Author + information
- Received October 22, 1999
- Revision received February 22, 2000
- Accepted April 5, 2000
- Published online August 1, 2000.
- Randy W Braith, PhD∗,* (, )
- Roger M Mills, MD, FACC§,
- Christopher S Wilcox, MD, PhD∥,
- Matthew J Mitchell, MS∗,
- James A Hill, MD, FACC† and
- Charles E Wood, PhD‡
- ↵*Reprint requests and correspondence: Dr. Randy W. Braith, P.O. Box 118206, Center for Exercise Science, University of Florida, Gainesville, Florida 32611
We sought to test the hypothesis that plasma volume (PV) expansion in heart transplant recipients (HTRs) is caused by failure to reflexively suppress the renin-angiotensin-aldosterone (RAA) axis.
Extracellular fluid volume expansion occurs in clinically stable HTRs who become hypertensive. We have previously demonstrated that the RAA axis is not reflexively suppressed by a hypervolemic stimulus in HTRs.
Plasma volume and fluid regulatory hormones were measured in eight HTRs (57 ± 6 years old) both before and after treatment with captopril (225 mg/day). Antihypertensive and diuretic agents were discontinued 10 days before. The HTRs were admitted to the Clinical Research Center (CRC), and, after three days of a constant diet containing 87 mEq/day of Na+, PV was measured by using the modified Evans blue dye dilution technique. After approximately four months (16 ± 5 weeks), the same HTRs again discontinued all antihypertensive and diuretic agents; they were progressed to a captopril dose of 75 mg three times per day over 14 days, and the CRC protocol was repeated.
Captopril pharmacologically suppressed (p < 0.05) supine rest levels of angiotensin II (−65%) and aldosterone (−75%). The reductions in vasopressin and atrial natriuretic peptide levels after captopril did not reach statistical significance. The PV, normalized for body weight (ml/kg), was significantly reduced by 12% when the HTRs received captopril.
Extracellular fluid volume is expanded (12%) in clinically stable HTRs who become hypertensive. Pharmacologic suppression of the RAA axis with high-dose captopril (225 mg/day) returned HTRs to a normovolemic state. These findings indicate that fluid retention is partly engendered by a failure to reflexively suppress the RAA axis when HTRs become hypervolemic.
Extracellular fluid volume expansion is well documented in clinically stable heart transplant recipients (HTRs) who become hypertensive (1–3). The mechanism of fluid retention in HTRs is not completely understood but has previously been attributed to the effects of the immunosuppressant agent cyclosporine. Indeed, both cyclosporine-mediated sympathoexcitation and nephrotoxicity could alter salt and water handling in these patients (4,5). Recent evidence suggests, however, that long-term fluid retention in HTRs may extend beyond cyclosporine-related mechanisms (1–3,6–8).
An alternative hypothesis is that denervation of cardiac volume receptors is the mechanism responsible for fluid retention in HTRs. In HTRs, arginine vasopressin (AVP) levels are not suppressed during the supine posture (6); plasma renin activity is not suppressed by water immersion (7); and abrupt volume expansion with isotonic saline infusion does not elicit suppression of the renin-angiotensin-aldosterone (RAA) system (8). In contrast, all of these maneuvers reflexively suppress AVP, plasma renin activity and the RAA axis in subjects with intact cardiac innervation. In a previous attempt to evaluate the role of cyclosporine in fluid retention, we measured plasma volume (PV) in both HTRs (n = 11) and in a cyclosporine control group consisting of age-matched liver transplant recipients (n = 6) (2). The HTRs demonstrated moderate PV expansion (14%), but the liver transplant recipients were normovolemic, despite receiving equivalent amounts of cyclosporine. The disparity in extracellular fluid volume between heart and liver transplant recipients was not accompanied by significant differences in either rest left ventricular or renal function between the two transplant groups (2).
The purpose of this study was to determine if long-term extracellular fluid volume expansion in HTRs is caused by an inability to suppress reflexively the RAA system when patients become hypervolemic. We speculated that pharmacologic suppression of the RAA axis could return HTRs to a normovolemic state. Accordingly, we employed a cross-over design and measured PV and fluid-regulatory hormones in HTRs under carefully standardized conditions both before and after high dose angiotensin-converting enzyme (ACE) inhibition with captopril.
