Abstract
BACKGROUND: Vancomycin treats serious Gram-positive infections. In St George's Hospital (SGH) intensive care unit (ICU) settings, vancomycin is administered intravenously by continuous infusion. Steady-state serum concentrations are monitored daily with a 20–25 mg/L therapeutic target. Non-therapeutic concentrations are associated with patient harm and prolonged stay.(1) A service evaluation revealed variable adherence to/effectiveness of the vancomycin prescribing/administration/monitoring protocol. Electronic prescribing and medicine administration (ePMA) system interface issues may have contributed.(2) Consequently, multifaceted interventions were devised and piloted on General ICU. OBJECTIVES: To (i) improve combined protocol prescribing/administration/monitoring adherence; (ii) enhance therapeutic protocol dosing; and (iii) ascertain accuracy of patients’ paper drug administration (PDA) charts compared with the ePMA system. METHODS: The quality improvement project (QIP) was approved by SGH clinical governance/audit teams. Over a 9 month period (September 2021 to May 2022) system/person-focused interventions were implemented. Protocol dosing(2) was revised; introducing a >90 kg patient 2 g loading dose, new renal-function categories and an increased maintenance dose for creatinine clearance (CL(CR)) >90 mL/min. Protocol accessibility was increased via ePMA and CliniBee/Microguide app integration. Educational protocol presentations were incorporated into medical/nursing induction training. Vancomycin prescribing/administration/monitoring data for non-renal replacement patients during the intervention period, was extracted retrospectively from the ePMA system. This was compared with baseline informing ICU data collected July 2020 to July 2021.(2) Patient's drug administration accuracy data (PDA charts/ePMA system) was extracted retrospectively from January to May 2022 and analysed. RESULTS: Compared with baseline, the proportion of patients receiving per protocol prescribing/administration of vancomycin loading/maintenance doses with daily monitoring increased [39% (7/18) to 68% (15/22)]. Within 48 h 54% (7/13) of vancomycin serum concentrations in all patients were therapeutic, demonstrating a baseline increase of 21% (3/9 to 7/13). In per protocol treated patients, serum concentrations increased 16% (2/7 to 4/9) therapeutically, decreased 20% (3/7 to 2/9) supra-therapeutically and increased 4% (2/7 to 3/9) sub-therapeutically. Supra-therapeutic concentrations were associated with CL(CR) <50 mL/min. Sub-therapeutic concentrations were associated with CL(CR) >90 mL/min and obesity. Compared with the ePMA system, there was 38% (5/13) less PDA charts recording both loading/maintenance doses. Administration time differences >60 min were recorded for 38% (3/8) of loading and 31% (4/13) maintenance doses. DISCUSSION: Staff turnover periods were associated with decreased protocol compliance. Further education is required around prescribing/administration of standardized vancomycin infusion bags, with associated rate changes. Integration of pharmacists into daily Microbiology ward rounds may alleviate these issues. Higher 20 mg/kg loading doses for >100 kg patients and maintenance dose revisions may reduce non-therapeutic concentrations.(3) Multidirectional variation in vancomycin administration timings recorded between PDA charts/ePMA system, requires further investigation. Infusion-pump data may offer the most accurate administration time for calculating pharmacokinetic variables. CONCLUSIONS: Multifaceted interventions were successful at improving adherence to/effectiveness of the vancomycin protocol. Findings will inform QIP roll-out across all three SGH ICUs, which will incorporate infusion-pump data collection to facilitate pharmacokinetic modelling. This will inform local dosing strategies and research into patient variability.