Muhammad A. Khan, Chameen Samarawickrama
Purpose: To assess the effect that two recently-developed cataract surgery risk allocation systems have on intraoperative complications of cataract surgery and compare these systems to allocation by clinician judgement.
Methods: Data was retrospectively collected for all cataract surgical patients at West-mead Hospital between 2017 and 2020. Patients’ risk factors were used to score them using the United Kingdom Cataract Complexity Scoring System (UKCCSS) and the New Zealand Cataract Risk Stratification System (NZCRSS). These systems were then compared to the classification of a senior surgeon, described as clinical judgement (CJ).
Results: A total of 3245 eyes had adequate data for risk stratification. CJ, the UKCCSS and NZCRSS categorised 34.8%, 10.1% and 6.5% as high-risk, respectively. The intra-operative complication rate for the high-risk group (15.0-17.8%) was twice that of their low-risk group (6.1-8.8%) for all three systems. All three systems had low sensitivities for intra-operative complications, with CJ’s sensitivity (56.7%) being considerably higher than the UKCCSS (19.5%) and NZCRSS (11.1%). However, CJ had poor specificity for detecting intra-operative complications (56.7%) compared to the UKCCSS (90.1%) and NZCRSS (93.4%).
Conclusion: This is the first study to externally validate the UKCCSS and NZCRSS as risk stratification systems for cataract surgery. CJ’s relatively high sensitivity led to many more patients being classified as high-risk, but yielded an almost equivalent intra-operative complication rate to the two systems. Thus, the two systems should be used as strong adjuncts to CJ for allocating cataract surgery cases to trainees, thereby minimising overcalling of risk or compromising safety.