Eight male HTRs (57 ± 6 years old) were entered into the study 22 ± 4 months (group mean) after transplantation. The HTRs were clinically stable and free of significant rejection, infection or other major illness, and all received immunosuppressive therapy with cyclosporine, prednisone and azathioprine. The HTRs were receiving maintenance doses of prednisone (5 to 10 mg/day) at the time of the study, and none had required enhanced corticosteroid doses within six months of the study. All HTRs had biatrial anastomosis at transplantation. The HTRs underwent endomyocardial biopsy within two months of the study, and there was no evidence of rejection at the time of biopsy in any of the subjects. The whole-blood cyclosporine trough level, calculated as an average of four determinations over six months before the study, was 225 ± 18 ng/ml (group mean). None of the transplant recipients were hypertensive before transplantation, but all required antihypertensive agents for management of hypertension after transplantation. None of the HTRs had evidence of cardiac allograft vascular disease.
This study was a cross-over design performed in the Clinical Research Center (CRC) at Shands Hospital at the University of Florida. Extracellular fluid volume and fluid-regulatory hormones were measured in the same cohort of HTRs under two conditions: visit 1) after discontinuation of all diuretic and antihypertensive agents; and visit 2) after a 14-day regimen of ACE inhibition consisting of a target dose of 225 mg/day of captopril. The protocol was approved by the Institutional Review Board for the protection of human subjects at the University of Florida, and all subjects provided written, informed consent to participate in the study.
Protocol without captopril
Diuretics and all antihypertensive agents were discontinued 10 days before the study. No alpha- or beta-blockers or other cardiac medications were used by the HTRs. The subjects did not add salt to their usual diet for four days before admission to the hospital. Thereafter, they were admitted to the CRC. The first three days in the CRC were an equilibration to the controlled diet, which consisted of sufficient calories to maintain the subjects’ current weight and provided 87 mEq/day of Na+ and 80 mEq/day of K+. This diet was served as three meals taken at 8:00 am, 12:00 pm and 5:00 pm. Water intake was standardized to 1,000 ml on day 3, but was provided as desired on all other days of the study. Consumption of alcohol, caffeine and tobacco products was not allowed.
Subjects were awakened each morning at 7:00 am, and while still supine, blood pressure was recorded in triplicate using an automated system (Datascope Corp., Paramus, New Jersey). Blood samples were drawn from an indwelling venous catheter for analysis of plasma concentrations of AVP, angiotensin II, aldosterone and atrial natriuretic peptide (ANP). A 24-h urine collection was started each morning at 7:00 am. All subjects were within 15% of Na+ balance before the PV measurements were taken on the morning of day 4.
On day 4, PV was measured. At 7:00 am, the subjects were awakened and blood samples were obtained for determination of fluid balance hormones and hematocrit; renal creatinine clearance was also determined. The subjects then stood and voided and returned to a supine posture. After 60 min of supine rest, the PV measurement was taken (9). After a 48-h period of observation, diuretic and antihypertensive agents were restarted, and the subjects were discharged from the CRC.
Protocol with captopril
After approximately four months (group mean 16 ± 5 weeks), the same cohort of HTRs (n = 8) again discontinued all diuretic and antihypertensive agents. The subjects were then progressed to a captopril target dose of 75 mg three times per day for 14 days and were admitted a second time to the CRC. The equilibration protocol and PV measurements described earlier were repeated for each subject.
Blood sample collection
Blood samples for PV measurements were drawn into vacutainers containing heparin. Blood samples for AVP, angiotensin II, aldosterone and ANP were drawn into vacutainers containing EDTA. Plasma was separated by centrifugation at 3,500 × g at 4°C for 20 min and the samples were frozen at −80°C until the assays were performed.
Plasma volume was determined by using a modified Evans blue dye (T-1824) dilution technique (9). Plasma samples were drawn at 10, 20 and 30 min after injection of blue dye, and the data were extrapolated to zero-time in calculating the PV. The dye from the plasma samples was extracted onto a wood-cellulose powder (Solka Floc SW 40A) chromatographic column after it had been separated from albumin by the action of detergent (Teepol 610 in 2% Na2HPO4). Interfering substances such as pigments, proteins and chylomicrons were washed from the column with 2% Na2HPO4. The dye was then eluted from the column with a 1:1 acetone-water mixture. The addition of KH2PO4 buffered the pH of the eluate to 7.0; absorbance of the eluate was read at 615 nm. Duplicate microhematocrit (Hct) determinations were made using a microhematocrit centrifuge (IEC, model MB, Needham Heights, Massachusetts) and a IEC Micro-Capillary Tube Reader. Raw Hct values were corrected (×0.91) for trapped plasma and for whole-body Hct (10). Blood volume was calculated as PV/(1 − corrected Hct).
