Differential Effects of Theophylline on Sympathetic Excitation, Hemodynamics, and Breathing in Congestive Heart Failure
Background— Patients with heart failure have high levels of central sympathetic outflow and also have a high prevalence of sleep-related breathing disorders, predominantly central sleep apnea. The options for treating central sleep apnea in heart failure are limited and include theophylline. Whether theophylline alters sympathetic activity in heart failure patients is not known.
Methods and Results— Using a single-blinded, randomized, placebo-controlled study design, we investigated the sympathetic, hemodynamic, neurohumoral, and ventilatory effects of theophylline in patients with congestive heart failure compared with healthy control subjects closely matched for age, sex, and body mass index. Theophylline increased muscle sympathetic nerve activity and lowered transcutaneous CO2 in the control subjects but only lowered transcutaneous CO2 in the heart failure patients. Theophylline nearly doubled plasma renin concentration in both the healthy subjects (P<0.01) and the heart failure patients (P<0.02).
Conclusions— Our study shows that in heart failure patients, there are differential effects of theophylline: in contrast to healthy subjects, theophylline does not increase sympathetic activity in heart failure, whereas increases in plasma renin and ventilation are still evident. These novel findings may have important implications for understanding the potential harmful and beneficial effects of theophylline and related substances in heart failure patients.
Received January 12, 2004; revision received April 15, 2004; accepted April 15, 2004.
There is an emerging epidemic of congestive heart failure (CHF).1 Therapeutic strategies in CHF patients, such as antagonism of angiotensin, aldosterone, and adrenergic activation, target those mechanisms that are known to contribute to the pathophysiology of the failing heart.
Patients with CHF have high levels of central sympathetic outflow and also have a high prevalence of sleep-related breathing disorders, predominantly central sleep apnea. The presence of central sleep apnea in CHF is highly predictive of increased mortality.2 Evidence indicates that prevention of the central sleep apnea with continuous positive airway pressure may result in improved transplant-free survival.3 Evidence of potential beneficial effects of preventing central sleep apnea, together with the difficulties inherent in any widespread administration of continuous positive airway pressure treatment to CHF patients, has prompted the exploration of alternative strategies for the treatment of central sleep apnea.4
One such option has been the use of theophylline, a phosphodiesterase inhibitor with excitatory effects on breathing. Javaheri et al5 have shown that short-term administration of theophylline to CHF patients with central sleep apnea will significantly attenuate the severity of apnea. Pesek et al6 have also successfully used theophylline to treat life-threatening central apneas in the setting of normal ventricular function.
Despite its therapeutic use for more than 6 decades, the mechanisms of action of theophylline are poorly understood. What is known is that it is a nonselective phosphodiesterase inhibitor at supratherapeutic concentrations, inhibits intracellular calcium release, inhibits inflammatory mediators, and is an adenosine receptor antagonist.7,8 CHF patients have a 3- to 4-fold increase in plasma adenosine levels.9 Englestein et al10 have demonstrated that augmentation of endogenous adenosine levels in normal humans increases heart rate, blood pressure, sympathetic activity, and ventilation. Adenosine may hence be implicated in both the breathing abnormalities and the increased sympathetic activation in CHF.
Adenosine intensely inhibits respiration via the nucleus tractus solitarius but also mediates excitatory effects by peripheral chemoreceptors.10 Theophylline, which crosses the blood-brain barrier, stimulates breathing by competitive antagonism of adenosine at central neuronal receptor sites.11 Adenosine may also be closely involved in the regulation of central sympathetic outflow, although the interactions between adenosine and sympathetic activation are complex. Adenosine inhibits release of norepinephrine from sympathetic nerve endings, causes vasodilation, and inhibits renin release. These inhibitory effects of adenosine are particularly prominent in efferent nerves and in the brainstem.12 However, adenosine also evokes strong sympathetic excitatory reflex effects, presumably mediated by chemically sensitive receptors and afferent nerves in the heart, skeletal muscle, and chemoreceptors.13
The effect of theophylline on attenuating central sleep apnea severity in CHF mandates a more detailed evaluation of the pressor and sympathetic excitatory effects of theophylline in the context of the pathophysiological milieu of the failing heart. Whether theophylline alters sympathetic activity in CHF patients has never been studied. The effects of theophylline on other neurohumoral mechanisms that affect prognosis in CHF, such as aldosterone and endothelin, are also unknown. In healthy subjects, theophylline14 and other methylxanthines, such as caffeine15 and aminophylline, increase sympathetic activity, although some studies16 do not support these findings.
