Knee Pain vs Structural Disease as Driver of Excess OA Mortality

Middle-aged women with knee pain, regardless of history of radiographic osteoarthritis (ROA), have an increased risk of all-cause and cardiovascular disease (CVD)-specific mortality, according to research published in Annals of the Rheumatic Diseases.

Stefan Kluzek, MRCP, MSc, from the Oxford NIHR Musculoskeletal Biomedical Research Unit and the Arthritis Research UK Centre for Sport, Exercise, and Osteoarthritis, and colleagues conducted a prospective, longitudinal cohort study to determine if either joint pain or radiographic osteoarthritis (ROA) of the knee and hand are associated with all-cause and disease-specific mortality in middle-aged women.

Researchers identified 4 cohort subgroups from the community-based Chingford Cohort Study— Pain-/ROA-; Pain+/ROA-; Pain-/ROA+; and Pain+/ROA+. Participants were female, between the ages of 45-64 years. Researchers included a total of 1629 women (knee osteoarthritis [OA] n=821, hand OA n=808) for final data analyses.

Primary outcomes of interest were all-cause and cause-specific mortality due to CVD and cancer. Detailed mortality information was provided by the Health and Social Care Information Center, based on information collected by the Office of National Statistics. Causes of death were collected from death certificates.

Dr Kluzek and colleagues performed a physical examination of all patients at baseline, and assessed anteroposterior (AP) radiographs of participants’ hands and weight-bearing AP radiographs of participants’ knees.

Participants were classified as having knee OA (KOA) if the researchers found a Kellgren-Lawrence (K/L) OA grade of 2 or more in at least one knee at baseline. Radiographic hand OA was defined as positive if at least one joint had a K/L grade of ≥2. 

Through a self-administered questionnaire, participants assessed both knee and hand pain at year 3; reports were made on the number of days that knee or hand pain had occurred in the previous month.

During the study follow-up period of 1993-2014 (median follow up: 21.7 years), 166 and 163 deaths were reported in the KOA and hand OA groups, respectively. Women with ROA (in both Pain+ and Pain- subgroups) were older and had higher systolic blood pressure levels compared to women without OA (Pain-/ROA-). Women who had painful KOA (ROA+ or ROA-) were more likely to use non-steroidal anti-inflammatory drugs, and women with knee or hand pain (Pain+/ROA-) were more likely to be using hormone replacement therapy medication.

In the age-adjusted model, the researchers found that women with knee pain and no ROA had a 49% increased risk of dying from all-cause mortality when compared with women who had neither knee pain nor ROA (hazard ratio [HR]: 1.49 95% confidence interval [CI], 1.04-2.14; P =.029).

Mortality risk was higher among women with painful knee ROA, with a 97% increased mortality risk in the age-adjusted model (HR: 1.97 95% CI, 1.23-3.17; P =.005). HRs increased slightly in the painful knee subgroups (ROA+, ROA-) after adjustment for Framingham Risk Score Factors (HR: 1.55 95% CI, 1.07-2.22 and HR: 2.06, 95% CI, 1.27-3.33, respectively).

Summary and Clinical Applicability

“In this prospective, community-based study of middle-aged women, we found a significant association between knee pain, with or without ROA, and an increased risk of all-cause and CVD-specific mortality,” noted Dr Kluzek and colleagues. “We found no association between knee ROA only and decreased survival, although subjects with symptomatic ROA had the highest risk of both all-cause and CVD-specific mortality.”

The researchers noted that the causal association is most likely due to sedentary behavior and poor cardiovascular fitness resulting from regular knee pain.

Limitations and Disclosures

• Generalizability of these results is limited to middle-aged, predominantly Caucasian women
• Hand and knee radiographs were taken 3 years before initial hand and knee pain assessment, so outcomes were measured from year 3 onwards. 
• Multivariable group analysis is likely to underestimate the absolute risk of the exposure groups