Blood Clots Kill Millions Worldwide, Yet Most Countries Cannot Even Count Them

Venous thromboembolism, the umbrella term for deep-vein thrombosis and pulmonary embolism, has long been described as the third most common cause of vascular death worldwide, trailing only heart attack and stroke. Yet until now, no contemporary global summary of its true toll has existed. A landmark collaborative review published in Nature Reviews Cardiology, led by Behnood Bikdeli of Brigham and Women’s Hospital and Harvard Medical School together with an international team spanning five continents, has assembled the most comprehensive picture to date of how often these clots occur, how often they kill, and how often they come back. The findings are sobering: in 2019 alone, an estimated 1.6 million new cases of pulmonary embolism and 2.5 million new cases of deep-vein thrombosis occurred worldwide, and the data underpinning those numbers come overwhelmingly from a handful of wealthy nations.

The technical challenge the authors faced is one that plagues much of global cardiovascular epidemiology. Reliable incidence data were available for only a limited number of countries, concentrated in Northern America, Western Europe and Oceania. Vast populations, including those of Sub-Saharan Africa and South Asia, are effectively invisible in the literature, not because clots do not form there, but because the diagnostic infrastructure, imaging capacity and registry systems needed to count them are absent or incomplete. Even where data exist, they are riddled with heterogeneity: different studies use different case definitions, different coding algorithms, different imaging thresholds and different follow-up windows, making direct comparison an exercise in careful caveats.

The headline incidence figures reveal striking regional variation. During 2012 to 2016, the global annual incidence of deep-vein thrombosis ranged from 25.1 to 131.5 cases per 100,000 population, a fivefold spread that reflects genuine biological and demographic differences as well as differences in detection. Pulmonary embolism incidence showed its own temporal dynamics: between the 2012 to 2016 window and the 2017 to 2022 window, incidence rose in Northern America and Asia but remained stable in Oceania. Rising incidence can mean more disease, but it can also mean more computed tomography pulmonary angiography, more sensitive D-dimer testing and more incidental findings on scans performed for other reasons. Disentangling true epidemiologic change from diagnostic drift is one of the central methodological puzzles the review confronts.

Mortality figures are equally variable and equally revealing. Among patients diagnosed with pulmonary embolism, 30-day all-cause mortality ranged globally from 1.0 percent to 9.0 percent, a ninefold difference across regions and studies. Perhaps the most provocative finding is that more than half of those deaths were not directly attributable to the pulmonary embolism itself. Patients who develop venous clots tend to be older, frailer and burdened with cancer, heart disease, infection or recent surgery, and competing causes of death claim many of them within the first month. This distinction matters enormously for interpreting regional comparisons: population-level pulmonary embolism-related mortality is generally higher in Northern America and Western Europe than in Asia, but whether that reflects true differences in disease severity, differences in age structure, or differences in how death certificates are completed remains unresolved.

Recurrence, the other key outcome, proved more consistent across the globe. The one-year cumulative incidence of recurrent venous thromboembolism ranged from 2.9 percent to 8.1 percent, and this figure remained fairly stable between the 2012 to 2016 and 2017 to 2022 periods. That stability is clinically meaningful. It suggests that despite the widespread adoption of direct oral anticoagulants, which have transformed outpatient treatment over the past decade, the fundamental recurrence risk has not dramatically shifted, and it reinforces the importance of risk-stratified decisions about how long patients should remain on anticoagulation. Extended-duration low-intensity apixaban strategies, currently under investigation in trials such as HI-PRO for patients with provoked clots and persistent risk factors, are one response to this persistent residual risk.

The review also catalogues the long shadow that these clots cast beyond the acute event. Survivors of deep-vein thrombosis face post-thrombotic syndrome, a chronic condition of leg pain, swelling and skin changes that can be debilitating. Survivors of pulmonary embolism can develop chronic thromboembolic pulmonary hypertension, a progressive and potentially fatal narrowing of the pulmonary arteries, as well as persistent functional impairment and reduced quality of life documented in cohorts across Europe. These non-fatal complications are a major driver of long-term disability and cost, yet they are precisely the outcomes least well captured in administrative data and least consistently reported in the literature, compounding the problem of underestimating the disease’s full burden.

