Dysregulation of Peripheral and Central Chemoreflex Responses in Chagas’ Heart Disease Patients Without Heart Failure
Background The peripheral and central chemoreflexes are important autonomic mechanisms for regulating breathing and cardiovascular function. Although pathological inflammatory infiltration of the peripheral chemoreceptors and central nervous system has been reported in Chagas’ disease, functional evaluation of chemoreflexes has not yet been performed.
Methods and Results The hypothesis that chemoreflex function is altered in patients with Chagas’ heart disease (CH) but normal left ventricle function was tested in 12 CH patients and 13 matched control subjects. The ventilatory rate, minute ventilation, heart rate, mean arterial pressure, forearm blood flow, forearm vascular resistance, and venous norepi-nephrine responses to hypoxia and hypercapnia were determined. During hypoxia, the decrease in oxygen saturation was smaller in CH patients, despite a similar ventilatory response between groups. Both groups showed an increase in heart rate during hypoxia, but this response was blunted in CH patients. Although the mean arterial pressure response to hypoxia was similar in both groups, forearm vascular resistance significantly decreased in control subjects while remaining unchanged in CH patients. Moreover, a significant increase in plasma norepinephrine levels elicited by stimulation of peripheral chemoreceptors was observed only in the CH group. During hypercapnia, the increase in minute ventilation was smaller in CH patients, who did not exhibit the increase in norepinephrine observed in control subjects.
Conclusions These data suggest that CH potentiates respiratory, cardiovascular, and autonomic responses to peripheral chemoreceptor activation by hypoxia in patients with normal left ventricular function. The ventilatory and sympathetic responses to central chemoreceptor activation by hypercapnia, however, are significantly blunted.
Received May 8, 2001; revision received July 24, 2001; accepted July 31, 2001.
American trypanosomiasis (Chagas’ disease, ChD) is associated with pathological changes involving the heart and autonomic nervous system.1–3 Impaired autonomic control of the cardiovascular system is one of the hallmarks of ChD.4–7 The role of the autonomic nervous system in the pathophysiological and clinical evolution of the cardiac alterations remains unclear.8,9 The peculiar involvement of the autonomic nervous system in Chagas’ heart disease (CH) causes abnormalities in neurogenic cardiovascular control before left ventricle dysfunction or overt heart failure even appears.4–9
Responses to hypoxia are mediated mostly by the peripheral chemoreceptors, located in the carotid sinus; responses to hypercapnia are mediated primarily by the central chemoreceptors, located on the ventral surface of the medulla.10–12 The autonomic nervous system constitutes a key component of the complex cardiovascular and respiratory responses to hypoxia and hypercapnia.10–12 Pathological changes of the peripheral chemoreceptors and central nervous system have been documented in chagasic patients.2,13 Baroreflex impairment has also been reported in patients with CH.4,9 The arterial baroreflex has profound modulating influence on chemoreceptor function.14 These facts support the hypothesis that chemoreflex function may be disturbed in patients with CH. Because no previous studies involving CH patients have investigated chemoreflex function or the neurogenic respiratory control, the purpose of the present study was to evaluate the respiratory, cardiovascular, and neurohormonal responses of such patients to peripheral and central chemoreceptor activation. Because of the potential for coexisting heart failure to disturb chemoreflex function,15,16 chemoreflex responses were studied only in asymptomatic patients with CH but normal left ventricular systolic function as well as carefully matched control subjects.
We studied 12 patients with CH (mean age 43±2 years, range 32 to 52 years; 8 women and 4 men; CH group) diagnosed by serological tests and typical ECG evidence of heart involvement. These patients had been referred to the Heart Institute of the University of São Paulo for evaluation of suspected CH on the basis of their epidemiological histories and abnormal ECG findings. On clinical examination, none of them had any signs of congestive heart failure, and all had a resting ejection fraction of ≥0.60 (mean ejection fraction 0.70±1.3, range 0.60 to 0.75) determined via echocardiography. Clinical and laboratory data showed no evidence of other disease. The risk of including a patient with endothelial dysfunction was minimized by exclusion of patients with hypertension and diabetes.
