People who take both the antidepressant Paxil and the cholesterol lowering drug Pravachol may be at risk for a sharp rise in blood sugar. Researchers found an average blood sugar increase of 19 units (mg/dl) among people taking the drug combination. The increase in diabetics was 48 units.

Neither drug taken alone is known to affect blood sugar.

The researchers estimate that between 500,000 and 1 million people in the country may be taking both drugs. Between 13 and 15 million people have prescriptions for at least one of them.

Many drug combinations can produce side effects that don't occur when one of the drugs is taken alone. Which combinations will do so are difficult to predict and these effects usually only become known well after the drugs have been marketed and taken by millions of people. The researchers were interested in finding out whether any drug pairs are unknowingly elevating blood sugar.

They started by looking at symptoms caused by drugs known to elevate blood sugar. Many of these are non-specific, such as fatigue and fever. By doing so, they were able to construct a list of symptoms, a fingerprint suggesting that a person might, without knowing, be suffering from increased blood sugar.

They then looked through the FDA's Adverse Event Reporting System (AERS). This is a database where physicians report adverse events and side effects experienced by patients, along with what drugs the patients are taking. Looking for their symptom fingerprint in patients taking pairs of drugs not known to increase blood sugar, they found four pairs of drugs that gave a good match. They focused on the Paxil/Pravachol pair because these drugs are so commonly prescribed.

There was no information in AERS indicating that any of these patients had elevated blood sugar. To find out whether the combination of Paxil and Pravachol actually raised blood sugar, the researchers had to turn to another source of information. They looked at the electronic medical records from patients in three separate medical institutions to see if patients taking the two drugs showed changes in blood sugar.

The records showed that 135 non-diabetic patients had an average rise in blood sugar of 19 mg/dl after they had begun taking the two drugs. And 104 patients with diabetes had an average increase in their blood sugar of 48 mg/dl after they had begun taking both drugs.

The researchers then conducted a study in mice to see if the two drugs taken together would cause a rise in blood sugar. The mice were fed a high-fat, high-calorie diet for several weeks. This caused the mice to become pre-diabetic — have a higher than normal blood sugar level and show resistance to insulin. These mice were then given Paxil and Pravachol, either singly or in combination.

Neither drug alone increased blood glucose. But when both were taken together for three weeks, blood glucose increased significantly — from about 128 mg/dl to 193 mg/dl.

The side effects produced by drug combinations do eventually come to light. But sooner would be better. The researchers are continuing to look for information suggesting that other drug combinations are causing previously unknown side effects.

The study was a joint venture involving Stanford University School of Medicine, Vanderbilt University and Harvard Medical School.

An ahead-of-print article detailing the study was published online by Clinical Pharmacology & Therapeutics on May 25, 2011. This article will also appear in a future print edition of the journal.