The normal circadian variation in blood pressure and heart rate is characterized by a maximum value at midmorning, decreasing throughout the remainder of the day, reaching a nadir in the early-morning sleep hours, and increasing abruptly just before waking (1). The mechanisms that control circadian variation in normal subjects include the degree of physical activity, posture, baroreflexes, and the autonomic nervous system (2). Sympathovagal balance is an important determinant of blood pressure and heart rate, and increased sympathetic activity and vagal withdrawal are associated with the early-morning increase in blood pressure and heart rate (3). Sympathetic nervous system activity also is reflected in a circadian variation in plasma norepinephrine and epinephrine levels, which parallel blood-pressure and heart-rate patterns (4). Additionally, a circadian variation has been described for plasma renin activity and atrial natriuretic peptide (ANP) in normal subjects (5).
Congestive heart failure alters physical performance and is associated with alterations in sympathetic and parasympathetic nervous system activity (6), the renin-angiotensin system, and vasopressin and ANP secretion (6-13). Baroreflexes also are abnormal in cogestive heart failure (14,15). As the severity of congestive heart failure progresses, as assessed by hemodynamic measurements at rest and during exercise, a corresponding reduction occurs in the variability of the time course of blood pressure and heart rate throughout the day (16,17).
The severity of congestive heart failure may be assessed by the plasma concentration of norepinephrine (reflecting the degree of sympathetic nervous system activity), the left ventricular ejection fraction (LVEF), and exercise capacity (18). Treatment with angiotensin-converting enzyme (ACE) inhibitors has been found to improve symptoms and prognosis in congestive heart failure (19-23), possibly because of cardiac unloading by inhibition with abnormally increased neurohumoral systems and restoration of reflex mechanisms (24,25).
We hypothesized that the degree of variability in the circadian pattern of blood pressure and heart rate recorded from ambulatory patients would correlate directly with cardiovascular indices of the severity of heart failure (i.e., LVEF) and inversely with neurohumoral activity (i.e., plasma norepinephrine and ANP). We further predicted that treatment with ACE inhibitors would restore some degree of autonomic control, as reflected by a more normal variability of circadian patterns of blood pressure. Accordingly, we recorded ambulatory 24-h blood pressures and heart rates in patients with congestive heart failure, who were treated only with digitalis and diuretics, to determine whether circadian patterns correlated with neurohumoral parameters, left ventricular systolic dysfunction, and clinical assessment of the severity of heart failure. We excluded patients with heart failure resulting from hypertension because alterations in the circadian patterns of blood pressure may be abnormal even before the appearance of heart failure (26,27). We then studied the effect of two ACE inhibitors with different pharmacodynamic profiles on these parameters to determine whether or not chronic treatment with ACE inhibitors altered the variation in blood pressure and heart rate, as previously shown after administration of the first dose (28).
Thirty patients of both sexes older than 21 years with a clinical diagnosis of nonhypertensive heart failure, class II-IV (New York Heart Association), and an LVEF ≤40% were recruited for the study. The Institutional Committee on the Use of Humans in Research approved the protocol, and patients who elected to participate signed informed consents. Patients were excluded if they had unstable ischemic heart disease, recent (within 8 weeks) myocardial infarction, primary valvular heart disease, supine or standing systemic arterial blood pressure <90/60 mm Hg, serum creatinine ≥2.5 mg/dl, known hypersensitivity to ACE inhibitors, or a history of angioedema. Patients who worked evening or night shifts also were excluded from the trial.
Qualifying patients discontinued any previous vasodilator therapy and entered a 7-day placebo period, during which they received stable doses of digoxin, diuretics, and placebo to match a single 5-mg lisinopril tablet daily and a single 12.5-mg captopril tablet administered every 8 h. Throughout the placebo period, patients' body weights could not vary by >1 kg. Patients whose stable digoxin or diuretic dose did meet these criteria could reenter the trial for an additional placebo period after being stabilized.
