The net benefit of thrombolysis in the management of intermediate risk pulmonary embolism: Systematic review and meta‐analysis

Abstract Background Benefit of thrombolytic therapy in patients with massive pulmonary embolism (PE) is evident. However, evidence supporting benefit in clinical outcomes of this approach in intermediate risk PE is lacking. Objective To determine the impact of thrombolysis on overall survival in intermediate risk PE patients. Methods We searched in MEDLINE (OVID), EMBASE, LILACS, and the Cochrane Central Register of Controlled Trials (CENTRAL) from present day. We also searched in other databases and unpublished literature. We included clinical trials without language restrictions. The risk of bias was evaluated with the Cochrane Collaboration's tool. The primary outcome was overall survival. Secondary outcomes were adverse events, including major bleeding, and all‐cause mortality. The measure of the effect was the risk ratio with a 95% confidence interval (CI). Results We included 11 studies in the qualitative and quantitative analysis, with a total of 1855 patients. Risk of bias was variable among the study items. There were no results reported about overall survival in any of the studies. The risk ratio (RR) for all‐cause mortality was 0.68 95% CI (0.40 to 1.16). The RR of overall bleeding, major bleeding and stroke were 2.72 95% CI (1.58 to 4.69), 2.17 95% CI (1.03 to 4.55), and 2.22 95% CI (0.17 to 28.73), respectively. Additionally, the RR for recurrent PE was 0.56 95% CI (0.23 to 1.37). Conclusions In patients with intermediate risk PE, the risk of bleeding is higher when thrombolysis is used. There was no significant difference between thrombolysis and anticoagulation in recurrence of PE, stroke, and all‐cause mortality.

A variety of therapeutic strategies have been evaluated for the acute management of high-risk situations, and they include predominantly thrombolytic procedures. Thrombolysis has proven to be a treatment modality with survival benefit in patients with massive PE [12]; however, the risk: benefit ratio is less clear in cases that present without hypotension [13]. In intermediate risk PE, there are data supporting the use of thrombolytic procedures to improve hemodynamic parameters, although the mortality benefit of these interventions in the light of the risk of bleeding remains uncertain [14].
The current standard of care for the management of acute PE with hemodynamic stability is systemic anticoagulation, and duration of treatment varies based on the presence of risks factors and the overall risk of bleeding in patient-specific populations [15]. The recommendation for the use of thrombolytic therapy is limited to acute PE with hypotension after excluding major contraindications associated with a higher risk of bleeding (intracranial metastases, thrombocytopenia, recent bleeding, etc.) [5,16].
There is no homogeneous consensus among different guidelines on the management of patients with intermediate risk PE [15,17]. The objective of this meta-analysis was to determine the clinical benefits and harms of thrombolysis followed by anticoagulation versus anticoagulation alone, in patients with intermediate risk PE.

METHODS
We performed this review according to the recommendations of the Cochrane Collaboration and following the PRISMA Statement. The PROSPERO registration number is CRD42019128229.

Information sources
Literature search was conducted in accordance to recommendations by Cochrane. We used medical subject headings (MeSh), Emtree language, Decs and text words related. We searched MEDLINE (OVID), EMBASE, LILACS, and the Cochrane Central Register of Controlled Trials (CENTRAL) from inception to nowadays (Appendix 1). To ensure literature saturation, we scanned references from relevant articles identified through the search, conferences, thesis databases, Open Gray database, Google scholar, and clinicaltrials.gov, among others. We contacted authors by e-mail in case of missing information.

Data collection
Each reference by title and abstract was reviewed by three researchers. Then, full-text of relevant studies were scanned for data extraction, applying prespecified inclusion, and exclusion criteria. Disagreements were resolved by consensus among the three authors. Using a standardized form, one reviewer extracted the following information from each article: study design, geographic location, author's names, title, objective, inclusion and exclusion criteria, number of patients included, losses to follow-up, timing, definition of outcomes, outcomes and association measures, and funding source.

Risk of bias
For clinical trials, the Cochrane Collaboration tool was utilized, which covers sequence generation, allocation concealment, blinding, incomplete outcome data, selective reporting, and other biases. Researchers judged about the possible risk of bias from the extracted information, rating it as "high risk," "low risk," or "unclear risk," Graphic representation of potential bias was computed using Review Manager 5.3 (RevMan ® 5.3).

Data analysis/synthesis of results
The statistical analysis was performed using RevMan ® 5.3. For categorical outcomes, information about risk ratios (RR) with 95% confidence intervals was reported, and the information was pooled with a random effect meta-analysis according to the heterogeneity expected. The results were reported in forest plots of the estimated effect of the included studies with a 95% confidence interval (95% CI). Heterogeneity was evaluated using the I 2 test. For the interpretation, it was determined that values of 25%, 50%, and 75% correspond to low, medium, and high levels of heterogeneity, respectively.

Publication bias
Potential publication bias was assessed using a funnel plot for all-cause mortality and overall bleeding since these were the outcomes evaluated by more than 10 studies.

Sensitivity analysis
We performed sensitivity analysis extracting weighted studies and running the estimated effect to find differences.

Subgroup analysis
We performed a subgroup analysis based on the specific type of thrombolytic used (r-TPA vs nonrecombinant TPA).

Risk of bias
Most studies had low risk of bias for random sequence generation, blinding of outcome assessment, selective reporting, and other bias. On the other side, regarding the blinding of participants and personnel, incomplete outcome data, and allocation concealment issues, there was unclear and high-risk bias for most of the studies (Figure 2A and 2B).

