Although acute diastolic dysfunction is an early sequela of the rejecting heart, reported sensitivities and specificities have varied widely when Doppler echocardiography is used for rejection surveillance. This study examines the temporal relationship between changes in Doppler echocardiographic indexes of diastolic function and sequential endomyocardial biopsies to identify possible factors accounting for false-positive and false-negative results. A total of 114 Doppler echocardiographic studies and biopsies were performed weekly in 39 patients aged 14 to 59 years during the initial 3 months after heart transplantation. All Doppler examinations were within 24 hours of biopsy and were analyzed in a blinded fashion. Onset of restrictive physiology, defined as a 15% decrease in either isovolumic relaxation time or pressure half-time, was determined by analysis of the Doppler mitral flow velocity curve. [GRAPHICS] Sensitivity, 82%; specificity, 79%; positive predictive value, 79%; negative predictive value, 82%; accuracy, 81%. chi-2 = 30.22; p = 0.001. Positive Doppler echocardiograms were associated with concurrent rejection in 80% of patients on biopsy. Of 12 false-positive Doppler studies, moderate rejection developed in 7 (58%) patients within 28 days, one patient died of progressive graft failure, and four (25%) had spontaneous improvement of diastolic function. Of 10 false-negative Doppler studies, five were concurrent with rejection already being treated, three with mild rejection not requiring treatment, and two with new moderate rejection on biopsy. We conclude: (1) Doppler echocardiographic false-negative studies occur most frequently during the course of rejection therapy; (2) restrictive physiology preceding biopsy evidence of rejection accounts for a large proportion of Doppler false-positive studies, and (3) positive Doppler echocardiography in the absence of concurrent rejection by biopsy is a strong indication for follow-up biopsy.