ZFOS Center for Orthopaedics and Sports Medicine, Munich, Germany; Department of Orthopedics and Sports Orthopedics, Technical University Munich, Munich, Germany
Auto-generated profile, based on publicly available information
Christian Suren
ZFOS Center for Orthopaedics and Sports Medicine, Munich, Germany; Department of Orthopedics and Sports Orthopedics, Technical University Munich, Munich, Germany
Background: Machine Learning (ML) in arthroplasty is becoming more popular, as it is perfectly suited for prediction models. However, results have been heterogeneous so far. We hypothesize that an accurate ML model for outcome prediction in THA must be able to compute arthroplasty-specific data. In this study, we evaluate a ML approach applying data from two German arthroplasty-specific registries to predict adverse outcomes after THA, after careful evaluations of ML algorithms, outcome and input variables by an interdisciplinary team of data scientists and surgeons. Methods: Data of 1217 cases of primary THA from a single center were derived from two German arthroplasty-specific registries between 2016 to 2019. The XGBoost algorithm was adjusted and applied. Accuracy, sensitivity, specificity and AUC were calculated. Results: For the prediction of complications, the ML algorithm achieved an accuracy of 80.3%, a sensitivity of 31.0%, a specificity of 89.4% and an AUC of 64.1%. For the prediction of surgery duration, the ML algorithm yielded an accuracy of 81.7%, a sensitivity of 58.2%, a specificity of 91.6% and an AUC of 89.1%. The feature importance indicated non-linear outcomes for age, height, weight and surgeon. No relevant linear correlations were found. Conclusion: The attunement of input and output data as well as the modifications of the ML algorithm permitted the development of a feasible ML model for the prediction of complications and surgery duration.
Purpose: The diagnostic criteria of prosthetic joint infection (PJI) recommended by the most commonly used diagnostic algorithms can be obscured or distorted by other inflammatory processes or aseptic pathology. Furthermore, the most reliable diagnostic criteria are garnered during revision surgery. A robust, reliable addition to the preoperative diagnostic cascade is warranted. Calprotectin has been shown to be an excellent diagnostic marker for PJI. In this study, we aimed to evaluate a lateral flow test (LFT) in the challenging patient cohort of a national referral centre for revision arthroplasty. Methods: Beginning in March 2019, we prospectively included patients scheduled for arthroplasty exchange of a total hip (THA) or knee arthroplasty (TKA). Synovial fluid samples were collected intra-operatively. We used the International Consensus Meeting of 2018 (ICM) score as the gold standard. We then compared the pre-operative ICM score with the LFT result to calculate its diagnostic accuracy as a standalone pre-operative marker and in combination with the ICM score as part of an expanded diagnostic workup. Results: A total of 137 patients with a mean age of 67 (± 13) years with 53 THA and 84 TKA were included. Ninety-nine patients (72.8%) were not infected, 34 (25.0) were infected, and four (2.9%) had an inconclusive final score and could not be classified after surgery. The calprotectin LFT had a sensitivity (95% confidence interval) of 0.94 (0.80–0.99) and a specificity of 0.87 (0.79–0.93). The area under the receiver operating characteristic curve (AUC) for the calprotectin LFT was 0.94 (0.89–0.99). In nine cases with an inconclusive pre-operative ICM score, the calprotectin LFT would have led to the correct diagnosis of PJI. Conclusions: The synovial fluid calprotectin LFT shows excellent diagnostic metrics both as a rule-in and a rule-out test, even in a challenging patient cohort with cases of severe osteolysis, wear disease, numerous preceding surgeries, and poor soft tissue conditions, which can impair the common diagnostic criteria. As it is available pre-operatively, this test might prove to be a very useful addition to the diagnostic algorithm.
30 March 2025 in Journal of Shoulder and Elbow Surgery