Absence of Parasympathetic Control of Heart Rate After Human Orthotopic Cardiac Transplantation
Background Partial reinnervation of cardiac sympathetic nerves has been observed after heart transplantation; we hypothesized that parasympathetic control to the heart after transplantation may return as well. To test this hypothesis, we examined heart rate responses produced by two cardiovascular reflexes whose efferent limbs are subserved by vagal fibers to the heart: (1) trigeminal reflex (simulated diving reflex) and (2) arterial baroreflex with phenylephrine injection.
Methods and Results An “early” group (n=31, <24 months after transplantation) and a “late group” (n=27, >45 months after transplantation) were studied and compared with a control group with intact cardiac innervation (n=32) and a renal transplant group with similar transplant immunosuppressive regimen (n=11). For trigeminal reflex testing, responses of the donor sinus node (DSN) (sinus node controlling heart rate) and recipient sinus node (RSN) in the innervated remnant right atrium in cardiac transplant patients were compared with heart rate responses in the control groups. For arterial baroreflex testing, baroreflex gains for the DSN and RSN in the cardiac transplant groups were compared with those of the control group. With engagement of the trigeminal reflex, the DSN rate of both transplant groups changed minimally (early, 1.2±1.2 bpm; late, 1.8±2.5 bpm) compared with the expected decrease in control subjects (−19.8±3.0 bpm) and renal transplant patients (−23.9±4.9 bpm) (P<.001 versus cardiac transplants). Changes in the RSN rate of both cardiac transplant groups (early, −13.0±4.0 bpm; late, −10.0±3.7 bpm) were similar to the control groups. Arterial baroreflex gains for the DSN were also depressed (early, 0.1±0.2 ms/mm Hg; late, 0.2±0.2 ms/mm Hg) compared with control (14.9±1.8 ms/mm Hg) and RSN (early, 9.9±1.3 ms/mm Hg; late, 10.9±1.3 ms/mm Hg; P<.001 versus DSN transplant).
Conclusions These data suggest that parasympathetic influences on donor heart rate are absent in the majority of patients up to 96 months after cardiac transplantation.
Cardiac transplantation results in complete afferent and efferent denervation of the donor atria and ventricles. This denervation includes both sympathetic and parasympathetic divisions of the autonomic nervous system. Although functional reinnervation of both sympathetic and parasympathetic fibers to the heart have been demonstrated in the canine transplant model within 6 months after transplantation,1 only limited functional sympathetic reinnervation has been demonstrated in humans.2,3 Several groups examined the return of parasympathetic innervation to the heart after transplantation; however, most of these have involved small numbers of patients early after transplantation.4–11 In general, these studies do not support vagal reinnervation of the heart after transplantation. A limited number of patients, however, have exhibited responses that were suggestive of the return of parasympathetic control of heart rate.4,5,8 No study has examined whether functional parasympathetic reinnervation occurs late (>48 months) after cardiac transplantation.
Parasympathetic responses appear to be important in human disease processes. Increased vagal tone is observed frequently during the early stages of acute myocardial infarction12 and during fainting.13 In patients with normally innervated hearts and a history of myocardial infarction, the absence of vagal mechanisms appear to be important in the development of ventricular arrhythmias.14,15 Thus, the presence or absence of parasympathetic reinnervation of the transplanted heart may be important, especially late after transplantation, when there is an increased incidence of coronary disease. Finally, the continued absence of parasympathetic innervation to the heart would dictate the continued ineffectiveness of digoxin and atropine and maneuvers that alter cardiac vagal tone.
The objective of these experiments, therefore, was to determine whether parasympathetic control of heart rate returns in both the short and long term after cardiac transplantation.
To assess the presence of parasympathetic influences on the donor sinus node after cardiac transplantation, the heart rate responses were measured for two reflexes whose efferent limbs are subserved by vagal fibers to the heart. Two separate experiments were performed: one in which the trigeminal reflex (simulated diving reflex) was examined and a second in which the arterial baroreflex was examined. These reflexes were chosen because their responses are well described and their afferent limbs subserve receptor areas that can be relatively selectively stimulated, thereby triggering responses simply and uniformly.
