EM – Critical Care Journal Club

Topic: Thrombolysis for sub-massive pulmonary embolism (PE)

Article: Konstantinides, S.V., et. al. Impact of Thrombolytic Therapy on the Long-Term Outcome of Intermediate-Risk Pulmonary Embolism. Journal of the American College of Cardiology. 2017. Volume 69, No. 12: 1536-1544.    

Resident Reviewer: Adam Weightman, MD, MBA

 

Case / Problem

While several previous studies have examined short- and even medium-term outcomes of thrombolysis for intermediate-risk PE, there has not been a robust study investigating long-term effects of this controversial intervention. Much of the support for use of thrombolytics in this subgroup of patients is the avoidance of theoretical long-term patient-centered complications, such as chronic thromboembolic pulmonary hypertension (CTEPH) or persistent debilitating symptoms.

 

How was article selected?

This article is the first to prospectively investigate the risk of long-term effects of thrombolysis in a large, randomized patient population with intermediate-risk PE. It is not only potentially practice-changing, but also “hot off the presses,” being published only about 6 months ago. It is important that we are familiar with emerging data in this area of study given the controversy surrounding management intermediate-risk PE .

 

Study Description and Research Question

Multicenter, double-blind, placebo-controlled randomized trial (intention-to-treat analysis)

to answer the question:

Does the use of tenecteplase for intermediate-risk PE increase long-term survival and/or prevent late sequelae of PE

 

Relevance

Acute PE is a common clinical entity with >100,000 cases annually in the US alone. Despite this high incidence, there is still debate regarding the appropriate treatment for acute PE, particularly in intermediate-risk patients. Previous studies demonstrated an improvement in all-cause mortality or hemodynamic decompensation within 7 days for patients receiving tenecteplase, as well as a higher incidence of hemorrhagic stroke and major non-intracranial bleeding. It has been theorized that rapidly and effectively reducing thrombotic burden may minimize the risk of persistent pulmonary hypertension and perhaps right heart failure, however, the evidence to substantiate this hypothesis is limited.

 

PICO

Population: 709 of 1,005 patients at 28 of the original 76 PEITHO study sites who were 18 years or older with objectively confirmed acute PE with first symptoms 15 days or less before randomization, RV dysfunction confirmed by echocardiography or spiral CT of the chest, and myocardial injury confirmed by a positive troponin I or T test result.

Intervention: Single weight-based IV bolus (administered over 5-10 seconds) of tenecteplase + unfractionated heparin (see below)

Comparison: Single IV bolus of the same volume and appearance + unfractionated heparin (as intravenous bolus followed by infusion at a rate adjusted to achieve and maintain an aPTT of 2.0-2.5x that of control)

Outcomes: Vital, clinical and hemodynamic status 24 months or later after randomization; included mortality rates, clinical and echocardiographic parameters

 

Appraisal of Internal Validity

For a complete evaluation of the internal validity of the initial PEITHO trial, please see Emily Zametkin’s review. This review will evaluate the relevant aspects of the follow-up study.

  • Did the experimental and control groups starts out with a similar prognosis? YES
    • No statistically significant differences in baseline parameters between intervention and control groups
    • Moreover, baseline parameters between patients followed over the long-term and those “not followed” were not statistically significantly different, with the exception of body weight and history of previous VTE
  • What outcome measures were used? Well-defined? Replicable? YES (except clinical status)
    • Mortality/cause of death
    • Clinical status (unclear exactly what measures they evaluated)
    • Echocardiographic parameters (RV end-diastolic diameter >30mm, RV:LV end-diastolic diameter >0.9, hypokinesia of the RV free wall, TAPSE reduced, tricuspid systolic velocity >2.6 m/s, systolic PA pressure)
  • How complete was follow up? Was it sufficient? How many dropouts? POOR for most measures
    • 709/1,005 (70.4%) of original PEITHO patients studied were followed long-term based on their site of randomization (only 28/76 sites participated)
      • Unlikely to introduce selection bias, as randomization was stratified by center and performed in blocks within centers to ensure balanced distribution of treatment groups
    • 353/359 (98.3%) patients in tenecteplase arm and 343/350 (98%) in placebo arm had survival rate and cause of death assessed, <2% unplanned loss to follow-up for mortality (for ~10% of patients, cause of death was unable to be determined)
    • 175/359 (48.7%) patients in tenecteplase arm and 183/350 (52.3%) patients in placebo arm had clinical status assessed
    • 144/359 (40.1%) patients in tenecteplase arm and 146/350 (41.7%) patients in placebo arm had echocardiography performed
  • Was outcome assessment blind? UNSURE
  • Was follow up time sufficiently long to detect important effects on outcomes of interest? YES

 

Given high rates of lack of follow-up, particularly for clinical status and echocardiographic parameters, there is a real risk of bias in the results of this study. Because of this, the internal validity is likely fair to poor for most measured outcomes, as the authors address.

 

Primary Results

  • Overall mortality rate was 19.2% in the followed patients with a median follow-up time of 37.8 months
  • No significant mortality differences were observed between patients who underwent thrombolysis with tenecteplase and patients randomized to heparin alone (20.3% tenecteplase, 18.0% placebo, p=0.43)
  • Most late deaths (between day 30 and long-term follow-up) resulted from cancer, acute or chronic respiratory failure, or other illnesses (mostly acute infections or chronic systemic inflammatory diseases)
  • No significant difference in persistence of clinical symptoms between groups (36% tenecteplase, 30.1% placebo, p=0.23), overall prevalence 33%
  • The leading persistent symptom was “mild exertional dyspnea”
  • 12% of patients randomized to tenecteplase and 10.9% of patients randomized to placebo were in NYHA functional class III or IV
  • <2% of patients in each treatment arm had symptoms or clinical signs indicating heart failure
  • Of patients who underwent echocardiography, no significant differences regarding the presence of residual pulmonary hypertension or RV dysfunction between groups (44.1% tenecteplase, 36.6% placebo, p=0.20), overall 44% of those screened demonstrated indicators of pulmonary hypertension or RV dysfunction
  • No significant difference in definitive diagnosis of CTEPH between groups (2.1% tenecteplase, 3.2% placebo, p=0.79), nor low or intermediate ultrasound-based probability of CTEPH (85% tenecteplase, 96% placebo)

 

Of note, because “clinical and echocardiographic examinations could not be performed in all survivors…the findings of this study cannot be considered definitive evidence of the frequency of persistence of pulmonary hypertension or RV dysfunction and cannot ascertain the possible impact of thrombolysis on the long-term hemodynamic course of these patients.” The authors also caution against using this study to resolve the debate on the true incidence of CTEPH or the role of thrombolysis in preventing this late complication.

 

External Validity

The results of this study are likely generalizable to our population based on the fact that the original study enrolled participants from several clinical sites across 13 European countries. It is unclear exactly what type of hospital (e.g. teaching vs. community, inner-city vs. suburban) each site was classified as in the original study, or which sites participated in the follow-up study. Thus, there is a risk that these results may not be applicable to our patients, who tend to be underserved and may have a higher burden of underlying and untreated disease (among other potential differences).

 

Utility to Practice

Consider early anticoagulation and close monitoring (with rescue reperfusion for cases of hemodynamic decompensation) in patients with intermediate-risk PE. Be on the lookout for forthcoming studies that more robustly assess long-term clinical, hemodynamic, and functional outcomes.

 

Resident Reviewer: Adam Weightman, MD, MBA
Faculty Review: Alex Sheng, MD