Background/Purpose: The interpretation of patient-reported outcomes, such as the Knee injury and Osteoarthritis Outcome Score (KOOS) questionnaire, can prove difficult if benchmarks are lacking. Therefore, we recently developed KOOS percentile curves with data from a Dutch population-based cohort (Loef et al. 2020. DOI:10.1016/j.joca.2020.03.014). With the present study, we aimed to investigate the application of the KOOS percentile curves using pre- and postoperative data of patients with knee osteoarthritis (OA) undergoing total knee arthroplasty (TKA).

Methods: We used data of the Longitudinal Leiden Orthopaedics Outcomes of Osteo-Arthritis study (LOAS) (van de Water et al. 2019. DOI: 10.2106/JBJS.18.00642). The current analyses are comprised of patients who have been included from June 2012 until June 2017, who were between 45 and 65 years of age, and undergoing primary TKA.KOOS subscale scores (0-100) were obtained preoperatively and 6, 12 and 24 months after TKA. Preoperative radiographs of the operated knee were assessed for radiographic OA severity according to the Kellgren-Lawrence (KL) grading system by an experienced musculoskeletal radiologist in a subset (37%) of patients. Comorbidities were self-reported using a standardized questionnaire. We stratified all analyses by sex. We plotted the median (interquartile range) population-level KOOS scores of all subscales on the previously developed population-based KOOS percentile curves(1). Additionally, we investigated differences in score trajectories between patients with preoperative KL scores ≤2 to ≥3, and presence (versus absence) of comorbidities.

Results: The study population consisted of 853 knee OA patients, of whom 62% were women. The mean (SD) age was 59 (5) years, mean BMI was 30 (5) kg/m2). Overall, 74% of the population had moderate to severe radiographic OA, a minority had no (4%), doubtful (6%) or minimal but definte  (17%) radiographic OA. Comorbidities were present in 75% of patients. We observed poor KOOS scores across all subscales, which improved greatly postoperatively, with stabilization of scores after 12 months (table 1). Preoperatively, median KOOS scores of all subscales were at or below the 2.5th percentile. Scores increased to approximately the 25th percentile 12 months postoperatively. Greater improvements were observed in the subscale pain, and less improvements in the subscales sport and recreational function, and quality of life (figure 1). Patients with higher preoperative KL scores and without comorbidities showed greater improvements in KOOS scores (figure 2).

Conclusion: We applied KOOS data of knee OA patients undergoing TKA on the KOOS percentile curves, showcasing multiple ways these curves may be used in research and clinical care. In comparison to the general population, we observed poor preoperative KOOS scores, with large interpatient variation. Although major improvements were observed postoperatively, median population scores did not normalize to scores of the general population. The percentile curves may aid patient-clinician communication, improve management of treatment expectations and support shared-decision making.

Figure 1. Preoperative and 12 months postoperative KOOS pain, symptoms, activities of daily living (ADL) function, sport and recreation function, and quality of life (QOL) subscale scores of men and women with knee OA undergoing TKA plotted on the KOOS percentile curves.

Figure 2. Preoperative and 12 months postoperative KOOS pain subscale scores of men and women with knee OA undergoing TKA, stratified by Kellgren-Lawrence score (upper quadrant), and by presence of comorbidities (lower quadrant).

« Back to ACR Convergence 2020

ACR Meeting Abstracts - https://acrabstracts.org/abstract/application-of-the-knee-injury-and-osteoarthritis-outcome-score-percentile-curves-on-preoperative-and-up-to-2-years-postoperative-data-of-patients-undergoing-total-knee-arthroplasty/