The control group consisted of 13 normal subjects (mean age 38±2 years, range 22 to 55 years; 7 women and 6 men) matched with the CH patients with regard to age, sex, race, body mass index, and blood pressure. All subjects provided informed written consent, and the study was approved by the Ethics Committee of the Heart Institute of the University of São Paulo.
Finger arterial pressure was continuously monitored by the volume-clamp method (Finapres, Ohmeda 2300, Ohmeda Monitoring Systems). Heart rate was monitored by electrocardiography. Forearm blood flow was measured in the right arm by venous occlusion plethysmography (D. E. Hokansen, model EC-4) with a mercury-filled Silastic strain gauge. Forearm flow measurements were recorded for intervals of ≈10 seconds every 20 seconds. Three readings were obtained for each mean minute value. Forearm vascular resistance was calculated as the mean arterial pressure during a 1-minute record divided by the mean minute forearm blood flow.
End-tidal CO2 and oxygen saturation were monitored with a capnograph and pulse oximeter (Novametrix, model 7100 CO2SMO ETCO2/SpO2 Monitor, Novametrix Medical Systems Inc). Ventilatory rate and minute ventilation were monitored with a pneumotachograph (Heated Pneumotachometer, Hans-Rudolph, Inc) and a differential pressure transducer (MP 45-871 Validyne, Engineering Corp) linked to a signal integrator amplifier (Gould Instruments Systems, Inc).
Plasma norepinephrine levels were measured via high-performance liquid chromatography.
All signals were recorded simultaneously on a Gould strip-chart recorder (RS 3800, Gould Instruments Systems, Inc) and on a computer using customized CODAS software (Computer Operated Data Acquisition Software; AT-CODAS; DATAQ Instruments).
Experimental Protocol and Procedures
We used an established protocol according to Somers et al.17,18 All 25 subjects participated in all components of the study, which was performed with subjects in the supine position in a quiet room. First, a catheter was inserted into the antecubital vein. Studies were initiated after a ≥30-minute rest period, during which all subjects were familiarized with the experimental techniques. Measurements were obtained for 3 minutes while the subjects breathed room air (baseline recording) and then for 5 minutes (analyzed with a minute-by-minute format) during exposure to the gas mixture (as follows) via a mouthpiece and a noseclip to ensure exclusive mouthbreathing. At the end of each recording session (baseline and chemoreceptor stimulation), a 4-mL blood sample was collected to measure the plasma norepinephrine level by a catheter previously inserted into the antecubital vein. We used the same sequence of stimulation for all subjects, namely (1) peripheral chemoreceptor stimulation: 10% O2 in N2 with titrated CO2 to maintain isocapnia; and (2) central chemoreceptor stimulation: 7% CO2 plus 93% O2. At least 20 minutes was allowed to elapse between peripheral chemoreceptor stimulation and initiation of the central chemoreflex protocol.
Statistical analyses were performed by an independent statistician with SAS software (SAS Institute Inc). Demographic data, baseline characteristics, the peripheral chemosensitivity index (PCI), and norepinephrine levels were compared by the unpaired t test. The sex distribution was compared by the Fisher exact test.
Responses to both hypoxia and hypercapnia were analyzed by repeated-measures ANOVA with time (baseline versus intervention) as the within factor and group (control versus CH) as the between factor. The key variable was the group-by-time interaction. Because of the significant difference in oxygen saturation response between group-by-time interactions during hypoxia, oxygen saturation was used as a covariate of the other responses to compare control versus CH. The PCI was calculated as the change in minute ventilation divided by the change in oxygen saturation corresponding to minutes 2, 3, 4, and 5 to determine differences between groups.19
Data are presented as the mean±SEM. A value of P<0.05 was considered significant.
The mean age, body mass index, sex distribution, and arterial blood pressure of CH patients were similar to those of control subjects (Table 1).