The PV and neurohormone assays used in the present study have been used reliably in our laboratory for numerous experiments in human subjects (2,3,8,11,12). Plasma AVP was measured by radioimmunoassay, as previously described (11). Plasma angiotensin II and aldosterone levels were determined by radioimmunoassay, as previously described by Braith et al. (12). Atrial natriuretic peptide was extracted from plasma using a modification of a technique also described by Braith et al. (8). Plasma (1 ml) was deproteinized by adding 750 μl of 0.1 mol/liter acetic acid and 1.25 ml methanol. The samples were placed on a rocking shaker for 10 min, followed by centrifugation for 20 min at 6,000 rpm at 4°C. The supernatant was dried by vacuum centrifugation. Radioimmunoassay was performed with a kit from Peninsula Laboratories (Belmont, California) by using ANP antiserum that has 0% cross-reactivity with human brain natriuretic peptide and C-type natriuretic peptide.
The Na+ and K+ concentrations in urine and plasma were measured using a Nova-1 electrolyte analyzer (Nova Biomedical, Waltham, Massachusetts). Plasma osmolality was measured with a vapor pressure osmometer (Wescor Inc., Logan, Utah). Standard methods were used for the measurement of serum creatinine and the calculation of creatinine clearance.
Hemodynamic data and left ventricular function
To evaluate the possible contribution of cardiac function on fluid volume status, the subjects underwent right heart catheterization using standard thermodilution techniques and two-dimensional echocardiography. These examinations were all performed during the interval between the two CRC admissions.
Plasma volume, neurohormone and creatinine values before and after ACE inhibition were compared using intragroup analysis of variance (ANOVA), with only two levels of classification (with and without ACE inhibition). Plasma volume values in HTRs before ACE inhibition and, after it, in age-matched healthy control subjects and in age-matched liver transplant recipients were compared using ANOVA and the Scheffé post-hoc multiple comparisons procedure. The ANOVAs were performed using the SAS general linear model procedure (SAS Institute Inc., Cary, North Carolina). An alpha level of p < 0.05 was required for statistical significance.
There were no changes in clinical status for any HTR between the two admissions to the CRC. There were no significant changes in body weight, and the immunosuppression regimen was not changed for any subject during the study.
The hemodynamic and left ventricular function data are presented in Table 1. The mean cardiac index was 2.80 ± 0.53 liters/min per m2, which falls within the lower one-third of the normal range of 2.4 to 4.2 liters/min per m2(13). Right atrial mean pressure for the group was slightly elevated at 6.1 ± 1.0 mm Hg, which was 22% greater than the accepted maximal right atrial mean pressure of 5 mm Hg, whereas the mean occluded pulmonary artery pressure of 10.3 ± 3.2 mm Hg fell within the stated normal range of 2 to 12 mm Hg (13). These observations are consistent with other investigators’ suggestions that the cardiac allograft exhibits mild or occult diastolic dysfunction, particularly of the right ventricle (14,15).
Blood volume, PV and electrolyte data are presented in Table 2. Blood volume and PV, normalized for body weight (ml/kg), were significantly reduced by 12% when the HTRs were receiving captopril, as compared with the identical CRC protocol and measurements performed when HTRs were not receiving captopril. The reduction in PV observed when HTRs were receiving high dose captopril was not associated with significantly different concentrations in plasma Na+ or plasma osmolality.
Urine volume and urinary salt excretion during the 24 h preceding the PV measurements were not significantly different (p > 0.05) before (0.9 ± 0.2 ml/kg per h; 3.6 ± 0.2 mEq/h) versus after captopril (0.8 ± 0.3 ml/kg per h; 3.4 ± 0.2 mEq/h). Likewise, serum creatinine levels and 24-h creatinine clearance were not significantly different before (1.6 ± 0.4 mg/dl; 80.0 ± 23 ml/min) versus after captopril (1.5 ± 0.5 mg/dl; 82.0 ± 19 ml/min).
The neurohormone data are also presented in Table 2. Supine angiotensin II and aldosterone levels were significantly (p < 0.05) greater before versus after captopril, with levels suppressed by 65% and 75%, respectively. The changes in AVP and ANP levels after captopril did not reach statistical significance (p > 0.05). The elevated angiotensin II and aldosterone levels before captopril, despite PV expansion (12%), suggest that the RAA axis in HTRs is not suppressed by a hypervolemic stimulus.