Primary outcome: overall survival
There were no results reported about overall survival among the included studies.

Secondary outcomes
Secondary outcomes were adverse events, including major bleeding, changes in parameters of RV strain, and all-cause mortality.
Six studies reported results on recurrent PE suggesting benefit from thrombolysis, but this difference was not statistically significant, RR 0.56 (0.23 to 1.37) I 2 = 0% ( Figure 3C). Only two studies included outcomes on risk of stroke and found no advantage of thrombolysis over anticoagulation-only intervention, RR 2.22 (0.17 to 28.73) I 2 = 56% ( Figure 3D).

Analysis of the impact different therapies had in RV compromise
was not feasible, since there was substantial heterogeneity in the variables used among studies (RV wall motion, septum dynamics, right/left ventricle end-diastolic dimension ratio) and in the methods used for evaluation (electrocardiogram, echocardiogram, pulmonary artery hemodynamics, markers of cardiac injury).
All-cause mortality was evaluated in all studies, favoring the group that received thrombolysis, yet difference was not statistically significant, RR 0.68 (0.40 to 1.16) I 2 = 0% ( Figure 3E).

Publication bias
We did not find any publication bias (Supplementary files 1 and 2).

Sensitivity analysis
When excluding the study by Meyer et al [27] for the outcome major bleeding, we found a RR of 1.33 95% CI (0.58 to 3.05). There were no differences between the two interventions; therefore, Meyer et al.
influenced on the final effect.
Regarding the outcome overall bleeding, we found a RR of 2.37 95% CI (1.06 to 5.29). Therefore, there was no change compared with the effect found in the general analysis.

Subgroup analysis
Analysis of results based on thrombolytic regimen used showed that the risk of overall and major bleeding is higher with thrombolysis, when compared to anticoagulation alone. However, this is statistically significant for tenecteplase, but not for alteplase. Regarding recurrent PE and all-cause mortality, there was no difference between thrombolysis and anticoagulation alone, regardless of thrombolytic chosen (Table 2). Four studies did not specify which specific thrombolytic agent was used; therefore, they were not included in the subgroup analysis [21,26,28,29].

Summary of the main results
Review of trials comparing thrombolytic therapy versus anticoagula-

Contrast with the literature
The net benefit of thrombolytic strategies in the management of intermediate risk PE continues to be an area of active debate. Based on the results of our meta-analysis, we agree with other authors that the risk of bleeding is higher in patients who receive thrombolysis [30][31][32][33][34][35][36][37].
However, our study did not show a benefit on all-cause mortality, risk of stroke, and recurrence of PE, when thrombolysis was added to anticoagulation.
The PEITHO trial, the largest study that has included patients with intermediate risk PE, showed a clear advantage of thrombolysis in hemodynamic status and need of life-sustaining measures, yet no difference in 30-day mortality was found [27]. Although other metaanalysis have reported an improvement in mortality in the intervention group [32,35], we believe these results are driven by the composite primary outcome of the PEITHO trial, which includes early death and hemodynamic decompensation; this observation was also made by another group of authors [36]. A most recent analysis of multiple systematic reviews also reported a reduction in mortality secondary to thrombolysis, but recognizes that results are inconsistent, the large use of antinomies, and the questionable quality of some of the studies included [38].  [30,33] and after long-term follow-up [39]. This emphasizes that PE patient population is heterogeneous, and several factors should be taken into account when choosing the optimal therapeutic regimen for every patient. Data seem to be consistent regarding the particular increased risk of bleeding in older patients, but a precise age cutoff has not been established [27,31], and risk stratification tools have not been validated and are not routinely used [40].
Optimal thrombolytic agent, dose, and route of administration continue to be a matter of debate; since historically, there is no consistency in treatment regimens used among different studies in PE patients. Authors have described a higher risk of bleeding among different thrombolytic agents, such as tenecteplase [30], which could be related to its higher potency. Our subgroup analysis corroborates this finding, showing a statistically significant higher risk of bleeding when tenecteplase is compared with anticoagulation alone; similarly, subgroup of patients who received alteplase seemed to have higher risk of bleeding, but the difference was not significant.
Safety of thrombolytics has been studied in more detail in patients with myocardial infarction (MI) and stroke, and evidence seems to differ from what studies have shown in PE patients. In acute ST segment elevation MI, tenecteplase has shown similar efficacy and safety profile compared to alteplase, without an increase in bleeding risk [41].
One group even showed that tenecteplase was associated with fewer noncerebral bleeding complications [42]. In stroke trials, clinical outcomes have also been similar between two groups, and tenecteplase has not shown increased risk of bleeding [43][44][45][46][47].   Our meta-analysis has limitations. Definitions of outcomes, particularly major bleeding, and follow-up times were not consistent among authors. Likewise, thrombolytic agents and doses used are different.

Strengths and limitations
Therefore, conclusions in regard to comparison of outcomes related to specific thrombolytic agents, dosage, and protocol of administration were not possible.
Other limitations are related to the high risk of bias associated with the unclear blinding of participants and personnel, incomplete outcome data, and allocation concealment issues of the studies included in the meta-analysis.
Although the primary objective of this study was to evaluate overall survival, we did not find any data in this specific subgroup of patients with intermediate risk PE. This is one of multiple questions that remain unanswered and requires further studies.

CONCLUSIONS
In

AUTHORS CONTRIBUTION
All authors contributed substantially to the study design, data analysis and interpretation, and the writing of the manuscript.

AVAILABILITY OF DATA AND MATERIAL
Data collected for this study will be available upon request.