Postoperative Outcomes Within an Enhanced Recovery After Surgery Protocol National Audit of Complications Following Elective Total Hip or Knee Replacement Within or Without an Enhanced Recovery After Surgery Protocol (POWER2)

Methods National audit of a 60-day prospective observational cohort in the that the pre-defined postoperative complications at 30 days of follow-up in adult patients electively undergoing hip replacement surgery (HRS) or knee replacement surgery (KRS) with or without a program of Enhanced Recovery (ERAS: Enhanced Recovery after Surgery) with any level of protocol compliance (0-100%)

Research Places Spanish Hospitals where these surgical interventions are performed usually

Objectives To determine the incidence of Postoperative Complications per patient and procedure, regardless of the degree of adherence to ERAS protocols and its impact on the Hospital Stay and on postoperative complications including 30-day mortality

Sample Size For an alpha error of 5% (95% confidence and an accuracy of 3% and estimating a number of patients with complications of 50%, the calculation of sample size yields 3012 patients, although the final sample size may be smaller depending on the proportion of complications detected.

Inclusion criteria Patients over 18 undergoing elective HRS or KRS PTC regardless of its affiliation to a ERAS program and the level of compliance with protocol (from 0-100%)

Statistical analysis Continuous variables will be described as mean and standard deviation, if it is a normal distribution, or median and interquartile range, if they are not distributed normally. The comparisons of continuous variables are performed by one-way ANOVA or the Mann-Whitney test, depending on proceed A univariate analysis will be carried out to test the factors associated with postoperative complications, hospital stay and death in Hospital. Univariate analyzes and logistic regression models will be constructed Hierarchical multivariate to identify the factors associated in a independent with these results and to adjust the differences in the factors of confusion. The factors will be introduced in the models based on their relationship with the univariate result (p <0.05), the biological plausibility and the low rate of missing data.