Sherry Boschert

SAN FRANCISCO — A clinical phase II/III trial of a potential oral medication for pouchitis will begin this year, following a Food and Drug Administration decision to grant orphan drug status to the experimental compound AST-120.

There are no approved treatments for pouchitis, a problem seen in approximately half of patients who undergo a colectomy and surgical creation of a j-pouch to treat ulcerative colitis. The pouch becomes inflamed and is associated with frequent and severe diarrhea, abdominal cramps, fever, and dehydration.

Pouchitis affects fewer than 100,000 patients in the United States, allowing AST-120 to earn its orphan drug status, which is designed to encourage companies to develop medicines for rare diseases. Orphan drug status grants 7 years of exclusive rights to the market of the drug, among other benefits.

Ocera Therapeutics also is conducting studies of AST-120 for the treatment of Crohn's disease, fistulizing Crohn's disease, and other GI diseases, Dr. Laurent Fischer said at a JP Morgan Healthcare conference. Dr. Fischer is president and CEO of the San Diego-based company, which licensed AST-120 from its Japanese developers.

The first 10 patients with pouchitis in an open-label proof-of-concept trial showed significant improvements after 4 weeks of taking 2-g sachets of AST-120 t.i.d., he said. Nine patients completed therapy and one dropped out because of an upper respiratory tract infection. Four patients had complete remission, and five had a clinical response.

AST-120 is an oral spherical adsorptive carbon that may adsorb bile acids and bacterial toxins associated with pouchitis, potentially protecting the intestinal mucosa of the j-pouch from inflammation. AST-120 does not absorb vitamins and digestive enzymes, and is not itself systemically absorbed in the GI tract.

The drug is marketed in Japan as a treatment to delay the time to dialysis and to decrease uremic symptoms in patients with chronic renal failure. AST-120 has been studied in Japan for the treatment of Crohn's disease in more than 2,600 patients, and appears to have a good safety profile, Dr. Fischer said.

A separate, ongoing phase III study in the United States of AST-120 to treat fistulizing Crohn's disease may produce results in time for presentation at the Digestive Disease Week conference in May, he added.

AST-120 is also being studied for the treatment of Crohn's disease, fistulizing Crohn's disease, and other GI diseases. DR. FISCHER

SAN FRANCISCO — A clinical phase II/III trial of a potential oral medication for pouchitis will begin this year, following a Food and Drug Administration decision to grant orphan drug status to the experimental compound AST-120.

There are no approved treatments for pouchitis, a problem seen in approximately half of patients who undergo a colectomy and surgical creation of a j-pouch to treat ulcerative colitis. The pouch becomes inflamed and is associated with frequent and severe diarrhea, abdominal cramps, fever, and dehydration.

Pouchitis affects fewer than 100,000 patients in the United States, allowing AST-120 to earn its orphan drug status, which is designed to encourage companies to develop medicines for rare diseases. Orphan drug status grants 7 years of exclusive rights to the market of the drug, among other benefits.

Ocera Therapeutics also is conducting studies of AST-120 for the treatment of Crohn's disease, fistulizing Crohn's disease, and other GI diseases, Dr. Laurent Fischer said at a JP Morgan Healthcare conference. Dr. Fischer is president and CEO of the San Diego-based company, which licensed AST-120 from its Japanese developers.

The first 10 patients with pouchitis in an open-label proof-of-concept trial showed significant improvements after 4 weeks of taking 2-g sachets of AST-120 t.i.d., he said. Nine patients completed therapy and one dropped out because of an upper respiratory tract infection. Four patients had complete remission, and five had a clinical response.

AST-120 is an oral spherical adsorptive carbon that may adsorb bile acids and bacterial toxins associated with pouchitis, potentially protecting the intestinal mucosa of the j-pouch from inflammation. AST-120 does not absorb vitamins and digestive enzymes, and is not itself systemically absorbed in the GI tract.

The drug is marketed in Japan as a treatment to delay the time to dialysis and to decrease uremic symptoms in patients with chronic renal failure. AST-120 has been studied in Japan for the treatment of Crohn's disease in more than 2,600 patients, and appears to have a good safety profile, Dr. Fischer said.

A separate, ongoing phase III study in the United States of AST-120 to treat fistulizing Crohn's disease may produce results in time for presentation at the Digestive Disease Week conference in May, he added.

AST-120 is also being studied for the treatment of Crohn's disease, fistulizing Crohn's disease, and other GI diseases. DR. FISCHER

SAN FRANCISCO — A clinical phase II/III trial of a potential oral medication for pouchitis will begin this year, following a Food and Drug Administration decision to grant orphan drug status to the experimental compound AST-120.

There are no approved treatments for pouchitis, a problem seen in approximately half of patients who undergo a colectomy and surgical creation of a j-pouch to treat ulcerative colitis. The pouch becomes inflamed and is associated with frequent and severe diarrhea, abdominal cramps, fever, and dehydration.

Pouchitis affects fewer than 100,000 patients in the United States, allowing AST-120 to earn its orphan drug status, which is designed to encourage companies to develop medicines for rare diseases. Orphan drug status grants 7 years of exclusive rights to the market of the drug, among other benefits.

Ocera Therapeutics also is conducting studies of AST-120 for the treatment of Crohn's disease, fistulizing Crohn's disease, and other GI diseases, Dr. Laurent Fischer said at a JP Morgan Healthcare conference. Dr. Fischer is president and CEO of the San Diego-based company, which licensed AST-120 from its Japanese developers.

The first 10 patients with pouchitis in an open-label proof-of-concept trial showed significant improvements after 4 weeks of taking 2-g sachets of AST-120 t.i.d., he said. Nine patients completed therapy and one dropped out because of an upper respiratory tract infection. Four patients had complete remission, and five had a clinical response.

AST-120 is an oral spherical adsorptive carbon that may adsorb bile acids and bacterial toxins associated with pouchitis, potentially protecting the intestinal mucosa of the j-pouch from inflammation. AST-120 does not absorb vitamins and digestive enzymes, and is not itself systemically absorbed in the GI tract.

The drug is marketed in Japan as a treatment to delay the time to dialysis and to decrease uremic symptoms in patients with chronic renal failure. AST-120 has been studied in Japan for the treatment of Crohn's disease in more than 2,600 patients, and appears to have a good safety profile, Dr. Fischer said.

A separate, ongoing phase III study in the United States of AST-120 to treat fistulizing Crohn's disease may produce results in time for presentation at the Digestive Disease Week conference in May, he added.

AST-120 is also being studied for the treatment of Crohn's disease, fistulizing Crohn's disease, and other GI diseases. DR. FISCHER

SAN FRANCISCO — Reports of difficulty falling asleep were associated with poorer performance on some cognitive measures in a study of 174 elderly community-dwelling blacks.

Self-reported sleep trouble appears to be a unique predictor of cognitive performance, even after controlling for age, gender, education, depression, and current health, reported Alyssa A. Gamaldo of the psychology department at North Carolina State University, Raleigh, and her associates.

When asked whether they'd had any trouble falling asleep in the past year, 29% of participants said they did, a rate that's consistent with the findings of previous, more rigorous studies of sleep difficulties, she said at the annual meeting of the Gerontological Society of America. From 10% to 40% of older adults reported sleep difficulties in earlier studies.

Investigators in the current study analyzed data from a subset of blacks in the Baltimore Longitudinal Study of Aging. Participants were living independently and had a mean age of 73 years, a mean education of 10 years, and a mean monthly income of $800. The cohort was 71% women.

Those who reported sleep trouble tended to perform worse on measures of short-term memory and working memory. Short-term memory was measured by the forward digit span task and the backward digit span task, and working memory was measured by the alpha span task.

Participants also completed the Mini-Mental State Examination to measure global cognitive status and the California Verbal Learning Test to measure episodic memory, but results for these were not significantly different between adults who did or did not have trouble sleeping.

Several previous studies have suggested that the relationship between self-reported sleep difficulties and cognitive performance may be moderated by depression. In the current study, however, neither depression nor older age appeared to exacerbate differences in cognitive performance, Ms. Gamaldo said.

The study employed the Center for Epidemiological Studies Depression Scale to measure depressive symptoms and these scores were used to categorize 36 participants as depressed (a score of 16 or greater) and 138 as not depressed (score of 0–15).

The analysis did find one significant interaction related to age and depression. Participants who were both older and depressed were more likely to have low scores on the Mini-Mental State Examination.

To assess current health, participants were asked to rate their health on a scale from 1 (“very poor”) to 4 (“very good”).

SAN FRANCISCO — Reports of difficulty falling asleep were associated with poorer performance on some cognitive measures in a study of 174 elderly community-dwelling blacks.

Self-reported sleep trouble appears to be a unique predictor of cognitive performance, even after controlling for age, gender, education, depression, and current health, reported Alyssa A. Gamaldo of the psychology department at North Carolina State University, Raleigh, and her associates.

When asked whether they'd had any trouble falling asleep in the past year, 29% of participants said they did, a rate that's consistent with the findings of previous, more rigorous studies of sleep difficulties, she said at the annual meeting of the Gerontological Society of America. From 10% to 40% of older adults reported sleep difficulties in earlier studies.

Investigators in the current study analyzed data from a subset of blacks in the Baltimore Longitudinal Study of Aging. Participants were living independently and had a mean age of 73 years, a mean education of 10 years, and a mean monthly income of $800. The cohort was 71% women.

Those who reported sleep trouble tended to perform worse on measures of short-term memory and working memory. Short-term memory was measured by the forward digit span task and the backward digit span task, and working memory was measured by the alpha span task.

Participants also completed the Mini-Mental State Examination to measure global cognitive status and the California Verbal Learning Test to measure episodic memory, but results for these were not significantly different between adults who did or did not have trouble sleeping.

Several previous studies have suggested that the relationship between self-reported sleep difficulties and cognitive performance may be moderated by depression. In the current study, however, neither depression nor older age appeared to exacerbate differences in cognitive performance, Ms. Gamaldo said.

