Background Hand osteoarthritis (HOA) is the most frequent form of osteoarthritis (60%). The main risk factors of hand osteoarthritis are: age above 40, women, obesity, heredity and work. A recent meta-analysis seems to indicate that obesity (through the systemic release of pro inflammatory mediators such adipokines) is a risk factor of HOA (OR=1.9). However, results of studies analyzing obesity as a risk factor of HOA are mixed and focused on radiological HOA. If obesity is a substantial risk factor of HOA, we hypothesize that patients with severe obesity may present a high prevalence of this disease.
Objectives The aim of the present was to study the prevalence of HOA which is defined as Heberden and Bouchard nodes in a prospective observational study in a population of patients with severe obesity (BMI>35 kg/m2) followed in specialized clinical nutrition department at a university hospital in Paris, France (Pitié Salpêtrière).
Methods A senior rheumatologist collected demographic information (personal and familial histories, current treatment, hand pain), questionnaires of hand disability (DREISER and Cochin) and performed a complete detailed physical examination. Heberden and Bouchard nodes were diagnosed by a physical examination of hand joints and knee OA was confirmed by previous medical history. We studied the association (Fisher’s exact test) between knee and hand osteoarthritis to explore if obesity is a risk factor of multiple osteoarthritis.
Results We studied 112 patients: 65% of women, average age 47.1 years and average BMI 45.9 kg/m2 (30.8-69.1). We found a high proportion of Heberden nodes (16.1%), noteworthy particularly in the second and third fingers and a prevalence of Bouchard nodes of 1.8%. A third of patients with clinical hand osteoarthritis had hand pain and did not report hand disability. Knee osteoarthritis was reported in 23.2% of subjects. We found a significant link between knee and hand osteoarthritis (p=0.006). Multivariate analysis (taking into account age, sex, BMI, diabetes mellitus, hypertension and cholesterol serum level) showed that age was a significant risk factor of HOA [OR=1.11 (IC 95%: 1.04-1.17)] and knee osteoarthritis [OR=1.08 (IC 95%: 1.02-1.15)]. Univariate analysis showed that patients with knee osteoarthritis more often had diabetes mellitus (57.6% vs 26.7%, p=0.003), hypertension (73.1% vs 32.6%, p=0.006) and dyslipidemia (42.3% vs 4.7%, p=5.10-5).
Conclusions In comparison with the known prevalence of clinical HOA in other series of patients aged over 55 years (4% to 14.9%), we found a high prevalence of Heberden nodes (16.1%) in this selected population of relatively young patients (mean age47 years) with severe obesity, suggesting a systemic participation of adipose tissue. Moreover disturbances that are part of the metabolic syndrome could participate in the development of knee osteoarthritis which is otherwise linked to HOA.
Disclosure of Interest None Declared