Responses of cardiac transplant patients’ donor sinus node (DSN) (actual transplant heart rate) were compared with the responses of the sinus node in control groups as well as with the responses of the recipient sinus node (RSN) of each cardiac transplant patient. Innervation to the RSN located in the right atrial remnant remains intact after cardiac transplantation.16,17 Thus, the RSN may respond to changes in vagal or sympathetic tone and serves as an internal control for each cardiac transplant patient.
Studies were performed in 58 cardiac transplant patients and 32 healthy control subjects. Eleven renal transplant patients also served as a control group because of their identical immunosuppressive regimen, including cyclosporine, which is known to affect sympathetic tone.18 The study protocol was approved by the Human Subjects Review Committee of the Virginia Commonwealth University and McGuire Veterans Affairs Medical Center, and written informed consent was obtained from all participants. Thirty-one cardiac transplant patients at <24 months after transplantation (mean, 12±1 months; range, 2 to 24 months) composed an “early group,” and 27 patients at >45 months after transplantation (mean, 66±3 months; range, 45 to 96 months) composed a “late group.” Cardiac transplant patients were without significant epicardial coronary artery disease and had normal left ventricular ejection fractions. Cardiac allograft rejection was excluded by right ventricular endomyocardial biopsy, usually obtained within 1 week of study. Control subjects were free of organic heart disease as determined by history and physical examination. In those >40 years old, exercise tolerance tests or cardiac catheterization was performed and found to be normal. Renal transplant patients were free of organic heart disease on the basis of history and physical examination. None of the study participants were taking β-adrenergic–blocking agents or had neurological or other concomitant diseases that would influence autonomic neural function. All transplant patients were receiving cyclosporine; all except 3 were receiving azathioprine; and 51 were receiving prednisone. Other medicines are listed in Table 1⇓. All vasculoactive medications were stopped ≥24 hours before the study.
Trigeminal Reflex Study
The diving reflex is an oxygen-conserving reflex present in both animals and humans consisting of striking bradycardia, peripheral vasoconstriction, diminished cardiac output, and well-maintained blood pressure.19,20 In humans, immersion of the face alone produces similar physiological changes and has been termed the simulated diving reflex or trigeminal reflex.21,22 This reflex is composed of an afferent limb consisting of facial cutaneous receptors subserved by the sensory division of the trigeminal nerve and an efferent limb consisting in part of vagal fibers to the heart. With immersion of the face in cold water (10° to 17°C), a decrease in heart rate occurs while blood pressure is maintained.
The trigeminal reflex was activated by having each subject immerse his or her face in cold water (10° to 15°C) for 25 to 30 sec. Blood pressure was measured at rest before immersion for baseline recordings and between 20 and 30 seconds during the immersion by sphygmomanometer. In control subjects and renal transplant patients, heart rate was measured from the surface ECG. In the cardiac transplant group, DSN rate was determined from the surface ECG, and RSN rate was determined from a recipient right atrial electrogram recorded using an esophageal pill electrode (Arzco Medical Electronics) The recipient right atrium is electrically isolated from the donor (transplanted) atrium and its sinus node, thereby allowing independent measurement of RSN and DSN rates. Eight cardiac transplant patients initially screened for study were found to have atrial flutter or fibrillation in the recipient atrium and were excluded from study. Both surface and right atrial electrograms were displayed continuously during the intervention on separate channels of a monitor/strip-chart recorder (VR-6; PPG Biomedical) and recorded at a paper speed of 50 mm/sec.
Each subject was studied in the Human Physiology Laboratory in the sitting position after fasting for ≥2 hours. After placement of surface and esophageal electrodes and sphygmomanometer cuff, each subject rested for 5 minutes before baseline recordings commenced. After baseline recordings and a small inspiration, each subject immersed his or her face in a basin of cold water placed before them for 25 to 30 seconds. On emerging from the water, subjects were instructed to avoid a large inspiration but to breathe calmly while recording was continued for an additional 30 seconds. Two interventions were done with a 5-minute recovery period between the interventions. The intervention with the maximum recorded change in heart rate was used for final data for each individual.
