Sara Freeman

ROME — Eye and renal complications were no less frequent in older, difficult-to-treat patients with type 2 diabetes who were randomized to intensive rather than standard glucose control in the Veterans Affairs Diabetes Trial.

Furthermore, there was no difference between the two treatment approaches in the development of any new diabetic neuropathy, according to results presented at the annual meeting of the European Association for the Study of Diabetes.

These latest findings from the 7.5-year trial followed on from the primary end point data presented at the American Diabetes Association meeting in May 2008 just 8 days after the trial had ended. The latter showed that there was no difference between intensive and standard lowering of glycated hemoglobin A_1c (HbA_1c) in terms of overall cardiovascular protection.

The two cochairs of the Veterans Affairs Diabetes Trial (VADT)—Dr. William C. Duckworth, of the Veterans Affairs Medical Center in Phoenix, and Dr. Carlos Abraira of the Miami Veterans Affairs Medical Center—discussed the new secondary outcomes data.

Compared with standard glucose control, the intensive approach was associated with nonsignificant differences in the number of new eye procedures, which included cataract surgery (10.2% vs. 11.6%), photocoagulation (8.8% vs. 7%), and vitrectomy (3% vs. 3.3%).

“There were no differences in proliferative diabetic retinopathy or macular edema,” said Dr. Abraira, and “no difference in new onset of retinopathy.”

However, he noted that slightly more patients in the intensive therapy arm, compared with the standard control arm, required laser eye surgery overall (8.3/100 patients per year vs. 7.5/100 patients per year). In addition, the percentage of patients who developed retinopathy early—defined as a two-step increase in the Early Treatment Diabetic Retinopathy Study scale—was marginally higher in the less intensively treated patients than in the intensively treated controls (22.1% vs. 17%).

With regards to kidney complications, Dr. Abraira said that the decline in renal function was similar in both groups of patients, and severe renal disease—evidenced by doubling of serum creatinine, serum creatinine greater than 3 mg/dL, or end-stage renal disease—was “unaffected by glucose control.”

Having a prior cardiovascular event, being a smoker, or having microalbuminuria or retinopathy at baseline were expected predictors of progression to macroalbuminuria, said Dr. Abraira. While progression from normal or microalbuminuria to overt proteinuria did not differ between standard and intensive glucose control, there was a significant decrease in progression from normal to micro- or macroalbuminuria in the intensively treated patients (31% vs. 38% for standard control, P = .02).

Dr. Abraira also showed that the percentage of patients in the standard vs. intensive treatment arms with neuropathy was similar: any neuropathy, 43.8% vs. 43.5%; mononeuropathy, 4% vs. 4.7%; and peripheral neuropathy, 40% vs. 38.4%. However, there was a slightly higher percentage of patients in the intensive glucose control arm that developed autonomic neuropathy (8.2% vs. 5.2%) though this was not significantly different.

Dr. Duckworth pointed out that the 1,791 patients enrolled in the VADT were a difficult-to-treat group, with a mean age of 60 years and a median duration of diabetes of 11.5 years. The mean HbA_1c at the start of the trial was 9.4% and fell to a median of 8.4% in the standard glucose control arm and 6.9% in the intensive glucose control arm.

Importantly, patients in the trial achieved good blood pressure control and “almost ideal” levels of LDL cholesterol and other lipids, Dr. Duckworth said. This suggests that in the presence of optimal risk factor management, it does little to add intensive glucose control in terms of the overall benefits that can be achieved.

'There were no differences in proliferative diabetic retinopathy or macular edema.' DR. ABRAIRA

ROME — Eye and renal complications were no less frequent in older, difficult-to-treat patients with type 2 diabetes who were randomized to intensive rather than standard glucose control in the Veterans Affairs Diabetes Trial.

Furthermore, there was no difference between the two treatment approaches in the development of any new diabetic neuropathy, according to results presented at the annual meeting of the European Association for the Study of Diabetes.

These latest findings from the 7.5-year trial followed on from the primary end point data presented at the American Diabetes Association meeting in May 2008 just 8 days after the trial had ended. The latter showed that there was no difference between intensive and standard lowering of glycated hemoglobin A_1c (HbA_1c) in terms of overall cardiovascular protection.

The two cochairs of the Veterans Affairs Diabetes Trial (VADT)—Dr. William C. Duckworth, of the Veterans Affairs Medical Center in Phoenix, and Dr. Carlos Abraira of the Miami Veterans Affairs Medical Center—discussed the new secondary outcomes data.

Compared with standard glucose control, the intensive approach was associated with nonsignificant differences in the number of new eye procedures, which included cataract surgery (10.2% vs. 11.6%), photocoagulation (8.8% vs. 7%), and vitrectomy (3% vs. 3.3%).

“There were no differences in proliferative diabetic retinopathy or macular edema,” said Dr. Abraira, and “no difference in new onset of retinopathy.”

However, he noted that slightly more patients in the intensive therapy arm, compared with the standard control arm, required laser eye surgery overall (8.3/100 patients per year vs. 7.5/100 patients per year). In addition, the percentage of patients who developed retinopathy early—defined as a two-step increase in the Early Treatment Diabetic Retinopathy Study scale—was marginally higher in the less intensively treated patients than in the intensively treated controls (22.1% vs. 17%).

With regards to kidney complications, Dr. Abraira said that the decline in renal function was similar in both groups of patients, and severe renal disease—evidenced by doubling of serum creatinine, serum creatinine greater than 3 mg/dL, or end-stage renal disease—was “unaffected by glucose control.”

Having a prior cardiovascular event, being a smoker, or having microalbuminuria or retinopathy at baseline were expected predictors of progression to macroalbuminuria, said Dr. Abraira. While progression from normal or microalbuminuria to overt proteinuria did not differ between standard and intensive glucose control, there was a significant decrease in progression from normal to micro- or macroalbuminuria in the intensively treated patients (31% vs. 38% for standard control, P = .02).

Dr. Abraira also showed that the percentage of patients in the standard vs. intensive treatment arms with neuropathy was similar: any neuropathy, 43.8% vs. 43.5%; mononeuropathy, 4% vs. 4.7%; and peripheral neuropathy, 40% vs. 38.4%. However, there was a slightly higher percentage of patients in the intensive glucose control arm that developed autonomic neuropathy (8.2% vs. 5.2%) though this was not significantly different.

Dr. Duckworth pointed out that the 1,791 patients enrolled in the VADT were a difficult-to-treat group, with a mean age of 60 years and a median duration of diabetes of 11.5 years. The mean HbA_1c at the start of the trial was 9.4% and fell to a median of 8.4% in the standard glucose control arm and 6.9% in the intensive glucose control arm.

