Pre-infection anticoagulant exposure and SARS-CoV-2 infection outcomes – Differential mortality by age
Itshak Amsalem a), Asher Shafrir b c) d), Yosef Kalish e), Ora Paltiel e) f)
- a) Jesselson Integrated Heart Center, Shaare Zedek Medical Center and Faculty of Medicine, Hebrew University of Jerusalem, Israel
- b) Faculty of Medicine, Hebrew University of Jerusalem, Jerusalem, Israel
- c) Meuhedet Health Services, Tel Aviv, Israel
- d) Institute of Gastroenterology and Liver Diseases, Hadassah-Hebrew University Medical Center, Jerusalem, Israel
- e) Department of Hematology, Hadassah University Medical Center, Jerusalem, Israel
- f) Braun School of Public Health and Community Medicine, Hadassah Medical Organization, Faculty of Medicine, Hebrew University of Jerusalem, Jerusalem, Israel
Abstract
Background
The risk of thrombosis increases after SARS-CoV-2 infection. This study aimed to assess associations between pre-infection anticoagulant exposure and SARS-CoV-2 infection-related outcomes in a population-based cohort.
Methods
Members of the “Meuhedet” health maintenance organization aged >45 years who tested positive for SARS-CoV-2 infection (03/2020–04/2022) were followed. Pre-infection anticoagulant exposure (PAE) was defined as any anticoagulant therapy prescribed ≥1 month prior to SARS-CoV-2 testing. Univariate analyses, multivariable models adjusting for confounders, propensity-score matching, and an age-stratified analysis were performed to assess associations between PAE and hospitalization, intensive care unit (ICU) admission, 30-day and one-year mortality.
Results
Of the 127,801 patients included, 2951(2.3 %) had PAE. Comorbidities including ischemic heart disease, diabetes mellitus, hypertension, heart failure, and atrial fibrillation were more common among anticoagulant-exposed than unexposed individuals (p < 0.001).
Patients with PAE experienced higher hospitalization (22.7 % vs 5.6 %), ICU admissions (1.9 % vs 0.5 %), 30-day and 1-year mortality rates (4.8 % vs. 0.6, and 8.8 % vs. 1.1 %, respectively), than unexposed individuals, but similar lengths-of-stay. In the multivariable analysis, PAE was independently associated only with hospitalizations (adjusted odds ratio (aOR) = 1.29 [95 % confidence interval (CI): 1.13–1.47]), whereas in the propensity-matched analysis, none of the outcomes differed significantly between the groups.
However, in the stratum aged >75 years, 30-day and one-year mortality were significantly reduced in those with PAE (aOR = 0.68 [CI:0.48–0.97], and aOR = 0.73 [CI:0.55–0.97], respectively).
Conclusion
SARS-CoV-2-infected individuals with prior exposure to anticoagulants have more comorbidities and experienced a higher incidence of hospitalization but not mortality compared to unexposed patients. Paradoxically, mortality risks decreased in the oldest stratum of anticoagulant-exposed individuals. Further research is required to assess mechanisms for this apparent protective effect.