Abstract Introduction The semi-automated compounding of intravenous chemotherapy is an evolving technology with previously identified benefits to patients and operator safety.[1] The introduction of a Gri-fill® 4 semi-automated unit to University Hospital Limerick (UHL) in 2022 marked the first use of semi-automated chemotherapy compounding in Ireland. To date, limited research exists regarding implementation of semi-automated units within Irish hospitals. Aim To compare dosing accuracy and preparation time of manually versus semi-automated compounded chemotherapy. To understand perceptions of pharmacy technicians regarding manual versus semi-automated compounding of chemotherapy. Methods Dose accuracy, defined as absolute percentage dose error versus the prescribed dose, was obtained gravimetrically for a total of 90 chemotherapy preparations. Preparations of carboplatin (22 manual, 22 semi-automated) and paclitaxel (23 manual, 23 semi-automated) were included. Dose accuracy via both procedures for each drug was compared statistically using the Mann-Whitney U-test, whereby p<0.05 denotes statistical significance. A dose accuracy cut-off, signalling need for repreparation, was >5%. Time in seconds for compounding via each procedure was recorded. An anonymous online survey consisting of both Likert-scale and free-text questions was disseminated to all ten pharmacy technicians who work in the UHL aseptic compounding unit (ACU). Likert-scale responses were analysed using descriptive statistics, while qualitative statements were analysed thematically. Results A statistically significant difference in dose accuracy distribution between manual (mean rank = 28.20) versus semi-automated (mean rank = 16.80) groups was found for the carboplatin preparations (p<0.05, U=367.5). Absolute percentage dose error tended to be higher using the manual procedure. No significant difference in dose accuracy distribution was observed between procedures for paclitaxel (p>0.05, U=285.0) (manual mean rank = 22.61, semi-automated mean rank = 24.39). Overall, zero semi-automated and 2 manual preparations exceeded the >5% dose accuracy limit. For both drugs, median preparation times were longer for the semi-automated (200.5, 199 seconds) versus the manual (116.5, 141 seconds) procedure. Nine out of ten pharmacy technicians responded to the survey. Over half (55%) found working with the semi-automated unit demanding. A higher percentage felt preparations compounded manually were safer than preparations from the semi-automated unit (100% versus 66%). Practical issues with the semi-automated unit, including lack of space within the isolator, staff resource allocation and reliance on machine calibration were highlighted. However, its capacity to produce large multiples of one preparation type was recognised. Conclusion This pilot study highlighted comparative (paclitaxel) or significantly improved (carboplatin) dosing accuracy of a semi-automated compounding unit versus a manual procedure. While the number of preparations analysed was limited and necessity for larger confirmatory studies across numerous ACUs is acknowledged, this data highlights capacity for semi-automated units to accurately prepare chemotherapy. The observed decreased repreparation requirements of the semi-automated procedure highlight potential for such units to aid sustainable chemotherapy production. However, to optimise implementation, future research should investigate solutions to concerns identified by pharmacy technicians and verify the accuracy results obtained. Only then can the benefits of such semi-automated units to both patients and staff be achieved. Reference 1. Seger AC, Churchill WW, Keohane CA, Belisle CD, Wong ST, Sylvester KW, Chesnick MA, Burdick E, Wien MF, Cotugno MC, Bates DW. Impact of robotic antineoplastic preparation on safety, workflow, and costs. Journal of Oncology Practice. 2012;8(6):344–9.
Read full abstract