The Impact of Obesity on Knee Osteoarthritis Symptoms and Related Biomarker Profiles in a Bariatric Surgery Cohort

Purpose: Obesity is a modifiable risk factor of knee osteoarthritis (KOA). Surgical weight loss may be more effective than diet and exercise in delaying or avoiding joint replacement, as retrospective studies have in fact shown sustained improvement in KOA pain after bariatric surgery. We prospectively evaluated painful KOA in the obese population, tracked whether weight loss after bariatric surgery affects KOA-related pain and function, and studied serum and plasma levels of adipokines and (published) OA biomarkers in this subset of KOA patients. Methods: We screened consecutive patients prior to laparoscopic adjustable gastric banding (LAGB), sleeve gastrectomy, or gastric bypass at NYU and Bellevue. Patients age ≥18 with knee pain for ≥1 month and a visual analog scale pain score ≥30 mm were enrolled, excluding those with lupus, rheumatoid arthritis, psoriatic arthritis, or psoriasis. Baseline pre-op assessments included X-Rays for OA severity by Kellgren–Lawrence (KL) grade and completion of the Knee Injury and Osteoarthritis Outcome Score (KOOS). They repeated the questionnaire and were re-weighed to calculate percent excess weight loss (%EWL) at 1, 3, 6 and 12 month post-op intervals. Patients were consented for optional tissue collection (blood, urine and intra-operative adipose samples) for biomarker analysis at all visits, with analyses using existing data from controls and a non-obese OA cohort for comparison. Results: Of 537 patients planning to have bariatric surgery, 309 (58%) reported knee pain. 176 enrolled having met criteria and consented for the study: 91.5% female, mean BMI 43.6 kg/m2 ± 7 (31.6–60.6), and mean age of 42.4 ± 11 (18–73). For radiographic severity, KL0 = 43 (25%), KL1 = 34 (19%), KL2 = 38 (22%), KL3 = 34 (19%), KL4 = 27 (15%). The mean pre-bariatric KOOS pain score was 45.4 (0 = worst, 100 = best), with more pain associating with radiographic severity (p = 0.001 for both KL0 vs. KL1-2 and KL1-2 vs. KL3-4) and BMI - and similar trends across other KOOS subscores. 150 patients successfully underwent surgery (18 LAGB, 97 sleeve, 35 bypass), with the most followups 12 months later (111 patients). While the majority of pain improvement occurred within the first month (mean delta KOOS = 26.4 and then continued to slowly trend upwards at 3, 6 and 12 months with mean delta KOOS =24.8, 30.6, 29.3), the mean %EWL at those time intervals increased more gradually and consistently. Overall, %EWL did correlate well with delta KOOS pain when examining the 12 month outcomes (R = 0.246, p = 0.009), and virtually all patients still with BMI ≥ 40 at 12 months improved only minimally. Neither presence nor severity of KOA severity affected knee pain improvement from weight loss. We obtained baseline pre-bariatric blood from 30 patients (mean BMI = 44.1) and compared them with stored samples from a cohort of non-obese patients with KOA (mean BMI = 26.9), and a true control cohort without obesity or KOA (mean BMI = 26.6). Leptin was higher in serum from our bariatric KOA cohort than the non-obese KOA patients (mean 110.2 + 61.7 ug/ml vs. 26.8 + 16.6, p = 0.0001), and the control samples (12.8 + 7.8, p < 0.0005, Mann Whitney test, vs. non-obese KOA). Within our bariatric cohort of 30 patients, those with KL0/1 had lower leptin than KL2/3/4 subjects (mean 75.6 + 36.0 ug/ml vs. 121.7 + 72.3, p = 0.077). These findings all support a role for leptin in OA pathogenesis, while similar studies with adiponectin (mean 58.3 + 29.4 ug/ml vs. 70.1 + 35.2 for obese vs. non-obese) failed to show any similar trends. Assays for KOA progression marker IL-1Ra were markedly elevated in our cohort vs. the non-obese KOA and control groups (1123 pg/ml vs. 324.9 vs 272.0, p < 0.0001), and without variation by KL severity. We also found elevations in IL-6 and PGE2, but not 15-hydroxyeicosatetraenoic acid (15-HETE), in the obese KOA patients (data not shown). Conclusions: Bariatric surgery improves patients’ KOA pain proportional to percent excess weight loss, with a majority of the pain reduction occurring over the first month. Weight loss appears to lag behind KOOS pain improvement in our cohort, further supporting the notion that metabolic or cytokine-driven factors lead to knee pain beyond the mechanics of increased load and stress. Our preliminary data suggests that leptin and IL-1Ra contribute to pain and potentially to joint damage in obesity-associated KOA.