At the end of the placebo period, patients were randomized to active treatment if they had a technically satisfactory ambulatory blood pressure recording and demonstrated placebo drug compliance of ≥80% (as determined by pill count). Patients then received 6 weeks of double-blind treatment with either 5 mg of lisinopril once daily or 12.5 mg of captopril every 8 h (the placebo was continued for the non-acting drug). Patients were assessed weekly, at which time the doses of lisinopril (10 mg, then 20 mg daily) or captopril (25 mg, then 50 mg, every 8 h) could be increased. Doses were increased to achieve maximal improvement in clinical status without causing serious adverse effects.
Diuretic doses could be adjusted upward until the final week of the double-blind treatment and were decreased if the serum blood urea nitrogen (BUN) increased to ≥50 mg/dl or twice the baseline concentration, if serum potassium decreased to ≤3.4 mEq/l, if serum creatinine increased to ≥2.4 mg/dl or twice the baseline concentration, or if the patient exhibited clinical signs of hypovolemia. Diuretics could be increased if a patient had clinical evidence of worsening heart failure, including increased rales, increased edema, or increased jugular venous pressure, and a weight gain ≥3 kg.
The use of calcium channel antagonists, β-adrenoceptor antagonists, any antihypertensive medication, vasodilators (prazosin, hydralazine, or other ACE inhibitors), or long-acting nitrates was prohibited during the study. Patients were counseled to avoid any dietary changes that could significantly affect body weight or fluid balance (i.e., major changes in consumption of calories, alcohol, or sodium).
Clinic measurement of blood pressure and heart rate
During patients' weekly office visits, their resting blood pressure and heart rate were measured after they had been in a supine position for 5 min and standing for 3 min. Blood pressure was measured with a mercury sphygmomanometer on the same arm (usually nondominant), to which the ambulatory blood pressure cuff was applied. Diastolic pressure was recorded as Korotkoff phase 5.
Twenty-four-hour ambulatory blood pressure and heart rate were recorded by using Oxford AccuTracker II Monitors (Suntech Medical Instruments, Inc., Clifton, NJ, U.S.A.). Recordings were made at the end of the placebo period just before randomization (baseline) and after 6 weeks of active drug treatment. The monitors were programmed to record blood pressure and heart rate measurements every 15 min from 0600 h to 2400 h, and every 30 min from 2400 h until 0600 h. Minimal criteria for acceptable recordings included (a) a minimum of 67 good readings from the total of 84 scheduled readings; (b) at least two good readings per hour during 0500 h until 22:59, with no more than 3 h with 1 data point each; and (c) one good reading per hour between 23:00 and 04:59 h, with no more than 1 h of missing data. Repeated 24-h recordings were performed until these criteria were satisfied.
The LVEF was measured by radionuclide angiography in the left anterior oblique position with 99Tc-labeled red blood cells, by using a standard gamma camera and a computer retrieval system with conservative R-wave gating.
Plasma norepinephrine, ANP, and renin activity
Blood samples were collected to determine plasma levels of norepinephrine, ANP, and renin activity before randomization (baseline). These samples were drawn before morning dosing. Time 0 corresponded with the initial daily dose of placebo or active drug (≈0900 h). Blood samples were drawn after 30 min of supine rest and after placement of an indwelling antecubital venous canula. Blood was collected in glass tubes containing EDTA and placed immediately in an ice bath. Samples were centrifuged at 4°C within 30 min of collection, and the plasma was frozen and stored at -80°C.
Plasma norepinephrine (in picograms per milliliter) was assayed by using high-performance liquid chromatography with electrochemical detection after solid-phase extraction from plasma by using acid-washed alumina (sensitivity, 15 fmol). Quantitative determination of plasma ANP (picograms ANP per milliliter) was performed by double-antibody radioimmunoassay (sensitivity 10 pg/ml; interassay variation, 10%). Plasma renin activity (nograms angiotensin I per milliliter per hour) was determined by radioimmunoassay of generated angiotensin I, by using a commercially available kit (Baxter Travenol Diagnostics) (sensitivity, 18 pg (angiotensin I); intrarun variability of ±10%; antisera cross-reactivity with either angiotensin II or III of ≤0.01%).
Linear regression was used to determine the significance of correlations between various measured parameters from the pooled lisinopril and captopril subject data. Intragroup comparisons were made by using Student's t test. Statistical significance was accepted at a p ≤ 0.05 level.