Importantly, patients in the trial achieved good blood pressure control and “almost ideal” levels of LDL cholesterol and other lipids, Dr. Duckworth said. This suggests that in the presence of optimal risk factor management, it does little to add intensive glucose control in terms of the overall benefits that can be achieved.

'There were no differences in proliferative diabetic retinopathy or macular edema.' DR. ABRAIRA

ROME — Eye and renal complications were no less frequent in older, difficult-to-treat patients with type 2 diabetes who were randomized to intensive rather than standard glucose control in the Veterans Affairs Diabetes Trial.

Furthermore, there was no difference between the two treatment approaches in the development of any new diabetic neuropathy, according to results presented at the annual meeting of the European Association for the Study of Diabetes.

These latest findings from the 7.5-year trial followed on from the primary end point data presented at the American Diabetes Association meeting in May 2008 just 8 days after the trial had ended. The latter showed that there was no difference between intensive and standard lowering of glycated hemoglobin A_1c (HbA_1c) in terms of overall cardiovascular protection.

The two cochairs of the Veterans Affairs Diabetes Trial (VADT)—Dr. William C. Duckworth, of the Veterans Affairs Medical Center in Phoenix, and Dr. Carlos Abraira of the Miami Veterans Affairs Medical Center—discussed the new secondary outcomes data.

Compared with standard glucose control, the intensive approach was associated with nonsignificant differences in the number of new eye procedures, which included cataract surgery (10.2% vs. 11.6%), photocoagulation (8.8% vs. 7%), and vitrectomy (3% vs. 3.3%).

“There were no differences in proliferative diabetic retinopathy or macular edema,” said Dr. Abraira, and “no difference in new onset of retinopathy.”

However, he noted that slightly more patients in the intensive therapy arm, compared with the standard control arm, required laser eye surgery overall (8.3/100 patients per year vs. 7.5/100 patients per year). In addition, the percentage of patients who developed retinopathy early—defined as a two-step increase in the Early Treatment Diabetic Retinopathy Study scale—was marginally higher in the less intensively treated patients than in the intensively treated controls (22.1% vs. 17%).

With regards to kidney complications, Dr. Abraira said that the decline in renal function was similar in both groups of patients, and severe renal disease—evidenced by doubling of serum creatinine, serum creatinine greater than 3 mg/dL, or end-stage renal disease—was “unaffected by glucose control.”

Having a prior cardiovascular event, being a smoker, or having microalbuminuria or retinopathy at baseline were expected predictors of progression to macroalbuminuria, said Dr. Abraira. While progression from normal or microalbuminuria to overt proteinuria did not differ between standard and intensive glucose control, there was a significant decrease in progression from normal to micro- or macroalbuminuria in the intensively treated patients (31% vs. 38% for standard control, P = .02).

Dr. Abraira also showed that the percentage of patients in the standard vs. intensive treatment arms with neuropathy was similar: any neuropathy, 43.8% vs. 43.5%; mononeuropathy, 4% vs. 4.7%; and peripheral neuropathy, 40% vs. 38.4%. However, there was a slightly higher percentage of patients in the intensive glucose control arm that developed autonomic neuropathy (8.2% vs. 5.2%) though this was not significantly different.

Dr. Duckworth pointed out that the 1,791 patients enrolled in the VADT were a difficult-to-treat group, with a mean age of 60 years and a median duration of diabetes of 11.5 years. The mean HbA_1c at the start of the trial was 9.4% and fell to a median of 8.4% in the standard glucose control arm and 6.9% in the intensive glucose control arm.

Importantly, patients in the trial achieved good blood pressure control and “almost ideal” levels of LDL cholesterol and other lipids, Dr. Duckworth said. This suggests that in the presence of optimal risk factor management, it does little to add intensive glucose control in terms of the overall benefits that can be achieved.

'There were no differences in proliferative diabetic retinopathy or macular edema.' DR. ABRAIRA

ROME — For every 1% rise in baseline hemoglobin A_1c, the risk that a patient with type 2 diabetes would later develop coronary heart disease increased by 11% in a large observational study.

Furthermore, there was a 25% decreased risk of cardiovascular disease (CVD) in patients with a median hemoglobin A_1c (HbA_1c) of 6.7%, compared with 7.9%, Dr. Katarina Eeg-Olofsson of the University of Gothenburg (Sweden) reported at the annual meeting of the European Association for the Study of Diabetes.

The study looked at the effect of glycemic control on the development of cardiovascular complications and mortality in 16,701 individuals with type 2 dia-betes who were logged in the Swedish National Diabetes Register and who had not yet experienced a cardio- or cerebrovascular event.

“The role of improved glycemic control in reducing cardiovascular risk is currently under debate,” Dr. Eeg-Olofsson said.

She added that recently presented data from three large randomized trials looking at this question—the Veterans Affairs Diabetes Trial, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, and Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation (ADVANCE)—had failed to provide conclusive evidence that lower HbA_1c levels are protective against the development of cardiovascular events.

“The aim of our study was to examine the longitudinal association between glycemic control and cardiovascular complications,” said Dr. Eeg-Olofsson. She noted that the patients studied were unselected and that the definition of type 2 diabetes used was treatment by diet or with oral hypoglycemic agents, alone or with insulin, and patient age of 40 years or older at diagnosis with diabetes.

The current study was unable to distinguish among specific drugs that were used, however, because this information was not initially logged in the database. More detailed analysis is perhaps possible in the future because more detailed information is now being collected.

The progress of the patients identified within the register as having type 2 diabetes was monitored for 6 years, linking HbA_1c levels to the number of first-incident fatal or nonfatal coronary heart disease (CHD) events, strokes, and CVD cases, and deaths due to any cause as defined by ICD codes. Within this time frame, there were 2,128 fatal or nonfatal coronary heart disease events, 1,289 fatal or nonfatal stroke events, 3,122 fatal or nonfatal CVD events, and 1,570 all-cause mortality events.

The mean age of the patients studied was 64 years and the mean duration of diabetes was 8 years. HbA_1c levels were an average of 7.6% at baseline. Systolic blood pressure was an average of 148 mm Hg, half of all patients were using antihypertensive medications, and 11% were taking lipid-lowering drugs.

Dr. Eeg-Olofsson and associates found that 13.9% of patients were current smokers, and just over one-fifth (22%) already had microalbuminuria. One-quarter of patients were treated by diet alone, 12% with an oral hypoglycemic agent alone, 37% with an oral hypoglycemic agent plus insulin, and a further 28% were using only insulin.

For 4,259 patients, the HbA_1c levels were known both at baseline and at follow-up, and the baseline characteristics of this subgroup of patients were similar to the entire cohort.

Event rates for the first-incident fatal and nonfatal CHD, stroke, and CVD, and all-cause mortality were 25.7, 15.5, 38.6, and 18.4 per 1,000 person-years.

