AJP - Heart Ad Instruments
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 274: H2177-H2187, 1998;
0363-6135/98 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Henry, R. A.
Right arrow Articles by Eckberg, D. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Henry, R. A.
Right arrow Articles by Eckberg, D. L.
Vol. 274, Issue 6, H2177-H2187, June 1998

MODELING IN PHYSIOLOGY
Interactions between CO2 chemoreflexes and arterial baroreflexes

Rebecca A. Henry, I-Li Lu, Larry A. Beightol, and Dwain L. Eckberg

Departments of Medicine and Physiology, Hunter Holmes McGuire Department of Veterans Affairs Medical Center, and Medical College of Virginia of Virginia Commonwealth University, Richmond, Virginia 23249

    ABSTRACT
Top
Abstract
Introduction
Methods
Results
Discussion
References

We studied interactions between CO2 chemoreflexes and arterial baroreflexes in 10 supine healthy young men and women. We measured vagal carotid baroreceptor-cardiac reflexes and steady-state fast Fourier transform R-R interval and photoplethysmographic arterial pressure power spectra at three arterial pressure levels (nitroprusside, saline, and phenylephrine infusions) and three end-tidal CO2 levels (3, 4, and 5%, fixed-frequency, large-tidal-volume breathing, CO2 plus O2). Our study supports three principal conclusions. First, although low levels of CO2 chemoreceptor stimulation reduce R-R intervals and R-R interval variability, statistical modeling suggests that this effect is indirect rather than direct and is mediated by reductions of arterial pressure. Second, reductions of R-R intervals during hypocapnia reflect simple shifting of vagally mediated carotid baroreflex responses on the R-R interval axis rather than changes of baroreflex gain, range, or operational point. Third, the influence of CO2 chemoreceptor stimulation on arterial pressure (and, derivatively, on R-R intervals and R-R interval variability) depends critically on baseline arterial pressure levels: chemoreceptor effects are smaller when pressure is low and larger when arterial pressure is high.

defense reaction; vagal baroreceptor; sympathetic innervation; hyperventilation

    INTRODUCTION
Top
Abstract
Introduction
Methods
Results
Discussion
References

CHEMOREFLEXES and baroreflexes are intimately intertwined on a variety of levels. Afferent fibers from peripheral chemoreceptors and arterial baroreceptors terminate in the solitary tract nucleus (28) [in some instances, on the same cells (22)], where there are important opportunities for reflex interactions to occur (28). Inputs from chemoreceptors and baroreceptors may vary simultaneously. Examples include high-altitude exercise (hypertension, hypoxemia, and hypocapnia), hemorrhage (hypotension and hypoxemia), and O2 treatment of patients with lung disease (hyperoxia and hypercapnia). Moreover, inputs from chemoreceptors may alter baroreflex function, and inputs from baroreceptors may alter chemoreflex function.

We evaluated chemoreflex-baroreflex interactions in healthy young men and women who voluntarily breathed at a constant respiratory rate and a large tidal volume, while we maintained end-tidal CO2 levels at about 3, 4, and 5% and arterial pressures at low, usual, and high levels. We evaluated our results with a statistical model and drew three main conclusions. First, although low levels of CO2 chemoreceptor stimulation reduce R-R intervals and R-R interval variability, this effect appears to be indirect, rather than direct, and is mediated by reductions of arterial pressure. Second, reductions of R-R intervals secondary to reduced CO2 chemoreceptor stimulation represent simple resetting of baroreflex responses rather than changes of baroreflex gain, range, or operational point. Third, the influence of CO2 chemoreceptor stimulation on arterial pressure (and, derivatively, on R-R intervals and R-R interval variability) depends critically on baseline arterial pressure levels: chemoreceptor effects are smaller when pressure is low and larger when arterial pressure is high.

    METHODS
Top
Abstract
Introduction
Methods
Results
Discussion
References

Subjects

Five men and five women (ages 23-26 yr), all healthy and nonsmoking, participated. The human research committees of the Medical College of Virginia of Virginia Commonwealth University and the Hunter Holmes McGuire Department of Veterans Affairs Medical Center approved this research. Each subject gave his or her informed written consent before participation. Volunteers were screened to exclude those with histories of diabetes and cardiovascular disorders. No subject had lung disease or breathing disorders, including sleep apnea. Subjects were taking no medications and had refrained from taking alcohol or caffeine and doing strenuous exercise during the 24 h preceding study sessions.

Measurements

R-R intervals were measured from digitized surface electrocardiograms. Beat-by-beat arterial pressure was measured continuously with a photoplethysmogram (Finapres model 2300, Ohmeda, Englewood, CO) in the middle phalanx of the middle finger, and cuff arterial pressure was measured by a Dinamap arterial pressure monitor (model 1846SX, Critikon, Tampa, FL) in the arm. Breath-by-breath tidal volume was measured continuously with a pneumotachograph (model 3, Fleisch; differential pressure transducer MP45-1, Validyne, Northridge, CA) connected to an airtight face mask by a two-way respiratory valve (model 7924, Hans Rudolph, Kansas City, MO). End-tidal CO2 concentrations were detected by an infrared gas analyzer (Engström Eliza CO2 Analyser, Gambro Engström AB, Bromma, Sweden) that sampled expiratory air from a small port in the face mask. Transcutaneous CO2 and O2 tensions were estimated by a noninvasive monitor (model 840, Novametrix, Wallingford, CT). Percent O2 saturation was detected in some subjects by a finger oximeter (Biox 3700, Ohmeda). Data were recorded continuously by digital audiotape (TEAC model RD-145T, Tokyo, Japan) and electrostatic strip-chart (Gould ES1000, Greenbelt, MD) recorders.

Experimental Protocols

The study was divided into two separate experiments conducted on different days. The order of experiments and the order of interventions within experiments were randomized. On one day, sustained decreases and increases of arterial baroreceptor inputs were provoked pharmacologically during controlled breathing at three levels of end-tidal CO2 (nominally 3, 4, and 5%). On another day, brief carotid baroreceptor stimuli were delivered at the same three levels of end-tidal CO2.

Steady-State Measurements

Breathing. Subjects were taught to control their respiratory rate and depth by breathing in synchrony with an audio cue that signaled the beginning of inspiration and expiration. Subjects maintained a constant inspiratory duration (2 s), inspiratory:expiratory duration ratio (1:1), respiratory frequency (15 breaths/min or 0.25 Hz), and tidal volume. Before any measurements were made, a unique tidal volume was determined for each subject to allow him or her to blow off CO2 and maintain an end-tidal CO2 level of 3%. Subjects maintained this target tidal volume throughout all trials by watching an oscilloscope that displayed their breath-by-breath inspired tidal volume excursions. Gas comprising 10% CO2 in 90% O2 was bled into the breathing line as necessary to maintain end-tidal CO2 levels at about 3, 4, or 5% .

