The Impact of Fellowship Training on Surgical Outcomes in Total Hip Arthroplasty: A Propensity Score-Matched Analysis.

Abstract

Fellowship training has become increasingly popular in the field of orthopaedics, with adult reconstruction being one of the most heavily pursued training pathways among residents. However, the value of this training is underexplored in the literature. This study sought to compare the incidences of complications, including periprosthetic joint infections, periprosthetic fractures, and dislocations, between patients who underwent elective total hip arthroplasty (THA) by an arthroplasty-trained surgeon versus a non-arthroplasty surgeon at 90 days, one year, and two years. We utilized a national database to identify a national cohort of patients who underwent elective THA from 2010 to 2021. From this set of patients, we obtained a physician report, which included the name, National Provider Identifier (NPI), and location of the operating surgeon. A web search was utilized to identify whether the physician had undergone fellowship training in adult reconstruction. Utilizing this information, patients were split into two cohorts based on their surgeon: arthroplasty-trained and non-arthroplasty-trained. A propensity score match was used to match patients based on demographics, comorbidities, and surgeon case volume, resulting in two cohorts of 177,777 patients each. At 90 days, patients in the non-arthroplasty cohort experienced higher incidences of surgical site infection (SSI) (P = 0.002), venous thromboembolism (VTE) (P = 0.006), periprosthetic joint infection (PJI) (P = 0.035), periprosthetic fracture (PPFx) (P = 0.001), aseptic revision (P < 0.001), and dislocation (P < 0.001). At one and two years, these trends continued, with patients in the non-arthroplasty cohort experiencing higher rates of PJI, PPFx, aseptic revision, and dislocation (all P < 0.05). Patients who underwent elective THA with arthroplasty-trained surgeons experienced fewer complications up to two years postoperatively.