Nationwide Results of Microorganism Antigen Testing as a Component of Preoperative Synovial Fluid Analysis

Probability Score for the Diagnosis of Periprosthetic Joint Infection: Development and Validation of a Practical Multi-analyte Machine Learning Model

Background and objective The diagnosis of periprosthetic joint infection (PJI) relies on established criteria-based systems requiring interpretation and combination of multiple laboratory tests into scoring systems. In routine clinical care, clinicians implement these algorithms to diagnose PJI. Existing literature indicates suboptimal adoption and implementation of these criteria in clinical practice, even among experts. Recognizing the need for accurate PJI diagnosis through proper synthesis of multiple laboratory parameters, this study aimed to develop and validate a machine learning (ML) model that generates a preoperative PJI probability score based solely on synovial fluid (SF) biomarkers within 24 hours. Materials and methods A two-stage ML model was constructed using 104,090 SF samples from 2,923 institutions (2018-2024). First, unsupervised learning identified natural clusters in the data to label samples as "infected" or "not infected." Then, these labels trained a supervised logistic regression model that generated PJI scores (0-100), categorizing cases as PJI positive (> 80), PJI negative (< 20), or equivocal (20-80). The model incorporated 10 SF biomarkers: specimen integrity markers (absorbance at 280 nm, red blood cell count), inflammatory markers (white blood cell count, percentage of neutrophils, SF C-reactive protein), a PJI-specific biomarker (alpha-defensin), and microbial antigen markers (Staphylococcus, Enterococcus, Candida, and Cutibacterium acnes). Notably, culture results were excluded to allow for a 24-hour diagnosis. After splitting data into training (n = 83,272) and validation (n = 20,818) cohorts, performance was assessed against modified 2018 International Consensus Meeting criteria, including evaluation with probabilistically reclassified "inconclusive" cases. Results The ML model and resulting PJI score showed high diagnostic accuracy in the validation cohort. The PJI score achieved 99.3% sensitivity and 99.5% specificity versus the clinical reference before reclassification of inconclusive cases and 98.1% sensitivity and 97.6% specificity after probabilistic reclassification. With a disease prevalence of 20.7%, the positive predictive value reached 91.5% and the negative predictive value 99.5%. The model resolved 95% (1,363/1,442) of samples deemed inconclusive by the clinical standard. The analysis identified alpha defensin, percentage of neutrophils, and white blood cell count as the most influential model features. The model performed well in culture-negative infections. Conclusions The ML model and resulting PJI score demonstrated exceptional diagnostic accuracy by leveraging unsupervised SF biomarker pattern clustering. The model substantially reduced diagnostic uncertainty by definitively classifying most inconclusive cases, revealing their natural alignment with infected or non-infected patterns. This performance was achieved without SF culture results, enabling definitive diagnostic information within 24 hours based solely on biomarkers. The clinical significance demonstrates that an ML algorithm can match the diagnostic accuracy of complex clinical standards while transferring analytical complexity from clinicians to laboratories, minimizing the implementation gap that hinders current criteria-based approaches.

The Application of artificial intelligence in periprosthetic joint infection

Periprosthetic joint infection (PJI) represents one of the most devastating complications following total joint arthroplasty, often necessitating additional surgeries and antimicrobial therapy, and potentially leading to disability. This significantly increases the burden on both patients and the healthcare system. Given the considerable suffering caused by PJI, its prevention and treatment have long been focal points of concern. However, challenges remain in accurately assessing individual risk, preventing the infection, improving diagnostic methods, and enhancing treatment outcomes. The development and application of artificial intelligence (AI) technologies have introduced new, more efficient possibilities for the management of many diseases. In this article, we review the applications of AI in the prevention, diagnosis, and treatment of PJI, and explore how AI methodologies might achieve individualized risk prediction, improve diagnostic algorithms through biomarkers and pathology, and enhance the efficacy of antimicrobial and surgical treatments. We hope that through multimodal AI applications, intelligent management of PJI can be realized in the future.

Commercial Synovial Antigen Testing is Inferior to Traditional Culture for the Diagnosis of Periprosthetic Joint Infection in Patients Undergoing Revision Total Knee Arthroplasty

BACKGROUND

Despite its limitations, a culture remains the "gold standard" for pathogen identification in patients who have periprosthetic joint infection (PJI). Recently, a synovial fluid antigen test has been introduced by a commercial entity. The purpose of this multicenter study was to determine the accuracy of the antigen test in the diagnosis of PJI.

