Mechanism of the ventilatory response to carbon monoxide. T V Santiago, N H Edelman J Clin Invest. 1976;57(4):977-986. https://doi.org/10.1172/JCI108374. Research Article The effects of carbon monoxide on ventilation were studied in unanesthetized goats. Responses to single breaths of 10- 25% CO in O2, which rapidly raised carboxyhemoglobin (COHb) from 5 to 60%, were considered to reflect peripheral chemoreceptor-mediated reflexes whereas responses to continuous inhalation of 1% CO in O2, which slowly raised COHb from 0 to 60%, were considered to reflect both peripheral chemoreceptor and nonperipheral chemoreceptor mechanisms. In each of six goats, single breaths of CO failed to elicit any immediate ventilatory response. However, slow buildup of carboxyhemoglobinemia in the same animals always elicited ventilatory stimulation (from a mean of 7.43 to 16.02 liter/min, P less than 0.001) beginning 5-6 min after onset of 1% CO in O2 inhalation when COHb saturation reached 50-60%. In eight studies of six animals HCO3- concentration fell (from 21.3 to 15.8 meq/liter; P less than 0.001) and lactate concentration rose (from 2.5 to 4.2 meq/liter; P less than 0.05) in the cisternal cerebrospinal fluid during the CO-induced hyperpnea. Additional studies ruled out ventilatory stimulation from left heart failure or enhanced chemo- sensitivity to carbon dioxide. Although the delayed hyperpnea was associated with a hyperdynamic cardiovascular response to CO, blockade of these circulatory effects with propranolol (2 mg/kg) failed to abolish the delayed hyperpnea; however, the propranolol did unmask an element of ventilatory depression which preceded the […] Find the latest version: https://jci.me/108374/pdf Mechanism of the Ventilatory Response to Carbon Monoxide TEODORO V. SANriAwO and NORMAN H. EDELMAN From the Pulmonary Diseases Division, Department of Medicine, College of Medicine and Dentistry of New Jersey-Rutgers Medical School, Piscataway, New Jersey 08854 A B S T R A C T The effects of carbon monoxide on to peripheral chemoreceptor stimulation. The ventilatory ventilation were studied in unanesthetized goats. Re- response is less clear when arterial oxygen content is sponses to single breaths of 10-25% CO in 02, which decreased in the presence of normal oxygen tensions. rapidly raised carboxyhemoglobin (COHb) from 5 to One such state, carboxyhemoglobinemia, has attracted 60%, were considered to reflect peripheral chemorecep- considerable attention in previous attempts to answer tor-mediated reflexes whereas responses to continuous this question. No consistent pattern has emerged from inhalation of 1% CO in 02, which slowly raised COHb these studies. John Haldane (1), for example, in a from 0 to 60%, were considered to reflect both periph- famous experiment performed on himself, hyperventi- eral chemoreceptor and nonperipheral chemoreceptor lated after carbon monoxide inhalation. This was con- mechanisms. In each of six goats, single breaths of CO firmed by Haggard and Henderson in dogs (2), and failed to elicit any immediate ventilatory response. more recently by Ayres et al. in man (3), but Chiodi However, slow buildup of carboxyhemoglobinemia in et al. (4) found no ventilatory changes in human sub- the same animals always elicited ventilatory stimulation jects with carboxyhemoglobin (COHb)1 levels as high (from a mean of 7.43 to 16.02 liter/min, P < 0.001) as 50%. Similarly, there is no agreement as to whether beginning 5-6 min after onset of 1% CO in 02 inhala- or not carboxyhemoglobinemia increases neuronal ac- tion when COHb saturation reached 50-60%. In eight tivity from the peripheral chemoreceptors (5-7). studies of six animals HCOs- concentration fell (from In a recent study of a related problem, the ventila- 21.3 to 15.8 meq/liter; P <0.001) and lactate concen- tory response to hypoxia in severe anemia (8), we tration rose (from 2.5 to 4.2 meq/liter; P < 0.05) in pointed out the complexities inherent in experiments of the cisternal cerebrospinal fluid during the CO-induced this nature. Our data indicated that while diminished hyperpnea. Additional studies ruled out ventilatory 02 content of arterial blood did not seem to be a stimulation from left heart failure or enhanced chemo- ventilatory stimulus per se, it could substantially alter sensitivity to carbon dioxide. Although the delayed both peripheral chemoreceptor- and central nervous hyperpnea was associated with a hyperdynamic cardio- system-mediated responses to hypoxic hypoxia. Thus, vascular response to CO, blockade of these circulatory the clinical significance of defining the ventilatory re- effects with propranolol (2 mg/kg) failed to abolish sponse to carbon monoxide goes beyond the occasional the delayed hyperpnea; however, the propranolol did case of carbon monoxide poisoning since it is likely to unmask an element of ventilatory depression which pre- shed light upon ventilatory adjustments to a variety of ceded the hyperpnea. Conclusions were: (a) hyper- situations in which the central nervous system is de- ventilation in response to