Objective: Female sex is considered a significant risk in cardiac surgery and is included in the majority of scores for risk assessment. However, the evidence is controversial and older women undergoing cardiac surgery have not specifically been investigated. We assessed the influence of female sex on surgical risk (30-day mortality) in a secondary analysis of the GOPCABE trial (German Off-Pump Coronary Artery Bypass grafting in the Elderly (GOPCABE) trial, comparing on- to off-pump) and also evaluated its impact on risk prediction from commonly used risk scores.
Methods: We performed logistic regression analyses on the GOPCABE trial population, where patients were randomized to either on- or off-pump CABG. The study was performed in 12 cardiac surgery centers in Germany and analyzed 2394 patients having undergone CABG at age ≥75 years (1187 on-pump, 1207 off-pump). Of the 2394 patients, 755 (32%) were women. The logistic EuroSCORE and the German KCH score were calculated as expected (E) mortality and values were compared to observed (O) 30-day mortality (O/E ratio).
Results: There was no difference in mortality or major cardiovascular adverse events after 30 days between men and women for both on- and off-pump CABG (men: on- vs. off-pump OR = 0.90, 95%-CI: [0.63;1.27]; women: on- vs. off-pump OR = 1.07, 95%-CI: [0.62;1.87]). Therefore, groups were combined for further analyses. Both men and women had considerable and similar comorbidities. Expected mortality was significantly higher for women than for men (logistic EuroSCORE: 8.88±6.71% vs. 7.99±6.69%, p = 0.003; KCH score: 4.42±3.97% vs. 3.57±3.65%, p = 0.001). However, observed mortality rates (O) tended to be even lower in women (2.1% vs. 3.0%). The O/E ratio was closer to 1 in men than in women (0.84 vs. 0.47). Excluding female sex from the risk models increased O/E ratio to 0.69.
Conclusions: Female sex is not a risk factor in coronary bypass surgery in the GOPCABE population. The result is the same for on- and off-pump surgery. Since female sex is a component of most risk scores, the findings may identify a potential inaccuracy in current surgical risk assessment, specifically for elderly women.
Trial registration: Clinicaltrials.gov GOPCABE trial No. NCT00719667.