Incidence of low-grade infection in aseptic loosening of total hip arthroplasty

Abstract

Purpose We investigated the hypothesis that many total hip arthroplasty revisions that are classified as aseptic are in fact low-grade infections missed with routine diagnostics.

Methods In 7 Dutch hospitals, 176 consecutive patients with the preoperative diagnosis of aseptic loosening of their total hip arthroplasty were enrolled. During surgery, between 14 and 20 tissue samples were obtained for culture, pathology, and broad-range 16S rRNA PCR with reverse line blot hybridization. Patients were classified as either not being infected, suspected of having infection, or infected according to strict, predefined criteria. Each patient had a follow-up visit after 1 year.

Results 7 patients were classified as infected, 4 of whom were not identified by routine culture. 15 additional patients were suspected of having infection. 20 of these 22 patients received a cemented prosthesis, fixated with antibiotic-loaded bone cement. All 22 patients received prophylactic systemic antibiotics. 7 of them reported complaints one year after surgery, but only one showed signs of early loosening. However, additional surgery was not performed in any of the patients.

Interpretation Although the proportions were not as high as previously reported in the literature, between 4% and 13% of patients with the preoperative diagnosis of aseptic loosening were infected. However, as thorough debridement was performed during surgery and prophylactic antibiotics were used, the diagnosis of infection did not have any obvious clinical consequences, as most patients performed well at the 1-year follow-up. Whether this observation has implications for long-term implant survival remains to be seen.

DJFM: study design, writing protocol, inclusion of patients, data analysis, writing of manuscript, principal investigator. GvHt: study design, surgery, writing of manuscript. CV: study design, writing protocol, surgery, writing of manuscript. BJB: study design, inclusion and surgery of patients, local coordinator. GHIMW: study design, surgery, writing of manuscript. NJAT: study design, inclusion and surgery of patients, local coordinator. BWS: study design, inclusion and surgery of patients, local coordinator, writing of manuscript. FRAJdM: study design, inclusion and surgery of patients, local coordinator. CSS: PCR optimization and analysis, writing of manuscript. IvdP: PCR optimization and analysis, writing of manuscript. TF: pathology, writing of manuscript. LMS: study design, writing of protocol, PCR supervisor, writing of manuscript. TWB: study design, writing of protocol, pathology, supervisor, writing of manuscript. WJAD: study design, writing of protocol, writing of manuscript, senior investigator.

The authors thank the following people for their contributions: Sanne Spijkers, Henriette de Gouw, Liesbeth Jutten, all participating orthopaedic surgeons, and the microbiology and pathology depaertments of all hospitals.

This study was supported by an institutional research grant from Stryker Orthopaedics (Mahwah, NJ). Stryker had no role in planning the study, data collection, analysis, interpretation of data, or in writing of the manuscript.