Severe Hypoglycemia From Clarithromycin-Sulfonylurea Drug Interaction

An 82-year-old white male with a 4-month history of type 2 diabetes was treated with diet and glyburide 5 mg daily. His blood glucose readings were in the low 100s (mg/dl). His comorbidities included atherosclerotic heart disease, hypertension, mild chronic renal failure with a creatinine of 1.6 mg/dl, emphysema, and bladder cancer. In a local emergency department (ED), he was diagnosed with “bronchitis” and prescribed clarithromycin, 1,000 mg daily. He became unresponsive with an Accucheck reading of 24 mg/dl, 48 h after starting clarithromycin. Administration of intravenous glucose in the ED resolved the problem. However, 12 h later, he again was found unresponsive. A chemstrip glucose test, administered by the paramedics, was in the 0–30 mg/dl range. Intravenous glucose resolved the problem a second time. There were no further episodes after stopping the sulfonylurea.

A 72-year-old white male with type 2 diabetes of >10 years’ duration was treated with diet and glipizide 15 mg daily. His most recently measured HbA_1c was 6.1%. His comorbidities included atherosclerotic heart disease, hypertension, and chronic renal failure with creatinine in the mid-3s (mg/dl). In a local urgent care center he was diagnosed with “bronchitis” and prescribed clarithromycin 1,000 mg daily. He was brought to the emergency department (ED) in a stupor 48 h after starting clarithromycin. His Accucheck reading was 20. Admission to the hospital and continuous intravenous glucose resolved the stupor. There were no further episodes after ceasing sulfonylurea therapy.

Common features of both cases included type 2 diabetes treated with diet and a sulfonylurea, age >70 years, male sex, and renal insufficiency. Common comorbidities included hypertension, atherosclerotic heart disease, and mildly low albumin levels. Both men were given identical clarithromycin dosages, and both experienced profound mental status changes 48 h after initiating the antibiotic.

Drug-induced hypoglycemia in elderly patients with type 2 diabetes is a major clinical concern. Several reviews have addressed this problem, and three general mechanisms have been implicated. First, hypoglycemia can be induced by a single hypoglycemic agent such as a sulfonylurea. Second, two or more hypoglycemic drugs can induce hypoglycemia. Specific examples known to cause hypoglycemia include a sulfonylurea plus insulin, a sulfonylurea and salicylates, and a sulfonylurea or insulin mixed with alcohol. Third, multiple drug-drug interactions have been reported to potentiate the effect of sulfonylureas. These include anti-inflammatory agents, sulfa antibiotics, bishydroxycoumarin, and antidepressants. Other agents such as propranolol and tetracyclines have been reported to potentiate the hypoglycemic effects of insulin (1).

All the sulfonylurea agents stimulate insulin release from pancreatic islet cells. The sulfonylurea agents with the longest half-lives cause the most problems and risk of hypoglycemia. Chlorpropamide has the longest half-life and has been reported to cause substantial hypoglycemia. Of the newer second-generation agents, glyburide has been reported to cause hypoglycemia more often than glipizide. Sulfonylurea-induced hypoglycemia can be particularly problematic in patients with impaired renal or hepatic function (2).

Various risk factors have been analyzed when looking at sulfonylurea-induced hypoglycemia. These include age >60 years, renal dysfunction, alcohol ingestion, sepsis, intentional overdose, and liver cirrhosis. The most common cause for sulfonylurea-induced hypoglycemia is an interaction between two medications (2).

Both glyburide and glipizide are well absorbed and are 90–99% protein bound. Glyburide and glipizide are extensively metabolized by the cytochrome P450 system in the liver—the principal enzyme seems to be CYP2C9 (3). Glyburide has weakly active metabolites, whereas the metabolites of glipizide are inactive. Glyburide is ∼50% excreted through feces and the remainder is renally excreted. Reduced renal function delays clearance of the parent drug and its metabolites. Glipizide and its metabolites are predominately excreted in the urine with less enteric excretion than glyburide. (4) The half-life of glipizide is ∼7 h, and the half-life of glyburide is ∼10 h. Both have durations of action between 12 and 24 h in the normal host.

Clarithromycin is well absorbed from the gastrointestinal tract. It is 40–70% protein bound and has first-pass hepatic metabolism, being converted to 14-OH clarithromycin. Clarithromycin inhibits intestinal p-glycoprotein as well as cytochrome CYP3A4.

Mechanisms of drug-drug interactions include one drug binding another, displacement from protein binding sites, alteration of drug metabolism, or alteration of drug excretion. Alterations of drug metabolism account for the majority of clinically relevant interactions (5). In these cases, we postulate that clarithromycin may have displaced glipizide and glyburide from protein binding sites, thereby increasing the unbound, or free, portion of the drug. It is unlikely that clarithromycin reduced metabolism of glipizide and glyburide because of the isoenzymes involved with drug metabolism. Decreased renal clearance may have also played a role in these cases. Although clarithromycin inhibits intestinal p-glycoprotein, we do not think this was the mechanism in these two cases, as secretion of glipizide or glyburide into the gut is not a major route of clearance of these two agents.

We conclude that clarithromycin should be cautiously prescribed to type 2 diabetic patients with mild renal impairment who are taking sulfonylurea medications.