Author + information
- Received March 23, 2004
- Revision received May 28, 2004
- Accepted June 7, 2004
- Published online September 15, 2004.
- Anne B. Taegtmeyer, BMBCh*,†,
- Angela M. Crook, MSc‡,
- Paul J.R. Barton, PhD† and
- Nicholas R. Banner, FRCP*,* ()
- ↵*Reprint requests and correspondence:
Dr. Nicholas R. Banner, Royal Brompton and Harefield NHS Trust, Hill End Road, Harefield, Middlesex UB9 6JH, United Kingdom
Objectives This study was designed to test the hypothesis that heterotopic heart transplant (HHT) patients have lower blood pressure than orthotopic cardiac transplant (OCT) patients because their native heart is involved in blood pressure homeostasis.
Background Hypertension occurs more frequently after OCT than after liver or lung transplantation, suggesting that transplantation of the heart itself contributes to post-transplant hypertension.
Methods Blood pressure and related measurements in 233 OCT and 38 HHT patients were studied retrospectively post-transplant.
Results Systolic blood pressure (SBP) was persistently lower among HHT patients (means 121 vs. 137, 126 vs. 137, 125 vs. 139, and 128 vs. 143 mm Hg at month 3 and years 1, 3, and 5 respectively, p < 0.005). Left ventricular and aortic systolic pressures were also lower (130 vs. 143 mm Hg, p = 0.01 and 129 vs. 142 mm Hg, p = 0.01). Multivariable analysis with age, gender, body mass index, creatinine, steroids, cyclosporine, use of antihypertensive medication, donor left ventricular ejection fraction, donor weight, and type of transplant as covariables showed HHT to be independently associated with a lower SBP at each time point (beta-coefficients −16.2, −12.1, −13.3, and −14.2 mm Hg, p < 0.01). The adjusted hazard ratio for the development of systolic hypertension among HHT compared with OCT patients was 0.59 (95% confidence interval 0.39 to 0.91, p = 0.017).
Conclusions Heterotopic heart transplant patients had lower SBP than OCT patients, consistent with the hypothesis that the native heart continues to contribute to blood pressure homeostasis.
Hypertension develops early after cardiac transplantation and can be difficult to manage (1,2). Cyclosporine contributes to post-transplant hypertension through activation of the sympathetic nervous system (3), nephrotoxicity (4), and inhibition of endothelium-dependent vasodilation (5). However, other factors such as cardiac innervation may play a role, because the incidence of hypertension in liver (45%) and lung (66%) transplant patients receiving cyclosporine is lower than in heart transplant patients (95% at 5 years) (1,6,7).
Evidence for the role of cardiac innervation in cardiovascular homeostasis comes from human and animal studies. Autotransplantation in dogs results in increased plasma volume (8), and human heart transplant recipients have blunted responses to salt or volume loading (9). They also respond to central blood volume reduction with an attenuated increase in sympathetic activity (10).
Two types of heart transplantation are performed. The orthotopic cardiac transplant (OCT), in which the recipient's heart is removed and replaced by a denervated donor heart, is the commonest. An alternative is the heterotopic heart transplant (HHT), in which the recipient heart and its innervation remain intact and the donor heart is placed in the right hemithorax with the donor and recipient left ventricles functioning in parallel (11). Heterotopic heart transplant is often performed when the donor organ is smaller than the recipient heart or when pulmonary artery pressures are elevated (12).
We hypothesized that because HHT leaves the native heart intact, its use may be associated with less hypertension than is OCT.
Clinical records and echocardiograms of 271 adult cardiac transplant patients transplanted at our institution between 1991 and 1999 who survived longer than three months were examined. Of these, 233 had undergone OCT and 38 HHT. The ethics committee gave approval.
The OCT and HHT groups were well matched in terms of age, gender, size, indication for transplantation, pre-transplant blood pressure, use of angiotensin-converting enzyme inhibitors, previous history of hypertension, and pre-transplant renal function calculated according to Cockcroft and Gualt (13) (Table 1).As expected, donor weight was lower in HHT than in OCT patients (mean 36 vs. 64 kg, p < 0.0001). Of the 38 HHT patients, 23 underwent HHT because of donor-recipient size mismatching and 15 because of elevated systolic pulmonary artery pressures. The latter patients' pre-transplant systolic blood pressures did not differ significantly from those of patients who received HHT on the basis of size mismatch alone (109 ± 19 mm Hg vs. 107 ± 12 mm Hg).
All patients received triple-therapy immunosuppression (cyclosporine, azathioprine, and corticosteroids). Patients unable to tolerate cyclosporine, or patients with several early rejection episodes, were converted to tacrolimus. When possible, steroids were reduced and eventually withdrawn from three months after transplant. Antihypertensive agents were angiotensin-converting enzyme inhibitors, doxazosin, calcium channel blockers, diuretics and, rarely, beta-blockers. Both OCT and HHT patients were treated if their blood pressure was persistently >140/90 mm Hg (14).
