A prediction-correction scheme for forcing alveolar gases along certain time courses.

A computerized prediction-correction scheme has been devised for the control of alveolar gases. First, a model is run off-line to predict the inspiratory gas tensions at each second that should yield the desired alveolar patterns. Second, during the experiment, there is feedback correction based on the deviation of the actual alveolar values from the desired alveolar values. The actual alveolar values are found by a second computer and passed to the controlling computer using interrupts. The controlling computer has four digital-toi-analog outputs for controlling CO2, O2, N2, and air flows so as to achieve the commanded inspiratory PCO2 and PO2 (CO2 and O2 partial pressures, respectively). The scheme is illustrated for the generation of sinusoidal alveolar PCO2 with alveolar PO2 held constant and for steps of alveolar PCO2 at constant alveolar PO2.

The respiratory effects in man of altering the time profile of alveolar carbon dioxide and oxygen within each respiratory cycle.

1. Breathing hypoxic gas through an external dead space (ca. 1200 c.c.) stimulated ventilation disproportionately. A loop (ca. 250 c.c.) in the inspiratory pathway reduced the effect.2. The alveolar time patterns of P(CO) (2) and P(O) (2) characteristic of tube breathing with or without the loop have been simulated in moderate hypoxia by changing the composition of inspired gas at selected intervals after the beginning of inspiration.3. Supplying CO(2)-free gas in late inspiration usually stimulated ventilation, but less than did real tube breathing. Supplying CO(2)-free gas early in inspiration usually depressed ventilation. The difference between the ;CO(2)-free late' and ;CO(2)-free early' effects was 20% of the control ventilation (P < 0.001), i.e. was nearly the same as between the effects of real tube breathing without and with the loop.4. Tube-like P(A, O) (2) time patterns had no effects.5. A-a P(CO) (2) and P(O) (2) gradients remained constant throughout.6. The V(E), f and V(T) relations were unaltered in tube breathing.7. The respiratory system can discriminate between small differences in time patterns of P(A, CO) (2) but not of P(A, O) (2); the signal is amplified by steady hypoxia. The arterial chemoreceptors are probably responsible for these effects.

Patterns of breathing in response to alternating patterns of alveolar carbon dioxide pressures in man.

The time profile of alveolar PCO2 within the respiratory cycle has been forced to follow contrasting patterns in alternate breaths, in two different ways. Within-breath changes (w.b.c., with a CO2-rich inspirate supplied early or late in alternate inspirations) involved minimal alternation of end-tidal PCO2. Between-breath changes (b.b.c., with whole inspirates of CO2-free or CO2-rich gas) involved large swings of end-tidal PCO2. As previously reported (Metias, Cunningham, Howson, Petersen & Wolff, 1981), both patterns of forcing were associated with alternation of ventilation, but only when hypoxia was present. The patterns of the alternating reflex responses in 118 runs on four human subjects in steady hypoxia are described in terms of alternation of inspiratory and expiratory tidal volume, time and mean flow. These patterns often disappeared, or changed unpredictably in mid-run. The inspiratory pattern of reflex alternation depended in part on the type of forcing, but alternation of inspiratory tidal volume was usually observed with both types. No single pattern of expiratory alternation emerged as predominant. The pattern of reflex expiratory alternation was surprisingly independent of the pattern of inspiratory alternation: indeed, in w.b.c., but not in b.b.c., alternation of mean expiratory flow and of mean inspiratory flow were mutually exclusive. It is concluded that in man, as in cats and dogs, the arterial chemoreceptor pathway has access to various parts of the respiratory pattern generator, the exact response depending to some extent on the timing within the respiratory cycle. In particular, expiratory variables may be influenced directly through the arterial chemoreceptor pathway, i.e. without any supposedly mediating inspiratory alternation being demonstrable. The results are discussed briefly in relation to some current views on the organization of respiratory control.

Reflex effects on human breathing of breath-by-breath changes of the time profile of alveolar PCO2 during steady hypoxia.

