Urinary hypoxia: an intraoperative marker of risk of cardiac surgery-associated acute kidney injury.

Background: Acute kidney injury (AKI) is common after cardiac surgery and profoundly affects postoperative mortality and morbidity. There are no validated methods to assess risk of AKI intraoperatively. Methods: We determined the association between postoperative AKI and intraoperative urinary oxygen tension (PO2), measured via a fiber optic probe in the tip of the urinary catheter, in 65 patients undergoing high-risk cardiac surgery requiring cardiopulmonary bypass (CPB). AKI was diagnosed by modified Kidney Disease: Improving Global Outcomes criteria. Results: Urinary PO2 fell during the operation, often reaching its nadir during rewarming or after weaning from CPB. Nadir urinary PO2 was lower in the 26 patients who developed AKI (mean ± SD, 8.9 ± 5.6 mmHg) than in the 39 patients who did not (14.9 ± 10.2 mmHg, P = 0.008). Patients who developed AKI had longer periods of urinary PO2 ≤15 and 10 mmHg than patients who did not. Odds of AKI increased when urinary PO2 fell to ≤10 mmHg {3.60 [95% confidence interval (CI) 1.27-10.21]} or ≤5 mmHg [3.60 (95% CI 1.04-12.42), P = 0.04] during the operation. When urinary PO2 fell to ≤15 mmHg, for more than or equal to the median duration for all patients (4.8 min/h surgery), the odds of AKI were 4.85 (95% CI 1.64-14.40), P = 0.004. The area under the receiver-operator curve for this parameter alone was 0.69, and was 0.89 when other variables with P ≤ 0.10 in univariable analysis were included in the model. Conclusion: Low urinary PO2 during adult cardiac surgery requiring CPB predicts AKI, so may identify patients in which intervention to improve renal oxygenation might reduce the risk of AKI.

Basal renal O2 consumption and the efficiency of O2 utilization for Na+ reabsorption.

We examined how the presence of a fixed level of basal renal O2 consumption (Vo2(basal); O2 used for processes independent of Na(+) transport) confounds the utility of the ratio of Na(+) reabsorption (TNa(+)) to total renal Vo2 (Vo2(total)) as an index of the efficiency of O2 utilization for TNa(+). We performed a systematic review and additional experiments in anesthetized rabbits to obtain the best possible estimate of the fractional contribution of Vo2(basal) to Vo2(total) under physiological conditions (basal percent renal Vo2). Estimates of basal percent renal Vo2 from 24 studies varied from 0% to 81.5%. Basal percent renal Vo2 varied with the fractional excretion of Na(+) (FENa(+)) in the 14 studies in which FENa(+) was measured under control conditions. Linear regression analysis predicted a basal percent renal Vo2 of 12.7-16.5% when FENa(+) = 1% (r(2) = 0.48, P = 0.001). Experimentally induced changes in TNa(+) altered TNa(+)/Vo2(total) in a manner consistent with theoretical predictions. We conclude that, because Vo2(basal) represents a significant proportion of Vo2(total), TNa(+)/Vo2(total) can change markedly when TNa(+) itself changes. Therefore, caution should be taken when TNa(+)/Vo2(total) is interpreted as a measure of the efficiency of O2 utilization for TNa(+), particularly under experimental conditions where TNa(+) or Vo2(total) changes.