Arterial pressure changes. To monitor how chemoreceptors affect baroreceptor control of R-R interval, we infused sodium nitroprusside (a direct vasodilator), saline (placebo), or phenylephrine hydrochloride (an alpha 1-adrenergic vasoconstrictor) intravenously at a rate of 1 µg · kg-1 · min-1 to vary arterial pressure and levels of arterial baroreceptor stimulation while subjects maintained expired end-tidal CO2 concentrations of 3, 4, or 5%. Subjects breathed at a constant tidal volume and frequency for 5 min at each end-tidal CO2 level. Readings of cuff arterial pressure and transcutaneous CO2 and O2 tensions were taken 3 min into each 5-min period; the electrocardiogram, arterial pressure, end-tidal CO2, and tidal volume were measured continuously. After the subject had breathed for 5 min at each CO2 level during the first infusion, the same procedure was repeated with the remaining two infusions. Both the order of intravenous infusions and the end-tidal CO2 concentrations during each infusion were randomized, and subjects were blinded to the intervention of the moment.

Carotid baroreflex testing. Carotid baroreceptor-cardiac reflex responses were elicited at 3, 4, and 5% end-tidal CO2 concentrations by application of graded, stepwise pressure and suction sequences to a neck chamber strapped to the anterior of the subject's neck, as described previously (10). The subject was equipped with a mouthpiece instead of a face mask, so that the neck collar would fit properly without interfering with the breathing apparatus. Subjects breathed deeply until their end-tidal CO2 concentrations reached the target level, at which time they were asked to stop breathing. As the subject held a normal end-expiratory volume, external neck pressure was then increased to ~40 mmHg for ~4 s and reduced by R-wave-triggered stepwise 15-mmHg decrements to approximately -65 mmHg. This pressure sequence was repeated seven times at each CO2 level and averaged to define reproducible (9) sigmoid R-R interval-carotid distending pressure response relationships. Carotid distending pressure was calculated as systolic minus applied neck pressure.

Carotid baroreceptor-cardiac reflex response relationships provoked with the neck chamber yielded discrete parameters that were analyzed statistically: baseline R-R interval, R-R interval range, maximum slope, and operational point (10). Baseline R-R intervals were taken as the average of the last two R-R intervals preceding the 40-mmHg pressure step. [Because R-R intervals are constant during brief held expiration (12), the last two intervals before neck pressure changes are representative of R-R intervals during the entire period of measurement.] R-R interval range was taken as the difference between the maximum and minimum R-R intervals on the sigmoid relation. Maximum slope was calculated from serial linear regression analyses of each three-pair sequence of carotid distending pressures and R-R intervals. Operational point was the baseline R-R interval minus the minimum R-R interval, expressed as a percentage of the R-R interval range.

Two male and two female subjects visited the laboratory on a third occasion to determine O2 saturation during the steady-state protocol. These subjects were equipped with the finger oximeter, and they breathed to the same tidal volumes that they used during their earlier studies. First, subjects breathed at 3% end-tidal CO2 for 5 min with room air. Transcutaneous CO2 and O2 partial pressures, finger O2 saturation, and end-tidal CO2 concentrations were recorded during each minute of the 5-min breathing period. Next, the same procedure was repeated, but with subjects maintaining 5% end-tidal CO2 levels with 10% CO2 in 90% O2 added to the breathing line. (This was the same gas mixture used in the previous studies.)

Although transcutaneous PCO2 (PtcCO2) and PO2 are not identical to arterial PCO2 and PO2, they correlate closely (30). We improved the correlation (34) between PtcCO2 and arterial PCO2 by applying the temperature correction
P<SC>co</SC><SUB>2</SUB> (at 37°C) = Ptc<SUB>CO<SUB>2</SUB></SUB> × (10<SUP>0.019 × 37 − <IT>T</IT><SUB>s</SUB></SUP>) − 4 (1)
where TS is sensor temperature.

Data Analyses

R-R interval spectral analysis. Data were digitized at a rate of 250 Hz and analyzed off-line with commercial signal-processing hardware and software (CODAS, Dataq Instruments, Akron, OH). R-R interval spectral power was estimated at low (0.05-0.15 Hz) and respiratory (0.20-0.30 Hz) frequencies with methods described previously (5), on the basis of the Welch algorithm of averaging periodograms (39) implemented according to the method of Rabiner et al. (31). We analyzed ~256 s data with a commercial software package (DADiSP, DSP Development, Cambridge, MA). The time series was spline interpolated at 4 Hz to yield equidistant time intervals that were divided into seven equal overlapping segments. Each segment was detrended and fast Fourier transformed to derive its frequency representation, the periodogram. The seven periodograms were averaged to yield a spectrum estimate for the entire R-R interval time series. The frequency resolution was 0.0156 Hz. Spectral power across the low (0.05-0.15 Hz) and respiratory (0.20-0.30 Hz) frequency bands was summed separately to produce single estimates of low-frequency and respiratory-frequency power.

Statistical Analyses

Analysis of covariance. We used several statistical methods to extract the greatest amount of information possible from our results. First, we used analysis of covariance to evaluate the effects of interventions on outcome variables. We then used multiple ANOVA to determine if there were differences among outcome variables among the different groups characterized by our interventions. If this analysis identified different behaviors among different groups, multivariate regression was used to identify possible causal relationships. (Because outcome variables such as arterial pressures and R-R intervals are interrelated, simple regression analysis might not be appropriate to identify possible causal relationships.)

Statistical modeling. We constructed a simultaneous linear statistical model that employed a two-stage least-squares approach (19). (We used a linear model because nonlinear models, including exponential and power models, did not improve the correlations we obtained with the linear model.) First, we divided variables into independent and dependent categories. Independent variables, those that could not be influenced by other variables because they were controlled externally or could not change, included end-tidal CO2, nitroprusside, phenylephrine, and gender. Dependent variables, those that could be influenced by independent variables and by each other, included R-R intervals, R-R interval spectral power at respiratory and low frequencies, and diastolic pressure. In our equations, end-tidal CO2, R-R intervals, R-R interval spectral power, and diastolic pressures were treated as continuous variables, and nitroprusside, phenylephrine, and gender were treated as dummy variables and assigned values of 0 or 1 (male = 0; female = 1). Second, we evaluated all combinations of variables to identify subsets of variables that most economically accounted for our results. If both selection methods identified the same model, we accepted that model as the best-fitting model. Otherwise, we used both forward selection and backward elimination models and set the entry level at 0.1. We ranked all combinations to identify subsets with the smallest number of variables and best goodness of fit according to Akaike's Information Criteria (1), adjusted R2 (18), and Mallows' CP (23).