The unadjusted hazard ratios (HRs) for these events were 1.17, 1.12, 1.14, and 1.17.

The HRs remained significant, even after adjustment for a variety of factors—age, sex, diabetes duration, body mass index, smoking, systolic blood pressure, microalbuminuria, and type of hypoglycemic, antihypertensive, or lipid-lowering therapy being used.

The hazard ratios for CHD, stroke, and CVD rose significantly for every 1% rise in baseline HbA_1c, said Dr. Eeg-Olofsson.

For example, the hazard ratio for CHD at a baseline HbA_1c of 6.0%–6.9% was 1.0 and for an HbA_1c level of 7.0%–7.9% was 1.26. At an HbA_1c of 8.0%–8.9% and 9.0%–9.9%, the hazard ratio for CHD had risen to 1.42, and when HbA_1c levels exceeded 10.0%, the hazard ratio was 1.62.

ROME — For every 1% rise in baseline hemoglobin A_1c, the risk that a patient with type 2 diabetes would later develop coronary heart disease increased by 11% in a large observational study.

Furthermore, there was a 25% decreased risk of cardiovascular disease (CVD) in patients with a median hemoglobin A_1c (HbA_1c) of 6.7%, compared with 7.9%, Dr. Katarina Eeg-Olofsson of the University of Gothenburg (Sweden) reported at the annual meeting of the European Association for the Study of Diabetes.

The study looked at the effect of glycemic control on the development of cardiovascular complications and mortality in 16,701 individuals with type 2 dia-betes who were logged in the Swedish National Diabetes Register and who had not yet experienced a cardio- or cerebrovascular event.

“The role of improved glycemic control in reducing cardiovascular risk is currently under debate,” Dr. Eeg-Olofsson said.

She added that recently presented data from three large randomized trials looking at this question—the Veterans Affairs Diabetes Trial, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, and Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation (ADVANCE)—had failed to provide conclusive evidence that lower HbA_1c levels are protective against the development of cardiovascular events.

“The aim of our study was to examine the longitudinal association between glycemic control and cardiovascular complications,” said Dr. Eeg-Olofsson. She noted that the patients studied were unselected and that the definition of type 2 diabetes used was treatment by diet or with oral hypoglycemic agents, alone or with insulin, and patient age of 40 years or older at diagnosis with diabetes.

The current study was unable to distinguish among specific drugs that were used, however, because this information was not initially logged in the database. More detailed analysis is perhaps possible in the future because more detailed information is now being collected.

The progress of the patients identified within the register as having type 2 diabetes was monitored for 6 years, linking HbA_1c levels to the number of first-incident fatal or nonfatal coronary heart disease (CHD) events, strokes, and CVD cases, and deaths due to any cause as defined by ICD codes. Within this time frame, there were 2,128 fatal or nonfatal coronary heart disease events, 1,289 fatal or nonfatal stroke events, 3,122 fatal or nonfatal CVD events, and 1,570 all-cause mortality events.

The mean age of the patients studied was 64 years and the mean duration of diabetes was 8 years. HbA_1c levels were an average of 7.6% at baseline. Systolic blood pressure was an average of 148 mm Hg, half of all patients were using antihypertensive medications, and 11% were taking lipid-lowering drugs.

Dr. Eeg-Olofsson and associates found that 13.9% of patients were current smokers, and just over one-fifth (22%) already had microalbuminuria. One-quarter of patients were treated by diet alone, 12% with an oral hypoglycemic agent alone, 37% with an oral hypoglycemic agent plus insulin, and a further 28% were using only insulin.

For 4,259 patients, the HbA_1c levels were known both at baseline and at follow-up, and the baseline characteristics of this subgroup of patients were similar to the entire cohort.

Event rates for the first-incident fatal and nonfatal CHD, stroke, and CVD, and all-cause mortality were 25.7, 15.5, 38.6, and 18.4 per 1,000 person-years.

The unadjusted hazard ratios (HRs) for these events were 1.17, 1.12, 1.14, and 1.17.

The HRs remained significant, even after adjustment for a variety of factors—age, sex, diabetes duration, body mass index, smoking, systolic blood pressure, microalbuminuria, and type of hypoglycemic, antihypertensive, or lipid-lowering therapy being used.

The hazard ratios for CHD, stroke, and CVD rose significantly for every 1% rise in baseline HbA_1c, said Dr. Eeg-Olofsson.

For example, the hazard ratio for CHD at a baseline HbA_1c of 6.0%–6.9% was 1.0 and for an HbA_1c level of 7.0%–7.9% was 1.26. At an HbA_1c of 8.0%–8.9% and 9.0%–9.9%, the hazard ratio for CHD had risen to 1.42, and when HbA_1c levels exceeded 10.0%, the hazard ratio was 1.62.

ROME — For every 1% rise in baseline hemoglobin A_1c, the risk that a patient with type 2 diabetes would later develop coronary heart disease increased by 11% in a large observational study.

Furthermore, there was a 25% decreased risk of cardiovascular disease (CVD) in patients with a median hemoglobin A_1c (HbA_1c) of 6.7%, compared with 7.9%, Dr. Katarina Eeg-Olofsson of the University of Gothenburg (Sweden) reported at the annual meeting of the European Association for the Study of Diabetes.

The study looked at the effect of glycemic control on the development of cardiovascular complications and mortality in 16,701 individuals with type 2 dia-betes who were logged in the Swedish National Diabetes Register and who had not yet experienced a cardio- or cerebrovascular event.

“The role of improved glycemic control in reducing cardiovascular risk is currently under debate,” Dr. Eeg-Olofsson said.

She added that recently presented data from three large randomized trials looking at this question—the Veterans Affairs Diabetes Trial, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, and Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation (ADVANCE)—had failed to provide conclusive evidence that lower HbA_1c levels are protective against the development of cardiovascular events.

“The aim of our study was to examine the longitudinal association between glycemic control and cardiovascular complications,” said Dr. Eeg-Olofsson. She noted that the patients studied were unselected and that the definition of type 2 diabetes used was treatment by diet or with oral hypoglycemic agents, alone or with insulin, and patient age of 40 years or older at diagnosis with diabetes.

The current study was unable to distinguish among specific drugs that were used, however, because this information was not initially logged in the database. More detailed analysis is perhaps possible in the future because more detailed information is now being collected.

The progress of the patients identified within the register as having type 2 diabetes was monitored for 6 years, linking HbA_1c levels to the number of first-incident fatal or nonfatal coronary heart disease (CHD) events, strokes, and CVD cases, and deaths due to any cause as defined by ICD codes. Within this time frame, there were 2,128 fatal or nonfatal coronary heart disease events, 1,289 fatal or nonfatal stroke events, 3,122 fatal or nonfatal CVD events, and 1,570 all-cause mortality events.

