EASD: Studies slam cardiovascular safety of sulfonylureas

STOCKHOLM – While they are effective glucose-lowering agents, sulfonylureas seem to substantially increase the risk for cardiovascular events relative to other antihyperglycemic agents, according to several presentations given at the annual meeting of the European Association for the Study of Diabetes.

In a systematic review and meta-analysis of 84 randomized controlled trials (RCTs) and 26 observational studies, accounting for more than 36,573 and more than 1.5 million subjects with type 2 diabetes mellitus, respectively, treatment with a sulfonylurea was found to significantly increase all-cause mortality and cardiovascular-related mortality when compared to no active treatment or placebo and to other antidiabetic drugs.

In the RCTs, the hazard ratios comparing sulfonylureas to all treatments combined were 1.26 (95% confidence interval, 1.10-1.44) for all-cause mortality and 1.46 (95% CI, 1.21-1.77) for cardiovascular mortality. HR for CV mortality was highest when comparing sulfonylureas with GLP-1 agonists and SGLT-2 inhibitors, at 45.4 (95% CI, 2.07-362.8) and 42.6 (95% CI, 1.71-359.1), with significantly elevated risk also seen when compared with DPP-4 inhibitors (HR, 4.42; 95% CI, 1.92-13.0), thiazolidinediones (HR, 3.05; 95% CI, 1.79-5.54), and insulin (HR, 1.30; 95% CI, 1.02-1.66).

There was also an increase in the risk for acute myocardial infarction and stroke in some comparisons of sulfonylureas versus other treatments.

“We think this study is the most comprehensive review of RCTs and observational studies to compare CV-related outcomes and all-cause mortality among patients receiving sulfonylureas versus [other] antidiabetic agents,” said Dr. Steve Bain of Abertawe Bro Morgannwg University NHS Trust in Swansea, England.

“We accept that the confidence intervals were wide, but there is a tendency for all significant results to trend in the same direction.” Results of the observational studies analyses generally supported the results from the RCTs, he added. A notable limitation was that there had been no differentiation between the risk according to different sulfonylureas. Dr. Bain said this was to try to keep the number of patients available for analysis as high as possible in the sulfonylurea group.

Data from the German CREST (CV Risk and Events on SU Treatment) study also showed that there was a higher CV risk associated with sulfonylureas when compared with metformin in patients newly diagnosed with type 2 diabetes mellitus. The retrospective cohort database study included more than 35,000 individuals, of whom 26,883 received no antihyperglycemic treatment or treatment with an antihyperglycemic agent other than a sulfonylurea or metformin between January 2010 and December 2011, leaving 7,874 who were given metformin monotherapy and 904 given sulfonylurea monotherapy. Propensity score matching was used to form the final study groups consisting of 730 patients each.

Crude, propensity-matched, and adjusted HRs for death comparing sulfonylurea with metformin as first-line treatment were 3.32 (95% CI, 2.6-4.3; P less than .001), 1.4 (95% CI, 0.9-2.3; P less than .12), and 2.01 (95% CI, 1.5-2.6; P less than .001), respectively. Corresponding HRs for major cardiovascular events (MACE) were 1.9 (95% CI, 1.4-2.4; P less than .001), 1.4 (95% CI, 0.9-2.2; P less than .14), and 1.3 (95% CI, 1.03-1.74; P less than .05).

Dr. Thomas Wilke of the University of Wismar (Germany) who presented the CREST study findings, noted that there were some differences between the groups in terms of the baseline demographics, with those receiving sulfonylureas being somewhat older and more likely to be women, have comorbid disease, and discontinue monotherapy earlier than patients treated with metformin, or possibly be more likely to have contraindications to the use of metformin.

“Taking these differences into account, patients treated with sulfonylureas seem to be at increased risk of death, MACE, and type 2 diabetes hospitalization, compared with metformin-treated patients,” in this German population, Dr. Wilke concluded.

Finally, Dr. Jan Eriksson of Uppsala (Sweden) University presented findings from an observational study of more than 50,000 patients with type 2 diabetes treated with a combination of two noninsulin antidiabetic agents between 2006 and 2013. Of these, 40,736 had received metformin in combination with a sulfonylurea and 12,024 had received a combination of metformin and a DPP-4 inhibitor and were studied further.

