Dataset Information

Diagnosis of early stage knee osteoarthritis based on early clinical course: data from the CHECK cohort.

ABSTRACT:

Background

Early diagnosis of knee osteoarthritis (OA) is important in managing this disease, but such an early diagnostic tool is still lacking in clinical practice. The purpose of this study was to develop diagnostic models for early stage knee OA based on the first 2-year clinical course after the patient's initial presentation in primary care and to identify whether these course factors had additive discriminative value over baseline factors.

Methods

We extracted eligible patients' clinical and radiographic data from the CHECK cohort and formed the first 2-year course factors according to the factors' changes over the 2 years. Clinical expert consensus-based diagnosis, which was made via evaluating patients' 5- to 10-year follow-up data, was used as the outcome factor. Four models were developed: model 1, included clinical course factors only; model 2, included clinical and radiographic course factors; model 3, clinical baseline factors + clinical course factors; and model 4, clinical and radiographic baseline factors + clinical and radiographic course factors. All the models were built by a generalized estimating equation with a backward selection method. Area under the receiver operating characteristic curve (AUC) and its 95% confidence interval (CI) were calculated for assessing model discrimination. Delong's method compared AUCs.

Results

Seven hundred sixty-one patients with 1185 symptomatic knees were included in this study. Thirty-seven percent knees were diagnosed as OA at follow-up. Model 1 contained 6 clinical course factors; model 2: 6 clinical and 3 radiographic course factors; model 3: 6 baseline clinical factors combined with 5 clinical course factors; and model 4: 4 clinical and 1 radiographic baseline factors combined with 5 clinical and 3 radiographic course factors. Model discriminations are as follows: model 1, AUC 0.70 (95% CI 0.67-0.74); model 2, 0.74 (95% CI 0.71-0.77); model 3, 0.77 (95% CI 0.74-0.80); and model 4, 0.80 (95% CI 0.77-0.82). AUCs of model 3 and model 4 were slightly but significantly higher than corresponding baseline-factor models (model 3 0.77 vs 0.75, p = 0.031; model 4 0.80 vs 0.76, p = 0.003).

Conclusions

Four diagnostic models were developed with "fair" to "good" discriminations. First 2-year course factors had additive discriminative value over baseline factors.

SUBMITTER: Wang Q

PROVIDER: S-EPMC8375192 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectiveWe aimed to describe the natural history leading to end-stage knee osteoarthritis (esKOA), focusing on knee symptoms, radiographic severity, and the presence of limited mobility or instability.MethodsWe performed knee-based analyses of Osteoarthritis Initiative data from 7691 knees (4165 participants). We used a validated definition of esKOA that relied on meeting one of two criteria: (1) severe radiographic knee osteoarthritis (Kellgren-Lawrence [KL] grade=4) with moderate-to-intense pain (Likert WOMAC pain+function>11/88) or (2) KL grade<4 with intense or severe pain (WOMAC pain+function>22) and limited mobility (flexion contracture≥5°) or instability (based on a varus and valgus stress test). We also introduced an alternate definition of esKOA that relied on meeting one of two criteria that omitted physical exam findings:(1) severe radiographic knee osteoarthritis (KL grade=4) with at least moderate symptoms or (2) KL grade=2 or 3 with intense or severe symptoms and persistent knee pain (frequent knee pain during three or more months in the past year). We used descriptive statistics to explore the frequency of components of esKOA at the index visit when they had incident esKOA, at the annual visit before developing esKOA, and the interval change between those visits.ResultsOur analytic sample was mostly female (58%), without radiographic knee osteoarthritis (KL grade=0 or 1; 60%), without stability or mobility concerns (91%), and without persistent knee pain (77%). At the visit before incident esKOA, most knees already had moderate-to-severe radiographic osteoarthritis using the original (62%) or alternate (50%) definition (versus <15% for either definition of no esKOA). Over 80% of knees that reached the criteria for esKOA achieved this based on increased knee symptom severity - typically without worsening radiographic severity (80%).ConclusionRadiographic severity predisposed a knee to esKOA. However, worsening knee symptoms led to the development of incident esKOA. If investigators want to increase the chance of identifying incident esKOA as an outcome, they should enrich their study samples with people with moderate-to-severe radiographic osteoarthritis. Our findings also highlight the potential reversibility of esKOA (a knee that is classified with esKOA but later is not classified with esKOA). Reversibility is not a flaw of an outcome defining esKOA but rather a desirable clinical outcome to demonstrate a therapeutic intervention can help people with esKOA improve their knee symptoms and delay a knee replacement.

