A three-year study of a first-generation chemotherapy-compounding robot

Abstract

PURPOSE: Results of a performance evaluation of an automated system for compounding antineoplastic preparations are reported. METHODS: Three years after the pharmacy department of a hospital in Saudi Arabia installed an i.v.-compounding robot (CytoCare, Health Robotics), data captured by the pharmacy information system and the machine's integrated software were analyzed to assess the performance of the robot in terms of compounding accuracy, days of operation, and downtime. RESULTS: The robot was used to prepare 3.82%, 10.80%, and 13.79% of selected antineoplastics compounded in 2010, 2011, and 2012, respectively. The robot failed to meet the specified dose accuracy range of ±5% in compounding 3 of 337 chemotherapy preparations (0.9%) in 2010, 349 of 1516 preparations (23%) in 2011, and 460 of 2993 preparations (15%) in 2012. The robot was operational on 40%, 39%, and 61% of available workdays in 2010, 2011, and 2012, respectively. Robot throughput relative to the pharmacy's manual compounding process was low, with substantial medication waste resulting from dose preparation failures. Implementation challenges included workflow disruptions due to robot downtime, mechanical issues (e.g., robot arm-clamping failures), difficulty obtaining gravimetric data for some drugs, and the need to recalibrate the device to accept i.v. bags, syringes, and medication vials incompatible with manufacturer specifications. CONCLUSION: The introduction of a chemotherapy-compounding robot for preparation of patient-specific i.v. antineoplastic drugs had a limited efficiency impact in practice. This solution, with its numerous limitations and technical difficulties, is not yet mature enough for universal adoption.