The study employed the Center for Epidemiological Studies Depression Scale to measure depressive symptoms and these scores were used to categorize 36 participants as depressed (a score of 16 or greater) and 138 as not depressed (score of 0–15).

The analysis did find one significant interaction related to age and depression. Participants who were both older and depressed were more likely to have low scores on the Mini-Mental State Examination.

To assess current health, participants were asked to rate their health on a scale from 1 (“very poor”) to 4 (“very good”).

SAN FRANCISCO — Reports of difficulty falling asleep were associated with poorer performance on some cognitive measures in a study of 174 elderly community-dwelling blacks.

Self-reported sleep trouble appears to be a unique predictor of cognitive performance, even after controlling for age, gender, education, depression, and current health, reported Alyssa A. Gamaldo of the psychology department at North Carolina State University, Raleigh, and her associates.

When asked whether they'd had any trouble falling asleep in the past year, 29% of participants said they did, a rate that's consistent with the findings of previous, more rigorous studies of sleep difficulties, she said at the annual meeting of the Gerontological Society of America. From 10% to 40% of older adults reported sleep difficulties in earlier studies.

Investigators in the current study analyzed data from a subset of blacks in the Baltimore Longitudinal Study of Aging. Participants were living independently and had a mean age of 73 years, a mean education of 10 years, and a mean monthly income of $800. The cohort was 71% women.

Those who reported sleep trouble tended to perform worse on measures of short-term memory and working memory. Short-term memory was measured by the forward digit span task and the backward digit span task, and working memory was measured by the alpha span task.

Participants also completed the Mini-Mental State Examination to measure global cognitive status and the California Verbal Learning Test to measure episodic memory, but results for these were not significantly different between adults who did or did not have trouble sleeping.

Several previous studies have suggested that the relationship between self-reported sleep difficulties and cognitive performance may be moderated by depression. In the current study, however, neither depression nor older age appeared to exacerbate differences in cognitive performance, Ms. Gamaldo said.

The study employed the Center for Epidemiological Studies Depression Scale to measure depressive symptoms and these scores were used to categorize 36 participants as depressed (a score of 16 or greater) and 138 as not depressed (score of 0–15).

The analysis did find one significant interaction related to age and depression. Participants who were both older and depressed were more likely to have low scores on the Mini-Mental State Examination.

To assess current health, participants were asked to rate their health on a scale from 1 (“very poor”) to 4 (“very good”).

SAN FRANCISCO — Medications that are not classically thought of as anticholinergic but that have anticholinergic properties contribute to functional declines in older patients who are on multiple drug therapy.

Investigators tabulated cumulative anticholinergic burden scores for 3,070 adults aged 70 years or older, with each drug scored 0–3 based on serum anticholinergic activity. Every 1-point increase in the anticholinergic burden score was equivalent to adding a year of age as far as patients' ability to complete activities of daily living (ADL) and instrumental activities of daily living (IADL), Dr. Kaycee M. Sink of Wake Forest University, Winston-Salem, N.C., and her associates reported in a poster.

“While one drug in and of itself may not be enough to cause someone disability or functional impairment, if you're on multiple drugs with anticholinergic activities, that might add up,” Dr. Sink said at the annual meeting of the Gerontological Society of America. The cross-sectional analysis used two previously existing lists of dozens of drugs with anticholinergic burden scores, including medications like digoxin and diltiazem that usually aren't thought of as anticholinergic.

The analysis was part of the Ginkgo Evaluation of Memory Study, a randomized, controlled trial of ginkgo to prevent dementia. Previous studies have shown an association between anticholinergic drug use and decreased muscle strength or poorer cognition in older adults, possibly related to the drugs' central effects on psychomotor slowing or their effects at the neuromuscular junction.

In the current study, 39% of participants were taking one or more drugs with anticholinergic activity. In this anticholinergic subgroup, 89% reported their health as good or better with the medications, compared with 96% of participants on no drugs with anticholinergic activity. Hypertension was present in 60% of those in the anticholinergic group and 51% of those in the nonanticholinergic group.

Performance-based assessments of ADL and IADL found that the likelihood of being dependent on help for ADL or IADL increased 10%–15% for every 1-point increase in the cumulative anticholinergic burden score, Dr. Sink said.

The results were adjusted for potentially confounding factors including age, sex, alcohol use, body mass index, hypertension, self-reported health, and clinic site. Effects on gait speed were adjusted for the person's height. Every 1-point increase in the anticholinergic burden score decreased gait speed by 0.009 m/sec.

The study was limited by an inability to completely control for the potentially confounding effects of the indications for the drug use. In addition, the anticholinergic burden scoring system did not consider dose effects.

The investigators will conduct a longitudinal study of whether anticholinergic burden predicts functional decline over time. Results should be available in a year or so, she said. If the anticholinergic burden score can predict who will decline faster, “then I think it would be useful to create a tool that clinicians can use in the office to calculate up each person's anticholinergic score,” Dr. Sink added. “I don't think it would be that hard” to create a computerized tool to assess anticholinergic burden.

Dr. Sink has no association with companies that make the drugs studied.

'If you'reon multipledrugs with anticholinergic activities, that might add up.' DR. SINK

SAN FRANCISCO — Medications that are not classically thought of as anticholinergic but that have anticholinergic properties contribute to functional declines in older patients who are on multiple drug therapy.

Investigators tabulated cumulative anticholinergic burden scores for 3,070 adults aged 70 years or older, with each drug scored 0–3 based on serum anticholinergic activity. Every 1-point increase in the anticholinergic burden score was equivalent to adding a year of age as far as patients' ability to complete activities of daily living (ADL) and instrumental activities of daily living (IADL), Dr. Kaycee M. Sink of Wake Forest University, Winston-Salem, N.C., and her associates reported in a poster.

“While one drug in and of itself may not be enough to cause someone disability or functional impairment, if you're on multiple drugs with anticholinergic activities, that might add up,” Dr. Sink said at the annual meeting of the Gerontological Society of America. The cross-sectional analysis used two previously existing lists of dozens of drugs with anticholinergic burden scores, including medications like digoxin and diltiazem that usually aren't thought of as anticholinergic.

The analysis was part of the Ginkgo Evaluation of Memory Study, a randomized, controlled trial of ginkgo to prevent dementia. Previous studies have shown an association between anticholinergic drug use and decreased muscle strength or poorer cognition in older adults, possibly related to the drugs' central effects on psychomotor slowing or their effects at the neuromuscular junction.

In the current study, 39% of participants were taking one or more drugs with anticholinergic activity. In this anticholinergic subgroup, 89% reported their health as good or better with the medications, compared with 96% of participants on no drugs with anticholinergic activity. Hypertension was present in 60% of those in the anticholinergic group and 51% of those in the nonanticholinergic group.

Performance-based assessments of ADL and IADL found that the likelihood of being dependent on help for ADL or IADL increased 10%–15% for every 1-point increase in the cumulative anticholinergic burden score, Dr. Sink said.

The results were adjusted for potentially confounding factors including age, sex, alcohol use, body mass index, hypertension, self-reported health, and clinic site. Effects on gait speed were adjusted for the person's height. Every 1-point increase in the anticholinergic burden score decreased gait speed by 0.009 m/sec.

The study was limited by an inability to completely control for the potentially confounding effects of the indications for the drug use. In addition, the anticholinergic burden scoring system did not consider dose effects.

The investigators will conduct a longitudinal study of whether anticholinergic burden predicts functional decline over time. Results should be available in a year or so, she said. If the anticholinergic burden score can predict who will decline faster, “then I think it would be useful to create a tool that clinicians can use in the office to calculate up each person's anticholinergic score,” Dr. Sink added. “I don't think it would be that hard” to create a computerized tool to assess anticholinergic burden.

Dr. Sink has no association with companies that make the drugs studied.

'If you'reon multipledrugs with anticholinergic activities, that might add up.' DR. SINK

SAN FRANCISCO — Medications that are not classically thought of as anticholinergic but that have anticholinergic properties contribute to functional declines in older patients who are on multiple drug therapy.

Investigators tabulated cumulative anticholinergic burden scores for 3,070 adults aged 70 years or older, with each drug scored 0–3 based on serum anticholinergic activity. Every 1-point increase in the anticholinergic burden score was equivalent to adding a year of age as far as patients' ability to complete activities of daily living (ADL) and instrumental activities of daily living (IADL), Dr. Kaycee M. Sink of Wake Forest University, Winston-Salem, N.C., and her associates reported in a poster.

“While one drug in and of itself may not be enough to cause someone disability or functional impairment, if you're on multiple drugs with anticholinergic activities, that might add up,” Dr. Sink said at the annual meeting of the Gerontological Society of America. The cross-sectional analysis used two previously existing lists of dozens of drugs with anticholinergic burden scores, including medications like digoxin and diltiazem that usually aren't thought of as anticholinergic.

The analysis was part of the Ginkgo Evaluation of Memory Study, a randomized, controlled trial of ginkgo to prevent dementia. Previous studies have shown an association between anticholinergic drug use and decreased muscle strength or poorer cognition in older adults, possibly related to the drugs' central effects on psychomotor slowing or their effects at the neuromuscular junction.

In the current study, 39% of participants were taking one or more drugs with anticholinergic activity. In this anticholinergic subgroup, 89% reported their health as good or better with the medications, compared with 96% of participants on no drugs with anticholinergic activity. Hypertension was present in 60% of those in the anticholinergic group and 51% of those in the nonanticholinergic group.

Performance-based assessments of ADL and IADL found that the likelihood of being dependent on help for ADL or IADL increased 10%–15% for every 1-point increase in the cumulative anticholinergic burden score, Dr. Sink said.

The results were adjusted for potentially confounding factors including age, sex, alcohol use, body mass index, hypertension, self-reported health, and clinic site. Effects on gait speed were adjusted for the person's height. Every 1-point increase in the anticholinergic burden score decreased gait speed by 0.009 m/sec.

