Original Contribution
Prognostic Value of Right Ventricular Dysfunction Markers for Serious Adverse Events in Acute Normotensive Pulmonary Embolism

https://doi.org/10.1016/j.jemermed.2016.09.002Get rights and content

Abstract

Background

Right ventricular dysfunction (RVD) in pulmonary embolism (PE) has been associated with increased morbidity. Tools for RVD identification are not well defined. The prognostic value of RVD markers to predict serious adverse events (SAE) during hospitalization is unclear.

Objective

Prospectively compare the incidence of SAE in normotensive emergency department patients with PE based upon RVD by goal-directed echocardiography (GDE), cardiac biomarkers, and right-to-left ventricle ratio by computed tomography (CT). Simplified Pulmonary Embolism Severity Index (sPESI) was calculated. Deaths and readmissions within 30 days were recorded.

Methods

Consecutive normotensive PE patients underwent GDE focused on RVD (RV enlargement, hypokinesis, or septal bowing), serum troponin, and brain natriuretic peptide (BNP), and evaluation of the CT ventricle ratio. In-hospital SAE and complications within 30 days were recorded.

Results

We enrolled 123 normotensive PE patients (median age 59 years, 49% female). Twenty-six of 123 (26%) patients had one or more SAE. RVD was detected in 26% by GDE, in 39% by biomarkers, and in 38% with CT. In-hospital SAE included one death, six respiratory interventions, six dysrhythmias, three major bleeding episodes, and 21 hypotension episodes. Forty-one percent of patients RVD positive by GDE had SAE, compared to the 18% RVD negative by GDE. Odds ratios for GDE, CT, BNP, troponin, and sPESI for SAE were 3.2 (95% confidence interval [CI] 1.2–8.5), 2.0 (95% CI 0.8–5.1), 3.3 (95% CI 1.3–8.6), 4.2 (95% CI 1.4–13.5), and 2.9 (95% CI 1.1–8.3), respectively. Five patients had non-PE-related deaths within 30 days.

Conclusion

The incidence of SAE within days of PE was significant in our cohort. Those with RVD had an increased risk of nonmortality SAE.

Introduction

The ability to diagnose, risk stratify, and initiate treatment in patients with acute pulmonary embolism (PE) is crucial for emergency physicians. One half of all PE cases in the United States are diagnosed in the emergency department (ED), and acute PE has a higher 3-month mortality rate than myocardial infarction (1). Patients with PE and overt signs and symptoms of hypotension and shock have “massive PE” and are at high risk for death and morbidity. The majority of PE cases, however, present without hypotension, and reported morbidity and mortality rates in normotensive PE cohorts vary. Clinical deterioration and death from PE usually involve right ventricular dysfunction (RVD). Occlusion of the pulmonary arteries may lead to increased RV afterload. Increased RV pressures can cause dilatation of the thin-walled RV, leading to RV wall strain, impaired RV systolic function, septal deviation to the left, and thus, impaired left ventricular (LV) filling and cardiac output. RV strain can deteriorate into RV failure. Signs of RVD are detected directly using echocardiography and computed tomography (CT), and indirectly with serum cardiac biomarkers.

Several studies have demonstrated that normotensive PE patients with RVD, determined by any modality, may have an increased risk for adverse events 2, 3, 4, 5, 6, 7, 8, 9. Conversely, a study by Beam et al. showed that PE patients with normal vital signs, no signs of RVD, and other low-risk criteria could be immediately discharged from the ED on oral anticoagulation without a significantly increased risk of major bleeding or recurrent venous thromboembolism (10). RVD, identified by echocardiogram or biomarkers, is not currently part of any of the low-risk clinical scoring criteria.

Previous reports on the prognostic performance of markers of RVD have used varying cutoff values for cardiac biomarkers, comprehensive echocardiography, and CT criteria 2, 3, 4. Cardiac biomarkers have diagnostic accuracy limitations for detection of PE-provoked RVD. Serum troponin is specific for ischemia and infarction but is not sensitive for RVD. Brain natriuretic peptide (BNP) is not sensitive or specific for RVD. The CT ventricle dimension ratio can evaluate for RV dilatation, but does not evaluate RV systolic function and has not been validated in acute PE patients without hemodynamic compromise. Comprehensive echocardiography evaluates both RV dilatation and RV systolic function but is not usually immediately available in EDs. The echocardiographic RVD criteria often involve quantitative assessments utilizing Doppler technology. Previous studies have used different thresholds and methods to determine RVD and were conducted in clinically diverse populations.

Most studies looking at the primary clinical outcomes of PE have focused on recurrence of venous thromboembolism, bleeding complications, and death over the course of months. However, the incidence of in-hospital serious adverse events (SAEs) during hospitalization is not often reported 11, 12. Other studies report that morbidity and mortality from PE are on the decline (13). However, a study by Grifoni et al. reported that 30% of normotensive PE patients had RVD, with a 10% rate of subsequent PE-related shock and 5% in-hospital mortality (6). Investigations into the prognostic significance of RVD in acute PE have used different short-term and long-term morbidity, mortality, and functional outcomes. Overall, PE-related clinical outcome reports do not usually focus on the emergency physician's concern for early clinical deterioration, or further interventions beyond anticoagulation during hospitalization.

