What is the Safe Distance Between Hip and Knee Implants to Reduce the Risk of Ipsilateral Metachronous Periprosthetic Joint Infection?

Abstract

Background Periprosthetic joint infection (PJI), the most common cause of revision after TKA and THA, is a devastating complication for patients that is difficult to diagnose and treat. An increase in the number of patients with multiple joint arthroplasties in the same extremity will result in an increased risk of ipsilateral PJI. However, there is no definition of risk factors, micro-organism patterns, and safe distance between knee and hip implants for this patient group. Questions/purposes (1) In patients with hip and knee arthroplasties on the same side who experience a PJI of one implant, are there factors associated with the development of subsequent PJI of the other implant? (2) In this patient group, how often is the same organism responsible for both PJIs? (3) Is a shorter distance from an infected prosthetic joint to an ipsilateral prosthetic joint associated with greater odds of subsequent infection of the second joint? Methods We designed a retrospective study of a longitudinally maintained institutional database that identified all one-stage and two-stage procedures performed for chronic PJI of the hip and knee at our tertiary referral arthroplasty center between January 2010 and December 2018 (n = 2352). Of these patients, 6.8% (161 of 2352) had an ipsilateral hip or knee implant in situ at the time of receiving surgical treatment for a PJI of the hip or knee. The following criteria led to the exclusion of 39% (63 of 161) of these patients: 4.3% (seven of 161) for incomplete documentation, 30% (48 of 161) for unavailability of full-leg radiographs, and 5% (eight of 161) for synchronous infection. With regard to the latter, per internal protocol, all artificial joints were aspirated before septic surgery, allowing us to differentiate between synchronous and metachronous infection. The remaining 98 patients were included in the final analysis. Twenty patients experienced ipsilateral metachronous PJI during the study period (Group 1) and 78 patients did not experience a same-side PJI (Group 2). We analyzed the microbiological characteristics of bacteria during the first PJI and ipsilateral metachronous PJI. Calibrated, full-length plain radiographs were evaluated. Receiver operating characteristic curves were analyzed to determine the optimal cutoff for the stem-to-stem and empty native bone distance. The mean time between the initial PJI and ipsilateral metachronous PJI was 8 ± 14 months. Patients were followed for a minimum of 24 months for any complications. Results The risk of ipsilateral metachronous PJI in the other joint secondary to a joint implant in which PJI develops can increase up to 20% in the first 2 years after the procedure. There was no difference between the two groups in age, sex, initial joint replacement (knee or hip), and BMI. However, patients in the ipsilateral metachronous PJI group were shorter and had a lower weight (1.6 ± 0.1 m and 76 ± 16 kg). An analysis of the microbiological characteristics of bacteria at the time of the initial PJI showed no differences in the proportions of difficult-to-treat, high virulence, and polymicrobial infections between the two groups (20% [20 of 98] versus 80% [78 of 98]). Our findings showed that the ipsilateral metachronous PJI group had a shorter stem-to-stem distance, shorter empty native bone distance, and a higher risk of cement restrictor failure (p < 0.01) than the 78 patients who did not experience ipsilateral metachronous PJI during the study period. An analysis of the receiver operating characteristic curve showed a cutoff of 7 cm for the empty native bone distance (p < 0.01), with a sensitivity of 72% and a specificity of 75%. Conclusion The risk of ipsilateral metachronous PJI in patients with multiple joint arthroplasties is associated with shorter stature and stem-to-stem distance. Appropriate position of the cement restrictor and native bone distance are important in reducing the risk of ipsilateral metachronous PJI in these patients. Future studies might evaluate the risk of ipsilateral metachronous PJI owing to bone adjacency. Level of Evidence Level III, therapeutic study.

CORR Insights®: What is the Safe Distance Between Hip and Knee Implants to Reduce the Risk of Ipsilateral Metachronous Periprosthetic Joint Infection?

“…In their current work in Clinical Orthopaedics and Related Research ® , Akkaya et al [1] present an interesting and very relevant perspective about periprosthetic joint infection (PJI) by looking at the unusual (but not rare) setting of ipsiplateral THA and TKA. The paper, from the laboratory of Professor Mustafa Citak, arises from an analysis of insitutional data collected at the world-renowned ENDO-Klinik in Germany.…”

“…Equally, how does the central European experience translate to the North American, Oceania, and Asian settings (acknowledging differences in populations, resource availability, healthcare system structures and clinical care paradigms)? Of equal importance, approaches to the diagnosis of PJI have evolved since 2010 (which is when retrospective patient inclusion began in this study [1]) —in particular, the trend toward adoption of standardized diagostic methods such as the updated Musculoskeletal Infection Society criteria (2018) [8]. This raises fair questions as to whether concurrent infections may have been missed in this report; future studies should therefore seek to replicate or refute these findings.…”

“…Given the demonstrated association of shorter intramedullary separation of implants and absence of an effective cement restrictor as factors linked to secondary infections in this study [1], the value of antibiotic-impregnanted cement in this setting seems worthy of further exploration. Equally, because the high-quality results of Akkaya et al [1] came from patients who mostly had cemented fixation, future studies will need to ascertain whether the use of uncemented stems or constructs carries a greater or smaller metachronous PJI risk burden for ipsilateral hip and knee arthroplasties.…”

“…I suggest our answers will instead come from the world’s increasingly sophisticated datasets captured by large national and regional registries. Despite the limitations inherent in the retrospective nature of such work, using these resources may quickly provide informative and meaningful data to support (or refute) the findings of Akkaya et al [1] and may allow us to focus on the development of prospective risk profiling approaches. Such screening tools might result in reductions in subsequent infections owing to better decision-making and perhaps earlier care interventions.…”

What’s New in Adult Reconstructive Knee Surgery

“…Unexpectedly, the investigators found that the associations (regressions) between the coefficient-of-variation tertiles and complication, PJI, mortality, and non-home discharge variables were not significant in revision TKA, although they argued that the relatively small sample size (n = 636) was responsible for these findings. Factors associated with the development of subsequent PJI among patients with ipsilateral hip and knee arthroplasties who initially experienced infection of 1 of the implants were addressed by Akkaya et al 58 . Patients who developed ipsilateral metachronous PJI had significantly shorter stem-to-stem (tip-to-tip) distances (8 compared with 14 cm; p < 0.01) and empty (no foreign body such as cement) native bone distances (5 compared with 11 cm; p < 0.01), as well as an elevated risk of cement extrusion beyond its restrictor (70% compared with 5%; p < 0.01) compared with patients who did not develop infection.…”