A nested case-control study on mortality in users of ibopamine.

Abstract

A recent interim analysis of the PRIME II placebo-controlled study showed a significantly higher mortality in the group treated with ibopamine than in the control group. The objective was to study mortality in patients on ibopamine, and to assess risk factors for death. All 2147 drug-dispensing outlets (DDO) in the Netherlands were asked to provide a printout of the complete medication history of users of ibopamine. A reaction was received from 92% of the DDO. From the 14,024 identified former or current users of ibopamine, a sample of 3148 patients (22%) was enrolled in the follow-up study. All general practitioners (GP) of these patients received an enquiry pertaining to the vital status of their patient, cause of death, primary cause and NYHA classification of heart failure, echo- and electrocardiographic data, serum creatinine, admissions and the effects of ibopamine. Cases were defined as patients who died during the follow-up period which ended on the day of return of the questionnaire or the day of decease (index date). Two random controls were obtained for each case from the non-deceased patients at the index date. The design was a follow-up study with risk factor assessment in a nested case-control design. Questionnaires were returned regarding almost 70% of the sample. Mortality in this group was 25%. A case-control analysis was performed with the first 104 cases and 208 random controls. Patients with NYHA class IV had a 3-times increased risk of dying. In patients with a serum-creatinine level in the highest quartile the risk of dying was increased threefold. Higher doses of ibopamine seemed to have a protective effect. Significantly more cases than controls used amiodarone. Also, opioids were used more often, which may be related to their use in terminal cardiac failure. NYHA classification and serum-creatinine levels were independent risk factors for death in patients with heart failure on ibopamine. Although there were increased risk estimates for current use of ibopamine and amiodarone, these did not reach statistical significance. This may be related, however, to the fact that this analysis was restricted to the first 20% of cases.