Finally, we performed a two-component analysis (19). The first component constructed a path diagram on the basis of results of the regression analysis, according to the equation
DBP = &bgr;<SUB>1</SUB> + &bgr;<SUB>2</SUB>(CO<SUB>2</SUB>) + &bgr;<SUB>3</SUB>(NP) + &bgr;<SUB>4</SUB>(PE) + &bgr;<SUB>5</SUB>(Gender) + &egr;<SUB>2</SUB> (2)
where DBP is diastolic blood pressure, beta 1-beta 5 are parameter coefficients, and epsilon 2 is the error representing the cumulative effect of all measured variables that could have affected measured diastolic pressure. This analysis revealed the direct effects of the independent variables on diastolic pressure.

The second component solved the equation
<IT>r</IT> = &agr;<SUB>1</SUB> + &agr;<SUB>2</SUB>(CO<SUB>2</SUB>) + &agr;<SUB>3</SUB>(NP) + &agr;<SUB>4</SUB>(PE) (3)
 + &agr;<SUB>5</SUB>(Gender) + &agr;<SUB>6</SUB>(DBP) + &egr;<SUB>1</SUB>
where r is a generic term representing R-R intervals or R-R interval spectral power at respiratory or low frequencies, alpha 1-alpha 6 are parameter coefficients, and epsilon 1 is error. This equation revealed direct effects of independent variables and diastolic pressures on R-R intervals and respiratory and low-frequency R-R interval spectral power. Because this equation could not deal with both nitroprusside and phenylephrine simultaneously, only one drug was included in each pair of equations at a time to determine the effect of that drug on r. Finally, the combination of these two equations revealed the indirect effects the independent variables exerted on r through their direct effects on diastolic pressure. (We performed analyses with systolic pressure as well as diastolic pressure and found that both documented the same types of effects in the path analysis. Because our analysis yielded slightly higher R2 values for diastolic than for systolic pressure, we report only results involving diastolic pressure.) We considered parameter coefficients to be significant when P < 0.05.

Carotid baroreflex responses. Each parameter (baseline R-R interval, R-R interval range, maximum slope, and operational point) was compared at each CO2 level (control, 3, 4, and 5%). A nonparametric sign-rank test was used to test the significance of each variable. A univariate testing procedure was used to determine if the variable was distributed normally for each parameter. If the variable was distributed normally, a Student's t-test was used to confirm the results of the sign-rank test. Differences were considered significant when P < 0.05.

    RESULTS
Top
Abstract
Introduction
Methods
Results
Discussion
References

Steady-State Measurements

Breathing and end-tidal CO2 control. Subjects maintained excellent control of tidal volume, respiratory frequency, inspiratory duration, and inspiratory-expiratory ratios throughout the entire protocol. This is illustrated in Fig. 1, which shows 15 consecutive breaths from one subject. The average tidal volume for all subjects, for all interventions, was 1.02 liter. Because all subjects maintained tidal volumes accurately, we were able to maintain average end-tidal CO2 very close to targeted levels. At low, normal, and high arterial pressures, end-tidal CO2 concentrations averaged (mean ± SE) 3.03 ± 0.03, 2.98 ± 0.04, and 3.05 ± 0.03%; 3.98 ± 0.08, 4.03 ± 0.04, and 4.03 ± 0.04%; and 5.06 ± 0.03, 5.07 ± 0.04, and 5.08 ± 0.04%. Average CO2 concentrations within each CO2 level were comparable (P = 0.25-0.90). (Subject baseline end-tidal CO2 concentrations during uncontrolled breathing ranged between 3.87 and 5.01% and averaged 4.54 ± 0.17%.)


View larger version (23K):
[in this window]
[in a new window]
  		Fig. 1.   Fifteen consecutive breaths from 1 subject. Data sweeps were triggered on onset of inspiration.

Transcutaneous CO2 and O2 measurements. As expected, calculated transcutaneous PCO2 levels were proportional to end-tidal CO2 concentrations; at low, normal, and high arterial pressures, transcutaneous PCO2 levels averaged 14.2, 20.2, and 26.8 mmHg, respectively. At low, normal, and high arterial pressures, PO2 levels averaged 74, 117, and 158 mmHg, respectively. Importantly, each of the four subjects studied with finger oximetry registered 100% O2 saturation during both 3% (room air) and 5% end-tidal CO2 (10% CO2 in 90% O2) breathing periods.

Hemodynamic measurements. Figure 2, A-C, shows recordings made during nitroprusside, saline, and phenylephrine infusions, respectively, in one subject (end-tidal CO2 = 3%). As expected (11), R-R intervals, R-R interval fluctuations, and arterial pressures were higher during phenylephrine than nitroprusside infusions. (In this and other subjects, arterial pulse pressure tended to be higher during nitroprusside than saline infusions; pulse pressures were comparable during nitroprusside and phenylephrine infusions.) Figure 3 shows mean diastolic pressures for all subjects at all levels of arterial pressure and end-tidal CO2. At each pressure level, mean diastolic pressure was significantly (P <=  0.05) lower at end-tidal CO2 levels of 3% than at 5%.


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 2.   Recordings from 1 subject at each arterial pressure level at low (3%) end-tidal CO2. Experimental records during low (A; nitroprusside), normal (B; saline), and high (C; phenylephrine) levels of arterial baroreceptor stimulation.


View larger version (37K):
[in this window]
[in a new window]
  		Fig. 3.   Mean + SE diastolic pressures from all 10 subjects at all levels of end-tidal CO2 (3, 4, and 5%) and at low (nitroprusside), normal (saline), and high (phenylephrine) arterial pressure.

Figure 4 shows mean R-R intervals for all subjects at all levels of end-tidal CO2. At each pressure level, mean R-R intervals were significantly lower at end-tidal CO2 levels of 3% than at 5%. Figure 5, A-C, shows fast Fourier transform R-R interval power spectra for one subject during infusions of nitroprusside, saline, and phenylephrine, respectively, at 3 and 4% end-tidal CO2 levels. In this subject, R-R interval power at the respiratory frequency (0.25 Hz) was proportional to arterial pressure (not shown) and end-tidal CO2 levels. Figure 6 shows mean R-R interval spectral power at the respiratory frequency (integrated between 0.20 and 0.30 Hz) for all subjects at all levels of arterial pressure and end-tidal CO2. At each pressure level, integrated R-R interval spectral power at the respiratory frequency was lower at end-tidal CO2 levels of 3% than 5%. No distinct trends or relationships were observed for R-R interval power at low frequencies (0.05-0.15 Hz) during chemoreceptor and baroreceptor stimulation.


View larger version (37K):
[in this window]
[in a new window]
  		Fig. 4.   Mean R-R intervals from all 10 subjects at all levels of end-tidal CO2 (3, 4, and 5%) and at low (nitroprusside), normal (saline), and high (phenylephrine) arterial pressure.


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 5.   R-R interval power spectra of 1 subject breathing at a constant tidal volume and respiratory rate (0.25 Hz) at low (A; nitroprusside), normal (B; saline), and high arterial pressures (C; phenylephrine) and 2 levels (3 and 4%) of end-tidal CO2.