The mean age of the patients studied was 64 years and the mean duration of diabetes was 8 years. HbA_1c levels were an average of 7.6% at baseline. Systolic blood pressure was an average of 148 mm Hg, half of all patients were using antihypertensive medications, and 11% were taking lipid-lowering drugs.

Dr. Eeg-Olofsson and associates found that 13.9% of patients were current smokers, and just over one-fifth (22%) already had microalbuminuria. One-quarter of patients were treated by diet alone, 12% with an oral hypoglycemic agent alone, 37% with an oral hypoglycemic agent plus insulin, and a further 28% were using only insulin.

For 4,259 patients, the HbA_1c levels were known both at baseline and at follow-up, and the baseline characteristics of this subgroup of patients were similar to the entire cohort.

Event rates for the first-incident fatal and nonfatal CHD, stroke, and CVD, and all-cause mortality were 25.7, 15.5, 38.6, and 18.4 per 1,000 person-years.

The unadjusted hazard ratios (HRs) for these events were 1.17, 1.12, 1.14, and 1.17.

The HRs remained significant, even after adjustment for a variety of factors—age, sex, diabetes duration, body mass index, smoking, systolic blood pressure, microalbuminuria, and type of hypoglycemic, antihypertensive, or lipid-lowering therapy being used.

The hazard ratios for CHD, stroke, and CVD rose significantly for every 1% rise in baseline HbA_1c, said Dr. Eeg-Olofsson.

For example, the hazard ratio for CHD at a baseline HbA_1c of 6.0%–6.9% was 1.0 and for an HbA_1c level of 7.0%–7.9% was 1.26. At an HbA_1c of 8.0%–8.9% and 9.0%–9.9%, the hazard ratio for CHD had risen to 1.42, and when HbA_1c levels exceeded 10.0%, the hazard ratio was 1.62.

GLASGOW, SCOTLAND — Data from the intensive glucose-lowering arm of the ADVANCE trial will be released earlier than expected, during the American Diabetes Association's 68th Annual Scientific Sessions in June.

Study investigator Dr. Neil R. Poulter of the National Heart and Lung Institute at Imperial College London, made the announcement during the annual professional conference of Diabetes U.K. The results had been planned for release at the European Association for the Study of Diabetes in September.

Only data from the intensive glucose-lowering arm of ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation) will be released, and these data follow as a direct result of the recent closure of the intensive glucose-lowering arm of the ACCORD (Action to Control Cardiovascular Risk in Diabetes) trial.

Dr. Poulter said that the press statement issued by the data safety monitoring board for the ADVANCE trial still stood and that there was no evidence of increased cardiovascular mortality in patients treated with an intensive glucose-lowering regimen as there had been in ACCORD. At the time this statement was released, the data analyses were 99% complete, so there is unlikely to be any change, he said.

“In terms of statistical robustness, ADVANCE is more powerful; it's got about twice as much patient years of follow-up as ACCORD does,” Dr. Poulter said. Commenting on the differences between the trials, and what they may mean, he added, “The difference in insulin use was really quite dramatic between the two [trials].”

Dr. Poulter is a member of the management committee and director of the north European region of the ADVANCE study.

GLASGOW, SCOTLAND — Data from the intensive glucose-lowering arm of the ADVANCE trial will be released earlier than expected, during the American Diabetes Association's 68th Annual Scientific Sessions in June.

Study investigator Dr. Neil R. Poulter of the National Heart and Lung Institute at Imperial College London, made the announcement during the annual professional conference of Diabetes U.K. The results had been planned for release at the European Association for the Study of Diabetes in September.

Only data from the intensive glucose-lowering arm of ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation) will be released, and these data follow as a direct result of the recent closure of the intensive glucose-lowering arm of the ACCORD (Action to Control Cardiovascular Risk in Diabetes) trial.

Dr. Poulter said that the press statement issued by the data safety monitoring board for the ADVANCE trial still stood and that there was no evidence of increased cardiovascular mortality in patients treated with an intensive glucose-lowering regimen as there had been in ACCORD. At the time this statement was released, the data analyses were 99% complete, so there is unlikely to be any change, he said.

“In terms of statistical robustness, ADVANCE is more powerful; it's got about twice as much patient years of follow-up as ACCORD does,” Dr. Poulter said. Commenting on the differences between the trials, and what they may mean, he added, “The difference in insulin use was really quite dramatic between the two [trials].”

Dr. Poulter is a member of the management committee and director of the north European region of the ADVANCE study.

GLASGOW, SCOTLAND — Data from the intensive glucose-lowering arm of the ADVANCE trial will be released earlier than expected, during the American Diabetes Association's 68th Annual Scientific Sessions in June.

Study investigator Dr. Neil R. Poulter of the National Heart and Lung Institute at Imperial College London, made the announcement during the annual professional conference of Diabetes U.K. The results had been planned for release at the European Association for the Study of Diabetes in September.

Only data from the intensive glucose-lowering arm of ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation) will be released, and these data follow as a direct result of the recent closure of the intensive glucose-lowering arm of the ACCORD (Action to Control Cardiovascular Risk in Diabetes) trial.

Dr. Poulter said that the press statement issued by the data safety monitoring board for the ADVANCE trial still stood and that there was no evidence of increased cardiovascular mortality in patients treated with an intensive glucose-lowering regimen as there had been in ACCORD. At the time this statement was released, the data analyses were 99% complete, so there is unlikely to be any change, he said.

“In terms of statistical robustness, ADVANCE is more powerful; it's got about twice as much patient years of follow-up as ACCORD does,” Dr. Poulter said. Commenting on the differences between the trials, and what they may mean, he added, “The difference in insulin use was really quite dramatic between the two [trials].”

Dr. Poulter is a member of the management committee and director of the north European region of the ADVANCE study.

GLASGOW, SCOTLAND — Fasting plasma glucose levels in pregnant women are associated with the development of preeclampsia, according to the latest data to be released from the Hyperglycemia and Adverse Pregnancy Outcome study.

Principal investigator Dr. Boyd L. Metzger gave a sneak preview of the latest results from the 7-year multinational trial, known as HAPO, during the annual professional conference of Diabetes U.K.

The new findings were on the secondary outcomes of the study. All three measures of glucose control—FPG at 1 hour and at 2 hours and the oral glucose tolerance test—were associated with preeclampsia. After adjusting for potentially confounding factors, the data showed that the odds of developing preeclampsia were 1.2–1.28 times higher for every 1-point standard deviation increase in maternal glucose concentration.

The key findings are the same as those previously released “in that outcomes are very strongly related to the mother's blood sugar level during her pregnancy and that the effects seem to occur at levels that are lower than we've been diagnosing and treating gestational diabetes at in the past,” Dr. Metzger of Northwestern University, Chicago, said in an interview.

