P2.Q323: NSQIP-Pediatric Analysis of Pancreatic Resection Outcomes in Children

Background: Pediatric pancreatic operations are uncommon. In this study, we aim to utilize a large database to assess postoperative outcomes of pancreatic resections in pediatric patients and the factors associated with early postoperative adverse events.

Methods: Patients (age≤18) undergoing pancreatectomy operations for pancreatic diagnoses between 2012-2021 were identified using National Surgical Quality Improvement Program-Pediatric. Preoperative demographics and comorbidities were compared between patients with and without postoperative adverse events (complications/readmissions/reoperations/death). Postoperative outcomes were compared across operation types. Multivariable logistic regression was used to identify factors associated with postoperative adverse events.

Results: There were 336 patients identified in the database, of which 260 (77.4%) underwent distal pancreatectomy, 5 (1.5%) underwent pancreatoduodenectomy, 3 (0.9%) underwent total pancreatectomy, 12 (3.6%) underwent total pancreatectomy with islet autotransplantation, 54 (16.1%) underwent enucleation, and 2 (0.6%) underwent debridement. Indications for operation included hypoglycemia (n=95, 28.3%), pancreatic neoplasms and cysts (n=134, 39.9%), pancreatitis and pseudocysts (n=36, 10.7%), trauma (n=12, 3.6%), and other pancreatic diagnoses (n=59, 17.6%). Overall, 16.4% of patients were being treated for cancer while this rate was 80% in patients undergoing pancreatoduodenectomy (p=0.007). Median length of stay was 7 days (5-15). In the 30-day postoperative period, 31 (9.2%) patients were readmitted, 20 (6.0%) patients underwent reoperation, 85 (25.3%) patients had complications, and there was one (0.3%) mortality. Overall, 109 (32.4%) patients had postoperative adverse events (Table 1). Patients with adverse events had lower median age (10.7 vs 11.9 years, p=0.018), longer median operative time (225 vs 185 minutes, p=0.013), and higher percentage of contaminated or dirty/infected wound classification (11% vs 2.6%, p=0.001). Furthermore, patients were more likely to have an adverse event if they were operated between 2012-2016 compared to 2017-2021 (39.4% vs 27.9%, p=0.029). Patients with adverse events more frequently were preterm (15.6% vs 6.2%, p=0.011), had ASA classification ≥Class III (68.8% vs 56.8%, p=0.035), and had preoperative comorbidities such as respiratory, gastrointestinal, hematologic, cardiac comorbidities (45% vs 22%, p< 0.001). They also more frequently required preoperative nutritional support (41.3% vs 12.8%, p< 0.001), steroid usage (6.4% vs 3.1%, p=0.004), and blood transfusions < 48 hours prior to surgery (8.3% vs 2.2%, p=0.016). Odds of any adverse event were independently higher in those with preoperative comorbidities (aOR: 1.97, 95% CI: 1.05-3.69) and need for preoperative nutritional support (aOR: 4.4, 95% CI: 2.15-9.3). Patients undergoing pancreatectomies between the ages 3-5 years had lower odds of adverse events (aOR: 0.12, 95% CI: 0.02, 0.7). Other factors such as race, diagnosis, operation type, ASA classification, and operation year were not associated with postoperative adverse events [Table 2].

Conclusion: In pediatric patients undergoing pancreatectomy operations, preoperative comorbidities and need for nutritional support are independently associated with higher odds of adverse events. Multidisciplinary preoperative optimization should be considered in these high-risk groups.