Compared with the patients who received a DPP-4 inhibitor on top of metformin, those who received a sulfonylurea had an increased risk of both fatal and nonfatal cardiovascular disease (HR, 1.17; 95% CI, 1.01-1.37) and death from any cause (HR, 1.25; 95% CI, 1.02-1.54) even in the presence of confounding factors such as age, gender, existing cardiovascular disease, and cardio-preventative medication use.

The risk for severe hypoglycemia was also greater with sulfonylurea than DPP-4 use (HR, 2.07; 95% CI, 1.11-3.86), and experiencing severe hypoglycemia was associated with an increased risk for cardiovascular events (HR, 1.51; 95% CI, 1.21-1.88).

“Hypoglycemia may contribute to the increased CVD risk associated with sulfonylureas,” Dr. Eriksson suggested. He concluded that the results of ongoing and future randomized trials would be important in order to understand the cardiovascular risk associated with sulfonylureas better and try to figure out any underlying mechanisms.

Dr. Raz Itamar, head of the diabetes unit at Hadassah Medical Center in Jerusalem, chaired the session during which all these study findings were presented and provided a general comment in an interview on why the cardiovascular safety of sulfonylureas was a particular issue that these studies looked at.

He noted that a study performed in the United States in the 1970s had raised suspicion that there might be an increased cardiovascular risk associated with the sulfonylureas, but the current concerns really started with findings from the UKPDS [United Kingdom Prospective Diabetes Study] and the Women’s Health Study (WHS), for example.

“In the UKPDS, patients who were on metformin and a sulfonylurea demonstrated a doubling in the cardiovascular event rate,” Dr. Itamar said, adding that the WHS data also showed a doubling of the risk.

There is a lack of head-to-head, randomized, controlled data, he emphasized and commented that there were many limitations to the analyses presented at the session. These included the small sample sizes and short duration of follow-up, among others. There is also evidence in the literature suggesting that the cardiovascular risk of individual sulfonylureas may be different. So the results need to be reviewed with caution, he advised.

“As a clinician, I try to give a sulfonylurea which seems to have a safer cardiovascular profile, like glimepiride and gliclazide,” Dr. Itamar said. He noted that there was a randomized controlled trial currently recruiting a projected 6,000 patients that would compare glimepiride with a DPP-4 inhibitor directly for several years that should provide additional and much needed information on the cardiovascular safety of sulfonylureas.

Merck UK funded the systematic review and meta-analysis. AstraZeneca supported the CREST study and the sulfonylureas versus DPP4 inhibitor comparison study. Dr. Bain disclosed he had received honoraria, teaching, and research sponsorship or grants from multiple pharmaceutical companies including Merck Sharp & Dohme and AstraZeneca. Dr. Wilke did not report his disclosures. Dr. Eriksson disclosed receiving research grants or honoraria from AstraZeneca and Merck Sharp & Dohme, among others, and that he had previously been an employee of AstraZeneca. Dr. Itamar did not report having any relevant disclosures.

STOCKHOLM – While they are effective glucose-lowering agents, sulfonylureas seem to substantially increase the risk for cardiovascular events relative to other antihyperglycemic agents, according to several presentations given at the annual meeting of the European Association for the Study of Diabetes.

In a systematic review and meta-analysis of 84 randomized controlled trials (RCTs) and 26 observational studies, accounting for more than 36,573 and more than 1.5 million subjects with type 2 diabetes mellitus, respectively, treatment with a sulfonylurea was found to significantly increase all-cause mortality and cardiovascular-related mortality when compared to no active treatment or placebo and to other antidiabetic drugs.

In the RCTs, the hazard ratios comparing sulfonylureas to all treatments combined were 1.26 (95% confidence interval, 1.10-1.44) for all-cause mortality and 1.46 (95% CI, 1.21-1.77) for cardiovascular mortality. HR for CV mortality was highest when comparing sulfonylureas with GLP-1 agonists and SGLT-2 inhibitors, at 45.4 (95% CI, 2.07-362.8) and 42.6 (95% CI, 1.71-359.1), with significantly elevated risk also seen when compared with DPP-4 inhibitors (HR, 4.42; 95% CI, 1.92-13.0), thiazolidinediones (HR, 3.05; 95% CI, 1.79-5.54), and insulin (HR, 1.30; 95% CI, 1.02-1.66).

