. The relationship between insulin levels and Aβ turned out to be age related, and was especially apparent in people over the age of 70.

The rosiglitazone trial compared 20 patients taking 4 mg of the drug daily for 6 months with 10 patients taking placebo. All of the patients had a confirmed diagnosis of AD or MCI and averaged 73 years of age. Patients taking rosiglitazone did have significantly lower fasting plasma insulin levels than control patients, but there were no differences between the groups on fasting glucose, lipids, or liver enzymes (Am. J. Geriatr. Psychiatry 2005;13:950–8).

At 4 and 6 months into the study, patients taking rosiglitazone performed significantly better than those taking placebo on a delayed memory task (the Buschke Selective Reminding Test). At the 6-month time point patients taking rosiglitazone performed significantly better on a measure of selective attention (the Stroop Color-Word Test).

The other trial, which has not yet been published, involved 71 patients over the age of 55 with IGT or T2DM. Patients were randomly assigned to receive placebo, pioglitazone (30 mg/day), or nateglinide (360 mg/day) for 4 months.

Compared with baseline, patients taking pioglitazone showed significant improvement in performance on a story recall test, while patients taking nateglinide or placebo showed no such improvement. Furthermore, among the patients taking pioglitazone, the improvement was directly proportional to the extent of their metabolic treatment response, as measured by a 2-hour oral glucose tolerance test.

A subset of these patients received PET scans at baseline and after 4 months of treatment. As expected, insulin-resistant patients showed hypometabolism in the left temporal lobe and the right parietal region, both areas that are known to be affected in the very early stages of AD.

In addition, while performing a task of word memory in the PET scanner, normal subjects showed a pattern of increased glucose metabolism in the right frontal lobe. Patients with IGT showed a significantly smaller increase in glucose metabolism in that area, and patients with T2DM showed a smaller increase still.

Pioglitazone and rosiglitazone were both tolerated well in the two trials.

Although these studies are small and preliminary, if confirmed in larger studies, they open the possibility that insulin sensitizers may provide an effective treatment for the cognitive decline seen in Alzheimer's disease.

SAN FRANCISCO — The optimal time to initiate antiretroviral therapy during pregnancy depends on a balance of factors, Dr. Deborah Cohan said at a meeting on HIV management sponsored by the University of California, San Francisco.

The primary goal is viral suppression by the third trimester to minimize the chances of HIV transmission to the fetus. At least one study shows that the median time to viral suppression is about 50 days in pregnant women, although 10% fail to achieve total suppression within 6.5 months.

“In the United States we tend to start antiretrovirals between 12 and 14 weeks or beyond,” said Dr. Cohan of the University of California, San Francisco. Many women “feel pretty bad in the first trimester, and the last thing we want is for them to…attribute their nausea and vomiting to the antiretrovirals.”

Women who have morning sickness may vomit up some of their medication, and that can create worries about whether and how to redose. Fortunately, the weight of the evidence is that transmission does not occur in the first trimester, so antiretroviral therapy may not be crucial during that time.

There are some situations, however, in which antiretroviral therapy would be appropriate during the first trimester. For example, if the woman is continuing her preconception regimen, and the regimen includes only nonteratogenic medications that are well tolerated, it need not be discontinued.

First-trimester antiretrovirals also are indicated in patients who need them immediately for their own health.

For women who fail to tolerate their preconception regimen during the first trimester despite the use of antiemetics, the recommendation is to discontinue all medications at once. The one exception is if the patient is on a regimen containing nonnucleoside reverse transcription inhibitors. In that case, discontinuation should be staggered.

The principles of determining a proper antiretroviral regimen are similar in pregnant and nonpregnant women, except for one major consideration: Is this for her own health, and is it going to be a long-term regimen, or is it strictly chemoprophylaxis to prevent transmission?

If it's strictly chemoprophylaxis, less potent regimens may be acceptable. These could include triple nucleoside reverse transcriptase inhibitors. “Triple nukes really have fallen out of favor in terms of chronic use in adults,” Dr. Cohan said. “In this setting it may be appropriate. If someone comes to you with a CD4 count of 600 [cells/mm

Nelfinavir, a less potent protease inhibitor that has fallen out of favor, also is an option. It tends to be quite well tolerated in pregnancy, and in fact can counter the constipation that pregnant women frequently experience.

Another question concerns whether antiretroviral therapy is needed in pregnant women with viral loads less than 1,000 cells/mL. One as yet unpublished study of more than 1,200 woman-infant pairs determined the transmission rate to be 9.8% among women with low viral loads who don't get antiretroviral therapy, compared with 1.0% for women who do, yielding a highly significant odds ratio of 0.10.

Finally, there's the question of what one should do for women who are unlikely to comply with an antiretroviral regimen because of their life circumstances. Another unpublished study looked at the cost-effectiveness of directly observed therapy, which requires keeping women in the hospital during the third trimester. This resulted in a greatly reduced transmission rate and a cost saving of $3,200 per pregnancy.

Dr. Cohan routinely orders directly observed therapy for women in difficult circumstances.

SAN FRANCISCO — The optimal time to initiate antiretroviral therapy during pregnancy depends on a balance of factors, Dr. Deborah Cohan said at a meeting on HIV management sponsored by the University of California, San Francisco.

The primary goal is viral suppression by the third trimester to minimize the chances of HIV transmission to the fetus. At least one study shows that the median time to viral suppression is about 50 days in pregnant women, although 10% fail to achieve total suppression within 6.5 months.

“In the United States we tend to start antiretrovirals between 12 and 14 weeks or beyond,” said Dr. Cohan of the University of California, San Francisco. Many women “feel pretty bad in the first trimester, and the last thing we want is for them to…attribute their nausea and vomiting to the antiretrovirals.”

Women who have morning sickness may vomit up some of their medication, and that can create worries about whether and how to redose. Fortunately, the weight of the evidence is that transmission does not occur in the first trimester, so antiretroviral therapy may not be crucial during that time.

There are some situations, however, in which antiretroviral therapy would be appropriate during the first trimester. For example, if the woman is continuing her preconception regimen, and the regimen includes only nonteratogenic medications that are well tolerated, it need not be discontinued.

First-trimester antiretrovirals also are indicated in patients who need them immediately for their own health.

For women who fail to tolerate their preconception regimen during the first trimester despite the use of antiemetics, the recommendation is to discontinue all medications at once. The one exception is if the patient is on a regimen containing nonnucleoside reverse transcription inhibitors. In that case, discontinuation should be staggered.

The principles of determining a proper antiretroviral regimen are similar in pregnant and nonpregnant women, except for one major consideration: Is this for her own health, and is it going to be a long-term regimen, or is it strictly chemoprophylaxis to prevent transmission?

If it's strictly chemoprophylaxis, less potent regimens may be acceptable. These could include triple nucleoside reverse transcriptase inhibitors. “Triple nukes really have fallen out of favor in terms of chronic use in adults,” Dr. Cohan said. “In this setting it may be appropriate. If someone comes to you with a CD4 count of 600 [cells/mm

Nelfinavir, a less potent protease inhibitor that has fallen out of favor, also is an option. It tends to be quite well tolerated in pregnancy, and in fact can counter the constipation that pregnant women frequently experience.

Another question concerns whether antiretroviral therapy is needed in pregnant women with viral loads less than 1,000 cells/mL. One as yet unpublished study of more than 1,200 woman-infant pairs determined the transmission rate to be 9.8% among women with low viral loads who don't get antiretroviral therapy, compared with 1.0% for women who do, yielding a highly significant odds ratio of 0.10.

Finally, there's the question of what one should do for women who are unlikely to comply with an antiretroviral regimen because of their life circumstances. Another unpublished study looked at the cost-effectiveness of directly observed therapy, which requires keeping women in the hospital during the third trimester. This resulted in a greatly reduced transmission rate and a cost saving of $3,200 per pregnancy.

Dr. Cohan routinely orders directly observed therapy for women in difficult circumstances.

SAN FRANCISCO — The optimal time to initiate antiretroviral therapy during pregnancy depends on a balance of factors, Dr. Deborah Cohan said at a meeting on HIV management sponsored by the University of California, San Francisco.

The primary goal is viral suppression by the third trimester to minimize the chances of HIV transmission to the fetus. At least one study shows that the median time to viral suppression is about 50 days in pregnant women, although 10% fail to achieve total suppression within 6.5 months.

“In the United States we tend to start antiretrovirals between 12 and 14 weeks or beyond,” said Dr. Cohan of the University of California, San Francisco. Many women “feel pretty bad in the first trimester, and the last thing we want is for them to…attribute their nausea and vomiting to the antiretrovirals.”

Women who have morning sickness may vomit up some of their medication, and that can create worries about whether and how to redose. Fortunately, the weight of the evidence is that transmission does not occur in the first trimester, so antiretroviral therapy may not be crucial during that time.

There are some situations, however, in which antiretroviral therapy would be appropriate during the first trimester. For example, if the woman is continuing her preconception regimen, and the regimen includes only nonteratogenic medications that are well tolerated, it need not be discontinued.

First-trimester antiretrovirals also are indicated in patients who need them immediately for their own health.

For women who fail to tolerate their preconception regimen during the first trimester despite the use of antiemetics, the recommendation is to discontinue all medications at once. The one exception is if the patient is on a regimen containing nonnucleoside reverse transcription inhibitors. In that case, discontinuation should be staggered.

The principles of determining a proper antiretroviral regimen are similar in pregnant and nonpregnant women, except for one major consideration: Is this for her own health, and is it going to be a long-term regimen, or is it strictly chemoprophylaxis to prevent transmission?

If it's strictly chemoprophylaxis, less potent regimens may be acceptable. These could include triple nucleoside reverse transcriptase inhibitors. “Triple nukes really have fallen out of favor in terms of chronic use in adults,” Dr. Cohan said. “In this setting it may be appropriate. If someone comes to you with a CD4 count of 600 [cells/mm

Nelfinavir, a less potent protease inhibitor that has fallen out of favor, also is an option. It tends to be quite well tolerated in pregnancy, and in fact can counter the constipation that pregnant women frequently experience.

