The diagnostic accuracy of bedside maneuvers in the evaluation of patients with systolic murmurs has not been assessed objectively. Therefore, we evaluated 50 patients with documented systolic murmurs and compared all standard bedside techniques. Murmurs originating within the right-sided chambers of the heart were best differentiated from all other murmurs by augmentation of their intensity with inspiration (100 percent sensitivity, 88 percent specificity) and diminution of their intensity with expiration (100 percent sensitivity, 88 percent specificity). The murmur of hypertrophic cardiomyopathy was distinguished from all other murmurs by an increase in intensity with the Valsalva maneuver (65 percent sensitivity, 96 percent specificity) and during squatting-to-standing action (95 percent sensitivity, 84 percent specificity), and by a decrease in intensity during standing-to-squatting action (95 percent sensitivity, 85 percent specificity), passive leg elevation (85 percent sensitivity, 91 percent specificity), and handgrip (85 percent sensitivity, 75 percent specificity). The murmurs of mitral regurgitation and ventricular septal defect had parallel responses to all maneuvers, but could be differentiated from other systolic murmurs by augmentation of their intensity with handgrip (68 percent sensitivity, 92 percent specificity) and during transient arterial occlusion (78 percent sensitivity, 100 percent specificity), and by a decrease in their intensity during the inhalation of amyl nitrite (80 percent sensitivity, 90 percent specificity). No single maneuver identified the murmur of aortic stenosis, but the diagnosis could be made by exclusion. Although no single maneuver is 100 percent accurate in diagnosing the cause of a systolic murmur, its origin can be determined accurately at the bedside by observation of the response to a combination of maneuvers. (N Engl J Med 1988; 318:1572–8.) © 1988, Massachusetts Medical Society. All rights reserved.