In a recent work published by the Journal, Du and Deng[1] explored the association of educational attainment (EA) with the risk of varicose vein (VV), venous thromboembolism (VTE), and phlebitis. The hypothesis is that features of obesity may be important mediators in this association. To do this, the authors used a Mendelian randomization design. The authors found that each 4.2 years of EA was associated with a 0.78 times lower risk of VV, 0.79 times lower risk of VTE, and a 0.57 times lower risk of phlebitis. After adjustment for EA, the authors found that obesity-related traits such as body mass index (BMI), basal metabolic rate, hip circumference, and waist circumference (WC) increased the risk of VTE (odds ratio: 1.42, 1.72, 1.53, and 1.69, respectively). In addition, EA was found to be inversely associated with the risk of developing VTE through obesity traits. In particular, BMI explained 18.84% of the effect of EA on VTE and basal metabolic rate of 10.04%, but these were not statistically significant. Conversely, hip circumference and WC significantly mediated 31.62 and 53.38% of the effect on VTE.

The assumption behind this study is on one side that both low EA[2] and obesity[3] [4] are associated with VTE risk, and that high EA is associated with a lower obesity degree. This study is perhaps reflective of the great interest into the impact of obesity on thrombosis-related complications clinically,[5] [6] as well as the impact on anticoagulant drug treatments.[7] Obesity is also not a binary (yes/no) diagnosis, with outcomes impacted by associated metabolic-related comorbidities.[8]

One interesting finding in the study by Du and Deng[1] is that WC but not BMI was the strongest mediator for VTE risk. This finding reinforces the concept that measures of obesity have not the same meaning in increasing thrombotic risk; however, sometimes they are used as equivalent. This is mainly due to the fact that most registries and studies do not collect data on WC but only on BMI, despite WC is very easy to measure.

Indeed, in a Swedish cohort, participants with normal BMI but increased WC had 53% higher risk of VTE.[9] A previous study also showed that per 1 standard deviation increase in WC there was an increased odds ratio of 1.803 of developing VTE.[10] These findings suggest that visceral adiposity drives VTE risk. Risk of VTE in obese patients is mediated by chronic inflammation, impaired fibrinolysis, and upregulated tissue factor pathway.[11]

As EA has a genetically determined component (20–40% according to twin and genome-wide association studies) that is constant across different populations,[12] the authors suggest that interventions on environmental factors influencing obesity-related traits may mitigate the association between EA and obesity-related traits and the subsequent risk of VV and VTE.

While there is evidence that weight gain after a first VTE is associated with increased risk of recurrent thrombotic events (6.6-fold increased risk for ≥7.5 kg weight gain),[13] the association of weight loss with risk of VTE is more controversial. During 9-year follow-up in the ARIC cohort, weight loss was associated with incident VTE (hazard ratio [HR]: 2.11), but it was not specified whether patients had malnutrition or were diagnosed with cancer.[14] The only evidence of an association between low body weight and reduced risk of VTE comes from the case–control EDITH study including 1,354 patients in whom underweight associated with a reduction in VTE compared with normal weight (odds ratio: 0.55).[15] In addition, in 30,171 patients with BMI ≥35 kg/m2 matched to 218,961 nonsurgical patients followed for a median 9.3 years, bariatric surgery was associated with a fivefold greater VTE risk at 30 days (HR: 5.01), but this risk decreased over time (41% lower VTE risk and a 55% lower pulmonary embolism risk at 5 years).[16] While it is arguable that this risk may be related to obesity reduction, weight loss magnitude during follow-up was not reported.

In conclusion, while there is evidence that obesity mediates the risk of VTE associated with EA, and that weight gain increases the risk of recurrent VTE, there is still uncertainty on the degree of weight loss necessary to reduce VTE risk as well as the optimal strategy to reach this target ([Fig. 1]). The impact of associated metabolic comorbidities and their changes over time (risk being dynamic rather than a static measure) also require consideration. Given the racial and sex differences in thrombotic (and bleeding) outcomes,[17] [18] the picture is very likely to be far more complex.