Predicting incident radiographic knee osteoarthritis in middle-aged women within 4 years: the importance of knee-level prognostic factors
Predicting incident radiographic knee osteoarthritis in middle-aged women within 4 years: the importance of knee-level prognostic factors
Objective: To develop and internally validate risk models and a clinical risk score tool to predict incident radiographic knee osteoarthritis (RKOA) in middle-aged women. Methods: We analyzed 649 women in the Chingford 1,000 Women study. The outcome was incident RKOA, defined as Kellgren/Lawrence grade 0–1 at baseline and ≥2 at year 5. We estimated predictors’ effects on the outcome using logistic regression models. Two models were generated. The clinical model considered patient characteristics, medication, biomarkers, and knee symptoms. The radiographic model considered the same factors, plus radiographic factors (e.g., angle between the acetabular roof and the ilium's vertical cortex [hip α-angle]). The models were internally validated. Model performance was assessed using calibration and discrimination (area under the receiver characteristic curve [AUC]). Results: The clinical model contained age, quadriceps circumference, and a cartilage degradation marker (C-terminal telopeptide of type II collagen) as predictors (AUC = 0.692). The radiographic model contained older age, greater quadriceps circumference, knee pain, knee baseline Kellgren/Lawrence grade 1 (versus 0), greater hip α-angle, greater spinal bone mineral density, and contralateral RKOA at baseline as predictors (AUC = 0.797). Calibration tests showed good agreement between the observed and predicted incident RKOA. A clinical risk score tool was developed from the clinical model. Conclusion: Two models predicting incident RKOA within 4 years were developed, including radiographic variables that improved model performance. First-time predictor hip α-angle and contralateral RKOA suggest OA origins beyond the knee. The clinical tool has the potential to help physicians identify patients at risk of RKOA in routine practice, but the tool should be externally validated.
88-97
Garriga, Cesar
5403565c-65fd-448f-904f-e41df634c888
Sanchez-Santos, Maria T.
04817dfd-fc86-4801-88f4-e3d54319fe39
Judge, Andrew
c6a83964-1d7c-4aa8-b2bf-9c264d1e487d
Hart, Deborah
1d1f15cd-05f4-488d-abdc-a5466eb55195
Spector, Tim
1fd13fc5-08aa-44c9-b920-c41e925e79d6
Cooper, Cyrus
e05f5612-b493-4273-9b71-9e0ce32bdad6
Arden, Nigel
23af958d-835c-4d79-be54-4bbe4c68077f
1 January 2020
Garriga, Cesar
5403565c-65fd-448f-904f-e41df634c888
Sanchez-Santos, Maria T.
04817dfd-fc86-4801-88f4-e3d54319fe39
Judge, Andrew
c6a83964-1d7c-4aa8-b2bf-9c264d1e487d
Hart, Deborah
1d1f15cd-05f4-488d-abdc-a5466eb55195
Spector, Tim
1fd13fc5-08aa-44c9-b920-c41e925e79d6
Cooper, Cyrus
e05f5612-b493-4273-9b71-9e0ce32bdad6
Arden, Nigel
23af958d-835c-4d79-be54-4bbe4c68077f
Garriga, Cesar, Sanchez-Santos, Maria T., Judge, Andrew, Hart, Deborah, Spector, Tim, Cooper, Cyrus and Arden, Nigel
(2020)
Predicting incident radiographic knee osteoarthritis in middle-aged women within 4 years: the importance of knee-level prognostic factors.
Arthritis Care & Research, 72 (1), .
(doi:10.1002/acr.23932).
