[Morphine tablets for chronic non-tumor-induced pain. Which factors modify the success or failure of a long-term therapy?].

In a prospective study 60 patients receiving morphine for treatment of mostly neuropathic and musculoskeletal pain of non-malignant origin were investigated. The aim of the study was to determine the characteristics of responders to morphine therapy and the frequency and severity of side effects. METHODS. Eligible patients suffered from chronic pain that had not been relieved despite all current therapy. All of them received controlled-release morphine tablets. Dose was increased in case of insufficient pain relief. Before morphine treatment and at the follow-ups pain intensity was rated on a numeric analogous scale, analgesia and side effects on a four-stage verbal rating scale. Intake of the prescribed and other drugs was randomly controlled by urine analysis. The patients were divided in three groups retrospectively: group I, non-responders who ceased the morphine therapy within 1 month due to weak analgesia or severe side effects; group II, patients in whom the therapy was stopped within the following months despite persisting pain; group III, patients who continued therapy. STATISTICS. ANOVA, chi-squared test, non-parametric tests (Kruskal-Wallis, Wilcoxon). Results were accepted as significant at p < 0.05. RESULTS. Twenty-three patients were non-responders. Fourteen other patients stopped the therapy after initial response because the side effects exceeded the benefit of analgesia (group II). Only 10 of the remaining 23 responders had good analgesia and minor side effects during the observation time. Constipation, despite prophylactic laxatives, and nausea were the most frequently reported events causing cessation of therapy. Analgesia and side effects in groups II and III were constant during the observation time of 241 (36-1486) days. In groups II and III, 43% of the patients needed an increase of the morphine dosage during therapy. The initially sufficient morphine dosage was significantly higher in group II than in group III; only one patient needed more than 90 mg/day in group III. Urine screening tests in 45 patients disclosed that 18 patients concealed the intake of other opioids or psychotropic drugs, mostly benzodiazepines. CONCLUSIONS. The study showed many problems with the patients' compliance and acceptance of oral morphine due to side effects and lack of analgesia. The discrepancy from other, positive reports might be explained by the extremely selected patients of this study and by the fact that not only responders were included for evaluation. Patients' compliance seems to have been overestimated in previous studies because no urinary controls were taken.