Complications after total hip arthroplasty : register-based studies on surgical approach and infections

Abstract

Background: The often excellent outcome of total hip arthroplasty (THA) is sometimes compromised by prosthesis related complications. There is an ongoing debate regarding the effect of surgical approach on prosthesis related complications and patient reported outcome. The incidence of periprosthetic joint infection (PJI) following THA in Sweden is unknown as well as the completeness of the Swedish Hip Arthroplasty Register (SHAR) to record reoperations due to infection.

Methods: Study 1 and 2 were solely based on data from the SHAR. In study 1, the risk of revision due to infection, dislocation and loosening of the prosthesis was compared between the posterior and the direct lateral approach for the most commonly used cemented THAs. Relative risks were calculated with a Cox regression analysis with respect to known confounders. In study 2, Patient Reported Outcome Measures (PROMs) were compared between patients that had received a THA for osteoarthritis by either the posterior or direct lateral approach. The postoperative pain, satisfaction and health related quality of life were compared by Chi2-test, T-test and multivariate regression analysis.

In study 3 and 4 a large cohort of patients that had undergone primary THA were selected from the SHAR and their antibiotic consumption was extracted from the Swedish Prescribed Drugs Register (SPDR). For patients with ≥28 days of antibiotic treatment within 2 years after the surgery a medical records review was conducted. From this survey the cumulative incidence of early and delayed PJI was calculated. The number of reoperated THAs due to infection were compared with the SHAR reoperation database determining the completeness of the register. In Study 5, PJI-cases were compared to matched controls regarding known and suspected risk factors from the SHAR and the National Patient Register and analyzed by multivariate regression analysis.

Results: The posterior approach lead to superior PROM values for patients with OA and lead to less aseptic loosening but increased risk of dislocation in some of the investigated all cemented THAs. The cumulative incidence of early and delayed PJIs in Sweden for the study period was 0.9 % and the trend was slightly increasing. The registration completeness of reoperations due to infection in the SHAR was 67 %. Patient comorbidity had more influence than surgical or environmental factors on PJI risk.

Discussion: The discussion regarding the advantages of different surgical approaches in THA is not ended by these studies, but it seems that the surgical approach has influence on the outcome. The surgical approach should therefore perhaps be a part of the preoperative planning just as the method of fixation and implant and related to both patient factors and the reason for the THA.

In order to reduce the infection incidence it is important to define the magnitude of the incidence. This study presents a novel approach to incidence estimation and evaluates the SHAR reoperation database which has been the most commonly used measure of infections after THA in Sweden. As patient comorbidity seems to be the most important factor on a population level it will be difficult to further decrease the incidence.