Background: Acute renal failure (ARF) is a recognised complication following cardiac surgery, but the incidence varies widely in the published literature and there are no Australian data available to help predict the risks of ARF in patients with pre-existing renal disease.
Aim: To determine the incidence, outcome and risk factors for ARF following cardiac surgery.
Methods: A retrospective case control analysis of 903 consecutive patients who had cardiac surgery (795 CABG, 68 valve/septal surgery, 40 combined valve/CABG) in 1992-93. ARF was defined as doubling of serum creatinine concentration (Cr) to > 0.13 mmol/L if serum Cr was < or = 0.13 mmol/L pre-operatively, or else a rise in serum Cr of > or = 0.10 mmol/L after cardiac surgery. For each subject with ARF, two case control subjects were matched for date of surgery, surgeon, age, sex, type of surgery and pre-operative serum Cr to permit analysis of the influence of pre-operative factors (hypertension, diabetes mellitus, left ventricular systolic dysfunction) and for the comparison of cardiopulmonary bypass time upon the development of ARF. Subsidiary endpoints were mortality, need for dialysis and length of hospital stay.
Results: ARF developed in only 1.1% of patients with 'normal' pre-operative renal function (creatinine < or = 0.13 mmol/L) and none required dialysis. ARF developed in 16% of those with impaired pre-operative renal function, 20% of whom required dialysis. Mortality from ARF was 13%. The risk of ARF rose from 10.4% in those with pre-operative serum Cr 0.14-0.20 mmol/L to 36.8% if the serum Cr was > 0.20 mmol/L (p < 0.01). Mortality was higher (4.2% vs 0.7%, p < 0.01) and length of hospital stay longer (14.5 vs nine days [median], p < 0.001) in those with impaired pre-operative renal function. ARF was more likely in those over 65 years, if valve surgery was included and where there was prolonged cardiopulmonary bypass time.
Conclusions: These data confirm that ARF following cardiac surgery is uncommon without pre-operative impairment of renal function but currently carries a mortality rate of 13%. Impaired renal function alone is associated with higher mortality and prolonged hospital stay. Studies to prevent ARF in this setting should focus on the high risk subsets described in this study.