But the new data show how the effects of maternal glucose during pregnancy appear to extend beyond an increased chance of having a high-birth-weight baby, fetal hyperinsulinemia, and delivery by cesarean section, and into outcomes such as preeclampsia. However, there was no specific number at which the risk of these adverse outcomes started to increase, he said. Although adverse outcomes appeared to occur at fasting plasma glucose (FPG) levels as low as 4.0–4.9 mg/dL, the data did not seem to agree as far as what level was needed to experience one adverse outcome over another.

Four outcomes were designated as “primary” at the start of the study, and were reported last June. These included the frequency of having a high-birth-weight baby, a first cesarean section, neonatal hypoglycemia, and fetal hyperinsulinemia. The last was assessed by measuring the level of C-peptide in the cord blood; a high level was defined as greater than the 90th percentile of what would be expected. Of these, Dr. Metzger said that there were strong associations between maternal glucose levels and both birth weight and fetal hyperinsulinemia; there were milder associations with having a first cesarean section and neonatal hypoglycemia.

“The relationship between maternal blood glucose and the outcome of [a woman's] pregnancy appears to be continuous across the whole spectrum of adverse outcomes, so how much risk is too much risk?” commented Dr. Metzger. This highlighted just one of the problems of applying the HAPO data as they stand to clinical practice. “If the clinical assessment of this information is that we should be intervening to prevent these risks, then we are going to have to lower the values that we have traditionally been using” to diagnose women with gestational diabetes—which will mean a that lot more women will be diagnosed with the disorder, he said.

This was the first time the HAPO findings have been discussed at a professional diabetes meeting outside the United States since their presentation at the American Diabetes Association 67th Annual Scientific Sessions last June.

The investigators are just starting to decipher all the data gathered from the more than 23,000 women who participated in the observational study, and it could be some time before there is an agreement on what these results mean for the diagnosis and management of gestational diabetes. There are also further analyses to be performed, such as the effect of maternal glucose on insulin sensitivity and triglyceride levels.

The first steps toward achieving a consensus on the clinical significance of the HAPO data will be made during a stand-alone meeting to be held in Pasadena, Calif. in a few months.

The effects seem to occur at levels lower than we've been diagnosing gestational diabetes in the past. DR. METZGER

GLASGOW, SCOTLAND — Fasting plasma glucose levels in pregnant women are associated with the development of preeclampsia, according to the latest data to be released from the Hyperglycemia and Adverse Pregnancy Outcome study.

Principal investigator Dr. Boyd L. Metzger gave a sneak preview of the latest results from the 7-year multinational trial, known as HAPO, during the annual professional conference of Diabetes U.K.

The new findings were on the secondary outcomes of the study. All three measures of glucose control—FPG at 1 hour and at 2 hours and the oral glucose tolerance test—were associated with preeclampsia. After adjusting for potentially confounding factors, the data showed that the odds of developing preeclampsia were 1.2–1.28 times higher for every 1-point standard deviation increase in maternal glucose concentration.

The key findings are the same as those previously released “in that outcomes are very strongly related to the mother's blood sugar level during her pregnancy and that the effects seem to occur at levels that are lower than we've been diagnosing and treating gestational diabetes at in the past,” Dr. Metzger of Northwestern University, Chicago, said in an interview.

But the new data show how the effects of maternal glucose during pregnancy appear to extend beyond an increased chance of having a high-birth-weight baby, fetal hyperinsulinemia, and delivery by cesarean section, and into outcomes such as preeclampsia. However, there was no specific number at which the risk of these adverse outcomes started to increase, he said. Although adverse outcomes appeared to occur at fasting plasma glucose (FPG) levels as low as 4.0–4.9 mg/dL, the data did not seem to agree as far as what level was needed to experience one adverse outcome over another.

Four outcomes were designated as “primary” at the start of the study, and were reported last June. These included the frequency of having a high-birth-weight baby, a first cesarean section, neonatal hypoglycemia, and fetal hyperinsulinemia. The last was assessed by measuring the level of C-peptide in the cord blood; a high level was defined as greater than the 90th percentile of what would be expected. Of these, Dr. Metzger said that there were strong associations between maternal glucose levels and both birth weight and fetal hyperinsulinemia; there were milder associations with having a first cesarean section and neonatal hypoglycemia.

“The relationship between maternal blood glucose and the outcome of [a woman's] pregnancy appears to be continuous across the whole spectrum of adverse outcomes, so how much risk is too much risk?” commented Dr. Metzger. This highlighted just one of the problems of applying the HAPO data as they stand to clinical practice. “If the clinical assessment of this information is that we should be intervening to prevent these risks, then we are going to have to lower the values that we have traditionally been using” to diagnose women with gestational diabetes—which will mean a that lot more women will be diagnosed with the disorder, he said.

This was the first time the HAPO findings have been discussed at a professional diabetes meeting outside the United States since their presentation at the American Diabetes Association 67th Annual Scientific Sessions last June.

The investigators are just starting to decipher all the data gathered from the more than 23,000 women who participated in the observational study, and it could be some time before there is an agreement on what these results mean for the diagnosis and management of gestational diabetes. There are also further analyses to be performed, such as the effect of maternal glucose on insulin sensitivity and triglyceride levels.

The first steps toward achieving a consensus on the clinical significance of the HAPO data will be made during a stand-alone meeting to be held in Pasadena, Calif. in a few months.

The effects seem to occur at levels lower than we've been diagnosing gestational diabetes in the past. DR. METZGER

GLASGOW, SCOTLAND — Fasting plasma glucose levels in pregnant women are associated with the development of preeclampsia, according to the latest data to be released from the Hyperglycemia and Adverse Pregnancy Outcome study.

Principal investigator Dr. Boyd L. Metzger gave a sneak preview of the latest results from the 7-year multinational trial, known as HAPO, during the annual professional conference of Diabetes U.K.

The new findings were on the secondary outcomes of the study. All three measures of glucose control—FPG at 1 hour and at 2 hours and the oral glucose tolerance test—were associated with preeclampsia. After adjusting for potentially confounding factors, the data showed that the odds of developing preeclampsia were 1.2–1.28 times higher for every 1-point standard deviation increase in maternal glucose concentration.

The key findings are the same as those previously released “in that outcomes are very strongly related to the mother's blood sugar level during her pregnancy and that the effects seem to occur at levels that are lower than we've been diagnosing and treating gestational diabetes at in the past,” Dr. Metzger of Northwestern University, Chicago, said in an interview.

But the new data show how the effects of maternal glucose during pregnancy appear to extend beyond an increased chance of having a high-birth-weight baby, fetal hyperinsulinemia, and delivery by cesarean section, and into outcomes such as preeclampsia. However, there was no specific number at which the risk of these adverse outcomes started to increase, he said. Although adverse outcomes appeared to occur at fasting plasma glucose (FPG) levels as low as 4.0–4.9 mg/dL, the data did not seem to agree as far as what level was needed to experience one adverse outcome over another.

