Scully William, Piuzzi Nicolas S, Sodhi Nipun, Sultan Assem A, George Jaiben, Khlopas Anton, Muschler George F, Higuera Carlos A, Mont Michael A
Cleveland Clinic, Department of Orthopedic Surgery, Cleveland, Ohio, USA.
University Institute of the Italian Hospital of Buenos Aires, Buenos Aires, Argentina.
Hip Int. 2020 Mar;30(2):125-134. doi: 10.1177/1120700019826482. Epub 2019 Feb 5.
Evaluating body mass index (BMI) as a continuous variable eliminates the potential pitfalls of only considering BMI as a binary or categorical variable, as most studies do when correlating BMI and total hip arthroplasty (THA) outcomes. Therefore, the objective of this study was to correlate the effect of continuous BMI on 30-day complications post-THA. Specifically, we correlated BMI to: (1) 30-day readmissions and reoperations; (2) medical complications; and (3) surgical complications in: (a) normal-weight; (b) over-weight; (c) obese; and (d) morbidly obese patients.
Using the NSQIP database, 93,598 primary THAs were identified. 30-day rates of readmissions, reoperations, and medical/surgical complications as well as patient BMI data were extrapolated. A comparative analysis using univariate, multivariate, and spline regression models adjusting for demographics and comorbidities were created to study the continuous effect of BMI on different outcomes.
Readmission ( 0.001), reoperation ( 0.007), superficial infection ( 0.003), prosthetic joint infection ( 0.001), and sepsis ( 0.026) had a J-shaped relationship with BMI, with the lowest rates seen in patients with BMI around 28 kg/m. The risks of mortality ( 0.007) and transfusion ( 0.001) had a reverse J-shaped relationship, with the risk steadily decreasing for BMIs in the normal weight and overweight range, and then flattening afterwards.
This data proposes a multifactorial effect of BMI on post-THA complications. Considering BMI as a continuous variable allows for a better assessment when considering the interplay between modifiable risk factors, such as smoking or alcohol use, as well as multiple comorbidities.
将体重指数（BMI）作为连续变量进行评估，可消除仅将BMI视为二元或分类变量时的潜在缺陷，而大多数研究在关联BMI与全髋关节置换术（THA）结果时都是如此。因此，本研究的目的是关联连续BMI对THA术后30天并发症的影响。具体而言，我们将BMI与以下方面进行关联：（1）30天再入院和再次手术；（2）医疗并发症；以及（3）手术并发症，涉及：（a）正常体重；（b）超重；（c）肥胖；以及（d）病态肥胖患者。
使用国家外科质量改进计划（NSQIP）数据库，确定了93598例初次THA。推算出30天再入院、再次手术以及医疗/手术并发症发生率以及患者BMI数据。创建了使用单变量、多变量和样条回归模型并针对人口统计学和合并症进行调整的比较分析，以研究BMI对不同结果的连续影响。
再入院（P = 0.001）、再次手术（P = 0.007）、浅表感染（P = 0.003）、人工关节感染（P = 0.001）和脓毒症（P = 0.026）与BMI呈J形关系，BMI约为28 kg/m²的患者发生率最低。死亡率（P = 0.007）和输血（P = 0.001）风险呈倒J形关系，在正常体重和超重范围内，BMI的风险稳步下降，然后趋于平稳。
该数据表明BMI对THA术后并发症具有多因素影响。将BMI视为连续变量在考虑可改变的风险因素（如吸烟或饮酒）以及多种合并症之间的相互作用时可进行更好的评估。