Our data confirm previous reports that extracellular fluid volume is expanded (12%) in HTRs (1–3,8). The results support the hypothesis that fluid retention in HTRs is partly engendered by failure to reflexively suppress the RAA axis when patients become volume expanded. Pharmacologic suppression of the RAA axis with high dose captopril (225 mg/day) returned HTRs to a euvolemic state. This conclusion is strengthened by the cross-over study design with the same cohort of HTRs studied before and after high dose ACE inhibition, with each subject serving as his or her own control, as well as the standardization of dietary salt, water and caloric intake for three days before the measurements.
Extracellular fluid volume in health and disease
In an attempt to interpret the physiologic significance of the relative reduction in fluid volume in our cohort of HTRs after captopril administration, we compared the PV levels of the present study with those levels in two reference groups previously studied in our laboratory under identical conditions (without captopril): 1) healthy age-matched subjects; and 2) age-matched liver transplant recipients (Fig. 1). The regimen of high dose ACE inhibition in HTRs contracted PV to euvolemic levels that are essentially the same (p > 0.05) as those observed in normal control subjects with intact cardiac innervation and in orthotopic liver transplant recipients with intact cardiac innervation who were immunosuppressed with cyclosporine.
Hemodynamic data and left ventricular function
Abnormal cardiac allograft systolic function is one possible alternative explanation for expanded extracellular fluid volume in HTRs. Although clinical evidence suggests occult diastolic dysfunction in long-term, clinically stable heart transplant recipients (14,15), previously published data from our laboratory (2,3,16), other published data (14,15) and the hemodynamic data for the subjects of this study all indicate that cardiac index at rest is maintained within the normal range, albeit generally toward the lower one-third of the normal range. The echocardiographic findings in the present cohort did not indicate abnormal hypertrophy and confirmed normal left ventricular systolic function (Table 1). Thus, with normal systolic function, the expanded extracellular fluid volume that we observed in HTRs before captopril likely cannot be attributed to differences in cardiac output. However, our hemodynamic and fluid homeostasis measurements were not simultaneous. Rather, the hemodynamic assessments in HTRs were performed during the interval (16 ± 5 weeks; group mean ± SD) between the first (without captopril) and second (with captopril) fluid balance measurements, and therefore somewhat limit the generalization of cardiac hemodynamic data to the physiologic state of the patient at the specific times that fluid volume was studied.
Cyclosporine and renal function
Cyclosporine-induced nephrotoxicity, including cyclosporine-mediated increases in renal vascular resistance and endothelial dysfunction (17), is another possible explanation for expanded extracellular fluid volume in HTRs. Indeed, most reports of fluid retention in organ transplant recipients have emphasized cyclosporine-induced nephrotoxicity (5,7,17). The adverse effects of cyclosporine include an increase in renal vascular resistance and a concomitant decrease in the glomerular filtration rate and renal blood flow; these effects are clearly associated with fluid retention. In the present study, we relied on the semiquantitative estimate of the glomerular filtration rate by using endogenous creatinine clearance. Although this technique may not be sufficiently sensitive to detect small differences in filtration in HTRs before and after captopril, our data indicate that the glomerular filtration rate was the same in our cohort of HTRs before and after high dose ACE inhibition. Our finding of normal extracellular fluid volume in HTRs after captopril, despite an unchanged cyclosporine dose and cyclosporine trough levels, clearly shows that the cyclosporine effect alone is an inadequate explanation for the fluid volume expansion detected in HTRs. Rather, our findings suggest that the moderate fluid retention that we have described in HTRs is derived from the combination of a deafferented heart (not allowing us to detect fluid volume expansion and appropriately suppress the RAA axis) and the use of cyclosporine, which further attenuates the abnormal renal salt and water handling.
Cardiac denervation and neuroendocrine function
Before captopril, plasma levels of angiotensin II, aldosterone and AVP in HTRs were within the normal range for subjects measured in our laboratory in the recumbent position. This finding was surprising, however, because the observed 12% blood volume expansion in HTRs before captopril could be expected to suppress angiotensin II, aldosterone and AVP. In various mammalian species, hypervolemia leads promptly to renal vasodilation, diuresis and natriuresis (18–21). These renal responses are mediated, in part, by cardiac mechanoreceptors located in atrial and ventricular chambers, which play a central role in sensing changes in blood volume (19–21). Normally, innervated cardiac mechanoreceptors respond to hypervolemia by suppressing AVP, the RAA axis, thirst and sympathetic traffic to the kidney (19–21). In concert, these compensatory responses increase diuresis and natriuresis, which return extracellular fluid volume to the euvoluemic state. Surgical denervation of cardiac afferent fibers in animal models is known to attenuate and delay both the diuretic and natriuretic response to hypervolemia (22–26).
