Document resource
Introduction In traditional cardiac catheterisation laboratory staffing models in the United Kingdom, dedicated radiographers are utilised to oversee radiation dose optimisation. There has been no direct comparison of procedural radiation doses between a radiographer assisted and operator lead image acquisition model. We compared two centres in the Southwest, the first with a traditional radiographer assisted service and the second without radiographers, with image acquisition performed entirely by the operating physician. This study aimed to compare a patient centred radiation measure; the Dose Area Product (DAP), stratified across a range of coronary procedures.Methods We conducted a retrospective, non-inferiority analysis comparing radiation data from two centres between January 2023 and July 2020: Centre A (reference centre with radiographer-assisted acquisition) and Centre B (operator acquisition). To ensure relevant comparison, data were stratified into three categories based on procedural urgency: Elective Left Heart Catheterisation (LHC), Urgent PCI (NSTEMI), and Emergency PCI (STEMI).A non-inferiority margin (Δ) of 1.10 (representing a 10% increase in radiation dose) was pre-specified. Skewness analysis confirmed rightward skew of both data sets, with a small number of complex cases resulting in outlier high-dose exposure. As a result, Dose Area Product (DAP) values (Gy.cm2) underwent natural log-transformation for parametric testing. The primary endpoint was the Ratio of Geometric Means (RGM) with corresponding 95% Confidence Intervals (CI). Non-inferiority was confirmed if the upper bound of the 95% CI did not exceed the 1.10 threshold.Results 4,494 procedures were analysed (Centre A: n=1,565, Centre B: n=2,929, table 1). For emergency and urgent cases, the operator-only model demonstrated non-inferiority compared to the reference centre (figure 1). For Emergency PCI, the RGM was 0.69 (95% CI: 0.64–0.73; p < 0.0001), and for Urgent PCI, the RGM was 0.87 (95% CI: 0.80–0.94; p < 0.0001). In both cases, the 95% CI remained below 1.0, meeting the criteria for non-inferiority.For Elective LHC (excluding graft studies), the RGM was 1.03 (95% CI: 0.96–1.11) with the upper bound of the 95% CI exceeding the 1.10 margin.Conclusion In our two-centre analysis operator-only image acquisition was non-inferior to radiographer-assisted imaging in high-acuity settings, including emergency and urgent PCI. This finding suggests that direct operator control of imaging parameters is safe during time-sensitive interventions.The borderline results in the elective LHC cohort may be attributed to unmeasured differences in patient and procedural characteristics, between the two institutions. These co-variates were not explored in this preliminary analysis, limiting the generalisability of these findings and are planned for future analyses.These initial findings support the safety of an operator-only image acquisition model during time-sensitive invasive coronary procedures.Abstract 535 Figure 1Non-inferiority analysis, using pre-specified non-inferiority threshold of 1.10Abstract 535 Table 1Total procedures performed and Mean DAP. LHC: Left heart catheterisation, PCI: Percutaneous coronary intervention, DAP: Dose Area Product