Maternal and Personal Smoking, Genetic Susceptibility, and Risk of Osteoarthritis: A Population‐Based Prospective Cohort Study

OBJECTIVE: To investigate the associations of maternal and personal smoking with the risk of knee, hip, hand, and overall osteoarthritis (OA) and the role of genetic predisposition to OA. METHODS: A total of 321,075 UK Biobank participants (mean age 56 years, 55% women) without OA at baseline were included; 29.9% reported maternal smoking, 24.3% reported former smoking, and 9.5% reported current smoking. Maternal smoking around childbirth (yes or no) and personal smoking status (never, former, or current) and dose (pack-years) were self-reported by the participants. Cox proportional hazards models were used to explore the associations of maternal and personal smoking status and personal smoking dose with the risk of OA. Genetic susceptibility of OA was assessed using polygenic risk scores (PRS). RESULTS: During follow-up, 18,470 (5.8%), 12,064 (3.8%), 6,605 (2.1%), and 52,853 (16.5%) participants developed knee, hip, hand, and overall OA, respectively. Both maternal and personal smoking were positively associated with the risk of knee, hip, hand, and overall OA (maternal smoking hazard ratio [HR] = 1.08-1.23; personal smoking HR = 1.18-1.40). There were significant dose-responsive associations between personal smoking dose and OA risk (P for trend <0.05). The association between maternal and personal smoking and OA risk was similar across different PRS strata. CONCLUSION: Both maternal and personal smoking are independently associated with an elevated risk of OA. These associations are universal across different genetic susceptibility of OA.