Failure to reflexively suppress the RAA axis when HTRs become hypervolemic may be responsible, in part, for the unique severity of de novo post-transplantation hypertension in some HTRs. In the present study, we observed a PV increase of 12% when, theoretically, a fluid volume increase of only 3% to 4% can result in sustained hypertension (27). In a previous study, we examined the arterial blood pressure and renal and endocrine responses to abrupt volume expansion in HTRs (8). Infusion of 0.154 mol/liter saline at 8 ml/kg per h for 4 h elicited a hypertensive response (15 ± 8/8 ± 5 mm Hg), which persisted for 48 h. In addition, saline infusion did not reflexively suppress AVP, angiotensin II and aldosteone in HTRs. Net urine flow and urinary Na+ excretion were blunted and delayed in HTRs, resulting in elimination of only 51% of the Na+ load over 48 h, which may be caused by failure to reflexively suppress fluid-regulatory hormones. These defects in blood pressure and fluid homeostasis were not seen in age-matched normal control subjects or in liver transplant recipients who were also immunosuppressed with cyclosporine (8).
Diuretic therapy may be ineffective at mitigating fluid retention in HTRs. Previous studies of hydrochlorothiazide and furosemide therapy for hypertension have shown that initial decreases in PV, glomerular filtration rate and renal blood flow return to pretreatment values within 6 to 12 weeks of continuous therapy (28). The first dose of a diuretic agent increases the urine flow rate and renal Na+ excretion and decreases body weight. However, with repeated administration, progressive losses of salt and water are soon curtailed, and the reduced body weight stabilizes (29). This “diuretic braking phenomenon” appears to be independent of the class of diuretic agent given and is achieved by a combination of renal Na+ retention in the postdiuretic period and by resistance to the natriuretic response to the diuretic agent (30).
This study was limited by a small sample size and the sequential order in which HTRs had their two in-patient fluid volume studies. However, the small sample and nonrandomized order of study patients with and without captopril is compensated by the cross-over design of the study, wherein each subject served as their own control.
We demonstrated, in a small cohort of HTRs, that extracellular fluid volume is expanded (12%) in HTRs, and that fluid retention is partly engendered by a failure to suppress the RAA axis when patients become hypervolemic. Pharmacologic suppression of the RAA axis with high dose captopril (225 mg/day) returned HTRs to a euvolemic state. Poor adaptation of the RAA axis to fluid retention may be partly responsible for the incidence and severity of post-transplant hypertension in some HTRs.
☆ This work was supported by Clinical Research Center Grant RR00082 from the National Institutes of Health, Bethesda, Maryland; National Research Service Award HL08777 (Dr. Braith); and a grant from Novartis, East Hanover, New Jersey (Dr. Braith).
- angiotensin-converting enzyme
- analysis of variance
- atrial natriuretic peptide
- arginine vasopressin
- Clinical Research Center
- heart transplant recipients
- plasma volume
- Received October 22, 1999.
- Revision received February 22, 2000.
- Accepted April 5, 2000.
- American College of Cardiology
- Braith R.W,
- Mills R.M,
- Wilcox C.S,
- Davis G.L,
- Wood C.E
- Greenleaf J.E,
- Convertino V.A,
- Mangseth G.R
- Raff H,
- Wood C.E,
- Kane C.W
- Braith R.W,
- Wood C.E,
- Limacher M.C,
- et al.
- Nichols W.W,
- Nicholini F.A
- Young J.B,
- Winters W.L,
- Bourge R,
- Uretsky B.F
- Braith RW, Limacher MC, Mills RM, Leggett SH, Pollock ML, Staples ED. Exercise-induced hypoxemia in heart transplant recipients. J Am Coll Cardiol 19934;22:768–76.
- Henry J.O,
- Gauer J,
- Reeves J
- Johnson J,
- Ward A,
- Moore W,
- Segar W
- Lee M.E,
- Thrasher T.N,
- Ramsay D.J
- Gilmore J.P,
- Daggett W.M
- Goetz KL, Fater DC, Schultz HD, Sundet WD. Diuretic response to left atrial distension is eliminated after cardiac denervation. In: Takacs L, editor. Kidney and Body Fluids. Adv Physiol Sci 1981;11:643–7.
- Vatner S.F,
- Manders W.T,
- Knight D.R
- Guyton A.C