Association of squatting with increased prevalence of radiographic tibiofemoral knee osteoarthritis: the Beijing Osteoarthritis Study

Abstract

OBJECTIVE: To examine the association between squatting, a common daily posture in China, and the prevalence of radiographic osteoarthritis (OA) in different knee compartments among Chinese subjects from Beijing, and to estimate how much of the difference in prevalence of knee OA between Chinese subjects in Beijing and white subjects in Framingham, Massachusetts is accounted for by the impact of squatting. METHODS: We recruited a random sample of Beijing residents age > or =60 years. Subjects answered questions on joint symptoms, and knee radiographs were obtained. Subjects were also asked to recall the average amount of time spent squatting each day at age 25 years. Radiographic films (weight-bearing anteroposterior and skyline views) were read for Kellgren/Lawrence (K/L) grade and individual radiographic features. Medial disease was defined when radiographs showed a K/L grade of > or =2 at the tibiofemoral joint and a medial joint space narrowing score of > or =1, and lateral disease was assessed in a comparable manner in the lateral compartments. We examined the association of squatting with the prevalence of tibiofemoral OA as well as with the prevalence of patellofemoral knee OA, while adjusting for age and other potential confounding factors. We used the same approach to assess the relationship between squatting and tibiofemoral OA in the medial compartment and in the lateral compartment. Finally, we estimated the impact of squatting at age 25 on the difference in prevalence of knee OA between Chinese subjects in Beijing and white subjects in the Framingham OA Study. RESULTS: Squatting was very common among the Chinese subjects: approximately 40% of men and approximately 68% of women reported squatting > or =1 hour per day at age 25. The prevalence of tibiofemoral OA increased as the time spent squatting at age 25 increased in both the men and the women. Compared with subjects who squatted <30 minutes per day at age 25, the multivariable-adjusted prevalence odds ratios of tibiofemoral OA were 1.1 for time spent squatting of 30-59 minutes/day, 1.0 for 60-119 minutes/day, 1.7 for 120-179 minutes/day, and 2.0 for > or =120 minutes/day among the men (P for trend = 0.074), and the respective odds ratios among the women were 1.4, 1.3, 1.2, and 2.4 (P for trend = 0.077). A weaker association with patellofemoral OA was found. Prolonged squatting in daily life was more strongly associated with medial knee OA than with lateral disease in the men, but had a similar effect on both knee compartments in the women. After adjusting for the impact of squatting, the age-adjusted difference in prevalence of tibiofemoral OA was reduced from an excess of 14.4% to 9.5% in the Chinese women, but the difference in prevalence of tibiofemoral OA in the Chinese men increased after adjustment for age and squatting, from 2.9% lower to 7.0% lower as compared with their white counterparts. CONCLUSION: Prolonged squatting is a strong risk factor for tibiofemoral knee OA among elderly Chinese subjects in Beijing, and accounts for a substantial proportion of the difference in prevalence of tibiofemoral OA between Chinese subjects in Beijing and white subjects in Framingham.