The study was limited by an inability to completely control for the potentially confounding effects of the indications for the drug use. In addition, the anticholinergic burden scoring system did not consider dose effects.

The investigators will conduct a longitudinal study of whether anticholinergic burden predicts functional decline over time. Results should be available in a year or so, she said. If the anticholinergic burden score can predict who will decline faster, “then I think it would be useful to create a tool that clinicians can use in the office to calculate up each person's anticholinergic score,” Dr. Sink added. “I don't think it would be that hard” to create a computerized tool to assess anticholinergic burden.

Dr. Sink has no association with companies that make the drugs studied.

'If you'reon multipledrugs with anticholinergic activities, that might add up.' DR. SINK

SAN FRANCISCO — People who lose at least 30 pounds and successfully keep the weight off for more than a year share eight key characteristics, Suzanne Phelan, Ph.D., said at a meeting sponsored by the American Diabetes Association.

These characteristics may translate into weight-loss and maintenance strategies for patients at risk for diabetes, said Dr. Phelan of Brown University, Providence, R.I. Previous studies have shown that even a 5%–10% weight loss in overweight or obese people can significantly reduce the risk for diabetes.

Conventional behavioral weight-loss programs incorporating changes in diet and exercise can produce weight loss over a 6-month period, but most participants usually regain a third of the lost weight within 1 year, and after 5 years are back to their baseline weight or have gained weight, she said.

The National Weight Control Registry, established in 1994 by researchers at Brown University and the University of Colorado, analyzed data on 5,585 people who reduced their weight by a minimum of 30 pounds and maintained that for more than a year. Detailed initial questionnaires and annual follow-up surveys tracked their progress.

This cohort of “successful losers” was 77% female, 85% college educated, 95% white, 65% married, and an average of 46 years old. It's a self-selected group—“they call us because they want to join the registry,” Dr. Phelan said.

Weight loss averaged 32 kg in women and 34 kg in men, losses that were maintained for an average of 6 years; 13% of participants had kept the weight off for 10 years or more. The weight loss was achieved through diet and exercise in 89% of cases, through diet alone in 10%, and by exercise alone in 1%.

Dr. Phelan noted that 45% of respondents said they lost the weight on their own. “This still surprises me. We're investigating them further,” she said. The other 55% joined formal weight-loss programs to shed the weight. To mimic the characteristics of successful losers, Dr. Phelan suggested these eight strategies:

▸ Eat a low-calorie, low-fat diet. Energy intake averaged 1,385 kcal/day, with 27% of energy coming from fat. Using meal-replacement products may help people not only to lose weight, but to keep it off, compared with people who eat conventional foods during weight loss, studies suggest. “I think they're underutilized,” said Dr. Phelan, who has no association with companies that make meal-replacement products.

▸ Engage in a high level of physical activity. Energy expenditures averaged 2,545 kcal/week for women and 3,293 kcal/week for men.

▸ Limit TV viewing. Only 12% watched 21 or more hours of TV a week, 25% watched 11–20 hours, and more than 62% watched fewer than 10 hours a week.

▸ Step on a scale often. In all, 31% of respondents weighed themselves weekly, and 44% weighed themselves daily. Those who stopped weighing regained weight.

▸ Maintain diet consistency. About half of successful losers eat the same way on weekends and holidays as they do on weekdays.

▸ Limit diet variety. Successful losers keep fewer different types of foods on hand. Dr. Phelan suggested selecting just one or two brands of crackers or other foods for the home.

▸ Eat breakfast. The day starts with breakfast for 78% of successful losers.

▸ Limit fast food. The weight-loss maintainers ate fast food less than once a week (0.77 times a week, on average).

ELSEVIER GLOBAL MEDICAL NEWS

SAN FRANCISCO — People who lose at least 30 pounds and successfully keep the weight off for more than a year share eight key characteristics, Suzanne Phelan, Ph.D., said at a meeting sponsored by the American Diabetes Association.

These characteristics may translate into weight-loss and maintenance strategies for patients at risk for diabetes, said Dr. Phelan of Brown University, Providence, R.I. Previous studies have shown that even a 5%–10% weight loss in overweight or obese people can significantly reduce the risk for diabetes.

Conventional behavioral weight-loss programs incorporating changes in diet and exercise can produce weight loss over a 6-month period, but most participants usually regain a third of the lost weight within 1 year, and after 5 years are back to their baseline weight or have gained weight, she said.

The National Weight Control Registry, established in 1994 by researchers at Brown University and the University of Colorado, analyzed data on 5,585 people who reduced their weight by a minimum of 30 pounds and maintained that for more than a year. Detailed initial questionnaires and annual follow-up surveys tracked their progress.

This cohort of “successful losers” was 77% female, 85% college educated, 95% white, 65% married, and an average of 46 years old. It's a self-selected group—“they call us because they want to join the registry,” Dr. Phelan said.

Weight loss averaged 32 kg in women and 34 kg in men, losses that were maintained for an average of 6 years; 13% of participants had kept the weight off for 10 years or more. The weight loss was achieved through diet and exercise in 89% of cases, through diet alone in 10%, and by exercise alone in 1%.

Dr. Phelan noted that 45% of respondents said they lost the weight on their own. “This still surprises me. We're investigating them further,” she said. The other 55% joined formal weight-loss programs to shed the weight. To mimic the characteristics of successful losers, Dr. Phelan suggested these eight strategies:

▸ Eat a low-calorie, low-fat diet. Energy intake averaged 1,385 kcal/day, with 27% of energy coming from fat. Using meal-replacement products may help people not only to lose weight, but to keep it off, compared with people who eat conventional foods during weight loss, studies suggest. “I think they're underutilized,” said Dr. Phelan, who has no association with companies that make meal-replacement products.

▸ Engage in a high level of physical activity. Energy expenditures averaged 2,545 kcal/week for women and 3,293 kcal/week for men.

▸ Limit TV viewing. Only 12% watched 21 or more hours of TV a week, 25% watched 11–20 hours, and more than 62% watched fewer than 10 hours a week.

▸ Step on a scale often. In all, 31% of respondents weighed themselves weekly, and 44% weighed themselves daily. Those who stopped weighing regained weight.

▸ Maintain diet consistency. About half of successful losers eat the same way on weekends and holidays as they do on weekdays.

▸ Limit diet variety. Successful losers keep fewer different types of foods on hand. Dr. Phelan suggested selecting just one or two brands of crackers or other foods for the home.

▸ Eat breakfast. The day starts with breakfast for 78% of successful losers.

▸ Limit fast food. The weight-loss maintainers ate fast food less than once a week (0.77 times a week, on average).

ELSEVIER GLOBAL MEDICAL NEWS

SAN FRANCISCO — People who lose at least 30 pounds and successfully keep the weight off for more than a year share eight key characteristics, Suzanne Phelan, Ph.D., said at a meeting sponsored by the American Diabetes Association.

These characteristics may translate into weight-loss and maintenance strategies for patients at risk for diabetes, said Dr. Phelan of Brown University, Providence, R.I. Previous studies have shown that even a 5%–10% weight loss in overweight or obese people can significantly reduce the risk for diabetes.

Conventional behavioral weight-loss programs incorporating changes in diet and exercise can produce weight loss over a 6-month period, but most participants usually regain a third of the lost weight within 1 year, and after 5 years are back to their baseline weight or have gained weight, she said.

The National Weight Control Registry, established in 1994 by researchers at Brown University and the University of Colorado, analyzed data on 5,585 people who reduced their weight by a minimum of 30 pounds and maintained that for more than a year. Detailed initial questionnaires and annual follow-up surveys tracked their progress.

This cohort of “successful losers” was 77% female, 85% college educated, 95% white, 65% married, and an average of 46 years old. It's a self-selected group—“they call us because they want to join the registry,” Dr. Phelan said.

Weight loss averaged 32 kg in women and 34 kg in men, losses that were maintained for an average of 6 years; 13% of participants had kept the weight off for 10 years or more. The weight loss was achieved through diet and exercise in 89% of cases, through diet alone in 10%, and by exercise alone in 1%.

Dr. Phelan noted that 45% of respondents said they lost the weight on their own. “This still surprises me. We're investigating them further,” she said. The other 55% joined formal weight-loss programs to shed the weight. To mimic the characteristics of successful losers, Dr. Phelan suggested these eight strategies:

▸ Eat a low-calorie, low-fat diet. Energy intake averaged 1,385 kcal/day, with 27% of energy coming from fat. Using meal-replacement products may help people not only to lose weight, but to keep it off, compared with people who eat conventional foods during weight loss, studies suggest. “I think they're underutilized,” said Dr. Phelan, who has no association with companies that make meal-replacement products.

▸ Engage in a high level of physical activity. Energy expenditures averaged 2,545 kcal/week for women and 3,293 kcal/week for men.

▸ Limit TV viewing. Only 12% watched 21 or more hours of TV a week, 25% watched 11–20 hours, and more than 62% watched fewer than 10 hours a week.

▸ Step on a scale often. In all, 31% of respondents weighed themselves weekly, and 44% weighed themselves daily. Those who stopped weighing regained weight.

▸ Maintain diet consistency. About half of successful losers eat the same way on weekends and holidays as they do on weekdays.

▸ Limit diet variety. Successful losers keep fewer different types of foods on hand. Dr. Phelan suggested selecting just one or two brands of crackers or other foods for the home.

▸ Eat breakfast. The day starts with breakfast for 78% of successful losers.

▸ Limit fast food. The weight-loss maintainers ate fast food less than once a week (0.77 times a week, on average).

ELSEVIER GLOBAL MEDICAL NEWS

SAN FRANCISCO — Men whose prostate-specific antigen levels were less than 3 ng/mL at their initial screenings had a 20-fold lower risk of dying of prostate cancer, compared with men who presented with higher PSA levels in a study that tracked 19,970 men for 12 years.