Arterial Baroreflex Study
The arterial baroreflex was activated by raising systemic arterial pressure with a bolus injection of phenylephrine in subjects or patients with systolic arterial pressure <160 mm Hg. In the control groups, heart rate was measured from surface ECG recordings. In the cardiac transplant groups, RSN rate was determined by recording the recipient right atrial ECG with a right atrial intracavitary electrode positioned posteriorly. The DSN rate was determined by the surface ECG. Systemic arterial pressure was measured by a femoral artery catheter with the use of saline-filled pressure transducers (Abbott Critical Care Systems). Respiratory movements were recorded with a bellows type respirometer, and all signals were displayed on separate channels of a monitor/strip-chart recorder (VR-12; PPG Biomedical) and recorded at a paper speed of 50 mm/sec.
Testing was performed in the cardiac catheterization laboratory after routine or annual (cardiac transplant patients) coronary arteriography for evaluation of possible coronary artery disease. Subjects were studied in the fasting state and were premedicated before catheterization with 50 mg diphenhydramine IV. After instrumentation was placed and a 5-minute rest period, baseline measurements were recorded; then, phenylephrine was injected intravenously over 5 to 10 sec followed by a 10-mL saline flush. An initial dose of 100 μg was given, and after blood pressure returned to baseline, subsequent increased doses were given until arterial pressure was increased by 20 to 30 mm Hg in a smooth ramp.
Trigeminal Reflex Study
For a given face immersion intervention, heart rate in control subjects or DSN (RR interval) and RSN (AA interval) rate in cardiac transplant patients were averaged over 10 beats for baseline values. Baseline values were compared with the maximum change occurring in these parameters during intervention. Baseline mean arterial pressure was compared with mean arterial pressure during intervention (calculated by adding one third of the pulse pressure to the diastolic pressure).
Changes in parameters from baseline to intervention within groups were tested for statistical significance by paired t tests. A general linear model one-way ANOVA was used to compare the change from baseline among groups. Estimate of the relation between the change in DSN rate during intervention and time after cardiac transplantation was determined by calculation of Pearson correlation coefficients.
Arterial Baroreflex Study
For each subject, systolic arterial pressure in response to phenylephrine measured during expiration was plotted against the subsequent RR interval (control subjects and DSN in cardiac transplant patients) or AA interval (RSN in cardiac transplant patients) and least-squares linear regression analysis was performed. The slope of each RR or AA interval–systolic pressure relation was used as a measurement of the baroreflex gain. This value was derived for each control subject or for the DSN and RSN for each cardiac transplant patient, respectively; therefore, a baroreflex gain was derived from the response of the DSN and RSN for each cardiac transplant patient. Comparison of baroreflex gains among groups was performed using a general linear model one-way ANOVA. All results are reported as mean±SEM.
Baseline values for the normal control group, the two groups of cardiac transplant patients, and the renal transplant patient control group are displayed in Table 2⇓. MAP was elevated in the transplant groups, probably secondary to immunosuppressive therapy with cyclosporine. DSN rates were elevated compared with heart rate in control subjects and renal transplant patients and were also elevated compared with RSN rates for early and late transplant groups.
Trigeminal Reflex Study
Fig 1A⇓ displays the mean maximum change in mean arterial pressure in response to face immersion in cold water for the groups studied. There was no significant change observed in the normal control or the early or late transplant groups. A significant increase in MAP was noted in the renal transplant group.
Fig 1B⇑ displays the mean maximum change in RSN rate in response to face immersion for the groups studied. There was a sizable and significant decrease in rates in the two control groups as well as both cardiac transplant groups, reflecting the normal increase in vagal tone from engagement of the trigeminal reflex. These findings were consistent with the fact that the RSN in cardiac transplant patients remains innervated after cardiac transplantation.