View larger version (31K):
[in this window]
[in a new window]
  		Fig. 6.   Mean R-R interval power spectra integrated over 0.20-0.30 Hz for all subjects at 3 levels of end-tidal CO2 (3, 4, and 5%) and at low (nitroprusside), normal (saline), and high (phenylephrine) arterial pressures.

Statistical Modeling of Steady-State Results

Analysis of covariance. We performed analysis of covariance on all 19 combinations of independent and dependent variables. Although all combinations showed some correlation with R-R intervals, only six variables were significant in the full model regression. The univariate model that best explained R-R interval responses was
<AR><R><C>RR = 0.840 +</C></R><R><C>(0.0001)</C></R></AR> <AR><R><C>0.036 (CO<SUB>2</SUB>)</C></R><R><C>(0.054)</C></R></AR><AR><R><C> − 0.99 (PE)  </C></R><R><C>(0.0053)</C></R></AR> (4)
<AR><R><C>+ 1.388 (Gender,PE)</C></R><R><C>(0.0021)</C></R></AR> <AR><R><C>+ 0.013 (DBP,PE) </C></R><R><C>(0.0005)</C></R></AR>
<AR><R><C>− 0.014 (DBP,Gender,PE)</C></R><R><C>(0.0033)</C></R></AR><AR><R><C> − 0.0019 (DBP,NP)</C></R><R><C>(0.0004)</C></R></AR>
where the first number (0.840) is the R-R interval intercept; CO2 is end-tidal CO2 level; PE is phenylephrine infusion; PE,Gender is phenylephrine infusion in female subjects; DBP,PE is diastolic pressure and phenylephrine infusion; and DBP,Gender,PE is diastolic pressure during phenylephrine infusions in female subjects. The numbers before each variable indicate the parameter estimate, and the numbers in parentheses below each variable indicate the probability. This model ranked first according to Akaike's Information Criteria and CP and ranked fourth according to adjusted R2. Its R2 was 0.64.

The univariate model that best explained diastolic pressure was
<AR><R><C>DBP = 69.88</C></R><R><C>(0.0001)</C></R></AR> <AR><R><C>+ 3.53 (CO<SUB>2</SUB>)</C></R><R><C>(0.0043)</C></R></AR> <AR><R><C>− 2.01 (CO<SUB>2</SUB>,NP)</C></R><R><C>(0.0001)</C></R></AR> (5)
 <AR><R><C>+ 14.62 (PE)</C></R><R><C>(0.0001)</C></R></AR> <AR><R><C>− 8.86(Gender)</C></R><R><C>(0.0001)</C></R></AR>
where the first number (69.88) is diastolic pressure intercept; CO2,NP is end-tidal CO2 level and nitroprusside infusion; and Gender is female. This model ranked first according to Akaike's Information Criteria, CP, and adjusted R2. Its R2 was 0.57. The analysis of covariance pointed toward significant interactions between the level of CO2 chemoreceptor stimulation and diastolic pressure, as shown in Figs. 3 and 4.

Path analysis. Table 1 lists the results of two-stage simultaneous least-squares statistical modeling. In Table 1, significant parameter coefficients (beta ) identify direct effects between the compared variables; the sign of the parameter coefficient indicates whether the direct effect is parallel (positive) or opposite (negative); and the numerical value indicates the strength of the influence. For example, the parameter coefficients of 2.676 and 14.716 for CO2 and phenylephrine indicate that both interventions significantly raise diastolic pressure, and the numbers indicate that phenylephrine increases diastolic pressure much more than CO2. Although Table 1 suggests that CO2, nitroprusside, and phenylephrine do not exert direct effects on R-R intervals and R-R interval spectral power at the respiratory frequency, it does suggest that all three variables exert indirect effects through their direct effects on diastolic pressure. The path analysis suggested that male and female subjects had similar reactions to the interventions; therefore, subject gender did not influence the results.

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Results of two-stage least-squares modeling

Arterial pressure. Influences due exclusively to chemoreceptor activity were identified by responses during saline infusion (i.e., at usual levels of baroreceptor stimulation). Under this condition, low CO2 levels decreased diastolic pressure. (Low CO2 refers to 3% end-tidal CO2 and is distinguished from 4 and 5% end-tidal CO2, which are referred to as normal CO2. As mentioned, average baseline end-tidal CO2 levels was 4.54%, or about midway between the 4 and 5% levels induced experimentally.) These results are depicted schematically in Fig. 7A. The path analysis indicated that the level of CO2 chemoreceptor stimulation exerts a direct, parallel effect on diastolic pressure; that is, increases of end-tidal CO2 concentrations from low to normal levels increase diastolic pressure.


View larger version (21K):
[in this window]
[in a new window]
  		Fig. 7.   Results of statistical modeling during usual (A; saline), low [B; nitroprusside (NP)], and high [C; phenylephrine (PE)] baroreceptor stimulation (see METHODS). Vertical arrows indicate direction of effect (up-arrow , positive; down-arrow , negative) on diastolic blood pressure (DBP); thickness of lines and darkness of fonts indicate relative magnitude of each influence.

The effects of adding baroreceptor to chemoreceptor input in the path analysis were determined by including measurements made at all arterial pressures. These results are depicted schematically in Fig. 7, B and C. The path analysis investigated the individual effects of baroreceptor input on diastolic pressure before it assessed possible chemoreflex and baroreflex interactions. As expected, nitroprusside decreased and phenylephrine increased diastolic pressures at all CO2 levels, from those measured during saline infusion (Fig. 3). The path analysis confirmed these expected results: nitroprusside exerted a direct, opposite effect on diastolic pressure (Fig. 7B), and phenylephrine exerted a direct, parallel effect on diastolic pressure (Fig. 7C).

Chemoreceptor influences during both nitroprusside and phenylephrine infusions were similar to those during the saline infusion: low levels of CO2 chemoreceptor stimulation decreased, and high levels increased diastolic pressures (Fig. 3). However, the path analysis indicated that when chemoreceptor inputs were varied at low diastolic pressures (Fig. 3, left, and Fig. 7B), the direct, opposite effect of nitroprusside diminished the direct, parallel effect of CO2 chemoreceptor stimulation compared with chemoreceptor influences at usual levels of baroreceptor stimulation. On the other hand, when chemoreceptor inputs were varied at high diastolic pressures (Fig. 3, right, and Fig. 7C), the direct, parallel effect of phenylephrine increased the direct, parallel effect of CO2 chemoreceptor stimulation and resulted in an augmented net effect of CO2 compared with chemoreceptor influences at usual levels of baroreceptor stimulation. This is indicated by the heavy arrow in Fig. 7C.

Arterial pressure spectral power. Changes of the levels of CO2 chemoreceptor and arterial baroreceptor stimulation did not systematically alter arterial pressure spectral power.