Mean oxygen saturation, end-tidal CO2, ventilatory rate, minute ventilation, mean arterial pressure, heart rate, forearm blood flow, forearm vascular resistance, and norepinephrine level in CH patients were similar to values in the control subjects during both baseline measurements (Table 2).
Responses to Isocapnic Hypoxia
The magnitude of the hypoxia-induced decrease in oxygen saturation was smaller in CH patients (P=0.01) (Table 3 and Figure 1); however, no significant difference in the end-tidal CO2 values was observed when those for baseline versus hypoxia or CH patients versus control subjects were compared (Table 3). Control subjects increased their ventilatory rates during the 5 minutes of hypoxia, but no changes occurred in CH patients (Table 3). CH patients and control subjects showed similar increases in minute ventilation during hypoxia (Table 3). The PCI at minute 2 of hypoxia (PCI2), however, was significantly higher in the CH group (0.660±0.13 L · min−1 · % So2−1) than the control group (0.283±0.03 L · min−1 · % So2−1) (P=0.01) (Figure 2).
The hypoxia-induced increase in heart rate was significantly blunted in CH patients compared with control subjects (P=0.02) (Table 3 and Figure 1); however, no changes in the mean arterial pressure were observed during hypoxia in either CH patients or control subjects (Table 3). Hypoxia induced a significant increase in forearm blood flow and decrease in forearm vascular resistance in control subjects; no changes in these parameters occurred in CH patients. Comparison between groups reached significance for both the forearm blood flow (P=0.02) and forearm vascular resistance (P=0.02) responses (Table 3) (Figure 1). Norepinephrine levels increased only in CH patients during hypoxia (P=0.01) (Table 3 and Figure 1).
Responses to Hypercapnia With Hyperoxia
The magnitudes of the oxygen saturation, end-tidal CO2, mean arterial pressure, heart rate, forearm blood flow, and forearm vascular resistance responses to hypercapnia with hyperoxia were similar for CH patients and control subjects (Table 4).
During hypercapnia, the ventilatory rate increased only in control subjects, but both groups showed increases in minute ventilation (Table 4). The increase in minute ventilation during hypercapnia, however, was significantly smaller in CH patients (P=0.02, Table 4 and Figures 3 and 4⇓). Norepinephrine levels increased in control subjects during hypercapnia (P=0.001) but did not change significantly in CH patients (P=0.8910) (Table 4 and Figure 3). Norepinephrine levels during hypercapnia decreased by −2.5±51.8 pg/mL in CH patients and increased by 51.8±41.3 pg/mL in control subjects (P=0.01) (Figure 3).
This study is the first to investigate the integrated respiratory, hemodynamic, and autonomic responses to chemoreflex activation in individuals with CH. The major finding is that CH patients without left ventricle dysfunction demonstrate marked dysregulation of both the peripheral and central chemoreflex responses.
During peripheral chemoreflex stimulation, the magnitude of the ventilatory response exhibited by CH patients during the 5-minute period of the protocol was similar to that of control subjects. Calculation of the PCI, however, demonstrated that the sensitivity at minute 2 was significantly greater in CH patients, suggesting an enhanced ventilatory response to peripheral chemoreflex stimulation in this group.
The potentiation of peripheral chemoreflex sensitivity may explain why plasma norepinephrine levels increased only in CH patients during hypoxia and why the hypoxia was not accompanied by forearm vasodilation, despite similar mean arterial pressure responses.
The forearm vascular response during hypoxia in humans is extremely complex and dependent on neural (primary and secondary reflexes activated during hypoxia) and endothelial factors.17,18,20–22 Despite the increase in muscle sympathetic nerve activity observed during hypoxia,17,18 the net response in normal humans is a decrease in vascular resistance in muscle territories,20 as observed in the control group.
