Defining Postoperative Weight Change Following Pancreatectomy: Factors Associated With Distinct And Dynamic Weight Trajectories

Presenter: Maxwell Trudeau BS | University of Pennsylvania Background: Weight change offers the simplest, global indication of a patients’ surgical recovery, and is aligned with long-term quality of life. Yet, no studies have thoroughly investigated postoperative weight dynamics following pancreatectomy, with wide disparity amongst surgeons’ perceptions of this concept. This study aims to define postoperative weight change dynamics following pancreatectomy and determine factors associated with optimal and poor weight trajectories. Methods: From 2004-2019, 1,156 proximal (PD; 66%) and distal (DP; 34%) pancreatectomies were performed at a single institution for benign and malignant pathologies. Patient weights were acquired preoperatively and at postoperative months 1, 3 and 12. Optimal (Top Quartile, weight restoration) and poor (Bottom Quartile, persistent weight loss) postoperative weight cohorts were identified at 1-year. Multivariable analyses (MVA) were used to identify pre-, intra- and postoperative factors associated with these particular courses. Results: The median percentage weight change (PWC) 1-year post-pancreatectomy was -7% (IQR: -2% to -13%), with similar median PWC between PD (-8.2%) and DP (-4.4%, p=0.568). Median weight loss primarily occurs within the first month and then plateaus thereafter. Just 19.4% of patients obtained their original weight within one year. No factors were significantly associated with postoperative PWC as a continuous variable, including: elderly age [>65] (p=0.495), obesity (p=0.409), preoperative diabetes or exocrine insufficiency (p=0.536; p=0.883, respectively), jaundice at presentation (p=0.492), malignancy (p=0.522), pre- or postoperative TPN/TEN (p=0.890; p=0.509, respectively), severe complications (p=0.627), adjuvant therapy (p=0.524), and tumor recurrence (p=0.368). Subsequently, weight restoration and persistent weight loss cohorts were analyzed (Figure 1). MVA revealed the independent factors associated with weight restoration were: non-elderly age (OR 1.73), non-obesity (OR 2.56), jaundice (OR 1.63), DP [reference: PD] (OR 2.34), not experiencing 30-day readmission (OR 1.61), and lack of tumor recurrence (OR 2.53). Conversely, persistent weight loss was associated with: elderly age (OR 1.51), obesity (OR 2.36), PD (OR 2.02), operative time>325minutes (OR 1.65), experiencing any complication (OR 1.68), receipt of adjuvant therapy (OR 1.44), and experiencing tumor recurrence (OR 5.46). Overall, 33% of the persistent weight loss cohort experienced tumor recurrence. Subset analysis of pancreatic adenocarcinoma patients (N=424, 37%) revealed factors associated with weight restoration to be: non-obesity (OR 2.95), DP (OR 1.74), estimated blood loss < 400 (OR 1.71), not experiencing a severe complication (OR 5.73), and lack of recurrence (OR 2.32). For all patients, weight restoration was associated with longer median overall survival (31 vs. 18 months, p=0.039), while patients with persistent weight loss lived a median of 16 months. Conclusion: These data define postoperative weight kinetics following major pancreatectomy, and identify optimal and poor weight trajectories. Overall, 80% of patients will not obtain their original weight, which may negatively impact their quality of life. Ultimately, weight restoration appears to be predetermined, driven by immutable characteristics. Persistent weight loss also appears to be inevitable due to demographic/pathological factors, particularly tumor recurrence, but may be mitigated by limiting operative time and complications. This study provides insight to post-pancreatectomy recovery, helping to guide patient expectations. Moreover, it offers a foundation for future studies directed towards nutritional optimization for pancreatic resections.