The mean time course of 24-h blood pressure and heart rate derived from the raw data, as well as curves derived from Fourier analysis (six harmonics), were constructed for inspection. Six harmonics were chosen because the resultant curve-smoothing effect preserves the shape of the circadian fluctuation of the raw data (29). The rate-pressure product was calculated, and the resultant time-course curves treated in a similar fashion. Fourier analysis (six harmonics) of the 24-h time course of systolic blood pressure, diastolic blood pressure, heart rate, and rate-pressure product was used. The analysis was carried out by using BMDP program 1T (BMDP Statistical Software, CA, U.S.A.), running on an IBM 3081-K computer under the CMS operating system. Program 1T uses the arbitrary radix discrete fast-Fourier transform algorithm to compute periodograms.
The absolute amplitude of the 24-h time course for each patient for systolic and diastolic blood pressure, heart rate, and rate-pressure product was calculated as 50% of the difference between the daytime acrophase and the nighttime nadir, expressed as millimeters of mercury. The coefficient of variation of the time course of systolic and diastolic blood pressure and heart rate also was calculated (coefficient of variation = standard deviation/mean).
Table 1 presents demographic characteristics, baseline blood pressure, heart rates, LVEF, and clinical classifications (New York Heart Association). The origin of heart failure was ischemic heart disease in 21 patients and idiopathic dilated cardiomyopathy in the remaining nine. Demographic characteristics and LVEF were comparable for lisinopril and captopril groups. However, 27% of the captopril patients were in clinical class IV, compared with 0% of the patients in the lisinopril group. The final daily dose of lisinopril was 16 ± 7 mg (mean ± SD) and 130 ± 38 mg for captopril.
Baseline ambulatory blood-pressure and heart-rate patterns, LVEF, and neurohumoral associations
Table 2 summarizes baseline 24-h blood pressure, heart rate, and rate-pressure product, as well as plasma norepinephrine, ANP, and renin activity. The absolute amplitude of the 24-h systolic blood-pressure measurements, at baseline, bore an inverse correlation to simultaneous measurements of plasma norepinephrine (R = 0.51; p = 0.004) and ANP (R = 0.51; p = 0.004) but not to renin activity (R = 0.34; NS; Fig. 1). Recordings from two patients illustrate the discordance between the amplitude of systolic blood pressure and plasma norepinephrine and ANP (Fig. 2). The absolute amplitude of diastolic blood pressure and heart rate did not correlate significantly with neurohormonal measurements.
LVEF did not correlate with the absolute amplitude of the 24-h systolic blood-pressure curve (R = 0.28; NS) but did not correlate inversely with plasma norepinephrine (R = 0.49; p = 0.005) and ANP concentrations (r = 0.44; p = 0.015). LVEF did not correlate with plasma renin activity (R = 0.19, NS), nor was there a correlation between LVEF and diastolic blood pressure, heart rate, and rate-pressure product. Although a strong correlation existed between the absolute amplitude of systolic blood pressure and the coefficient of variation for 24-h systolic blood pressure (r = 0.832; p = 0.0001), the coefficient of variation of the time-course curves for systolic and diastolic blood pressure, heart rate, and rate-pressure product and plasma neurohormones and LVEF correlated only weakly.
Effect of treatment with ACE inhibitors
A nonsignificant decrease in the office blood pressure and heart rate was observed in both lisinopril and captopril treatment groups (Table 3). Mean 24-h systolic blood pressure decreased by 9 mm Hg after treatment with lisinopril (p = 0.01) and by 4 mm Hg after captopril (p = NS); diastolic blood pressure decreased 1 and 2 mm Hg after treatment with lisinopril and captopril, respectively (both p = NS). Treatment with lisinopril decreased average systolic blood pressure more during the first 8 h after dosing (14 ± 14 mm Hg) than did treatment with captopril (5 ± 10 mm Hg; p = 0.05). Eight h after dosing, reductions in systolic blood pressure, heart rate, and rate-pressure product were greater in the lisinopril group than in the captopril group (p < 0.05). Treatment with ACE inhibitors increased the absolute amplitude of the systolic blood pressure. This posttreatment increase correlated inversely with the pretreatment amplitude (Fig. 3).