R-R intervals. At usual levels of arterial baroreceptor stimulation (during saline infusion), low CO2 concentrations decreased R-R intervals (Fig. 4). However, the path analysis (Table 1) suggested that CO2 did not exert this effect on R-R intervals directly. Rather, this analysis revealed that CO2 chemoreceptor stimulation exerted its effect on R-R intervals indirectly, through its parallel effect on diastolic pressure. These results are depicted schematically in Fig. 8.


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 8.   Path analysis revealed that CO2 chemoreceptor stimulation exerted its effect on R-R intervals (RR) indirectly, through its parallel effect on DBP. Vertical arrows indicate direction of effect, and thickness of lines and darkness of fonts indicate relative magnitude of each influence. +, Parallel effect; -, opposite effect; Ø, negligible effect. See RESULTS for details.

Because the path analysis suggested that CO2 chemoreceptors modulate R-R intervals indirectly through their influence on diastolic pressure, any alteration in the effects of CO2 on diastolic pressure should translate into alterations in the effects of CO2 on R-R intervals. We found that baroreceptor stimulation augmented the effects of CO2 chemoreceptor stimulation on R-R intervals (Fig. 4). At low end-tidal CO2 concentrations (3%), the direct, opposite effect of nitroprusside on diastolic pressure was minimized by the direct, parallel effect of CO2 on diastolic pressure. The path analysis indicated that this reduction of CO2 chemoreceptor influence on diastolic pressure greatly diminished the indirect influence of CO2 chemoreceptor stimulation on R-R intervals (Fig. 8B). At normal end-tidal CO2 concentrations (4, 5%), however, the direct, parallel effect of CO2 on diastolic pressure tended to outweigh the direct, opposite effect of nitroprusside on diastolic pressure. This, in turn, augmented the parallel effect of diastolic pressure on R-R intervals and produced a small but noticeable indirect, parallel influence on R-R interval (Fig. 8C). During phenylephrine infusion, when the influence of CO2 chemoreceptor stimulation on diastolic pressure was large, the influence of CO2 on R-R intervals also was large (Fig. 8D).

R-R interval spectral power. Changes of R-R interval spectral power at the respiratory frequency paralleled those of R-R intervals, and the statistical analysis (Table 1) yielded a path diagram that was identical to that for R-R intervals (Fig. 8).

Carotid baroreflex responses. Figure 9 shows a recording made during one neck pressure/suction sequence (Fig. 9A) and average (of 7 sequences) carotid baroreceptor-cardiac reflex responses for the same subject, at end-tidal CO2 concentrations of 3 and 4% (Fig. 9B). In this subject during hypocapnia, the sigmoid baroreflex relation was shifted to operate at lower R-R intervals, but over the same carotid distending pressure range. The shift of the baroreceptor-cardiac reflex relationship signifies simple resetting because the baroreflex slope, range, and operational point were unchanged. Figure 10 shows and Table 2 lists average carotid baroreflex responses for all subjects. Because average stimulus-response relationships at 4 and 5% end-tidal CO2 were almost superimposable, only responses at 4% end-tidal CO2 concentrations are depicted. In general, average relationships displayed similar shapes at each end-tidal CO2 level. However, the response relationship for 3% end-tidal CO2 was shifted downward to operate over shorter R-R intervals. The position of stimulus-response relationships on the pressure axis did not appear to be affected by end-tidal CO2 concentration. Average range, slope, and operational point were comparable at each level of end-tidal CO2 (Table 2).


View larger version (32K):
[in this window]
[in a new window]
  		Fig. 9.   Recording from 1 subject during a single neck pressure/suction sequence (A) and average responses of that same subject to 7 such pressure sequences at each of end-tidal (ET) CO2 levels of 3 and 4% (B). Larger filled and open circles on average relationship curves indicate operational points (see METHODS).


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 10.   Average vagally mediated carotid baroreceptor-cardiac reflex responses for all 10 subjects at 2 end-tidal CO2 levels. Larger filled and open circles on average relationship curves indicate operational points.

                              
View this table:
[in this window]
[in a new window]
  		Table 2.   Baroreflex characteristics

    DISCUSSION
Top
Abstract
Introduction
Methods
Results
Discussion
References

We studied chemoreflex-baroreflex interactions in healthy young men and women who voluntarily breathed at a constant respiratory rate and a large tidal volume, while we maintained end-tidal CO2 levels at 3, 4, and 5%, and arterial pressures at low, usual, and high levels. We evaluated R-R interval and arterial pressure changes with both time- and frequency-domain methods and analyzed our global findings with a two-stage least-squares linear statistical model. Our study yielded three principal conclusions. First, although low levels of CO2 chemoreceptor stimulation reduce R-R intervals and R-R interval variability, this effect appears to be indirect, rather than direct, and appears to be mediated by reductions of arterial pressure. Second, the influence of CO2 chemoreceptor stimulation on arterial pressure (and, derivatively, on R-R intervals and R-R interval variability) depends critically on baseline arterial pressure levels: chemoreceptor effects are smaller when arterial pressure is low and larger when arterial pressure is high. Third, the effects of CO2 chemoreceptor stimulation on R-R intervals and R-R interval variability are not mediated by changes of arterial baroreflex function. Although reduced CO2 chemoreceptor stimulation shifts, or resets, carotid baroreceptor-cardiac reflex relationships to operate at lower R-R intervals, it does not alter the gain, range, or operational point of these relationships.

Chemoreceptors are responsive to changes of O2, CO2, and pH (16) and are located both peripherally, in the carotid and aortic bodies and abdomen (26), and centrally, in the ventrolateral medulla and elsewhere (27). Because our measurements were made during hyperoxia, it is likely that the hemodynamic changes we observed were mediated primarily by central CO2 chemoreceptors (13). Our study of chemoreflex and baroreflex interactions differs from earlier studies in several respects. We studied conscious, cooperative subjects, and minimized (8) [but possibly did not abolish (3)] contributions from O2 chemoreceptors, by obtaining our measurements with 100% O2 saturation. We studied subjects at constant respiratory rates and tidal volumes. Finally, we studied CO2 chemoreceptor reflex responses over a range of chemoreceptor stimulation from subnormal to high normal levels.

Chemoreflex and Baroreflex Interactions

Many studies show that arterial baroreceptor inputs modulate responses to chemoreceptor stimulation. Heistad et al. (14) reported that in anesthetized dogs, hypotension enhances, and hypertension inhibits, vasoconstrictor responses to chemoreceptor stimulation with nicotine or hypoxic hypercapnic (asphyxic) blood. Similarly, Wennergren et al. (40) found that in anesthetized cats, chemoreceptor vasoconstriction in skeletal muscle is more powerful at low or medium carotid sinus pressures than at high carotid sinus pressures. Although Mancia (24) found that in anesthetized dogs, peripheral responses to chemoreceptor stimulation by hypoxic hypercapnic blood are reduced at high levels of baroreceptor stimulation, they also found that chemoreflex responses are reduced at low levels of baroreceptor stimulation.

