SUN-185 EARLY PREDICTION OF CARDIAC SURGERY ASSOCIATED ACUTE KIDNEY INJURY BY MEASUREMENT OF URINARY OXYGEN TENSION IN THE INTENSIVE CARE UNIT

Abstract

Continuous measurement of bladder urinary oxygen tension (PO2) may provide a real-time surrogate measure of oxygenation in the renal medulla. Therefore, urinary PO2 may be a useful tool for early prediction of acute kidney injury (AKI) and potentially also the management of renal oxygenation in patients with AKI. We examined the relationship between urinary PO2, measured in the intensive care unit (ICU), and development of AKI in patients who had undergone cardiac surgery requiring cardiopulmonary bypass (CPB). In 74 patients undergoing on-pump cardiac surgery, bladder urinary PO2 was measured continuously in the ICU using a fibre optic probe placed in the patient’s bladder catheter. AKI was diagnosed by modified ‘Kidney Disease: Improving Global Outcomes’ (KDIGO) criteria, excluding the urine flow criterion. Twenty-eight out of 74 patients (37.8%) developed post-operative AKI (18 Stage 1, 7 Stage 2, and 3 Stage 3). The median time to first diagnosis of AKI, using serum creatinine, was 16.9 hours after entry into the ICU (interquartile range (IQR): 13, 32 h). The median duration of monitoring of urinary PO2 was 37.8 hours (IQR: 24.0, 47.3 h). During the first 3 hours after arrival in the ICU, overall mean urinary PO2 (AKI: 19.4 ± 9.3 vs. non-AKI: 25.3 ± 9.7 mmHg, p = 0.01), and the nadir (lowest recorded) urinary PO2 (AKI: 12.1 ± 6.8 vs. non-AKI: 16.0 ± 5.7 mmHg, p = 0.01) were lower in patients who developed AKI than those who did not. Furthermore, within the first 3 hours of entry to the ICU, 17 of the 28 patients (61%) who developed AKI had urinary PO2 less than or equal to 15 mmHg while this threshold was only reached in 16 of the 46 patients (35%) who did not develop AKI . Patients who developed AKI also experienced longer periods with urinary PO2 less than or equal to 15 mmHg (median time of 10.9 min per hour of measurement) than those who did not (median 0.0 min per hour), p = 0.01. If urinary PO2 fell to 10 mmHg or less at any time during the first 3 hours of the patient’s ICU stay, the odds of developing AKI were 5.6 times greater (95% CI 1.9 – 16.6; p = 0.002) than if it remained greater than 10 mmHg. The area under the receiver-operator curve for this parameter was 0.67 (95% CI 0.54 – 0.81; p = 0.01). Similar observations were made when urinary PO2 was considered over the first 6 hours, or the first 12 hours, after ICU admission, as well as for the entire duration of available monitoring of urinary PO2 in the ICU. In patients who have undergone cardiac surgery requiring CPB, increased risk of AKI can be detected within 3 hours after entry into the ICU, by continuous measurement of bladder urinary PO2. Thus, urinary PO2 has the potential to be a real-time index of risk of AKI in these patients.