OP0226 TRAJECTORIES OF BODY MASS INDEX FROM EARLY ADULTHOOD TO LATE MIDLIFE AND INCIDENCE OF TOTAL KNEE ARTHROPLASTY FOR OSTEOARTHRITIS: FINDINGS FROM A PROSPECTIVE COHORT STUDY

Abstract

BackgroundThere is limited evidence regarding the association between trajectories of body mass index (BMI) across adulthood and knee osteoarthritis.ObjectivesWe examined the association between body mass index (BMI) trajectories across early adulthood to midlife and risk of total knee arthroplasty (TKA) for osteoarthritis.MethodsThis study examined 24,368 participants (40-70 years at recruitment) in the Melbourne Collaborative Cohort Study who had weight collected at 1990–1994, 1995–1998, and 2003–2007 and recalled weight at age 18–21 years. BMI trajectories were derived using weight data at the four timepoints. Incidence of TKA after 2003–2007 until December 2018 was determined by linking cohort records to the National Joint Replacement Registry.ResultsUsing group-based trajectory modelling, six distinct trajectories (TR) of BMI from early adulthood to late midlife were identified: lower normal to normal BMI (TR1: BMI at age 18-21 years to BMI at approximately 62 years (kg/m2), 20.0±1.9 to 22.1±1.7; 19.7%); normal BMI to borderline overweight (TR2: 21.5±2.3 to 25.8±1.7; 36.7%), normal BMI to overweight (TR3: 22.0±2.2 to 29.5±1.9; 26.8%), overweight to borderline obese (TR4: 28.5±2.7 to 30.5±2.3; 3.5%), normal BMI to class 1 obesity (TR5: 22.8±2.5 to 34.3±2.3; 10.1%), and overweight to class 2 obesity (TR6: 25.6±3.9 to 39.2±2.9; 3.2%). Over 12.4 years, 1,328 (5.4%) participants had TKA. The hazard ratios for TKA increased in all TR compared with TR1: TR2 2.03 (95% CI 1.64-2.52), TR3 4.00 (3.19-4.91), TR4 5.17 (3.77-7.10), TR5 7.00 (5.54-8.80), and TR6 8.59 (6.44-11.46). It is estimated that 28.4% TKA would be reduced if individuals followed the trajectory that was one lower, a national health system savings of $AUD 373 million. Most of this reduction would occur in TR2 (population attributable fraction 37.9% (26.7%-47.3%) and TR3 PAF 26.8% (20.0%, 31.2%) (Table 1).Table 1.Reduction in TKA if individuals followed the trajectory that was one lowerPopulation counterfactualsPopulation at risk, n (%)TKA under the original scenario, n (%)TKA under the new scenario*, n (%)Difference in risk, n (%)PAF** (95% CI), %TR14811 (19.7)124 (2.6%)No change 124 (2.6%)0-If TR2 followed TR1 trajectory and rate of TKA*8943 (36.7)378 (4.2%)2.6% of 8943 = 233145 (10.9)37.9 (26.7, 47.3)If TR3 followed TR2 trajectory and rate of TKA*6526 (26.8)416 (6.4%)4.2% of 6526 = 274142 (10.7)26.8 (20.0, 31.2)If TR4 followed TR3 trajectory and rate of TKA*845 (3.5)64 (7.6%)6.4% of 845 = 5410 (0.8)3.1 (0, 6.0)If TR5 followed TR4 trajectory and rate of TKA*2466 (10.1)253 (10.3%)7.6% of 2466 =18766 (5.0)20.2 (0, 36.3)If TR6 followed TR5 trajectory and rate of TKA*777 (3.2)93 (12.0%)10.3% of 777 = 8013 (1.0)0.4 (0.0, 10.0)Total population243681328952376 (28.4)-*if the trajectory is changed, ** PAFs and related 95%CIs were calculated by the Stata punafcc package using the formula ∑pKRi[(HRi − 1)/HRi], where pKRi is the proportion of total knee replacements observed in the ith obesity trajectory and HRi is the hazard ratio (HR) associated with that category. PAFs were calculated using pKRi and HRi estimated from the entire sample. All HRi values were generated from Cox proportional hazards regression models adjusted for covariates (age at baseline, sex, country of birth, physical activity, smoking history, and comorbidity) and postestimation analyses.ConclusionOur study suggests that prevention of weight gain from young adulthood to midlife in order to reduce overweight and obesity could have a major impact on reducing the burden of severe knee osteoarthritis and associated healthcare costs.Figure 1.A. Proportion of Total Knee Arthroplasties in each trajectory category B. Speculated patters and associated percentages represents the proportion of Total Knee Arthroplasties that could be avoided in each trajectory category if the participants followed the lower trajectory category i.e. TR2 followed TR1Disclosure of InterestsNone declared