Statistical analysis
Statistical analysis was performed using GraphPad Prism 5.03 (GraphPad Software, Inc., San Diego, CA), and R 3.4.4 (R Core Team, Vienna, Austria). Propensity matching was performed to identify a two to one control population for the patients who experienced a stroke. This was performed using the ‘nearest-neighbour’ method whereby a control patient whose propensity score is closest to that of a patient returning to theatre is identified. Patients were matched on: age, sex, LV function, BMI, operation priority, operation category and logistic EuroSCORE. If multiple control patients have propensity scores that are equally close, one of these control subjects is selected at random. Standardised difference of means was calculated for both continuous and categorical variables in order to ensure that the frequency of a variable was equally balanced between the mini-bypass and matched populations.
The Kaplan-Meier method was used to plot the patient survival rates, with the log-rank (Mantel-Cox) test used to compare groups. Univariate analyses were performed. For comparison of groups, continuous variables were analysed with the Mann-Whitney U test if not normally distributed and with the Students t-test if normally distributed. Categorical variables were analysed with Fisher’s exact test. p < 0.05 was considered statistically significant.