Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB, Canada.
Department of Surgery, University of Alberta, Edmonton, AB, Canada.
Ann Surg Oncol. 2018 Sep;25(9):2669-2680. doi: 10.1245/s10434-018-6652-x. Epub 2018 Jul 13.
Sarcopenia, visceral obesity (VO), and reduced muscle radiodensity (myosteatosis) are suggested risk factors for postoperative morbidity in colorectal cancer (CRC), but usually are not concurrently assessed. Published thresholds used to define these features are not CRC-specific and are defined in relation to mortality, not postoperative outcomes. This study aimed to evaluate body composition in relation to length of hospital stay (LOS) and postoperative outcomes.
Pre-surgical computed tomography (CT) images were assessed for total area and radiodensity of skeletal muscle and visceral adipose tissue in a pooled Canadian and UK cohort (n = 2100). Sex- and age-specific values for these features were calculated. For 1139 of 2100 patients, LOS data were available, and sex- and age-specific thresholds for sarcopenia, myosteatosis, and VO were defined on the basis of LOS. Association of CT-defined features with LOS and readmissions was explored using negative binomial and logistic regression models, respectively.
In the multivariable analysis, the predictors of LOS (P < 0.001) were age, surgical approach, major complications (incidence rate ratio [IRR] 2.42; 95% confidence interval [CI] 2.18-2.68), study cohort, and three body composition profiles characterized by myosteatosis combined with either sarcopenia (IRR, 1.27; 95% CI 1.12-1.43) or VO (IRR, 1.25; 95% CI 1.10-1.42), and myosteatosis combined with both sarcopenia and VO (IRR, 1.58; 95% CI 1.29-1.93). In the multivariable analysis, risk of readmission was associated with VO alone (odds ratio [OR] 2.66; 95% CI 1.18-6.00); P = 0.018), VO combined with myosteatosis (OR, 2.72; 95% CI 1.36-5.46; P = 0.005), or VO combined with myosteatosis and sarcopenia (OR, 2.98; 95% CI 1.06-5.46; P = 0.038). Importantly, the effect of body composition profiles on LOS and readmission was independent of major complications.
The findings showed that CT-defined multidimensional body habitus is independently associated with LOS and hospital readmission.
肌少症、内脏肥胖(VO)和肌肉放射性密度降低(肌脂病)被认为是结直肠癌(CRC)术后发病率的危险因素,但通常不能同时评估。用于定义这些特征的已发表阈值与死亡率有关,而不是与术后结果有关,并且不是 CRC 特异性的。本研究旨在评估与住院时间(LOS)和术后结果相关的身体成分。
在加拿大和英国的队列中(n=2100),对术前计算机断层扫描(CT)图像进行了骨骼肌和内脏脂肪组织的总面积和放射性密度评估。计算了这些特征的性别和年龄特异性值。对于 2100 名患者中的 1139 名,有 LOS 数据,根据 LOS 确定了肌少症、肌脂病和 VO 的性别和年龄特异性阈值。使用负二项式和逻辑回归模型分别探讨 CT 定义特征与 LOS 和再入院的相关性。
在多变量分析中, LOS 的预测因素(P<0.001)为年龄、手术方式、主要并发症(发生率比[IRR]2.42;95%置信区间[CI]2.18-2.68)、研究队列以及三种体成分特征,肌脂病与肌少症(IRR,1.27;95%CI 1.12-1.43)或 VO(IRR,1.25;95%CI 1.10-1.42)相结合,以及肌脂病与肌少症和 VO 相结合(IRR,1.58;95%CI 1.29-1.93)。在多变量分析中,再入院的风险与 VO 单独相关(比值比[OR]2.66;95%CI 1.18-6.00;P=0.018),与 VO 合并肌脂病(OR,2.72;95%CI 1.36-5.46;P=0.005),或与 VO 合并肌脂病和肌少症(OR,2.98;95%CI 1.06-5.46;P=0.038)。重要的是,体成分特征对 LOS 和再入院的影响独立于主要并发症。
研究结果表明,CT 定义的多维身体形态与 LOS 和住院再入院独立相关。