Will Denq – Urgent Matters

Will Denq, MD & Evan Kuhl, MD

May 31, 2017

In 2016, an interesting and potentially practice-changing articles was published: the PESIT trial by Prandoni et al.. The prospective study utilized 11 Italian hospitals and evaluated all patients for pulmonary embolism (PE) who were admitted to the hospital for first-time syncope. The results made for headlines; pulmonary emboli were found in 1 in 6 patients, and was broadcast by major news outlets including NPR and Medscape. With the headlines came a debate amongst physicians about just how useful--if not harmful this study might be.

Prior to the PESIT trial, there was scant data on the prevalence of PE in patients admitted for syncope. The authors created a multicenter, cross-sectional study with 11 hospitals in Italy to evaluate the frequency of PEs in patients admitted for syncope. Each patient was risk stratified with a Simplified Wells Score (see Table 1) Those with a score greater than 4 or with a positive D-dimer underwent CT angiography or V/Q scan to evaluate for pulmonary emboli.

Of 2,584 patients who were evaluated by the Emergency Department for syncope, 560 patients were included in the study. Of these, 230 patients received CTA or V/Q scans, and 97 patients were found to have PEs.

The Objective Results:

2584 patients assessed
1. 560 included (Mean age = 76 )
2. 1867 patients discharged and therefore excluded
3. 157 inpatients further excluded
  1. Prior episodes of syncope, anticoagulated, pregnant excluded
97/560 (17.3%) were found to have PE
1. 230/560 were high risk -> CTA or V/Q scan
2. 180 had CTA, 1 had autopsy
  1. 73/180 had PE
    1. 49/73 (67%) had a main pulmonary artery or lobar artery PE
    2. 19/73 (26%) had segmental PE
    3. 5/73 (7%) had subsegmental PE
3. 49 had V/Q
330/560 were low risk -> no imaging

The result of this study carries a bold statement. 1 in 6 of patients hospitalized for syncope have a PE, therefore perhaps we should consider testing syncope patients for PE even when we think the symptoms are caused by something else given the high prevalence of disease. However, without reading the paper in detail may result in lowering the threshold to test for PE and potentially an overutilization of D-dimer, CTA, and/or V/Q scans. Below,we present the pros and cons of this paper to better inform whether to consider PE in your next syncope patient.

Practice Changing:

Bottom Line: The PESIT trial is a well designed study that should increase our consideration of PE during syncope workups, especially in the geriatric population. When a pulmonary embolism causes syncope, a missed diagnosis may cause significant complications and patient safety issues.

Although we are getting better at catching PEs, they can often be subtle and difficult to diagnose. If a casual reader were to browse through the abstract or to base their judgement on a tertiary news source, they would miss the true value behind this study. It actually calls for a closer examination of PE in older patients that are admitted for first-time syncope, not for testing everyone. Those who are low risk and considered safe for discharge were not included in this study, and this study should not impact our practice in these patients. Those at high enough risk to warrant admission may need a second look. In those with prior PEs, undetermined cause of syncope, or clinical features of PE (respiratory distress, tachycardia, hypotension, or signs of DVT) we should have an increased suspicion of pulmonary embolism.

This conclusion results from the question of, how exactly do we translate an inpatient study and apply it to the ED setting? What population of syncope seen in the ED do we even consider PE in? Based on the population you see and ED you work in, admission criteria for syncope is likely to differ. For appropriate reasons, there wasn’t a standard for syncope admission in this study. However, if we refer to Table 1 of the article, a useful data point is the age of the study population. Not so surprisingly, the mean and median age of the study population was 76 and 80, respectively. It makes sense. Geriatric patients tend to have higher admission rates since they have more comorbidities, less reserve, and a larger number of possible etiologies for their symptoms.

Although there is a different discussion to be had about the causality, it is still standard of care to treat PE even if it exists as a non-etiologic diagnosis for their syncope admission. 67% of patients with a CT positive for PE had a significant embolus that could be related to the syncope they experienced. Only 7% of CT-PE positive patients had subsegmental PEs. The best we can take from this data is to consider PE as an association in the geriatric population that presents as first-time syncope.

Not so fast:

Bottom Line: This inpatient study has the dangerous implication of resource overutilization and the results are difficult to translate to other countries and the ED.

There are radiation and potential contrast risks, workflow, and healthcare costs to consider especially if patients or providers do not understand the implications behind this study. The study does not imply that 1 in 6 patients syncopized because of a PE. It just reports that of the patients admitted for syncope, 1 in 6 were found to have a PE. Association, but not causality is what this is. 12.7% of those found to have pulmonary emboli had already been given an alternative diagnosis. Many of these patients may have presented to the ED with an emboli, and others may have been asymptomatic as the clinical significance of these findings is not known.