Project description:Background The methods for assessing knee osteoarthritis (OA) do not provide enough comprehensive information to make robust and accurate outcome predictions. Purpose To develop a deep learning (DL) prediction model for risk of OA progression by using knee radiographs in patients who underwent total knee replacement (TKR) and matched control patients who did not undergo TKR. Materials and Methods In this retrospective analysis that used data from the OA Initiative, a DL model on knee radiographs was developed to predict both the likelihood of a patient undergoing TKR within 9 years and Kellgren-Lawrence (KL) grade. Study participants included a case-control matched subcohort between 45 and 79 years. Patients were matched to control patients according to age, sex, ethnicity, and body mass index. The proposed model used a transfer learning approach based on the ResNet34 architecture with sevenfold nested cross-validation. Receiver operating characteristic curve analysis and conditional logistic regression assessed model performance for predicting probability and risk of TKR compared with clinical observations and two binary outcome prediction models on the basis of radiographic readings: KL grade and OA Research Society International (OARSI) grade. Results Evaluated were 728 participants including 324 patients (mean age, 64 years ± 8 [standard deviation]; 222 women) and 324 control patients (mean age, 64 years ± 8; 222 women). The prediction model based on DL achieved an area under the receiver operating characteristic curve (AUC) of 0.87 (95% confidence interval [CI]: 0.85, 0.90), outperforming a baseline prediction model by using KL grade with an AUC of 0.74 (95% CI: 0.71, 0.77; P < .001). The risk for TKR increased with probability that a person will undergo TKR from the DL model (odds ratio [OR], 7.7; 95% CI: 2.3, 25; P < .001), KL grade (OR, 1.92; 95% CI: 1.17, 3.13; P = .009), and OARSI grade (OR, 1.20; 95% CI: 0.41, 3.50; P = .73). Conclusion The proposed deep learning model better predicted risk of total knee replacement in osteoarthritis than did binary outcome models by using standard grading systems. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Richardson in this issue.

Project description:BackgroundWe aimed to evaluate the prevalence of hip and knee osteoarthritis (HOA and KOA) according to American College of Rheumatology (ACR) criteria among participants with suspected early symptomatic osteoarthritis (OA) in the CHECK cohort. We also assessed whether participants not fulfilling ACR criteria at baseline develop ACR-defined OA at 2-year and/or 5-year follow up, and which baseline factors are associated with this development.MethodsThe CHECK cohort included 1002 subjects with first presentation of knee and/or hip complaints. The primary outcome was onset of HOA and/or KOA according to the ACR criteria, including the clinical classification criteria and the combined clinical and radiographic classification criteria at 2-year and/or 5-year follow up.ResultsOf the participants with hip complaints, 63% (n?=?370) were classified as having HOA at baseline according to the ACR criteria. Of those not classified with HOA at baseline, 40% developed HOA according to the clinical or combined clinical/radiographic ACR criteria after 2 and/or 5?years. Up to 92% of participants (n?=?829) with knee complaints were classified as having KOA at baseline; of those not classified with KOA at baseline, 55% developed KOA according to the clinical ACR criteria or the clinical/radiographic ACR criteria after 2 and/or 5?years. The following factors were associated with development of HOA: morning stiffness (OR 2.39; 95% CI 1.14-4.98), painful internal rotation (OR 2.53; 95% CI 1.23-5.19), hip flexion <?115° (OR 2.33; 95% CI 1.17-4.64) and erythrocyte sedimentation rate (ESR)?<?20?mm/h (OR 2.94; 95% CI 1.13-7.61). No variables were associated with development of KOA at 2-year and/or 5-year follow up.ConclusionsA large proportion of persons with hip complaints not fulfilling the ACR criteria at baseline develop HOA after 2 and/or 5 years of follow up. Almost all persons with knee complaints already fulfill the clinical and/or radiographic ACR criteria for OA, and half of the persons not fulfilling criteria at baseline will do so after 5?years of follow up. Several individual ACR criteria for HOA at baseline were associated with the development of HOA at follow up. This association was not proven for KOA, probably because of the small number of subjects developing KOA in this study.