Using a single-blinded, randomized, placebo-controlled study design, we investigated the sympathetic, neurohumoral, and ventilatory effects of theophylline in patients with CHF compared with healthy control subjects closely matched for age, sex, and body mass index.
Patients with CHF (NYHA functional class II–III), and healthy control subjects of both sexes, from 18 to 69 years old, were included. Patients with impaired left ventricular function (left ventricular ejection fraction ≤35%) because of chronic ischemic heart disease or dilated cardiomyopathy were eligible. Exclusion criteria were myocardial infarction or pulmonary edema within 6 months of entry; obesity (body mass index >28); a history suggestive of obstructive sleep apnea; renal, liver, or lung disease (forced expiratory volume in 1 second <70% of predicted); polyneuropathy or other neuron diseases; and treatment with drugs having sympathomimetic activity (theophylline, moxonidine, clonidine, β2 sympathomimetics). Healthy control subjects were closely matched for age, sex, and body mass index. No control subjects were on regular medication. The institutional ethics committee approved the study. Informed written consent was obtained from all patients and control subjects before enrollment in the protocol.
Microneurography and Respiration
Sympathetic activity was measured by use of microneurographic recordings of efferent muscle sympathetic nerve activity (MSNA) as described previously.17 Blood pressure was measured noninvasively by sphygmomanometry (Dinamap XL Monitor, model 9302; Johnson & Johnson Medical).
Respiratory rate and tidal volume were approximated by calibrated respiratory inductive plethysmography (Respitrace Systems, Ambulatory Monitoring Inc) as previously described.17,18 CO2 levels were monitored transcutaneously (TINA, Radiometer), with an arterial blood sample taken for in vivo calibration (ABL 3, Radiometer).17 Arterial oxygen saturation was measured transcutaneously on the index finger by pulse oximetry (Andos Oxycount MCC).
After insertion of an intravenous catheter in an antecubital vein, blood samples were obtained, without applying a tourniquet, before the infusion of theophylline or placebo and after the resting period thereafter. Endothelin-1, pro-atrial natriuretic peptide, aldosterone, and renin were determined by commercially available immunoassays. Catecholamines were analyzed by liquid chromatography.
Patients and control subjects were studied in the morning, in a supine position, 2 hours after a low-calorie breakfast free of caffeine-containing beverages. After a satisfactory nerve signal had been obtained, subjects were randomized to theophylline or placebo. Placebo or theophylline infusion began after a 10-minute baseline recording period, with a loading dose of 5 mg/kg over a period of 20 minutes followed by a continuous dose of 0.5 mg · kg−1 · h−1 over a period of 10 minutes and a 15-minute period without drug infusion. The subjects were blinded to the intervention.
Lung function tests were performed in the supine position before and after the experiment.19 The maximum inspiratory pressure as an index of respiratory muscle strength was also evaluated.
Sympathetic bursts were quantified manually by a single observer (A.D.) blinded to subject and intervention. In our institution, the intraobserver variation in identifying bursts is 5%, and the interobserver variation is 11%.17 Data for MSNA, heart rate, arterial blood pressure, tidal volume, minute volume, oxygen saturation, and transcutaneous Pco2 were averaged during the last 5 minutes of the baseline and continuous dose and during minutes 8 to 12 of the loading dose. Changes in MSNA are expressed as percent of baseline values. All variables are given as mean±SEM. Demographic, pulmonary function, and baseline data in patients and control subjects were compared by use of Student’s unpaired t test. Repeated-measures ANOVA with time as within-factor and group (placebo versus theophylline) as between-factor was used to analyze the effects of theophylline. Although the ANOVA was calculated with baseline, loading, and continuous doses, the tables give only baseline and continuous dose for clarity. Two-tailed tests were used, and significance was accepted at a value of P<0.05.