METHODS

This retrospective study identified 613 patients undergoing revision total knee arthroplasty who had undergone preoperative synovial fluid analysis. A PJI was defined using the 2018 International Consensus Meeting (ICM) criteria. Patients who had an extended period (> 180 days) from aspiration to revision procedure (n = 62), those presenting within 90 days of their index arthroplasty procedure (n = 17), and patients who had an inconclusive ICM score (n = 8) were excluded. Using receiver operator characteristic curve analyses, we examined the utility of the microbial identification (MID) antigen test and any positive culture (either preoperative or intraoperative) in the diagnosis of PJI.

RESULTS

A total of 526 patients were included. Of these, 125 (23.8%) were ICM positive and 401 (76.2%) were ICM negative. Culture demonstrated an area under the curve (AUC) of 0.864, sensitivity of 75.2%, and specificity of 97.5%. On the other hand, the MID test exhibited an AUC of 0.802, sensitivity of 61.6%, and specificity of 98.8%. The AUC of culture was significantly higher than that of the MID test (P = .037). The MID test was positive in 41.9% of culture-negative PJI cases. We also observed a high rate of discordance (29.7%) when both culture and the MID test were positive in the ICM-positive group.

CONCLUSIONS

Synovial fluid antigen testing does not provide additional clinical benefit when compared to traditional cultures for the diagnosis of PJI. The antigen test had low sensitivity in the diagnosis of PJI and a relatively high rate of discordance with culture.

LEVEL OF EVIDENCE

Level III.

Update on the role of pathology and laboratory medicine in diagnosing periprosthetic infection

Technological and implant design advances have helped reduce the frequency of aseptic total joint arthroplasty failure, but periprosthetic joint infections (PJI) remain a clinical important problem with high patient morbidity. Misinterpreting PJI as aseptic mechanical loosening commonly leads to unsatisfactory revision arthroplasty, persistent infection, and poor long-term results. While there is no single "gold standard" diagnostic test for PJI, recent collaborative efforts by Orthopaedic and Infectious Disease Societies have developed algorithms for diagnosing PJI. However, the efficacy of individual tests as well as diagnostic thresholds are controversial. We review the recommended thresholds for commonly used screening tests as well as tissue histopathology and confirmatory tests to diagnose periprosthetic infection. We also update lesser-known laboratory tests, and we briefly summarize rapidly evolving molecular tests to diagnose periprosthetic infection. Pathologists hold a critical role in assisting with PJI diagnosis, maintaining laboratory test quality and interpreting test results. Collaboration between clinicians and pathologists is essential to provide optimal patient care and reduce the burden of PJI.

The False-Positive Rate of Synovial Fluid Culture at a Single Clinical Laboratory Using Culture Bottles

Introduction Synovial fluid (SF) cultures can yield false-positive or negative results when diagnosing periprosthetic joint infection (PJI). False-positives may arise during sample collection or from laboratory contamination. Understanding false-positive SF culture rates is crucial for interpreting PJI laboratory data, yet clinical laboratories rarely report these rates. This study aimed to define the false-positive SF culture rate at a major specialized clinical laboratory. Methods This study retrospectively analyzed prospectively collected data at a single clinical laboratory that receives SF for clinical testing for PJI. A total of 180,317 periprosthetic SF samples from the hip, knee, and shoulder were identified from January 2016 to December 2023, which met the inclusion criteria for this study. Samples were classified by both a modified 2018 International Consensus Meeting (ICM) score and an inflammation score that combined the SF-C-reactive protein, alpha-defensin, SF-white blood cell count, and SF-polymorphonuclear% into one standardized metric. Logistic regression was utilized to evaluate the impact of various collection-based characteristics on culture positivity, including inflammation biomarkers, the source joint, quality control metrics, and days of specimen transport to the laboratory. SF culture false-positivity was calculated based on the ICM category of "not-infected" or low inflammation score. Results Overall, 13.3% (23,974/180,317) of the samples were associated with a positive culture result: 12.5% for knee samples, 20.3% for hip samples, and 14.7% for shoulder samples. The false-positive SF culture rate among 131,949 samples classified as "not-infected" by the modified 2018 ICM definition was 0.47% (95%CI: 0.43 to 0.51%). Stratification by joint revealed a false-positive rate of 0.34% (95%CI: 0.31 to 0.38%) for knee samples, 1.24% (95%CI: 1.05 to 1.45%) for hip samples, and 3.02% (95%CI: 2.40 to 3.80%) for shoulder samples, with p < 0.0001 for all comparisons. The false-positive SF culture rate among 90,156 samples, representing half of all samples with the lowest standardized inflammation scores, was 0.47% (95%CI: 0.43 to 0.52%). Stratification by joint revealed a false-positive rate of 0.33% (95%CI: 0.29 to 0.37%) for knee samples, 1.45% (95%CI: 1.19 to 1.77%) for hip samples, and 3.09% (95%CI: 2.41 to 3.95%) for shoulder samples, with p<0.0001 for all comparisons. Multivariate logistic regression demonstrated the joint source (hip, shoulder) and poor sample quality as collection-based factors associated with a false-positive culture. Evaluation of a cohort of samples selected to minimize collection-based causes of false-positive culture demonstrated a false-positive rate of 0.30%, representing the ceiling limit for laboratory-based SF culture false-positivity. Conclusions This study utilizes two methods to estimate the false-positive SF culture rate at a single specialized clinical laboratory, demonstrating an overall false-positive rate of approximately 0.5%. Stratification of samples by source joint demonstrated that periprosthetic SF from the shoulder and hip have a substantially higher false-positive culture rate than that of the knee. The lowest false-positive SF culture rate (0.30%) was observed among samples from the knee-passing quality control. Culture positivity due to contamination at this specific laboratory is less than 0.30% because all specimens undergo identical processing.