Other studies show that chemoreceptor inputs modulate responses to arterial baroreceptor stimulation. Brunner et al. (6) reported that heart rate responses to baroreflex stimulation are augmented by hypercapnia. Marshall (25) explored this physiology in cats and used the anesthetic althesin to preserve the alerting stage of the defense reaction (15) and, as expected, found that baroreceptor stimulation (elevation of pressure in a blind carotid pouch) led to arterial pressure and heart rate reductions. Chemoreceptor stimulation (lingual artery injections of a phosphate mixture or CO2-equilibrated saline) led to arterial pressure and heart rate increases. Simultaneous chemoreceptor and baroreceptor stimulation led to arterial pressure and heart rate increases; that is, when chemoreceptors and baroreceptors were stimulated simultaneously, baroreflex responses were overridden.

Although we cannot say how our subjects would have responded to frank hypercapnia, the reduction of arterial pressure that we documented during hypocapnia (Fig. 3) is what would have been predicted from the results of Marshall (25). However, our finding that hypocapnia reduces R-R intervals would not have been predicted from the study of Marshall (which showed that hypercapnia reduces R-R intervals). If the tachycardia Marshall documented reflects the alerting stage of the defense reaction, our findings suggest that defense reactions are not operative at usual or low levels of CO2 chemoreceptor stimulation. We evaluated hypocapnia rather than hypercapnia, because during hypercapnia, even highly motivated human subjects cannot override the stimulus to breathe and maintain constant ventilation (38).

Some human studies are similar, but not identical, to ours. Bristow and co-workers (4) found, as did Marshall (25), that hypercapnia provokes tachycardia. They also found that hypercapnia reduces (but does not abolish) baroreflex-mediated R-R interval prolongation. Our findings do not necessarily contradict those of Bristow et al. (4); they merely suggest that modulation of baroreflexes by CO2 chemoreceptors does not extend to subnormal, hypocapnic end-tidal CO2 levels. Our study may be closer to that of Al-Ani and co-workers (2), who calculated R-R interval spectral power in healthy women during hypercapnia and compensatory hyperventilation and during hypocapnia and voluntary hyperventilation. Al-Ani et al. reported, as we do, that hypocapnia shortens R-R intervals and tends to reduce respiratory frequency R-R interval spectral power. In our study, the reduction of respiratory frequency R-R interval spectral power was highly significant (Fig. 6).

Physiological Implications

It is highly likely that the R-R interval shortening we documented during hypocapnia reflects withdrawal of vagal restraint. R-R interval shortening during hyperventilation is not affected by beta -adrenergic blockade (33) and is largely prevented by large-dose atropine (37). Moreover, reductions of R-R interval spectral power at the respiratory frequency (Figs. 5 and 6) probably reflect reductions of vagal-cardiac neuronal activity (11). If this inference is correct, our results support the notion that usual levels of CO2 chemoreceptor stimulation contribute to usual levels of human vagal-cardiac nerve activity. Kollai et al. (21), on the basis of a critical assumption regarding the threshold pressure for human baroreceptor-cardiac reflex responses, postulated the existence of a baroreflex-independent component of resting vagal-cardiac nerve traffic. Our findings at once support and challenge the speculation of Kollai et al. Although we document a CO2 chemoreflex contribution to resting vagal-cardiac nerve traffic (Figs. 2 and 4-6), our statistical model (Table 1) suggests that this contribution is indirect and is mediated by arterial baroreflex responses to chemoreceptor-mediated increases of arterial pressure. Our data suggest, further, that the influence of CO2 chemoreceptor stimulation on vagal activity is small, relative to that of arterial baroreceptors (Table 1, Figs. 4 and 6).

Our results do not indicate how chemoreceptor input increases arterial pressure. It is established that, in humans, supranormal levels of CO2 chemoreceptor stimulation increase muscle sympathetic nerve activity (35, 38), and, in animals, preganglionic sympathetic nerve firing is proportional to levels of CO2 chemoreceptor stimulation below, as well as above usual levels. If proportionality between sympathetic nerve activity and CO2 chemoreceptor stimulation extends below, as well as above, usual end-tidal CO2 levels, then changes of sympathetic nerve activity are sufficient to explain our findings. However, Burnum et al. (7) showed that forearm vasodilation during hypocapnia persists after brachial blockade of efferent nerve traffic. Therefore, changes of efferent sympathetic nerve activity are not necessary to explain the proportionality between CO2 chemoreceptor stimulation and arterial pressure that we documented.

Limitations

Our study has several potential limitations. First, we evaluated our results with a statistical model, which we hoped would provide convincing arguments for causality on the basis of statistical probability. Our reliance on the model (19) may be at once a weakness and a strength of our study. The weakness of the statistical model is that the conclusions are grounded purely in mathematics and not necessarily in physiology. Conversely, the strength of the model is that it is free of author biases and preconceptions. The R2 values we obtained, 0.64 and 0.57, indicate that a high proportion of the hemodynamic changes we measured can be explained by the statistical model. Moreover, in our discussion, we used the results of the model to provide a structure for our physiological interpretations of the results.

Second, we did not structure our analysis in a way that would discover possible influences of R-R interval changes on arterial pressure changes. [An earlier study showed that respiratory-frequency R-R interval fluctuations contribute to respiratory-frequency arterial pressure changes (36).] Although our results do not exclude influences of R-R intervals on arterial pressure, they make it unlikely; neither the level of CO2 chemoreceptor stimulation nor the level of arterial baroreceptor stimulation altered arterial pressure fluctuations significantly.

Third, we used an indirect measure to estimate arterial pressure. Although the photoplethysmograph we used has been validated carefully against intra-arterial pressure under baseline conditions (17, 29), its measurements may not be as reliable during phenylephrine infusions (17). However, our conclusions are based primarily on diastolic pressures, which are accurately registered by photoplethysmography, even during phenylephrine infusions (29).

Fourth, we were unable to quantitate end-tidal CO2 levels during carotid baroreceptor stimulation. (Neck pressure changes were applied during held expiration.) End-tidal CO2 levels cannot be measured in absence of respiration, and the transcutaneous measurements we made change were too sluggish to track changes occurring during 15-s periods. The consequence of this unavoidable problem was that arterial PCO2 levels probably were higher at the time neck pressure changes were applied than they were when subjects were breathing with end-tidal CO2 levels maintained at 3%. However, the major downward shift of baroreceptor-cardiac reflex relationships on the R-R interval axis after hyperventilation (Figs. 9 and 10) provides strong evidence that subjects were hypocapnic at the time baroreflex stimuli were applied. Moreover, despite the increases of CO2 that must have occurred during held expiration, some proportionality should have been preserved; that is, baroreflex responses were elicited at three graded levels of PCO2.

Fifth, voluntary control of breathing may have influenced autonomic neural outflow. However, our subjects breathed at the same rate and to the same tidal volume throughout all infusions; therefore, it is unlikely that voluntary control of breathing influenced results during one measurement period and not another. Finally, hypocapnia may produce cerebral artery constriction (20) [or dilation at low arterial pressures (32)]. Although we assume that cerebral metabolism was normal during the rather modest perturbations we used, we did not measure cerebral blood flow.

