• 
• 

Three Part Question

In [patients with confirmed non-massive pulmonary embolism], does [high sensitivity cardiac troponin (hs-TnT or hs-TnI) perform better than brain natriuretic peptide (NT-proBNP or BNP)] to [identify patients at risk of death or serious adverse events].

Clinical Scenario

You are taking care of a 36 year-old female with confirmed pulmonary embolism (PE) on CT angiography. She complains of a moderate pleuretic chest pain and is slightly short of breath. Her vitals signs and her ECG are normal. She is hemodynamically stable since her arrival in the ED. You know that right ventricular strain in patients with PE is associated with significant mortality and morbidity and that high-sensitivity troponin and BNP could help identify these patients. While planning to potentially discharge the patient after appropriate PE treatment, you wonder if high sensitivity troponin would better perform than BNP to identify individuals with high-risk of complications or serious adverse events.

Search Strategy

The search was performed on January 8th and 9th, 2016.

A. No Best BETs comparing BNP and troponin in pulmonary embolism was found. Although, one Best BETs already addressed prognostic value of BNP in pulmonary embolism.

B. Cochrane database of systematic reviews
['(natriuretic peptide) OR troponin] AND (pulmonary embolism) in Title, Abstract, Keywords in Cochrane Reviews'
No relevant Cochrane review was found on this subject.

C. www.clinicaltrials.org was searched for an ongoing trial on this topic. No relevant trial was found.

D. MEDLINE (trough Pubmed)

1) natriuretic peptide[MeSH Terms] Filters: Humans : 16888
2) troponin[MeSH Terms] Filters: Humans : 9363
3) pulmonary embolism[MeSH Terms] Filters: Humans : 31152
4) 1+2+3 : (("natriuretic peptides"[MeSH Terms] AND "humans"[MeSH Terms]) AND ("troponin"[MeSH Terms] AND "humans"[MeSH Terms])) AND ("pulmonary embolism"[MeSH Terms] AND "humans"[MeSH Terms]) AND "humans"[MeSH Terms] : 66


E. EMBASE

1) 'brain natriuretic peptide derivative'/exp AND 'troponin'/exp AND 'lung embolism'/exp :354
2) #1 AND 'human'/de AND [adult]/lim :91

Search Outcome

Of the 66 articles found on Medline, 13 were potentially relevant. Of these 13 papers, one was excluded because it was written in chinese.

Of the 91 papers found on Embase, 10 articles were potentially relevant. After cross-referencing, 7 of these articles were duplicates of those found on MEDLINE. One was not available in his full text version. The other 2 articles were considered potentially relevant.

References of these 14 articles were screened for relevant papers and no new study was found. Of the 14 potentially relevant articles, 2 were kept for final analysis as they answer the clinical question under study. They are presented below. Other papers were excluded mainly because they did not assess high-sensitivity troponin or compare both biomarkers.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Lankeit M. et al. 2010 Germany	156 consecutive normotensive patients with confirmed PE	Prospective observational study	Pronostic value of hs-TnT for adverse 30-day outcome (≥ 1: death, need for cathecolamines, endotracheal intubation, or CPR).	hs-TnT < 14 pg/mL and NT-proBNP< 1000 pg/mL had 100% pronostic sensitivity and negative predictive value (NPV) for the absence of adverse event. hs-TnT ≥ 14 pg/mL on admission was the only biomarker significantly associated with long term mortality ( HR 3.57 (95% CI:1.05-12.14, p=0.041).	Low number of adverse events in the acute phase of the study. Study was not designed and powered to compare hs-TnT and NT-proBNP head-to-head and thus detecting superiority.
Vuilleumier N. et al. 2014 Switzerland	230 patients aged ≥ 65 years with normotensive PE	Prospective observational study	Pronostic value of NT-proBNP (cut-off value of 300 pg/mL), hs-TnT (cut-off value of 14 pg/mL) and clinical scores, for a composite endpoint of PE-related complications at 30 days defined as PE-related death, recurrent venous thromboembolism or major bleed.	hs-Tnt had the highest prognostic accuracy (area under the ROC curve 0.75, CI 0.63-0.86, p<0.001) compared to NT-proBNP (0.69 CI 0.58-0.79, p=0.003 NPV for hs-Tnt vs NT-proBNP were similar (95.9, (CI 90.0-98.4) vs 98.6 (CI 92.6-99.8)).	Patients aged of ≥ 65 years, decreasing external validity. Non-consecutive recrutement because of missing data on biomarker values. Small number of adverse events. Study was not designed and powered to compare hs-TnT and NT-proBNP head-to-head and thus detecting superiority.

Comment(s)

We found 2 cohort studies evaluating prognostic accuracy of hs-TnT and NT-proBNP in normotensive PE. Cut-off values were different for NT-proBNP between those studies. None was powered and designed to evaluate head-to-head hs-TnT and NT-proBNP. Both cardiac biomarkers have a very good NPV, thus correctly identifying low-risk patients. However, high sensitivity cardiac troponin over 14 pg/mL seems to better predict long term mortality. When combined to echocardiographic findings of right ventricular dysfunction, hs-TnT ≥ 14 pg/mL (OR 11.87, CI: 1.42-99.56) and NT-proBNP ≥ 1000 pg/mL (OR 17.8, CI: 2.11-150.01) were associated with an increased risk of complications (Lankeit et al. 2010). Abbreviations: hs-TnT, high-sensitivity cardiac troponins T; hs-TnI, high-sensitivity cardiac troponins I; NT-proBNP, N-terminal pro-brain natriuretic peptide; BNP, brain natriuretic peptide; CPR, cardiopulmonary resuscitation; HR, hazard ratio; CI, Confidence interval (95% unless specified); ROC, receiver operating curve; NPV, negative predictive value; OR, odd ratio.

Clinical Bottom Line

In patients with confirmed non-massive pulmonary embolism, high sensitivity cardiac troponin seems to perform slightly better than brain natriuretic peptide (NT-proBNP) to accurately identify patients at risk of complications. However, no study was appropriately powered to compare these two biomarkers head-to-head. While waiting for further studies, prognostic values of these biomarkers should be considered with care and combined with other findings such as right ventricular function on echocardiography to identify patients at low risk of complications that are securely dischargeable.

References

1. Lankeit M, Friesen D, Aschoff J, et al. Highly sensitive troponin T assay in normotensive patients with acute pulmonary embolism European heart journal 2010;31(15):1836-44
2. Vuilleumier N, Limacher A, Mean M, et al. Cardiac biomarkers and clinical scores for risk stratification in elderly patients with non-high-risk pulmonary embolism Journal of internal medicine 2015 Jun;277(6):707-16