CHF was caused by ischemic heart disease in 10 patients and idiopathic cardiomyopathy in 12 patients. Eleven patients were in NYHA class II and 11 patients in class III. Left ventricular ejection fraction was 29.0±1.4%. ACE inhibitors were used in 16 patients, β-blockers in 13 patients, diuretics in 11 patients, and digitalis glycosides in 9 patients. Comparison between the CHF patients and healthy subjects, closely matched for age, sex, or body mass index, are shown in Table 1. At baseline, CHF patients had higher MSNA compared with the healthy control subjects (Table 2).
Hemodynamic and Neurohumoral Effects of Theophylline
Plasma concentration of theophylline after infusion was 10.1±0.4 mg/L for the CHF patients and 9.0±0.8 mg/L for the healthy subjects (P=NS). Compared with placebo, in healthy subjects, theophylline increased sympathetic activity whether expressed as bursts per minute or percentage increase in burst amplitude (Figure 1). There was also an increase in plasma norepinephrine (Table 3).
By contrast, compared with the effects of placebo, theophylline had no effect on increasing any of the microneurographic measures of sympathetic nerve activity in CHF patients (Figure 1). Median norepinephrine levels in the CHF patients tended to decrease after theophylline (Table 3).
Hemodynamics and Ventilation
In healthy subjects, theophylline increased heart rate and systolic blood pressure, but this effect was not significant compared with placebo (Table 4). In CHF, there was no significant effect of theophylline on heart rate.
Although theophylline did not significantly change levels of endothelin-1, atrial natriuretic peptide, aldosterone, or erythropoietin, there was a significant increase in plasma renin levels both in the healthy subjects and in CHF patients after theophylline (Table 3). In the CHF patients, the magnitude of increase in plasma renin activity was independent of the presence or absence of ACE inhibitor therapy.
Theophylline significantly lowered transcutaneous Pco2 in the healthy subjects and in the CHF patients (Figure 2). Despite the fall in CO2 during theophylline, indirect estimation of minute ventilation still tended to increase in the CHF patients (Table 4). There was no significant effect of theophylline on vital capacity, forced expiratory volume, or maximal inspiratory pressure in either patients or control subjects (data not shown).
NYHA class, ejection fraction, or medications did not significantly influence the effects of theophylline in CHF. There was no correlation between changes in MSNA and the fall in transcutaneous CO2. For the patients and control subjects combined, there was a significant positive correlation between the increase of MSNA expressed in bursts per minute and the increase in norepinephrine (r=0.47; P<0.05) and endothelin-1 (r=0.61; P<0.01) after theophylline infusion.
Adenosine, which is markedly increased in CHF, exerts important effects on sleep, on breathing, and on sympathetic outflow. Adenosine nucleotides also inhibit renin secretion from juxtaglomerular cells.20 Abnormalities in sleep, sleep-related breathing, sympathetic activation, and the renin-angiotensin system are highly prevalent in CHF. Although theophylline may attenuate the severity of central sleep apnea in CHF, perhaps by its adenosine antagonism effects, the effects of theophylline on blood pressure, sympathetic activity, and plasma renin activity in CHF have never previously been studied.
In this single-blinded, randomized, placebo-controlled study comparing the effects of theophylline in healthy control subjects and in CHF patients, we note that theophylline increases MSNA, plasma renin, and ventilation in healthy subjects. The increased sympathetic activity is evident even in the setting of increased ventilation, which would be expected to inhibit central sympathetic outflow.21 By contrast, in patients with CHF, theophylline did not increase heart rate or sympathetic nerve traffic, although the excitatory effects on renin and breathing were still present.