Another question concerns whether antiretroviral therapy is needed in pregnant women with viral loads less than 1,000 cells/mL. One as yet unpublished study of more than 1,200 woman-infant pairs determined the transmission rate to be 9.8% among women with low viral loads who don't get antiretroviral therapy, compared with 1.0% for women who do, yielding a highly significant odds ratio of 0.10.

Finally, there's the question of what one should do for women who are unlikely to comply with an antiretroviral regimen because of their life circumstances. Another unpublished study looked at the cost-effectiveness of directly observed therapy, which requires keeping women in the hospital during the third trimester. This resulted in a greatly reduced transmission rate and a cost saving of $3,200 per pregnancy.

Dr. Cohan routinely orders directly observed therapy for women in difficult circumstances.

SAN FRANCISCO — The liver biopsy remains the preferred method for diagnosing nonalcoholic fatty liver disease, but biopsy candidates should be chosen with care. Not all patients with signs of the disease will require a biopsy, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

Patients who are eventually diagnosed with nonalcoholic fatty liver disease (NAFLD) present initially in a variety of ways, said Dr. Bass of the University of California, San Francisco.

For example, an insurance exam can turn up an incidental aminotransferase elevation or an enlarged liver. An abdominal imaging study may reveal a fatty liver. A patient may have a complication of cirrhosis. Or NAFLD patients may be identified by screening high-risk populations with liver enzyme tests or liver ultrasound. An increasing number of NAFLD patients are also being identified by liver biopsy during weight-reduction surgery, he said.

But it's not practical or desirable to screen all at-risk patients with a biopsy, and there are some good reasons not to do so. (See box.) About 25% of patients will experience significant pain during the biopsy, and 1%–3.5% of patients will have morbidities such as hypotension, pneumothorax, hemoperitoneum, hemobilia, and gall bladder penetration. About 0.1% of patients will die from the procedure.

There are five situations in which a liver biopsy is essential: when a patient's liver enzymes show an unusual pattern or are 3–5 times normal; when other liver disease cannot be excluded; when the patient does not have metabolic syndrome; to confirm a clinical suspicion of cirrhosis; and for qualifying a patient for entry into a clinical trial.

Although a definitive diagnosis still requires a biopsy, there are several alternatives for assessing the liver, Dr. Bass said.

Elevated liver enzymes can be suggestive of NAFLD, but in a phenomenon Dr. Bass called “The Silence of the Labs,” some patients with NAFLD have normal liver enzymes. He cited one study of patients undergoing gastric bypass in which 68% had normal ALT and AST, but only 52% had a normal liver biopsy. In the remaining 48% with abnormal biopsy results, about 27% had nonalcoholic fatty liver, and the others had nonalcoholic steatohepatitis.

An NAFLD diagnosis is often made by exclusion—after alcoholic liver disease, drug-induced liver injury, iron overload, hepatitis B and C, and autoimmune hepatitis have been excluded.

It's difficult to exclude a significant contribution from alcohol, because patients are not always truthful. For inclusion in clinical trials, NIH defines “nonalcoholic” as less than 14 units of alcohol per week for men or less than 7 units per week for women. A unit is one can of beer, one glass of wine, or one shot of hard liquor.

Combining ultrasound evidence of fatty liver and liver enzyme elevation without markers for hepatitis C or B yields a 96% positive predictive value for NAFLD, according to one study. However, ultrasound is sensitive, but not very specific. CT imaging is somewhat more specific. In CT, a normal liver has about the same density as the spleen; in NAFLD, the spleen is brighter. But CT is too costly for routine screening. At least three serological tests for hepatic fibrosis are being developed, Dr. Bass said. Transient elastography, combining 5-MHz ultrasound and 50-Hz elastic waves, may also help diagnosis.

Biopsy's Pros And Cons

Pros

▸ Grade and stage of NAFLD are determined.

▸ Confidence in the diagnosis is 100%.

▸ Patients are motivated to lose weight.

▸ Biopsy is essential for enrollment in clinical trials of treatments.

Cons

▸ Risk of morbidity is increased with biopsy.

▸ Noninvasive diagnosis is quite accurate.

▸ Natural history of NAFLD is benign in most patients.

▸ NAFLD is a common disorder.

▸ There is no proven, specific treatment for NAFLD.

Source: Dr. Bass

SAN FRANCISCO — The liver biopsy remains the preferred method for diagnosing nonalcoholic fatty liver disease, but biopsy candidates should be chosen with care. Not all patients with signs of the disease will require a biopsy, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

Patients who are eventually diagnosed with nonalcoholic fatty liver disease (NAFLD) present initially in a variety of ways, said Dr. Bass of the University of California, San Francisco.

For example, an insurance exam can turn up an incidental aminotransferase elevation or an enlarged liver. An abdominal imaging study may reveal a fatty liver. A patient may have a complication of cirrhosis. Or NAFLD patients may be identified by screening high-risk populations with liver enzyme tests or liver ultrasound. An increasing number of NAFLD patients are also being identified by liver biopsy during weight-reduction surgery, he said.

But it's not practical or desirable to screen all at-risk patients with a biopsy, and there are some good reasons not to do so. (See box.) About 25% of patients will experience significant pain during the biopsy, and 1%–3.5% of patients will have morbidities such as hypotension, pneumothorax, hemoperitoneum, hemobilia, and gall bladder penetration. About 0.1% of patients will die from the procedure.

There are five situations in which a liver biopsy is essential: when a patient's liver enzymes show an unusual pattern or are 3–5 times normal; when other liver disease cannot be excluded; when the patient does not have metabolic syndrome; to confirm a clinical suspicion of cirrhosis; and for qualifying a patient for entry into a clinical trial.

Although a definitive diagnosis still requires a biopsy, there are several alternatives for assessing the liver, Dr. Bass said.

Elevated liver enzymes can be suggestive of NAFLD, but in a phenomenon Dr. Bass called “The Silence of the Labs,” some patients with NAFLD have normal liver enzymes. He cited one study of patients undergoing gastric bypass in which 68% had normal ALT and AST, but only 52% had a normal liver biopsy. In the remaining 48% with abnormal biopsy results, about 27% had nonalcoholic fatty liver, and the others had nonalcoholic steatohepatitis.

An NAFLD diagnosis is often made by exclusion—after alcoholic liver disease, drug-induced liver injury, iron overload, hepatitis B and C, and autoimmune hepatitis have been excluded.

It's difficult to exclude a significant contribution from alcohol, because patients are not always truthful. For inclusion in clinical trials, NIH defines “nonalcoholic” as less than 14 units of alcohol per week for men or less than 7 units per week for women. A unit is one can of beer, one glass of wine, or one shot of hard liquor.

Combining ultrasound evidence of fatty liver and liver enzyme elevation without markers for hepatitis C or B yields a 96% positive predictive value for NAFLD, according to one study. However, ultrasound is sensitive, but not very specific. CT imaging is somewhat more specific. In CT, a normal liver has about the same density as the spleen; in NAFLD, the spleen is brighter. But CT is too costly for routine screening. At least three serological tests for hepatic fibrosis are being developed, Dr. Bass said. Transient elastography, combining 5-MHz ultrasound and 50-Hz elastic waves, may also help diagnosis.

Biopsy's Pros And Cons

Pros

▸ Grade and stage of NAFLD are determined.

▸ Confidence in the diagnosis is 100%.

▸ Patients are motivated to lose weight.

▸ Biopsy is essential for enrollment in clinical trials of treatments.

Cons

▸ Risk of morbidity is increased with biopsy.

▸ Noninvasive diagnosis is quite accurate.

▸ Natural history of NAFLD is benign in most patients.

▸ NAFLD is a common disorder.

▸ There is no proven, specific treatment for NAFLD.

Source: Dr. Bass

SAN FRANCISCO — The liver biopsy remains the preferred method for diagnosing nonalcoholic fatty liver disease, but biopsy candidates should be chosen with care. Not all patients with signs of the disease will require a biopsy, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

Patients who are eventually diagnosed with nonalcoholic fatty liver disease (NAFLD) present initially in a variety of ways, said Dr. Bass of the University of California, San Francisco.

For example, an insurance exam can turn up an incidental aminotransferase elevation or an enlarged liver. An abdominal imaging study may reveal a fatty liver. A patient may have a complication of cirrhosis. Or NAFLD patients may be identified by screening high-risk populations with liver enzyme tests or liver ultrasound. An increasing number of NAFLD patients are also being identified by liver biopsy during weight-reduction surgery, he said.

But it's not practical or desirable to screen all at-risk patients with a biopsy, and there are some good reasons not to do so. (See box.) About 25% of patients will experience significant pain during the biopsy, and 1%–3.5% of patients will have morbidities such as hypotension, pneumothorax, hemoperitoneum, hemobilia, and gall bladder penetration. About 0.1% of patients will die from the procedure.

There are five situations in which a liver biopsy is essential: when a patient's liver enzymes show an unusual pattern or are 3–5 times normal; when other liver disease cannot be excluded; when the patient does not have metabolic syndrome; to confirm a clinical suspicion of cirrhosis; and for qualifying a patient for entry into a clinical trial.

Although a definitive diagnosis still requires a biopsy, there are several alternatives for assessing the liver, Dr. Bass said.

Elevated liver enzymes can be suggestive of NAFLD, but in a phenomenon Dr. Bass called “The Silence of the Labs,” some patients with NAFLD have normal liver enzymes. He cited one study of patients undergoing gastric bypass in which 68% had normal ALT and AST, but only 52% had a normal liver biopsy. In the remaining 48% with abnormal biopsy results, about 27% had nonalcoholic fatty liver, and the others had nonalcoholic steatohepatitis.

An NAFLD diagnosis is often made by exclusion—after alcoholic liver disease, drug-induced liver injury, iron overload, hepatitis B and C, and autoimmune hepatitis have been excluded.

It's difficult to exclude a significant contribution from alcohol, because patients are not always truthful. For inclusion in clinical trials, NIH defines “nonalcoholic” as less than 14 units of alcohol per week for men or less than 7 units per week for women. A unit is one can of beer, one glass of wine, or one shot of hard liquor.