While the authors focus on the number of patients admitted, they do not account for the 2584 patients screened in the ED. Additionally, many of the patients found to have PEs showed clear evidence of DVT and would have a high Well’s score, which should be identified in the ED prior to admission. When all 2584 patients are taken into consideration, you see a much more typical rate of PEs that many of us are more used to seeing, near 5-10%. It is difficult to translate the algorithm and results of the study across the world from Italy to the US. Do we also discharge 72% of our first time syncope patients? Do we D-dimer all of our syncope patients? The clear answer is no. Despite the authors’ claim that the “standardized protocol” for the diagnostic workup of syncope was based on international guidelines - there is no mention in either guideline about the use of D-dimer or even Well’s in the initial consideration for syncope (1,2).

Summary:

Bottom Line: This study is not practice-changing, but it should increase our awareness for the prevalence of PE in the geriatric population that presents with first-time syncope. If you decide to change your practice, consider the downstream consequences.

There may be some benefit in considering PE in the differential in the admitted geriatric patient with first-time syncope. However, the guidelines (1,2) from both cardiology societies in the United States and Europe do not even discuss the use of D-dimer or Well’s criteria in the initial workup for syncope. Although these tools are an acceptable method for ruling out PE, PE may not even be on the differential for a first-time syncope patient.

This study potentially opens a can of worms. If we start to screen the geriatric population more for PEs, what are the downstream consequences? We will begin to adopt age-adjusted D-dimers more? Will we seen an increase in V/Q scans in patients with kidney disease? Will we see an increase in kidney disease from the contrast used in CTAs? What is the appropriate management for the subsegmental PEs that we find? Will we see an increase in bleeding complications such as GI bleeds or head bleeds in the fall-prone geriatric population? What implications will this have for ER workflow? This article, like many others, is merely a beginning step towards more inquiry. It is not practice-changing, but should increase awareness for the prevalence of PE especially in our geriatric population.

What this paper does do is to open an avenue to have discussions with patients. In the world of the internet and immediate access to information, patients come with new expectations. While we may see an increase in patients or other physicians requesting CT scans and D dimer testing, we have the opportunity to better inform our patients. Identifying your patient’s concerns and addressing them with evidence-based medicine should be the foundation of our practice. Being able to translate the evaluation for DVT and PE for a lay person can be difficult, but may produce a more informed and empowered patient. This paper, and the media attention it received, is a launch point for a larger discussion we should all be having with our patients.

Task Force for the Diagnosis and Management of Syncope, European Society of Cardiology, European Heart Rhythm Association, Heart Failure Association, Heart Rhythm Society. Guidelines for the diagnosis and management of syncope (version 2009). Eur Heart J 2009;30:2631-2671
Strickberger SA, Benson DW, Biaggioni I, et al. AHA/ACCF Scientific Statement on the evaluation of syncope: from the American Heart Association Councils on Clinical Cardiology, Cardiovascular Nursing, Cardiovascular Disease in the Young, and Stroke, and the Quality of Care and Outcomes Research Interdisciplinary Working Group; and the American College of Cardiology Foundation: in collaboration with the Heart Rhythm Society: endorsed by the American Autonomic Society. Circulation 2006;113:316-327

Will Denq, MD & Evan Kuhl, MD are both Emergency Medicine Residents at The George Washington University Hospital

Emergency physicians commonly use a combination of history, physical exam, EKG, labs, and imaging to help diagnose congestive heart failure (CHF) in the emergency department in patients who present with shortness of breath. However, sometimes it can be confusing and what initially looks like heart failure ends up being something else, such as pneumonia (1-3). Misdiagnosis can have real consequences, particularly when diuretics used to improve CHF result in worsened conditions such as a pulmonary embolus or other serious pathologies.

In a recent paper, Basset et al derived and validated a CHF clinical prediction rule called the Brest score (4) which is intended to improve the accuracy of a heart failure diagnosis in ED patients. They looked at 927 undifferentiated dyspneic patients in their ED to create it and externally validated it in 206 patients. The score relies on 11 variables based only on history, physical, and EKG to risk stratify patients into low, intermediate, and high probability groups.

The prevalence of confirmed CHF patients was 6.7% in the low probability group, 58% in the intermediate, and 91.5% in the high. The authors argue that these results are accurate enough when weighing risks and benefits to begin early treatment in the high risk group. Unlike many diagnoses made in the ED, a healthcare provider using this score may not necessarily have to wait for labs or radiology results to initiate management. On the other spectrum, with so few CHF patients in the low risk group, patients who are low risk by the Brest score should prompt the provider to search for other etiologies of dyspnea.

