建立不同職業肥胖患者減重手術治療成效之預測模型

目的：全世界開發中及已開發中國家的肥胖問題已達到流行病的程度。肥胖危險因子包含與職業相關的社會決定因子，如工作環境、工作性質。對於不同職業別患者參與減重手術之成效，在過去文獻中未盡探討。本文針對不同職業背景之患者，分析術後成效與預測因子之關聯性，期望建立預測術後成效之模型。方法：本研究是回顧性病例對照研究，收錄101位接受減重手術之病態性肥胖患者並追蹤一年。以多餘流失體重百分比作為減重手術成功與否的指標，並記錄病患之職業類別、勞動辛勞度職業別、身體組成及生化數值。先使用曼－惠特尼U考驗法以及卡方檢定做單變相無母數分析，建立獨立相關變相，再進一步利用多變項羅吉斯迴歸模型來預測減重成效的預測因子，並以ROC曲線下面積大小來決定預測之正確性。結果：無母數分析顯示減重手術一年後成功與否與勞動辛勞度職業別無統計上顯著差異（p=0.165），而與職業類別有顯著差異（p=0.009）。在病患術前身體組成、生化數值等眾多因子當中，僅臀圍（p=0.008）、脂肪重量（p=0.001）以及身體質量指數（p=0.004）有顯著差異。進一步利用多變項羅吉斯迴歸模型，以向後選取法來預測減重成效的預測因子，顯示僅術前身體質量指數達統計上顯著差異（p=0.026, OR=0.817），其ROC曲線下面積為0.733。結論：減重手術對於病態性肥胖確實是有效的減重方法之一，而手術一年後的成效不會因不同職業別、體力勞動多寡而有所差異，而術前身體質量指數為影響手術成功否與的唯一預測因子。

關鍵字

職業；減重手術；病態性肥胖；多餘流失體重百分比；身體質量指數

並列摘要

Objective: Obesity prevalence in developed and developing countries around the world has reached epidemic proportions. Risk factors of obesity includes social determinants related to occupations, such as the working environment and the characteristics of occupation. The effectiveness of bariatric surgery for patients from different occupations has not been explored in the previous studies. This article analyzes patients in different occupational backgrounds, including factors that may predict postoperative outcomes: different occupational categories, different occupational labor levels, preoperative body compositions and biochemical markers. We aim to establish a model for predicting the effectiveness of bariatric surgery. Design: This study is a retrospective case-control study. We enrolled 101 patients with morbid obesity undergoing laparoscopic sleeve gastrectomy and followed for one year. The percentage of excess weight loss (%EWL) after one year of surgery is used as an indicator of the success of bariatric surgery. We analyze factors that may affect the outcome of the surgery including occupational categories (six items), occupational labor levels (three items), preoperative body compositions and biochemical markers. The occupational categories include six groups, and the occupational labor levels include three groups according to the different degrees of labor hardship, which are divided into mild, moderate, and severe. First, we use the Mann-Whitney U test and the Chi-square test to perform nonparametric analysis, find independent variables, and further use the multivariate logistic regression to predict the factors of effectiveness of bariatric surgery. Then we calculate the area under the ROC curve to determine the correctness of the prediction. Setting: One hospital, Taipei Medical University Hospital(TMUH). Patients: Patients with morbid obesity who underwent bariatric surgery in TMUH were included. Intervention: Bariatric surgery. Measurements and Main Results: Nonparametric analysis shows that there is no statistically significant difference between the occupational labor levels and the success rate of one-year bariatric surgery (p=0.165), but significantly difference between the occupational categories and the success rate of one-year bariatric surgery is noted (p=0.009). Among factors of preoperative body compositions and biochemical markers, there is significant differences in hip circumference (p=0.008), fat mass (p=0.001), and body mass index (BMI) (p=0.004) (Table 1). Then we use the multivariate logistic regression to model these significant factors by backward selection, the result shows that only the preoperative BMI has a statistically significant difference (p=0.026, OR=0.817). The ROC curve of BMI shows that the area under the curve is 0.73. (Figure 1). Conclusions:Bariatric surgery is indeed one of the most effective weight loss methods to treat morbid obesity. Different occupations and occupational labor levels will not affect the first year success rate of bariatric surgery. The preoperative BMI is the only predictor of the first year success rate of bariatric surgery.

並列關鍵字

occupation ； bariatric surgery ； morbid obesity ； excess lost weight percentage ； body mass index

國際替代計量

建立不同職業肥胖患者減重手術治療成效之預測模型

全文下載

主題瀏覽