Multivariate use of MRI biomarkers to classify histologically confirmed necrosis in symptomatic total hip arthroplasty

Abstract

The failure of total hip arthroplasty (THA) is commonly associated with the necrosis of the periprosthetic tissue. To date, there is no established method to noninvasively quantify the progression of such necrosis. Magnetic resonance imaging (MRI) of soft tissues near implants has undergone a recent renaissance due to the development of multispectral metal‐artifact reduction techniques. Advanced analysis of multispectral MRI has been shown capable of detecting small magnetism effects of metallic debris in periprosthetic tissue. The purpose of this study is to demonstrate the diagnostic utility of these MRI‐based tissue‐magnetism signatures. Together with morphological MRI metrics, such as synovial volume and thickness, these measurements are utilized as biomarkers to noninvasively detect soft‐tissue necrosis in symptomatic THA patients (N=78). All subjects underwent an advanced MRI scan before revision surgery and tissue biopsies utilized for necrosis grading. Statistical analyses demonstrated a weak, but significant positive correlation (P = .04) between MRI magnetism signatures and necrosis scores, while indicating no meaningful association between the latter and serum cobalt and chromium ion levels. Receiver‐operating characteristic (ROC) analyses were then performed based on uni‐ and multivariate logistic regression models utilizing the measured MRI biomarkers as predictors of severe necrosis. The area under the curve of the ROC plots for MRI biomarkers as combined predictors were found to be 0.70 and 0.84 for cross‐validation and precision‐recall tests, respectively.

Artificial Learning and Machine Learning Decision Guidance Applications in Total Hip and Knee Arthroplasty: A Systematic Review

“…Sixteen studies (32.7%) evaluated AI/ML applications to accurately predict patient reoperations, operating time, hospital LOS, discharges, readmissions, or surgical and inpatient costs [ [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] , [27] , [28] , [29] ]. In addition, 16 studies (32.7%) used patients’ preoperative risk factors and other patient-specific variables to optimize the patient selection and surgical planning process through the use of AI/ML-based predictions of surgical outcomes and postoperative complications [ [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] , [39] , [40] , [41] , [42] , [43] , [44] ]. The majority of the decision support studies evaluated AI/ML model performance using receiver operating characteristic/AUC, accuracy, sensitivity, and specificity.…”

“…Hip and knee arthroplasty typically involve an older and highly comorbid patient population, and these tools can be especially helpful in identifying patient-specific needs and risks within this population. Examples of how these models can enable providers to create and optimize personalized treatment plans include accurate identification of an implant from a previous surgery for revision procedures and classifying total knee arthroplasty (TKA) surgical candidates based on patient-specific risk factors [29][30][31][33][34][35][37][38][39][40][41][42]44,62]. Hyer et al demonstrated an AI/ML model which classified TKA and total hip arthroplasty patients based on surgical complexity scores [19].…”

MRI Characteristics of Adverse Local Tissue Reactions Associated With Intraoperative Tissue Damage and Poor Revision Surgery Outcomes: A Systematic Review

“…24 Three studies did not discuss how the volume was calculated. 5,21,22 Of the five studies evaluating intraoperative histology, three identified synovial volume as a predictor of ALVAL grade. 10,12,19 Burge et al 11 were unable to show a relationship between synovial volume and ALVAL score.…”

“…Both studies evaluating intraoperative tissue necrosis identified a correlation synovial volume. 14,19 In addition, Sherafati et al 22 performed a logistical regression analysis suggesting that synovial volume was likely the key imaging metrics related to increasing tissue necrosis. Both studies looking at postoperative complications and re-revision found no correlation with synovial volume and these dependent variables.…”

“…Metal artifact reduction MRI was performed in all 15 studies. 5,[9][10][11][12][13][14][15][16][17][18][19][20][21][22] Four studies used the Hauptfleisch grading system for ALTR classification. 9,13,15,16 This defined cystic lesions as either type I with synovial wall thickness ,3 mm or type II with synovial wall thickness .3 mm.…”