In a prospective clinical study, routine endomyocardial biopsy (EMB) for early detection of cardiac allograft rejection was replaced by two noninvasive diagnostic techniques. In 32 patients who underwent transplantation during a 6-month period, intramyocardial electrogram (IMEG) amplitude was recorded with the telemetry function of a dual-chamber pacemaker system. An amplitude decrease of more than 8% below the individual range of variability in overnight long-term recordings served as an indicator of rejection. A prolongation of the time interval Te--the time span between maximal posterior wall contraction and peak retraction velocity--of more than 20 msec was used as an additional indicator of rejection-related left ventricular dysfunction. For patient safety, routine EMBs were still performed but remained undisclosed to the transplantation team. The pathologist was asked to intervene in cases of discrepancies between biopsy results and medical decisions, but no such intervention was necessary. Twenty-seven rejection episodes were treated medically, of which 22 were diagnosed noninvasively. In four patients, EMB, ordered because noninvasive parameters remained inconclusive, led to medical treatment. No false negative IMEG results were observed. Retrospective correlation with rejection gave a 100% negative predictive value for IMEG and a 96.9% negative predictive value for echocardiographic diagnosis. We conclude that omission of a schedule of routine biopsies is justified, if the described techniques of IMEG and echocardiography are meticulously applied and indicate normal cardiac function. EMB remains indicated if noninvasive parameters are not conclusive or if false-positive IMEG results, which were observed in the presence of infection or arrhythmias, are suspected. The frequency of EMB can be reduced by 90%, compared to conventional routine biopsy schedules. Diagnostic safety is increased because the additional information gained from IMEG and echocardiography is helpful in the interpretation of EMBs.