Data were available for 253, 209, and 126 patients at years 1, 3, and 5 respectively, owing to death or follow-up <1, 3, or 5 years. Figure 1shows survival beyond three months. Survival after HHT was slightly worse than after OCT, related in part to donor-recipient size mismatch (12).
Noninvasive blood pressure was measured with an oscillometric technique using the Datascope Accutorr Plus (Montvale, New Jersey), recommended by the European Society of Hypertension (15). Readings >140 mm Hg were measured again for confirmation. Native aortic valve opening in HHT patients was assessed by echocardiography. Hemodynamic data measured at cardiac catheterization were available for 46 OCT and 24 HHT cases. Acute rejection was diagnosed histologically and graded according to International Society of Heart and Lung Transplantation criteria (16). Rejection episodes were treated with augmented immunosuppression (three daily doses of 1 g methylprednisolone). Rabbit antithymocyte globulin and/or OKT3 were also used in cases of rejection associated with hemodynamic compromise.
For univariate analyses, two-tailed Student ttest was employed to test for differences between means and Pearson's chi-square or Fisher exact test (if expected frequencies were five or less) to determine differences in proportions. Paired ttests were employed for analysis of matched data.
Multivariable regression models were constructed with SBP as the dependent variable using a stepwise backwards method. Results were expressed as beta-coefficients and a p value of <0.05 was considered significant.
Survival analysis was performed using life tables and Kaplan-Meier methods and the log-rank test was used to determine significance. The development of hypertension (SBP ≥140 mm Hg or treatment with antihypertensive agent[s]) over time was also examined using the Kaplan-Meier technique (17); Cox method was used to generate a hazard ratio (18). Analyses were performed using STATA 7 (Stata Corp., College Station, Texas).
Systolic blood pressure was lower in HHT than in OCT patients at all time points (Fig. 2,Table 2). When corrected for multiple testing (Bonferroni correction), all p values remained <0.05. Other than at one year, diastolic blood pressure did not differ between the two groups. Invasively measured left ventricular and aortic systolic blood pressures were also lower in HHT patients (Table 3).
Heterotopic heart transplant patients required fewer antihypertensive agents than OCT patients, particularly at month 3 and year 1. Thus, the higher SBPs seen in OCT patients were in spite of greater use of antihypertensive agents.
Use of tacrolimus (associated with less hypertension than cyclosporin) (19) was similar between the two groups. Acute rejection and treatment with augmented immunosuppression occurred more frequently among HHT patients (mean number of rejection episodes greater than grade 1B 2.13 vs. 1.53, p < 0.05) (Table 2). Heterotopic heart transplant patients also received high-dose steroids more often than did OCT patients (mean number of treatments 2.68 vs. 1.94, p = 0.012) (Table 2).
In keeping with heterotopic grafts and their stroke volumes being smaller than orthotopic grafts, cardiac index (CI) was significantly lower among HHT compared with OCT patients (Table 3). Because a limited number of CI measurements were available (28 OCT and 12 HHT at year 1 or 3), CI was not included in the multivariable analysis. Subjects for whom CI data were available were also analyzed separately: CI and SBP were both lower among HHT patients, even when age, gender, and year after transplantation were matched between HHT and OCT subjects (mean CI 2.02 ± 0.38 l/m/m2vs. 2.97 ± 0.5 l/m/m2[paired ttest p = 0.006], mean SBP 124 ± 13 mm Hg vs. 142 ± 16 mm Hg [paired ttest p = 0.002]). After matching HHT and OCT patients on the basis of CI, SBP was still lower in HHT patients than in OCT patients: mean SBP 127 ± 13 mm Hg vs. 143 ± 20 mm Hg (p = 0.02) in 13 HHT and 13 OCT patients matched for CI (mean CI 2.22 ± 0.31 and 2.2 ± 0.32), indicating that the lower SBP seen among HHT patients was not due to lower CI. Subgroups of HHT patients were examined at year 1 and results are shown in Table 4.
Adjusted beta coefficients for the effect of HHT compared to OCT on SBP are shown in Table 5and did not differ greatly from unadjusted values. In patients receiving cyclosporine, 12 h post-dose cyclosporine levels did not significantly affect SBP (data not shown).
The proportion of patients with systolic hypertension was always less in HHT than OCT patients (Fig. 3).The hazard ratio for the development of hypertension (adjusted for age, gender, and body mass index using the Cox method) in HHT compared with OCT patients was 0.59 (95% confidence interval 0.39 to 0.91, p = 0.017).
Figure 4shows SBP in two patients who underwent HHT followed some months later by OCT in the native heart position. The findings are discussed later.