Four outcomes were designated as “primary” at the start of the study, and were reported last June. These included the frequency of having a high-birth-weight baby, a first cesarean section, neonatal hypoglycemia, and fetal hyperinsulinemia. The last was assessed by measuring the level of C-peptide in the cord blood; a high level was defined as greater than the 90th percentile of what would be expected. Of these, Dr. Metzger said that there were strong associations between maternal glucose levels and both birth weight and fetal hyperinsulinemia; there were milder associations with having a first cesarean section and neonatal hypoglycemia.

“The relationship between maternal blood glucose and the outcome of [a woman's] pregnancy appears to be continuous across the whole spectrum of adverse outcomes, so how much risk is too much risk?” commented Dr. Metzger. This highlighted just one of the problems of applying the HAPO data as they stand to clinical practice. “If the clinical assessment of this information is that we should be intervening to prevent these risks, then we are going to have to lower the values that we have traditionally been using” to diagnose women with gestational diabetes—which will mean a that lot more women will be diagnosed with the disorder, he said.

This was the first time the HAPO findings have been discussed at a professional diabetes meeting outside the United States since their presentation at the American Diabetes Association 67th Annual Scientific Sessions last June.

The investigators are just starting to decipher all the data gathered from the more than 23,000 women who participated in the observational study, and it could be some time before there is an agreement on what these results mean for the diagnosis and management of gestational diabetes. There are also further analyses to be performed, such as the effect of maternal glucose on insulin sensitivity and triglyceride levels.

The first steps toward achieving a consensus on the clinical significance of the HAPO data will be made during a stand-alone meeting to be held in Pasadena, Calif. in a few months.

The effects seem to occur at levels lower than we've been diagnosing gestational diabetes in the past. DR. METZGER

GLASGOW, SCOTLAND — Metformin may protect against the development of cancer in patients with type 2 diabetes, according to findings from a large cohort study.

Among patients with diabetes who had been treated with metformin, 7.9% developed cancer, compared with 12.9% of diabetics who had never used the drug, reported Dr. Josie Evans of the University of Dundee (Scotland).

“Since the publication of the United Kingdom Prospective Diabetes Study (UKPDS), metformin is now one of the most commonly used oral antidiabetic medications in the world,” Dr. Evans said at the annual professional conference of Diabetes U.K.

Interest in an association between metformin and the development of cancer stems from the observation that the oral antidiabetic agent acts via the enzyme AMP-activated protein kinase, which is affected by the presence of LKB1, a well-known tumor suppressor. Previous research by Dr. Evans and her associates showed that of 3,828 patients with type 2 diabetes receiving hospital treatment for cancer, 1,276 had used metformin and 2,552 had not (BMJ 2005;330:1304-5).

Building on these pilot study data, the researchers performed a larger cohort study involving 8,170 patients diagnosed with type 2 diabetes, 4,085 of whom had been treated with metformin.

“When we excluded people who had only received one prescription of metformin, we still found a big difference in the proportions of users and nonusers who developed cancer,” Dr. Evans said. The cancer rate was 7.3% in patients who had used metformin more than once and 11.9% in those who had never used the antidiabetic drug.

“Of course one of the problems with this study is that it is an observational study, and it may be that the metformin users were at lower baseline risk of cancer than the comparators,” she commented.

One of the most significant differences in the patients at baseline was that the metformin users were an average of 6 years younger than the nonusers. But after adjusting for age, smoking history, and body mass index, among other variables, outcomes remained in favor of metformin use. Similar results were obtained for other outcomes, including all-cause mortality and mortality from cancer, as well as the incidence of three common cancers—colorectal, breast, and lung.

“We wanted to be sure that it's the metformin that is associated with a reduced risk of cancer and not something to do with people who are taking metformin,” Dr. Evans said, “so we looked to see if we could find a dose effect associated with metformin.” Although the data initially seemed to show a dose effect, this appeared to be confounded by the duration of treatment. Nevertheless, stratifying the results by dose and by duration suggested that there is “some evidence of a dose effect of metformin.” (See box.)

Dr. David Matthews of the Oxford (England) Center of Diabetes, Endocrinology and Metabolism commented that there was no evidence that the cardiovascular protection afforded by metformin in the UKPDS was dose related, so perhaps the same will hold true for metformin and cancer.

The study was supported by Tenovus Scotland, a charitable organization.

ELSEVIER GLOBAL MEDICAL NEWS

GLASGOW, SCOTLAND — Metformin may protect against the development of cancer in patients with type 2 diabetes, according to findings from a large cohort study.

Among patients with diabetes who had been treated with metformin, 7.9% developed cancer, compared with 12.9% of diabetics who had never used the drug, reported Dr. Josie Evans of the University of Dundee (Scotland).

“Since the publication of the United Kingdom Prospective Diabetes Study (UKPDS), metformin is now one of the most commonly used oral antidiabetic medications in the world,” Dr. Evans said at the annual professional conference of Diabetes U.K.

Interest in an association between metformin and the development of cancer stems from the observation that the oral antidiabetic agent acts via the enzyme AMP-activated protein kinase, which is affected by the presence of LKB1, a well-known tumor suppressor. Previous research by Dr. Evans and her associates showed that of 3,828 patients with type 2 diabetes receiving hospital treatment for cancer, 1,276 had used metformin and 2,552 had not (BMJ 2005;330:1304-5).

Building on these pilot study data, the researchers performed a larger cohort study involving 8,170 patients diagnosed with type 2 diabetes, 4,085 of whom had been treated with metformin.

“When we excluded people who had only received one prescription of metformin, we still found a big difference in the proportions of users and nonusers who developed cancer,” Dr. Evans said. The cancer rate was 7.3% in patients who had used metformin more than once and 11.9% in those who had never used the antidiabetic drug.

“Of course one of the problems with this study is that it is an observational study, and it may be that the metformin users were at lower baseline risk of cancer than the comparators,” she commented.

One of the most significant differences in the patients at baseline was that the metformin users were an average of 6 years younger than the nonusers. But after adjusting for age, smoking history, and body mass index, among other variables, outcomes remained in favor of metformin use. Similar results were obtained for other outcomes, including all-cause mortality and mortality from cancer, as well as the incidence of three common cancers—colorectal, breast, and lung.

“We wanted to be sure that it's the metformin that is associated with a reduced risk of cancer and not something to do with people who are taking metformin,” Dr. Evans said, “so we looked to see if we could find a dose effect associated with metformin.” Although the data initially seemed to show a dose effect, this appeared to be confounded by the duration of treatment. Nevertheless, stratifying the results by dose and by duration suggested that there is “some evidence of a dose effect of metformin.” (See box.)

