Anticoagulants vs thrombolytic therapy for the management of right heart thrombus in pulmonary embolism: A systematic review

Abstract

BACKGROUND

Right heart thrombi (RHT) are associated with high mortality in patients with pulmonary embolism (PE). Thrombolytic therapy (TLT) and anticoagulation therapy (ACT) are the main treatment options. However, the optimal treatment strategy remains controversial due to a lack of randomized controlled trials.

AIM

To compare the efficacy and safety of TLT vs isolated ACT in patients with RHT and PE.

METHODS

We searched PubMed, Scopus, and Wiley Online Library from database inception to January 20, 2026, using a comprehensive strategy combining terms for RHT (including “right heart thrombus”, “right ventricular thrombus”, “atrial thrombus”, “thrombus in transit”), PE (“pulmonary embolism” OR “PE”), and thrombolysis (“thrombolysis”, “fibrinolysis”, “alteplase”, “tenecteplase”, and “streptokinase”). Inclusion criteria involved the following: Adults with confirmed RHT in acute PE; direct comparison of TLT + ACT vs ACT alone; intention-to-treat design; and case series with n > 5 per group.

RESULTS

Five observational studies (n = 205 patients) were included. TLT did not significantly reduce early all-cause mortality during hospitalization (OR = 0.36, 95%CI: 0.10-1.29, and P = 0.12). The heterogeneity of the results across the included studies was low and not statistically significant (I² = 18% and P = 0.30). TLT also did not significantly reduce all-cause mortality at 30 days (OR = 0.26, 95%CI: 0.04-1.64, and P = 0.15). The heterogeneity of the results across the included studies was moderate and not statistically significant (I² = 51% and P = 0.13). No subgroup analyses by hemodynamic status or thrombus type were possible.

CONCLUSION

Routine thrombolysis does not significantly reduce mortality compared with ACT alone in the overall cohort of RHT and PE patients.

Key Words: Right heart thrombi; Pulmonary embolism; Thrombolytic therapy; Anticoagulants; Mortality; Heparin; Hemodynamic instability

Core Tip: Management of patients with right heart thrombi in pulmonary embolism (PE) remains challenging. This meta-analysis of 205 patients did not reveal a significant advantage of routine thrombolytic therapy over isolated anticoagulation in reducing mortality. Our data do not support universal use of either strategy. Treatment decisions should be individualized, and clinicians should follow current guidelines, which recommend thrombolysis for hemodynamically unstable patients with massive PE.

INTRODUCTION

Right heart thrombi (RHT) are a relatively rare clinical phenomenon. They can be a consequence of deep vein thrombosis of the lower extremities and venous embolism or may develop due to reduced blood flow velocity in the context of atrial fibrillation, severe right ventricular (RV) dysfunction, or pulmonary hypertension, or due to the development of hypercoagulable syndrome in primary cardiac tumors or metastatic cardiac involvement. Patients with mechanical heart valves, pacemakers, central venous catheters, and endocarditis also have an increased risk of thrombus formation[1-5]. RHT can be located in the right atrium (RA) or the RV, and they can prolapse from the proximal part of the inferior vena cava into the RA or extend from the RV outflow tract into the pulmonary artery trunk, including floating through the pulmonary valve[5]. RHT are most commonly encountered in pulmonary embolism (PE), with the detection rate in PE ranging from 2.6% to 23%[6-10].

There are three types of RHT. Type A thrombus is the most common. It occurs when a thrombus from the deep veins of the lower extremities migrates into the right heart chambers. Type A thrombi lie freely within the heart chambers and have an elongated, worm-like shape. They pose the highest risk of embolism to pulmonary circulation. They are associated with PE and lower extremity deep vein thrombosis in 79%-98% of cases[10]. Type B thrombi have an oval shape, a broad base, and are attached to the heart chamber wall. Their source is presumed to be the atrium or ventricle. An association with PE is noted in 38%-40% of cases and is characterized by less severe clinical manifestations[10]. Type C thrombi combine the features of type A and B thrombi. They are rare, extremely mobile clots that resemble cardiac myxomas[2] and have an intermediate degree of association with PE (62%-67%)[10].

