Mechanical Thrombectomy Linked to Higher Costs in PE Study

A large national database analysis published in CVIR Endovascular indicates that mechanical thrombectomy is associated with higher in-hospital mortality, discharge other than home, intracranial hemorrhage, and total costs compared to catheter-directed thrombolysis for the management of acute intermediate- and high-risk pulmonary embolism. The study, which utilized data from the Nationwide Readmissions Database, examined 28,376 adult patients who underwent either procedure between 2016 and 2022. The findings suggest that while both interventions are used to treat severe blood clots in the lungs, the mechanical approach carries a higher burden of adverse events and financial expense in real-world clinical practice.

The research team, led by Waseem Wahood and colleagues, aimed to compare the unplanned six-month readmissions and in-hospital outcomes of two primary endovascular strategies: catheter-directed thrombolysis and mechanical thrombectomy. Catheter-directed thrombolysis involves the localized delivery of clot-dissolving drugs directly to the site of the pulmonary embolism, typically using lower dosages than systemic administration to minimize bleeding risks. Mechanical thrombectomy, on the other hand, utilizes large-bore aspiration systems or other mechanical devices to physically remove the clot, often avoiding the use of thrombolytic agents entirely. These therapies are generally reserved for patients with intermediate-high or high-risk pulmonary embolism, where the clot burden is significant enough to threaten hemodynamic stability or right ventricular function.

The study design relied on administrative claims data from the Nationwide Readmissions Database, which records approximately 35 million discharges annually from over 2,000 hospitals across 30 states. This database accounts for roughly 60% of all hospitalizations in the United States, providing a nationally representative sample. Researchers identified adult patients with acute, nonseptic pulmonary embolism who received either catheter-directed thrombolysis or mechanical thrombectomy. To ensure accurate classification, the team excluded patients with septic pulmonary embolism, acute ST-elevation myocardial infarction, and ischemic stroke. Additionally, patients who received intravenous thrombolytics in combination with the endovascular procedures were excluded to maintain mutually exclusive treatment groups.

The cohort comprised 18,168 patients who underwent catheter-directed thrombolysis and 10,208 patients who underwent mechanical thrombectomy. Among these, 3,945 patients were classified as having high-risk pulmonary embolism, representing 13.9% of the total sample. The remaining patients had intermediate-risk pulmonary embolism. To account for differences in patient characteristics between the two groups, the researchers employed inverse probability weighted regression adjustment. This statistical method estimates average treatment effects by reducing bias from measured confounders, such as age, sex, insurance status, income, and comorbidity indices. Covariate balance was assessed using standardized mean differences, with adequate balance defined as values below 0.1.

The primary outcomes measured included in-hospital mortality, discharge other than home, intracranial hemorrhage, gastrointestinal bleeding, blood transfusion, total cost at index admission, and unplanned six-month readmissions. Discharge other than home includes discharges to short-term care, skilled nursing facilities, or home health care, indicating a level of post-hospitalization care beyond standard home recovery. The study also tracked recurrent pulmonary embolism and mortality at 7 days, 30 days, and 6 months. By analyzing these multiple endpoints, the researchers sought to provide a comprehensive view of the short-term and medium-term consequences of choosing one endovascular intervention over the other.

The results indicated that mechanical thrombectomy was associated with higher average treatment effects for several adverse outcomes at index admission. Specifically, the average treatment effect for in-hospital mortality was 0.007, with a p-value of 0.048, suggesting a statistically significant increase in death during the hospital stay. The rate of discharge other than home was also higher for the mechanical thrombectomy group, with an average treatment effect of 0.031 and a p-value less than 0.001. Intracranial hemorrhage, a severe bleeding complication, occurred more frequently in the mechanical thrombectomy group, with an average treatment effect of 0.006 and a p-value less than 0.001. These findings highlight the potential safety concerns associated with the mechanical approach, particularly regarding bleeding risks and the need for extended care after the procedure.

Economic analysis revealed that mechanical thrombectomy was associated with significantly higher total hospitalization costs. The average treatment effect for cost was 0.118, corresponding to approximately 12.5% higher expenses compared to catheter-directed thrombolysis. The mean total hospitalization cost was $41,884 for mechanical thrombectomy versus $34,788 for catheter-directed thrombolysis. This cost differential likely reflects the higher per-case costs of large-bore mechanical thrombectomy devices, which can range from $8,000 to $12,000, compared to the generically available pharmaceuticals used in catheter-directed thrombolysis. Additionally, the mechanical thrombectomy group had a longer mean length of stay, at 6.4 days compared to 5.9 days for the catheter-directed thrombolysis group, further contributing to the overall financial burden.