Inspection of 24-h recordings for systolic and diastolic blood pressure, heart rate, and rate-pressure product, transformed by Fourier analysis, revealed differences in the influence of lisinopril and captopril on the time course of these parameters (Figs. 4 and 5). After 6 weeks of treatment, systolic and diastolic blood pressure, heart rate, and rate-pressure product were slightly lower in both the lisinopril and captopril groups. Lisinopril decreased systolic and diastolic blood pressure (Fig. 4) throughout the 24-h period, approaching baseline before the next daily morning dose (Fig. 4). In contrast, the effect of captopril on systolic and diastolic blood pressure was characterized by more fluctuation (Fig. 5). Systolic and diastolic blood pressure returned to control values within 8 h after the first dose of captopril; a second dose of captopril again decreased systolic blood pressure, which again returned to baseline within 8 h. A similar pattern was observed during the third dosing period of the day. Heart rates were relatively unaffected by both lisinopril and captopril after 6 weeks of treatment. Thus changes in rate-pressure product followed a pattern similar to that of systolic blood pressure. The nighttime nadir of blood pressure and heart rate curves in both the lisinopril and captopril groups was preserved.
Four adverse events classified as “severe” occurred in two patients, both randomized to captopril treatment. One patient had fatigue, orthopnea, and dyspnea after 8 days of randomized treatment but was able to continue in the trial. The other patient had nocturnal sudden death after 26 days of randomized treatment; the patient had voiced no complaints on retiring the previous evening. One patient randomized to lisinopril was withdrawn from the study after having dizziness after 10 days of randomized treatment. Adverse experiences occurring in more than two study patients during randomized treatment, in order of decreasing frequency, included dizziness in six patients in the lisinopril group and chest pain in four patients in the captopril group.
Clinical laboratory results
Serum creatinine increased by ≥0.3 mg/dl in two patients taking lisinopril (1.1 to 1.5 to 2.0 mg/dl) and in two patients taking captopril (1.2-1.6 and 1.3-1.6 mg/dl). One patient taking captopril had an increase in BUN from 29 mg/dl at baseline to 96 mg/dl at the end of the study. This patient was thought to have prerenal azotemia.
Twenty-four-hour ambulatory blood pressure readings from patients in this study demonstrated an inverse correlation between the absolute amplitude of systolic blood pressure and rate-pressure product and plasma norepinephrine concentration. These data are consistent with previous observations in hospitalized patients (16,17), indicating that increased activity of the sympathetic nervous system is associated with worsening cardiac function. Similarly, the increase in plasma ANP concentration is consistent with the counterregulatory effect of this hormone and is probably related to increased atrial tension induced by volume loading in left ventricular heart failure (19). The significant but moderate correlation between plasma norepinephrine (r = 0.5; p = 0.004) and ANP (r = 0.51; p = 0.004) with the absolute amplitude of 24-h systolic blood pressure is indicative of the multiplicity of interactive factors-neurohumoral and others-that are altered in patients with congestive heart failure and that influence blood pressure. The lack of association with plasma renin activity is expected, because activation of the renin-angiotensin system is not as consistent in patients with heart failure (9). In our study, the absolute amplitude of systolic blood pressure correlated better with indices of severity of heart failure than did the coefficient of variation of systolic blood pressure.
A direct correlation between LVEF and the absolute amplitude of the circadian variation in both systemic systolic arterial blood pressure and rate-pressure product has been reported previously (16,17). Data from this study did not show a significant correlation between absolute amplitude of the systolic blood pressure and LVEF and may be related to greater variation in blood pressure in ambulatory patients as compared with those confined to the hospital. Therefore in active patients, neurohumoral modulation is more responsible for the progressive lack of systemic arterial variation in patients with heart failure, underscoring the need to interrupt these mechanisms by pharmacologic unloading. Thus the circadian variation in blood pressure may represent a more integrative reflection of the heart and peripheral vasculature than does the LVEF. The absolute amplitude of systolic blood pressure overlaps greatly between normal subjects and patients with congestive heart failure (16,17), such that separation on this basis alone is not possible.