The greater incidence of hypertension among OCT recipients may be due to the loss of cardiac inputs to blood pressure homeostasis. In keeping with this hypothesis, we found SBP to be lower in HHT patients (whose native hearts remain in situ) than in OCT patients until five years after transplant.
Possible mechanisms by which the native heart contributes to blood pressure homeostasis are through cardiopulmonary baroreceptors and the release of natriuretic peptides. Cardiopulmonary baroreceptors are located within the myocardium and provide tonic inhibition of sympathetic outflow to the heart and peripheral circulation, and lower blood pressure when filling volumes are adequate (20). When this tonic inhibitory input is disrupted, baroreflexes are impaired (10). Natriuretic peptides are released in response to myocardial stretch or when the myocardium has undergone remodeling and actions include natriuresis, diuresis, and vasodilation (21). Both atrial natriuretic peptide and brain-type natriuretic peptide concentrations are elevated after OCT (22). As HHT patients possess two hearts, they may have higher natriuretic peptide levels.
Vascular pathology could predispose to the development of post-transplant hypertension; however, when post-transplant blood pressures in patients with pre-transplant ischemic heart disease was compared with blood pressure in patients of other pre-transplant diagnoses, there was no difference (data not shown).
Arterial blood pressure in a HHT recipient could be influenced by the parallel function of the donor and native left ventricles. In practice, however, the native left ventricle of patients who have undergone HHT usually does not contribute significantly to cardiac output, and the native aortic valve often does not open at all (23). Subgroup analysis of HHT patients to determine whether aortic valve opening might have contributed to the lower SBP showed there was no difference in SBP between those whose aortic valve opened and those whose valve did not. Similarly, paced linkage (24) did not cause any difference in SBP. These observations support the view that the lower blood pressure observed in the heterotopic group was not due to HHT itself, but was related to the presence of the native heart.
Diastolic blood pressure did not differ significantly between the two groups. This could result from the physiology of HHT where asynchronous beating of donor and recipient hearts prevents blood pressure from returning to minimum after donor diastole because recipient systole has already commenced.
The heterotopic transplant patients as a whole experienced more rejection episodes requiring treatment with augmented immunosuppression than the OCT group. This may be due to the fact that more HHT patients had been weaned from maintenance steroids by three months after transplantation (41% of HHT patients vs. 27% of OCT patients) (Table 2).
Analysis of hemodynamic data for a subgroup of patients demonstrated that CI was lower in the heterotopic patients, probably owing to the smaller size of the donor heart and a greater incidence of rejection in the HHT group. However, a comparison of HHT and OCT patients matched for CI still demonstrated the difference in SBP.
In two cases, we had the opportunity to examine blood pressure after HHT and then after OCT in the same patient. As the heterotopic organ remained in situ after OCT, this resulted in two denervated hearts. Immunosuppression was with cyclosporine throughout and both patients experienced a rise in SBP after OCT (Fig. 4).
The retrospective nature of the study means that complete hemodynamic data were not available for all patients and natriuretic peptide levels were not measured. Differences in donor-recipient size mismatching between the OCT and HHT groups mean they were not directly comparable; however, donor weight was included in the multivariable model. Although treated blood pressure was studied, the difference in SBP was consistent at all time points and the number of antihypertensive agents used was included in the multivariable model. Also, the higher SBP seen in OCT patients was despite greater treatment with antihypertensive agents.
Further studies are required to determine the exact mechanism by which the native heart acts to reduce blood pressure in the HHT group.
Heterotopic heart transplant patients had lower SBP, measured both invasively and noninvasively than OCT patients, managed with the same immunosuppression protocol in the same center. Multivariable analysis confirmed that HHT was independently associated with lower SBP compared to OCT. This is consistent with the hypothesis that the native heart continues to play a role in blood pressure homeostasis after HHT and that its removal in OCT contributes to the development of post transplant hypertension.
The authors thank Jane Breen, Paula Rogers, and Michael Octave for assistance with data collection and analysis.
This study was supported by the Harefield Research Foundation and the Royal Brompton and Harefield NHS Trust.
- Abbreviations and acronyms
- cardiac index
- heterotopic heart transplant
- orthotopic cardiac transplant
- systolic blood pressure
- Received March 23, 2004.
- Revision received May 28, 2004.
- Accepted June 7, 2004.
- American College of Cardiology Foundation
- Ozdogan E.,
- Banner N.,
- Fitzgerald M.,
- Musumeci F.,
- Khaghani A.,
- Yacoub M.
- Parent R.,
- Stanley P.,
- Chartrand C.
- Mohanty P.K.,
- Thames M.D.,
- Arrowood J.,
- Sowers J.R.,
- McNamara C.,
- Szentpetry S.
- Banner N.R.,
- Khaghani A.,
- Fitzgerald M., et al
- O'Brien E.,
- Wateber B.,
- Parati G.,
- Staessen J.,
- Myers G.
- Cox D.R.
- Mancia G.,
- Donald D.