Type A thrombi are associated with high early mortality, reaching 42%, whereas for type B thrombi, the mortality rate is 4%[11]. Overall mortality for RHT is 31%. It is suggested that the adverse outcome is not so much due to the mobility of the thrombus itself, but rather to its combination with severe PE and the chosen treatment strategy[12].

To date, there is no single consensus regarding the optimal treatment method for RHT. Available data are based mainly on descriptions of clinical case series, which are often subject to selection bias in favor of a particular treatment method. The main therapeutic strategies include anticoagulant therapy (ACT) with heparin, systemic thrombolytic therapy (TLT), surgical embolectomy, and modern minimally invasive catheter-directed techniques[13-16].

Due to the fact that access to high-tech surgical care and catheter-directed techniques is limited in many medical centers, our meta-analysis is dedicated to a comparative analysis of the efficacy and safety of TLT (in combination with subsequent ACT) vs ACT alone.

MATERIALS AND METHODS

This systematic review and meta-analysis were performed in accordance with the PRISMA guidelines.

Search strategy

Two investigators independently conducted a search of electronic databases up to January 20, 2026. Relevant articles were identified in PubMed, Scopus, and the Wiley Online Library from database inception to January 20, 2026, using the following search query: (“right heart thrombi” OR “right heart thrombus” OR “right ventricular thrombus” OR “atrial thrombus” OR “thrombus in transit”) AND (“pulmonary embolism” OR “PE”) AND (“thrombolysis” OR “thrombolytic” OR “fibrinolysis” OR “alteplase” OR “tenecteplase” OR “streptokinase”).

Selection process

After screening titles and abstracts, two authors independently performed full-text assessments to determine whether the studies met the eligibility criteria. Disagreements were resolved through discussion with a third investigator. Studies were included if they met the following parameters (Figure 1).

Figure 1 Study selection algorithm.

Population: Adult patients (≥ 18 years) with confirmed RHT in the setting of acute PE or suspected PE.

Intervention and comparison: A group receiving systemic TLT followed by ACT and a group receiving ACT alone. An inclusion criterion was the ability to distinguish these groups from the presented clinical series.

Outcomes: Early (in-hospital) and 30-day mortality, PE recurrence rate, major bleeding rate, and thrombus resolution rate. The analysis was performed on an intention-to-treat basis: Outcomes were assessed based on the initially assigned therapy, regardless of whether the treatment was completed in full or changed during the process.

Study type: Randomized controlled trials, prospective and retrospective cohort studies, registries, and case series with n > 5 patients. We included small retrospective case series (n > 5 patients) because this pathology is rare and large prospective studies are limited. This threshold was chosen to minimize publication bias, which is characteristic of single descriptions of unique cases. When selecting case series, studies with ≤ 2 patients in one of the comparison groups were excluded due to the difficulty of statistical outcome assessment.

Access: Access to the full text of the article.

Exclusion criteria: Tumor thrombi (e.g., renal cancer invasion), vegetations in infective endocarditis, thrombi on pacemaker leads, and isolated left heart thrombi. The following were excluded: Systematic reviews and meta-analyses by other authors (used exclusively to search for primary sources to avoid double-counting data); single clinical cases and series with fewer than 5 patients, which do not allow for adequate statistical evaluation; and studies with discrepancies in numerical data between the abstract, main text, tables, or figures.

An additional exclusion criterion was publications describing patients with confirmed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. This decision was based on the specific pathogenesis of SARS-CoV-2-associated coagulopathy (immunothrombosis), which differs from classical mechanisms of venous thromboembolism, as well as differences in anticoagulation protocols and the hemorrhagic risk profile in this cohort.

Additional search

A manual search of reference lists of selected articles and systematic reviews was also conducted to identify relevant primary sources.