Our data are consistent with several earlier studies. The increase in MSNA in healthy older subjects that we noted after theophylline is comparable to that seen by Corti et al15 in healthy young subjects in response to caffeine administration. Also, in studies using oral caffeine administration, Notarius et al16 found no change in MSNA or heart rate in patients with heart failure. Our data showing increased plasma renin activity after theophylline administration are also consistent with studies in healthy young men, showing that caffeine increases plasma renin activity.22 We now extend these findings, reporting a similar effect of theophylline in healthy subjects and also showing that plasma renin activity is increased by theophylline even in CHF. Thus, our data suggest the interesting speculation that heightened adenosine levels in CHF may act to inhibit or buffer excessive increases in plasma renin activity and that adenosine antagonism by theophylline blunts this inhibitory effect of adenosine, resulting in a more than 2-fold increase in plasma renin activity in CHF.
In our study, theophylline raised renin but not aldosterone plasma concentration. This is most likely explained by the fact that aldosterone plasma concentrations start to rise only about 30 minutes after increased renin secretion. Because there was no significant effect of theophylline on aldosterone, even in the control subjects, it is unlikely that the administration of an ACE inhibitor in the patients accounts for this finding.
The differential neural circulatory responses we observed in CHF patients compared with healthy subjects are also consistent with precedents established in other high adenosine situations, such as exercise. Exercise causes an increase in adenosine concentrations in contracting human skeletal muscle, thereby stimulating muscle afferents.23 Muscle afferent activation is an important mechanism for sympathetic excitation. Intrabrachial theophylline, with only minor systemic concentrations, did not affect resting MSNA but abolished the increase in MSNA with exercise.23 Thus, with increasing adenosine plasma concentrations during exercise, adenosine antagonism by theophylline plays a sympathoinhibitory role even in healthy subjects.23
The increased ventilation that we observed in healthy subjects and in CHF patients is also consistent with previous studies of the effects of theophylline in animals,11,24 in patients with brain damage,25 in hypercapnic central sleep apnea without heart failure,6 and in healthy subjects given caffeine.22 The effects of theophylline on ventilation in CHF patients has not been studied previously, although Javaheri et al5 clearly demonstrated an improvement in nocturnal oxygen saturation. Our findings of increased ventilation in CHF patients given theophylline are relevant to their observations regarding the efficacy of theophylline in CHF patients with central sleep apnea.5 Nevertheless, because lower CO2 in CHF is associated with an increased prevalence of central sleep apnea,4 any attenuation of central sleep apnea with theophylline would occur despite the lower CO2 levels.
The plasma theophylline concentrations achieved in our study were in the low therapeutic range (the therapeutic range is 8 to 20 mg/L) that is recommended for the treatment of respiratory disorders.8 Levels achieved in the present study are also comparable to those noted in previous studies.5,6,25 Thus, the results of theophylline administration observed in our present study relate well to a better understanding of the results from these previous investigations. Other important strengths of our study include the blinded, randomized, placebo-controlled study design and the comprehensive evaluation of sympathetic neural, humoral, hemodynamic, and ventilatory measures.
Limitations include, first, the single-blinded study design. In its defense, this approach was adopted so as to maximize patient safety. Also, even though the data were obtained in a single-blinded manner, quantification of microneurographic measures of sympathetic activity was performed by a single observer blinded to subject and intervention. Second, despite the striking fall in transcutaneous CO2 with theophylline in the CHF patients, increases in estimated minute ventilation fell short of significance. This is most likely because of the method used, in which qualitative measures of chest wall movement are extrapolated to derive quantitative approximate measures of minute ventilation.