There was also an increase in the risk for acute myocardial infarction and stroke in some comparisons of sulfonylureas versus other treatments.

“We think this study is the most comprehensive review of RCTs and observational studies to compare CV-related outcomes and all-cause mortality among patients receiving sulfonylureas versus [other] antidiabetic agents,” said Dr. Steve Bain of Abertawe Bro Morgannwg University NHS Trust in Swansea, England.

“We accept that the confidence intervals were wide, but there is a tendency for all significant results to trend in the same direction.” Results of the observational studies analyses generally supported the results from the RCTs, he added. A notable limitation was that there had been no differentiation between the risk according to different sulfonylureas. Dr. Bain said this was to try to keep the number of patients available for analysis as high as possible in the sulfonylurea group.

Data from the German CREST (CV Risk and Events on SU Treatment) study also showed that there was a higher CV risk associated with sulfonylureas when compared with metformin in patients newly diagnosed with type 2 diabetes mellitus. The retrospective cohort database study included more than 35,000 individuals, of whom 26,883 received no antihyperglycemic treatment or treatment with an antihyperglycemic agent other than a sulfonylurea or metformin between January 2010 and December 2011, leaving 7,874 who were given metformin monotherapy and 904 given sulfonylurea monotherapy. Propensity score matching was used to form the final study groups consisting of 730 patients each.

Crude, propensity-matched, and adjusted HRs for death comparing sulfonylurea with metformin as first-line treatment were 3.32 (95% CI, 2.6-4.3; P less than .001), 1.4 (95% CI, 0.9-2.3; P less than .12), and 2.01 (95% CI, 1.5-2.6; P less than .001), respectively. Corresponding HRs for major cardiovascular events (MACE) were 1.9 (95% CI, 1.4-2.4; P less than .001), 1.4 (95% CI, 0.9-2.2; P less than .14), and 1.3 (95% CI, 1.03-1.74; P less than .05).

Dr. Thomas Wilke of the University of Wismar (Germany) who presented the CREST study findings, noted that there were some differences between the groups in terms of the baseline demographics, with those receiving sulfonylureas being somewhat older and more likely to be women, have comorbid disease, and discontinue monotherapy earlier than patients treated with metformin, or possibly be more likely to have contraindications to the use of metformin.

“Taking these differences into account, patients treated with sulfonylureas seem to be at increased risk of death, MACE, and type 2 diabetes hospitalization, compared with metformin-treated patients,” in this German population, Dr. Wilke concluded.

Finally, Dr. Jan Eriksson of Uppsala (Sweden) University presented findings from an observational study of more than 50,000 patients with type 2 diabetes treated with a combination of two noninsulin antidiabetic agents between 2006 and 2013. Of these, 40,736 had received metformin in combination with a sulfonylurea and 12,024 had received a combination of metformin and a DPP-4 inhibitor and were studied further.

Compared with the patients who received a DPP-4 inhibitor on top of metformin, those who received a sulfonylurea had an increased risk of both fatal and nonfatal cardiovascular disease (HR, 1.17; 95% CI, 1.01-1.37) and death from any cause (HR, 1.25; 95% CI, 1.02-1.54) even in the presence of confounding factors such as age, gender, existing cardiovascular disease, and cardio-preventative medication use.

The risk for severe hypoglycemia was also greater with sulfonylurea than DPP-4 use (HR, 2.07; 95% CI, 1.11-3.86), and experiencing severe hypoglycemia was associated with an increased risk for cardiovascular events (HR, 1.51; 95% CI, 1.21-1.88).

“Hypoglycemia may contribute to the increased CVD risk associated with sulfonylureas,” Dr. Eriksson suggested. He concluded that the results of ongoing and future randomized trials would be important in order to understand the cardiovascular risk associated with sulfonylureas better and try to figure out any underlying mechanisms.

Dr. Raz Itamar, head of the diabetes unit at Hadassah Medical Center in Jerusalem, chaired the session during which all these study findings were presented and provided a general comment in an interview on why the cardiovascular safety of sulfonylureas was a particular issue that these studies looked at.

He noted that a study performed in the United States in the 1970s had raised suspicion that there might be an increased cardiovascular risk associated with the sulfonylureas, but the current concerns really started with findings from the UKPDS [United Kingdom Prospective Diabetes Study] and the Women’s Health Study (WHS), for example.

