Total Joint Arthroplasty in Immunocompromised Patients

A Matched Pair Analysis for Comorbidities

Morteza Meftah, MD; Grace Plassche, BS; Ariel Silverman, BS; Peter B. White, BS; Ira H. Kirschenbaum, MD


Curr Orthop Pract. 2019;30(3):246-249. 

In This Article


In recent time, the number of IC patients undergoing TJA has been increasing.[7] As these patients are at a theoretical risk of increased perioperative complications and postoperative infections, the true risk has yet to be fully established. In the current study, we failed to find that IC patients are at an increased risk for perioperative complications or postoperative infections; however, we did find that IC patients are at an increased risk for reoperation.

This study is not without limitations. First, this is a retrospective study with inherent limitations associated with such designs. Second, although this is matched-pair study, the sample size may be too small to detect significant differences in infections and complication rates. Since we have an unusually high rate of immunocompromised patients in our geographical location, sample-size analysis was not performed. Third, not all HIV and HCV patients are the same, as their immunosuppression can be correlated with their CD4 count and viral loads, respectively. Our minimum acceptance level for CD4 count was 500; it is possible that lower numbers affect the outcomes of TJA in HIV patients. Finally, this is a single-institution study in an urban area that is highly prevalent with these conditions, and therefore the results of this study may not be applicable to other areas. Strengths of this study include the match-pair design, which included matching a cohort based on medical comorbidities.

Several studies have attempted to evaluate the risk of perioperative complications and postoperative infection in immunocompromised patients and have found mixed results. Although some have found that immunocompromised patients are not at an increased risk for complications,[12–14] others have found that they are.[11,20] Dimitriou et al.[11] performed a large review and found that the overall complication rate for patients with HIV was 3.3%. However, in the present study we found that the risk of a perioperative medical complication was much larger at 20% for the IC group. This difference is likely attributable to our more liberal definition of a complication, which included perioperative medical complications such as urinary tract infections and atrial fibrillations.

One of the most concerning aspects of perioperative complications in IC patients is the increased theoretical risk for postoperative joint infections. In the large meta-analysis by Dimitriou et al.[11] they found that patients with HIV are at an increased risk of infection, around 7.3%, compared to N-IC patients, at 3.3%. In a recent study, that included patients with HIV, HCV, and HBV,[20] it was found that HBV and HCV are independent risk factors for periprosthetic joint infections. In this current study, we failed to detect a similar finding. We found that the incidence of deep infections (2.1% vs. 0.9%) and superficial infections (5.5% vs. 2.7%) was similar among the IC and N-IC groups. We believe that this finding is likely attributable to our sample size, and a larger number of patients may be needed to detect a significant difference. Interestingly, in our cohort of patients who had comorbidities controlled for other independent risk factors for perioperative infections,[22] we continued to see no significant difference. One conclusion from this observation is that with current available treatments for immunocompromised patients, the risk of complications is related to other coexisting conditions such as diabetes. Although use of opioids has been shown to be a risk factor for readmission, infection, and complications,[24] this was not significant in our study.

The revision and reoperation rates between IC and N-IC were statistically significant; however, when matched for comorbidities such as diabetes, they were no longer significant. Also, treating IC patients, such as HCV prior to joint replacement potentially has different outcomes as compared to non-treated HCV. The presence of cirrhosis or advanced liver disease also effects the outcomes. We had no cases of advanced liver disease in this cohort.

In conclusion, we were unable to detect differences between IC and N-IC and patients in complications or infection risks and were unable to identify a medical comorbidity that increased the risk of infection. We did find that IC patients are at an increased risk for reoperation during the first 2 yr postoperatively, most notably for infection.