Accuracy of Predictive Algorithms in Total Hip and Knee Arthroplasty Acute Periprosthetic Joint Infections Treated With Debridement, Antibiotics, and Implant Retention (DAIR)

KLIC Score Does Not Predict Failure After Early Prosthetic Joint Infection: An External Validation With 153 Knees and 130 Hips

BACKGROUND

A preoperative risk score, the KLIC score (chronic renal failure [K], liver cirrhosis [L], indication of the index surgery [I], cemented prosthesis [C], and C-reactive protein >115 mg/L), has been developed to predict the risk of treatment failure after early prosthetic joint infection (PJI). This study aimed to validate the KLIC score for the debridement, antibiotics, and implant retention (DAIR) procedure and one-stage revisions in a Northern European cohort.

METHODS

Revisions due to early PJI of the hip or knee between January 1, 2008, and September 12, 2021, were identified retrospectively. The primary outcome was early failure, which was considered when the patient needed an unscheduled surgery, the patient died, or the patient was prescribed long-term suppressive antibiotics. To examine the association between KLIC score and failure risk, univariable logistic regression with area under the curve (AUC) was used. In addition, models were calibrated to assess prognostic ability and clinical utility was examined with decision-curve analyses.

RESULTS

An increase in KLIC score had a moderate predictive value for early failure after DAIR (odds ratio [OR] 1.45; confidence interval [CI] 1.13 to 1.90). For one-stage revision, it was only slightly predictive of failure (OR 1.20; CI 0.93 to 1.56). After 60 days, the AUC for DAIR was 0.63 (CI 0.55 to 0.72) and 0.56 (CI 0.46 to 0.66) for one-stage revisions, indicating poor discriminative ability. The decision-curve analyses revealed that the model did not offer a remarkable net benefit across a range of threshold probabilities.

CONCLUSIONS

We demonstrated that the KLIC score is not a reliable predictor of early failure after early PJI in a Northern European cohort. Using the model to guide treatment decisions does not provide any additional clinical utility beyond the baseline strategies.

Perioperative Demographic and Laboratory Characteristics of Failed Debridement, Antibiotics, and Implant Retention: Can We Determine Which Patients Will Fail?

BACKGROUND

Debridement, antibiotics, and implant retention (DAIR) are the mainstays surgical treatment for acute periprosthetic joint infection (PJI). However, reoperation following DAIR is common and the risk factors for DAIR failure remain unclear. This study aimed to assess the perioperative characteristics of patients who failed initial DAIR treatment.

METHODS

A retrospective review was conducted on 83 patients who underwent DAIR for acute PJI within 3 months following index surgery from 2011 to 2022, with a minimum one-year follow-up. Surgical outcomes were categorized using the Musculoskeletal Infection Society outcome reporting tool (Tiers 1 to 4). Patient demographics, laboratory data, and perioperative outcomes were compared between patients who had failed (Tiers 3 and 4) (n = 32) and successful (Tiers 1 and 2) (n = 51) DAIR treatment. Logistic regression was also performed.

RESULTS

After logistic regression, Charlson Comorbidity Index (odds ratio [OR]: 1.57; P = .003), preoperative C-reactive protein (OR: 1.06; P = .014), synovial white blood cell (OR: 1.14; P = .008), and polymorphonuclear cell (PMN%) counts (OR: 1.05; P = .015) were independently associated with failed DAIR. Compared with total hip arthroplasty, total knee arthroplasty patients (OR: 6.08; P = .001) were at increased risk of DAIR failure. The type of organism and time from primary surgery were not correlated with DAIR failure.

CONCLUSIONS

Patients who had failed initial DAIR tended to have significantly higher Charlson Comorbidity Index, C-reactive protein, synovial white blood cell, and PMN%. The total knee arthroplasty DAIRs were more likely to fail than the total hip arthroplasty DAIRs. These characteristics should be considered when planning acute PJI management, as certain patients may be at higher risk for DAIR failure and may benefit from other surgical treatments.

LEVEL OF EVIDENCE

III.

Debridement, antibiotics, and implant retention (DAIR) for the early prosthetic joint infection of total knee and hip arthroplasties: a systematic review

PURPOSE

Early periprosthetic joint infection (PJI) represents one of the most fearsome complications of joint replacement. No international consensus has been reached regarding the best approach for early prosthetic knee and hip infections. The aim of this updated systematic review is to assess whether debridement, antibiotics, and implant retention (DAIR) is an effective choice of treatment in early postoperative and acute hematogenous PJI.

METHODS

This systematic review was performed using the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. The diagnostic criteria defining a PJI, the most present pathogen, and the days between the index procedure and the onset of the PJI were extracted from the selected articles. Additionally, the mean follow-up, antibiotic regimen, and success rate of the treatment were also reported.

RESULTS

The articles included provided a cohort of 970 patients. Ten studies specified the joint of their cohort in PJIs regarding either hip prostheses or knee prostheses, resulting in 454 total knees and 460 total hips. The age of the patients ranged from 18 to 92 years old. Success rates for the DAIR treatments in the following cohort ranged from 55.5% up to a maximum of 90% (mean value of 71%).