In conclusion, we studied arterial baroreceptor-central CO2 chemoreceptor interactions in healthy young adults. Our study supports three main conclusions. First, the level of CO2 chemoreceptor stimulation in part determines the level of arterial pressure; however, this influence is proportional to arterial pressure, such that CO2 chemoreceptors exert a small effect when arterial pressure is low and a large effect when arterial pressure is high. Second, although low levels of CO2 chemoreceptor stimulation reduce R-R intervals and R-R interval variability, these effects appear to be indirect, rather than direct, and to be mediated by reductions of arterial pressure. Third, vagally mediated arterial baroreceptor-cardiac reflex relationships are not altered by reduced levels of CO2 chemoreceptor stimulation; rather, they are reset to operate normally, but at lower baseline R-R intervals and arterial pressures.

    ACKNOWLEDGEMENTS

We thank our human volunteers who made this research possible.

    FOOTNOTES

This study was supported by grants and contracts from the Department of Veterans Affairs; National Heart, Lung, and Blood Institute Grants HL-30506, HL-22296, and HL-07556; and National Aeronautics and Space Administration Grants NAG-2-408 and NAS-17720.

Address for reprint requests: D. L. Eckberg, Cardiovascular Physiology, Hunter Holmes McGuire Dept. of Veterans Affairs Medical Center, 1201 Broad Rock Blvd., Richmond, VA 23249.

Received 24 February 1997; accepted in final form 2 March 1998.

    REFERENCES
Top
Abstract
Introduction
Methods
Results
Discussion
References

1.   Akaike, H. Fitting autoregressive models for prediction. Ann. Inst. Stat. Math. 21: 243-247, 1969.

2.   Al-Ani, M., A. S. Forkins, J. N. Townend, and J. H. Coote. Respiratory sinus arrhythmia and central respiratory drive in humans. Clin. Sci. (Colch.) 90: 235-241, 1996[Medline].

3.   Biscoe, T. J., M. J. Purves, and S. R. Sampson. The frequency of nerve impulses in single carotid body chemoreceptor afferent fibres recorded in vivo with intact circulation. J. Physiol. (Lond.) 208: 121-131, 1970[Abstract/Free Full Text].

4.   Bristow, J. D., E. B. Brown, Jr., D. J. C. Cunningham, M. G. Howson, M. J. R. Lee, T. G. Pickering, and P. Sleight. The effects of raising alveolar PCO2 and ventilation separately and together on the sensitivity and setting of the baroreceptor cardiodepressor reflex in man. J. Physiol. (Lond.) 243: 401-425, 1974[Abstract/Free Full Text].

5.   Brown, T. E., L. A. Beightol, J. Koh, and D. L. Eckberg. Important influence of respiration on human R-R interval power spectra is largely ignored. J. Appl. Physiol. 75: 2310-2317, 1993[Abstract/Free Full Text].

6.   Brunner, M. J., A. Wallace, and C. L. MacAnespie. Interaction of carotid chemoreceptor and baroreceptor reflexes in anesthetized dogs. Am. J. Physiol. 254 (Regulatory Integrative Comp. Physiol. 23): R1-R10, 1988[Abstract/Free Full Text].

7.   Burnum, J. F., J. B. Hickam, and H. D. McIntosh. The effect of hypocapnia on arterial blood pressure. Circulation 9: 89-95, 1954[Medline].

8.   Dejours, P. Chemoreflexes in breathing. Physiol. Rev. 42: 335-358, 1962[Free Full Text].

9.   Eckberg, D. L., V. A. Convertino, J. M. Fritsch, and D. F. Doerr. Reproducibility of human vagal carotid baroreceptor-cardiac reflex responses. Am. J. Physiol. 263 (Regulatory Integrative Comp. Physiol. 32): R215-R220, 1992[Abstract/Free Full Text].

10.   Eckberg, D. L., and J. M. Fritsch. How should human baroreflexes be tested? News Physiol. Sci. 8: 7-12, 1993.[Abstract/Free Full Text]

11.   Eckberg, D. L., R. F. Rea, O. K. Andersson, T. Hedner, J. Pernow, J. M. Lundberg, and B. G. Wallin. Baroreflex modulation of sympathetic activity and sympathetic neurotransmitters in humans. Acta Physiol. Scand. 133: 221-231, 1988[Medline].

12.   Fritsch, J. M., M. L. Smith, D. T. F. Simmons, and D. L. Eckberg. Differential baroreflex modulation of human vagal and sympathetic activity. Am. J. Physiol. 260 (Regulatory Integrative Comp. Physiol. 29): R635-R641, 1991[Abstract/Free Full Text].

13.   Gesell, R., J. Lapides, and M. Levin. The interaction of central and peripheral chemical control of breathing. Am. J. Physiol. 130: 155-170, 1940.

14.   Heistad, D. D., F. M. Abboud, A. L. Mark, and P. G. Schmid. Interaction of baroreceptor and chemoreceptor reflexes. Modulation of the chemoreceptor reflex by changes in baroreceptor activity. J. Clin. Invest. 53: 1226-1236, 1974.

15.   Hilton, S. M., and J. M. Marshall. The pattern of cardiovascular response to carotid chemoreceptor stimulation in the cat. J. Physiol. (Lond.) 326: 495-513, 1982[Abstract/Free Full Text].

16.   Hornbein, T. F., and A. Roos. Specificity of H ion concentration as a carotid chemoreceptor stimulus. J. Appl. Physiol. 18: 580-584, 1963[Abstract/Free Full Text].

17.   Imholz, B. P. M., G. A. van Montfrans, J. J. Settels, G. M. A. van der Hoeven, J. M. Karemaker, and W. Wieling. Continuous non-invasive blood pressure monitoring: reliability of Finapres device during the Valsalva manoeuvre. Cardiovasc. Res. 22: 390-397, 1988[Medline].

18.   Jobson, J. D. Applied Multivariate Data Analysis. New York: Springer-Verlag, 1991, p. 254-274.

19.   Johnson, R. A., and D. W. Wichern. Applied Multivariate Statistical Analysis. Englewood Cliffs, NJ: Prentice Hall, 1992, p. 340-347.

20.   Kety, S. S., and C. F. Schmidt. The effects of altered arterial tensions of carbon dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. J. Clin. Invest. 27: 484-492, 1948.

21.   Kollai, M., G. Jokkel, I. Bonyhay, J. Tomcsanyi, and A. Naszlady. Relation between baroreflex sensitivity and cardiac vagal tone in humans. Am. J. Physiol. 266 (Heart Circ. Physiol. 35): H21-H27, 1994[Abstract/Free Full Text].

22.   Lipski, J., R. M. McAllen, and A. Trzebski. Carotid baroreceptor and chemoreceptor inputs onto single medullary neurones. Brain Res. 107: 132-136, 1976[Medline].