Combining ultrasound evidence of fatty liver and liver enzyme elevation without markers for hepatitis C or B yields a 96% positive predictive value for NAFLD, according to one study. However, ultrasound is sensitive, but not very specific. CT imaging is somewhat more specific. In CT, a normal liver has about the same density as the spleen; in NAFLD, the spleen is brighter. But CT is too costly for routine screening. At least three serological tests for hepatic fibrosis are being developed, Dr. Bass said. Transient elastography, combining 5-MHz ultrasound and 50-Hz elastic waves, may also help diagnosis.

Biopsy's Pros And Cons

Pros

▸ Grade and stage of NAFLD are determined.

▸ Confidence in the diagnosis is 100%.

▸ Patients are motivated to lose weight.

▸ Biopsy is essential for enrollment in clinical trials of treatments.

Cons

▸ Risk of morbidity is increased with biopsy.

▸ Noninvasive diagnosis is quite accurate.

▸ Natural history of NAFLD is benign in most patients.

▸ NAFLD is a common disorder.

▸ There is no proven, specific treatment for NAFLD.

Source: Dr. Bass

SAN FRANCISCO — With no specific treatment available for nonalcoholic fatty liver disease, the best current strategy centers on monitoring the patient's condition and managing the patient's lifestyle and metabolic syndrome, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

The patient's liver enzymes, liver function (bilirubin levels, albumin levels, prothrombin time), and platelet count should be monitored. Each patient also should undergo regular ultrasound exams. Patients should be instructed to avoid hepatotoxins—most notably, alcohol—and should be advised to pursue gradual weight loss with diet and exercise.

“Weight loss remains the simplest advice you can give,” said Dr. Bass of the University of California, San Francisco, citing a study showing even modest weight loss (less than 10% of initial body weight) can reduce intrahepatic fat while leaving intramuscular fat unchanged. Such weight loss also improved basal and insulin-stimulated glucose metabolism (Diabetes 2005;54:603–8).

Bariatric surgery can be helpful for some patients with nonalcoholic fatty liver disease (NAFLD), but it should be the newer restrictive surgery involving gastric banding, which tends to decrease steatosis, fibrosis, and nonalcoholic steatohepatitis. Older malabsorptive surgical strategies can be dangerous; they can lead to increased steatosis, fibrosis, nonalcoholic steatohepatitis, and liver failure. The insulin-sensitizing agent metformin appears to be helpful in NAFLD; but the published studies tend to be small and open label, so the evidence base is not overwhelming. The thiazolidinediones pioglitazone and troglitazone seem to improve liver enzymes and fibrosis measured histologically, but again, the evidence is from open-label trials.

Dr. Bass noted some caveats with thiazolidinediones: They can cause weight gain and relapse upon discontinuation, and some patients experience serious side effects such as congestive heart failure and hepatotoxicity.

SAN FRANCISCO — With no specific treatment available for nonalcoholic fatty liver disease, the best current strategy centers on monitoring the patient's condition and managing the patient's lifestyle and metabolic syndrome, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

The patient's liver enzymes, liver function (bilirubin levels, albumin levels, prothrombin time), and platelet count should be monitored. Each patient also should undergo regular ultrasound exams. Patients should be instructed to avoid hepatotoxins—most notably, alcohol—and should be advised to pursue gradual weight loss with diet and exercise.

“Weight loss remains the simplest advice you can give,” said Dr. Bass of the University of California, San Francisco, citing a study showing even modest weight loss (less than 10% of initial body weight) can reduce intrahepatic fat while leaving intramuscular fat unchanged. Such weight loss also improved basal and insulin-stimulated glucose metabolism (Diabetes 2005;54:603–8).

Bariatric surgery can be helpful for some patients with nonalcoholic fatty liver disease (NAFLD), but it should be the newer restrictive surgery involving gastric banding, which tends to decrease steatosis, fibrosis, and nonalcoholic steatohepatitis. Older malabsorptive surgical strategies can be dangerous; they can lead to increased steatosis, fibrosis, nonalcoholic steatohepatitis, and liver failure. The insulin-sensitizing agent metformin appears to be helpful in NAFLD; but the published studies tend to be small and open label, so the evidence base is not overwhelming. The thiazolidinediones pioglitazone and troglitazone seem to improve liver enzymes and fibrosis measured histologically, but again, the evidence is from open-label trials.

Dr. Bass noted some caveats with thiazolidinediones: They can cause weight gain and relapse upon discontinuation, and some patients experience serious side effects such as congestive heart failure and hepatotoxicity.

SAN FRANCISCO — With no specific treatment available for nonalcoholic fatty liver disease, the best current strategy centers on monitoring the patient's condition and managing the patient's lifestyle and metabolic syndrome, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

The patient's liver enzymes, liver function (bilirubin levels, albumin levels, prothrombin time), and platelet count should be monitored. Each patient also should undergo regular ultrasound exams. Patients should be instructed to avoid hepatotoxins—most notably, alcohol—and should be advised to pursue gradual weight loss with diet and exercise.

“Weight loss remains the simplest advice you can give,” said Dr. Bass of the University of California, San Francisco, citing a study showing even modest weight loss (less than 10% of initial body weight) can reduce intrahepatic fat while leaving intramuscular fat unchanged. Such weight loss also improved basal and insulin-stimulated glucose metabolism (Diabetes 2005;54:603–8).

Bariatric surgery can be helpful for some patients with nonalcoholic fatty liver disease (NAFLD), but it should be the newer restrictive surgery involving gastric banding, which tends to decrease steatosis, fibrosis, and nonalcoholic steatohepatitis. Older malabsorptive surgical strategies can be dangerous; they can lead to increased steatosis, fibrosis, nonalcoholic steatohepatitis, and liver failure. The insulin-sensitizing agent metformin appears to be helpful in NAFLD; but the published studies tend to be small and open label, so the evidence base is not overwhelming. The thiazolidinediones pioglitazone and troglitazone seem to improve liver enzymes and fibrosis measured histologically, but again, the evidence is from open-label trials.

Dr. Bass noted some caveats with thiazolidinediones: They can cause weight gain and relapse upon discontinuation, and some patients experience serious side effects such as congestive heart failure and hepatotoxicity.

SAN FRANCISCO — Diabetes treatment is about to acquire several more layers of complexity as new classes of drugs become available, Dr. David M. Kendall said at a conference sponsored by the American Diabetes Association.

Incretin mimetics, dipeptidyl peptidase IV (DPP-IV) inhibitors, dual peroxisome proliferator-activated receptor (PPAR) activators, amylin analogues, and cannabinoid receptor blockers—not to mention inhaled forms of insulin—are all wending their way through clinical trials.

Some drugs in these classes already have been approved, most notably the incretin mimetic exenatide (Byetta), an inhaled insulin (Exubera), and the amylin analogue pramlintide (Symlin). Others are in the late stages of clinical trials and may be approved later this year.

A cookbook approach to using these new drugs is unlikely to emerge, said Dr. Kendall, a diabetologist currently serving as executive director for medical affairs at Amylin Pharmaceuticals, San Diego. Instead, clinicians will have to consider which classes of compounds, and which drugs within those classes, are likely to benefit individual patients, depending on their clinical characteristics.

In his talk, Dr. Kendall commented on how these compounds work and what clinical evidence is available to assist in matching the drug to the patient.

The “incretin effect” refers to gut-related factors that increase insulin levels rapidly in response to oral glucose. Several drugs are being developed that are analogues of one of the two incretins known—glucagon-like peptide 1 (GLP-1).

In addition to their direct effect on pancreatic islet cells, GLP-1 analogues can help restore appropriate suppression of glucagon secretion, and thus limit hepatic glucose output after a meal. Furthermore, these drugs slow the rate of gastric emptying and increase satiety, leading to moderate weight loss.

Several GLP-1 analogues are under development. Exenatide already has been approved for patients with type 2 diabetes who fail to achieve glycemic control with metformin or a sulfonylurea. Similar compounds are not far behind, including liraglutide and a compound currently designated CJC-1134.

Dr. Kendall referred to exenatide and other incretin mimetics as “smart bombs for diabetes,” since they lead to a “profound restoration of phasic insulin secretion,” a decrease in postmeal glucose excursions, and reductions in hemoglobin A_1c (HbA_1c) along with weight loss. About 75%–80% of patients respond to these compounds.

DPP-IV rapidly degrades incretins, and Dr. Kendall mentioned at least four DPP-IV inhibitors under development. Several, including vildagliptin and sitagliptin, have reached phase III clinical trials. Orally administered, these compounds lead to sustained glucose control, and HbA_1c reductions of about 0.7%. They appear to be weight neutral, and they're well tolerated, but the response rate is not yet known.

Dual PPAR activators are agonists for both PPAR-a and PPAR-g, and so may have beneficial effects on both insulin sensitivity and lipid metabolism. Several are under development, including muraglitazar, tesaglitazar, and naveglitazar.

Dr. Kendall mentioned a head-to-head study comparing muraglitazar and pioglitazone. They appear to have similar efficacy in improving insulin sensitivity, but muraglitazar was significantly better at reducing HbA_1c and improving levels of triglycerides and HDL cholesterol.

“This is a two for one,” Dr. Kendall said, that's particularly suited for “anyone with type 2 diabetes and insulin resistance who's at higher risk for cardiovascular disease, particularly those who have a specific need to target abnormalities in HDL cholesterol and triglycerides.”

Amylin is a pancreatic islet hormone that's cosecreted with insulin. It's been demonstrated to lower plasma glucagon levels in patients with both type 1 and type 2 diabetes, to regulate rates of gastric emptying, and to be associated with reductions in food intake and moderate weight loss. Amylin levels are deficient in patients with type 1 and type 2 diabetes.

Pramlintide, the first amylin analogue, was approved by the Food and Drug Administration in 2005 as an insulin adjunct in patients with either type 1 or type 2 diabetes.

It appears to lower the dose of bolus insulin required for glycemic control by 50%, it leads to a stabilization of plasma glucose, it decreases HbA_1c by 0.4%–0.7%, and it decreases food intake, leading to weight loss of 2 kg or more.

While not directed specifically at diabetes, cannabinoid receptor blockers are likely to be useful in this population as an adjunct to diet and exercise for the management of obesity. Rimonabant (Acomplia) is the furthest along in clinical development.