“In the UKPDS, patients who were on metformin and a sulfonylurea demonstrated a doubling in the cardiovascular event rate,” Dr. Itamar said, adding that the WHS data also showed a doubling of the risk.

There is a lack of head-to-head, randomized, controlled data, he emphasized and commented that there were many limitations to the analyses presented at the session. These included the small sample sizes and short duration of follow-up, among others. There is also evidence in the literature suggesting that the cardiovascular risk of individual sulfonylureas may be different. So the results need to be reviewed with caution, he advised.

“As a clinician, I try to give a sulfonylurea which seems to have a safer cardiovascular profile, like glimepiride and gliclazide,” Dr. Itamar said. He noted that there was a randomized controlled trial currently recruiting a projected 6,000 patients that would compare glimepiride with a DPP-4 inhibitor directly for several years that should provide additional and much needed information on the cardiovascular safety of sulfonylureas.

Merck UK funded the systematic review and meta-analysis. AstraZeneca supported the CREST study and the sulfonylureas versus DPP4 inhibitor comparison study. Dr. Bain disclosed he had received honoraria, teaching, and research sponsorship or grants from multiple pharmaceutical companies including Merck Sharp & Dohme and AstraZeneca. Dr. Wilke did not report his disclosures. Dr. Eriksson disclosed receiving research grants or honoraria from AstraZeneca and Merck Sharp & Dohme, among others, and that he had previously been an employee of AstraZeneca. Dr. Itamar did not report having any relevant disclosures.

STOCKHOLM – While they are effective glucose-lowering agents, sulfonylureas seem to substantially increase the risk for cardiovascular events relative to other antihyperglycemic agents, according to several presentations given at the annual meeting of the European Association for the Study of Diabetes.

In a systematic review and meta-analysis of 84 randomized controlled trials (RCTs) and 26 observational studies, accounting for more than 36,573 and more than 1.5 million subjects with type 2 diabetes mellitus, respectively, treatment with a sulfonylurea was found to significantly increase all-cause mortality and cardiovascular-related mortality when compared to no active treatment or placebo and to other antidiabetic drugs.

In the RCTs, the hazard ratios comparing sulfonylureas to all treatments combined were 1.26 (95% confidence interval, 1.10-1.44) for all-cause mortality and 1.46 (95% CI, 1.21-1.77) for cardiovascular mortality. HR for CV mortality was highest when comparing sulfonylureas with GLP-1 agonists and SGLT-2 inhibitors, at 45.4 (95% CI, 2.07-362.8) and 42.6 (95% CI, 1.71-359.1), with significantly elevated risk also seen when compared with DPP-4 inhibitors (HR, 4.42; 95% CI, 1.92-13.0), thiazolidinediones (HR, 3.05; 95% CI, 1.79-5.54), and insulin (HR, 1.30; 95% CI, 1.02-1.66).

There was also an increase in the risk for acute myocardial infarction and stroke in some comparisons of sulfonylureas versus other treatments.

“We think this study is the most comprehensive review of RCTs and observational studies to compare CV-related outcomes and all-cause mortality among patients receiving sulfonylureas versus [other] antidiabetic agents,” said Dr. Steve Bain of Abertawe Bro Morgannwg University NHS Trust in Swansea, England.

“We accept that the confidence intervals were wide, but there is a tendency for all significant results to trend in the same direction.” Results of the observational studies analyses generally supported the results from the RCTs, he added. A notable limitation was that there had been no differentiation between the risk according to different sulfonylureas. Dr. Bain said this was to try to keep the number of patients available for analysis as high as possible in the sulfonylurea group.

Data from the German CREST (CV Risk and Events on SU Treatment) study also showed that there was a higher CV risk associated with sulfonylureas when compared with metformin in patients newly diagnosed with type 2 diabetes mellitus. The retrospective cohort database study included more than 35,000 individuals, of whom 26,883 received no antihyperglycemic treatment or treatment with an antihyperglycemic agent other than a sulfonylurea or metformin between January 2010 and December 2011, leaving 7,874 who were given metformin monotherapy and 904 given sulfonylurea monotherapy. Propensity score matching was used to form the final study groups consisting of 730 patients each.

