Abstract / Summary
Serum calcium is an important indicator for assessing the severity of acute pancreatitis. The physiological processes following pancreatitis and pancreaticoduodenectomy (PD) are similar; however, it is unknown whether serum calcium is associated with outcomes after PD. The present study analyzed serum calcium on postoperative days 3, 6, 9, and 12. Logistic regression was used to identify correlations between serum calcium, clinical postoperative pancreatic fistula (POPF), and other complications. Generalized additive models (GAM) were used to determine serum calcium thresholds. Restricted cubic spline (RCS) was used to visualize linear or non-linear relationships between serum calcium and outcomes. Latent class analysis (LCA) was used to identify potential categories based on serum calcium levels at various time points. Generalized additive mixed-effects models (GAMM) were combined to analyze the relationship between serum calcium and outcomes at four-time points. The study analyzed 447 PD patients. Serum calcium levels at four-time points were negatively associated with clinical POPF, with thresholds of 1.96 and 1.98 calculated using GAM. RCS was used to visualize the relationship between serum calcium and outcomes in a possible L-type. We identified four potential classes through LCA, and further logistic regression demonstrated that persistently lower serum calcium levels were associated with an increased risk of clinical POPF. GAMM-exposed postoperative serum calcium was negatively associated with clinical POPF. In this retrospective cohort, lower postoperative serum calcium levels were associated with higher odds of clinical POPF and other postoperative complications. These findings are exploratory and require prospective validation before serum calcium can be considered for clinical risk assessment.