Dyspnea may be a sign of cardiac disease, but it is not typically included as a prognostic predictor of cardiac risk. Abidov and colleagues studied 17,991 patients referred for cardiac stress testing and myocardial perfusion imaging to see if dyspnea provided prognostic information beyond the typical risk markers. In addition to typical clinical information, such as age, sex, cardiac risk factors, and nature of chest pain, researchers also asked patients if they had experienced shortness of breath, which is a symptom of dyspnea. The patients were then followed for a mean of 2.7 (±1.7) years.

Investigators prospectively enrolled this cohort and categorized patients into five groups: typical angina, atypical angina, non-anginal chest pain, no chest pain with dyspnea, and asymptomatic. Patients also were subdivided based on whether or not they had known coronary disease. The groups then underwent cardiac stress testing that included 11,888 exercise stress tests and 6,103 pharmacologic stress tests. Of the enrolled patients, 5,804 also had myocardial perfusion imaging performed as part of their tests. Patients with dyspnea accounted for 6 percent of the total group and were more likely to be older, have left ventricular enlargement, and have hypertension or diabetes. They also had less inducible ischemia than patients with typical angina.

During the follow-up period, 786 patients without known coronary disease died, including 224 who died from cardiac causes. Another 720 patients with known disease died, including 347 who died from cardiac causes. Within these groups, patients with dyspnea died at a proportionally higher rate than patients in the other categories. Dyspnea independently predicted death from cardiac and all causes after multivariable and propensity analysis; it carried a fourfold risk versus asymptomatic patients and a twofold risk versus those with typical angina. The authors conclude that dyspnea is a potentially important marker of mortality risk in patients referred for exercise testing. They caution that these results deserve further study to see how dyspnea can be incorporated properly into cardiac risk assessment.