Dr. David Matthews of the Oxford (England) Center of Diabetes, Endocrinology and Metabolism commented that there was no evidence that the cardiovascular protection afforded by metformin in the UKPDS was dose related, so perhaps the same will hold true for metformin and cancer.

The study was supported by Tenovus Scotland, a charitable organization.

ELSEVIER GLOBAL MEDICAL NEWS

GLASGOW, SCOTLAND — Metformin may protect against the development of cancer in patients with type 2 diabetes, according to findings from a large cohort study.

Among patients with diabetes who had been treated with metformin, 7.9% developed cancer, compared with 12.9% of diabetics who had never used the drug, reported Dr. Josie Evans of the University of Dundee (Scotland).

“Since the publication of the United Kingdom Prospective Diabetes Study (UKPDS), metformin is now one of the most commonly used oral antidiabetic medications in the world,” Dr. Evans said at the annual professional conference of Diabetes U.K.

Interest in an association between metformin and the development of cancer stems from the observation that the oral antidiabetic agent acts via the enzyme AMP-activated protein kinase, which is affected by the presence of LKB1, a well-known tumor suppressor. Previous research by Dr. Evans and her associates showed that of 3,828 patients with type 2 diabetes receiving hospital treatment for cancer, 1,276 had used metformin and 2,552 had not (BMJ 2005;330:1304-5).

Building on these pilot study data, the researchers performed a larger cohort study involving 8,170 patients diagnosed with type 2 diabetes, 4,085 of whom had been treated with metformin.

“When we excluded people who had only received one prescription of metformin, we still found a big difference in the proportions of users and nonusers who developed cancer,” Dr. Evans said. The cancer rate was 7.3% in patients who had used metformin more than once and 11.9% in those who had never used the antidiabetic drug.

“Of course one of the problems with this study is that it is an observational study, and it may be that the metformin users were at lower baseline risk of cancer than the comparators,” she commented.

One of the most significant differences in the patients at baseline was that the metformin users were an average of 6 years younger than the nonusers. But after adjusting for age, smoking history, and body mass index, among other variables, outcomes remained in favor of metformin use. Similar results were obtained for other outcomes, including all-cause mortality and mortality from cancer, as well as the incidence of three common cancers—colorectal, breast, and lung.

“We wanted to be sure that it's the metformin that is associated with a reduced risk of cancer and not something to do with people who are taking metformin,” Dr. Evans said, “so we looked to see if we could find a dose effect associated with metformin.” Although the data initially seemed to show a dose effect, this appeared to be confounded by the duration of treatment. Nevertheless, stratifying the results by dose and by duration suggested that there is “some evidence of a dose effect of metformin.” (See box.)

Dr. David Matthews of the Oxford (England) Center of Diabetes, Endocrinology and Metabolism commented that there was no evidence that the cardiovascular protection afforded by metformin in the UKPDS was dose related, so perhaps the same will hold true for metformin and cancer.

The study was supported by Tenovus Scotland, a charitable organization.

ELSEVIER GLOBAL MEDICAL NEWS

GLASGOW, SCOTLAND — Fasting plasma glucose levels in pregnant women are associated with the development of preeclampsia, according to the latest data to be released from the Hyperglycemia and Adverse Pregnancy Outcome study.

Principal investigator Dr. Boyd L. Metzger gave a sneak preview of the latest results from the 7-year multinational trial, known as HAPO, during the annual professional conference of Diabetes U.K.

The new findings were on the secondary outcomes of the study. All three measures of glucose control—FPG at 1 hour and at 2 hours and the oral glucose tolerance test—were associated with preeclampsia. After adjusting for potentially confounding factors, the data showed that the odds of developing preeclampsia were 1.2–1.28 times higher for every 1-point standard deviation increase in maternal glucose concentration.

The key findings are the same as those previously released “in that outcomes are very strongly related to the mother's blood sugar level during her pregnancy and that the effects seem to occur at levels that are lower than we've been diagnosing and treating gestational diabetes at in the past,” Dr. Metzger of Northwestern University, Chicago, said in an interview.

But the new data show how the effects of maternal glucose during pregnancy appear to extend beyond an increased chance of having a high- birth-weight baby, fetal hyperinsulinemia, and delivery by cesarean section, and into outcomes such as preeclampsia. However, there was no specific number at which the risk of these adverse outcomes started to increase, he said. Although adverse outcomes appeared to occur at fasting plasma glucose (FPG) levels as low as 4.0–4.9 mg/dL, the data did not seem to agree as far as what level was needed to experience one adverse outcome over another.

Four outcomes were designated as “primary” at the start of the study, and were reported last June. These included the frequency of having a high-birth-weight baby, a first cesarean section, neonatal hypoglycemia, and fetal hyperinsulinemia. The last was assessed by measuring the level of C-peptide in the cord blood; a high level was defined as greater than the 90th percentile of what would be expected. Of these, Dr. Metzger said that there were strong associations between maternal glucose levels and both birthweight and fetal hyperinsulinemia; there were milder associations with having a first cesarean section and neonatal hypoglycemia.

“The relationship between maternal blood glucose and the outcome of [a woman's] pregnancy appears to be continuous across the whole spectrum of adverse outcomes, so how much risk is too much risk?” commented Dr. Metzger. This highlighted just one of the problems of applying the HAPO data as they stand to clinical practice. “If the clinical assessment of this information is that we should be intervening to prevent these risks, then we are going to have to lower the values that we have traditionally been using” to diagnose women with gestational diabetes—which will mean that a lot more women will be diagnosed with the disorder, he said.

This was the first time the HAPO findings have been discussed at a professional diabetes meeting outside the United States since their presentation at the American Diabetes Association 67th Annual Scientific Sessions last June.

The investigators are just starting to decipher all the data gathered from the more than 23,000 women who participated in the observational study, and it could be some time before there is an agreement on what these results mean for the diagnosis and management of gestational diabetes. There are also further analyses to be performed, such as the effect of maternal glucose on insulin sensitivity and triglyceride levels.

The first steps toward achieving a consensus on the clinical significance of the HAPO data will be made during a stand-alone meeting to be held in Pasadena, Calif. in a few months. The International Workshop Conference on Gestational Diabetes: Diagnosis and Classification (www.iadpsg.org

“We are now faced with a lot of information,” said Dr. Metzger. “What change will occur is going to be determined by consensus, and by weighing up all the clinical implications.”

The effects from blood sugar occur at levels that are lower than have been treated in the past. DR. METZGER

GLASGOW, SCOTLAND — Fasting plasma glucose levels in pregnant women are associated with the development of preeclampsia, according to the latest data to be released from the Hyperglycemia and Adverse Pregnancy Outcome study.