SAN FRANCISCO — Diabetes treatment is about to acquire several more layers of complexity as new classes of drugs become available, Dr. David M. Kendall said at a conference sponsored by the American Diabetes Association.

Incretin mimetics, dipeptidyl peptidase IV (DPP-IV) inhibitors, dual peroxisome proliferator-activated receptor (PPAR) activators, amylin analogues, and cannabinoid receptor blockers—not to mention inhaled forms of insulin—are all wending their way through clinical trials.

Some drugs in these classes already have been approved, most notably the incretin mimetic exenatide (Byetta), an inhaled insulin (Exubera), and the amylin analogue pramlintide (Symlin). Others are in the late stages of clinical trials and may be approved later this year.

A cookbook approach to using these new drugs is unlikely to emerge, said Dr. Kendall, a diabetologist currently serving as executive director for medical affairs at Amylin Pharmaceuticals, San Diego. Instead, clinicians will have to consider which classes of compounds, and which drugs within those classes, are likely to benefit individual patients, depending on their clinical characteristics.

In his talk, Dr. Kendall commented on how these compounds work and what clinical evidence is available to assist in matching the drug to the patient.

The “incretin effect” refers to gut-related factors that increase insulin levels rapidly in response to oral glucose. Several drugs are being developed that are analogues of one of the two incretins known—glucagon-like peptide 1 (GLP-1).

In addition to their direct effect on pancreatic islet cells, GLP-1 analogues can help restore appropriate suppression of glucagon secretion, and thus limit hepatic glucose output after a meal. Furthermore, these drugs slow the rate of gastric emptying and increase satiety, leading to moderate weight loss.

Several GLP-1 analogues are under development. Exenatide already has been approved for patients with type 2 diabetes who fail to achieve glycemic control with metformin or a sulfonylurea. Similar compounds are not far behind, including liraglutide and a compound currently designated CJC-1134.

Dr. Kendall referred to exenatide and other incretin mimetics as “smart bombs for diabetes,” since they lead to a “profound restoration of phasic insulin secretion,” a decrease in postmeal glucose excursions, and reductions in hemoglobin A_1c (HbA_1c) along with weight loss. About 75%–80% of patients respond to these compounds.

DPP-IV rapidly degrades incretins, and Dr. Kendall mentioned at least four DPP-IV inhibitors under development. Several, including vildagliptin and sitagliptin, have reached phase III clinical trials. Orally administered, these compounds lead to sustained glucose control, and HbA_1c reductions of about 0.7%. They appear to be weight neutral, and they're well tolerated, but the response rate is not yet known.

Dual PPAR activators are agonists for both PPAR-a and PPAR-g, and so may have beneficial effects on both insulin sensitivity and lipid metabolism. Several are under development, including muraglitazar, tesaglitazar, and naveglitazar.

Dr. Kendall mentioned a head-to-head study comparing muraglitazar and pioglitazone. They appear to have similar efficacy in improving insulin sensitivity, but muraglitazar was significantly better at reducing HbA_1c and improving levels of triglycerides and HDL cholesterol.

“This is a two for one,” Dr. Kendall said, that's particularly suited for “anyone with type 2 diabetes and insulin resistance who's at higher risk for cardiovascular disease, particularly those who have a specific need to target abnormalities in HDL cholesterol and triglycerides.”

Amylin is a pancreatic islet hormone that's cosecreted with insulin. It's been demonstrated to lower plasma glucagon levels in patients with both type 1 and type 2 diabetes, to regulate rates of gastric emptying, and to be associated with reductions in food intake and moderate weight loss. Amylin levels are deficient in patients with type 1 and type 2 diabetes.

Pramlintide, the first amylin analogue, was approved by the Food and Drug Administration in 2005 as an insulin adjunct in patients with either type 1 or type 2 diabetes.

It appears to lower the dose of bolus insulin required for glycemic control by 50%, it leads to a stabilization of plasma glucose, it decreases HbA_1c by 0.4%–0.7%, and it decreases food intake, leading to weight loss of 2 kg or more.

While not directed specifically at diabetes, cannabinoid receptor blockers are likely to be useful in this population as an adjunct to diet and exercise for the management of obesity. Rimonabant (Acomplia) is the furthest along in clinical development.

SAN FRANCISCO — Diabetes treatment is about to acquire several more layers of complexity as new classes of drugs become available, Dr. David M. Kendall said at a conference sponsored by the American Diabetes Association.

Incretin mimetics, dipeptidyl peptidase IV (DPP-IV) inhibitors, dual peroxisome proliferator-activated receptor (PPAR) activators, amylin analogues, and cannabinoid receptor blockers—not to mention inhaled forms of insulin—are all wending their way through clinical trials.

Some drugs in these classes already have been approved, most notably the incretin mimetic exenatide (Byetta), an inhaled insulin (Exubera), and the amylin analogue pramlintide (Symlin). Others are in the late stages of clinical trials and may be approved later this year.

A cookbook approach to using these new drugs is unlikely to emerge, said Dr. Kendall, a diabetologist currently serving as executive director for medical affairs at Amylin Pharmaceuticals, San Diego. Instead, clinicians will have to consider which classes of compounds, and which drugs within those classes, are likely to benefit individual patients, depending on their clinical characteristics.

In his talk, Dr. Kendall commented on how these compounds work and what clinical evidence is available to assist in matching the drug to the patient.

The “incretin effect” refers to gut-related factors that increase insulin levels rapidly in response to oral glucose. Several drugs are being developed that are analogues of one of the two incretins known—glucagon-like peptide 1 (GLP-1).

In addition to their direct effect on pancreatic islet cells, GLP-1 analogues can help restore appropriate suppression of glucagon secretion, and thus limit hepatic glucose output after a meal. Furthermore, these drugs slow the rate of gastric emptying and increase satiety, leading to moderate weight loss.

Several GLP-1 analogues are under development. Exenatide already has been approved for patients with type 2 diabetes who fail to achieve glycemic control with metformin or a sulfonylurea. Similar compounds are not far behind, including liraglutide and a compound currently designated CJC-1134.

Dr. Kendall referred to exenatide and other incretin mimetics as “smart bombs for diabetes,” since they lead to a “profound restoration of phasic insulin secretion,” a decrease in postmeal glucose excursions, and reductions in hemoglobin A_1c (HbA_1c) along with weight loss. About 75%–80% of patients respond to these compounds.

DPP-IV rapidly degrades incretins, and Dr. Kendall mentioned at least four DPP-IV inhibitors under development. Several, including vildagliptin and sitagliptin, have reached phase III clinical trials. Orally administered, these compounds lead to sustained glucose control, and HbA_1c reductions of about 0.7%. They appear to be weight neutral, and they're well tolerated, but the response rate is not yet known.

Dual PPAR activators are agonists for both PPAR-a and PPAR-g, and so may have beneficial effects on both insulin sensitivity and lipid metabolism. Several are under development, including muraglitazar, tesaglitazar, and naveglitazar.

Dr. Kendall mentioned a head-to-head study comparing muraglitazar and pioglitazone. They appear to have similar efficacy in improving insulin sensitivity, but muraglitazar was significantly better at reducing HbA_1c and improving levels of triglycerides and HDL cholesterol.

“This is a two for one,” Dr. Kendall said, that's particularly suited for “anyone with type 2 diabetes and insulin resistance who's at higher risk for cardiovascular disease, particularly those who have a specific need to target abnormalities in HDL cholesterol and triglycerides.”

Amylin is a pancreatic islet hormone that's cosecreted with insulin. It's been demonstrated to lower plasma glucagon levels in patients with both type 1 and type 2 diabetes, to regulate rates of gastric emptying, and to be associated with reductions in food intake and moderate weight loss. Amylin levels are deficient in patients with type 1 and type 2 diabetes.

Pramlintide, the first amylin analogue, was approved by the Food and Drug Administration in 2005 as an insulin adjunct in patients with either type 1 or type 2 diabetes.

It appears to lower the dose of bolus insulin required for glycemic control by 50%, it leads to a stabilization of plasma glucose, it decreases HbA_1c by 0.4%–0.7%, and it decreases food intake, leading to weight loss of 2 kg or more.

While not directed specifically at diabetes, cannabinoid receptor blockers are likely to be useful in this population as an adjunct to diet and exercise for the management of obesity. Rimonabant (Acomplia) is the furthest along in clinical development.

SAN FRANCISCO — A high risk for sleep apnea was common in women with polycystic ovary syndrome and was linked to high fasting insulin levels, Dr. Esra Tasali reported at a conference sponsored by the American Diabetes Association.

Women with PCOSand normal glucose tolerance who were at high risk for sleep apnea and were given oral glucose showed insulin levels that were twice those of women with lower risk for sleep apnea.

This finding suggests that sleep apnea might worsen the metabolic consequences of insulin resistance, accelerating the conversion from normal to impaired glucose tolerance, Dr. Tasali said.

Although the study does not establish causation, Dr. Tasali recommended that women with PCOS be systematically evaluated for sleep apnea, because its treatment might improve glucose metabolism.

A high risk for sleep apnea was observed in 30 of 40 women with PCOS, and 92% of the women had sleep problems, according to Dr. Tasali and her colleagues at the University of Chicago (J. Clin. Endocrinol. Metab. 2006;91:36–42).

Of the 40 women, 32 had previously been given an oral glucose tolerance test. Glucose tolerance was normal in 19 women. In 22 women at high sleep apnea risk, average fasting insulin levels were significantly higher (168 pmol/L) than they were in the 10 women at low apnea risk (97 pmol/L).

Among the 13 women with impaired glucose tolerance, glucose and insulin levels did not differ depending on the level of apnea risk.

Another cohort of eight women with PCOS underwent overnight polysomnography for symptoms suggestive of obstructive sleep apnea. Mean sleep efficiency was 80% in the women with PCOS, compared with 92% in a control group of age-matched, nonobese women. The women with PCOS also had significantly longer mean sleep latency (41 minutes vs. 10 minutes), and significantly shorter total sleep time (323 minutes vs. 442 minutes, a difference of almost 2 hours).

“Sleep apnea might be an intrinsic component of the metabolic disturbances that appear with polycystic ovary syn- drome,” said Dr. Tasali.

