Risk factors for diagnostic delay in acute aortic dissection.
ABSTRACT In acute aortic dissection (AAD), timely diagnosis is challenging. However, dedicated studies of the entity and determinants of delay are currently lacking. We surveyed pre-/in-hospital time to diagnosis and explored risk factors for diagnostic delay. We analyzed the dedicated database of a metropolitan AAD network (161 patients diagnosed since 1996; 115 Stanford type A) in terms of hospital arrival times (from pain to presentation at any hospital) and in-hospital diagnostic times (presentation to final diagnosis). Median (interquartile range) in-hospital diagnostic times were approximately twofold greater than hospital arrival times (177 minutes, 644, vs 75 minutes, 124, p = 0.0001, Wilcoxon test). Median annual in-hospital diagnostic times were most often approximately 3 hours (spread was wide, but decreased after 2001; rho = -0.94, p = 0.005). Risk factors (univariate analysis) for in-hospital diagnostic time >75th percentile (12 hours) included pleural effusion (odds ratio 3.96, 95% confidence interval 1.80 to 8.69), dyspneic presentation (odds ratio 3.33, 95% confidence interval 1.93 to 8.59), and age <70 years (odds ratio 2.34, 95% confidence interval 1.03 to 5.36). Systolic arterial pressure < or =105 mm Hg decreased the likelihood of lengthy diagnosis (odds ratio 0.08, 95% confidence interval 0.01 to 0.59). In patients (n = 82) with routine values (since 2000), troponin positivity (odds ratio 3.63, 95% confidence interval 1.12 to 11.84) and an acute coronary syndrome-like electrocardiogram (odds ratio 2.88, 95% confidence interval 1.01 to 8.17) were also risk factors. In conclusion, in a metropolitan setting, most of the diagnostic delay may occur in hospital. At presentation, pleural effusion, troponin positivity, acute coronary syndrome-like electrocardiogram, and dyspnea are possible "clinical confounders" associated with particularly long in-hospital diagnostic times.
- [show abstract] [hide abstract]
ABSTRACT: We investigated frequency/characteristics of acute coronary syndrome-like (ACS-like) electrocardiographic (ECG) profiles among patients with a final diagnosis of acute aortic syndrome (AAS), and explored pathophysiologic determinants and prognostic relevance within each Stanford subtype. We blindly reviewed presentation electrocardiograms of 233 consecutive patients with final diagnosis of AAS (164 Stanford type A) at a regional treatment center. Prevalence of ACS-like ECG findings was 27% (type A, 26%, type B, 29%); most were non-ST-elevation myocardial infarction-like. Patients with ACS-like ECG findings more often had coronary ostia involvement (p=0.002), pleural effusion (p=0.02), significant aortic regurgitation (p=0.01), and troponin positivity (p=0.001). ACS-like ECG profile in type A disease was independently associated with coronary ostia involvement (odds ratio [OR] 5.27, 95% confidence interval [CI] 1.75 to 15.88). ACS-like ECG profile predicted in-hospital mortality (OR 2.90, 95% CI 1.24 to 6.12), as did age (each incremental 10-year: OR 1.59, 95% CI 1.14 to 2.22), and syncope at presentation (OR 2.90, 95% CI 1.16 to 7.24). In conclusion, about 25% of our AAS patients (in either Stanford subtype) presented ACS-like ECG patterns-often with non-ST-elevation myocardial infarction characteristics-which could cause misdiagnosis. ACS-like ECG profile was associated with more complicated disease, and in type A disease was a strong independent predictor of in-hospital mortality.The American Journal of Cardiology 10/2007; 100(6):1013-9. · 3.21 Impact Factor
Article: Acute cardiac tamponade.New England Journal of Medicine 09/2003; 349(7):684-90. · 51.66 Impact Factor
Article: [Cardiogenic shock].Annales Françaises d Anesthésie et de Réanimation 02/1985; 4(4):43A-52A. · 0.84 Impact Factor