Association of intensive morphine treatment and increased stroke incidence in prostate cancer patients: a population-based nested case­control study.

OBJECTIVES: We address the potential problem of stroke induced by morphine exposure by comparing the incidence of stroke in cancer patients treated with and without morphine. METHODS: We performed a population-based nested case­control retrospective analysis on the Longitudinal Health Insurance Database 2000 and Registry for Catastrophic Illness Patients of Taiwan. This study is based on a malignancy cohort of 31 611 patients without a history of stroke, and 1208 patients who subsequently developed stroke served as the stroke group. Four controls of matched age, sex, entry year and entry month for each case were selected from the malignancy cohort from the non-stroke group. We used logistic regression to estimate the odds ratios and 95% confidence intervals, and applied the multivariable model to control for age, sex, hypertension, diabetes, hyperlipidemia and cardiovascular disease. RESULTS: Cancer patients who received morphine had a 12% higher risk of developing stroke than non-morphine users. However, the difference was nonsignificant. A significant difference only appears in prostate cancer patients, where morphine users have a 3.02-fold (4.24- and 2.90-fold for hemorrhagic and ischemic strokes, respectively) higher risk of suffering from stroke. The risk increased significantly as the morphine dosage increased to 170 mg/year of treatment. CONCLUSIONS: Intense morphine treatment may be associated with an increased stroke incidence in patients with malignancy, and the association is particularly significant for prostate cancer patients.