LO075: Clinical exam for acute aortic dissection: a systematic review and meta-analysis

Abstract

Introduction: Acute aortic dissection (AAD) is difficult to diagnose and if missed carries a significant mortality rate. Our aim was to assess the accuracy of history, physical exam and plain radiographs compared to advanced imaging in the diagnosis of AAD in adults presenting to the ED with a clinical suspicion of AAD. Methods: We conducted a librarian assisted systematic review. Databases searched: Pubmed, Medline, Embase and the Cochrane database from 1968 to January 2016. No restrictions for language were imposed. Studies were reviewed and data extracted by two independent reviewers. AAD was defined by CTA, MRI or TEE Prospective and retrospective studies of patients presenting with a clinical suspicion of AAD were included. Case series were excluded. Studies were combined if low clinical and statistical heterogeonity (I2<30%). Study quality was assessed using the QUADAS tool. Bivariate random effects meta analyses using Revman 5 and SAS 9.3 was performed. Results: We identified 792 records: 61 selected for full text review, 13 included and a further 7 from reference searches. 20 studies with 4721 participants were included (mean QUADAS score 12/14 SD 1.2, Kappa 0.8). Prevalence of AAD ranged from 9.6-76.1% (mean 39.1% SD 17.1%). Mean diagnosis in those without AAD varied between studies with ACS (30.3% SD 30.1%), Anuerysm(12.4% SD 9.8%), Chest wall pain(18.1% SD 13.3%) and PE(7.9% SD 7.85%) being the most common. The clinical findings most suggestive of AAD were, neurological deficit (specificity 94% LR 4.1 [95% CI, 3.1-5.2], I2 0%, n=9), hypotension(specificity 94% LR 2.6 [95% CI 1.6-4.2], I2 0%, n=8), pulse deficit (specificity 92% LR 3.4 [95% CI 1.8-6.4], I2 0%, n=9) and syncope (specificity 92% LR 1.4 [95% CI 1.1-1.8], I2 10%, n=6). The most useful for identifying patients less likely to have AAD were an absence of a widened mediastinum (sensitivity 80% LR 0.3 [95% CI 0.2-0.5], I2 20%, n=13) and an AHA Aortic dissection risk score <1 (n=2 sensitivity 91%,99% LR 0.02,0.22, [95% CI 0.003-0.128, 95%CI 0.2-0.3]). Conclusion: Suspicion for AAD should be raised with syncope, hypotension and pulse or neurological deficit in the appropriate clinical setting. Conversely the absence of a widened mediastinum and a low ADD score decreases likelihood. Clinical exam alone cannot rule out acute aortic dissection but it can help risk stratify for further testing.