Search results

During screening based on titles and abstracts, 188 articles were identified in PubMed, 165 in Scopus, and 351 in the Wiley Online Library; after removing duplicates, 488 articles remained. At the abstract reading stage, 433 articles were excluded: 2 articles described cases of RHT in patients aged < 18 years (Supplementary material), 36 articles focused on left heart thrombi and strokes, 20 articles described tumor and infectious thrombi, 15 articles described chronic thromboembolic pulmonary hypertension, 13 articles were dedicated exclusively to the diagnosis of RHT, 244 articles described single clinical cases, 98 articles did not match the research topic, 4 articles were about animals, and 1 article was retracted.

Thus, 55 articles were selected for full-text review. During detailed analysis, 50 studies were excluded for the following reasons: 27 articles were systematic analyses and reviews (Supplementary material); 7 articles described small patient groups within treatment arms, precluding adequate statistical evaluation; 6 studies were excluded due to the impossibility of clearly assigning patients to therapy groups (TLT vs ACT); 3 studies contained overlapping data; 3 articles were re-analyses of previously published clinical cases; 3 studies lacked a comparison group (described only one treatment strategy); and 1 article was removed due to critically poor methodological quality.

Handling overlapping data and final selection

We identified three studies from the Registro Informatizado de Enfermedad Trombo Embólica (RIETE) registry with overlapping data[8,17,18]. To avoid double-counting patients, preference was given to the study by Bikdeli et al[18] as the most complete and up-to-date. This study includes follow-up data for 546 patients (up to 2019), significantly expanding the sample compared to the previously published analysis of the same registry (Barrios et al[17]), which included 325 cases. To obtain missing primary data from the Bikdeli et al[18], an official request was sent to the study authors. The missing data were provided by the authors through personal correspondence, allowing the full inclusion of this analysis in our statistical calculations. However, the authors could not provide the requested level of detail for data from the post hoc analysis (number of outcomes in the subgroup of patients who received TLT); thus, we used calculated data.

We did not include the studies by Rose et al[13], Athappan et al[19], and Burgos et al[20] in the quantitative analysis, as these studies represent pooled analyses of clinical cases published in the literature. Their inclusion would lead to data duplication and artificially inflate statistical power. Although the study by Burgos et al[20] analyzed registered clinical cases of RHT from 2006 to 2016, it was not included because the studies in our meta-analysis rely on the intention-to-treat principle, while the Burgos et al’s study[20] classifies patients according to the “as-treated” principle. This approach can lead to selection bias, as patients with the most severe disease who do not survive to receive reperfusion therapy are automatically assigned to the conservative therapy group. For the same reason, we excluded the pooled analysis by Islam et al[21], which also lacks an intention-to-treat design and would lead to data duplication with our primary studies, as it covers the period from 1956 to 2017. We also did not include the institutional data from the Islam et al[21] because outcomes in that cohort were analyzed according to the astreated principle.

Risk of bias assessment

The validity and methodological quality of non-randomized studies were assessed using the Quality In Prognosis Studies tool[22], which includes six domains of potential bias: (1) Study participation; (2) Study attrition; (3) Prognostic factor measurement; (4) Outcome measurement; (5) Study confounding; and (6) Statistical analysis and reporting[22]. When interpreting individual domains, the study design, period of conduct, and clinical context were considered. In ambiguous cases, the final domain assessment was performed in favor of reflecting the actual capabilities of the study rather than formal compliance with modern requirements. All included studies were judged to have a moderate risk of bias (Figure 2). Formal assessment of publication bias (e.g., funnel plot, Egger’s test) was not performed due to the small number of included studies (n < 10).

Figure 2 Risk of bias assessment in prognostic studies using the Quality In Prognosis Studies tool.

Statistical analysis

Statistical data processing was performed using Review Manager software, version 5.4.1 (the Cochrane Collaboration, 2020). The meta-analysis was conducted using a random-effects model with the inverse variance method. Statistical heterogeneity was assessed using the χ²-based Q test and the I² heterogeneity index. Interpretation of I² followed the Cochrane Collaboration’s recommendations with nonoverlapping thresholds: I² = 0%-40% indicated insignificant heterogeneity; 30%-60%-moderate heterogeneity; 50%-90%-substantial heterogeneity; and 75%-100%-considerable heterogeneity.