Why does the global map look so lopsided? The authors point to a convergence of factors. Risk factors themselves are shifting: aging populations, rising obesity, growing cancer prevalence, increasing use of erythropoiesis-stimulating agents, and the thrombotic legacy of COVID-19 all push incidence upward, and these forces operate with different intensity in different regions. Genetic ancestry influences clot susceptibility, with well-documented variation in the prevalence of factor V Leiden, prothrombin mutations and other thrombophilias across populations. The so-called East Asian paradox, in which lower observed thrombosis rates coexist with different bleeding risks on anticoagulation, complicates the export of Western treatment algorithms. Meanwhile, misclassification, underdiagnosis and under-reporting systematically bias observed rates downward in resource-limited settings, so the absence of data should never be mistaken for the absence of disease.

The methodological machinery behind the review deserves attention in its own right. The authors aggregated population-based studies, national registries and WHO mortality database analyses, applying quality assessment tools such as the Newcastle-Ottawa Scale and harmonizing estimates where possible. They also drew on emerging validation science: recent work has shown that International Classification of Diseases coding algorithms for pulmonary embolism can be validated against chart review, and that natural language processing applied to electronic health records, including artificial intelligence triage of CT pulmonary angiograms, can improve case identification beyond what billing codes alone capture. These tools, the review suggests, will be essential for building the coordinated global surveillance system that the current data gaps demand, because standardized, validated case definitions are the precondition for any meaningful cross-country comparison.

The clinical implications extend to prevention as much as to counting. Hospitalization is a potent trigger for venous thromboembolism, and prophylaxis in medical and surgical inpatients remains underused worldwide, a gap the American Heart Association has flagged in formal calls to action. The review’s regional estimates give health systems a benchmark against which to measure their own performance: if 30-day mortality after pulmonary embolism can be 1 percent in one well-organized system and 9 percent in another, the difference is unlikely to be biology alone. Risk stratification tools, outpatient treatment pathways such as those validated in the Hestia study, contemporary guideline recommendations from both the European Society of Cardiology and the new 2026 American joint guidelines, and attention to disparities in care all represent levers that can be pulled where the will and the data exist.

What emerges from this effort is less a finished atlas than an urgent research agenda. The authors call for coordinated surveillance, standardized reporting and dedicated investigation into local and global drivers of venous thromboembolism, particularly in the regions where an estimated four million people develop these clots each year but where the evidence base is thinnest. For a condition that is largely preventable with prophylaxis and largely treatable when caught, the persistence of such stark blind spots is itself a global health failure. The review’s most important message may be that the world cannot manage what it does not measure, and that filling the empty spaces on the clot map of Sub-Saharan Africa, South Asia and beyond is now one of the most consequential unfinished tasks in cardiovascular medicine.

Subject of Research: Global epidemiology of venous thromboembolism, including deep-vein thrombosis and pulmonary embolism incidence, mortality and recurrence

Article Title: Global burden of venous thromboembolism: incidence and outcome estimates of deep-vein thrombosis and pulmonary embolism

Article References: Bikdeli, B., Rashedi, S., Pfeferman, M., Amalan, C., Leyva, H., Krishnathasan, D., Bukhari, S., Shen, B. H., Secemsky, E. A., Zhai, Z., Yamashita, Y., Sawano, M., Angchaisuksiri, P., Jimenez, D., Monreal, M., Sadeghipour, P., Zuin, M., Bertoletti, L., Ramacciotti, E., … Piazza, G. (2026). Global burden of venous thromboembolism: incidence and outcome estimates of deep-vein thrombosis and pulmonary embolism. Nature Reviews Cardiology. https://doi.org/10.1038/s41569-026-01334-3

Image Credits: AI Generated

DOI: 10.1038/s41569-026-01334-3

Keywords: venous thromboembolism, pulmonary embolism, deep-vein thrombosis, global burden of disease, epidemiology, mortality, recurrence, anticoagulation, cardiovascular disease, health surveillance, thrombosis registries, Nature Reviews Cardiology