Sulfonylurea compounds, especially tolbutamide and chlorpropamide, are known to interact with pheny

Fråga: Sulfonylurea compounds, especially tolbutamide and chlorpropamide, are known to interact with phenylbutazone and sometimes give rise to severe hypoglycemic reactions. What about the status of other NSAIDs, eg azapropazone?

Sammanfattning: The interactions between tolbutamide and especially phenylbutazone are still the most significant and well-reported examples of sulfonylurea - NSAID interaction. Less convincing interactions have been found for tolbutamide/oxyphenbutazone and tolbutamide/salicylates. Two case reports concerning tolbutamide and azapropazone in healthy volunteers have been found in the literature, while no interactions have appeared with respect to naproxen, sulindac and ibuprofen. Chlorpropamide interacts with phenylbutazone, low doses of salicylates probably exert a minimal hypoglycemic effect, while no interaction concerning ibuprofen has been found. No certain interactions between glibenclamide/piroxicam and tenoxicam were found in the literature, nor were there any results in favour of any interaction between glipizide and indoprofen.

Svar: Tolbutamide and NSAID: Already in 1959 it was reported that phenylbutazone in three patients with NIDDM increased the hypoglycemic effect of tolbutamide (1). The same phenomenon was observed in half a dozen studies. Phenylbutazone pretreatment increased tolbutamide half-life, whereas tolbutamide pretreatment decreased phenylbutazone half-life in man (2). In a recent overview of antidiabetic drug interactions (3) the mechanism behind the interaction between tolbutamide and phenylbutazone is explained by: 1) Increased tolbutamide concentrations due to inhibition of its oxidation; 2) Displacement of tolbutamide from plasma protein binding by phenylbutazone; 3) Potentiation of the action of insulin by tolbutamide, but the clinical significance of this cannot yet be determined. This clinical interaction is of major importance as serious and even fatal hypoglycemic reactions have been reported (eg 4,5).

Oxyphenbutazone has an effect similar to that of phenylbutazone, and 300 mg oxyphenbutazone daily increases the half-life of tolbutamide. Studies in normal volunteers have shown that oxyphenbutazone considerably prolongs tolbutamide half-life (6). However, it has not been possible to find any reports of a clinically significant interaction between oxyphenbutazone and tolbutamide.

Phenazone 300 mg and aminophenazone 3000 mg were administered daily to diabetic patients on tolbutamide. No changes were found following phenazone and aminophenazone treatment (7). However, no definite conclusions can be drawn from this limited study.

Naproxen does not appear to exert any effects on tolbutamide metabolism or plasma concentration (8). Sulindac 400 mg/day for one week was administered to 12 diabetics stabilized on tolbutamide. No significant changes were noted in tolbutamide plasma concentrations or half-life (9).

Azapropazone was reported to interact with tolbutamide in a 77-year-old woman, who had been taking tolbutamide 500 mg bid and who arrived in a comatose state 5 days after starting azapropazone 300 mg tid. Her plasma glucose was 2 mmol/l and she recovered after 40 ml of 50 per cent glucose was given intravenously (10). Tolbutamide pharmacokinetics was studied in three healthy subjects who were given a 500 mg intravenous dose following azapropazone 900 mg daily for four days. The mean serum half-life of tolbutamide increased from 7.7 to 25.2 hours with a proportional decrease in plasma clearance (10). In 1984 there was another case report concerning azapropazone and tolbutamide describing hypoglycemia. The mechanism behind was thought to be caused by displacement of tolbutamide from protein binding sites, and it is recommended that azapropazone should be avoided in patients receiving tolbutamide (11).

A single dose of ibuprofen (60 mg/kg) was reported to antagonize the hypoglycemic action of tolbutamide (250 mg/kg per os) in non-diabetic rabbits (12). However, no data concerning any interaction with respect to ibuprofen and any sulfonylurea in humans could be found presently like as in earlier question to Drugline (13).

Salicylates are supposed to exert a beneficial hypoglycemic effect in some mild maturity-onset diabetics and hypoglycemia has been noted as a consequence in salicylate intoxication. The reason of the hypoglycemic effect seems to be explained by an increase in plasma insulin levels. In vitro studies have further shown that sodium salicylate displaces tolbutamide from plasma protein binding, thus increasing the concentration of unbound active tolbutamide (14,15).

Chlorpropamide and NSAID: The effect of phenylbutazone and ibuprofen during four weeks was studied in 22 well controlled type-2-diabetics on chlorpropamide. Subjects taking phenylbutazone 300 mg per day showed a reduction in fasting blood sugar, which became statistically significant during the 3rd and 4th weeks. Clinical hypoglycemia, however, was never observed. No significant reduction was seen in fasting blood sugar in subjects taking ibuprofen 1200 mg per day. There was no significant change in post lunch blood sugar values in either group (16). The ability of three gram sodium salicylate to enhance the blood glucose lowering action of 1200 mg chlorpropamide was confirmed in healthy male volunteers who received an oral test dose of 50 g glucose. Salicylate raised the plasma concentration of insulin, but did not alter the concentration of chlorpropamide (17).

Glibenclamide and NSAID: In a clinical trial of 20 NIDDM-patients on glibenclamid, it was found that the administration of 300 mg mofebutazone daily during three weeks did not result in any clinically relevant changes of blood glucose (18). Focusing on piroxicam and glibenclamide, it was found that piroxicam significantly reduced blood glucose level in diabetics, while no significant influence was recorded in rats and healthy volunteers (19). Tenoxicam 20 mg daily in combination with glibenclamide 2.5 mg did not affect the overall glycoregulation in healthy subjects (20,21).