Treatment with the ACE inhibitors lisinopril and captopril increased the absolute amplitude of the systolic blood pressure in those patients with a low amplitude of systolic blood pressure at baseline, whereas the amplitude decreased in some patients with normal or high amplitudes at baseline, perhaps defining an optimal window of variability. Thus an increase in the low variability of the circadian systolic blood pressure after treatment may be used as an index of the effectiveness of therapy for heart failure.
Fourier analysis of the circadian variation in blood pressure, heart rate, and rate-pressure product did reveal pharmacodynamic differences in the two ACE-inhibitor treatment regimens (Figs. 4 and 5). These pharmacodynamic patterns could be predicted from what is known about these ACE inhibitors but does demonstrate for the first time the usefulness of ambulatory blood pressure monitoring to assess these effects in patients with congestive heart failure. It remains to be seen whether different pharmacodynamic patterns of blood-pressure and heart-rate changes correlate with improvement in morbidity or mortality.
Limitations of the study
Our study included patients receiving digitalis and diuretics, and the active treatment period included no placebo group. Although our results are similar to those in previously published reports, including those obtained from various combinations of placebo, digitalis, diuretics, and ACE inhibitors, we do not know what effect each of these drugs had on the circadian patterns of blood pressure and neurohormonal status. For example, digitalis influences baroreceptor reflexes (30,31), and diuretics have important effects on the renin-angiotensin and sympathetic nervous systems.
This study demonstrates the feasibility of recording ambulatory 24-h blood pressure in patients with congestive heart failure. The absolute amplitude of 24-h systolic blood pressure, as a simple measure, correlated inversely with neurohumoral indices of the severity of heart failure and increased after treatment with ACE inhibitors, if baseline variability was low. Thus absolute amplitude of 24-h systolic blood pressure may be a useful parameter for evaluating a different aspect of drug treatment for congestive heart failure. Ambulatory blood-pressure monitoring also appears to be useful for determining pharmacodynamic patterns of therapeutic agents in patients with heart failure.
Acknowledgment: We thank Terrence Flannigan, Michael B. Given, Ph.D., Robert A. Jones, and David Rice, Ph.D., for their advice and assistance in presenting these findings. This study was supported in part by a grant from Zeneca Pharmaceuticals Group, Inc.
1. Millar-Craig MW, Bishop CN, Raftery EB. Circadian variation
of blood pressure. Lancet
2. Mann S, Millar-Craig M, Mellville D, Balasubraminian V, Raftery EB. Physical activity and the circadian rhythm of blood pressure. Clin Sci
3. Furlan R, Guzzetti S, Crivellaro W, et al. Continuous 24-hour assessment of the neural regulation of systemic arterial pressure and RR variabilities in ambulant subjects. Circulation
4. Linsell CR, Lightman SL, Muller PE, Brown MJ, Causon RD. Circadian rhythm of epinephrine and norepinephrine in man. J Clin Endocrinol Metab
5. Portaluppi F, Bagni KB, Uberti E, et al. Circadian rhythm of atrial natriuretic peptide, renin, aldosterone, cortisol, blood pressure and heart rate in normal and hypertensive subjects. J Hypertens
6. Porter TR, Eckberg DL, Fritsch JM, et al. Autonomic pathophysiology in heart failure patients. J Clin Invest
7. Levine TB, Francis GS, Goldsmith SR, Simon AB, Cohn JN. Activity of the sympathetic nervous system and renin-angiotensin system assessed by plasma hormone levels and their relationship to hemodynamic abnormalities in congestive heart failure. Am J Cardiol
8. Viquerat CE, Daly P, Swedberg K, et al. Endogenous catecholamine levels in chronic heart failure: relation to the severity of hemodynamic abnormalities. Am J Med
9. Dzau VJ, Colucci WS, Hollenberg NK, Williams GH. Relation of the renin-angiotensin-aldosterone system to clinical state in congestive heart failure. Circulation