There are strong arguments for a cautious and circumspect approach to the administration of theophylline to CHF patients. In our study, we noted an increase in plasma renin activity. Whether the potentially harmful effects of this response to theophylline may be attenuated by ACE inhibitor or angiotensin receptor blocker administration remains to be determined.26 In a large retrospective epidemiological study, the use of theophylline for the treatment of lung disease was independently related to increased cardiovascular death in subjects with acute coronary insufficiency or congestive cardiomyopathy.27 Furthermore, a prospective study showed increased mortality in CHF patients treated with the phosphodiesterase inhibitor milrinone. Thus, our results should not be interpreted as providing a rational basis for the use of theophylline in the CHF population. Nevertheless, it is important that the neural circulatory, respiratory, and humoral effects of widely used agents such as theophylline be carefully documented, not only in healthy subjects but also in patients with disease conditions. It is conceivable, but very speculative, that although theophylline may adversely affect outcome in CHF patients generally, there may be beneficial effects in highly selected CHF patients, for example, those with severe central sleep apnea and on maximal therapy with agents opposing the effects of renin-angiotensin activation. In cardiovascular disease, especially in CHF patients, the promise of selective beneficial effects of theophylline (or other adenosine receptor antagonists), perhaps in treating central sleep apnea, argues strongly for a careful categorization of the other effects of theophylline. Our study shows that in the heart failure milieu, there are differential effects of theophylline: in contrast to healthy subjects, theophylline does not increase sympathetic activity in CHF, whereas increases in plasma renin and ventilation are still present. These novel findings have important implications for understanding the potential harmful and beneficial effects of theophylline and related substances on the pathophysiology of heart failure.
This study was supported by the Deutsche Forschungsgemeinschaft (An 260/2-2).
Braunwald E. The Denolin lecture. Congestive heart failure: a half century perspective. Eur Heart J. 2001; 22: 825–836.
Lanfranchi PA, Braghiroli A, Bosimini E, et al. Prognostic value of Cheyne-Stokes respiration in chronic heart failure. Circulation. 1999; 99: 1435–1440.
Sin DD, Logan AG, Fitzgerald FS, et al. Effects of continuous positive airway pressure on cardiovascular outcomes in heart failure patients with and without Cheyne-Stokes respiration. Circulation. 2000; 102: 61–66.
Bradley TD, Floras JS. Sleep apnea and heart failure, II: central sleep apnea. Circulation. 2003; 107: 1822–1826.
Hardman JG, Limbird LE, Molinoff PB, et al. Drugs used in the treatment of asthma. In: Goodman Gilman A, ed. The Pharmacological Basis of Therapeutics. 9th ed. New York, NY: McGraw-Hill; 1996: 672–682.
Funaya H, Kitakaze M, Node K, et al. Plasma adenosine levels increase in patients with chronic heart failure. Circulation. 1997; 95: 1363–1365.
Engelstein ED, Lerman BB, Somers VK, et al. Role of arterial chemoreceptors in mediating the effects of endogenous adenosine on sympathetic nerve activity. Circulation. 1994; 90: 2919–2926.
Javaheri S, Teppema LJ, Evers JA. Effects of aminophylline on hypoxemia-induced ventilatory depression in the cat. J Appl Physiol. 1988; 64: 1837–1843.
Suteparuk S, Nies AS, Andros E, et al. The role of adenosine in promoting cardiac beta-adrenergic subsensitivity in aging humans. J Gerontol A Biol Sci Med Sci. 1995; 50: B128–B134.
Corti R, Binggeli C, Sudano I, et al. Coffee acutely increases sympathetic nerve activity and blood pressure independently of caffeine content: Circulation. 2002; 106: 2935–2940.
Notarius CF, Atchison DJ, Rongen GA, et al. Effect of adenosine receptor blockade with caffeine on sympathetic response to handgrip exercise in heart failure. Am J Physiol. 2001; 281: H1312–H1318.
Quanjer PH, Tammeling GJ, Cotes JE, et al. Lung volumes and forced ventilatory flows: official position of the European Respiratory Society. Eur Respir J. 1993; 6 (suppl 16): 5–40.
Javaheri S, Evers JA, Teppema LJ. Increase in ventilation caused by aminophylline in the absence of changes in ventral medullary extracellular fluid pH and carbon dioxide tension. Thorax. 1989; 44: 121–125.
Sigurdsson A, Amtorp O, Gundersen T, et al. Neurohormonal activation in patients with mild or moderately severe congestive heart failure and effects of ramipril. Br Heart J. 1994; 72: 422–427.