CONCLUSION

Even though the DAIR procedure is quite limited, it is still considered an effective option for patients developing an early post-operative or acute hematogenous PJI. However, there is a lack of studies, in particular randomized control trials (RCTs), comparing DAIR with one-stage and two-stage revision protocols in the setting of early PJIs, reflecting the necessity to conduct further high-quality studies to face the burden of early PJI.

Outcomes After Successful Debridement, Antibiotic, and Implant Retention Therapy for Periprosthetic Joint Infection in Total Knee Arthroplasty

BACKGROUND

Controversy surrounds debridement, antibiotic and implant retention (DAIR) for treatment of acute periprosthetic joint infection (PJI). Data regarding DAIR's rate of infection resolution is variable with little investigation of functional outcomes.

METHODS

We identified 191 DAIR cases at a single institution from 2008 to 2020. Knee Injury and Osteoarthritis Outcome Score for Joint Replacement (KOOS-JR) and Patient Reported Outcome Measurement Information System-10 (PROMIS) scores were collected. Patient Reported Outcome Scores were obtained from 60 cases. Median follow-up (IQR) was 4.5 (2.1 to 7.6) versus 3.0 (1.2 to 5.3) years for the control group. Mean scores were compared to a cohort of uncomplicated total knee arthroplasties matched by age, body mass index, and sex using generalized linear models adjusted for follow-up duration. Kaplan-Meier survivorship curves for PJI were constructed.

RESULTS

Mean Patient Reported Outcome Scores for the DAIR cohort were 57.2 ± 19.7 for KOOS-JR, 41.6 ± 7.1 for PROMIS physical health (PH), and 46.6 ± 8.7 for PROMIS mental health (MH). Mean control group values were 65.8 ± 21.0 for KOOS-JR, 44.6 ± 8.4 for PROMIS PH, and 49.2 ± 9.2 for PROMIS MH. No difference was observed in KOOS-JR (P = .83) or PROMIS MH (P = .11). PROMIS PH was lower in the DAIR cohort compared to the control group (P = .048). Median follow-up (years) for all 191 cases was 5.9 (range, 0.5 to 13.1). Survivorship (years) without subsequent operation for infection was 84% at 1, 82% at 2, and 79% at 5.

CONCLUSION

Knee and MH outcomes in successful DAIR procedures were similar to uncomplicated total knee arthroplasty. DAIR's success rate was 79% in treating acute PJI at 5 years.

LEVEL OF EVIDENCE

Level III.

A machine learning-based model for "In-time" prediction of periprosthetic joint infection

Background

Previous criteria had limited value in early diagnosis of periprosthetic joint infection (PJI). Here, we constructed a novel machine learning (ML)-derived, "in-time" diagnostic system for PJI and proved its validity.

Methods

We filtered "in-time" diagnostic indicators reported in the literature based on our continuous retrospective cohort of PJI and aseptic prosthetic loosening patients. With the indicators, we developed a two-level ML model with six base learners including Elastic Net, Linear Support Vector Machine, Kernel Support Vector Machine, Extra Trees, Light Gradient Boosting Machine and Multilayer Perceptron), and one meta-learner, Ensemble Learning of Weighted Voting. The prediction performance of this model was compared with those of previous diagnostic criteria (International Consensus Meeting in 2018 (ICM 2018), etc.). Another prospective cohort was used for internal validation. Based on our ML model, a user-friendly web tool was developed for swift PJI diagnosis in clinical practice.

Results

A total of 254 patients (199 for development and 55 for validation cohort) were included in this study with 38.2% of them diagnosed as PJI. We included 21 widely accessible features including imaging indicators (X-ray and CT) in the model. The sensitivity and accuracy of our ML model were significantly higher than ICM 2018 in development cohort (90.6% vs. 76.1%, P = 0.032; 94.5% vs. 86.7%, P = 0.020), which was supported by internal validation cohort (84.2% vs. 78.6%; 94.6% vs. 81.8%).

Conclusions

Our novel ML-derived PJI "in-time" diagnostic system demonstrated significantly improved diagnostic potency for surgical decision-making compared with the commonly used criteria. Moreover, our web-based tool greatly assisted surgeons in distinguishing PJI patients comprehensively.

Level of evidence

Diagnostic Level III.

Accuracy of a Novel Preoperative Failure Risk Model for Debridement Antibiotics and Implant Retention (DAIR) in Acute Prosthetic Joint Infection

Given the variable success of the debridement, antibiotics and implant retention (DAIR) procedure in patients with acute prosthetic joint infection (PJI), an accurate selection of candidates is critical. In this study, we set about calculating the predictive value of a novel algorithm for predicting outcome following DAIR developed by Shohat et al. Sixty-four patients who underwent debridement for (early and late) acute PJI in a tertiary-level university hospital were selected, and the aforementioned algorithm was retrospectively applied. Patients with model scores of 40−50%, 50−60%, 60−70%, 70−80% and 80−90% displayed success rates of 33.34%, 41.18%, 57.9%, 78.27% and 100%, respectively. The receiver operating characteristic curve showed an area under the curve of 0.69. The calibration intercept value was 0, and the calibration slope value was 1. Failure rates were significantly higher for the following variables: revision surgery (p = 0.012) index surgery for reasons other than osteoarthritis (p = 0.01), and C-reactive protein level >30 mg/L (p = 0.042). This analysis demonstrated that the Shohat algorithm is associated with an optimal calibration value and a moderate predictive value for failure of a DAIR procedure in patients with acute PJI. Its validation is recommended before it can be routinely applied in daily practice.

