Papalia Rocco, Zampogna Biagio, Torre Guglielmo, Papalia Giuseppe Francesco, Vorini Ferruccio, Bravi Marco, Albo Erika, De Vincentis Antonio, Denaro Vincenzo
Department of Orthopedics and Trauma Surgery, Campus Bio-Medico University of Rome, 00128 Rome, Italy.
Multi-Specialist Clinical Institute for Orthopaedic Trauma Care (COT), 98124 Messina, Italy.
J Clin Med. 2021 Oct 28;10(21):5053. doi: 10.3390/jcm10215053.
The aim of this retrospective investigation is to evaluate the correlation between several preoperative and perioperative factors and the length of hospital stay in patients that underwent elective total hip arthroplasty with overnight admission. Medical records of patients that underwent THA from the beginning of 2016 to the end of 2018 were retrospectively screened. Demographics, comorbidities, whole blood count, intraoperative details, and length of postoperative stay were retrieved. The association between clinical, laboratory and surgical factors and the length of hospital stay was explored by means of negative binomial and logistic regression models. The median length of postoperative hospital stay was four days (Inter Quartile Range, IQR 3, 5). After univariate regression a stepwise multivariate regression showed that operative time ( = 0.001), the preoperative serum creatinine ( < 0.001), the intraoperative blood loss ( = 0.04) and the use of an anterolateral approach ( < 0.001) were found to correlate significantly with the increase of the hospitalization length, while no significant correlation was found for all the other features. Multivariable model fitted through logistic regression (LOS below or over the median value of four days) had an Area Under the Curve (AUC) of 0.748. Our analysis suggests a significant role played by different preoperative and perioperative variables in influencing the length of hospital stay.
本回顾性研究的目的是评估择期全髋关节置换术患者过夜入院时，术前和围手术期的几个因素与住院时间之间的相关性。对2016年初至2018年底接受全髋关节置换术（THA）患者的病历进行回顾性筛选。收集患者的人口统计学资料、合并症、全血细胞计数、术中细节及术后住院时间。通过负二项回归和逻辑回归模型探讨临床、实验室及手术因素与住院时间之间的关联。术后住院时间的中位数为4天（四分位间距，IQR 3, 5）。单因素回归后逐步多因素回归显示，手术时间（P = 0.001）、术前血清肌酐（P < 0.001）、术中失血量（P = 0.04）及采用前外侧入路（P < 0.001）与住院时间延长显著相关，而其他所有特征均未发现显著相关性。通过逻辑回归拟合的多变量模型（住院时间低于或超过4天的中位数）的曲线下面积（AUC）为0.748。我们的分析表明，术前和围手术期的不同变量在影响住院时间方面发挥了重要作用。