23.   Mallows, C. L. Some comments on CP. Technometrics 15: 661-675, 1973.

24.   Mancia, G. Influence of carotid baroreceptors on vascular responses to carotid chemoreceptor stimulation in the dog. Circ. Res. 36: 270-276, 1975[Abstract/Free Full Text].

25.   Marshall, J. M. Interaction between the responses to stimulation of peripheral chemoreceptors and baroreceptors: the importance of chemoreceptor activation of the defence areas. J. Auton. Nerv. Syst. 3: 389-400, 1981[Medline].

26.   Marshall, J. M. Peripheral chemoreceptors and cardiovascular regulation. Physiol. Rev. 74: 543-594, 1994[Free Full Text].

27.   Mitchell, R. A., H. H. Loeschcke, J. W. Severinghaus, B. W. Richardson, and W. H. Massion. Regions of respiratory chemosensitivity on the surface of the medulla. Ann. NY Acad. Sci. 109: 661-681, 1963.

28.   Miura, M., and D. J. Reis. The role of the solitary and paramedian reticular nuclei in mediating cardiovascular reflex responses from carotid baro- and chemoreceptors. J. Physiol. (Lond.) 223: 525-548, 1972[Abstract/Free Full Text].

29.   Omboni, S., G. Parati, A. Frattola, E. Mutti, M. Di Rienzo, P. Castiglioni, and G. Mancia. Spectral and sequence analysis of finger blood pressure variability. Comparison with analysis of intra-arterial recordings. Hypertension 22: 26-33, 1993[Abstract/Free Full Text].

30.   Peabody, J. L., M. M. Willis, G. A. Gregory, W. H. Tooley, and J. F. Lucey. Clinical limitations and advantages of transcutaneous oxygen electrodes. Acta Anaesthesiol. Scand. Suppl. 68: 76-82, 1978[Medline].

31.   Rabiner, L. R., R. W. Schafer, and D. Dlugos. Periodogram method for power spectrum estimation. In: Programs for Digital Signal Processing, edited by Digital Signal Processing Committee. New York: IEEE Press, 1979, p. 2.1-1-2.1-10.

32.   Radü, E. W., and G. H. du Boulay. Paradoxical dilatation of the large cerebral arteries in hypocapnia in man. Stroke 7: 569-572, 1976[Abstract/Free Full Text].

33.   Richardson, D. W., H. A. Kontos, A. J. Raper, and J. L. Patterson, Jr. Systemic circulatory responses to hypocapnia in man. Am. J. Physiol. 223: 1308-1312, 1972.

34.   Severinghaus, J. W. Transcutaneous blood gas analysis. Resp. Care 27: 152-159, 1982.

35.   Somers, V. K., A. L. Mark, D. C. Zavala, and F. M. Abboud. Contrasting effects of hypoxia and hypercapnia on ventilation and sympathetic activity in humans. J. Appl. Physiol. 67: 2101-2106, 1989[Abstract/Free Full Text].

36.   Taylor, J. A., and D. L. Eckberg. Fundamental relations between short-term RR interval and arterial pressure oscillations in humans. Circulation 93: 1527-1532, 1996[Abstract/Free Full Text].

37.   Thompson, H. K., Jr., J. N. Berry, and H. D. McIntosh. Circulatory responses to hyperventilation and exercise in normal subjects. Am. Heart J. 63: 106-114, 1962.

38.   Trzebski, A., M. L. Smith, L. A. Beightol, J. M. Fritsch-Yelle, R. F. Rea, and D. L. Eckberg. Modulation of human sympathetic periodicity by mild, brief hypoxia and hypercapnia. J. Physiol. Pharmacol. 46: 17-35, 1995[Medline].

39.   Welch, P. D. The use of fast Fourier transform for the estimation of power spectra: a method based on time averaging over short, modified periodograms. IEEE Transact. Audio Electroaccoust. AU-15: 70-73, 1967.

40.   Wennergren, G., R. Little, and B. Öberg. Studies on the central integration of excitatory chemoreceptor influences and inhibitory baroreceptor and cardiac receptor influences. Acta Physiol. Scand. 96: 1-18, 1976[Medline].

Am J Physiol Heart Circ Physiol 274(6):H2177-H2187
0002-9513/98 $5.00 Copyright © 1998 the American Physiological Society


This article has been cited by other articles:


Home page
 J. Appl. Physiol.Home page
R. D. Thijs, J. G. van den Aardweg, R. H. A. M. Reijntjes, J. G. van Dijk, and J. J. van Lieshout
Contrasting effects of isocapnic and hypocapnic hyperventilation on orthostatic circulatory control
J Appl Physiol, October 1, 2008; 105(4): 1069 - 1075.
[Abstract] [Full Text] [PDF]

Home page
 J. Physiol.Home page
G. H. Simmons, J. M. Manson, and J. R. Halliwill
Mild central chemoreflex activation does not alter arterial baroreflex function in healthy humans
J. Physiol., September 15, 2007; 583(3): 1155 - 1163.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
Y. C. Tzeng, P. D. Larsen, and D. C. Galletly
Effects of hypercapnia and hypoxemia on respiratory sinus arrhythmia in conscious humans during spontaneous respiration
Am J Physiol Heart Circ Physiol, May 1, 2007; 292(5): H2397 - H2407.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
J. M. Stewart, M. S. Medow, N. S. Cherniack, and B. H. Natelson
Postural hypocapnic hyperventilation is associated with enhanced peripheral vasoconstriction in postural tachycardia syndrome with normal supine blood flow
Am J Physiol Heart Circ Physiol, August 1, 2006; 291(2): H904 - H913.
[Abstract] [Full Text] [PDF]

Home page
 J. Physiol.Home page
D. L Eckberg and T. A Kuusela
Human vagal baroreflex sensitivity fluctuates widely and rhythmically at very low frequencies
J. Physiol., September 15, 2005; 567(3): 1011 - 1019.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
J. J. Van Lieshout, W. Wieling, J. M. Karemaker, and N. H. Secher
Syncope, cerebral perfusion, and oxygenation
J Appl Physiol, March 1, 2003; 94(3): 833 - 848.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
J. K. Shoemaker, D. D. O'Leary, and R. L. Hughson
PETCO2 inversely affects MSNA response to orthostatic stress
Am J Physiol Heart Circ Physiol, September 1, 2001; 281(3): H1040 - H1046.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
J. A. Taylor, C. W. Myers, J. R. Halliwill, H. Seidel, and D. L. Eckberg
Sympathetic restraint of respiratory sinus arrhythmia: implications for vagal-cardiac tone assessment in humans
Am J Physiol Heart Circ Physiol, June 1, 2001; 280(6): H2804 - H2814.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Henry, R. A.
Right arrow Articles by Eckberg, D. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Henry, R. A.
Right arrow Articles by Eckberg, D. L.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online