Abstract
Objective: To develop and internally validate risk models and a clinical risk score tool to predict incident radiographic knee osteoarthritis (RKOA) in middle-aged women. Methods: We analyzed 649 women in the Chingford 1,000 Women study. The outcome was incident RKOA, defined as Kellgren/Lawrence grade 0–1 at baseline and ≥2 at year 5. We estimated predictors’ effects on the outcome using logistic regression models. Two models were generated. The clinical model considered patient characteristics, medication, biomarkers, and knee symptoms. The radiographic model considered the same factors, plus radiographic factors (e.g., angle between the acetabular roof and the ilium's vertical cortex [hip α-angle]). The models were internally validated. Model performance was assessed using calibration and discrimination (area under the receiver characteristic curve [AUC]). Results: The clinical model contained age, quadriceps circumference, and a cartilage degradation marker (C-terminal telopeptide of type II collagen) as predictors (AUC = 0.692). The radiographic model contained older age, greater quadriceps circumference, knee pain, knee baseline Kellgren/Lawrence grade 1 (versus 0), greater hip α-angle, greater spinal bone mineral density, and contralateral RKOA at baseline as predictors (AUC = 0.797). Calibration tests showed good agreement between the observed and predicted incident RKOA. A clinical risk score tool was developed from the clinical model. Conclusion: Two models predicting incident RKOA within 4 years were developed, including radiographic variables that improved model performance. First-time predictor hip α-angle and contralateral RKOA suggest OA origins beyond the knee. The clinical tool has the potential to help physicians identify patients at risk of RKOA in routine practice, but the tool should be externally validated.
Text
Manuscript ACR reviewed clean 2
- Accepted Manuscript
Text
Figure 2 cal_dis_Mar 1 2019
Restricted to Repository staff only
Request a copy
Text
Supplementary Text S1 clean
Restricted to Repository staff only
Request a copy
Text
Supplementary Tables ACR cg
Restricted to Repository staff only
Request a copy
Show all 5 downloads.
More information
Accepted/In Press date: 21 May 2019
e-pub ahead of print date: 25 May 2019
Published date: 1 January 2020
Additional Information:
Funding Information:
The authors thank the patients and professionals who participated in data collection in the Chingford 1,000 Women study. We acknowledge the data management system developed by Alison Turner, as well as the selection of candidate predictors for RKOA in a Delphi process carried out by Dr. Kirsten M. Leyland, Prof. Daniel Prieto Alhambra, and Dr. Kassim Javaid. We thank R. N. Batra for the former development of the validation code and Dr. Anjali Shah and Dr. Thomas Perry for proofreading the manuscript. We also thank Dr. Jennifer A. de Beyer (Centre for Statistics in Medicine, University of Oxford) for English language editing and the addition of her informative suggestions, Katherine Edwards of the Botnar Research Centre (University of Oxford) for providing her expertise in how the hip ?-angle could affect the development of incident RKOA, registered pharmacist Marta Rodriguez-Alarcon for contributing to the categorization of medication types, and Senior Research Fellow/Statistician Dr. Millie Parsons for providing the data set and support for external validation using the Hertfordshire cohort (MRC Lifecourse Epidemiology Unit, University of Southampton, Southampton General Hospital, UK). We also acknowledge the research teams of the Osteoarthritis Initiative for providing the data set with the biomarker CTX-II for the external validation of our clinical model: Marc Hochberg (University of Maryland School of Medicine, Baltimore), Rebecca Jackson (The Ohio State University, Columbus), C. Kent Kwoh (University of Pittsburgh, Pennsylvania), Charles Eaton (Memorial Hospital of Rhode Island, Pawtucket), and Michael Nevitt (University of California, San Francisco, data coordinating center). A steering committee, comprised of representatives from these centers, the NIH, and the pharmaceutical partners, advised on the scientific aspects of the study. A representative from the US Food and Drug Administration advised the steering committee.
Publisher Copyright:
© 2019, American College of Rheumatology
Identifiers
Local EPrints ID: 432438
URI: http://eprints.soton.ac.uk/id/eprint/432438
ISSN: 0893-7524
PURE UUID: 50210fb1-ad52-48c7-a317-f38071e5c98a
Catalogue record
Date deposited: 12 Jul 2019 16:31
Last modified: 18 Mar 2024 05:09
Export record
Altmetrics
Contributors
Author:
Cesar Garriga
Author:
Maria T. Sanchez-Santos
Author:
Andrew Judge
Author:
Deborah Hart
Author:
Tim Spector
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
