Association Between Preoperative Mental Health and Clinically Meaningful Outcomes After Osteochondral Allograft for Cartilage Defects of the Knee: A Machine Learning Analysis

Abstract

Background: Fresh osteochondral allograft transplantation (OCA) is an effective method of treating symptomatic cartilage defects of the knee. This cartilage restoration technique involves the single-stage implantation of viable, mature hyaline cartilage into the chondral or osteochondral lesion. Predictive models for reaching the clinically meaningful outcome among patients undergoing OCA for cartilage lesions of the knee remain under investigation. Purpose: To apply machine learning to determine which preoperative variables are predictive for achieving the minimal clinically important difference (MCID) and substantial clinical benefit (SCB) at 1 and 2 years after OCA for cartilage lesions of the knee. Study Design: Case-control study; Level of evidence, 3. Methods: Data were analyzed for patients who underwent OCA of the knee by 2 high-volume fellowship-trained cartilage surgeons before May 1, 2018. The International Knee Documentation Committee questionnaire (IKDC), Knee Outcome Survey–Activities of Daily Living (KOS-ADL), and Mental Component (MCS) and Physical Component (PCS) Summaries of the 36-Item Short Form Health Survey (SF-36) were administered preoperatively and at 1 and 2 years postoperatively. A total of 84 predictive models were created using 7 unique architectures to detect achievement of the MCID for each of the 4 outcome measures and the SCB for the IKDC and KOS-ADL at both time points. Data inputted into the models included previous and concomitant surgical history, laterality, sex, age, body mass index (BMI), intraoperative findings, and patient-reported outcome measures (PROMs). Shapley Additive Explanations (SHAP) analysis identified predictors of reaching the MCID and SCB. Results: Of the 185 patients who underwent OCA for the knee and met eligibility criteria from an institutional cartilage registry, 135 (73%) patients were available for the 1-year follow-up and 153 (83%) patients for the 2-year follow-up. In predicting outcomes after OCA in terms of the IKDC, KOS-ADL, MCS, and PCS at 1 and 2 years, areas under the receiver operating characteristic curve (AUCs) of the top-performing models ranged from fair (0.72) to excellent (0.94). Lower baseline mental health (MCS), higher baseline physical health (PCS) and knee function scores (KOS-ADL, IKDC Subjective), lower baseline activity demand (Marx, Cincinnati sports), worse pain symptoms (Cincinnati pain, SF-36 pain), and higher BMI were thematic predictors contributing to failure to achieve the MCID or SCB at 1 and 2 years postoperatively. Conclusion: Our machine learning models were effective in predicting outcomes and elucidating the relationships between baseline factors contributing to achieving the MCID for OCA of the knee. Patients who preoperatively report poor mental health, catastrophize pain symptoms, compensate with higher physical health and knee function, and exhibit lower activity demands are at risk for failing to reach clinically meaningful outcomes after OCA of the knee.

Effect of Preoperative Imaging and Patient Factors on Clinically Meaningful Outcomes and Quality of Life After Osteochondral Allograft Transplantation: A Machine Learning Analysis of Cartilage Defects of the Knee

“…28 To determine preoperative factors associated with achieving clinically meaningful benefit after restorative cartilage procedures, Wang et al 27 performed a limited multivariate analysis on a combined cohort of patients treated with autograft mosaicplasty and OCA and reported that patients with lower preoperative activity levels and higher pain symptoms were more likely to fail to achieve meaningful clinical benefit. Although this analysis did not account for mental health, analyze preoperative factors simultaneously, or report predictive model performance, Ramkumar et al 23 applied ML to an OCA cohort of 185 patients and found that those with worse preoperative mental health and higher preoperative knee function were less likely to achieve meaningful clinical benefit in terms of QOL and knee function metrics, consistent with our findings. In the present study, ML was similarly chosen as the ideal process to simultaneously analyze a multitude of preoperative factors (anthropometric data, baseline PROMs, examination findings, imaging findings) reported heterogeneously (binary variables, categorical variables, continuous variables, etc) with unknown or differential weighted value in assessing postoperative clinical benefit.…”

“…Patients who demonstrate signs of disproportionately low mental health in light of their injuries warrant counseling that the surgery and rehabilitation may not necessarily translate to perceived clinical benefit owing to the underlying mental state. 23 Moreover, this population may be less motivated to commit to the postoperative rehabilitation. Similarly, patients who catastrophize pain symptoms may have alternative sources of pain or require counseling that surgical treatment of the cartilage lesion is unlikely to resolve all pain symptoms.…”

“…Similarly, patients who catastrophize pain symptoms may have alternative sources of pain or require counseling that surgical treatment of the cartilage lesion is unlikely to resolve all pain symptoms. 23 Relative importance factor analysis (SHAP analysis) of the models predictive of achieving meaningful functional improvement (IKDC, KOS-ADL) 2 years after OCA illustrated several important factors related to preoperative knee function, alignment, and BMI. Patients with higher preoperative knee function were less likely to achieve meaningful clinical benefit.…”

“…Ramkumar et al 23 first employed ML in cartilage restoration to demonstrate that several models had excellent predictive power in simultaneously assessing demographic, anthropometric, and baseline PROMs to identify patients likely to reach clinically meaningful outcomes. Patients who preoperatively report poor mental health, catastrophize pain symptoms, compensate with higher physical health and knee function, and exhibit lower activity demands were found to be at risk of failing to reach clinically meaningful outcomes (MCID or SCB) after OCA of the knee.…”

Can minimal clinically important differences in patient reported outcome measures be predicted by machine learning in patients with total knee or hip arthroplasty? A systematic review

“…From the 18 articles screened for full text, two studies were excluded due to including the wrong study population from the perspective of this review. One study [ 48 ] included knee arthroscopies and one [ 49 ] patients with osteochondral allograft for cartilage defects. Next, five studies [ 50 – 54 ] did not calculate MCIDs, even though they researched PROMs in TJA patients.…”