Although RR interval variability appears to be an ideal method for assessing reinnervation after heart transplantation, it has been shown that respiratory sinus arrhythmia is caused by the mechanical effect of respiration on the right atrium. The neck-suction induces heart rate changes only by means of nervous reflex and its hemodynamic effect is local and hence appears as a useful method for assessing reinnervation. We tested the presence of autonomic reinnervation in 18 heart transplant recipients, compared to 12 donor-age-matched controls. We measured the power of RR interval low- (LF, around 0.1 Hz) and respiratory fluctuations (HF) before and during rhythmic neck-suction stimulation at 0.1 Hz and at a frequency (0.20 Hz) similar to, but distinct from, that of respiration (controlled at 0.25 Hz), before and during 0.04 mg/kg atropine infusion, using autoregressive spectral analysis of RR interval, respiration and neck pressure signals. The relationship between pairs of signals at each frequency was quantitatively assessed by bivariate coherence function. All transplanted subjects showed low-amplitude HF, related to respiration. Detectable LF (whose power was lower than in controls: 1.15 +/- 0.39 versus 6.08 +/- 0.27 1n-ms2, p < 0.001), non coherent with respiration, were present in 11/18 transplanted subjects, and correlated with months since transplantation (r = +0.59, p < 0.05). HF neck suction induced the presence of a 0.20 Hz fluctuation in 12/12 controls, distinct from and greater than the 0.25 Hz respiratory component (7.28 +/- 0.26 versus 6.69 +/- 0.74 1n-ms2, p < 0.01); this was not seen in any of the transplanted subjects.(ABSTRACT TRUNCATED AT 250 WORDS)