Sebastian Unizony1, Na Lu2, Gunnar Tomasson3, Yuqing Zhang4, Peter A Merkel5, John H Stone1, J Antonio Aviña-Zubieta6, Hyon K Choi2. 1. Massachusetts General Hospital, Harvard Medical School, Boston. 2. Massachusetts General Hospital, Harvard Medical School, and Boston University School of Medicine, Boston. 3. Boston University School of Medicine, Boston, Massachusetts, and University of Iceland, Reykjavik, Iceland. 4. Boston University School of Medicine, Boston, Massachusetts. 5. University of Pennsylvania, Philadelphia. 6. Arthritis Research Canada, University of British Columbia, Richmond, British Columbia, Canada.
Abstract
OBJECTIVE: Giant cell arteritis (GCA) and the use of glucocorticoids have both been associated with increased risk of venous thromboembolism (VTE). However, the possibility of confounding by indication has not been investigated. We undertook this study to examine the temporal risk of VTE in GCA patients before and after GCA diagnosis, accounting for confounders including glucocorticoid treatment. METHODS: We conducted a matched cohort study using an electronic medical record database representative of the UK population (1990-2013). We calculated age-, sex-, and entry time-matched and multivariate relative risks (RRs) of VTE, comparing 6,441 patients with new-onset GCA (defined by corresponding diagnosis codes and prescribed glucocorticoid treatment) to 63,985 controls before and after GCA diagnosis. Analysis before GCA diagnosis was stratified by oral glucocorticoid use to account for confounding. RESULTS: There were 27 incident VTE events during the 12 months preceding GCA diagnosis and 195 afterward. Compared to controls, during the 12, 9, 6, and 3 months preceding GCA diagnosis, the age-, sex-, and entry time-matched RRs for VTE among patients with imminent GCA not treated with glucocorticoids were 1.8, 2.2, 2.4, and 3.6, respectively. In the first 3, 6, 12, 24, 48, and 96 months after GCA diagnosis, the corresponding RRs were 9.9, 7.7, 5.9, 4.4, 3.3, and 2.4. Multivariate analyses including several common VTE risk factors showed similar trends. CONCLUSION: The risk of VTE increases shortly before GCA diagnosis, peaks at the time of diagnosis, and then progressively declines thereafter. This risk is apparent in patients with imminent GCA unexposed to oral glucocorticoids, suggesting a role for inflammation-associated thrombosis that is independent of glucocorticoid use.
OBJECTIVE:Giant cell arteritis (GCA) and the use of glucocorticoids have both been associated with increased risk of venous thromboembolism (VTE). However, the possibility of confounding by indication has not been investigated. We undertook this study to examine the temporal risk of VTE in GCA patients before and after GCA diagnosis, accounting for confounders including glucocorticoid treatment. METHODS: We conducted a matched cohort study using an electronic medical record database representative of the UK population (1990-2013). We calculated age-, sex-, and entry time-matched and multivariate relative risks (RRs) of VTE, comparing 6,441 patients with new-onset GCA (defined by corresponding diagnosis codes and prescribed glucocorticoid treatment) to 63,985 controls before and after GCA diagnosis. Analysis before GCA diagnosis was stratified by oral glucocorticoid use to account for confounding. RESULTS: There were 27 incident VTE events during the 12 months preceding GCA diagnosis and 195 afterward. Compared to controls, during the 12, 9, 6, and 3 months preceding GCA diagnosis, the age-, sex-, and entry time-matched RRs for VTE among patients with imminent GCA not treated with glucocorticoids were 1.8, 2.2, 2.4, and 3.6, respectively. In the first 3, 6, 12, 24, 48, and 96 months after GCA diagnosis, the corresponding RRs were 9.9, 7.7, 5.9, 4.4, 3.3, and 2.4. Multivariate analyses including several common VTE risk factors showed similar trends. CONCLUSION: The risk of VTE increases shortly before GCA diagnosis, peaks at the time of diagnosis, and then progressively declines thereafter. This risk is apparent in patients with imminent GCA unexposed to oral glucocorticoids, suggesting a role for inflammation-associated thrombosis that is independent of glucocorticoid use.