At the six-month follow-up, mechanical thrombectomy was associated with small absolute increases in mortality, gastrointestinal bleeding, and blood transfusions. The average treatment effect for mortality at six months was 0.009, with a p-value of 0.025. Gastrointestinal bleeding also showed a higher rate in the mechanical thrombectomy group, with an average treatment effect of 0.004 and a p-value of 0.023. However, no significant differences were observed in unplanned readmission, intracranial hemorrhage, or recurrent pulmonary embolism at six months. This suggests that while the mechanical approach may carry higher immediate risks and costs, the long-term rates of clot recurrence and readmission may not differ significantly from catheter-directed thrombolysis in this real-world population.

Subgroup analyses provided further insights into the differences between intermediate- and high-risk pulmonary embolism. In the intermediate-risk subgroup, mechanical thrombectomy was associated with higher rates of intracranial hemorrhage, blood transfusion, and total cost at index admission. At six months, this subgroup also experienced higher rates of unplanned readmission, gastrointestinal bleeding, recurrent pulmonary embolism, blood transfusion, and mortality. In contrast, the high-risk subgroup showed different patterns, with mechanical thrombectomy associated with higher discharge other than home and total cost, but no significant difference in in-hospital mortality. Catheter-directed thrombolysis was associated with higher six-month pulmonary embolism readmission in the high-risk group, suggesting that treatment-associated patterns may vary depending on the severity of the initial presentation.

The discussion of the findings emphasizes the importance of considering residual confounding by indication. Patients undergoing mechanical thrombectomy often have higher baseline illness severity, including massive pulmonary embolism and hemodynamic instability, which may independently increase the risk of adverse outcomes. Although inverse probability weighting improved covariate balance, unmeasured clinical severity variables and selection factors may still influence the results. The study notes that administrative claims data cannot capture specific physiological markers such as right ventricular dysfunction, clot burden, or laboratory values, which are crucial for contemporary risk stratification. Therefore, the observed associations should be interpreted with caution, as they reflect real-world patterns rather than causal effects.

The study also compares its findings with those from randomized controlled trials, such as the PEERLESS trial and the HI-PEITHO trial. The PEERLESS trial reported favorable outcomes for mechanical thrombectomy using a hierarchical composite endpoint, but no significant differences were identified for mortality or intracranial hemorrhage when individual components were examined. The HI-PEITHO trial demonstrated that ultrasound-facilitated catheter-directed thrombolysis reduced the composite outcome of hemodynamic decompensation or death compared to anticoagulation alone. These trials were conducted in selected populations under controlled conditions, which may limit the generalizability of their findings to the broader spectrum of patients encountered in routine clinical practice. The current study provides valuable real-world insights that complement these trial-based findings.

Limitations of the study include the use of administrative claims data, which is prone to misclassification of International Classification of Diseases codes. The stratification of pulmonary embolism risk was based on these codes, which may not fully replicate contemporary guideline-based risk stratification. Additionally, the database does not capture in-hospital fibrinolytic dosing or the specific devices used for thrombectomy, making it difficult to distinguish between different technical profiles. The study period from 2016 to 2022 spans a time of evolving thrombectomy technology, meaning the results reflect a composite real-world experience rather than a specific device or technique. Despite these limitations, the study offers a comprehensive national-level analysis that highlights the need for prospective randomized data to define the optimal role of each strategy in contemporary pulmonary embolism care.

Subject of Research: Comparison of endovascular interventions for acute pulmonary embolism

Article Title: Unplanned 6-month readmissions and in-hospital outcomes for the management of acute intermediate- and high-risk pulmonary embolism: comparisons of endovascular interventions

Article References: Wahood, W., Hofmann, H., Seyam, O., Iyer, D., Monteleone, P., Paul, J. D., & Ahmed, O. (2026). Unplanned 6-month readmissions and in-hospital outcomes for the management of acute intermediate- and high-risk pulmonary embolism: comparisons of endovascular interventions. CVIR Endovascular, 9(1), Article 99. https://doi.org/10.1186/s42155-026-00752-1

Image Credits: AI Generated

DOI: 10.1186/s42155-026-00752-1

Keywords: pulmonary embolism, mechanical thrombectomy, catheter-directed thrombolysis, endovascular intervention, hospital readmission, healthcare costs, clinical outcomes, observational study, Unplanned, month, readmissions, in-hospital