10. Burnett JC Jr, Koa PC, Hu DC, et al. Atrial natriuretic peptide elevation in congestive heart failure in the human. Science
11. Cody RJ, Atlas SA, Laragh JH, et al. Atrial natriuretic factor in normal subjects and heart failure patients: plasma levels and renal, hormonal, and hemodynamic responses to peptide infusion. J Clin Invest
12. Kluger J, Cody RJ, Laragh JH. The contribution of sympathetic tone and the renin-angiotensin system to severe chronic congestive heart failure: response to specific inhibitors (prazosin and captopril). Am J Cardiol
13. Creager MA, Faxon DP, Cutler SS, Kohlmann O, Ryan TJ, Gavras H. Contribution of vasopressin to vasoconstriction in patients with congestive heart failure: comparison with the renin-angiotensin system and the sympathetic nervous system. J Am Coll Cardiol
14. Ferguson DW, Berg WJ, Roach PJ, Oren RM, Mark AL. Effects of heart failure on baroreflex control of sympathetic neural activity. Am J Cardiol
15. Ellenbogen DA, Mohanty PK, Szentpetery S, Thames MD. Arterial baroreflex abnormalities in heart failure: reversal after orthotopic cardiac transplantation. Circulation
16. Caruana MP, Lahiri KA, Cashman PMM, Altman DG, Raftery EB. Effects of chronic congestive heart failure secondary to coronary artery disease on the circadian rhythm of blood pressure and heart rate. Am J Cardiol
17. van de Borne P, Abramowicz M, Degre S, Degaute JP. Effects of chronic congestive heart failure on 24-hour blood pressure and heart rate patterns: a hemodynamic approach. Am Heart J
18. Cohn HN, Rector TS. Prognosis of congestive heart failure and predictors of mortality. Am J Cardiol
19. Cannon PS and the Captopril Multicenter Research Group. A placebo-controlled trial of captopril in refractory congestive heart failure. J Am Coll Cardiol
20. Kramer BL, Massie MB, Topic N. Controlled trial of captopril in chronic heart failure; a rest and exercise hemodynamic study. Circulation
21. Sharpe ND, Murphy J, Coxon R, Hannan S. Enalapril in patients with chronic heart failure: a placebo-controlled, randomized, double-blind study. Circulation
22. SOLVD: (Studies of Left Ventricular Dysfunction) Investigators. Effects of enalapril on survival of patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med
23. Pfeffer MA, Braunwald E, Moyel A, et al. Effects of captopril on mortality and morbidity in patients with left-ventricular dysfunction after myocardial infarction: results of the survival and ventricular enlargement trial. N Engl J Med
24. Kondowe GB, Deering AH, Riddell JG, Johnston GD, Harron DWG. The effect of acute and chronic captopril therapy on baroreflex function in man. Br J Clin Pharmacol
25. Osterziel KHN, Rohrig R, Dietz J, Manthey J, Hect J, Kubler W. Influence of captopril on the arterial baroreceptor reflex in patients with heart failure. Eur Heart J
26. Korbin I, Oigman W, Kumar A, et al. Diurnal variation of blood pressure in elderly patients with essential hypertension. J Am Geriatr Soc
27. Murphy MB, Nelson KS, Oliner CM, Elliott WJ. Higher nocturnal blood pressure in normal and hypertensive blacks compared with whites [Abstract]. Circulation
28. Giles TD, Katz R, Sullivan JM, et al. for the Multicenter Lisinopril-Captopril CHF study group. Short- and long-acting angiotensin-converting enzyme inhibitors: a randomized trial of lisinopril versus captopril in the treatment of congestive heart failure. J Am Coll Cardiol
29. Maching T, Engels G, Schobel S, Gellert J, Bachmann K. A new approach for evaluating circadian blood pressure levels and variability from 24-hour ambulatory blood pressure profiles. Am J Hypertens
30. Mason DT, Braunwald E. Studies on digitalis. X. Effects of ouabain on forearm vascular resistance and venous tone in normal subjects and in patients in heart failure. J Clin Invest
31. Ferguson DW, Berg WJ, Sanders JS, Roach PJ, Kempf JS, Kienzle MG. Sympathoinhibitory responses to digitalis glycosides in heart failure patients: direct evidence from sympathetic neural recordings. Circulation