Objective:To investigate the influencing factors of rectus abdominis atrophy after laparoscopic rectal cancer surgery and establish a predictive model.Methods:From Mar.2020 to Mar.2023,229 patients with rectal cancer who underwent radical laparo-scopic surgery were retrospectively selected,and relevant clinical data were collected.According to whether the patients suffered from rectus abdominis atrophy after surgery,the patients were divided into atrophic group(n=47)and control group(n=182).The factors of rectus atrophy after laparoscopic rectal cancer surgery were analyzed by logistic regression.The predictive model of rectus abdominis atrophy after laparoscopic rectal cancer surgery was constructed by receiver operating characteristic curve.The calibration degree of the predictive model was tested by Hosmer-Lemeshow.Results:Multivariate logistic regression analysis showed that older age,postoperative long-term bed rest,prophylactic ostomy,postoperative chemotherapy and postoperative surgical site infection were risk factors for rectus abdominis atrophy after laparoscopic surgery for rectal cancer(P<0.05),and high albumin level was protective factor(P<0.05).Receiver operating characteristic curve analysis showed that the area under the curve of the predictive model for predicting rectus abdo-minis atrophy after laparoscopic rectal cancer surgery was 0.868(95%CI:0.817-0.909),and the sensitivity and specificity were 85.11%and87.91%,respectively.Hosmer-Lemeshow test showed that prediction model had a good coincidence(χ2=3.251,P>0.05).Conclusions:Older age,postoperative long-term bed rest,prophylactic ostomy,postoperative chemotherapy,surgical site infection and low albumin are the related factors of rectus abdominis atrophy after laparoscopic rectal cancer surgery,according above factors,a pre-diction model can be established to better predict the risk of rectus abdominis atrophy.
关键词
直肠肿瘤/腹腔镜检查/预防性造口/腹直肌萎缩/影响因素分析/预测模型
Key words
Rectal neoplasms/Laparoscopy/Prophylactic ostomy/Rectus abdominis atrophy/Root cause analysis/Prediction model