RESULTS

Characteristics of included studies

A total of 205 patients were included in the analysis, whose mortality analysis covered the hospitalization, 14-day follow-up, and 30-day follow-up periods. The studies by Chartier et al[23], Torbicki et al[14], and Pierre-Justin and Pierard[15] were included in the analysis of early mortality. For the analysis of 30-day mortality, we included the studies by Pierre-Justin and Pierard[15], Seghda et al[7], and Bikdeli et al[18], since the former stated that all discharged patients were alive at the 30-day follow-up, while the latter two studies did not specify the exact timing of death for patients who died during the long-term follow-up period. Consequently, five studies were selected for the final analysis; their characteristics are presented in Table 1.

Table 1 Characteristics of studies included in the analysis.

Ref.	Chartier et al[23]	Torbicki et al[14]	Pierre-Justin and Pierard[15]	Seghda et al[7]	Bikdeli et al[18]
Design	Retrospective analysis of a hospital registry	Subgroup analysis of the ICOPER international prospective registry	Prospective single-center study	Prospective cohort study	Analysis of data from the international prospective RIETE registry
Country/center	France (Dijon/Lille)	Multicenter (52 centers, 7 countries)	Martinique (France), Fort-de-France	Burkina Faso, Ouagadougou	Multicenter (international)
Inclusion period	January 1986-January 1998 January	1995-November 1996 November	1997-June 1999	March 2012-September 2015	March 2001-September 2019
Total patients with PE (n)	341 (for the period 1992–1997)	2454	335	250	42620
RHT (n)	38, 28 from 1992 to 1997	42 of 1135 in whom echocardiography was performed	12	11	443 of 18803 in whom echocardiography was performed. Post hoc analysis n = 124
Prevalence of RHT	8.2% (28/341) in 1992-1997	3.7% (42/1135)	4.0% (12/335)	4% (11/250)	2.4% (443/18803)
RHT morphology	Worm-like (36/38), spherical (2/38)	Not described	Mobile: 9 (75%)-coil, 3 (25%)-ball	Mobile thrombi. Specific shape not described	Not described
Location of RHT	RA-30, RV-6, both chambers-2	RA/RV	All-RA (free-floating, could prolapse into RV)	RA in all cases (100%)	Right heart chambers
PE severity	Severe: NYHA IV-84% (32/38), cardiogenic shock-53% (20/38), cardiac arrest-47% (18/38)	More severe course than without RHT: More frequent hypotension, tachycardia, RV dysfunction, and heart failure	Massive PE: All in ICU, syncope (100%), hypotension (mean SBP 96 mmHg), acute pulmonary heart disease in all	All-high/intermediate-high risk	In those who received reperfusion therapy-more frequent hypotension (SBP < 100 mmHg), tachycardia (HR ≥ 110/minutes), syncope, hypoxemia, and RV hypokinesis
Anticoagulant monotherapy (1st line) (n)	8	17	5 (including 3 with absolute contraindications to thrombolysis)	5	62 (post hoc analysis)
Mortality with anticoagulant monotherapy (as first-line treatment)	In-hospital: 5/8 (62.5%)	14-day: 4/17 (23.5%); 3-month: 5/17 (29.4%)	In-hospital: 3/5 (60%)-all with contraindications to thrombolysis	30-day: 4/5 (80%)	30-day post hoc analysis-7/62 (11.3%)
Thrombolysis (1st line) (n)	9	24	7	6	62 (post hoc analysis)
Mortality with thrombolysis (as first-line treatment)	2/9 (22.2%)	14-day: 5/24 (20.8%); 3-month: 7/24 (29.2%)	1/7 (14.3%)	30-day: 1/6 (16.7%)	30-day post hoc analysis-6/62 (9.7%)
Confirmed thrombus lysis	Echocardiographic monitoring was used, but data on the frequency of thrombus lysis are not presented	Not described	In 7 of 9 patients (77.8%) who received thrombolysis, complete thrombus lysis and resolution of RV overload were documented on echocardiography at 12 hours	In all 6 patients who received thrombolysis, complete thrombus lysis was documented on follow-up echocardiography	Not described
Major bleeding	Not described	Not described	None reported	2/11 (18.2%) in the RHT group	Absolute values not provided in the article. Post hoc analysis: 6/62 (9.7%). Reperfusion: 4/62 (6.5%)
Recurrent PE	Not observed during the follow-up period	Not described	0% during hospitalization and 1-year follow-up in survivors	Not described	Not described
Early mortality (14-day/in-hospital)	17/38 (44.7%)	9/42 (21.4%)	4/12 (33.3%)
30-day mortality			4/12 (33.3%)	5/11 (45.4%)	40/443 (9.0%), in post hoc analysis-13/124 (10.5%)
Follow-up period	Mean 47.2 months (range 1-70 months)	3 months	12 months	30 days	30 days
Mortality during the follow-up period	20/38 (52.6%). Among the discharged: 3/21 (14.2%)	12/42 (28.6%)	4/12 (33.3%), among the discharged: 0%	5/11 (45.4%)	40/443 (9.0%) from PE: 24/443 (5.4%)
Authors’ key conclusion on treatment	No significant difference in mortality between methods. Thrombolysis is recommended as a rapid and accessible first-line therapy, especially in the absence of contraindications. Surgery remains the classic method, and catheter-based techniques are an alternative when contraindications exist	Patients with RA/RV thrombi have a higher risk of death, especially if treated with heparin alone. The authors suggest that anticoagulation alone may be insufficient even in clinically stable patients, and more aggressive methods (thrombolysis, embolectomy) should be considered, although this requires further research	Thrombolysis is an effective, rapid, and safe first-line therapy for patients with mobile thrombi. Heparin alone (monotherapy) in severe forms is ineffective and leads to 100% mortality. Surgery is a rescue method when thrombolysis fails	Thrombolysis significantly reduced mortality in the group with thrombi. In the absence of thrombolysis, 30-day survival was < 25%. Thrombolysis is the preferred option when surgical or percutaneous embolectomy is not available	Reperfusion therapy (predominantly thrombolysis) in patients with acute PE and concomitant RHT did not demonstrate a statistically significant reduction in 30-day mortality compared with anticoagulant monotherapy; however, the point estimate of effect (OR = 0.65 for PE-related death) does not rule out clinically meaningful benefit, and the wider confidence intervals indicate the need for further research

