Written by: Samantha Stark, MD (NUEM ‘20) Edited by: Jesus Trevino (NUEM ‘19) Expert Commentary by: Keith Hemmert, MD
The diagnosis of aortic dissections and other acute aortic syndromes (AAS) has long plagued the emergency physician due to the non-specific nature of their presenting symptoms and the potentially catastrophic consequences of a missed diagnosis. While there is increasing interest and a growing body of evidence regarding the use of D-dimer in diagnosis, guidance on how to use the D-dimer and comfort in doing so are lacking. In the ADvISED study (Diagnostic Accuracy of the Aortic Dissection Detection Risk Score Plus D-Dimer for Acute Aortic Syndromes), the safety and efficiency of integrating a pretest probability assessment with D-dimer testing is evaluated.
This was a multicenter prospective observational study that involved 6 hospitals in 4 different countries from 2014-2016, including 1850 patients. They observed the failure rate and efficiency of a diagnostic strategy for ruling out AAS that involved determining pretest probability and combining this with a D-dimer test. The tool used for assessing pretest probability was the aortic dissection detection risk score (ADD-RS, see below) and the D-dimer was considered negative if <500 ng/mL. As above, primary and secondary outcomes were the failure rate and efficiency of this strategy, respectively.
*For each risk category, one point is assigned if one or more risk factors is present. The ADD-RS can therefore vary from 0-3.
For the purposes of this study, it is appropriately assumed that anyone with ADD-RS>1 would need conclusive testing (CTA, TEE, or MRA) to evaluate for AAS regardless of D-dimer level. Therefore, the investigators looked primarily at the integration of negative D-dimer (DD-) testing with ADD-RS=0 or ADD-RS<1 as a possible rule out strategy for AAS. They found that among ADD-RS=0/DD- patients, the failure rate was 0.3% (1/294 patients, 95% CI, 0.1-1.9) and the efficiency in ruling out AAS was 15.9% (1/6 patients, 95% CI, 14.3-17.6); efficiency was computed as the number of patients with negative D-dimer within a risk category divided by the number of enrolled patients). Among ADD-RS<1/DD- patients, the failure rate was also 0.3% (3/924 patients, 95% CI, 0.1-1) and the efficiency was 49.9% (1/2 patients, 95% CI, 47.7-52.2). Of note, as mentioned above, in patients with ADD-RS>1/DD-, the failure rate was 4.4%, corresponding to 1 missed case for every 22 patients, an unacceptable failure rate for a potentially lethal condition.
While the study’s statistical methods were thorough and sound, and the results quite compelling, there were some issues with the study. Perhaps the most significant of these was that about half of the patients involved in the study did not undergo conclusive diagnosis for AAS (CTA, TEE, MRA, surgery, or autopsy), and their “case adjudication” was based on 2-week follow up data. Of the cases where presence or absence of AAS was determined based on follow up data alone, 13% were determined to have AAS, while 87% were determined to have other explanations for their symptoms. The authors make the arguably legitimate, but unvalidated, assumption that patients with undiagnosed symptomatic AAS would experience some significant clinical event in the 2-week time period from presentation to follow-up. This is further supported by the fact that they did identify 7 cases of AAS during said follow-up period. However, they state that, “nonetheless, we cannot exclude with certainty that in 731 study patients with ADD-RS<1/DD- and a negative 14-day follow-up, few cases of AAS with mild or atypical manifestations might have been missed.”
The authors also point out that there is no established acceptable failure rate of a rule out strategy for AAS. They extrapolate based on prior studies that showed a) the threshold clinical probability of AAS above which the benefits of testing outweigh the risks was 3% and b) similar strategies for PE rule out have been considered acceptable if the upper limit of the 95% CI around the failure rate was <3%, to suggest that the failure rate of 0.3% for both the ADD-RS=0/DD- and the ADD<1/DD- strategies (with upper limits of the 95% CI 1.9% and 1% respectively) could be considered acceptable. That suggestion is intriguing when there is data showing the misdiagnosis rate of AAS reaching as high as 40% and that a mere 2.7% of CTAs obtained to evaluate for AAS yield a positive result.
They conclude that integration of ADD-RS=0 or ADD-RS<1 with negative D-dimer testing may be considered to standardize the diagnostic rule out of AAS, and that expert evaluation and debate are needed to determine whether the outlined strategies are safe and efficient to be recommended and implemented in clinical practice. Below is a flowchart from the paper summarizing their proposed diagnostic approach.
Expert Commentary
Thank you for this thoughtful review of the ADvISED trial. Acute Aortic Syndromes (AAS), including aortic dissection, intramural hematoma, ulcer, and rupture, are a challenging set of pathologies for the Emergency Physician. As Drs. Stark and Trevino note in this excellent post, the clinical presentations are often vague, and the mortality rate is high if the diagnosis is missed. The approach to testing for AAS has historically not been based on any validated risk score, but rather on clinical gestalt. As a result of these factors the yield on diagnostic testing for AAS is low, raising concerns about resource use and radiation exposure. Hence, a validated approach to risk stratification, with an acceptably low miss rate, would be a great aid to the Emergency Physician.
It is always important to bear in mind the current standard of care for AAS. The American College of Emergency Medicine currently does not recommend the use of a clinical decision rule alone to rule out AAS, nor does it recommend the use of d dimer alone to rule out AAS. CTA, MRA and TEE are the recommended modalities to diagnose of rule out AAS. [1] While this study and others may ultimately be a factor in changing ACEP clinical policies, Emergency Medicine house staff should be mindful of the current standard of care when evaluating new diagnostic or risk stratification strategies.
