ABSTRACT & COMMENTARY
Echo in Acute Pulmonary Embolism
By Michael H. Crawford, MD, Editor
Pruszycyk P, et al. Prognostic value of echocardiography in normotensive patients with acute pulmonary embolism. J Am Coll Cardiol Img 2014;7:553-560.
It is believed that right ventricular (RV) performance in acute pulmonary embolism patients is of prognostic value, but specific RV function parameters are not agreed upon and there are little outcome data in this area. Thus, these investigators from Poland studied 411 consecutive patients with symptomatic acute pulmonary embolism (APE) who were hemodynamically stable on admission (systemic systolic blood pressure > 90 mm Hg). Echocardiograms were done as soon as possible after admission (immediately in 193, within 24 hours in 159, and in < 72 hours in 59). RV dysfunction was defined as: 1) RV free wall hypokinesis and an RV/LV diastolic diameter ratio of >0.9; or 2) an elevated systolic pressure gradient across the tricuspid valve of > 30 mm Hg and a pulmonary flow velocity acceleration time of < 80 msec. Patients without these findings were considered low risk. The primary endpoint was a combination of 30-day APE-related mortality or rescue thrombolysis in patients with cardiac arrest or shock.
Fifty-nine percent of the patients had RV dysfunction (sub-massive APE); the rest were low risk. Only nine in the sub-massive group received thrombolysis and seven of them survived. The primary endpoint was observed in 21 patients (5%). These patients had higher heart rates, troponins, and brain natriuretic peptide values. Many of the RV functional parameters on echocardiogram were significantly different in those who exhibited the primary endpoint vs those who didn’t. Also, these patients had lower LV ejection fractions as well. However, multivariate analysis showed that tricuspid annulus plane systolic excursion (TAPSE) was the only independent predictor of the primary endpoint (hazard ratio [HR], 0.64; 95% confidence interval [CI], 0.54-0.7; P < 0.0001). A TAPSE < 1.6 cm had a HR of 27.9 (95% CI, 6.2-124.6; P < 0.0001), a positive predictive value of 21%, and a negative predictive value of 99%. Also, the receiver operating curve (ROC) area for TAPSE was the highest of any parameter (0.9 at < 1.6 cm). The authors concluded that TAPSE is superior to other measures of RV and LV performance for predicting 30-day mortality or rescue thrombolysis.
COMMENTARY
Initially, normotensive patients with APE usually do well on anticoagulant therapy. However, there is a reluctance to discharge them from the hospital because some deteriorate and APE is commonly found at autopsy in hospitalized patients. It is known that RV dysfunction is associated with a worse outcome, but there is little agreement on which parameters are the most useful in risk prediction. If a low-risk group of APE patients could be identified, perhaps they could be discharged from the hospital earlier. Obviously, hypotensive patients would not be candidates for early discharge, so this group from Warsaw, Poland, studied 411 patients with APE who were normotensive on admission and performed echocardiography on them as soon as feasible (86% in 24 hours and all by 72 hours). They analyzed 15 RV and LV performance parameters to determine their ability to predict mortality or need for rescue thrombolysis. There was one clear winner, TAPSE. It was the only independent predictor by multivariate analysis and it had the highest ROC area of any parameter (0.9 at < 1.6 cm). At a TAPSE < 1.6, the HR for mortality or rescue thrombolysis was an astonishing 28. The nearest competitor was RV/LV diameter in the four-chamber view at HR = 7.3 and a ROC area of 0.6. Other findings, such as McConnell’s sign, had much lower HRs. However, the positive predictive value of TAPSE was only 21% at < 1.6 cm with a negative predictive value of 99%. In fact, at a TAPSE > 2 cm, there were no complications from APE with a negative predictive value 100%. Thus, TAPSE is better at defining the low-risk group among normotensive patients with APE.
Could we use TAPSE to decrease length of stay in APE? In this study, about half the patients with normotensive APE had no RV abnormalities on echo and no events, so perhaps they could go home earlier. In those with some signs of RV dysfunction, if the TAPSE was > 2.0 cm, perhaps they could go early as well. Of course, this would have to be tested in a large clinical trial. In those with a TAPSE < 1.6, perhaps earlier consideration of thrombolytic therapy is reasonable; of course, any clinical decisions would have to consider other factors such as comorbidities, bleeding risk, etc.
Finally, the study has significant limitations as the authors acknowledge. It is an observational study from one center and the causes of death were not adjudicated. Also, there was considerable delay in obtaining echocardiograms in about 15% (24-72 hours). In addition, there was a low incidence of events in this subgroup of APE; 5% for the primary combined endpoint and 3.4% mortality. Thus, extended hospital stays for any APE patient who does well initially may not make much sense.
