2015 ISAKOS Biennial Congress ePoster #203

Predicting The Progression Of Surgical Knee Arthritis – A 1000 Patient, Twenty-Year Longitudinal Radiographic Study

Nicholas Bottomley, MBBS, FRCS, Oxford, Oxon UNITED KINGDOM
Luke Jones, FRCS, Oxford, Oxfordshire UNITED KINGDOM
Kirsten Leyland, PhD, Oxford UNITED KINGDOM
William Jackson, FRCS(Orth), MD, Oxford UNITED KINGDOM
Nigel Arden, FRCP, MD, Oxford UNITED KINGDOM
David J. Beard, MA, MSc, DPhil, FRCS(Hon), Prof., Oxford, Oxfordshire UNITED KINGDOM
Andrew J. Price, DPhil, FRCS(Orth), Oxford, Oxfordshire UNITED KINGDOM

Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Science, University of Oxford, Oxford, Oxfordshire, UNITED KINGDOM

FDA Status Not Applicable

Summary: This study investigated 1000 subjects radiologically over 20 years to determine the timescale of progression of arthritis from an early radiological stage to a radiological end-stage, which may require joint replacement. More than 50% of subjects developed arthritis during the study and of those with early or moderate arthritis, 40% progressed to end-stage disease by 20-years.




Despite much resource and effort into the study of individual factors important to the progression of knee arthritis, as clinicians we are still not able to tell our patients the timescale over which early knee arthritis might progress to an advanced stage requiring knee replacement. This study investigates 1000 women over a twenty-year period to define the timescale of progression of knee arthritis to a stage requiring knee replacement.


One thousand women, part of the Chingford Study, underwent knee radiographs every five years for twenty years. The radiographs were assessed using the Kellgren and Lawrence (KL) score, with a score of 0 or 1 defined as no arthritis, a score of 2 or 3 defined as early-moderate surgical arthritis and a score of 4 as end-stage surgical arthritis. By convention, this stage would be radiologically suitable for arthroplasty. The time of progression between scores for each subject was calculated and assessed using life table analysis to define the percentage progression of each stage of arthritis at each 5-year timepoint.

Two sub-groups were assessed. The first was the incident group, those who developed arthritis during the study. The second was a sub-group of all those with early or moderate surgical arthritis (KL2 or 3), whose timescale of progression to end-stage arthritis was assessed. The set terminal event for the life table analysis was development of arthritis (KL=2) in the incident group and development of end-stage arthritis (KL4) in the moderate arthritis group.


The mean age at the beginning of the study was 54.2 years (SD 5.9 years). Of the 1000 subjects, 53.7% developed arthritis during the study (the incident group). Nineteen percent of this group developed arthritis by 5 years, 45% by ten years, 64% by 15 years and the entire incident group by 20 years. End-stage arthritis accounted for 6.5% of disease in the incident group.

Five-hundred and forty-two knees were categorised into the early-moderate surgical arthritis sub-group during the study. The life table analysis showed 3% progression to end-stage arthritis at 5 years, 11% at ten years, 17% at 15 years and 37% at 20 years.


This 20-year longitudinal study of 1000 subjects provides for the first time data for a surgeon to counsel a patient with early arthritis about the time over which their arthritis may progress. A patient with no arthritis at age 54 years has a 54% risk of developing it by age 74, 6.5% of this will be end stage disease.

Those subjects with early-moderate arthritis have an 11% chance of developing end-stage arthritis within 10 years and an almost 40% chance by 20 years.

This provides valuable information both to the clinician when counseling patients with early changes of arthritis and for the planning of resources in the management of the burden of arthritis within a healthcare system.