A Drug Utility Analysis of the Withdrawal of the Propoxyphene-acetaminophen Combination from a Teaching Hospital in Taiwan

Purpose: In March 2011, Taiwan's Food and Drug Administration announced license withdrawal of propoxyphene-containing products in response to the finding of propoxyphene-related fatal cardiotoxicity. This study was conducted to review the usage pattern of propoxyphene and acetaminophen combination dosage form (PA) and to assess the clinical impact in association to the withdrawal. Methods: This retrospective observational study was conducted in an 800-bed regional teaching hospital in northern Taiwan accredited by the Joint Institutional Review Board (JIRB). Study participants were outpatients older than 18 years for whom PA was prescribed between September 2010 and March 2011. Data collection was done through review of medical charts and electronic medical records from the health informatics system. Our primary outcomes were the usage pattern of PA and the efficacy of alternative analgesic agents used in replacement before and after the withdrawal of PA as an analgesic agent for pain management. Data analysis was done using descriptive statistics and Wilcoxon signed-rank test. Results: PA was mainly prescribed for pain control in patients with diseases of the musculoskeletal system and connective tissue (60.8%, n = 463) and neoplasms (16.0%, n = 122). Of the 761 included study participants, most of them were short-term users (82.1%; n = 625) and 17.9% (n = 136) were chronic users of PA. Dosages prescribed were within the recommended dosing range. No cardiovascular events were reported due to PA. The most frequently prescribed alternative analgesic agents in PA replacement were nonsteroidal antiinflammatory drugs (NSAIDs) (38.7%; n = 63), acetaminophen (22.1%; n = 36), and nonopioid transdermal analgesic agents (14.7%; n = 24) in short-term users. For chronic users, the most frequently prescribed analgesic agent was acetaminophen (35.2%; n = 25), followed by NSAIDs (31.0%; n = 22) and tramadol (11.3%; n = 8). There was no significant difference between the mean numerical rating score with prescription of PA and prescription of alternative analgesic agents in replacement of PA for pain management. Conclusion: PA was mainly prescribed for the management of mild to moderate musculoskeletal- and neoplasm-associated pain. Acetaminophen and NSAIDs were used most often in replacement of PA. There was no significant clinical effect associated with formulary withdrawal of PA because it was mainly prescribed for short-term use and other comparable alternative agents were available.