Fig 1B⇑ also displays the mean maximum change in DSN rate for both transplant groups compared with the maximum change in heart rate in control and renal transplant patients during face immersion. The DSN rate in both transplant groups did not change significantly from baseline. Compared with the other groups, this response was markedly attenuated. Examination of the responses of individual subjects for the early and late cardiac transplant groups are displayed as a function of time after transplantation in Fig 2⇓. Most patients in the early group (Fig 2A⇓) experienced only a minimal change in heart rate in response to face immersion, whereas two patients in this group experienced a more sizeable increase in rate. For the late group (Fig 2B⇓), the majority of patients experienced similar changes in rate compared with the early group except for two patients who experienced sizeable decreases in rate of −27 and −14 bpm, respectively. Corresponding changes in RSN rates were similar at −34 and −10 bpm for these subjects who were 72 and 60 months post-transplantation, respectively.
In an attempt to confirm that these observations were caused by the return of vagal influences on the DSN, we repeated the trigeminal reflex experiments before and after atropine administration in one of the subjects (initially 60 months after transplantation). This subject was now 144 months post-transplantation and had developed diabetes mellitus. The other subject had died in the intervening time period. Before atropine, the average change in rates for two face immersions were 9 bpm for DSN and −15 bpm for RSN. After atropine (1.5 mg), baseline DSN and RSN rates increased by 6 and 31 bpm, respectively. With face immersion, the DSN rate again increased by 6 bpm, whereas the RSN response was essentially abolished with only a slight decrease of −1 bpm. Thus, the DSN response in these experiments was not consistent with a return of vagal control to the heart.
The correlation between the changes in DSN rate and time after transplantation was poor (r=.23) and not significant (P=.38) for the early group. For the late group, this correlation was fair (r=.49) and just significant (P=.04).
Arterial Baroreflex Study
Baroreflex gains derived from phenylephrine injection for individual subjects as well as the group mean data are displayed in Fig 3⇓. In all subjects, regression lines were obtained with correlation coefficients of >.80 and values of P<.05. Group mean gains for control subjects (14.9±1.8 ms/mm Hg) and group mean gains derived from the response of the RSN in early (9.9±1.3 ms/mm Hg) and late (10.9±1.3 ms/mm Hg) cardiac transplant patients were similar (P=NS), which is in keeping with the fact that the RSN remains innervated. Baroreflex gains derived from the response of the DSN in cardiac transplant patients were essentially flat (early, 0.1±0.2 ms/mm Hg; late, 0.2±0.2 ms/mm Hg), indicating little or no increase in vagal outflow to the DSN in response to a rise in arterial pressure with phenylephrine. These gains were significantly less than those from control subjects and those derived from responses of the RSN (P<.001).
The heart is innervated by both sympathetic and parasympathetic divisions of the autonomic nervous system. Postganglionic parasympathetic fibers are distributed throughout the heart, but in mammals, the sinoatrial and atrioventricular nodes and the atria are more richly innervated than the ventricles and coronary vessels.23 The experiments in this study were specifically directed to examine whether functional parasympathetic control of the donor sinus node returns after cardiac transplantation. The results overall suggest that parasympathetic control was absent in the majority of patients up to 96 months after cardiac transplantation.
Several reports examine the question of whether parasympathetic reinnervation occurs after cardiac transplantation. Functional studies (vagal and stellate ganglion nerve stimulation and heart rate response to arterial baroreflex engagement) in animals with autotransplanted hearts suggest that both sympathetic and parasympathetic reinnervation occurs 9 to 12 months after transplantation.1,24,25 Studies in animals with allografted hearts found less consistent evidence for reinnervation, especially when episodes of rejection had occurred.1,24
Assessment of parasympathetic reinnervation in humans has been limited. Most studies have involved small numbers of patients relatively early after transplantation.4–11 A minority describe responses in isolated patients that are consistent with increasing parasympathetic influences on the donor sinus node,4,5,8 but most do not provide consistent evidence for parasympathetic reinnervation. A recent large study, however, measured heart period variability in response to carotid baroreceptor stimulation in 26 heart transplant recipients at 2 to 63 months after transplantation and found no evidence for return of parasympathetic control of heart rate.11 Given the results of these studies and the animal studies described above, experiments were undertaken to examine whether parasympathetic influences on heart rate reappeared in cardiac transplant patients who were <2 and >4 years post-transplantation.