The 15,582 men who initially presented with PSA levels less than 3 ng/mL and a normal digital rectal exam did not undergo biopsy, Dr. Monique J. Roobol told a symposium on genitourinary cancers.

Biopsy was offered to men with higher screening PSA levels or positive digital rectal exams in the Rotterdam section of the European Randomized Study of Screening for Prostate Cancer. Rescreening was offered after 1, 4, and 8 years. The analysis included all prostate cancers detected at or between screenings.

Of men biopsied because of screening PSA levels of 3 ng/mL or greater, 1% died of prostate cancer. In the group with initial PSA levels less than 3 ng/mL, 700 prostate cancers were subsequently detected, but only eight men (0.05%) died of the disease. To prevent one prostate cancer death, 1,981 men with initial PSA levels less than 3 ng/mL need to be biopsied, an unacceptable rate, said Dr. Roobol of the urology department at Erasmus University, Rotterdam, the Netherlands.

Mounting evidence suggests prostate cancer in men with low PSA levels may be indolent disease with better outcomes than cancer detected in men with higher PSA levels, the researchers said.

New risk markers need to be developed if the few prostate cancer deaths in men with screening PSA levels less than 3 ng/mL are to be prevented, said Dr. Roobol at the symposium, sponsored by the American Society of Clinical Oncology, the American Society for Therapeutic Radiology and Oncology, and the Society of Urologic Oncology. None of 139 men who presented with a PSA level less than 1 ng/mL died of prostate cancer.

Three of the eight men in the no-initial-biopsy group who died of prostate cancer were diagnosed with the disease within a year of initial screening; in two of the three, the PSA level did not change between screening and diagnosis. The other five in the no-initial-biopsy group who died of prostate cancer were diagnosed 2–8 years after initial screening. The time from diagnosis to death for these men ranged from 6 months to 8 years.

SAN FRANCISCO — Men whose prostate-specific antigen levels were less than 3 ng/mL at their initial screenings had a 20-fold lower risk of dying of prostate cancer, compared with men who presented with higher PSA levels in a study that tracked 19,970 men for 12 years.

The 15,582 men who initially presented with PSA levels less than 3 ng/mL and a normal digital rectal exam did not undergo biopsy, Dr. Monique J. Roobol told a symposium on genitourinary cancers.

Biopsy was offered to men with higher screening PSA levels or positive digital rectal exams in the Rotterdam section of the European Randomized Study of Screening for Prostate Cancer. Rescreening was offered after 1, 4, and 8 years. The analysis included all prostate cancers detected at or between screenings.

Of men biopsied because of screening PSA levels of 3 ng/mL or greater, 1% died of prostate cancer. In the group with initial PSA levels less than 3 ng/mL, 700 prostate cancers were subsequently detected, but only eight men (0.05%) died of the disease. To prevent one prostate cancer death, 1,981 men with initial PSA levels less than 3 ng/mL need to be biopsied, an unacceptable rate, said Dr. Roobol of the urology department at Erasmus University, Rotterdam, the Netherlands.

Mounting evidence suggests prostate cancer in men with low PSA levels may be indolent disease with better outcomes than cancer detected in men with higher PSA levels, the researchers said.

New risk markers need to be developed if the few prostate cancer deaths in men with screening PSA levels less than 3 ng/mL are to be prevented, said Dr. Roobol at the symposium, sponsored by the American Society of Clinical Oncology, the American Society for Therapeutic Radiology and Oncology, and the Society of Urologic Oncology. None of 139 men who presented with a PSA level less than 1 ng/mL died of prostate cancer.

Three of the eight men in the no-initial-biopsy group who died of prostate cancer were diagnosed with the disease within a year of initial screening; in two of the three, the PSA level did not change between screening and diagnosis. The other five in the no-initial-biopsy group who died of prostate cancer were diagnosed 2–8 years after initial screening. The time from diagnosis to death for these men ranged from 6 months to 8 years.

SAN FRANCISCO — Men whose prostate-specific antigen levels were less than 3 ng/mL at their initial screenings had a 20-fold lower risk of dying of prostate cancer, compared with men who presented with higher PSA levels in a study that tracked 19,970 men for 12 years.

The 15,582 men who initially presented with PSA levels less than 3 ng/mL and a normal digital rectal exam did not undergo biopsy, Dr. Monique J. Roobol told a symposium on genitourinary cancers.

Biopsy was offered to men with higher screening PSA levels or positive digital rectal exams in the Rotterdam section of the European Randomized Study of Screening for Prostate Cancer. Rescreening was offered after 1, 4, and 8 years. The analysis included all prostate cancers detected at or between screenings.

Of men biopsied because of screening PSA levels of 3 ng/mL or greater, 1% died of prostate cancer. In the group with initial PSA levels less than 3 ng/mL, 700 prostate cancers were subsequently detected, but only eight men (0.05%) died of the disease. To prevent one prostate cancer death, 1,981 men with initial PSA levels less than 3 ng/mL need to be biopsied, an unacceptable rate, said Dr. Roobol of the urology department at Erasmus University, Rotterdam, the Netherlands.

Mounting evidence suggests prostate cancer in men with low PSA levels may be indolent disease with better outcomes than cancer detected in men with higher PSA levels, the researchers said.

New risk markers need to be developed if the few prostate cancer deaths in men with screening PSA levels less than 3 ng/mL are to be prevented, said Dr. Roobol at the symposium, sponsored by the American Society of Clinical Oncology, the American Society for Therapeutic Radiology and Oncology, and the Society of Urologic Oncology. None of 139 men who presented with a PSA level less than 1 ng/mL died of prostate cancer.

Three of the eight men in the no-initial-biopsy group who died of prostate cancer were diagnosed with the disease within a year of initial screening; in two of the three, the PSA level did not change between screening and diagnosis. The other five in the no-initial-biopsy group who died of prostate cancer were diagnosed 2–8 years after initial screening. The time from diagnosis to death for these men ranged from 6 months to 8 years.

SAN FRANCISCO — For patients who need large doses of insulin (more than 200 U/day), U-500 insulin is the best choice because of more predictable pharmacokinetics and lower cost per unit.

“[With] a huge volume of insulin—60 or 80 U—you're going to have more variability in the absorption” with conventional insulins such as U-100, lispro, or glargine, Dr. Irl B. Hirsch told a meeting sponsored by the American Diabetes Association.

Although U-500 insulin is called “regular” insulin, “it ain't like regular insulin” because it's five times more concentrated and has longer pharmacokinetics, said Dr. Hirsch, professor of medicine at the University of Washington, Seattle. This has caused confusion in some hospitals regarding dosing.

For clarity, some physicians refer to milliliter or cubic centimeter measurements when referring to U-500 insulin, but “there's no formal consensus on this,” he said.

Twenty U of U-500 insulin in a U-100 syringe is 0.2 mL—the same dose as 100 U of U-100 insulin in a volume of 1 mL. U-500 insulin is available only from Eli Lilly & Co., for which Dr. Hirsch is a consultant.

The duration of action of U-500 insulin is up to 24 hours. Large doses of insulin can be given with one-fifth the volume using U-500 insulin, so there's less day-to-day variation in absorption and less variability of absorption in different body regions.

U-500 insulin can be helpful especially in those needing large depots of insulin but who have little subcutaneous tissue. A smaller volume of insulin is less painful for these patients.

The pharmacokinetic and pharmacodynamic characteristics of huge doses of either conventional or NPH insulin per injection have not been well studied since not many patients need such high doses, Dr. Hirsch noted.

With U-500 insulin, “Don't think about this as giving prandial insulin. The basal/prandial distinction we make with insulin components for a typical basal bolus sort of goes away when we're talking about U-500 insulin, since it is really both.”

The National Institutes of Health published an algorithm suggesting that insulin-resistant patients who need less than 200 U/day use U-100 insulin, and that U-500 insulin be considered for severely insulin-resistant patients who need more (Diabetes Care 2005;28:1240–4).

A twice-a-day regimen of U-500 insulin would be used for patients who need 200–300 U/day, and a three-times-a-day regimen would apply to patients who need 300–750 U/day. For 750–2,000 U/day, patients would use U-500 insulin t.i.d. plus a fourth dose at bedtime. Above 2,000 U/day, an insulin pump is best, Dr. Hirsch said.

Although a 20-mL vial of U-500 insulin costs about $260, compared with $43-$89 for a 10-mL vial of U-100 insulin, lispro, or aspart, the cost per unit is cheapest with U-500—3 cents a unit rather than 4 to 9 cents “This is the economical way,” he said.

ELSEVIER GLOBAL MEDICAL NEWS

SAN FRANCISCO — For patients who need large doses of insulin (more than 200 U/day), U-500 insulin is the best choice because of more predictable pharmacokinetics and lower cost per unit.

“[With] a huge volume of insulin—60 or 80 U—you're going to have more variability in the absorption” with conventional insulins such as U-100, lispro, or glargine, Dr. Irl B. Hirsch told a meeting sponsored by the American Diabetes Association.

Although U-500 insulin is called “regular” insulin, “it ain't like regular insulin” because it's five times more concentrated and has longer pharmacokinetics, said Dr. Hirsch, professor of medicine at the University of Washington, Seattle. This has caused confusion in some hospitals regarding dosing.

For clarity, some physicians refer to milliliter or cubic centimeter measurements when referring to U-500 insulin, but “there's no formal consensus on this,” he said.

Twenty U of U-500 insulin in a U-100 syringe is 0.2 mL—the same dose as 100 U of U-100 insulin in a volume of 1 mL. U-500 insulin is available only from Eli Lilly & Co., for which Dr. Hirsch is a consultant.

The duration of action of U-500 insulin is up to 24 hours. Large doses of insulin can be given with one-fifth the volume using U-500 insulin, so there's less day-to-day variation in absorption and less variability of absorption in different body regions.