Furthermore, severity of sleep apnea as measured by the apnea-hypopnea index, and the degree of oxygen desaturations during rapid-eye-movement sleep, accounted for more than 90% of the variability in measures of glucose tolerance including hemoglobin A_1c levels.

Together, these findings could mean that both glucose tolerance and sleep apnea are strongly influenced by a common mechanism in women with PCOS.

Dr. Tasali disclosed no conflicts of interest related to her presentation.

SAN FRANCISCO — A high risk for sleep apnea was common in women with polycystic ovary syndrome and was linked to high fasting insulin levels, Dr. Esra Tasali reported at a conference sponsored by the American Diabetes Association.

Women with PCOSand normal glucose tolerance who were at high risk for sleep apnea and were given oral glucose showed insulin levels that were twice those of women with lower risk for sleep apnea.

This finding suggests that sleep apnea might worsen the metabolic consequences of insulin resistance, accelerating the conversion from normal to impaired glucose tolerance, Dr. Tasali said.

Although the study does not establish causation, Dr. Tasali recommended that women with PCOS be systematically evaluated for sleep apnea, because its treatment might improve glucose metabolism.

A high risk for sleep apnea was observed in 30 of 40 women with PCOS, and 92% of the women had sleep problems, according to Dr. Tasali and her colleagues at the University of Chicago (J. Clin. Endocrinol. Metab. 2006;91:36–42).

Of the 40 women, 32 had previously been given an oral glucose tolerance test. Glucose tolerance was normal in 19 women. In 22 women at high sleep apnea risk, average fasting insulin levels were significantly higher (168 pmol/L) than they were in the 10 women at low apnea risk (97 pmol/L).

Among the 13 women with impaired glucose tolerance, glucose and insulin levels did not differ depending on the level of apnea risk.

Another cohort of eight women with PCOS underwent overnight polysomnography for symptoms suggestive of obstructive sleep apnea. Mean sleep efficiency was 80% in the women with PCOS, compared with 92% in a control group of age-matched, nonobese women. The women with PCOS also had significantly longer mean sleep latency (41 minutes vs. 10 minutes), and significantly shorter total sleep time (323 minutes vs. 442 minutes, a difference of almost 2 hours).

“Sleep apnea might be an intrinsic component of the metabolic disturbances that appear with polycystic ovary syn- drome,” said Dr. Tasali.

Furthermore, severity of sleep apnea as measured by the apnea-hypopnea index, and the degree of oxygen desaturations during rapid-eye-movement sleep, accounted for more than 90% of the variability in measures of glucose tolerance including hemoglobin A_1c levels.

Together, these findings could mean that both glucose tolerance and sleep apnea are strongly influenced by a common mechanism in women with PCOS.

Dr. Tasali disclosed no conflicts of interest related to her presentation.

SAN FRANCISCO — A high risk for sleep apnea was common in women with polycystic ovary syndrome and was linked to high fasting insulin levels, Dr. Esra Tasali reported at a conference sponsored by the American Diabetes Association.

Women with PCOSand normal glucose tolerance who were at high risk for sleep apnea and were given oral glucose showed insulin levels that were twice those of women with lower risk for sleep apnea.

This finding suggests that sleep apnea might worsen the metabolic consequences of insulin resistance, accelerating the conversion from normal to impaired glucose tolerance, Dr. Tasali said.

Although the study does not establish causation, Dr. Tasali recommended that women with PCOS be systematically evaluated for sleep apnea, because its treatment might improve glucose metabolism.

A high risk for sleep apnea was observed in 30 of 40 women with PCOS, and 92% of the women had sleep problems, according to Dr. Tasali and her colleagues at the University of Chicago (J. Clin. Endocrinol. Metab. 2006;91:36–42).

Of the 40 women, 32 had previously been given an oral glucose tolerance test. Glucose tolerance was normal in 19 women. In 22 women at high sleep apnea risk, average fasting insulin levels were significantly higher (168 pmol/L) than they were in the 10 women at low apnea risk (97 pmol/L).

Among the 13 women with impaired glucose tolerance, glucose and insulin levels did not differ depending on the level of apnea risk.

Another cohort of eight women with PCOS underwent overnight polysomnography for symptoms suggestive of obstructive sleep apnea. Mean sleep efficiency was 80% in the women with PCOS, compared with 92% in a control group of age-matched, nonobese women. The women with PCOS also had significantly longer mean sleep latency (41 minutes vs. 10 minutes), and significantly shorter total sleep time (323 minutes vs. 442 minutes, a difference of almost 2 hours).

“Sleep apnea might be an intrinsic component of the metabolic disturbances that appear with polycystic ovary syn- drome,” said Dr. Tasali.

Furthermore, severity of sleep apnea as measured by the apnea-hypopnea index, and the degree of oxygen desaturations during rapid-eye-movement sleep, accounted for more than 90% of the variability in measures of glucose tolerance including hemoglobin A_1c levels.

Together, these findings could mean that both glucose tolerance and sleep apnea are strongly influenced by a common mechanism in women with PCOS.

Dr. Tasali disclosed no conflicts of interest related to her presentation.

SAN FRANCISCO — Although nonnucleoside reverse transcriptase inhibitors form one of the mainstays of highly active antiretroviral therapy for HIV, they have a prominent Achilles heel: HIV can develop resistance to the entire NNRTI class with simple, single-point mutations.

Several second-generation nonnucleoside reverse transcriptase inhibitors (NNRTIs) are now in clinical trials, Dr. C. Bradley Hare said at a meeting on HIV management sponsored by the University of California, San Francisco. These drugs have been designed to require HIV to develop multiple mutations for resistance to take place.

Dr. Hare of the university discussed the results of current trials for the following compounds:

▸ Etravirine (TMC125). Developed by Tibotec, etravirine showed good in vitro activity against viruses with one of several NNRTI-resistant mutations. In 8-day, phase I, monotherapy studies in NNRTI-naive patients, etravirine resulted in viral loads declining by two orders of magnitude. In NNRTI-resistant patients treated for 8 days, the decline was smaller—one order of magnitude—but still impressive. Dosing is being worked out, but etravirine is likely to require twice-daily dosing. At a dosage of 800 mg b.i.d. in phase II research, etravirine is beating placebo in patients with NNRTI resistance and in patients with three-class drug failure. No hepatotoxicity or CNS toxicity has been observed, but 17% of patients have developed a rash.

▸ TMC278 (not yet named). TMC278, also from Tibotec, has not progressed beyond phase I studies. It has a longer half-life than etravirine, which may allow once-daily dosing. In a 7-day monotherapy trial in 47 patients, all three dose levels of the drug resulted in viral-load declines of more than one order of magnitude. Adverse events were mostly headache and nausea. None of the patients developed NNRTI mutations during the short trial.

▸ Capavirine (AG1549). In early studies, capavirine, developed by Pfizer Inc., was active against some viruses resistant to the first-generation NNRTIs efavirenz and navirapine. It seemed effective as monotherapy in NNRTI-naive patients. But a randomized, phase II study with 179 patients who had failed other NNRTIs was disappointing. Development of the drug was discontinued in 2005.

Dr. Hare disclosed financial ties to Pfizer and other pharmaceutical firms, but not to Tibotec.

SAN FRANCISCO — Although nonnucleoside reverse transcriptase inhibitors form one of the mainstays of highly active antiretroviral therapy for HIV, they have a prominent Achilles heel: HIV can develop resistance to the entire NNRTI class with simple, single-point mutations.

Several second-generation nonnucleoside reverse transcriptase inhibitors (NNRTIs) are now in clinical trials, Dr. C. Bradley Hare said at a meeting on HIV management sponsored by the University of California, San Francisco. These drugs have been designed to require HIV to develop multiple mutations for resistance to take place.

Dr. Hare of the university discussed the results of current trials for the following compounds:

▸ Etravirine (TMC125). Developed by Tibotec, etravirine showed good in vitro activity against viruses with one of several NNRTI-resistant mutations. In 8-day, phase I, monotherapy studies in NNRTI-naive patients, etravirine resulted in viral loads declining by two orders of magnitude. In NNRTI-resistant patients treated for 8 days, the decline was smaller—one order of magnitude—but still impressive. Dosing is being worked out, but etravirine is likely to require twice-daily dosing. At a dosage of 800 mg b.i.d. in phase II research, etravirine is beating placebo in patients with NNRTI resistance and in patients with three-class drug failure. No hepatotoxicity or CNS toxicity has been observed, but 17% of patients have developed a rash.

▸ TMC278 (not yet named). TMC278, also from Tibotec, has not progressed beyond phase I studies. It has a longer half-life than etravirine, which may allow once-daily dosing. In a 7-day monotherapy trial in 47 patients, all three dose levels of the drug resulted in viral-load declines of more than one order of magnitude. Adverse events were mostly headache and nausea. None of the patients developed NNRTI mutations during the short trial.

▸ Capavirine (AG1549). In early studies, capavirine, developed by Pfizer Inc., was active against some viruses resistant to the first-generation NNRTIs efavirenz and navirapine. It seemed effective as monotherapy in NNRTI-naive patients. But a randomized, phase II study with 179 patients who had failed other NNRTIs was disappointing. Development of the drug was discontinued in 2005.

Dr. Hare disclosed financial ties to Pfizer and other pharmaceutical firms, but not to Tibotec.

SAN FRANCISCO — Although nonnucleoside reverse transcriptase inhibitors form one of the mainstays of highly active antiretroviral therapy for HIV, they have a prominent Achilles heel: HIV can develop resistance to the entire NNRTI class with simple, single-point mutations.

Several second-generation nonnucleoside reverse transcriptase inhibitors (NNRTIs) are now in clinical trials, Dr. C. Bradley Hare said at a meeting on HIV management sponsored by the University of California, San Francisco. These drugs have been designed to require HIV to develop multiple mutations for resistance to take place.