ACT: Anticoagulant therapy; ICU: Intensive care unit; OR: Odds ratio; RV: Right ventricle; RA: Right atrium; SBP: Systolic blood pressure; HF: Heart failure; TLT: Thrombolytic therapy; RHT: Right heart thrombi; PE: Pulmonary embolism; HR: Heart rate; EchoCG: Echocardiography; ICOPER: International Cooperative Pulmonary Embolism Registry; NYHA: New York Heart Association; RIETE: Registro Informatizado de Enfermedad Trombo Embólica.

Thrombus morphology

Formal classification into types A, B, or C was not available in the original studies; most reported thrombi were mobile or wormlike, which corresponds to the type A morphology.

Mortality outcomes

According to the presented forest plot (Figure 3A), the use of TLT did not statistically significantly reduce the risk of overall mortality in patients with RHT and PE during the hospitalization period [odds ratio (OR) = 0.36, 95%CI: 0.10-1.29, and P = 0.12]. The heterogeneity of the results of the included studies was low and statistically non-significant (I² = 18% and P = 0.30).

Figure 3 Results of the meta-analysis of odds ratios. A: For all-cause mortality in patients with right heart thrombi (RHT) and pulmonary embolism (PE); B: For 30-day all-cause mortality in patients with RHT and PE. Red squares represent the weight of each individual study (the size of the squares corresponds to the study weight); black lines represent the 95%CI; the black diamond represents the pooled odds ratio.

In the analysis of 30-day mortality (Figure 3B), the use of TLT did not statistically significantly reduce the overall risk of mortality in patients with RHT and PE (OR = 0.26, 95%CI: 0.04-1.64, and P = 0.15). The heterogeneity of the results of the included studies was moderate and not statistically significant (I² = 51% and P = 0.13).

