Radiographic Knee OA Not Correlated With Diabetes, Altered Glucose Metabolism
Investigators assessed associations of diabetes and biomarkers of abnormal glucose metabolism with incident radiographic knee osteoarthritis, controlling for obesity.
After adjustment for body mass index (BMI), no association was found in older individuals between incident radiographic knee osteoarthritis (OA) and either the presence of diabetes mellitus (DM) or elevated abnormal glucose metabolism biomarkers, according to a report published in Arthritis Care & Research.
Both OA and DM are prevalent in the older adult population, and a connection between the two conditions has been suggested; however, previous studies have offered conflicting evidence, with most not accounting for BMI. Investigators sought to clarify whether such a connection exists independent of BMI and hypothesized that dysfunctional glucose metabolism and DM would correlate with greater odds for radiographic knee OA incidence.
A secondary analysis of participants from the Multicenter Osteoarthritis Study examined 987 (mean age, 60.6; mean BMI, 29.1 kg/m2) randomly stratified patients (1972 knees) according to BMI and evaluated them for evidence of tibiofemoral joint radiographic knee OA. All participants were free of radiographic knee OA at baseline, with a bilateral Kellgren-Lawrence grade <2.
Insulin resistance — using the homeostasis model of assessment (HOMA-IR) — and fasting glucose were measured at baseline whereas a DM diagnosis was established via self-report, medication use, or fasting glucose ≥126 mg/dl. There were 94 (9.5%) people (187 knees) with DM at baseline, in whom the mean BMI was significantly higher compared with people without DM (P <.0001).
Follow-up was conducted at 30, 60, and 84 months and incident radiographic knee OA was documented per knee, using Kellgren-Lawrence criteria ≥2 or a history of total knee replacement. Logistic regression analysis at the knee level was used to calculate adjusted odds ratios (aOR) for the potential associations.
After multivariable adjustment that considered BMI along with age, sex, and race, no association was detected between incident radiographic knee OA and baseline DM status (aOR 0.79; 95% CI, 0.53-1.18; P =.246), HOMA-IR (aOR 0.89; 95% CI, 0.79-1.01; P =.077), or fasting glucose (aOR 0.98; 95% CI, 0.87-1.11; P =.751), either overall or in men; however, in women, an inverse correlation was found between HOMA-IR and incident radiographic knee OA (aOR 0.80; 95% CI, 0.69-0.94; P =.005), indicating a possible protective relationship unexplained by the use of oral hypoglycemics.
At 84 months, no significant difference was found in the cumulative incidence of radiographic knee OA between people with (24.0%) or without (23.2%) baseline DM. Removing 54 patients from analysis who developed DM during the course of the study did not alter the results substantially.
Study strengths included a prospective design that eliminated the possibility of reverse causality, the use of a well-characterized cohort, 7-year follow-up, and confirmation of DM status via multiple methods.
Study limitations included possible unreliability of fasting information that classified patients as incorrectly having DM, lack of DM type data, using only radiographic evidence of radiographic knee OA, examination of only tibiofemoral OA, sample size that may not be large enough to detect associations, and potential lack of generalizability.
"The findings of the present analysis, along with the conflicting reports in the literature, illustrate the complexity of determining whether an independent relationship between DM and OA exists after accounting for the potentially confounding role of obesity," opined the investigators. They recommended that future studies explore the apparent protective effect of elevated insulin resistance in women.
Rogers-Soeder TS, Lane NE, Walimbe M, et al. Association of diabetes mellitus and biomarkers of abnormal glucose metabolism with incident radiographic knee osteoarthritis [published online November 12, 2018]. Arthritis Care Res (Hoboken). doi:10.1002/acr.23809