Responses of the DSN
The group mean responses of the DSN for the trigeminal reflex and the arterial baroreflex gains for the DSN were markedly diminished compared with control and with the responses of the RSN (Figs 1B⇑ and 3⇑). Individual subject responses for the trigeminal reflex in the early group were also minimal (Fig 2A⇑). Two subjects in the late group, however, did demonstrate a decrease in the DSN rate with face immersion that was similar to the RSN response in these subjects (see “Results”) as well as similar to the mean response in the control group. One of these subjects also was studied in the arterial baroreflex experiments, but the gain observed for this subject was minimal (0.7 ms/mm Hg) compared with the RSN gain (13.8 ms/mm Hg) and compared with the mean for control subjects (14.9±1.8 ms/mm Hg) (Fig 3⇑). Thus, it was unclear whether an increase in vagal traffic to the DSN was actually achieved in this patient. To further investigate this question, this particular subject had repeat trigeminal reflex testing before and after atropine, as described in “Results.” Unfortunately, repeat testing was performed 6 years after the initial study, and diabetes mellitus had supervened. Trigeminal reflex activation again resulted in slowing of the RSN. This response was abolished after atropine, indicating the presence of an increase in vagal tone. In contrast, the DSN exhibited a small increase in rate both before and after atropine. This response did not confirm the earlier observed rate decrease but was similar to the majority of responses in the late group (Fig 2B⇑). In addition, the response was consistent with the arterial baroreflex result in this patient and suggests no return of vagal control to the donor heart. The effects of the intervening 6-year time period and the development of diabetes mellitus since the initial study on these results are unknown. Overall, the DSN responses confirmed the findings from the previous systematic study in this area11 and found no return of functional parasympathetic control of heart rate after cardiac transplantation in humans.
Effect of Time After Transplantation
We considered the possibility that as time after transplantation lengthened, increasing parasympathetic influences might occur on the DSN. In the trigeminal reflex study for the early group, no such trend was observed (Fig 2A⇑). For the late group, a modest trend was observed (r=−.49, P=.04), suggesting that with time, parasympathetic influences on the DSN may increase (Fig 2B⇑). Baroreflex gains as noted above were uniformly near zero, suggesting no effect of time on these responses (Fig 3⇑). Overall, these data suggest that parasympathetic control of donor heart rate does not return for the majority of patients, even in the long-term post–cardiac transplantation period.
Responses of the RSN
The group mean responses of the RSN for both early and late transplant groups for the trigeminal and arterial baroreflex experiments, although statistically similar to responses for control subjects, tended to be slightly less than control values (Figs 1B⇑ and 3⇑). It is known that parasympathetic control of heart rate is reduced in patients with congestive heart failure.16,26,27 Because the RSN remains innervated after cardiac transplantation,16,17 these results indicated that parasympathetic control to the innervated portion of the heart returns toward normal after transplantation. These findings confirm and extend the results of Smith and Ellenbogen.16,26 Moreover, these findings indicated that the reduced parasympathetic control of the DSN that we observed was due to lack of reinnervation to the DSN and not due to persistently reduced parasympathetic control after correction of congestive heart failure by cardiac transplantation.
Trigeminal Reflex After Renal Transplant
It is known that cyclosporine treatment is accompanied by sustained sympathetic activation, which may be accentuated by cardiac dennervation.18 An additional concern was that cyclosporine might negatively affect parasympathetic control, resulting in the diminished-to-absent responses we observed in the cardiac transplant patients. We therefore tested 11 renal transplant patients receiving immunosuppressive regimens similar to those of the cardiac transplant patients. Group mean heart rate responses to trigeminal reflex testing were similar to those of the control subjects (Fig 1B⇑), suggesting that cyclosporine had little effect on parasympathetic outflow to the sinus node.
Atropine sulfate was not administered routinely in our experiments as a means of confirming efferent parasympathetic traffic. This was not done because (1) the reflexes studied have been well described and are known to result in an increase in parasympathetic outflow and (2) the response of the RSN was measured and found to be similar to control responses, which confirms the presence of efferent parasympathetic outflow for a given subject in the cardiac transplant groups. The response to atropine administered with the initial experiments for the two transplant patients who exhibited DSN slowing would have been helpful in determining whether this response was indeed due to parasympathetic influences.