U-500 insulin can be helpful especially in those needing large depots of insulin but who have little subcutaneous tissue. A smaller volume of insulin is less painful for these patients.

The pharmacokinetic and pharmacodynamic characteristics of huge doses of either conventional or NPH insulin per injection have not been well studied since not many patients need such high doses, Dr. Hirsch noted.

With U-500 insulin, “Don't think about this as giving prandial insulin. The basal/prandial distinction we make with insulin components for a typical basal bolus sort of goes away when we're talking about U-500 insulin, since it is really both.”

The National Institutes of Health published an algorithm suggesting that insulin-resistant patients who need less than 200 U/day use U-100 insulin, and that U-500 insulin be considered for severely insulin-resistant patients who need more (Diabetes Care 2005;28:1240–4).

A twice-a-day regimen of U-500 insulin would be used for patients who need 200–300 U/day, and a three-times-a-day regimen would apply to patients who need 300–750 U/day. For 750–2,000 U/day, patients would use U-500 insulin t.i.d. plus a fourth dose at bedtime. Above 2,000 U/day, an insulin pump is best, Dr. Hirsch said.

Although a 20-mL vial of U-500 insulin costs about $260, compared with $43-$89 for a 10-mL vial of U-100 insulin, lispro, or aspart, the cost per unit is cheapest with U-500—3 cents a unit rather than 4 to 9 cents “This is the economical way,” he said.

ELSEVIER GLOBAL MEDICAL NEWS

SAN FRANCISCO — For patients who need large doses of insulin (more than 200 U/day), U-500 insulin is the best choice because of more predictable pharmacokinetics and lower cost per unit.

“[With] a huge volume of insulin—60 or 80 U—you're going to have more variability in the absorption” with conventional insulins such as U-100, lispro, or glargine, Dr. Irl B. Hirsch told a meeting sponsored by the American Diabetes Association.

Although U-500 insulin is called “regular” insulin, “it ain't like regular insulin” because it's five times more concentrated and has longer pharmacokinetics, said Dr. Hirsch, professor of medicine at the University of Washington, Seattle. This has caused confusion in some hospitals regarding dosing.

For clarity, some physicians refer to milliliter or cubic centimeter measurements when referring to U-500 insulin, but “there's no formal consensus on this,” he said.

Twenty U of U-500 insulin in a U-100 syringe is 0.2 mL—the same dose as 100 U of U-100 insulin in a volume of 1 mL. U-500 insulin is available only from Eli Lilly & Co., for which Dr. Hirsch is a consultant.

The duration of action of U-500 insulin is up to 24 hours. Large doses of insulin can be given with one-fifth the volume using U-500 insulin, so there's less day-to-day variation in absorption and less variability of absorption in different body regions.

U-500 insulin can be helpful especially in those needing large depots of insulin but who have little subcutaneous tissue. A smaller volume of insulin is less painful for these patients.

The pharmacokinetic and pharmacodynamic characteristics of huge doses of either conventional or NPH insulin per injection have not been well studied since not many patients need such high doses, Dr. Hirsch noted.

With U-500 insulin, “Don't think about this as giving prandial insulin. The basal/prandial distinction we make with insulin components for a typical basal bolus sort of goes away when we're talking about U-500 insulin, since it is really both.”

The National Institutes of Health published an algorithm suggesting that insulin-resistant patients who need less than 200 U/day use U-100 insulin, and that U-500 insulin be considered for severely insulin-resistant patients who need more (Diabetes Care 2005;28:1240–4).

A twice-a-day regimen of U-500 insulin would be used for patients who need 200–300 U/day, and a three-times-a-day regimen would apply to patients who need 300–750 U/day. For 750–2,000 U/day, patients would use U-500 insulin t.i.d. plus a fourth dose at bedtime. Above 2,000 U/day, an insulin pump is best, Dr. Hirsch said.

Although a 20-mL vial of U-500 insulin costs about $260, compared with $43-$89 for a 10-mL vial of U-100 insulin, lispro, or aspart, the cost per unit is cheapest with U-500—3 cents a unit rather than 4 to 9 cents “This is the economical way,” he said.

ELSEVIER GLOBAL MEDICAL NEWS

SAN FRANCISCO — The heterogeneous nature of diabetes in the elderly makes it imperative to assess each patient individually before deciding whether to use aggressive or more conservative therapy, Dr. Hermes Florez said.

Some older diabetes patients have newly diagnosed disease and are quite functional, whereas others have long-standing disease and significant functional decline. Older adults are more likely to have multiple comorbidities and to be taking multiple medications.

It is also important to consider life expectancy, noted Dr. Florez, an endocrinologist at the University of Miami and the Miami Veterans Affairs Medical Center, at a meeting sponsored by the American Diabetes Association.

To help chart an individual's management plan, one should balance the potential benefits of aggressive glycemic control against the risks from comorbidities, medication side effects, and geriatric syndromes such as dementia, incontinence, and depression, advised Dr. Florez. He described the following sample cases to highlight treatment choices:

▸ Low risk, high benefit. Aggressive treatment was an easy decision for a 70-year-old woman with a 20-year history of diabetes who also had hypertension, lipid abnormalities, and early appearance of retinopathy but who functioned well independently and had no other comorbidities.

▸ High risk, low benefit. The opposite was true for a 68-year-old man with a 4-year history of diabetes who also had severe cardiomyopathy with ventricular tachycardia and couldn't walk. He already was taking 14 medications. Intensifying treatment for better blood pressure, lipid levels, or blood-sugar control could pose greater risks than benefits.

▸ Low risk, low benefit. Less easy to manage was a 75-year-old woman with new-onset diabetes, none of the associated cardiovascular risk factors, no other comorbidities, and no functional impairment. She's at low risk, but evidence is lacking that she would benefit from intensive therapy to lower her HbA_1c level below 6.5.

▸ High risk, low benefit. A 72-year-old man with long-standing diabetes of 18 years' duration, a history of multiple hypoglycemic episodes, and complications related to diabetes. Intensive therapy for blood glucose levels, lipids, and blood pressure probably would seem indicated, but he also had major cognitive deficits. Unless a relative or caregiver can monitor therapy, intensive treatment poses too much risk for side effects, falls, or further cognitive decline. Treatreat should be conservative.

Dr. Florez has received research funding from Merck & Co., a maker of diabetes medications.

SAN FRANCISCO — The heterogeneous nature of diabetes in the elderly makes it imperative to assess each patient individually before deciding whether to use aggressive or more conservative therapy, Dr. Hermes Florez said.

Some older diabetes patients have newly diagnosed disease and are quite functional, whereas others have long-standing disease and significant functional decline. Older adults are more likely to have multiple comorbidities and to be taking multiple medications.

It is also important to consider life expectancy, noted Dr. Florez, an endocrinologist at the University of Miami and the Miami Veterans Affairs Medical Center, at a meeting sponsored by the American Diabetes Association.

To help chart an individual's management plan, one should balance the potential benefits of aggressive glycemic control against the risks from comorbidities, medication side effects, and geriatric syndromes such as dementia, incontinence, and depression, advised Dr. Florez. He described the following sample cases to highlight treatment choices:

▸ Low risk, high benefit. Aggressive treatment was an easy decision for a 70-year-old woman with a 20-year history of diabetes who also had hypertension, lipid abnormalities, and early appearance of retinopathy but who functioned well independently and had no other comorbidities.

▸ High risk, low benefit. The opposite was true for a 68-year-old man with a 4-year history of diabetes who also had severe cardiomyopathy with ventricular tachycardia and couldn't walk. He already was taking 14 medications. Intensifying treatment for better blood pressure, lipid levels, or blood-sugar control could pose greater risks than benefits.

▸ Low risk, low benefit. Less easy to manage was a 75-year-old woman with new-onset diabetes, none of the associated cardiovascular risk factors, no other comorbidities, and no functional impairment. She's at low risk, but evidence is lacking that she would benefit from intensive therapy to lower her HbA_1c level below 6.5.

▸ High risk, low benefit. A 72-year-old man with long-standing diabetes of 18 years' duration, a history of multiple hypoglycemic episodes, and complications related to diabetes. Intensive therapy for blood glucose levels, lipids, and blood pressure probably would seem indicated, but he also had major cognitive deficits. Unless a relative or caregiver can monitor therapy, intensive treatment poses too much risk for side effects, falls, or further cognitive decline. Treatreat should be conservative.

Dr. Florez has received research funding from Merck & Co., a maker of diabetes medications.

SAN FRANCISCO — The heterogeneous nature of diabetes in the elderly makes it imperative to assess each patient individually before deciding whether to use aggressive or more conservative therapy, Dr. Hermes Florez said.

Some older diabetes patients have newly diagnosed disease and are quite functional, whereas others have long-standing disease and significant functional decline. Older adults are more likely to have multiple comorbidities and to be taking multiple medications.

It is also important to consider life expectancy, noted Dr. Florez, an endocrinologist at the University of Miami and the Miami Veterans Affairs Medical Center, at a meeting sponsored by the American Diabetes Association.

To help chart an individual's management plan, one should balance the potential benefits of aggressive glycemic control against the risks from comorbidities, medication side effects, and geriatric syndromes such as dementia, incontinence, and depression, advised Dr. Florez. He described the following sample cases to highlight treatment choices:

▸ Low risk, high benefit. Aggressive treatment was an easy decision for a 70-year-old woman with a 20-year history of diabetes who also had hypertension, lipid abnormalities, and early appearance of retinopathy but who functioned well independently and had no other comorbidities.

▸ High risk, low benefit. The opposite was true for a 68-year-old man with a 4-year history of diabetes who also had severe cardiomyopathy with ventricular tachycardia and couldn't walk. He already was taking 14 medications. Intensifying treatment for better blood pressure, lipid levels, or blood-sugar control could pose greater risks than benefits.