The respiratory effects of forced changes of alveolar PCO2 were studied in four healthy human subjects and in one anaesthetized cat. Solenoid valves, triggered by changes in mouth pressure, allowed changes from one inspiratory gas mixture to another, either during expiration (between-breath changes, BBC) or in the middle of inspiration (within-breath changes, WBC). In BBC the subject breathed CO2-free gas in one inspiration, CO2-rich gas in the next, and so on; end-tidal PCO2 alternated regularly from breath to breath by 1.1 kPa. In WBC CO2-free gas was given early in one inspiration and late in the next, with CO2-rich gas late in the former and early in the latter, and so on end-tidal PCO2 was nearly constant from breath to breath. Eight respiratory output variables were analysed. WBC induced small but significant alternation in most of the variables; these effects occurred almost exclusively in runs in hypoxia. The responses were not very different from those seen in BBC. The experiment on the cat showed that the alveolar PCO2 changes predicted during WBC are reflected by changes in pH in the arterial blood. The results confirm predictions based upon observations in the steady state of tube- and reversed-tube breathing in man. It seems likely that the responses are mediated by the arterial chemoreceptors responding to small changes in the profile of the (CO2, H+) oscillation.

The effects of hypercapnia, hypoxia and ventilation on the baroreflex regulation of the pulse interval.

1. The effects of moderate degrees of hypercapnia in hypoxia and in hyperoxia on the baroreceptor-cardiodepressor reflex have been studied in nine normal men.2. The beat-by-beat relation between pulse interval (I) and systolic pressure (P) during transient elevations of arterial pressure induced by intravenous injections of angiotensin II and phenylephrine was used to assess the sensitivity (DeltaI/DeltaP) and setting (I at a single reference arterial pressure) of the reflex.3. There was no consistent change in reflex sensitivity in any of the conditions studied.4. In hyperoxia (P(A, O2) approximately 200 torr) hypercapnia was associated with significant re-setting of the reflex in the direction of tachycardia. The extent of the re-setting was correlated with the degree of hypercapnia and with the accompanying increase in breathing.5. When hyperoxia with hypercapnia was replaced by hypoxia (P(A, O2) approximately 55 torr) with hypercapnia (which causes substantial arterial chemoreceptor activity), pulse interval at constant arterial pressure was further decreased.6. The tachycardia of hypoxia could not be accounted for by change of arterial pressure, P(A, CO2) or pulmonary ventilation, since it was most clearly demonstrable at constant values of pressure and either P(A, CO2) or ventilation.

The effects of raising alveolar PCO2 and ventilation separately and together on the sensitivity and setting of the baroreceptor cardiodepressor reflex in man.

1. The effects of changes in ventilation and/or alveolar P(CO2) on the baroreflex control of heart rate have been studied in seven experiments on six young men and women who had been trained to control tidal volume and respiratory frequency at various levels independent of alveolar P(CO2), during hyperoxia.2. Intravenous phenylephrine provoked transient rises of directly measured arterial pressure during which individual systolic pressures (P) were linearly related to the following pulse interval (I). Baroreflex sensitivity was expressed as the slope of the regression of I on P, and reflex setting (I(ref)) as I at a single reference arterial pressure (= mean P for the experiment).3. Voluntary control of breathing had little effect on heart rate and arterial pressure (baroreflex setting), but diminished reflex sensitivity.4. Hypercapnia regularly caused tachycardia at the reference pressure (i.e. baroreflex setting lowered). The response was completely or partly reproduced by change of P(A, CO2) at constant ventilation in four subjects but not in two others; in them change of ventilation at constant P(A, CO2) completely mimicked the effect of free-breathing hypercapnia.5. Values of baroreflex sensitivity were relatively scattered. Hypercapnia caused a fall in baroreflex sensitivity in three subjects whether ventilation was fixed or free to rise. After separating the effect of voluntarily controlling ventilation, ventilation per se was without effect on reflex sensitivity.6. It is concluded that hypercapnia and hyperpnoea have separate effects on the baroreflex, the relative magnitudes of which differ from one subject to another. Baroreflex setting and sensitivity vary independently in response to change of ventilation and of P(A, CO2).