Application of Machine Learning for Predicting Clinically Meaningful Outcome After Arthroscopic Femoroacetabular Impingement Surgery
Application of Machine Learning for Predicting Clinically Meaningful Outcome After Arthroscopic Femoroacetabular Impingement Surgery
复制标题
DOI：
10.1177/0363546519892905
复制
发表时间：
2019-12-23
影响因子：
4.8
通讯作者：
Nho, Shane J.
中科院分区：
文献类型：
--
作者：
Nwachukwu, Benedict U.;Beck, Edward C.;Nho, Shane J.
背景：髋关节镜已成为关节内髋关节病变手术治疗的重要工具。对于因股骨髋臼撞击综合征（FAIS）接受髋关节镜检查的患者，具有临床意义的结局的预测模型尚不清楚。目的：应用机器学习模型确定术前变量，预测FAIS髋关节镜术后2年达到最小临床重要差异（MCID）。研究设计：病例对照研究;证据等级3。研究方法：分析了2012年1月至2016年7月期间由接受过大量奖学金培训的外科医生接受髋关节镜检查治疗FAIS的患者的数据。髋关节结局评分-日常生活活动（HOS-ADL）、HOS-Sport Specific（HOS-SS）和改良Harris髋关节评分（mHHS）的MCID截止值分别为9.8、14.4和9.14。预测模型实现MCID相对于每个建立与LASSO算法（最小绝对收缩和选择算子）的功能选择，然后对选定的功能进行逻辑回归。研究数据使用PatientIQ进行分析，这是一个基于云的医疗保健研究和分析平台。结果如下：在符合入选标准的1103例患者中，898例（81.4%）至少报告了2年结局，并进入建模算法。分别有74.0%、73.5%和79.9%的患者达到了HOS-ADL、HOS-SS和mHHS阈值评分，以达到MCID。未达到HOS-ADL MCID的预测因素包括焦虑/抑郁、术前症状持续时间>2年、较高的体重指数、术前HOS-ADL评分较高和术前髋关节注射（均P <0.05）。未达到HOS-SS MCID的预测因素包括焦虑/抑郁、术前症状持续时间>2年、术前HOS-SS评分高和术前髋关节注射，而至少在娱乐水平上跑步是达到HOS-SS MCID的预测因素（所有P <0.05）。未达到mHHS MCID的预测因素包括焦虑或抑郁史、术前mHHS评分高和髋关节注射，而女性则预测达到MCID（所有P <0.05）。结论：本研究确定了FAIS髋关节镜术后获得有临床意义结局的预测变量。患者因素包括焦虑/抑郁、症状持续时间>2年、术前关节内注射和术前结局评分高，这些因素最能预测无法获得有临床意义的结局。这些发现对髋关节镜治疗FAI后的共同决策算法和术前预期管理具有重要意义。
Background: Hip arthroscopy has become an important tool for surgical treatment of intra-articular hip pathology. Predictive models for clinically meaningful outcomes in patients undergoing hip arthroscopy for femoroacetabular impingement syndrome (FAIS) are unknown. Purpose: To apply a machine learning model to determine preoperative variables predictive for achieving the minimal clinically important difference (MCID) at 2 years after hip arthroscopy for FAIS. Study Design: Case-control study; Level of evidence, 3. Methods: Data were analyzed for patients who underwent hip arthroscopy for FAIS by a high-volume fellowship-trained surgeon between January 2012 and July 2016. The MCID cutoffs for the Hip Outcome Score-Activities of Daily Living (HOS-ADL), HOS-Sport Specific (HOS-SS), and modified Harris Hip Score (mHHS) were 9.8, 14.4, and 9.14, respectively. Predictive models for achieving the MCID with respect to each were built with the LASSO algorithm (least absolute shrinkage and selection operator) for feature selection, followed by logistic regression on the selected features. Study data were analyzed with PatientIQ, a cloud-based research and analytics platform for health care. Results: Of 1103 patients who met inclusion criteria, 898 (81.4%) had a minimum of 2-year reported outcomes and were entered into the modeling algorithm. A total of 74.0%, 73.5%, and 79.9% met the HOS-ADL, HOS-SS, and mHHS threshold scores for achieving the MCID. Predictors of not achieving the HOS-ADL MCID included anxiety/depression, symptom duration for >2 years before surgery, higher body mass index, high preoperative HOS-ADL score, and preoperative hip injection (all P < .05). Predictors of not achieving the HOS-SS MCID included anxiety/depression, preoperative symptom duration for >2 years, high preoperative HOS-SS score, and preoperative hip injection, while running at least at the recreational level was a predictor of achieving HOS-SS MCID (all P < .05). Predictors of not achieving the mHHS MCID included history of anxiety or depression, high preoperative mHHS score, and hip injections, while being female was predictive of achieving the MCID (all P < .05). Conclusion: This study identified predictive variables for achieving clinically meaningful outcome after hip arthroscopy for FAIS. Patient factors including anxiety/depression, symptom duration >2 years, preoperative intra-articular injection, and high preoperative outcome scores are most consistently predictive of inability to achieve clinically meaningful outcome. These findings have important implications for shared decision-making algorithms and management of preoperative expectations after hip arthroscopy for FAI.