Achieving High Accuracy in Predicting the Probability of Periprosthetic Joint Infection From Synovial Fluid in Patients Undergoing Hip or Knee Arthroplasty: The Development and Validation of a Multivariable Machine Learning Algorithm

Background and objective The current periprosthetic joint infection (PJI) diagnostic guidelines require clinicians to interpret and integrate multiple criteria into a complex scoring system. Also, PJI classifications are often inconclusive, failing to provide a clinical diagnosis. Machine learning (ML) models could be leveraged to reduce reliance on these complex systems and thereby reduce diagnostic uncertainty. This study aimed to develop an ML algorithm using synovial fluid (SF) test results to establish a PJI probability score. Methods We used a large clinical laboratory's dataset of SF samples, aspirated from patients with hip or knee arthroplasty as part of a PJI evaluation. Patient age and SF biomarkers [white blood cell count, neutrophil percentage (%PMN), red blood cell count, absorbance at 280 nm wavelength, C-reactive protein (CRP), alpha-defensin (AD), neutrophil elastase, and microbial antigen (MID) tests] were used for model development. Data preprocessing, principal component analysis, and unsupervised clustering (K-means) revealed four clusters of samples that naturally aggregated based on biomarker results. Analysis of the characteristics of each of these four clusters revealed three clusters (n=13,133) with samples having biomarker results typical of a PJI-negative classification and one cluster (n=4,032) with samples having biomarker results typical of a PJI-positive classification. A decision tree model, trained and tested independently of external diagnostic rules, was then developed to match the classification determined by the unsupervised clustering. The performance of the model was assessed versus a modified 2018 International Consensus Meeting (ICM) criteria, in both the test cohort and an independent unlabeled validation set of 5,601 samples. The SHAP (SHapley Additive exPlanations) method was used to explore feature importance. Results The ML model showed an area under the curve of 0.993, with a sensitivity of 98.8%, specificity of 97.3%, positive predictive value (PPV) of 92.9%, and negative predictive value (NPV) of 99.8% in predicting the modified 2018 ICM diagnosis among test set samples. The model maintained its diagnostic accuracy in the validation cohort, yielding 99.1% sensitivity, 97.1% specificity, 91.9% PPV, and 99.9% NPV. The model's inconclusive rate (diagnostic probability between 20-80%) in the validation cohort was only 1.3%, lower than that observed with the modified 2018 ICM PJI classification (7.4%; p<0.001). The SHAP analysis found that AD was the most important feature in the model, exhibiting dominance among >95% of "infected" and "not infected" diagnoses. Other important features were the sum of the MID test panel, %PMN, and SF-CRP. Conclusions Although defined methods and tools for diagnosis of PJI using multiple biomarker criteria are available, they are not consistently applied or widely implemented. There is a need for algorithmic interpretation of these biomarkers to enable consistent interpretation of the results to drive treatment decisions. The new model, using clinical parameters measured from a patient's SF sample, renders a preoperative probability score for PJI which performs well compared to a modified 2018 ICM definition. Taken together with other clinical signs, this model has the potential to increase the accuracy of clinical evaluations and reduce the rate of inconclusive classification, thereby enabling more appropriate and expedited downstream treatment decisions.