The NICE classification for 'Ultra-radical (extensive) surgery for advanced ovarian cancer' guidance does not meaningfully predict postoperative complications: a cohort study

Abstract

Objective: To determine which descriptors of cytoreductive surgical extent in advanced ovarian cancer (AOC) best predict postoperative morbidity.

Design: Retrospective notes review.

Setting: A gynaecological cancer centre in the UK.

Population: Six hundred and eight women operated on for AOC over a period of 114 months at a tertiary cancer centre, between 16 August 2007 and 16 February 2017.

Methods: Outcome data were analysed by six approaches to classify the extent of surgery: standard/ultra-radical surgery; standard/radical/supra-radical surgery; presence/absence of gastrointestinal resections; low/intermediate/high surgical complexity score (SCS); presence of bowel anastomoses and/or diaphragmatic surgery; and the presence/absence of multiple bowel resections.

Main outcome measures: Major (grades 3-5) postoperative morbidity and mortality.

Results: Forty-three (7.1%) patients experienced major complications. Grade-5 complications occurred in six patients (1.0%). Patients who underwent multiple bowel resections had a relative risk (RR) of 7.73 (95% confidence interval, 95% CI 3.92-15.26), patients with a high SCS had an RR of 6.12 (95% CI 3.25-11.52), patients with diaphragmatic surgery and gastrointestinal anastomosis had an RR of 5.57 (95% CI 2.65-11.72), patients with 'any gastrointestinal resection' had an RR of 4.69 (95% CI 2.66-8.24), patients with ultra-radical surgery had an RR of 4.65 (95% CI 2.26-8.79), and patients with supra-radical surgery had an RR of 4.20 (95% CI 2.35-7.51) of grades 3-5 morbidity, compared with patients undergoing standard surgery as defined by the National Institute for Health and Care Excellence (NICE) in the UK. No significant difference was seen in the rate of major morbidity between standard (6/59, 10.2%) and ultra-radical (9/81, 11.1%) surgery within the cohort who had intermediate complex surgery (P > 0.05).

Conclusions: The numbers of procedures performed significantly correlate with major morbidity. The number of procedures performed better predicted major postoperative morbidity than the performance of certain 'high risk' procedures. We recommend using SCS to define a higher risk operation. NICE should re-evaluate the use of the term 'ultra-radical' surgery.

Tweetable abstract: Multiple bowel resection is the best predictor of morbidity and is more predictive than 'ultra-radical surgery'.

Keywords: NICE; Morbidity; ovarian cancer; surgery; ultra-radical.