Is treatment with glibenclamide (glyburide) associated with a higher incidence of hypoglycaemia tha

Fråga: Is treatment with glibenclamide (glyburide) associated with a higher incidence of hypoglycaemia than other sulphonylureas? Does treatment with glibenclamide more frequently induce hypoglycaemia in older as compared to younger patients?

Sammanfattning: Although a thorough literature search has been performed, only a limited number of relevant references have been found with respect to sulphonylureas and hypoglycaemia. A few studies show that hypoglycaemia reactions were significantly more common with glibenclamide compared to other sulphonylureas. However, in a majority of the few studies obtained only glibenclamide and glipizide have been compared. The median age of fatal/serious cases of hypoglycaemia with glibenclamide and glipizide was higher than in random samples on the same treatment. The data from the Swedish AR-reporting cannot be used for any conclusions.

Svar: Although a thorough literature search including Medline, Drugline and common pharmacological handbooks has been performed, few relevant references have been found. In a retrospective cohort study involving older patients (n = 13963 medicaid enrols, aged 65 years or older) the serious hypoglycaemia risk caused by various sulphonylureas was studied. The crude rate was highest in glibenclamide users, 16.6 per 1000 person-years (95 per cent CI, 13.2-19.9) and lowest among users of tolbutamide, 3.5 (95 per cent CI, 1.2-5.9). Users of tolbutamide, tolazamide, and glipizide showed lower risks of serious hypoglycaemia than users of chlorpropamide. Among the second generation of sulphonylureas, the adjusted relative risk of serious hypoglycaemia among glibenclamide users, compared with glipizide users, was 1.9 (95 per cent CI, 1.2-2.9). An increased risk of serious hypoglycaemia associated with use of glibenclamide compared to glipizide occurred in all strata including gender, race, residence, dose and duration of dose (1). In a German study, 237 type II diabetics treated with glibenclamide (mean dose 6.7 mg, age 65 years and duration of diabetes 9 years) were followed with respect to severe episodes of hypoglycaemia. Among the patients treated with glibenclamide, three (1.3 per cent) experienced one episode of severe hypoglycaemia each (incidence 0.013/patient year). Old age, maximum dosage of glibenclamide (15 mg/day) and multimorbidity were characteristics of these patients (2).

Glycemic control in elderly patients with NIDDM (type II diabetes, n = 22 randomized in a double-blind fashion) were followed during 6 months with either glibenclamide or gliclazide (a sulphonylurea related to glipizide). Glycemic control was equivalent with the two drugs, but hypoglycaemic reactions were significantly more common with glibenclamide than with gliclazide (p < 0.01) (3). In the years 1980-1987, 19 cases of severe hypoglycaemia during treatment of NIDDM with glipizide were reported to SADRAC. Patient age was 75 +/- 9 years, which was significantly higher than in all patients on glipizide in Sweden. An uneventful early recovery occurred in 14 patients. The remaining five patients had prolonged or recurrent hypoglycaemia for up to 60 hours. Two of the patients, both with complicating disorders died. The hypoglycaemic patients had renal impairment more often than age- and sex-matched controls treated with glipizide (odds ratio 4.0). There was no control group including patients treated with glibenclamide (4).

In a report 1990 focusing on sales and adverse reactions of oral hypoglycaemic drugs in Sweden glibenclamide dominated and in 1989 the sales were sufficient to treat 60000 of 82000 patients throughout the year. The ADR profile of the sulphonylureas at the same time by far was dominated by hypoglycaemia (n = 150/310 reports), a reaction which was fatal in 11 cases. The corresponding number for glibenclamide and glipizide (within parentheses) was 102 (21) and 10 (1) fatal cases, respectively. If these numbers are related to sales with respect to glibenclamide there was one report of hypoglycaemia/2.65 million DDD and one fatal case/27 million DDD, while the same result for glipizide was one report of hypoglycaemia/3.16 million DDD and one fatal case/66 million DDD. The results with respect to the other sulphonylureas were not evaluated.

Eleven patients, ten of which had been treated with glibenclamide, died due to their hypoglycaemic adverse reaction. Median age of these fatal cases was 75 years and two were more than 83 years. For comparison the median age at a random sample (one in 288 of all patients prescribed glibenclamide, 6) was 70 years and 5 per cent were 85 years or older (5). This information indicates that older patients on glibenclamide seem to have a more serious outcome than the younger age groups.

A current contact with the Swedish Adverse Drug Reactions Committee (SADRAC, 7) have provided a recent number of reports and sales statistics for now available sulphonylureas in Sweden until September 1996. With respect to hypoglycaemia the following results can be calculated: Glibenclamide, one case of hypoglycaemia/3.6 were, in million DDD (fatal ratio, 1/23 million DDD): glipizide 1/4.6 (fatal 1/43), chlorpropamide 1/6.1 (fatal 1/133) and totalazamide 1/2.2 (no fatal cases). However, if the differences are not of a major magnitude, these data from a spontaneous reporting system - mainly aimed as a generator of signals of rare serious reactions - cannot be used to draw any conclusions of the incidence of hypoglycaemia with different sulphonylureas.