Crude, propensity-matched, and adjusted HRs for death comparing sulfonylurea with metformin as first-line treatment were 3.32 (95% CI, 2.6-4.3; P less than .001), 1.4 (95% CI, 0.9-2.3; P less than .12), and 2.01 (95% CI, 1.5-2.6; P less than .001), respectively. Corresponding HRs for major cardiovascular events (MACE) were 1.9 (95% CI, 1.4-2.4; P less than .001), 1.4 (95% CI, 0.9-2.2; P less than .14), and 1.3 (95% CI, 1.03-1.74; P less than .05).

Dr. Thomas Wilke of the University of Wismar (Germany) who presented the CREST study findings, noted that there were some differences between the groups in terms of the baseline demographics, with those receiving sulfonylureas being somewhat older and more likely to be women, have comorbid disease, and discontinue monotherapy earlier than patients treated with metformin, or possibly be more likely to have contraindications to the use of metformin.

“Taking these differences into account, patients treated with sulfonylureas seem to be at increased risk of death, MACE, and type 2 diabetes hospitalization, compared with metformin-treated patients,” in this German population, Dr. Wilke concluded.

Finally, Dr. Jan Eriksson of Uppsala (Sweden) University presented findings from an observational study of more than 50,000 patients with type 2 diabetes treated with a combination of two noninsulin antidiabetic agents between 2006 and 2013. Of these, 40,736 had received metformin in combination with a sulfonylurea and 12,024 had received a combination of metformin and a DPP-4 inhibitor and were studied further.

Compared with the patients who received a DPP-4 inhibitor on top of metformin, those who received a sulfonylurea had an increased risk of both fatal and nonfatal cardiovascular disease (HR, 1.17; 95% CI, 1.01-1.37) and death from any cause (HR, 1.25; 95% CI, 1.02-1.54) even in the presence of confounding factors such as age, gender, existing cardiovascular disease, and cardio-preventative medication use.

The risk for severe hypoglycemia was also greater with sulfonylurea than DPP-4 use (HR, 2.07; 95% CI, 1.11-3.86), and experiencing severe hypoglycemia was associated with an increased risk for cardiovascular events (HR, 1.51; 95% CI, 1.21-1.88).

“Hypoglycemia may contribute to the increased CVD risk associated with sulfonylureas,” Dr. Eriksson suggested. He concluded that the results of ongoing and future randomized trials would be important in order to understand the cardiovascular risk associated with sulfonylureas better and try to figure out any underlying mechanisms.

Dr. Raz Itamar, head of the diabetes unit at Hadassah Medical Center in Jerusalem, chaired the session during which all these study findings were presented and provided a general comment in an interview on why the cardiovascular safety of sulfonylureas was a particular issue that these studies looked at.

He noted that a study performed in the United States in the 1970s had raised suspicion that there might be an increased cardiovascular risk associated with the sulfonylureas, but the current concerns really started with findings from the UKPDS [United Kingdom Prospective Diabetes Study] and the Women’s Health Study (WHS), for example.

“In the UKPDS, patients who were on metformin and a sulfonylurea demonstrated a doubling in the cardiovascular event rate,” Dr. Itamar said, adding that the WHS data also showed a doubling of the risk.

There is a lack of head-to-head, randomized, controlled data, he emphasized and commented that there were many limitations to the analyses presented at the session. These included the small sample sizes and short duration of follow-up, among others. There is also evidence in the literature suggesting that the cardiovascular risk of individual sulfonylureas may be different. So the results need to be reviewed with caution, he advised.

“As a clinician, I try to give a sulfonylurea which seems to have a safer cardiovascular profile, like glimepiride and gliclazide,” Dr. Itamar said. He noted that there was a randomized controlled trial currently recruiting a projected 6,000 patients that would compare glimepiride with a DPP-4 inhibitor directly for several years that should provide additional and much needed information on the cardiovascular safety of sulfonylureas.

Merck UK funded the systematic review and meta-analysis. AstraZeneca supported the CREST study and the sulfonylureas versus DPP4 inhibitor comparison study. Dr. Bain disclosed he had received honoraria, teaching, and research sponsorship or grants from multiple pharmaceutical companies including Merck Sharp & Dohme and AstraZeneca. Dr. Wilke did not report his disclosures. Dr. Eriksson disclosed receiving research grants or honoraria from AstraZeneca and Merck Sharp & Dohme, among others, and that he had previously been an employee of AstraZeneca. Dr. Itamar did not report having any relevant disclosures.