A profile of patient response across surgical end-stage hip, knee and spine osteoarthritis

Abstract

Purpose: Orthopaedic surgical procedures for osteoarthritis (OA) are among the most frequently performed scheduled procedures in North America. Volumes are highest for hip and knee replacements and lumbar spinal surgery. However, patient reported response is variable with 10-40% of patients reporting persistent pain and subthreshold minimal clinical improvement. Given the potential patient and economic impact, an ability to predict whether a patient will ‘respond’ to surgery or not is of considerable interest. Unfortunately, predictive factors have been variable across studies and limited in dimension, making direct comparison across hip, knee and spine surgery for OA (HKS-OA) difficult. The Longitudinal Evaluation in the Arthritis Program (LEAP) is a large prospective observational cohort study of patients undergoing surgical intervention for HKS-OA (LEAP-OA). A bio-psycho-social prognostic framework was used to select patient factors/outcomes to profile patients and recovery trajectories over a 1-year period. This report represents the first ‘bird’s eye view’ of congruent and contemporaneous surgical HKS-OA patient profiles, examining similarities and differences across groups, and outcome response using a generic health status outcome. Methods: LEAP-OA recruited consecutive patients undergoing primary hip or knee replacement for OA or decompression (with/without fusion) surgery for back and leg symptoms due to lumbar spinal stenosis due to facet OA from Nov 2015 to Dec 2018. All patients provided written informed consent. Eligibility criteria included ≥35 years of age and English fluency. Exclusion criteria included acute trauma/injury, inflammatory arthritis, revision surgery. Clinical and self-reported sociodemographic, physical, mental and social health-related data were captured pre-surgery. The generic SF-12 health measure was collected pre-surgery and at 3 and 12 months post-surgery. Follow-up response rates were >80%. Descriptive analyses assessed baseline characteristics and change in the SF-12 physical component scores (PCS) over time across groups. The primary outcome was the proportion of responders at 3 and 12 months across groups achieving a minimally clinically important improvement (MCII) using PCS (>=4.6). Results: There were 547 hip, 625 knee and 140 spine OA patients with complete data. Mean ages were 65.1, 66.2, and 67.1 years, and 46.3%, 43.4% and 57.5% were female, among hip, knee and spine patients, respectively. The proportion of patients who were overweight or obese was highest for knee patients. Mean number of concurrent symptomatic joints were 2.3, 3.2 and 3.6 for hip, knee and spine patients, respectively. Clinically relevant levels of pain catastrophizing were reported by 18.8% of spine patients vs. 13.9% and 14.5% of hip and knee patients, respectively. While mean depressive symptom scores and anxiety scores were similar, rates of moderate/severe depressive symptoms were somewhat higher among spine patients, and moderate/severe anxiety symptoms among knee patients (Table 1). Approximately half of spine patients, and 35.2% and 29.6% of knee and hip patients had neuropathic pain scores indicative of possible/likely neuropathic pain using the painDETECT questionnaire. Average pain intensity ratings were similar across groups. Pre-surgery SF-12 PCS scores were similar across groups. Mean changes in PCS scores at 3 and 12 months post-surgery were greatest for hip patients and similar for knee and spine patients (Table 2). The majority of improvement in PCS scores occurred during the first 3 months, with continuing improvement to 12 months for all groups. By 3 months, hip patients were most likely to have achieved a SF-12 PCS MCII, and by 12 months achievement rates were 80.3%, 64.6% and 62.9% for hip, knee and spine, respectively. Reported rates of satisfaction with surgical results and willingness to have the surgery again were consistently high across groups at 90% or more. By 12 months, approximately 92% of hip patients, and 82% of knee and spine patients reported being able to cope with their current symptom state. Conclusions: A comprehensive set of patient and health-related factors provided the first opportunity to identify similarities and differences across a contemporaneous group of SHK-OA patients. Our Results are consistent with current literature and reflect that a significant proportion of patients, especially those undergoing knee and spine surgery for OA, do not reach a minimal response to improvement in physical health-related quality of life, despite a very high satisfaction rate. The baseline variables, response trajectories and rates varied between OA groups suggesting that there may be unique prognostic factors across groups that explain differences in responder rates. They may also represent distinct ‘classes’ of OA patients generally. Applied to the over 1.6 million inpatient spinal fusions, hip and knee replacements performed in the US in 2014, our non-response rates translate to approximately 500,000 SHK-OA surgeries a year at cost of $10billion USD. This represents a tremendous economic argument for improved pre-operative prognostic decision making (i.e. precision surgery). Advanced analytics techniques are the next steps to determine predictive phenotypes and improve point-of-care decision making.View Large Image Figure ViewerDownload Hi-res image Download (PPT)