▸ Low risk, low benefit. Less easy to manage was a 75-year-old woman with new-onset diabetes, none of the associated cardiovascular risk factors, no other comorbidities, and no functional impairment. She's at low risk, but evidence is lacking that she would benefit from intensive therapy to lower her HbA_1c level below 6.5.

▸ High risk, low benefit. A 72-year-old man with long-standing diabetes of 18 years' duration, a history of multiple hypoglycemic episodes, and complications related to diabetes. Intensive therapy for blood glucose levels, lipids, and blood pressure probably would seem indicated, but he also had major cognitive deficits. Unless a relative or caregiver can monitor therapy, intensive treatment poses too much risk for side effects, falls, or further cognitive decline. Treatreat should be conservative.

Dr. Florez has received research funding from Merck & Co., a maker of diabetes medications.

SAN FRANCISCO — Diabetes patients whose lives are frequently disrupted by glycemic episodes requiring hospitalization need a multifaceted work-up to identify the cause of their brittle diabetes, Dr. Irl B. Hirsch said.

Compared with diabetes patients who have relatively stable and well-controlled glucose levels, patients with brittle diabetes tend to be younger and are more likely to be female, Dr. Hirsch said at a meeting sponsored by the American Diabetes Association. They also tend to have higher hemoglobin A_1c values and to use more insulin, and are more likely to have psychiatric disorders or psychosocial problems, including family disruption, adolescent crises, and personality disturbances, said Dr. Hirsch, professor of medicine at the University of Washington, Seattle.

Both physical and psychiatric problems can result in brittle diabetes. Look for the following to identify contributing factors:

▸ Counterregulatory hormone excess. This can cause insulin resistance and make diabetes very difficult to control. Cushing's disease, acromegaly, pheochromocytoma, and glucagonoma are rare causes of insulin resistance to consider in the work-up of brittle diabetes.

▸ Insulin antibodies. These are worth measuring but have a low yield in these patients. “Unless the physical exam suggests a diagnosis of Cushing's disease or acromegaly or some endocrine disorder like that,” go ahead and measure insulin antibodies, he said.

A potentially more productive approach is to measure both free and total insulin levels in a patient. A discrepancy in the results suggests that there's something binding up the insulin.

Be sure to use a clinical laboratory that knows how to measure free insulin, Dr. Hirsch cautioned. After the blood is drawn, several preparatory steps must be completed within 30 seconds before the sample is sent out to be assayed. “Not all labs do that,” he said.

▸ Celiac disease. This occurs in approximately 7%–8% of childhood-onset type 1 diabetes patients and can lead to brittle diabetes. Screening for celiac disease typically focuses on identifying the transglutaminase IgA antibody, but 5% of the population lacks IgA, Dr. Hirsch warned. Before looking for the antibody, measure IgA levels to make sure they're normal. If IgA is absent, the antibody test “becomes worthless,” he said.

▸ Gastroparesis. Considering this “makes sense. If you have a mismatching of food and insulin, it makes the diabetes that much more difficult to control,” Dr. Hirsch said.

▸ Injection and pump sites. Check these, because lipodystrophy may cause severe insulin resistance. Although the use of protease inhibitors is the most common cause of lipodystrophy in the general population, congenital lipodystrophy is the most frequent cause in Dr. Hirsch's patient population. Patients with congenital lipodystrophy typically get referred to him from a lipid clinic for severe hypertriglyceridemia and insulin resistance.

▸ Timing of insulin injections. Bad timing in relation to food intake probably is the most common mistake made in insulin therapy, he said. Problems with timing of injections usually won't disrupt a patient's life to the point of frequent hospitalizations, but they “certainly will cause glycemic instability,” he said.

▸ Psychological or psychiatric problems. These can interfere with diabetes management, so a mental health evaluation is a key part of the work-up for most patients with brittle diabetes. Frequent episodes of diabetic ketoacidosis usually result from patients' not taking their insulin, Dr. Hirsch said. Severe depression can make it difficult for a patient to manage diabetes.

The mental health evaluation “is best performed by someone knowledgeable about diabetes, particularly with experience in eating disorders, because that's the big issue, especially in adolescents,” he said.

Patients with brittle diabetes are more likely to have psychiatric disorders or psychosocial problems. DR. HIRSCH

SAN FRANCISCO — Diabetes patients whose lives are frequently disrupted by glycemic episodes requiring hospitalization need a multifaceted work-up to identify the cause of their brittle diabetes, Dr. Irl B. Hirsch said.

Compared with diabetes patients who have relatively stable and well-controlled glucose levels, patients with brittle diabetes tend to be younger and are more likely to be female, Dr. Hirsch said at a meeting sponsored by the American Diabetes Association. They also tend to have higher hemoglobin A_1c values and to use more insulin, and are more likely to have psychiatric disorders or psychosocial problems, including family disruption, adolescent crises, and personality disturbances, said Dr. Hirsch, professor of medicine at the University of Washington, Seattle.

Both physical and psychiatric problems can result in brittle diabetes. Look for the following to identify contributing factors:

▸ Counterregulatory hormone excess. This can cause insulin resistance and make diabetes very difficult to control. Cushing's disease, acromegaly, pheochromocytoma, and glucagonoma are rare causes of insulin resistance to consider in the work-up of brittle diabetes.

▸ Insulin antibodies. These are worth measuring but have a low yield in these patients. “Unless the physical exam suggests a diagnosis of Cushing's disease or acromegaly or some endocrine disorder like that,” go ahead and measure insulin antibodies, he said.

A potentially more productive approach is to measure both free and total insulin levels in a patient. A discrepancy in the results suggests that there's something binding up the insulin.

Be sure to use a clinical laboratory that knows how to measure free insulin, Dr. Hirsch cautioned. After the blood is drawn, several preparatory steps must be completed within 30 seconds before the sample is sent out to be assayed. “Not all labs do that,” he said.

▸ Celiac disease. This occurs in approximately 7%–8% of childhood-onset type 1 diabetes patients and can lead to brittle diabetes. Screening for celiac disease typically focuses on identifying the transglutaminase IgA antibody, but 5% of the population lacks IgA, Dr. Hirsch warned. Before looking for the antibody, measure IgA levels to make sure they're normal. If IgA is absent, the antibody test “becomes worthless,” he said.

▸ Gastroparesis. Considering this “makes sense. If you have a mismatching of food and insulin, it makes the diabetes that much more difficult to control,” Dr. Hirsch said.

▸ Injection and pump sites. Check these, because lipodystrophy may cause severe insulin resistance. Although the use of protease inhibitors is the most common cause of lipodystrophy in the general population, congenital lipodystrophy is the most frequent cause in Dr. Hirsch's patient population. Patients with congenital lipodystrophy typically get referred to him from a lipid clinic for severe hypertriglyceridemia and insulin resistance.

▸ Timing of insulin injections. Bad timing in relation to food intake probably is the most common mistake made in insulin therapy, he said. Problems with timing of injections usually won't disrupt a patient's life to the point of frequent hospitalizations, but they “certainly will cause glycemic instability,” he said.

▸ Psychological or psychiatric problems. These can interfere with diabetes management, so a mental health evaluation is a key part of the work-up for most patients with brittle diabetes. Frequent episodes of diabetic ketoacidosis usually result from patients' not taking their insulin, Dr. Hirsch said. Severe depression can make it difficult for a patient to manage diabetes.

The mental health evaluation “is best performed by someone knowledgeable about diabetes, particularly with experience in eating disorders, because that's the big issue, especially in adolescents,” he said.

Patients with brittle diabetes are more likely to have psychiatric disorders or psychosocial problems. DR. HIRSCH

SAN FRANCISCO — Diabetes patients whose lives are frequently disrupted by glycemic episodes requiring hospitalization need a multifaceted work-up to identify the cause of their brittle diabetes, Dr. Irl B. Hirsch said.

Compared with diabetes patients who have relatively stable and well-controlled glucose levels, patients with brittle diabetes tend to be younger and are more likely to be female, Dr. Hirsch said at a meeting sponsored by the American Diabetes Association. They also tend to have higher hemoglobin A_1c values and to use more insulin, and are more likely to have psychiatric disorders or psychosocial problems, including family disruption, adolescent crises, and personality disturbances, said Dr. Hirsch, professor of medicine at the University of Washington, Seattle.

Both physical and psychiatric problems can result in brittle diabetes. Look for the following to identify contributing factors:

▸ Counterregulatory hormone excess. This can cause insulin resistance and make diabetes very difficult to control. Cushing's disease, acromegaly, pheochromocytoma, and glucagonoma are rare causes of insulin resistance to consider in the work-up of brittle diabetes.

▸ Insulin antibodies. These are worth measuring but have a low yield in these patients. “Unless the physical exam suggests a diagnosis of Cushing's disease or acromegaly or some endocrine disorder like that,” go ahead and measure insulin antibodies, he said.

A potentially more productive approach is to measure both free and total insulin levels in a patient. A discrepancy in the results suggests that there's something binding up the insulin.

Be sure to use a clinical laboratory that knows how to measure free insulin, Dr. Hirsch cautioned. After the blood is drawn, several preparatory steps must be completed within 30 seconds before the sample is sent out to be assayed. “Not all labs do that,” he said.

▸ Celiac disease. This occurs in approximately 7%–8% of childhood-onset type 1 diabetes patients and can lead to brittle diabetes. Screening for celiac disease typically focuses on identifying the transglutaminase IgA antibody, but 5% of the population lacks IgA, Dr. Hirsch warned. Before looking for the antibody, measure IgA levels to make sure they're normal. If IgA is absent, the antibody test “becomes worthless,” he said.

▸ Gastroparesis. Considering this “makes sense. If you have a mismatching of food and insulin, it makes the diabetes that much more difficult to control,” Dr. Hirsch said.

▸ Injection and pump sites. Check these, because lipodystrophy may cause severe insulin resistance. Although the use of protease inhibitors is the most common cause of lipodystrophy in the general population, congenital lipodystrophy is the most frequent cause in Dr. Hirsch's patient population. Patients with congenital lipodystrophy typically get referred to him from a lipid clinic for severe hypertriglyceridemia and insulin resistance.

