P022: ACUTE KIDNEY INJURY (AKI) AND PANCREATECTOMY, A SINGLE-CENTER RETROSPECTIVE COHORT
Saurin Shah, MD1; Kai L Mongan, MS2; Jose Humanez, MD1; Stefan Braunecker, MD1; Paul Mongan, MD1; Amie L Hoefnagel1; 1Anesthesiology, University of Florida - Jacksonville; 2Northeast Ohio Medical University
Introduction/Background: Acute kidney injury (AKI) in the perioperative period is a common complication and is associated with increased length of stay, more complex management, and an increase in morbidity, and mortality. Intraoperative recommendations for avoiding AKI include optimizing hemodynamics, restoring the circulating volume, and avoiding diuretics and nephrotoxic agents.[1,2] Reported outcomes for postoperative KDIGO AKI in intrabdominal surgery ranges from 15-25% with no reports specific to pancreatic surgery, while the ASC-NSQIP AKI for pancreatectomy is 1.3%. We report a single-center experience with pancreatectomies for AKI, perioperative variables, and complications.
Methods: We retrospectively evaluated 200 consecutive pancreatectomy surgeries completed from 2018-21 at the University of Florida-Jacksonville with IRB approval. All data were extracted from the EPIC electronic health care record. The risk of AKI risk was assessed using the AKI risk index  and the MPOG risk model . AKI was defined using both the KDIGO and ACS-NSQIP criteria (Table). Data were analyzed using SPSS 27.
Results: Table delineates important data differences between patients with (n=20) and without (n=180) KDIGO defined AKI. Preoperative differences between the groups were related to gender, increased BMI, increased disease burden (Charlson Scale), and higher risk estimation (AKI and MPOG). Surgical procedures were equally distributed between groups, but the procedural time was longer in the AKI group. While crystalloid, colloid, and OR net fluid volumes were similar, there was increased blood loss, pRBC transfusions, and lower hemoglobin on POD 1 in the AKI cohort (median [IQR], 10.6 [9.3-11.6] vs. 8.5 [7.6-9.7], p<0.001). The incidence and duration of intraoperative hypotension, phenylephrine administration (median [IQR], 41 [18-189] vs. 48 [25-268] mcg/kg/min, p=0.27), and low urine output (<0.05ml/kg/hr, 15% vs. 30% p=0.09) were similar between groups. While the OR net fluid balance was similar in both groups, the AKI cohort had increased differences in net fluid balance through POD 3 with equilibration at day 7. In addition, the postoperative management, infections, and complications were higher in the AKI group with a longer LOS, discharge level of care, and death.
Discussion/Conclusion: While our observed incidence of AKI (10% KDIGO, 1.5% ACS-NSQIP) in this cohort was at or below expected, AKI was associated with preoperative measures of disease and risk, blood loss, transfusions, and hemoglobin levels, and not intraoperative variables under anesthetic control (volume administration, blood pressure, and vasopressor therapy). The AKI cohort did experience significantly more complicated postoperative management, more procedural interventions, a longer LOS, and was more likely to die. Future investigations should focus on the reduction of AKI and the association with other complications/critical interventions.
1. Update on Perioperative Acute Kidney Injury, A&A,2018, V127, p1236
2. Perioperative acute kidney injury. BJA,2015, V115, pii3
3. Preoperative Risk and the Association between Hypotension and Postoperative Acute Kidney Injury. Anesthesiology,2020; V132, p461
4. Development and Validation of an Acute Kidney Injury Risk Index for Patients Undergoing General Surgery. Anesthesiology,2009; V110, p505