Studies estimating risks following cardiac surgery for patients receiving kidney replacement therapy have been limited by the size and generalizability of those cohorts. This study used data linked between registries to estimate short-term postoperative outcomes for large patient cohorts receiving kidney replacement therapy at the time of surgery. This population-based observational cohort study included adult patients who had undergone cardiac surgery in Australia between 2010 and 2019. Patient data were linked with a kidney replacement therapy registry to accurately identify cohorts and extract relevant data. Multivariable logistic regression estimated risk of operative (30-day) mortality and other postoperative outcomes for long-term dialysis and functioning kidney transplant cohorts compared with each other and the general cardiac surgical population. Of 114,496 surgeries, 1,241 were for patients receiving long-term dialysis and 298 for those with a kidney transplant. The mortality rate was highest for valve-with-coronary artery bypass grafting for dialysis (18.78 per 100 surgeries (95% CI 13.37,25.25) and transplant patients (14.00 [5.82,26.74]). Dialysis patients had higher adjusted odds of mortality (odds ratio [OR] 4.17 [95% CI 3.31,5.25]) and all other measured outcomes than the general population. Kidney transplant recipients had similarly elevated odds of mortality (OR 3.52 [95% CI 2.16,5.72]). Despite the younger age of dialysis and transplant cohorts at surgery, operative mortality rates were higher, and for valve-with-coronary artery bypass grafting were 3.7- to 5-fold those of the general population. Dialysis patients were a high risk for cardiac surgery, and the prognosis for kidney transplant recipients was similarly poor.