▸ Timing of insulin injections. Bad timing in relation to food intake probably is the most common mistake made in insulin therapy, he said. Problems with timing of injections usually won't disrupt a patient's life to the point of frequent hospitalizations, but they “certainly will cause glycemic instability,” he said.

▸ Psychological or psychiatric problems. These can interfere with diabetes management, so a mental health evaluation is a key part of the work-up for most patients with brittle diabetes. Frequent episodes of diabetic ketoacidosis usually result from patients' not taking their insulin, Dr. Hirsch said. Severe depression can make it difficult for a patient to manage diabetes.

The mental health evaluation “is best performed by someone knowledgeable about diabetes, particularly with experience in eating disorders, because that's the big issue, especially in adolescents,” he said.

Patients with brittle diabetes are more likely to have psychiatric disorders or psychosocial problems. DR. HIRSCH

SAN FRANCISCO — Give e-mail correspondence with patients the same care and attention you'd give to paper records, faxes, or phone calls to minimize medicolegal liability, Dr. Jeffrey L. Brown said.

Physicians should be reasonably certain that the person requesting information by e-mail is authorized to receive it, just as would be done with phone calls, he said at the annual meeting of the American Academy of Pediatrics.

Your e-mail system should include an automated response to any e-mails received from patients, acknowledging that an e-mail message has been received and saying that you will respond within a set period of time, such as 24 or 48 hours, said Dr. Brown, of Cornell University, New York, and in private practice in Rye Brook, N.Y. He has no association with companies that market e-mail systems or services.

The automated response should alert patients that confidentiality cannot always be assured in e-mail correspondence, and that you cannot respond to urgent questions posed by e-mail. Patients should contact your office by phone for urgent matters.

The response also should inform patients that if they do not get your reply to any e-mail message within a reasonable period of time—“usually 48 hours,” Dr. Brown said—the patient should call your office because you may not have received the e-mail. If you are away from the office when patients e-mail, the automated response should give the date of your return.

In the other direction, e-mails sent by physicians must be compliant with the Health Insurance Portability and Accountability Act (HIPAA). As with faxes, conventional e-mails must protect the confidentiality of sensitive information such as Social Security numbers, medical identification numbers, laboratory results, diagnoses, medications, and more.

To ensure confidentiality in e-mails, use an encrypted message system, Dr. Brown advised. Solo practitioners or small practices may want to do an Internet search for the term “encrypting e-mail systems” to find a list of encryption providers, he said. Typically, an outgoing e-mail would be sent to the provider, encrypted, and returned to the physician's system before going out to a patient.

Confidential e-mail from physicians should contain a warning disclaimer similar to those used on fax transmissions. A typical disclaimer says the following: “This e-mail contains confidential and privileged information. It is intended only for the individual or entity to whom it is addressed. If you are not the intended recipient, or if you have received this transmission in error, you are hereby instructed to notify the sender and to erase its content and all attachments immediately. Copying, disseminating, or otherwise utilizing any of its content is unlawful and strictly prohibited.”

Treat e-mail messages like other patient correspondence, and file them appropriately, he added. Before erasing e-mail, save the patient's original e-mail and your response as hard copies in the patient's chart or electronically if you use electronic charts. Take precautions to protect confidential information on laptop computers and hard drives, as you would for other records. Use encryption software or change passwords frequently to prevent unauthorized access. Erase all confidential information from hard drives before disposing of them.

Other suggestions include not using your personal e-mail address to answer patient e-mails, not answering a new patient's e-mailed medical questions without first establishing a formal relationship, and not using an indiscreet topic in the heading of your response. “Don't write, 'Your pregnancy test is positive' in the subject line,” he said. “Say, 'I have your lab work,' or something like that.

“Even if you do all the right things, there is still a possibility that you will be subject to suits,” Dr. Brown said. “In the end, the best defense against legal action is practicing good medicine.”

SAN FRANCISCO — Give e-mail correspondence with patients the same care and attention you'd give to paper records, faxes, or phone calls to minimize medicolegal liability, Dr. Jeffrey L. Brown said.

Physicians should be reasonably certain that the person requesting information by e-mail is authorized to receive it, just as would be done with phone calls, he said at the annual meeting of the American Academy of Pediatrics.

Your e-mail system should include an automated response to any e-mails received from patients, acknowledging that an e-mail message has been received and saying that you will respond within a set period of time, such as 24 or 48 hours, said Dr. Brown, of Cornell University, New York, and in private practice in Rye Brook, N.Y. He has no association with companies that market e-mail systems or services.

The automated response should alert patients that confidentiality cannot always be assured in e-mail correspondence, and that you cannot respond to urgent questions posed by e-mail. Patients should contact your office by phone for urgent matters.

The response also should inform patients that if they do not get your reply to any e-mail message within a reasonable period of time—“usually 48 hours,” Dr. Brown said—the patient should call your office because you may not have received the e-mail. If you are away from the office when patients e-mail, the automated response should give the date of your return.

In the other direction, e-mails sent by physicians must be compliant with the Health Insurance Portability and Accountability Act (HIPAA). As with faxes, conventional e-mails must protect the confidentiality of sensitive information such as Social Security numbers, medical identification numbers, laboratory results, diagnoses, medications, and more.

To ensure confidentiality in e-mails, use an encrypted message system, Dr. Brown advised. Solo practitioners or small practices may want to do an Internet search for the term “encrypting e-mail systems” to find a list of encryption providers, he said. Typically, an outgoing e-mail would be sent to the provider, encrypted, and returned to the physician's system before going out to a patient.

Confidential e-mail from physicians should contain a warning disclaimer similar to those used on fax transmissions. A typical disclaimer says the following: “This e-mail contains confidential and privileged information. It is intended only for the individual or entity to whom it is addressed. If you are not the intended recipient, or if you have received this transmission in error, you are hereby instructed to notify the sender and to erase its content and all attachments immediately. Copying, disseminating, or otherwise utilizing any of its content is unlawful and strictly prohibited.”

Treat e-mail messages like other patient correspondence, and file them appropriately, he added. Before erasing e-mail, save the patient's original e-mail and your response as hard copies in the patient's chart or electronically if you use electronic charts. Take precautions to protect confidential information on laptop computers and hard drives, as you would for other records. Use encryption software or change passwords frequently to prevent unauthorized access. Erase all confidential information from hard drives before disposing of them.

Other suggestions include not using your personal e-mail address to answer patient e-mails, not answering a new patient's e-mailed medical questions without first establishing a formal relationship, and not using an indiscreet topic in the heading of your response. “Don't write, 'Your pregnancy test is positive' in the subject line,” he said. “Say, 'I have your lab work,' or something like that.

“Even if you do all the right things, there is still a possibility that you will be subject to suits,” Dr. Brown said. “In the end, the best defense against legal action is practicing good medicine.”

SAN FRANCISCO — Give e-mail correspondence with patients the same care and attention you'd give to paper records, faxes, or phone calls to minimize medicolegal liability, Dr. Jeffrey L. Brown said.

Physicians should be reasonably certain that the person requesting information by e-mail is authorized to receive it, just as would be done with phone calls, he said at the annual meeting of the American Academy of Pediatrics.

Your e-mail system should include an automated response to any e-mails received from patients, acknowledging that an e-mail message has been received and saying that you will respond within a set period of time, such as 24 or 48 hours, said Dr. Brown, of Cornell University, New York, and in private practice in Rye Brook, N.Y. He has no association with companies that market e-mail systems or services.

The automated response should alert patients that confidentiality cannot always be assured in e-mail correspondence, and that you cannot respond to urgent questions posed by e-mail. Patients should contact your office by phone for urgent matters.

The response also should inform patients that if they do not get your reply to any e-mail message within a reasonable period of time—“usually 48 hours,” Dr. Brown said—the patient should call your office because you may not have received the e-mail. If you are away from the office when patients e-mail, the automated response should give the date of your return.

In the other direction, e-mails sent by physicians must be compliant with the Health Insurance Portability and Accountability Act (HIPAA). As with faxes, conventional e-mails must protect the confidentiality of sensitive information such as Social Security numbers, medical identification numbers, laboratory results, diagnoses, medications, and more.

To ensure confidentiality in e-mails, use an encrypted message system, Dr. Brown advised. Solo practitioners or small practices may want to do an Internet search for the term “encrypting e-mail systems” to find a list of encryption providers, he said. Typically, an outgoing e-mail would be sent to the provider, encrypted, and returned to the physician's system before going out to a patient.

Confidential e-mail from physicians should contain a warning disclaimer similar to those used on fax transmissions. A typical disclaimer says the following: “This e-mail contains confidential and privileged information. It is intended only for the individual or entity to whom it is addressed. If you are not the intended recipient, or if you have received this transmission in error, you are hereby instructed to notify the sender and to erase its content and all attachments immediately. Copying, disseminating, or otherwise utilizing any of its content is unlawful and strictly prohibited.”

Treat e-mail messages like other patient correspondence, and file them appropriately, he added. Before erasing e-mail, save the patient's original e-mail and your response as hard copies in the patient's chart or electronically if you use electronic charts. Take precautions to protect confidential information on laptop computers and hard drives, as you would for other records. Use encryption software or change passwords frequently to prevent unauthorized access. Erase all confidential information from hard drives before disposing of them.

Other suggestions include not using your personal e-mail address to answer patient e-mails, not answering a new patient's e-mailed medical questions without first establishing a formal relationship, and not using an indiscreet topic in the heading of your response. “Don't write, 'Your pregnancy test is positive' in the subject line,” he said. “Say, 'I have your lab work,' or something like that.

“Even if you do all the right things, there is still a possibility that you will be subject to suits,” Dr. Brown said. “In the end, the best defense against legal action is practicing good medicine.”