In our study, no such decrease in forearm vascular resistance was observed in CH patients, indicative of an abnormal vascular response to hypoxia, with a predominance of sympathetic activity over local mechanisms. Unfortunately, we did not evaluate the sympathetic response to hypoxia by measuring muscle sympathetic nerve activity, as described by Somers et al17,19; instead, we measured plasma norepinephrine levels.23 A result similar to ours, however, has been reported by Rowell et al,24 who demonstrated that plasma norepinephrine levels do not increase in normal humans during hypoxia. Collectively, these findings, as well as the reported good correlation between acute venous plasma norepinephrine changes and sympathetic traffic measured by microneurography,25 strongly suggest that enhanced sympathetic activity during hypoxia may have counteracted the effects of local vasodilatory factors in the CH group.
Classically, ChD is associated with a blunted heart rate response to several physiological stimuli.6,7,26 Guzetti et al,27 studying heart rate variability responses to different stimuli with power spectral analysis, found indirect signs of a reduced capability for activation of cardiac sympathetic and also an impairment of vagal modulatory influences.
With regard to the divergence between a blunted heart rate and an enhanced sympathetic activity observed during hypoxia, it is noteworthy that patients with CH and a chronotropic deficiency increase their systemic vascular resistance during hand-grip maneuvers to compensate for the absence of an increase in cardiac output.26 We hypothesize that patients with CH lack the capacity to increase their cardiac output during the hypoxic stress (our patients exhibited a blunted tachycardia) and thus need to increase sympathetic outflow (which prevents vasodilation) to maintain a stable mean arterial pressure. The potential long-range implications of such autonomic dysregulation of acute afterload adjustments in regard to cardiac function and the evolution of heart failure deserve further attention.
In contrast to the increased response to hypoxia, patients with CH demonstrated depressed respiratory rate and minute ventilation responses to hypercapnia compared with control subjects. An earlier study by Guz et al28 suggests that the vagal and glossopharyngeal nerve damage associated with ChD 29 may be implicated in the decreased central chemosensitivity observed in the present study. They showed a significant depression of both the ventilatory response and respiratory discomfort after bilateral blockade of the vagus and glossopharyngeal nerves with lidocaine in a normal subject.28
Atropine blockade of the rostral ventrolateral medulla muscarinic receptors, which help regulate central chemosensitivity, produced a significant decrease in the ventilatory response to hypercapnia in one study.30 Because human chagasic IgG can block muscarinic receptors in the heart,31 the observed depression of central chemosensitivity in our CH patients could also be explained by the blockade of rostral ventrolateral medulla muscarinic receptors by similar antibodies.
Finally, it should be stressed that, as observed during hypoxia, the alteration in ventilatory response during hypercapnia changed in the same direction as the alteration in the sympathetic response, a finding that has already been reported by Narkiewicz et al.15,32 Thus, the ventilatory and autonomic responses in the CH group were enhanced with peripheral chemoreflex stimulation and depressed with central chemoreflex stimulation.
In summary, the increased peripheral chemosensitivity at minute 2, the absence of vasodilation, and the significant increase in plasma norepinephrine levels observed only in the CH patients during hypoxia suggest that individuals with CH without heart failure (or lesser left ventricular dysfunction) experience peripheral chemoreflex potentiation. In contrast, the blunted ventilatory response and absent sympathetic response during hypercapnia observed in our CH patients indicate that CH is associated with diminished ventilatory and autonomic responses to central chemoreceptor activation. Moreover, the abnormal chemoreflex responses already present in patients without significant left ventricular damage may contribute to the development of heart failure.
The study was supported by the E.J. Zerbini Foundation and Fundação de Amparo à Pesquisa do Estado de São Paulo.
Brandão HJS, Zulian R. Nerve cell depopulation in chronic Chagas’ disease: a quantitative study in the cerebellum. Inst Med Trop São Paulo. 1966; 8: 281–286.
Lopes ER, Tafuri WL. Involvement of the autonomic nervous system in Chagas’ heart disease. Rev Soc Bras Med Trop. 1983; 16: 206–212.