Principal investigator Dr. Boyd L. Metzger gave a sneak preview of the latest results from the 7-year multinational trial, known as HAPO, during the annual professional conference of Diabetes U.K.

The new findings were on the secondary outcomes of the study. All three measures of glucose control—FPG at 1 hour and at 2 hours and the oral glucose tolerance test—were associated with preeclampsia. After adjusting for potentially confounding factors, the data showed that the odds of developing preeclampsia were 1.2–1.28 times higher for every 1-point standard deviation increase in maternal glucose concentration.

The key findings are the same as those previously released “in that outcomes are very strongly related to the mother's blood sugar level during her pregnancy and that the effects seem to occur at levels that are lower than we've been diagnosing and treating gestational diabetes at in the past,” Dr. Metzger of Northwestern University, Chicago, said in an interview.

But the new data show how the effects of maternal glucose during pregnancy appear to extend beyond an increased chance of having a high- birth-weight baby, fetal hyperinsulinemia, and delivery by cesarean section, and into outcomes such as preeclampsia. However, there was no specific number at which the risk of these adverse outcomes started to increase, he said. Although adverse outcomes appeared to occur at fasting plasma glucose (FPG) levels as low as 4.0–4.9 mg/dL, the data did not seem to agree as far as what level was needed to experience one adverse outcome over another.

Four outcomes were designated as “primary” at the start of the study, and were reported last June. These included the frequency of having a high-birth-weight baby, a first cesarean section, neonatal hypoglycemia, and fetal hyperinsulinemia. The last was assessed by measuring the level of C-peptide in the cord blood; a high level was defined as greater than the 90th percentile of what would be expected. Of these, Dr. Metzger said that there were strong associations between maternal glucose levels and both birthweight and fetal hyperinsulinemia; there were milder associations with having a first cesarean section and neonatal hypoglycemia.

“The relationship between maternal blood glucose and the outcome of [a woman's] pregnancy appears to be continuous across the whole spectrum of adverse outcomes, so how much risk is too much risk?” commented Dr. Metzger. This highlighted just one of the problems of applying the HAPO data as they stand to clinical practice. “If the clinical assessment of this information is that we should be intervening to prevent these risks, then we are going to have to lower the values that we have traditionally been using” to diagnose women with gestational diabetes—which will mean that a lot more women will be diagnosed with the disorder, he said.

This was the first time the HAPO findings have been discussed at a professional diabetes meeting outside the United States since their presentation at the American Diabetes Association 67th Annual Scientific Sessions last June.

The investigators are just starting to decipher all the data gathered from the more than 23,000 women who participated in the observational study, and it could be some time before there is an agreement on what these results mean for the diagnosis and management of gestational diabetes. There are also further analyses to be performed, such as the effect of maternal glucose on insulin sensitivity and triglyceride levels.

The first steps toward achieving a consensus on the clinical significance of the HAPO data will be made during a stand-alone meeting to be held in Pasadena, Calif. in a few months. The International Workshop Conference on Gestational Diabetes: Diagnosis and Classification (www.iadpsg.org

“We are now faced with a lot of information,” said Dr. Metzger. “What change will occur is going to be determined by consensus, and by weighing up all the clinical implications.”

The effects from blood sugar occur at levels that are lower than have been treated in the past. DR. METZGER

GLASGOW, SCOTLAND — Fasting plasma glucose levels in pregnant women are associated with the development of preeclampsia, according to the latest data to be released from the Hyperglycemia and Adverse Pregnancy Outcome study.

Principal investigator Dr. Boyd L. Metzger gave a sneak preview of the latest results from the 7-year multinational trial, known as HAPO, during the annual professional conference of Diabetes U.K.

The new findings were on the secondary outcomes of the study. All three measures of glucose control—FPG at 1 hour and at 2 hours and the oral glucose tolerance test—were associated with preeclampsia. After adjusting for potentially confounding factors, the data showed that the odds of developing preeclampsia were 1.2–1.28 times higher for every 1-point standard deviation increase in maternal glucose concentration.

The key findings are the same as those previously released “in that outcomes are very strongly related to the mother's blood sugar level during her pregnancy and that the effects seem to occur at levels that are lower than we've been diagnosing and treating gestational diabetes at in the past,” Dr. Metzger of Northwestern University, Chicago, said in an interview.

But the new data show how the effects of maternal glucose during pregnancy appear to extend beyond an increased chance of having a high- birth-weight baby, fetal hyperinsulinemia, and delivery by cesarean section, and into outcomes such as preeclampsia. However, there was no specific number at which the risk of these adverse outcomes started to increase, he said. Although adverse outcomes appeared to occur at fasting plasma glucose (FPG) levels as low as 4.0–4.9 mg/dL, the data did not seem to agree as far as what level was needed to experience one adverse outcome over another.

Four outcomes were designated as “primary” at the start of the study, and were reported last June. These included the frequency of having a high-birth-weight baby, a first cesarean section, neonatal hypoglycemia, and fetal hyperinsulinemia. The last was assessed by measuring the level of C-peptide in the cord blood; a high level was defined as greater than the 90th percentile of what would be expected. Of these, Dr. Metzger said that there were strong associations between maternal glucose levels and both birthweight and fetal hyperinsulinemia; there were milder associations with having a first cesarean section and neonatal hypoglycemia.

“The relationship between maternal blood glucose and the outcome of [a woman's] pregnancy appears to be continuous across the whole spectrum of adverse outcomes, so how much risk is too much risk?” commented Dr. Metzger. This highlighted just one of the problems of applying the HAPO data as they stand to clinical practice. “If the clinical assessment of this information is that we should be intervening to prevent these risks, then we are going to have to lower the values that we have traditionally been using” to diagnose women with gestational diabetes—which will mean that a lot more women will be diagnosed with the disorder, he said.

This was the first time the HAPO findings have been discussed at a professional diabetes meeting outside the United States since their presentation at the American Diabetes Association 67th Annual Scientific Sessions last June.

The investigators are just starting to decipher all the data gathered from the more than 23,000 women who participated in the observational study, and it could be some time before there is an agreement on what these results mean for the diagnosis and management of gestational diabetes. There are also further analyses to be performed, such as the effect of maternal glucose on insulin sensitivity and triglyceride levels.

The first steps toward achieving a consensus on the clinical significance of the HAPO data will be made during a stand-alone meeting to be held in Pasadena, Calif. in a few months. The International Workshop Conference on Gestational Diabetes: Diagnosis and Classification (www.iadpsg.org

“We are now faced with a lot of information,” said Dr. Metzger. “What change will occur is going to be determined by consensus, and by weighing up all the clinical implications.”

The effects from blood sugar occur at levels that are lower than have been treated in the past. DR. METZGER