EPIDEMIOLOGY OF REVISION PROCEDURES FOLLOWING PRIMARY HIP AND KNEE ARTHROPLASTY FOR OSTEOARTHRITIS

Abstract

Revision surgery is an important outcome of hip and knee arthroplasty among patients with osteoarthritis (OA). The objective of this study was to determine the risk of revision over time in a large, geographically defined population in North America, and to assess the effects of age, sex, and socioeconomic status (SES) on knee and hip revision rates in this population.

We used data on admissions to all acute care hospitals in British Columbia (BC), Canada (population of four million people) from 1986 until 2004 (eighteen years of follow-up). Primary knee and hip replacement surgery for OA and revision procedures were identified using diagnostic (ICD-9, ICD-10) codes and surgical procedure codes. We excluded cases admitted due to injury or neoplasm. SES was assigned based on residential address linked to census data at the level of enumeration area (ecological variable), and analyzed according to quintiles or deciles of the distribution. In the analysis, we used Kaplan-Meier survival curves to describe the risk of first revision following first primary replacement surgery over time and parametric (Weibull) survival regression models to analyze the effects of joint (hip vs. knee) age, sex, and SES on revision rates. Death, emigration, and another primary joint replacement were treated as censoring events.

Between 1986 and 2004, we observed 24,169 first primary hip replacements and 22,875 first primary knee replacements. In these patients, there were 1,313 hip revisions and 914 knee revisions following a primary replacement. The risk of revision at five, ten and fifteen years after primary replacement according to joint and sex were identified The overall risks were 10.1% for the hip and 8.7% for the knee at ten years, and 15.5% and 14.7%, respectively, at fifteen years. Risk of revision (%) following first primary joint replacement surgery Hip Knee Years Men Women Men Women 5 5.0 3.9 4.6 3.8 10 11.0 9.4 9.7 8.0 15 17.8 14.1 18.5 12.4 In a multivariate survival regression model including joint, age, sex, and SES, only age and sex were significantly associated with revisions. The hazard ratio was 1.2 for men compared with women (p< 0.0001). Revision rates were higher in younger persons, with hazard ratios increasing from 1.7 for age 70–79 (p< 0.0001) to 3.9 for age < 49 (p< 0.0001) compared with age 80+. After adjusting for age and sex, SES was not significantly associated with revision risk following primary hip or knee replacement surgery in this population (p=0.75).

This is one of the largest and longest cohort studies in North America to look at the epidemiology of revision procedures following hip and knee replacement for OA. Higher rates in men and in younger persons are consistent with previous reports. The lack of association between revision risk and SES in BC is an important finding given current concerns about socioeconomic disparities in access to and outcomes of treatment for OA.