Early risk stratification of PE patients is important because consideration for advanced PE therapy should ideally take place prior to potential hemodynamic deterioration. Patients can be assigned to higher levels of monitoring if deemed to be at higher risk for clinical deterioration. We performed an a priori specified secondary analysis of a study of normotensive ED patients with acute PE comparing emergency physician-performed goal-directed echocardiography to comprehensive echocardiography for the detection of RVD. In this portion of the study, our goal was to contemporaneously determine the prognostic value of predefined RVD tests for development of SAE in ED patients with PE presenting without hemodynamic compromise based on emergency physician-performed echo. This study's hypothesis was that normotensive PE patients with signs of RVD would have an increased incidence of SAEs during hospitalization.

Section snippets

Study Design and Setting

This was an a priori specified secondary analysis of a study comparing emergency physician-performed goal-directed echocardiography to comprehensive echocardiography for the detection of RVD in consecutive normotensive ED patients diagnosed with PE by either computed tomography pulmonary angiogram (CTPA) or ventilation-perfusion (V/Q) lung scan (14). This study was conducted between July 2014 and August 2015 at an academic medical center with 118,000 annual ED visits. Our hospital's

Enrollment and Patient Characteristics

From June 2014 to August 2015, 135 patients with acute diagnosis of PE were identified and approached by investigators. Twelve patients were later excluded; 4 were excluded because no new clot burden was detected on CT upon further review by radiologists, 4 were hypotensive at presentation in the ED, 3 had unobtainable echocardiographic images, and 1 patient refused consent for follow-up. See Figure 2 for Standards for Reporting Diagnostic Accuracy diagram representation of included patients. A

Discussion

We prospectively evaluated normotensive PE patients for signs of RVD. There was a 26% prevalence of SAEs during hospitalization in our cohort. PE patients with RVD had a twofold increase in SAEs by any of our predefined RVD-based criteria. Over 80% of patients without RVD by either GDE, cardiac biomarker, or CT ratio did not have SAEs. The in-hospital mortality of our PE cohort was <1%, and this single death was in a patient without RVD. Two patients at low risk by sPESI criteria required

Conclusion

The incidence of nonmortality clinical deterioration within days of PE diagnosis was significant in our cohort of normotensive PE patients. The presence of RVD in normotensive PE patients was associated with an increased risk for short-term clinical deterioration or need for hospital-based interventions, and having this information in the ED may guide emergency physician decisions on patient disposition, level of monitoring, and consideration for reperfusion therapy.

Article Summary

1. Why is this article important?

  1. Few studies report on

References (36)

  • S. Dresden et al.

    Right ventricular dilatation on bedside echocardiography performed by emergency physicians aids in the diagnosis of pulmonary embolism

    Ann Emerg Med

    (2014)
  • R.A. Taylor et al.

    Accuracy of emergency physician-performed limited echocardiography for right ventricular strain

    Am J Emerg Med

    (2014)
  • D. Aujesky et al.

    Outpatient versus inpatient treatment for patients with acute pulmonary embolism: an international, open-label, randomised, non-inferiority trial

    Lancet

    (2011)
  • W. Zondag et al.

    Hestia criteria can safely select patients with pulmonary embolism for outpatient treatment irrespective of right ventricular function

    J Thromb Haemost

    (2013)
  • W. Zondag et al.

    Outpatient treatment in patients with acute pulmonary embolism: the Hestia Study

    J Thromb Haemost

    (2011)
  • J.A. Kline et al.

    Prospective evaluation of right ventricular function and functional status 6 months after acute submassive pulmonary embolism: frequency of persistent or subsequent elevation in estimated pulmonary artery pressure

    Chest

    (2009)
  • J.H. Cho et al.

    Right ventricular dysfunction as an echocardiographic prognostic factor in hemodynamically stable patients with acute pulmonary embolism: a meta-analysis

    BMC Cardiovasc Disord

    (2014)
  • G. Coutance et al.

    The prognostic value of markers of right ventricular dysfunction in pulmonary embolism: a meta-analysis

    Crit Care

    (2011)
  • Cited by (24)

    • Investigation of TEMPO partitioning in different skin models as measured by EPR spectroscopy – Insight into the stratum corneum

      2020, Journal of Magnetic Resonance
      Citation Excerpt :

      They are composed of a nitroxide group (N-O⋅) which is stabilized by different substituents (methyl, ethyl or spiro-cyclohexyl) at α-positions [2–4]. Cyclic nitroxide radicals with a five- (pyrrolidine) or six-membered (piperidine) ring are used in particular for EPR investigations in biological systems [2,4]. Due to their diverse physicochemical properties, nitroxide radicals can provide information on different microenvironmental properties like pH, temperature, viscosity and polarity [4].

    • Resuscitative Cardiopulmonary Ultrasound and Transesophageal Echocardiography in the Emergency Department

      2019, Emergency Medicine Clinics of North America
      Citation Excerpt :

      RV function can be evaluated with qualitative or quantitative assessments. RV dilation and septal kinetics are the main qualitative screening tools that can be easily and reliably performed by ED physicians.14–16 Weekes and colleagues14 prospectively demonstrated 100% sensitivity and 99% specificity for identifying RV dysfunction in normotensive patients with PE, and that FOCUS had better diagnostic accuracy identifying RV dysfunction than troponin or brain natriuretic peptide alone.

    View all citing articles on Scopus
    View full text