There were several limitations of the paper. The score still needs a prospective impact analysis validation, it was from a single center, and several pathologies may coexist in a single patient. Although the risk-benefit ratio skews towards treatment in the high risk group, the subsequent management of CHF is not benign especially in volume-depleted patients.

The Brest score represents an interesting adjunct in the early diagnosis of CHF in the undifferentiated dyspneic ED patient. With time being an important factor in management and disposition of CHF, the score stands alone in quickly risk stratifying the patient without waiting for labs or radiology results. Other CHF decision rules, such as the Boston criteria (5), Steingart and PRIDE (6,7) decision tools, rely on time-delaying lab results such as BNP or radiology results such as a CXR. The Framingham (8) and Gothenburg (9) scores were epidemiologic studies and utilize certain factors that are not readily obtainable at the bedside.

Along with decision rules, other data at the bedside can be helpful to diagnose CHF, such as lung ultrasound. A 2015 multi-center prospective study (n=1005) by Pivetta et al demonstrated a significantly higher accuracy than all other standards when using the clinical evaluation and bedside lung ultrasound. Sensitivity was reported at 97% with specificity being 97.4%. Perhaps a completely validated Brest score could improve the diagnosis when used in conjunction with lung ultrasound and move away from using data such as BNP or chest xray. It could result in an accurate, bedside diagnosis that does not rely on labs or imaging. Although not ready for prime time, it could significantly increase the accuracy of CHF diagnosis and accelerate management in the undifferentiated dyspneic ED patient.

Brest score:

*Variables*	Points (+2)
Age > 65	1
*Anamnesis variables*
Sudden dyspnea	2
Onset of symptoms at night	1
Orthopnea	1
*Risk Factor variables*
Prior CHF episode	2
Myocardial infarction	1
Chronic pulmonary disease	-2
*Clinical Examination variables*
Pulmonary crackles	2
Pitting leg edema	1
*EKG Abnormality variables*
ST segment abnormalities	1
Atrial fibrillation/flutter	1

*Max score*	15

Low Risk: ≤ 3

Medium Risk: 4 - 8

High Risk: ≥ 9

The Brest score starts automatically with 2 points – perhaps a superstitious move by the authors to avoid an unlucky total of 13 points.

References:

Remes J, Miettinen H, Reunanen A, Pyörälä K. Validity of clinical diagnosis of heart failure in primary health care. Eur Heart J 1991;12:315–21.
Fuat A, Hungin AP, Murphy JJ. Barriers to accurate diagnosis and effective management of heart failure in primary care: qualitative study. BMJ 2003;326:196.
Ansari M, Massi BM. Heart failure: how big is the problem? Who are the patients? What does the future hold? Am Heart J 2003;146:1–4.
Basset A, et al, Development of a clinical prediction score for congestive heart failure diagnosis in the emergency care setting: The Brest score, Am J Emerg Med (2016), http://dx.doi.org/10.1016/j.ajem.2016.08.023
Schocken DD, Arrieta MI, Leaverton PE, Ross EA. Prevalence and mortality rate of congestive heart failure in the United States. J Am Coll Cardiol 1992;20:301–6.
Baggish AL, Sieber U, Lainchbury JG, Cameron R, Anwaruddin S, Chen A, et al. A validated clinical and biochemical score for the diagnosis of acute heart failure. The ProBNP Investigation of Dyspnea in the Emergency Department (PRIDE) Acute Heart Failure Score. Am Heart J 2006;151:48–54.
Steinhart B, Thorpe KE, Bayoumi AM, Moe G, Januzzi Jr JL, Mazer CD. Improving the diagnosis of acute heart failure using a validated prediction model. J Am Coll Cardiol 2009;54:1515–21.
McKee PA, Castel WP, McNamara PM, KannelWB. The natural history of congestive heart failure: the Framingham Study. N Engl J Med 1971;285:1441–6.
Eriksson H, Caidahl K, Larsson B, Ohlson LO,Welin L,Wilhelmsen L, et al. Cardiac and pulmonary causes of dyspnoea–validation of a scoring test for clinical-epidemiological use: the Study of Men Born in 1913. Eur Heart J 1987;8:1007–14.

Will Denq, MD is an Emergency Medicine Resident at The George Washington University Hospital

Tag: Will Denq

Prevalence of Pulmonary Embolism among Patients Hospitalized for Syncope

The Brest Score – a new, quick CHF clinical prediction score at the bedside

Subscribe By Email