We thank Dr Chaix and colleagues for their interest in our publication. 1 In their letter to the editor, they question whether the higher prevalence of important atherosclerotic risk factors in those with venous thromboembolism (VTE) who subsequently developed arterial thrombotic disease (ATD) compared with the VTE-only group could be partially responsible for the ATD. First, we want to remind the readers that the aim of the present study was to investigate whether subjects with VTE had higher risk of ATD compared with those without VTE. Thus, with regard to confounding, the interrogating question would be whether a higher prevalence of atherosclerotic risk factors in the VTE group overall compared with those without VTE could explain the higher risk of ATD. To assess the impact of potential confounders on the association between VTE and ATD, we applied a multivariable Cox regression model that included traditional atherosclerotic risk factors, and the results were presented in Table 3 in the article. Adjustments for body mass index, diabetes mellitus, hypertension, hypercholesterolemia, smoking, physical activity, and education level had a slight impact on the risk estimate (17% change) in that it changed from 1.42 (95% confidence interval, 1.15-1.75) in the age-and sex-adjusted model to 1.35 (95% confidence interval, 1.09-1.66) in the multivariable model.We completely agree that atherosclerotic risk factors such as diabetes mellitus, hypertension, and hypercholesterolemia are modifiable and that the individual atherosclerotic risk profile consequently may have changed during follow-up (median, 12.2 years) in some subjects in our study population. Unfortunately, repeat measurements on atherosclerotic risk factors were not available for all participants in our study, and as discussed in our article, residual confounding by modifiable risk factors cannot be completely ruled out. However, because most large cohort studies, including the Tromsø study, 2 found no association between atherosclerotic risk factors and VTE, 3-5 we do not think it is likely that the degree of confounder misclassification during follow-up would be different among those who did and did not develop VTE.
DisclosuresNone.
Caroline Lind, BSc