Aortic Dissection Detection Risk Score: A Clinical Decision Rule That Needs Some Parenting

Clinical decision rules and risk scores can be useful adjuncts to the evaluation of patients in the emergency department (ED). These decision aids work best either when they prompt the ordering of diagnostic tests, improving the ability of the clinician to detect a disease process, or when they suggest a diagnostic test is unnecessary thereby saving important health care resources. Many of the most commonly used clinical decision rules are constructed to perform the latter role. The Pulmonary Embolism Rule out Criteria (PERC) rule can obviate the need to work up pulmonary embolism (PE) in low-risk patients. The Wells criteria, when applied to patients with suspected PE in whom imaging testing will be ordered, can identify low-risk patients in whom D-dimer can be used to avoid CT or V/Q imaging. The Aortic Dissection Detection Risk Score (ADD-RS) has been proposed as a clinical decision aid for the evaluation of acute aortic syndrome (AAS). Acute aortic syndrome, a spectrum of pathologies including penetrating aortic ulceration, intramural hematoma, and dissection, is both deadly and variable in its presentation. Its characteristic symptoms are often vague or nonspecific, overlapping with a number of other clinical entities. Approximately 25% of cases are either missed or misdiagnosed.1-3 The search for a diagnostic adjunct to aid in the diagnosis of AAS has thus far been focused on D-dimer testing or the development of a risk stratification score. The authors of the 2010 ACC/AHA Thoracic Aortic Disease Guidelines introduced the ADD-RS as a tool to aid in the diagnosis and risk stratification of patients with suspected AAS, despite the conspicuous absence of an emergency physician on the authorship or review team.4 The ADD-RS has been evaluated in several studies and does have the ability to classify patients into a low- (ADD-RS = 0), moderate- (ADD-RS = 1), and high-risk (ADD-RS > 1) groups where the prevalence of AAS in these categories ranges from 2.8 to 6.2, 7.9 to 27.3, and 7.8% to 39.1%, respectively.5-8 The generalizability of these proposed validation studies has been called in to question as the prevalence of AAS in the study populations (notably all performed at European and/or South American study sites) was quite high, ranging from 13% to 22.5%. This high prevalence of disease is likely a result of the inclusion criteria of many of these trials which included clinician ordering of diagnostic imaging. Even in the only prospective study to date, which did not require the ordering of diagnostic imaging, the prevalence of AAS was 13%.5 Ohle and colleagues,9 in this issue of Academic Emergency Medicine, find a somewhat expected result of applying a risk score validated in a high-prevalence population to a lower-prevalence population. With a prevalence of AAS of 3.2% in their cohort, they found that the ADD-RS = 0 had a sensitivity of 100%. However, the specificity of 12.3% is much lower than in previous studies. Stiell and Wells10 outline six questions that fully mature decision rules should be able to answer. In examining the ADD-RS through the lens described above, one can see there is a long way to go before this risk score can be safely implemented into clinical practice. Given variation in clinical practice, a clinical decision rule would be undoubtedly helpful in the evaluation of possible AAS. The methodologic derivation of this particular rule is, however, unclear and appears to have been through an expert consensus statement based on a thorough literature review.4 The means by which consensus was reached and the process by which the rule was derived has not been fully described. The rule has been evaluated primarily in a retrospective fashion in cohorts of patients where a high suspicion of disease exists. As noted above, the sole prospective validation study has questionable external validity compared with a North American patient population given the high prevalence of AAS. As of yet, there have been no descriptions of AAD-RS implementation into clinical practice and the questions of cost-effectiveness and its potential impacts on health care utilization are, as of yet, unexplored. In reference to the “fully mature” decision rule, the ADD-RS is squarely in its awkward preteen years. Ohle and colleagues9 rightly bring up a number of concerns about applying this risk score to a low-prevalence patient population. The low specificity and high sensitivity will result in unacceptably high false-positive rates, resulting in increased rates of low-yield, high-cost diagnostic imaging. If the risk score were used broadly with any patient with undifferentiated chest pain in the ED (as opposed to those with higher pretest probability of AAS), one could expect an even greater number of false-positives and the potential for greatly increased downstream testing. Furthermore, the true sensitivity of ADD-RS = 0 in a North American patient population is unknown. Ohle and colleagues,9 with a small retrospective patient population, report a sensitivity of 73.5% to 100%. In the higher prevalence cohort of the European validation studies sensitivity of ADD-RS = 0 ranged from 91.1% to 98.9%.5-8, 11 The authors also discuss the role of D-dimer in the diagnosis of AAS. Based on the results of the ADvISED trial and a subsequent retrospective validation trial by Gorla et al.,8 the combination of negative D-dimer and ADD-RS = 0 demonstrates potential in the evaluation of AAS.5, 8 However, given the lack of a large prospective validation study, the question of whether or not an ADD-RS of 0 (with or without negative D-dimer) is sufficient to rule out AAS in a low-prevalence patient population is still unanswered. The evaluation of the ADD-RS has thus far focused on the ability of the risk score to identify low-risk patients in whom diagnostic evaluation may be unnecessary. If prospective studies in a lower prevalence patient population show that an ADD-RS = 0 (with or without D-dimer) obviates the need for diagnostic imaging, it would be a helpful addition our clinical practice. What is not known is whether the benefits of decreased imaging utilization for ADD-RS of 0 patients will be offset by increased diagnostic evaluation of patients with ADD-RS ≥ 1. Furthermore, given the challenge of diagnosing AAS, the ideal clinical decision rule for AAS would allow us to identify the cases AAS we currently are missing. There is no evidence that the ADD-RS decreases the rates of missed diagnoses. Given the current evidence, the ADD-RS is not the risk score we are looking for.