Contemporary insights into the management of pulmonary embolism in women

This is an invited commentary on Newman et al. ⁶ The influence of patient sex on pulmonary embolism evaluation, treatment modality, and outcomes. Vasc Med 2025; 30: 68–76. and Bikdeli et al. ⁷ Sex differences in treatment strategies for pulmonary embolism in older adults: The SERIOUS-PE study of RIETE participants and US Medicare beneficiaries. Vasc Med 2025; 30: 59–67.

Venous thromboembolism (VTE) is the third most common cause of vascular-associated mortality, with increases since the COVID-19 pandemic. ¹ In the United States, more than one million adults are hospitalized with VTE events each year including 40% who have pulmonary embolism (PE). ² National surveillance data show a higher PE incidence in men than women, but prior studies have been variable regarding the relation of sex and PE outcomes.^3,4 Management of PE has evolved rapidly over the past decade with increasing implementation of multidisciplinary PE Response Teams (PERTs) along with advanced therapies. In other arenas of cardiovascular care, sex-based differences have been identified in management strategies, but detailed information for contemporary PE care have been lacking. ⁵ Two articles in this issue of Vascular Medicine provide complementary insights from large observational datasets regarding how women compared to men are managed with acute PE.

Newman and colleagues leveraged data from the PERT Consortium, which includes over 5000 adult PE cases from 35 US centers between 2015 and 2020, to evaluate the relation of sex to treatment approaches as well as risk factors. ⁶ In the second article, Bikdeli and colleagues evaluated PE management in older adults (over age 65 years) from two data sources: the European RIETE (Registro Informatizado de la Enfermedad TromboEmbólica) registry of more than 30,000 patients from 2001 to 2021, and US Medicare inpatient data from 2019 with more than 100,000 patients. ⁷ These studies should be lauded for reexamining a pressing topic to address whether or not recent changes in PE management have been applied similarly across different populations. Though in many cases the authors emphasize sex-based differences, it should be highlighted that, reassuringly, more similarities were observed than differences, including key elements such as type and duration of anticoagulation, rates of in-hospital bleeding, and overall use of advanced therapies.

For interpreting how sex-based treatment differences in PE may arise, a helpful framework is the World Health Organization (WHO) factors for evaluating sex inequalities. ⁸ This WHO model incorporates biological, clinical, and societal drivers.

From a biological standpoint, female-specific risk factors for VTE include pregnancy and hormonal therapies as well as differences in standard risk factors such as older age, prevalence of malignancy, and potentially bleeding risk. Both studies by Newman et al. ⁶ and Bikdeli et al. ⁷ report that women with PE were older. In the PERT Consortium, information was provided about the relation of the PE event with pregnancy (1.8%) and hormonal therapies (2.5% postmenopausal therapies, 6.7% oral contraceptives). There is no information from either study about gender assignments or the use of hormonal therapies by transgender patients. Further evidence from both clinical and preclinical models is needed to understand at a mechanistic level how the drivers of VTE may differ based on biological sex and the impact of female-specific factors. Hypothesized sex-based differences in bleeding risk did not translate into differences in bleeding rates or the choice and duration of anticoagulation therapy.^6,7 Interestingly, depression and use of antidepressant medications as well as reduced mobility were more common in women with PE, warranting additional studies about behavioral and psychological risk factors for thrombosis in women.^6,7,9

Clinical factors describe how interactions with the healthcare system drive sex-based differences including at the research, evaluation, and therapeutic phases. Regarding diagnosis, findings from the PERT Consortium registry suggest that echocardiography and initiation of anticoagulation may occur earlier in men compared to women. ⁶ However, the investigators only had information about whether these diagnostic tests and treatments happened before or after the PERT consultant saw the patient without any data about the timing from presentation. Thus, it is not possible to conclude that there were any treatment delays for women. It remains possible that symptom constellations or clinician interpretation of symptoms differ by sex, but further studies are needed to evaluate the intersection of presentation and timeliness of therapies. ¹⁰