Other prespecified outcomes

PE recurrence, major bleeding, and thrombus resolution were prespecified for assessment but could not be meta-analyzed due to insufficient data.

DISCUSSION

The problem of choosing a treatment strategy for RHT in PE has been debated for decades. In the first descriptions of case series (Boulay et al[24]), the authors noted the successful use of either TLT or ACT alone, but the small sample sizes did not allow for definitive conclusions.

An additional difficulty in managing such patients is the need for rapid decision-making, since RHT are associated with an extremely high risk of fatal complications in the first hours after detection. For instance, according to Chapoutot et al[25], in-hospital mortality reached 29%, with a significant proportion of patients dying even before the initiation of specific therapy or during transportation, which dictates the need for the fastest possible diagnosis and selection of a treatment strategy.

The conducted systematic review and meta-analysis, which included 205 patients with RHT in the setting of PE from five observational studies, did not reveal a statistically significant advantage of systemic thrombolysis over ACT alone with respect to early (OR = 0.36; 95%CI: 0.10-1.29; and P = 0.12) and 30-day mortality (OR = 0.26; 95%CI: 0.04-1.64; and P = 0.15). Although the point estimates suggest a trend towards reduced mortality with TLT, the confidence intervals are wide and cross unity, precluding a definitive conclusion about its superiority.

Our results contrast with the findings of a number of previously published pooled analyses, which reported a significant reduction in mortality with the use of TLT. For example, in the pooled analysis by Rose et al[13], which included 177 patients (1966-2000), TLT was associated with lower mortality compared to ACT (OR = 0.33; 95%CI: 0.11-0.98). However, this study did not account for baseline hemodynamic status or the reasons for therapy selection, which may have led to selection bias. Athappan et al[19], in a pooled analysis of 328 patients (1992-2013), also showed an advantage of TLT over ACT after adjusting for age and hemodynamic instability (survival OR = 4.83; 95%CI: 1.52-15.36). Burgos et al[20], based on 207 cases (2006-2016), confirmed that anticoagulant monotherapy is an independent predictor of death (OR = 2.4; 95%CI: 1.07-5.4) compared to TLT or surgery.

Similar results were obtained in the pooled analysis by Islam et al[21], which included 316 patients (1956-2017): Systemic thrombolysis was associated with higher odds of survival compared to anticoagulation alone (OR = 2.72; 95%: 1.11-6.64) after adjustment for shock. However, all these studies had significant selection bias: Patients in the most severe, unstable condition (cardiogenic shock or hypotension) were more often prescribed thrombolysis as life-saving therapy. At the same time, the comparison group (ACT) could automatically include either stable patients or those with absolute contraindications to TLT, which in itself is a negative prognostic factor. Moreover, these studies analyze therapy according to the astreated principle (treatment actually received) rather than intention-to-treat. Interestingly, in the institutional case series by Islam et al[21], also analyzed on an as-treated basis, mortality in the TLT group was 75% (three out of four), whereas in the ACT group it was 0% (0 out of 8).

More recent large registries (such as RIETE) using statistical matching methods (propensity matching) provide important adjustments, mitigating this effect. Thus, the most extensive data from the RIETE registry[18] did not reveal a statistically significant advantage of TLT in terms of 30-day survival. Moreover, a higher risk of PE recurrence was noted in the reperfusion group (5.8% vs 1.2% in the anticoagulant group). This is extremely important, as, according to the literature, patients with RHT initially have a higher rate of PE recurrence compared to patients without RHT (6.6% vs 2.3%)[26]. However, it should be noted that an in-depth analysis of the impact of the chosen therapy on the risk of PE recurrence could not be performed in our meta-analysis due to limited data on this outcome.

At the same time, a recent meta-analysis by Maqsood et al[27], which included 492 patients with floating RHT and PE, presented pooled mortality rates for each strategy (a single-group meta-analysis) without performing a direct comparison between them. Systemic thrombolysis was associated with the lowest weighted mortality-12% (95%CI: 5%-19%)-whereas for ACT alone this figure was 35% (95%CI: 21%-49%). The authors did not calculate OR for a direct comparison of the methods, but the presented data also indicate a potential benefit of reperfusion therapy.

