OBJECTIVE: Complication rates after pancreatic resections remain high despite improvement in perioperative management. The effects of body composition and the relationship among different body compartments on surgical morbidity are not comprehensively investigated. The aim of this study was to assess whether the evaluation of different body compartments and their relationship was associated with the development of major postoperative complications after pancreatoduodenectomy (PD) for cancer.
METHODS: We retrospectively analyzed 124 patients who underwent PD and had a staging computed tomography (CT) scan at our center. CT scan was used to measure abdominal skeletal muscle area and volume, as well as visceral fat area (VFA) and volume. The total abdominal muscle area (TAMA) was then normalized for height. The severity of complications was assessed. Univariate and multivariate analyses were performed to investigate correlations between the above variables and postoperative complications. The receiver operating characteristic curve methodology was used to investigate the predictive ability of each parameter.
RESULTS: Major complications occurred in 42 patients (33.9%). The prevalence of sarcopenia was 24.2%. Regression analyses revealed no correlation between abdominal muscular and adipose tissue areas. Univariate analysis showed that the depletion of muscle area normalized for height was not per se predictive of complications (P = 0.318). Multivariate logistic regression showed that the VFA/TAMA was the only determinant of major complications (odds ratio, 3.20; 95% confidence interval, 1.35-7.60; P = 0.008). The model predictive performance was 0.735 (area under the curve) with a sensitivity of 64.3% and a specificity of 74.4%.
CONCLUSION: Sarcopenic obesity is a strong predictor of major complications after PD for cancer.
- Body Composition
- Body Mass Index
- Intra-Abdominal Fat
- Middle Aged
- Multidetector Computed Tomography
- Muscle, Skeletal
- Pancreatic Neoplasms
- Postoperative Complications
- Retrospective Studies
- Risk Factors
- Journal Article