Amorim DS, Godoy RA, Manço JC, et al. Effects of acute elevation in blood pressure and atropine on heart rate in Chagas’ disease: a preliminary report. Circulation. 1968; 38: 289–294.
Sousa AC, Marin-Neto JA, Maciel BC, et al. Cardiac parasympathetic impairment in gastrointestinal Chagas’ disease. Lancet. 1987; 1: 985.
Köberle F. Cardiopatia parasimpaticopriva. Munch Med Wochenschr. 1959; 101: 1308–1310.
Morris AS, Tanowitz HB, Wittner M, et al. Pathophysiological insights into the cardiomyopathy of Chagas’ disease. Circulation. 1990; 82: 1900–1909.
Marshall JM. Peripheral chemoreceptors and cardiovascular regulation. Physiol Rev. 1994; 74: 543–594.
Bianchi AL, Denavit-Saubié M, Champgnat J. Central control of breathing in mammals: neuronal circuitry, membrane properties, and neurotransmitters. Physiol Rev. 1995; 75: 1–45.
Wade JG, Larson CP Jr, Hickley RF, et al. Effect of carotid endarterectomy on carotid chemoreceptor and baroreceptor function in man. N Engl J Med. 1970; 282: 823–829.
Alcântara FG. Pesquisa dos Corpos Caróticos na Moléstia de Chagas. Rev Bras de Med. 1970; 27: 291–296.
Somers VK, Mark AL, Abboud FM. Interaction of baroreceptor and chemoreceptor reflex control of sympathetic nerve activity in normal humans. J Clin Invest. 1991; 87: 1953–1957.
Narkiewicz K, Pesek CA, van de Borne PJH, et al. Enhanced sympathetic and ventilatory responses to central chemoreflex activation in heart failure. Circulation. 1999; 100: 226–267.
Somers VK, Mark AL, Zavalla DC, et al. Influence of ventilation and hypocapnia on sympathetic nerve response to hypoxia in normal humans. J Appl Physiol. 1989; 67: 2095–2100.
Somers VK, Mark AL, Zavalla DC, et al. Contrasting effects of hypoxia and hypercapnia on ventilation and sympathetic activity in humans. J Appl Physiol. 1989; 67: 2101–2106.
Blitzer ML, Lee SD, Creager MA. Endothelial-derived nitric oxide mediates hypoxic vasodilation of resistance vessels in humans. Am J Physiol. 1996; 271: H1182–H1185.
Leuenberg U, Gray K, Herr MD. Adenosine contributes to hypoxia-induced forearm vasodilation in humans. J Appl Physiol. 1999; 87: 2218–2224.
Rowell LB, Johson DG, Chase PB, et al. Hypoxemia raises muscle sympathetic activity but not norepinephrine in resting humans. J Appl Physiol. 1989; 66: 1736–1743.
Rea FR, Eckberg DL, Fritsch JM, et al. Relation of plasma norepinephrine and sympathetic traffic during hypotension in humans. Am J Physiol. 1990; 258: R982–R986.
Marin-Neto JA, Maciel BC, Gallo L Jr, et al. Effect of parasympathetic impairment on the haemodynamic response to handgrip in Chagas’ heart disease. Br Heart J. 1986; 55: 204–210.
Lopes ER, Tafuri WL, Chapadeiro E. Estudo morfológico e quantitativo dos núcleos dorsal do vago e hipogloso em chagásicos crônicos, com e sem megaesôfago. Rev Inst Med Trop São Paulo. 1969; 11: 123–129.
Nattie EE, Wood J, Mega A, et al. Rostral ventrolateral medulla muscarinic receptor involvement in central ventilatory chemosensitivity. J Appl Physiol. 1989; 66: 1462–1470.
Narkiewicz K, van de Borne PJH, Pesek CA, et al. Selective potentiation of peripheral chemoreflex sensitivity in obstructive sleep apnea. Circulation. 1999; 99: 1183–1189.