Methodological Challenges in Predicting Periprosthetic Joint Infection Treatment Outcomes: A Narrative Review

The management of periprosthetic joint infection (PJI) generally requires both surgical intervention and targeted antimicrobial therapy. Decisions regarding surgical management–whether it be irrigation and debridement, one-stage revision, or two-stage revision–must take into consideration an array of factors. These include the timing and duration of symptoms, clinical characteristics of the patient, and antimicrobial susceptibilities of the microorganism(s) involved. Moreover, decisions relating to surgical management must consider clinical factors associated with the health of the patient, alongside the patient's preferences. These decisions are further complicated by concerns beyond mere eradication of the infection, such as the level of improvement in quality of life related to management strategies. To better understand the probability of successful surgical treatment of a PJI, several predictive tools have been developed over the past decade. This narrative review provides an overview of available clinical prediction models that aim to guide treatment decisions for patients with periprosthetic joint infection, and highlights key challenges to reliably implementing these tools in clinical practice.

Prosthetic Knee Infection With Coagulase-Negative Staphylococcus: A Harbinger of Poor Outcomes

BACKGROUND

Coagulase-negative staphylococci (CoNS) are biofilm-producing pathogens whose role in periprosthetic joint infection (PJI) is increasing. There is little data on the prognosis and treatment considerations in the setting of PJI. We sought to evaluate the clinical characteristics, outcomes, and complications in these patients.

METHODS

This is a retrospective cohort study of adult patients at a single tertiary medical center from 2009 to 2020 with culture-proven CoNS PJI after total knee arthroplasty, as diagnosed by Musculoskeletal Infection Society criteria. The primary outcome was treatment success, with failure defined as recurrent CoNS PJI, recurrent PJI with a new pathogen, and/or chronic oral antibiotic suppression at one year postoperatively.

RESULTS

We identified 55 patients with a CoNS total knee arthroplasty PJI with a mean follow-up of 29.8 months (SD: 16.3 months). The most commonly isolated organism was Staphylococcus epidermidis (n = 36, 65.5%). The overall prevalence of methicillin resistance was 63%. Surgical treatment included surgical debridement, antibiotics, and implant retention in 25 (45.5%) cases and two-stage revision (22 articulating and eight static antibiotic-impregnated spacers). At one-year follow-up, only 47% of patients had successful management of their infection. The surgical debridement, antibiotics, and implant retention cohort had the higher rate of treatment failure (60.0%) compared to two-stage revision (46.7%).

CONCLUSION

These results indicate a poor rate of success in treating CoNS PJI. This likely represents the interplay of inherent virulence through biofilm formation and decreased antibiotic efficacy.

Risk Factors of DAIR Failure and Validation of the KLIC Score: A Multicenter Study of Four Hundred Fifty-Five Patients

Background: Debridement, antibiotic agents, and implant retention (DAIR) is a currently accepted approach for the treatment of early prosthetic joint infections (PJI). The success of a DAIR procedure has shown variable results throughout the published literature. Scoring systems such as the Kidney, Liver, Index surgery, Cemented prosthesis, and C-reactive protein value (KLIC) score for the selection of patients that are likely to benefit from DAIR have proved to be helpful in decision making. Our study aims to further validate the KLIC score using a large external multicentric cohort and to evaluate other risk factors for failure. Patients and Methods: A retrospective analysis of patients with an early acute PJI who were treated with DAIR and recorded in a database of eight Spanish university hospitals was performed. According to pre-operative variables of the KLIC study, patients were categorized into five groups: group A, ≤2 points; group B, 2.5-3.5 points; group C, 4-5 points; group D, 5.5-6.5 points; and group E, ≥7 points. Failure rates were compared between groups at 60 days and after 60 days of DAIR. Further variables for risk of failure were also analyzed. Results: A total of 455 patients with early acute PJI were included in the analyses. At 60 days, patients presenting with pre-operative elevated C-reactive protein serum levels, Staphylococcus aureus, and polymicrobial infections were associated with failure. Failure rates recorded were 12% for group A (n = 210), 18% for group B (n = 83), 26% for group C (n = 89), 24% for group D (n = 66), and 0% for group E (n = 7). Univariable analysis between consecutive groups of the KLIC score showed no differences for failure before 60 days of the DAIR procedure. Scheduled surgery and having the procedure performed by a specialized unit were also identified as important factors for DAIR success. Conclusions: Our results suggest the KLIC score was not useful for predicting failure in our cohort. Furthermore, our results indicate a specialized unit should conduct DAIR procedures.