(1601) Kidney NIRS Correlates Poorly With Renal Medulla Oxygen Concentration in Porcine Hemorrhage Model

Introduction: Near-infrared spectroscopy (NIRS) monitoring over the kidney to measure tissue oxygenation is used to assess acute kidney injury (AKI) risk in pediatric populations. Its utility in adults, however, remains less defined. NIRS monitoring is commonly used to measure oxygen saturation of tissue 1-3 cm below the skin. In adults (BMI < 35 kg/m2) undergoing cardiac surgery, measurements from NIRS sensors placed on the skin over the kidney predict postoperative AKI. However, the correlation between kidney NIRS (kNIRS) and renal tissue oximetry (rSO2) has never been validated. We hypothesized that kidney NIRS (kNIRS) correlates better with direct renal medulla oxygen concentration measurements than NIRS placed over the thigh (tNIRS).

Methods: Pigs (N=4) were anesthetized and subjected to hemorrhagic shock/hemodilution via stepwise removal of 25% of the animal's blood volume over 60 minutes. This was accomplished with 3 hemorrhage-hemodilution periods, each 20 minutes: blood was removed over 15 minutes followed by an isotonic crystalloid bolus (2 times the volume of blood removed) over 5 minutes. Shed blood was returned after 60-minute. We continuously monitored mean arterial pressure (MAP), renal blood flow (RBF), renal medulla oxygen concentration, and rSO2 with NIRS sensors placed on the skin superficial to the kidney and the thigh. Ultrasound guidance was used to confirm NIRS sensors placement and measure skin-to-kidney surface distance.

Results: The skin-to-kidney surface distance was < 2cm in all pigs. The averages of individual correlation coefficients were similar when kNIRS and tNIRS were compared to renal medullary oxygen concentration (r = .15 v.s .27; p=0.51), RBF (r = .57 vs. .58; p=0.99), and MAP (r = .71 vs .75; p = 0.39). There was a strong correlation between kNIRS and tNIRS measurements (r = .76; p < 0.0001).

Conclusions: In our sample of four pigs, we found that NIRS placed on the skin's surface over the kidney correlates poorly with renal medulla oxygen concentration and was not different from NIRS measurements over the thigh, even when the kidney was < 2cm from the skin surface.