As both the authors and Drs. Stark and Trevino point out, roughly half of the patients in the study did not have gold standard testing for AAS, whether by CTA, MRA, TEE, or autopsy; they were simply assumed to be negative for AAS. While it is difficult to find reliable data on the true mortality rate of untreated AAS, we can use mortality for acute aortic dissections as a proxy. The cumulative 14-day mortality for treated acute aortic dissections (including type A and type B, and both medical management and surgical management) approaches 50%. [2] One can presume that the mortality rate for untreated disease would be even higher. While this does lend a modicum of logic to the authors’ approach (if they aren’t dead at 14 days, they probably don’t have AAS), it is far from scientific, and fails to meet the bar for the level of evidence required to forego conclusive diagnostic testing for such a lethal pathology.
The highly lethal nature of AAS also raises the question of the acceptable miss rate for a risk stratification tool such as the one proposed in the ADvISED study. As mentioned, a tool like this will inevitably draw comparisons to the PERC rule, which has a failure rate of <2%. However, AAS is a substantially more lethal disease than PE; the 30-day mortality rate of PE is 4%, and the 1-year mortality rate is 13%. [3] While an apples to apples comparison of the mortality rates of PE and AAS is challenging, the aforementioned 14 day mortality rate for treated AAS (nearly 50% at 14 days) provides a stark contrast. Additionally, AAS encompasses a variety of discrete pathologies, some of which are extraordinarily lethal (e.g., the mortality of an ascending aortic dissection is 1% to 2% per hour after symptom onset). [2] To use another proxy, the in-hospital mortality rate for type A dissections is 22%, and for type B 13%. [4] These numbers, of course, discount the (presumably not insignificant) number of patients who die before completing transfer to a quaternary care facility.
All of this is to say that AAS is a much more lethal set of pathologies than PE, and therefore the acceptable failure rate for a risk stratification strategy must be correspondingly lower. The authors report a failure rate for the most conservative option (ADD-RS=0/DD-) of 0.3% - but this must be interpreted in light of the lack of conclusive diagnostic imaging in roughly half of the patients enrolled in the study. Lastly, a comparison of this strategy to clinical gestalt would enable us to evaluate the superiority of this approach to the current one; this is a ripe area for future investigation. In summary, the approach to AAS proposed in the ADvISED study is not ready for widespread implementation, although it is a promising step toward a usable risk stratification strategy.
References
1. ACEP. ACEP Clinical Policy on Thoracic Aortic Dissection
2. Tsai TT, Nienaber CA, Eagle KA. Acute aortic syndromes. Circulation. 2005;112(24):3802-3813. doi:10.1161/CIRCULATIONAHA.105.534198
3. Alotaibi GS, Wu C, Senthilselvan A, McMurtry MS. Secular Trends in Incidence and Mortality of Acute Venous Thromboembolism: The AB-VTE Population-Based Study. Am J Med. 2016;129(8):879.e19-879.e25. doi:10.1016/j.amjmed.2016.01.041
4. Evangelista A, Isselbacher EM, Bossone E, et al. Insights from the international registry of acute aortic dissection: A 20-year experience of collaborative clinical research. Circulation. 2018;137(17):1846-1860. doi:10.1161/CIRCULATIONAHA.117.031264
Keith Hemmert, MD
Assistant Professor of Emergency Medicine
Hospital of the University of Pennsylvania
Other Posts You May Enjoy
References
1. Nazerian P, Mueller C, Matos Soeiro A, Leidel B, Savadeo SAT, Giacino F, Vanni S, Grimm K, Oliveira MT, Pivetta E, Lupia E, Grifoni S, Morello F. Diagnostic Accuracy of the Aortic Dissection Detection Risk Score Plus D-Dimer for Acute Aortic Syndromes: The ADvISED Prospective Multicenter Study. Circulation. 2018;137:250-258.
2. Righini M et. al. Age-adjusted D-dimer cutoff levels to rule out pulmonary embolism: the ADJUST-PE study. JAMA. 2014;311:1117-1124.
3. Perrier A et. al. Multidetector-row computed tomography in suspected pulmonary embolism. N Engl J Med. 2005;352:1760-1768.
4. Van Belle A et. al. Effectiveness of managing suspected pulmonary embolism using an algorithm combining clinical probability, D-dimer testing, and computed tomography. JAMA. 2006;295:172-179.
5. Sarasin FP et. al. Detecting acute thoracic aortic dissection in the emergency department: time constraints and choice of the optimal diagnostic test. Ann Emerg Med. 1996;28:278-288.
6. Hansen MS et. al. Frequency of an inappropriate treatment of misdiagnosis of acute aortic dissection. Am J Cardiol. 2007;99:852-856.
7. Kurabayashi M et. al. Factors leading to failure to diagnose acute aortic dissection in the emergency room. J Cardiol. 2011;58:287-293.
8. Zhan S et. al. Misdiagnosis of aortic dissection: experience of 361 patients. J Clin Hypertens (Greenwich). 2012;14:256-260.
9. Lovy AJ et. al. Preliminary development of a clinical decision rule for acute aortic syndromes. Am J Emerg Med. 2013;31:1546-1550.