Evaluation of Syncope
Abstract & Commentary
Synopsis: The presence of suspected or certain heart disease is a strong predictor of cardiac syncope, and in patients unlikely to have heart disease, only prior palpitation predicted cardiac syncope.
Source: Alboni P, et al. J Am Coll Cardiol. 2001;37: 1921-1928.
Although it is widely held that the history is of paramount importance in the establishment of the etiology of syncope, there are little data on the relative diagnostic value of particular symptoms. Thus, Alboni and associates studied the predictive value of specific historical features for identifying the cause of syncope in 341 patients studied in a syncope unit. The patients were referred from 3 hospitals in Italy. Evaluation of the patients followed an algorithm suggested by the Associazione Nationale Medicine Cardiologi Ospedalieri, with the addition of an adenosine triphosphate (ATP) bolus test, if the cause was still unexplained at the end of the standard work-up. Patients were classified into 1 of 10 diagnoses: orthostatic; typical vasovagal; situational (ie, micturation, tussive); tilt induced; carotid sinus; drug induced; cardiac mechanical (ie, aortic stenosis); adenosine sensitive; arrhythmic; neuropsychiatric or metabolic. The clinical evaluation categorized 46 findings related to the initial syncopal episode. The 341 patients studied represent 96% of the enrolled patients; their mean age was 61 years and 54% were men. Broadly categorized, 58% of the patients had neurally mediated syncope (tilt 27%, carotid 14%, situational 7%, vasovagal 6%, orthostatic 2%, ATP 2%); 23% had cardiac causes (mechanical, ischemic, or arrhythmic), the majority of whom had AV block; 18% were unexplained; and 1% were neuropsychiatric. Almost half of the cardiac causes were discovered in the initial evaluation, but only 26% of the neurally mediated were found.
Multivariate analysis of the historical variables that were not part of the characteristics of the syncopal episode showed that only heart disease predicted a cardiac cause (odds ratio, 16; 95% confidence interval 5-48). Heart disease identified 95% of the patients with the final diagnosis of cardiac syncope but was present in 45% without a cardiac cause. By contrast, the absence of heart disease identified 97% of patients without a cardiac cause. Among those with cardiac syncope the most predictive indicators were supine syncope, blurred vision, or symptoms for less than 4 years. Predictors of neurally mediated syncope in patients with heart disease were symptoms greater than 4 years, history of presyncope, and recovery nausea. In patients without heart disease, the only predictor of cardiac syncope was palpitation preceding the syncope and of neurally mediated syncope was prodromal symptoms for greater than 10 seconds. Alboni et al concluded that the presence of suspected or certain heart disease is a strong predictor of cardiac syncope and, in patients unlikely to have heart disease, only prior palpitation predicted cardiac syncope.
Comment by Michael H. Crawford, MD
This study confirms the pivotal role of diagnosing heart disease in the evaluation of the cause of syncope. In this study, a cardiac cause was largely established by history, physical examination, and the resting ECG. Although echocardiography is in the algorithm, it appears after the presumptive diagnosis of heart disease as a confirmatory test. In the United States, most physicians consider echocardiography as an integral step in the initial diagnosis of a cardiac structural cause of syncope. Thus, had this study been done in the United States the results may have been different. Regardless, a high likelihood of established cardiac disease had a high sensitivity but a low specificity for cardiac syncope. Thus, the absence of heart disease on initial evaluation has a high negative predictive value (97%) for cardiac syncope unless palpitation is a prodromal symptom.
Since 45% of the patients with heart disease had neurologically mediated syncope, it is worth noting that certain symptoms suggested this cause: symptom duration greater than 4 years, history of presyncope, and recovery nausea were independent predictors. Other prodromal symptoms or their lack was not helpful, nor was the duration of recovery. The latter finding is contrary to my clinical experience. Neurally mediated syncope usually results in a recovery phase of hours, whereas the recovery phase of a typical Stokes-Adams attack (cardiac arrhythmia) is often minutes in my experience. Of course there are exceptions to this rule and the mix of patients in this study did not support this distinction.
There are some weaknesses of this study. The diagnosis of the true cause of syncope is naturally imprecise, so there is no true gold standard for this study. Also, since the study was done in Italy, cultural differences may make particular symptoms more or less useful vs. other countries. This may explain some of the differences between the findings of this study and previous ones and my clinical experience. On the other hand, there are some real strengths in this study. The study population is large and extensively evaluated with tilt testing, electrophysiologic testing, and ATP bolus testing. Also, the stepwise approach promulgated here could save money and reduce the patient inconvenience associated with a syncope evaluation. For example, patients with a likely cardiac cause should probably be admitted to the hospital and those with neurally mediated syncope will usually be evaluated as outpatients.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content