Dr. Hare of the university discussed the results of current trials for the following compounds:

▸ Etravirine (TMC125). Developed by Tibotec, etravirine showed good in vitro activity against viruses with one of several NNRTI-resistant mutations. In 8-day, phase I, monotherapy studies in NNRTI-naive patients, etravirine resulted in viral loads declining by two orders of magnitude. In NNRTI-resistant patients treated for 8 days, the decline was smaller—one order of magnitude—but still impressive. Dosing is being worked out, but etravirine is likely to require twice-daily dosing. At a dosage of 800 mg b.i.d. in phase II research, etravirine is beating placebo in patients with NNRTI resistance and in patients with three-class drug failure. No hepatotoxicity or CNS toxicity has been observed, but 17% of patients have developed a rash.

▸ TMC278 (not yet named). TMC278, also from Tibotec, has not progressed beyond phase I studies. It has a longer half-life than etravirine, which may allow once-daily dosing. In a 7-day monotherapy trial in 47 patients, all three dose levels of the drug resulted in viral-load declines of more than one order of magnitude. Adverse events were mostly headache and nausea. None of the patients developed NNRTI mutations during the short trial.

▸ Capavirine (AG1549). In early studies, capavirine, developed by Pfizer Inc., was active against some viruses resistant to the first-generation NNRTIs efavirenz and navirapine. It seemed effective as monotherapy in NNRTI-naive patients. But a randomized, phase II study with 179 patients who had failed other NNRTIs was disappointing. Development of the drug was discontinued in 2005.

Dr. Hare disclosed financial ties to Pfizer and other pharmaceutical firms, but not to Tibotec.

SAN FRANCISCO — Using nitrite inhalants, being uncircumcised, and engaging in certain sexual practices all increase the risk of HIV seroconversion among HIV-negative men who have sex with men, Dr. Susan P. Buchbinder reported at a meeting on HIV management sponsored by the University of California, San Francisco.

The results of her published study, along with related unpublished data, suggest a number of behavioral strategies to reduce the risk of HIV transmission among men who have sex with men (MSM).

“Some people say, 'We know what causes HIV [transmission], so why don't men change behavior?'” said Dr. Buchbinder, director of the university's HIV research section. “This is a response that one of my colleagues gave many years ago: She said, 'If behavior change were easy, I'd be thin.' I think we all recognize that behavior change is difficult. It's difficult to sustain over time. … We're trying to modify sexual risk and those are the kinds of behaviors that are often most difficult to change.”

Dr. Buchbinder's study involved 3,257 MSM from six U.S. cities who were HIV negative when they enrolled in the study in 1995. Participants were seen every 6 months for an 18-month period. During that time, 72 men became infected with HIV, yielding an annualized HIV seroincidence of 1.55 per 100 person-years (J. Acquir. Immune Defic. Syndr. 2005;39:82–9).

Taking into account the odds ratios of various risk factors (adjusted for sexual behaviors), as well as the prevalence of those risk factors in the population studied, Dr. Buchbinder and her colleagues calculated the population-attributable risks (PARs) of various behaviors and characteristics. (See table.)

The highest PARs were seen in men who had greater numbers of HIV-negative sex partners. The risk of seroconversion increased by 14% with each additional HIV-negative partner reported in the prior 6 months. Dr. Buchbinder explained this by noting that if one has a lot of HIV-negative partners, the chances increase that one of those partners may have recently become infected and is unaware of this.

These findings suggest that “we need to further develop new HIV-testing strategies, [such as] the implementation of rapid testing to allow people to know their serostatus more quickly,” she said.

The use of nitrite inhalants (known as poppers) also carried a high PAR in Dr. Buchbinder's published study. She mentioned other unpublished data that implicated other drugs, including crystal methamphetamine and sildenafil (Viagra).

“How is it that these substances augment the risk of acquiring HIV?” Dr. Buchbinder asked. “We can assume that for poppers and Viagra, [which are] often used in conjunction to enhance sexual pleasure, we might see an association with more anal sex. But in this study, we found that these three drugs were associated not just with having an increase in anal sex, and not even just having an increase in unprotected anal sex. [These men are] having an increase in unprotected anal sex with a partner whose serostatus was different.”

Furthermore, there are probably ways in which these drugs may enhance transmission biologically. For example, limited animal, in vitro, and human studies suggest that crystal methamphetamine is associated with increased HIV replication and perturbations in immune function.

Poppers are also associated with an effect on immune function as well as vasodilation in mucosal surfaces, which might facilitate transmission across those surfaces. And crystal methamphetamine, poppers, and Viagra may additionally be associated with more prolonged sexual activities, thereby increasing the risk of transmission across mucosal barriers.

It's difficult to know what to do about substance use in this population, Dr. Buchbinder said. Most studies on substance use address people who are addicted, and that model may not apply in a population that uses these drugs intermittently to enhance sexual pleasure. The few related studies on this population are limited by small numbers and short follow-up periods.

Another notable finding was that the risk of seroconversion was significantly higher in uncircumcised men, a finding that Dr. Buchbinder and her colleagues described as biologically plausible, with several possible mechanisms.

One somewhat surprising finding was the significant risk associated with receptive oral sex, even after controlling for receptive and insertional anal sex practices. Other studies have failed to find an independent contribution of receptive oral sex to HIV transmission. The investigators could not rule out the possibility that this apparent association may simply be a marker for riskier sex practices in general, or that it may reflect unmeasured confounders.

Although she acknowledged that behavior change is difficult to achieve, Dr. Buchbinder said that it's important not to give up.

“The first thing we have to remember is that behavior change works,” she said. “In the beginning of the HIV epidemic in the United States, seroincidence in gay men was as high as 20% per year. And now, even in the most heavily impacted subpopulations of men who have sex with men, we're talking about incidence rates of 2%–5% per year. So we've had a substantial decline in the rate of new infections, and that's all because of behavior change.”

SAN FRANCISCO — Using nitrite inhalants, being uncircumcised, and engaging in certain sexual practices all increase the risk of HIV seroconversion among HIV-negative men who have sex with men, Dr. Susan P. Buchbinder reported at a meeting on HIV management sponsored by the University of California, San Francisco.

The results of her published study, along with related unpublished data, suggest a number of behavioral strategies to reduce the risk of HIV transmission among men who have sex with men (MSM).

“Some people say, 'We know what causes HIV [transmission], so why don't men change behavior?'” said Dr. Buchbinder, director of the university's HIV research section. “This is a response that one of my colleagues gave many years ago: She said, 'If behavior change were easy, I'd be thin.' I think we all recognize that behavior change is difficult. It's difficult to sustain over time. … We're trying to modify sexual risk and those are the kinds of behaviors that are often most difficult to change.”

Dr. Buchbinder's study involved 3,257 MSM from six U.S. cities who were HIV negative when they enrolled in the study in 1995. Participants were seen every 6 months for an 18-month period. During that time, 72 men became infected with HIV, yielding an annualized HIV seroincidence of 1.55 per 100 person-years (J. Acquir. Immune Defic. Syndr. 2005;39:82–9).

Taking into account the odds ratios of various risk factors (adjusted for sexual behaviors), as well as the prevalence of those risk factors in the population studied, Dr. Buchbinder and her colleagues calculated the population-attributable risks (PARs) of various behaviors and characteristics. (See table.)

The highest PARs were seen in men who had greater numbers of HIV-negative sex partners. The risk of seroconversion increased by 14% with each additional HIV-negative partner reported in the prior 6 months. Dr. Buchbinder explained this by noting that if one has a lot of HIV-negative partners, the chances increase that one of those partners may have recently become infected and is unaware of this.

These findings suggest that “we need to further develop new HIV-testing strategies, [such as] the implementation of rapid testing to allow people to know their serostatus more quickly,” she said.

The use of nitrite inhalants (known as poppers) also carried a high PAR in Dr. Buchbinder's published study. She mentioned other unpublished data that implicated other drugs, including crystal methamphetamine and sildenafil (Viagra).

“How is it that these substances augment the risk of acquiring HIV?” Dr. Buchbinder asked. “We can assume that for poppers and Viagra, [which are] often used in conjunction to enhance sexual pleasure, we might see an association with more anal sex. But in this study, we found that these three drugs were associated not just with having an increase in anal sex, and not even just having an increase in unprotected anal sex. [These men are] having an increase in unprotected anal sex with a partner whose serostatus was different.”

Furthermore, there are probably ways in which these drugs may enhance transmission biologically. For example, limited animal, in vitro, and human studies suggest that crystal methamphetamine is associated with increased HIV replication and perturbations in immune function.

Poppers are also associated with an effect on immune function as well as vasodilation in mucosal surfaces, which might facilitate transmission across those surfaces. And crystal methamphetamine, poppers, and Viagra may additionally be associated with more prolonged sexual activities, thereby increasing the risk of transmission across mucosal barriers.

It's difficult to know what to do about substance use in this population, Dr. Buchbinder said. Most studies on substance use address people who are addicted, and that model may not apply in a population that uses these drugs intermittently to enhance sexual pleasure. The few related studies on this population are limited by small numbers and short follow-up periods.

Another notable finding was that the risk of seroconversion was significantly higher in uncircumcised men, a finding that Dr. Buchbinder and her colleagues described as biologically plausible, with several possible mechanisms.

One somewhat surprising finding was the significant risk associated with receptive oral sex, even after controlling for receptive and insertional anal sex practices. Other studies have failed to find an independent contribution of receptive oral sex to HIV transmission. The investigators could not rule out the possibility that this apparent association may simply be a marker for riskier sex practices in general, or that it may reflect unmeasured confounders.

Although she acknowledged that behavior change is difficult to achieve, Dr. Buchbinder said that it's important not to give up.

“The first thing we have to remember is that behavior change works,” she said. “In the beginning of the HIV epidemic in the United States, seroincidence in gay men was as high as 20% per year. And now, even in the most heavily impacted subpopulations of men who have sex with men, we're talking about incidence rates of 2%–5% per year. So we've had a substantial decline in the rate of new infections, and that's all because of behavior change.”

SAN FRANCISCO — Using nitrite inhalants, being uncircumcised, and engaging in certain sexual practices all increase the risk of HIV seroconversion among HIV-negative men who have sex with men, Dr. Susan P. Buchbinder reported at a meeting on HIV management sponsored by the University of California, San Francisco.

The results of her published study, along with related unpublished data, suggest a number of behavioral strategies to reduce the risk of HIV transmission among men who have sex with men (MSM).

