Acute pancreatitis associated with different combination therapies in patients infected with human immunodeficiency virus.

STUDY OBJECTIVE: To assess the risk of acute pancreatitis in patients receiving various combinations of protease inhibitors, nucleoside reverse transcriptase inhibitors (NRTIs), and nonnucleoside reverse transcriptase inhibitors (NNRTIs) for treatment of human immunodeficiency virus (HIV) infection.

DESIGN: Retrospective cohort study.

DATA SOURCE: Ohio Medicaid claims database, January 1997-December 2002.

PATIENTS: Four thousand nine hundred seventy-two patients with HIV infection who had received at least one antiretroviral drug.

MEASUREMENTS AND MAIN RESULTS: Three combination regimens were evaluated: didanosine plus other antiretroviral agents, protease inhibitors plus NRTIs or NNRTIs, and NRTI combinations (no didanosine) or NRTIs plus NNRTIs. We used Cox proportional hazard regression and Kaplan-Meier plots to examine the risk for acute pancreatitis. We identified 159 (3.2%) cases of acute pancreatitis during the study period. For patients who were newly treated for HIV, the incidence of acute pancreatitis was 1.95/100 person-years. Half of these cases developed within 500 days of the start of drug therapy. Hazard ratios (HRs) for acute pancreatitis were 39-54% higher for nonwhite patients than Caucasians and 240-290% higher for symptomatic versus nonsymptomatic patients. Hazard ratios also were significantly associated with increased age, liver injuries (HR 2.94, 6.73), and cardiovascular diseases (HR 1.68, 2.36), respectively, for both newly treated and previously diagnosed patients with HIV. The risk for patients receiving either protease inhibitors plus an NRTI or an NNRTI, or NRTI plus NNRTI combinations was not significantly different from the risk associated with didanosine combination therapy (p>0.10).

CONCLUSION: The risk of acute pancreatitis was significantly associated with age, race, symptomatic HIV infection, and liver and cardiovascular diseases. However, risk did not differ significantly among patients with different antiretroviral regimens. Our results can be used by the medical community to enhance patient safety and minimize costly adverse drug reactions among patients with HIV infection.