Director Regeneron Pharmaceuticals Tarrytown, United States
Background: The relationship between osteoarthritis (OA) and carpal tunnel syndrome (CTS) is not well characterized despite the two conditions sharing many risk factors. Specifically, there is sparse data on the incidence and risk of CTS in the OA population. Such data are important to contextualize trial results and post-marketing reports when evaluating therapies for OA.
Objectives: To estimate the incidence of CTS in patients with OA, overall and by type of OA, and compare to the incidence of CTS in the non-OA population.
Methods: Patients ≥ 45 years diagnosed with OA (index date) were identified and followed from Jan 1, 2018 to June 30, 2021 in Optum Claims. A random sample of patients ≥ 45 years who were not diagnosed with OA were selected and assigned a random index date (Guthrie 2012). Patients with a CTS diagnosis or undergoing carpal tunnel release surgery within the past 12 months were excluded. Patients were followed from index date to the date of first CTS event, end of continuous enrollment, death, OA diagnosis (for non-OA cohort only), or end of study period, whichever came first. OA type included knee, hand/wrist, hip, shoulder, other, and unspecified. Demographic (age, gender, race) and clinical (Charlson co-morbidity index, diabetes, hypothyroidism, rheumatoid arthritis) characteristics at baseline were balanced using propensity score matching (1:1 ratio). The risk of CTS in the OA and non-OA groups were evaluated using incidence rates and adjusted hazard ratios that were estimated using Cox regression.
Results: A total of 947,255 OA patients were matched to 947,255 non-OA patients during the study period. The mean (SD) age was 66 (11.08) years; 53% were female and 73% Caucasian. There were 11,408 CTS cases in the OA population, and 2,348 cases in the non-OA population. Incidence rates, and corresponding 95% CI, were 5.68 (5.58, 5.78) and 1.27 (1.22, 1.32) per 1,000 person years for the OA and non-OA population, respectively (HR = 2.62 [2.50, 2.74]). The incidence of CTS was elevated for all types of OA but was highest among those with hand or wrist OA (15.34 [14.84, 15.83]) compared to those with knee (5.56 [5.42, 5.69]) or hip (6.39 [6.09, 6.69]) OA.
Conclusions: Patients with OA are at increased risk of CTS compared to those without OA, with the incidence varying by type of OA. Further analyses are required to better understand the potential impact of unmeasured confounding due to occupation and obesity, and to assess whether findings are replicated in different patient populations and regions.