“Some people say, 'We know what causes HIV [transmission], so why don't men change behavior?'” said Dr. Buchbinder, director of the university's HIV research section. “This is a response that one of my colleagues gave many years ago: She said, 'If behavior change were easy, I'd be thin.' I think we all recognize that behavior change is difficult. It's difficult to sustain over time. … We're trying to modify sexual risk and those are the kinds of behaviors that are often most difficult to change.”

Dr. Buchbinder's study involved 3,257 MSM from six U.S. cities who were HIV negative when they enrolled in the study in 1995. Participants were seen every 6 months for an 18-month period. During that time, 72 men became infected with HIV, yielding an annualized HIV seroincidence of 1.55 per 100 person-years (J. Acquir. Immune Defic. Syndr. 2005;39:82–9).

Taking into account the odds ratios of various risk factors (adjusted for sexual behaviors), as well as the prevalence of those risk factors in the population studied, Dr. Buchbinder and her colleagues calculated the population-attributable risks (PARs) of various behaviors and characteristics. (See table.)

The highest PARs were seen in men who had greater numbers of HIV-negative sex partners. The risk of seroconversion increased by 14% with each additional HIV-negative partner reported in the prior 6 months. Dr. Buchbinder explained this by noting that if one has a lot of HIV-negative partners, the chances increase that one of those partners may have recently become infected and is unaware of this.

These findings suggest that “we need to further develop new HIV-testing strategies, [such as] the implementation of rapid testing to allow people to know their serostatus more quickly,” she said.

The use of nitrite inhalants (known as poppers) also carried a high PAR in Dr. Buchbinder's published study. She mentioned other unpublished data that implicated other drugs, including crystal methamphetamine and sildenafil (Viagra).

“How is it that these substances augment the risk of acquiring HIV?” Dr. Buchbinder asked. “We can assume that for poppers and Viagra, [which are] often used in conjunction to enhance sexual pleasure, we might see an association with more anal sex. But in this study, we found that these three drugs were associated not just with having an increase in anal sex, and not even just having an increase in unprotected anal sex. [These men are] having an increase in unprotected anal sex with a partner whose serostatus was different.”

Furthermore, there are probably ways in which these drugs may enhance transmission biologically. For example, limited animal, in vitro, and human studies suggest that crystal methamphetamine is associated with increased HIV replication and perturbations in immune function.

Poppers are also associated with an effect on immune function as well as vasodilation in mucosal surfaces, which might facilitate transmission across those surfaces. And crystal methamphetamine, poppers, and Viagra may additionally be associated with more prolonged sexual activities, thereby increasing the risk of transmission across mucosal barriers.

It's difficult to know what to do about substance use in this population, Dr. Buchbinder said. Most studies on substance use address people who are addicted, and that model may not apply in a population that uses these drugs intermittently to enhance sexual pleasure. The few related studies on this population are limited by small numbers and short follow-up periods.

Another notable finding was that the risk of seroconversion was significantly higher in uncircumcised men, a finding that Dr. Buchbinder and her colleagues described as biologically plausible, with several possible mechanisms.

One somewhat surprising finding was the significant risk associated with receptive oral sex, even after controlling for receptive and insertional anal sex practices. Other studies have failed to find an independent contribution of receptive oral sex to HIV transmission. The investigators could not rule out the possibility that this apparent association may simply be a marker for riskier sex practices in general, or that it may reflect unmeasured confounders.

Although she acknowledged that behavior change is difficult to achieve, Dr. Buchbinder said that it's important not to give up.

“The first thing we have to remember is that behavior change works,” she said. “In the beginning of the HIV epidemic in the United States, seroincidence in gay men was as high as 20% per year. And now, even in the most heavily impacted subpopulations of men who have sex with men, we're talking about incidence rates of 2%–5% per year. So we've had a substantial decline in the rate of new infections, and that's all because of behavior change.”

SAN FRANCISCO — With no specific treatment available for nonalcoholic fatty liver disease, the best current strategy centers on monitoring the patient's condition and managing the patient's lifestyle and metabolic syndrome, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

The patient's liver enzymes, liver function (bilirubin levels, albumin levels, and prothrombin time), and platelet count should be monitored. Each patient also should undergo regular ultrasound exams.

Patients with nonalcoholic fatty liver disease (NAFLD) should be instructed to avoid hepatotoxins—especially alcohol—and should be advised to pursue gradual weight loss with diet and exercise.

“Weight loss remains the simplest advice you can give,” said Dr. Bass of the University of California, San Francisco. He pointed to a study showing that even modest weight loss (less than 10% of the patient's initial body weight) can reduce intrahepatic fat while leaving intramuscular fat unchanged. This level of weight loss also improved basal and insulin-stimulated glucose metabolism (Diabetes 2005;54:603–8).

Bariatric surgery can be helpful for some patients with NAFLD, but it should be the newer restrictive surgery that involves gastric banding; this approach tends to decrease steatosis, fibrosis, and nonalcoholic steatohepatitis. The older malabsorptive surgical strategies can be dangerous; they can lead to increased steatosis, fibrosis, nonalcoholic steatohepatitis, and liver failure.

The insulin-sensitizing agent metformin appears to be helpful in patients with NAFLD; however, the published studies tend to be small and open label, and thus the available evidence base is not overwhelming.

The thiazolidinedionespioglitazone and troglitazone appear to improve liver enzymes and fibrosis measured histologically, but once again, the evidence comes from open-label trials.

Dr. Bass noted some caveats to be observed regardingthiazolidinediones: They can cause weight gain and relapse upon discontinuation, and some patients experience serious side effects, such as heart failure and hepatotoxicity.

SAN FRANCISCO — With no specific treatment available for nonalcoholic fatty liver disease, the best current strategy centers on monitoring the patient's condition and managing the patient's lifestyle and metabolic syndrome, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

The patient's liver enzymes, liver function (bilirubin levels, albumin levels, and prothrombin time), and platelet count should be monitored. Each patient also should undergo regular ultrasound exams.

Patients with nonalcoholic fatty liver disease (NAFLD) should be instructed to avoid hepatotoxins—especially alcohol—and should be advised to pursue gradual weight loss with diet and exercise.

“Weight loss remains the simplest advice you can give,” said Dr. Bass of the University of California, San Francisco. He pointed to a study showing that even modest weight loss (less than 10% of the patient's initial body weight) can reduce intrahepatic fat while leaving intramuscular fat unchanged. This level of weight loss also improved basal and insulin-stimulated glucose metabolism (Diabetes 2005;54:603–8).

Bariatric surgery can be helpful for some patients with NAFLD, but it should be the newer restrictive surgery that involves gastric banding; this approach tends to decrease steatosis, fibrosis, and nonalcoholic steatohepatitis. The older malabsorptive surgical strategies can be dangerous; they can lead to increased steatosis, fibrosis, nonalcoholic steatohepatitis, and liver failure.

The insulin-sensitizing agent metformin appears to be helpful in patients with NAFLD; however, the published studies tend to be small and open label, and thus the available evidence base is not overwhelming.

The thiazolidinedionespioglitazone and troglitazone appear to improve liver enzymes and fibrosis measured histologically, but once again, the evidence comes from open-label trials.

Dr. Bass noted some caveats to be observed regardingthiazolidinediones: They can cause weight gain and relapse upon discontinuation, and some patients experience serious side effects, such as heart failure and hepatotoxicity.

SAN FRANCISCO — With no specific treatment available for nonalcoholic fatty liver disease, the best current strategy centers on monitoring the patient's condition and managing the patient's lifestyle and metabolic syndrome, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

The patient's liver enzymes, liver function (bilirubin levels, albumin levels, and prothrombin time), and platelet count should be monitored. Each patient also should undergo regular ultrasound exams.

Patients with nonalcoholic fatty liver disease (NAFLD) should be instructed to avoid hepatotoxins—especially alcohol—and should be advised to pursue gradual weight loss with diet and exercise.

“Weight loss remains the simplest advice you can give,” said Dr. Bass of the University of California, San Francisco. He pointed to a study showing that even modest weight loss (less than 10% of the patient's initial body weight) can reduce intrahepatic fat while leaving intramuscular fat unchanged. This level of weight loss also improved basal and insulin-stimulated glucose metabolism (Diabetes 2005;54:603–8).

Bariatric surgery can be helpful for some patients with NAFLD, but it should be the newer restrictive surgery that involves gastric banding; this approach tends to decrease steatosis, fibrosis, and nonalcoholic steatohepatitis. The older malabsorptive surgical strategies can be dangerous; they can lead to increased steatosis, fibrosis, nonalcoholic steatohepatitis, and liver failure.

The insulin-sensitizing agent metformin appears to be helpful in patients with NAFLD; however, the published studies tend to be small and open label, and thus the available evidence base is not overwhelming.

The thiazolidinedionespioglitazone and troglitazone appear to improve liver enzymes and fibrosis measured histologically, but once again, the evidence comes from open-label trials.

Dr. Bass noted some caveats to be observed regardingthiazolidinediones: They can cause weight gain and relapse upon discontinuation, and some patients experience serious side effects, such as heart failure and hepatotoxicity.

SAN FRANCISCO — The liver biopsy remains the preferred method for diagnosing nonalcoholic fatty liver disease, but biopsy candidates should be chosen with care. Not all patients with signs of the disease will require a biopsy, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

Patients who are eventually diagnosed with nonalcoholic fatty liver disease (NAFLD) present initially in a variety of ways, said Dr. Bass of the University of California, San Francisco.

For example, an insurance exam can turn up an incidental aminotransferase elevation or an enlarged liver. An abdominal imaging study may reveal a fatty liver. A patient may have a complication of cirrhosis. Or NAFLD patients may be identified by screening high-risk populations with liver enzyme tests or liver ultrasound. In addition, an increasing number of NAFLD patients are being identified by liver biopsy during weight-reduction surgery, Dr. Bass said.

But it's not practical or desirable to screen all at-risk patients with a biopsy, and there are some good reasons not to do so. (See box.) About 25% of patients will experience significant pain during the biopsy. In addition, 1%–3.5% of patients will have morbidities such as hypotension, pneumothorax, hemoperitoneum, hemobilia, and gall bladder penetration. About 0.1% of patients will die from the procedure.