SAN FRANCISCO — Diabetes patients whose lives are frequently disrupted by glycemic episodes requiring hospitalization need a multifaceted work-up to identify the cause of their brittle diabetes, Dr. Irl B. Hirsch said.

Compared with diabetes patients who have relatively stable and well-controlled glucose levels, patients with brittle diabetes tend to be younger and are more likely to be female, Dr. Hirsch said at a meeting sponsored by the American Diabetes Association. They also tend to have higher hemoglobin A_1c values and to use more insulin, and are more likely to have psychiatric disorders or psychosocial problems, including family disruption, adolescent crises, and personality disturbances, said Dr. Hirsch, professor of medicine at the University of Washington, Seattle.

Both physical and psychiatric problems can result in brittle diabetes. Look for the following to identify contributing factors, he said:

▸ Counterregulatory hormone excess. This can cause insulin resistance and make diabetes very difficult to control. Cushing's disease, acromegaly, pheochromocytoma, and glucagonoma are rare causes of insulin resistance to consider in the work-up of brittle diabetes.

▸ Insulin antibodies. These are worth measuring but have a low yield in these patients. “Unless the physical exam suggests a diagnosis of Cushing's disease or acromegaly or some endocrine disorder like that,” go ahead and measure insulin antibodies, he said.

A potentially more productive approach is to measure both free and total insulin levels in a patient. A discrepancy in the results suggests that there's something binding up the insulin.

Be sure to use a clinical laboratory that knows how to measure free insulin, Dr. Hirsch cautioned. After the blood is drawn, several preparatory steps must be completed within 30 seconds before the sample is sent out to be assayed. “Not all labs do that,” he said.

▸ Celiac disease. This occurs in approximately 7%-8% of childhood-onset type 1 diabetes patients and can lead to brittle diabetes. Screening for celiac disease typically focuses on identifying the transglutaminase IgA antibody, but 5% of the population lacks IgA, Dr. Hirsch warned. Before looking for the antibody, measure IgA levels to make sure they're normal. If IgA is absent, the antibody test “becomes worthless,” he said.

▸ Gastroparesis. Considering this “makes sense. If you have a mismatching of food and insulin, it makes the diabetes that much more difficult to control,” Dr. Hirsch said.

▸ Injection and pump sites. Check these, because lipodystrophy may cause severe insulin resistance. Although the use of protease inhibitors is the most common cause of lipodystrophy in the general population, congenital lipodystrophy is the most frequent cause in Dr. Hirsch's patient population. Patients with congenital lipodystrophy typically get referred to him from a lipid clinic for severe hypertriglyceridemia and insulin resistance.

▸ Timing of insulin injections. Bad timing in relation to food intake probably is the most common mistake made in insulin therapy, he said. Problems with timing of injections usually won't disrupt a patient's life to the point of frequent hospitalizations, but they “certainly will cause glycemic instability,” he said.

▸ Psychological or psychiatric problems. These can interfere with diabetes management, so a mental health evaluation is a key part of the work-up for most patients with brittle diabetes. Frequent episodes of diabetic ketoacidosis usually result from patients' not taking their insulin, Dr. Hirsch said. Severe depression can make it difficult for a patient to manage diabetes.

The mental health evaluation “is best performed by someone knowledgeable about diabetes, particularly with experience in eating disorders, because that's the big issue, especially in adolescents,” he said.

SAN FRANCISCO — Diabetes patients whose lives are frequently disrupted by glycemic episodes requiring hospitalization need a multifaceted work-up to identify the cause of their brittle diabetes, Dr. Irl B. Hirsch said.

Compared with diabetes patients who have relatively stable and well-controlled glucose levels, patients with brittle diabetes tend to be younger and are more likely to be female, Dr. Hirsch said at a meeting sponsored by the American Diabetes Association. They also tend to have higher hemoglobin A_1c values and to use more insulin, and are more likely to have psychiatric disorders or psychosocial problems, including family disruption, adolescent crises, and personality disturbances, said Dr. Hirsch, professor of medicine at the University of Washington, Seattle.

Both physical and psychiatric problems can result in brittle diabetes. Look for the following to identify contributing factors, he said:

▸ Counterregulatory hormone excess. This can cause insulin resistance and make diabetes very difficult to control. Cushing's disease, acromegaly, pheochromocytoma, and glucagonoma are rare causes of insulin resistance to consider in the work-up of brittle diabetes.

▸ Insulin antibodies. These are worth measuring but have a low yield in these patients. “Unless the physical exam suggests a diagnosis of Cushing's disease or acromegaly or some endocrine disorder like that,” go ahead and measure insulin antibodies, he said.

A potentially more productive approach is to measure both free and total insulin levels in a patient. A discrepancy in the results suggests that there's something binding up the insulin.

Be sure to use a clinical laboratory that knows how to measure free insulin, Dr. Hirsch cautioned. After the blood is drawn, several preparatory steps must be completed within 30 seconds before the sample is sent out to be assayed. “Not all labs do that,” he said.

▸ Celiac disease. This occurs in approximately 7%-8% of childhood-onset type 1 diabetes patients and can lead to brittle diabetes. Screening for celiac disease typically focuses on identifying the transglutaminase IgA antibody, but 5% of the population lacks IgA, Dr. Hirsch warned. Before looking for the antibody, measure IgA levels to make sure they're normal. If IgA is absent, the antibody test “becomes worthless,” he said.

▸ Gastroparesis. Considering this “makes sense. If you have a mismatching of food and insulin, it makes the diabetes that much more difficult to control,” Dr. Hirsch said.

▸ Injection and pump sites. Check these, because lipodystrophy may cause severe insulin resistance. Although the use of protease inhibitors is the most common cause of lipodystrophy in the general population, congenital lipodystrophy is the most frequent cause in Dr. Hirsch's patient population. Patients with congenital lipodystrophy typically get referred to him from a lipid clinic for severe hypertriglyceridemia and insulin resistance.

▸ Timing of insulin injections. Bad timing in relation to food intake probably is the most common mistake made in insulin therapy, he said. Problems with timing of injections usually won't disrupt a patient's life to the point of frequent hospitalizations, but they “certainly will cause glycemic instability,” he said.

▸ Psychological or psychiatric problems. These can interfere with diabetes management, so a mental health evaluation is a key part of the work-up for most patients with brittle diabetes. Frequent episodes of diabetic ketoacidosis usually result from patients' not taking their insulin, Dr. Hirsch said. Severe depression can make it difficult for a patient to manage diabetes.

The mental health evaluation “is best performed by someone knowledgeable about diabetes, particularly with experience in eating disorders, because that's the big issue, especially in adolescents,” he said.

SAN FRANCISCO — Diabetes patients whose lives are frequently disrupted by glycemic episodes requiring hospitalization need a multifaceted work-up to identify the cause of their brittle diabetes, Dr. Irl B. Hirsch said.

Compared with diabetes patients who have relatively stable and well-controlled glucose levels, patients with brittle diabetes tend to be younger and are more likely to be female, Dr. Hirsch said at a meeting sponsored by the American Diabetes Association. They also tend to have higher hemoglobin A_1c values and to use more insulin, and are more likely to have psychiatric disorders or psychosocial problems, including family disruption, adolescent crises, and personality disturbances, said Dr. Hirsch, professor of medicine at the University of Washington, Seattle.

Both physical and psychiatric problems can result in brittle diabetes. Look for the following to identify contributing factors, he said:

▸ Counterregulatory hormone excess. This can cause insulin resistance and make diabetes very difficult to control. Cushing's disease, acromegaly, pheochromocytoma, and glucagonoma are rare causes of insulin resistance to consider in the work-up of brittle diabetes.

▸ Insulin antibodies. These are worth measuring but have a low yield in these patients. “Unless the physical exam suggests a diagnosis of Cushing's disease or acromegaly or some endocrine disorder like that,” go ahead and measure insulin antibodies, he said.

A potentially more productive approach is to measure both free and total insulin levels in a patient. A discrepancy in the results suggests that there's something binding up the insulin.

Be sure to use a clinical laboratory that knows how to measure free insulin, Dr. Hirsch cautioned. After the blood is drawn, several preparatory steps must be completed within 30 seconds before the sample is sent out to be assayed. “Not all labs do that,” he said.

▸ Celiac disease. This occurs in approximately 7%-8% of childhood-onset type 1 diabetes patients and can lead to brittle diabetes. Screening for celiac disease typically focuses on identifying the transglutaminase IgA antibody, but 5% of the population lacks IgA, Dr. Hirsch warned. Before looking for the antibody, measure IgA levels to make sure they're normal. If IgA is absent, the antibody test “becomes worthless,” he said.

▸ Gastroparesis. Considering this “makes sense. If you have a mismatching of food and insulin, it makes the diabetes that much more difficult to control,” Dr. Hirsch said.

▸ Injection and pump sites. Check these, because lipodystrophy may cause severe insulin resistance. Although the use of protease inhibitors is the most common cause of lipodystrophy in the general population, congenital lipodystrophy is the most frequent cause in Dr. Hirsch's patient population. Patients with congenital lipodystrophy typically get referred to him from a lipid clinic for severe hypertriglyceridemia and insulin resistance.

▸ Timing of insulin injections. Bad timing in relation to food intake probably is the most common mistake made in insulin therapy, he said. Problems with timing of injections usually won't disrupt a patient's life to the point of frequent hospitalizations, but they “certainly will cause glycemic instability,” he said.

▸ Psychological or psychiatric problems. These can interfere with diabetes management, so a mental health evaluation is a key part of the work-up for most patients with brittle diabetes. Frequent episodes of diabetic ketoacidosis usually result from patients' not taking their insulin, Dr. Hirsch said. Severe depression can make it difficult for a patient to manage diabetes.

The mental health evaluation “is best performed by someone knowledgeable about diabetes, particularly with experience in eating disorders, because that's the big issue, especially in adolescents,” he said.