In our meta-analysis, we were also unable to compare the efficacy of TLT and ACT on thrombus lysis. In several of the included studies[7,15], thrombolysis ensured rapid thrombus lysis (often within 12 hours), but such information was not provided in other included studies, nor were thrombus types described. Due to limited data, we were unable to perform an analysis of the impact of the assigned therapy on the risk of bleeding.

Our results show that in the pooled cohort of patients with RHT and PE, adding thrombolysis to anticoagulation did not lead to a statistically significant reduction in mortality compared with anticoagulation alone. This means that routine thrombolysis in all patients is not justified. However, due to the lack of randomization and subgroup analyses, we cannot exclude a potential benefit of TLT in certain patient categories (e.g., those with hemodynamic instability or mobile thrombi). Treatment decisions should be individualized based on clinical judgment. Current clinical guidelines recommend thrombolysis for hemodynamically unstable patients with massive PE, which is not contradicted by our findings. Creager et al[28] also emphasize that RHT are associated with a high risk of mortality, but they do not provide clear recommendations regarding the management of such patients.

The main feature of this systematic review and meta-analysis, unlike previous pooled analyses that assessed outcomes according to the treatment actually received (as-treated), is that it analyzed outcomes according to the initially assigned treatment (intention-to-treat). This approach was chosen because, based on the available data, crossovers between treatment groups are frequent in this patient population. Thus, we fill a gap in the literature by providing an estimate of treatment effectiveness that is free from this selection bias. In addition, we attempted to exclude overlapping data from pooled analyses and to minimize selection bias. Our findings do not support routine thrombolysis in all patients with RHT and PE, and they confirm the need for prospective studies dedicated to this question. However, given the low incidence of RHT, such studies are unlikely to be conducted in the near future.

Limitations

The main limitations of our work, like those of most studies in this field, are related to the nature of the pathology under study-its rarity precludes a large-scale randomized controlled trial. We were forced to include observational studies, which inevitably introduces a risk of bias. The key methodological challenges are as follows: (1) The meta-analysis is based on only five studies. In some of them (for example, the studies by Pierre-Justin and Pierard[15] and Seghda et al[7]), the patient groups are extremely small (11-12 patients), which limits statistical power and does not allow for extrapolating the conclusions to the entire population of patients with RHT; (2) Due to the impossibility of obtaining complete primary data details for one of the key studies (Bikdeli et al[18]), we were forced to use calculated estimates. This introduces a certain degree of assumption into the final statistical calculation and may slightly influence the final risk ratios; (3) Selection bias. Patients with the most severe condition were more likely to undergo reperfusion; (4) We analyzed patient outcomes according to the initially assigned therapy; (5) The lack of individual patient data did not allow for a more in-depth stratified analysis by thrombus morphology (A, B, and C), although it is reasonable to assume that the treatment strategy for a mobile thrombus (type A) and a mural thrombus (type B) may differ; (6) We did not perform subgroup analyses (e.g., by hemodynamic status or thrombus type) because the original studies did not provide data in such a stratified manner. Therefore, our conclusions apply to the pooled cohort as a whole and cannot be directly used for decision-making in specific patient subgroups (e.g., stable or unstable); and (7) The small number of included studies (n = 5) precluded formal testing for publication bias (e.g., funnel plot asymmetry or Egger’s regression test). While we attempted to minimize bias by including only studies with n > 5 per group and using an intention-to-treat approach, the possibility of unpublished negative studies cannot be excluded.

CONCLUSION

This meta-analysis does not support a statistically significant advantage of thrombolysis over anticoagulation alone in reducing mortality in patients with RHT and PE. Wide confidence intervals preclude definitive conclusions about the superiority of either strategy. No stratified analysis by hemodynamic status or thrombus type was performed. Therefore, our findings apply to the pooled cohort as a whole and should not be used to guide therapy in specific subgroups without additional evidence. Current guidelines recommend thrombolysis for hemodynamically unstable patients with massive PE. Prospective randomized trials with clinical subgroup stratification are needed.