There are five situations in which a liver biopsy is essential: when a patient's liver enzymes show an unusual pattern or are 3–5 times normal; when other liver disease can't be excluded; when the patient doesn't have metabolic syndrome; to confirm a clinical suspicion of cirrhosis; and for qualifying a patient for entry into a clinical trial.

Although a definitive diagnosis still requires a biopsy, there are several alternatives for assessing the liver, Dr. Bass said.

Elevated liver enzymes can be very suggestive of NAFLD, but in a phenomenon Dr. Bass called “The Silence of the Labs,” some patients with NAFLD have normal liver enzymes. He pointed to one study of patients undergoing gastric bypass in which 68% had normal ALT and AST, but only 52% had a normal liver biopsy. Among the remaining 48% with abnormal biopsy results, about 27% were found to have nonalcoholic fatty liver, and the others had nonalcoholic steatohepatitis.

The diagnosis of NAFLD is often made by exclusion. After the physician excludes alcoholic liver disease, drug-induced liver injury, iron overload, hepatitis B and C, and autoimmune hepatitis, NAFLD is the diagnosis that remains.

It can be difficult, however, to exclude a significant contribution from alcohol, especially since patients are not always truthful. For the purposes of inclusion in clinical trials, NIH defines “nonalcoholic” as less than 14 units of alcohol per week for men or less than 7 units per week for women. A unit is one can of beer, one glass of wine, or one shot of hard liquor, but there's considerable variation in the alcoholic content of drinks within each class.

The combination of ultrasound evidence of fatty liver plus liver enzyme elevation without markers for hepatitis C or B yields a 96% positive predictive value for NAFLD, according to one study. The problem is that although ultrasound is sensitive, it's not very specific.

Pros

▸ Grade and stage of NAFLD are determined.

▸ Confidence in the diagnosis is 100%.

▸ Patients are motivated to lose weight.

▸ Biopsy is essential for enrollment inclinical trials of treatments.

Cons

▸ Risk of morbidity is increased with biopsy.

▸ Noninvasive diagnosis is quite accurate.

▸ Natural history of NAFLD is benign in most patients.

▸ NAFLD is a common disorder.

▸ There is no proven, specific treatment for NAFLD.

Source: Dr. Bass

CT imaging is somewhat more specific. In CT, a normal liver has about the same density as the spleen, whereas in NAFLD, the spleen will be quite a bit brighter than the liver. Unfortunately, CT is too expensive for routine screening.

At least three serological tests for hepatic fibrosis are under development, Dr. Bass said. In addition, transient elastography—a combination of 5 MHz ultrasound and 50 Hz elastic waves—may make the diagnosis simpler in the near future.

SAN FRANCISCO — The liver biopsy remains the preferred method for diagnosing nonalcoholic fatty liver disease, but biopsy candidates should be chosen with care. Not all patients with signs of the disease will require a biopsy, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

Patients who are eventually diagnosed with nonalcoholic fatty liver disease (NAFLD) present initially in a variety of ways, said Dr. Bass of the University of California, San Francisco.

For example, an insurance exam can turn up an incidental aminotransferase elevation or an enlarged liver. An abdominal imaging study may reveal a fatty liver. A patient may have a complication of cirrhosis. Or NAFLD patients may be identified by screening high-risk populations with liver enzyme tests or liver ultrasound. In addition, an increasing number of NAFLD patients are being identified by liver biopsy during weight-reduction surgery, Dr. Bass said.

But it's not practical or desirable to screen all at-risk patients with a biopsy, and there are some good reasons not to do so. (See box.) About 25% of patients will experience significant pain during the biopsy. In addition, 1%–3.5% of patients will have morbidities such as hypotension, pneumothorax, hemoperitoneum, hemobilia, and gall bladder penetration. About 0.1% of patients will die from the procedure.

There are five situations in which a liver biopsy is essential: when a patient's liver enzymes show an unusual pattern or are 3–5 times normal; when other liver disease can't be excluded; when the patient doesn't have metabolic syndrome; to confirm a clinical suspicion of cirrhosis; and for qualifying a patient for entry into a clinical trial.

Although a definitive diagnosis still requires a biopsy, there are several alternatives for assessing the liver, Dr. Bass said.

Elevated liver enzymes can be very suggestive of NAFLD, but in a phenomenon Dr. Bass called “The Silence of the Labs,” some patients with NAFLD have normal liver enzymes. He pointed to one study of patients undergoing gastric bypass in which 68% had normal ALT and AST, but only 52% had a normal liver biopsy. Among the remaining 48% with abnormal biopsy results, about 27% were found to have nonalcoholic fatty liver, and the others had nonalcoholic steatohepatitis.

The diagnosis of NAFLD is often made by exclusion. After the physician excludes alcoholic liver disease, drug-induced liver injury, iron overload, hepatitis B and C, and autoimmune hepatitis, NAFLD is the diagnosis that remains.

It can be difficult, however, to exclude a significant contribution from alcohol, especially since patients are not always truthful. For the purposes of inclusion in clinical trials, NIH defines “nonalcoholic” as less than 14 units of alcohol per week for men or less than 7 units per week for women. A unit is one can of beer, one glass of wine, or one shot of hard liquor, but there's considerable variation in the alcoholic content of drinks within each class.

The combination of ultrasound evidence of fatty liver plus liver enzyme elevation without markers for hepatitis C or B yields a 96% positive predictive value for NAFLD, according to one study. The problem is that although ultrasound is sensitive, it's not very specific.

Pros

▸ Grade and stage of NAFLD are determined.

▸ Confidence in the diagnosis is 100%.

▸ Patients are motivated to lose weight.

▸ Biopsy is essential for enrollment inclinical trials of treatments.

Cons

▸ Risk of morbidity is increased with biopsy.

▸ Noninvasive diagnosis is quite accurate.

▸ Natural history of NAFLD is benign in most patients.

▸ NAFLD is a common disorder.

▸ There is no proven, specific treatment for NAFLD.

Source: Dr. Bass

CT imaging is somewhat more specific. In CT, a normal liver has about the same density as the spleen, whereas in NAFLD, the spleen will be quite a bit brighter than the liver. Unfortunately, CT is too expensive for routine screening.

At least three serological tests for hepatic fibrosis are under development, Dr. Bass said. In addition, transient elastography—a combination of 5 MHz ultrasound and 50 Hz elastic waves—may make the diagnosis simpler in the near future.

SAN FRANCISCO — The liver biopsy remains the preferred method for diagnosing nonalcoholic fatty liver disease, but biopsy candidates should be chosen with care. Not all patients with signs of the disease will require a biopsy, Dr. Nathan M. Bass said at the Third World Congress on Insulin Resistance Syndrome.

Patients who are eventually diagnosed with nonalcoholic fatty liver disease (NAFLD) present initially in a variety of ways, said Dr. Bass of the University of California, San Francisco.

For example, an insurance exam can turn up an incidental aminotransferase elevation or an enlarged liver. An abdominal imaging study may reveal a fatty liver. A patient may have a complication of cirrhosis. Or NAFLD patients may be identified by screening high-risk populations with liver enzyme tests or liver ultrasound. In addition, an increasing number of NAFLD patients are being identified by liver biopsy during weight-reduction surgery, Dr. Bass said.

But it's not practical or desirable to screen all at-risk patients with a biopsy, and there are some good reasons not to do so. (See box.) About 25% of patients will experience significant pain during the biopsy. In addition, 1%–3.5% of patients will have morbidities such as hypotension, pneumothorax, hemoperitoneum, hemobilia, and gall bladder penetration. About 0.1% of patients will die from the procedure.

There are five situations in which a liver biopsy is essential: when a patient's liver enzymes show an unusual pattern or are 3–5 times normal; when other liver disease can't be excluded; when the patient doesn't have metabolic syndrome; to confirm a clinical suspicion of cirrhosis; and for qualifying a patient for entry into a clinical trial.

Although a definitive diagnosis still requires a biopsy, there are several alternatives for assessing the liver, Dr. Bass said.

Elevated liver enzymes can be very suggestive of NAFLD, but in a phenomenon Dr. Bass called “The Silence of the Labs,” some patients with NAFLD have normal liver enzymes. He pointed to one study of patients undergoing gastric bypass in which 68% had normal ALT and AST, but only 52% had a normal liver biopsy. Among the remaining 48% with abnormal biopsy results, about 27% were found to have nonalcoholic fatty liver, and the others had nonalcoholic steatohepatitis.

The diagnosis of NAFLD is often made by exclusion. After the physician excludes alcoholic liver disease, drug-induced liver injury, iron overload, hepatitis B and C, and autoimmune hepatitis, NAFLD is the diagnosis that remains.

It can be difficult, however, to exclude a significant contribution from alcohol, especially since patients are not always truthful. For the purposes of inclusion in clinical trials, NIH defines “nonalcoholic” as less than 14 units of alcohol per week for men or less than 7 units per week for women. A unit is one can of beer, one glass of wine, or one shot of hard liquor, but there's considerable variation in the alcoholic content of drinks within each class.

The combination of ultrasound evidence of fatty liver plus liver enzyme elevation without markers for hepatitis C or B yields a 96% positive predictive value for NAFLD, according to one study. The problem is that although ultrasound is sensitive, it's not very specific.

Pros

▸ Grade and stage of NAFLD are determined.

▸ Confidence in the diagnosis is 100%.

▸ Patients are motivated to lose weight.

▸ Biopsy is essential for enrollment inclinical trials of treatments.

Cons

▸ Risk of morbidity is increased with biopsy.

▸ Noninvasive diagnosis is quite accurate.

▸ Natural history of NAFLD is benign in most patients.

▸ NAFLD is a common disorder.

▸ There is no proven, specific treatment for NAFLD.

Source: Dr. Bass

CT imaging is somewhat more specific. In CT, a normal liver has about the same density as the spleen, whereas in NAFLD, the spleen will be quite a bit brighter than the liver. Unfortunately, CT is too expensive for routine screening.

At least three serological tests for hepatic fibrosis are under development, Dr. Bass said. In addition, transient elastography—a combination of 5 MHz ultrasound and 50 Hz elastic waves—may make the diagnosis simpler in the near future.