Given that studies in other cardiovascular fields have suggested lower rates of interventional approaches in women, ¹¹ the results from Newman et al. ⁶ and Bikdeli et al. ⁷ regarding use of catheter-directed therapies have particular importance. In the PERT Consortium registry, lower rates of recommendation and use of catheter-based interventions in women (19.3% vs 21.9%) were reported. ⁶ In the RIETE and Medicare cohorts, no overall difference was observed in use of catheter-directed fibrinolysis (0.3% in RIETE, 3.6% vs 4.1% in Medicare). ⁷ It may appear that these findings are contradictory; however, it is important to note key distinctions in the study populations and the authors’ approach to the question of clinically meaningful treatment differences. The PERT Consortium registry had higher levels of acuity compared to RIETE (51.8% vs 38.1% intermediate risk and 25.1% vs 3.4% high risk).^6,7 In the intermediate–high-risk subgroup there was an observed higher rate of catheter-directed therapies in men. ⁶ The SERIOUS-PE investigators defined difference not based on p-value alone but on a standardized difference using a cut-off of 10% as clinically meaningful. ⁷ In our calculation of the standardized differences in the PERT Consortium registry, the observed differences in catheter-directed fibrinolysis would not meet this threshold (8.1% for recommendation and 6.4% for use), suggesting that there is concordance between the two studies.

Thus, the largest sex-based differences in PE treatment practice patterns appears to be restricted to the group of patients with the greatest clinical uncertainty: optimal patient selection for interventional techniques among those with intermediate-risk PE. It may be that women with intermediate-risk PE differ clinically from men. This is supported by the finding in RIETE that older age and active cancer were also predictors of lower rates of advanced therapies and these factors were more prevalent in women. ⁷ Sex also influences the clinical decision-making process (for both clinicians and patients). Intriguingly, in the RIETE study, there was an interaction between the sex of the treating clinician and patient in the use of fibrinolysis for intermediate–high-risk PE. ⁷ Women treated by women had a lower chance of fibrinolysis, consistent with the possibility that subtle factors are weighted differently by clinician–patient pairs.

Societal factors that may impact sex-based differences in PE include geography, gender roles, access to care, and clinician workforce diversity. As discussed above, the sex of the clinician may influence treatment choices and there remains a marked paucity of women in vascular interventional fields. ¹² Comparison of the RIETE registry with Medicare data in the SERIOUS-PE study suggests higher utilization of advanced therapies in the US compared to Europe (with the caveat that time periods are quite different), indicating geographical differences that may interact with sex differences. ⁷

The follow-up question is whether the sex differences lead to clinical disparities—is there undertreatment of women leading to poorer outcomes? Unfortunately, these two studies are limited in their ability to answer this crucial question. There is no outcome information in SERIOUS-PE, and the PERT Consortium only reported on in-hospital mortality. Univariate models showed no difference by sex, but multivariable models adjusting for many clinical covariates showed an association of sex with higher mortality.^6,7 There is no information available regarding the interaction of sex and treatment modalities with mortality nor evaluation of whether treatment mediated the relation of sex with mortality. Therefore, there is a need to better evaluate the decision for the use of advanced therapies, the relation to outcomes, and optimal outcome measure (beyond mortality). There are multiple factors that drive thresholds for advanced therapies in the background of clinical uncertainty about the best threshold. In addition to patient- and physician-related factors that may influence treatment decisions, creation of better risk models specific to both men and women for treatment strategies and posttherapy complications could help overcome external biases, if present. Forthcoming randomized control trials may provide a clearer picture of the best use of interventional therapies for the management of PE and implications for discrepant rates of usage between sexes. ¹³

The first step toward reducing sex-based health disparities is to reduce the knowledge gap about women and PE. ¹⁴ The articles by Newman et al. ⁶ and Bikdeli et al. ⁷ expand our understanding through the study of registry and administrative data derived from multiple centers and countries. One possible next step would be to leverage existing global VTE registries as platforms to conduct quality improvement studies. Implementation studies are needed to reduce sex-based disparities in the context of the larger goal of optimizing patient-centered VTE care by enhancing clinical decision-making tools.