The data presented in this study suggest that parasympathetic control of donor heart rate was absent in the majority of patients up to 96 months after cardiac transplantation.
This study was supported in part by the American Heart Association, Virginia Affiliate, Research Award VA-91-G-38 and Medical College of Virginia Clinical Research Center, National Institutes of Health Grant M01-RR-00065. We are grateful for the continued encouragement and support of Pramod K. Mohanty, MD, and J.V. Nixon, MD. We thank Dorothy Thompson and Sylvia Converse for preparation of the manuscript; Catherine Murphy, RN, for assistance with experiments; and Marian Schlutz, RN, BSN, and Dan Kirchberg, PA, Cardiac Transplantation, for assistance with recruiting.
- Received April 14, 1997.
- Revision received July 14, 1997.
- Accepted August 1, 1997.
- Copyright © 1997 by American Heart Association
Burke MN, McGinn AL, Homans DC, Christensen BV, Kubo SH, Wilson RF. Evidence for functional sympathetic reinnervation of left ventricle and coronary arteries after orthotopic cardiac transplantation in humans. Circulation. 1995;91:72–78.
Kaye DM, Esler M, Kingwell B, McPherson G, Esmore D, Jennings G. Functional and neurochemical evidence for partial cardiac sympathetic reinnervation after cardiac transplantation in humans. Circulation. 1993;88:1110–1118.
Stinson EB, Griepp RB, Schroeder JS, Dong E, Shumway NE. Hemodynamic observations one and two years after cardiac transplantation in man. Circulation. 1972;45:1183–1194.
Bernardi L, Bianchini B, Spadacini G, Leuzzi S, Valle F, Marchesi E, Passino C, Calciati A, Vigano M, Rinaldi M. Demonstrable cardiac reinnervation after human heart transplantation by carotid baroreflex modulation of RR interval. Circulation. 1995;92:2895–2903.
Webb SW, Adgey AAJ, Pantridge JF. Autonomic disturbances at onset of acute myocardial infarction. Br Med J. 1972;3:89–92.
Lewis T. Vasovagal syncope and the carotid sinus mechanism. Br Med J. 1932;1:873–874.
Farrell TG, Odemuyiwa O, Bashir A, Cripps TR, Malik M, Ward DE, Camm AJ. Prognostic value of baroreflex sensitivity testing after acute myocardial infarction. Br Heart J. 1992;67:129–137.
Schwartz PJ, Vanoli E, Stramba-Badiale M, De Ferrari GM, Billman GE, Foreman RD. Autonomic mechanisms and sudden death: new insights from analysis of baroreceptor reflexes in conscious dogs with and without a myocardial infarction. Circulation. 1988;78:969–979.
Ellenbogen KA, Mohanty PK, Szentpetery S, Thames MD. Arterial baroreflex abnormalities in heart failure: reversal after orthotopic cardiac transplantation. Circulation. 1989;79:51–58.
Andersen HT. Physiological adaptations in diving vertebrates. Physiol Rev. 1966;46:212–243.
Olsen CR, Fanestil DD, Scholander PF. Some effects of breath holding and apneic underwater diving on cardiac rhythm in man. J Appl Physiol. 1962;17:461–466.
Kawakami Y, Natelson BH, DuBois AB. Cardiovascular effects offace immersion and factors affecting diving reflex in man. J Appl Physiol. 1967;23:964–970.
Levy MN, Martin PJ. Neural control of the heart. In Berne RM, Sperelakis N, eds. Handbook of Physiology: Section 2: Cardiovascular System. Bethesda, Md: American Physiological Society; 1979:581–620.
Kaye MP, Randall WC, Hageman GR, Geis WP, Priola DV. Chronology and mode of reinnervation of the surgically denervated canine heart: functional and chemical correlates. Am J Physiol. 1977;233:H431–H437.
Eckberg DL, Drabinsky M, Braunwald E. Defective cardiac parasympathetic control in patients with heart disease. N Engl J Med. 1971;285:877–883.