CO inhalation is not mediated prived of its normal supply of oxygen. by the carotid bodies; (b) the delayed hyperpnea in In the current study, we attempted to understand the response to CO inhalation is primarily due to brain- mechanisms that determine the ventilatory responses to cerebrospinal fluid acidosis; (c) mobilization of body carbon monoxide of intact unanesthetized animals. We CO2 stores due to the circulatory response to CO may used a methodologic approach which allowed for dis- obscure an initial depression of ventilation by CO. tinction between peripheral chemoreceptor- and central nervous system-mediated responses. This was made INTRODUCTION possible by taking advantage of the previously demon- The increase in ventilation caused by reduction in strated differences in time of onset of these effects. arterial 02 tension is well known and clearly ascribed Peripheral chemoreceptor-mediated responses, which Received for publication 12 August 1975 and in revised 1Abbreviations used in this paper: COHb, carboxyhemo- form 6 November 1975. globin; CSF, cerebrospinal fluid. The Journal of Clinical Investigation Volume 57 April 1976 -977-986 977 take less than 10 s to become manifest (9), were tested parison. In these studies, involving the same six animals, by inhalations of single breaths of CO in 02. The com- decreases in arterial 02 saturation were produced by tran- bined peripheral chemoreceptor- and central nervous sient inhalations of nitrogen. During each period of nitrogen breathing (2-8 breaths), ventilation and oxygen saturation system-mediated responses (the latter take at least a (0-500 cuvette oximeter, Waters Instruments, Inc., Ro- minute to develop) (10) were tested by continuous in- chester, Minn.) were continuously recorded. To quantitate halation of 1% CO in 40% 02. To confirm the validity the responses to transient N2 inhalation the maximal in- of this approach, and to elucidate the mechanisms in- crease in ventilation was related to the maximal fall in O2 saturation for each period of inhalation of the test gas. The volved in the ventilatory responses that were observed, rationale and details of this method have been previously additional studies were done. They involved denerva- described (8). tion of the carotid bodies, monitoring of cerebrospinal The results were plotted together, comparing the ventila- fluid (CSF) acid-base changes, and evaluation of the tory responses to decreases in arterial 02 saturation achieved with either nitrogen or carbon monoxide inhalation. Failure effects of the concomitant circulatory changes during to increase ventilation within 30 s of inhalation of the test inhalation of carbon monoxide. The data support the gas was considered to signify lack of response to transient view that acute carboxyhemoglobinemia does not stimu- hypoxia. late peripheral chemoreceptors and provide a basis for Extended carbon monoxide breathing. On separate days, the same six animals breathed 40% 02 in N2 for 10 min a multifactorial concept of the ventilatory effects of followed by 1% CO in 40% 02 plus N2 for 7-13 min. Ven- central nervous system hypoxia. tilation, arterial blood pressure, and heart rate were moni- tored as before. Arterial blood was sampled every minute and measurements of pH, gas tensions, arterial So2, and METHODS COHb were carried out as described above. While the 28 goats were studied unanesthetized and in the standing total duration of the study varied from animal to animal position. During the week before performance of the ex- the end point was consistent in each animal; this was the periments, an indwelling arterial catheter was inserted into development of sustained hyperpnea (see Results). This end a femoral artery under general anesthesia. During the point was chosen after preliminary studies showed that if study, this catheter was connected by means of a three-way animals were allowed to breathe CO beyond the establish- stopcock to a calibrated blood pressure transducer (model ment of sustained hyperventilation, apnea ensued within 2-4 P23 Dd, Statham Instruments Div., Gould, Inc., Oxnard, min. After the inhalation of 1% CO they resumed breath- Calif.). This system enabled continuous monitoring of heart ing 40% 02 in N2. rate and arterial blood pressure; it also provided easy ac- We then tried to determine the mechanisms responsible cess to arterial blood with no disturbance to the animal. The for the delayed hyperpnea which was observed during ex- animals were studied standing and lightly restrained by the tended inhalation of CO. The following possibilities were horns to a stock. They were allowed to breathe through a entertained: (a) heightened medullary chemoreceptor sensi- snugly fitting mask with an attached Rudolph one-way tivity causing an enhanced response to C02; this was con- valve. A Fleisch pneumotachograph was interposed in the sidered as a possibility since another form of hypoxia (in- inspiratory line, and attached to a (Statham PM 15) differ- halation of gases with low oxygen tension) is known to en- ential pressure transducer.