|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Intensive Care Unit (Dr. Nakos) and Department of Pulmonary Medicine (Dr. Costantopoulos), Ioannina University Hospital, Ioannina, Greece; and the Department of Intensive Care Unit (Dr. Lahana), Athens Veterans Hospital, Athens, Greece.
Correspondence to: George Nakos, MD, FCCP, Director, Intensive Care Unit, Ioannina University Hospital, 45500 Ioannina, Greece; e-mail: gnakos{at}cc.uoi.gr
| Abstract |
|---|
|
|
|---|
Design: Experimental study.
Setting: A university hospital.
Interventions: A simulated central apnea model achieved by a particular setting of the mechanical ventilator in 10 brain-dead patients.
Measurements: Hemodynamic studies using right-heart catheterization and continuous monitoring of arterial blood gas levels.
Results: Hypercapnic hypoxic apneas were associated with no change in heart rate, fall in mean systemic arterial pressure and systemic vascular resistance (from 83 ± 9 to 68 ± 7 mm Hg and 1,115 ± 82 to 768 ± 58 dyne·s·cm5, respectively; each p < 0.05), and rise in mean pulmonary artery pressure, pulmonary vascular resistance, and pulmonary capillary wedge pressure (PCWP) [from 17 ± 1.5 to 26 ± 3 mm Hg, 102 ± 27 to 166 ± 43 dyne·s·cm5, and 10 ± 1 to 14 ± 2 mm Hg, respectively; each p < 0.05].
Conclusion: Our results suggest that in the absence of central autonomic regulation in humans, apnea-induced hypoxemia and/or hypercapnia are associated with peripheral vasodilatation and pulmonary vasoconstriction, which are probably local in origin, as well as a significant increase in PCWP indicating cardiac dysfunction.
Key Words: acute cardiovascular response brain death central apnea hypercapnia hypoxemia
| Introduction |
|---|
|
|
|---|
In the present study, we used a model of central apnea in brain-dead patients. In brain death, the central neural drive to the respiratory muscles, sympathetic neural activity, as well as resting vagal tone are abolished irreversibly.6 We used this model to induce apneas without fluctuations in intrathoracic pressure and central autonomic activity and to investigate the local effects of hypoxemia and/or hypercapnia on systemic and pulmonary circulations. We performed hemodynamic studies during apnea-ventilation cycles, and we posed the following questions: (1) what is the local effect of hypoxemia and/or hypercapnia on the systemic and pulmonary circulations during nonobstructive apneas? and (2) in the absence of central autonomic regulation, is there any influence of these stimuli on the heart?
| Materials and Methods |
|---|
|
|
|---|
|
Before the beginning of the hemodynamic studies, the assessment of the stability of the apnea model was done by allowing eight apnea-ventilation cycles. The variations in arterial blood gas levels, mean systemic arterial pressure (SAPm), and temperature, as well as cardiac output (CO) and mixed venous oxygen saturation (S
O2) before and after the apnea-ventilation cycles were monitored. Patients were excluded from the protocol when serious adverse effects, such as cardiac arrhythmia or hemodynamic collapse, were observed.
Measurements
Values for arterial blood gases and hemodynamic parameters were obtained during three sequential apnea-ventilation cycles, at the following four specified periods: baseline, before any apneic intervention; late apnea, the last 60 s of the disruption of ventilation; postapnea, the first 60 s following the end of apnea; and recovery, 3 min after stabilization under baseline conditions. Continuous monitoring of PaO2 and PaCO2, pH, and arterial oxygen saturation (SaO2) were obtained by a multiparameter intraarterial fiberoptic sensor that was passed via a 20-gauge catheter in radial artery (Seldinger Paratrend 7 system; Biomedical Sensors; High Wycombe, UK). Concomitantly with the arterial blood gas measurement, the monitoring of SAPm was done through the same artery.
Right-heart catheterization was performed using a standard technique through the subclavian vein. A 7.5F, balloon-tipped, flow-directed continuous CO pulmonary artery catheter (CCOmboV; Edwards Lifesciences LLC; Invire, CA) was pushed forward under the control of the pressure curve on the screen of a multichannel monitor (HP 78353B; Hewlett-Packard; Palo Alto, CA). Mean right arterial pressure (RAPm), and mean pulmonary artery pressure (PAPm) were recorded on the monitor screen. Mean pressures were obtained by electronic averaging of the values for systolic and diastolic pressures. Pulmonary capillary wedge pressure (PCWP) was measured after inflation of the distal balloon. All of the measurements were obtained at the end of expiration.
Continuous CO monitoring was done using the thermodilution principle8 by a continuous CO computer (Vigilance Volumetric CEDV; Edwards Life Sciences; Irvine, CA). The values of CO and S
O2 were presented on the STAT mode screen of the continuous CO computer and were updated every 18 s. We obtained the values of CO that corresponded to the specified periods of the apnea model. The mean average of three recorded values was calculated. Stroke volume (SV), cardiac index (CI), systemic vascular resistance (SVR), pulmonary vascular resistance (PVR), arterial oxygen content (CaO2), venous oxygen content (CvO2), and oxygen consumption (
O2) were calculated using the following standard equations: (1) SV = CO/heart rate (HR) x 103 (mL/beat); (2) CI = CO/body surface area (BSA) (L/min/m2); (3) SVR = (SAPm RAPm) x 80/CI (dyne·s·cm5); (4) PVR = (PAPm PCWP) x 80/CI (dyne·s·cm5); (5) CaO2 = 1.34 x hemoglobin (Hb) x SaO2 (mL/100 mL); (6) CvO2 = 1.34 x Hb x S
O2) (mL/100 mL); and (7)
O2 = CO x (CaO2 CvO2) (mL/min).
Statistics
Data are presented as mean ± SD. Statistical comparisons the between groups were performed using two-way analysis of variance for repeated measurements. The significance of differences between the values was determined with the Bonferroni correction for multiple comparisons. A p value of < 0.05 was considered to be significant.
| Results |
|---|
|
|
|---|
O2 throughout the apnea-ventilation cycles.
Blood Gases
In comparison with baseline and recovery, apneas led to a statistically significant decrease in PaO2 and arterial pH, and an increase in PaCO2. At postapnea, hypoxemia was corrected, but mild hypercapnia persisted (Table 2
).
|
|
Postapnea vs Late Apnea:
At postapnea, mean SAPm and SVR were significantly greater in comparison with late apnea (82 ± 9 vs 68 ± 7 mm Hg, respectively; and 1,128 ± 94 vs 768 ± 58 dyne·s·cm5, respectively [each p < 0.05]). When compared with late apnea, mean CO and SV were significantly lower (5.1 ± 1 vs 6.2 ± 1 L/min, respectively; and from 73 ± 14 vs 90 ± 15 mL/beat, respectively [each p < 0.05]). Mean PAPm, PVR, and PCWP were found significantly lower at postapnea than in late apnea (19 ± 2 vs 26 ± 3 mm Hg, respectively; 126 ± 27 vs 166 ± 43 dyne·s·cm5, respectively; and 11 ± 3 vs 14 ± 2 mm Hg, respectively [each p < 0.05]).
Postapnea vs Baseline:
Although PVR decreased at postapnea in comparison with late apnea (126 ± 27 vs 166 ± 43 dyne·s·cm5, respectively [p < 0.05]), it remained higher compared with baseline values (126 ± 27 vs 102 ± 26 dyne·s·cm5, respectively [p < 0.05]). At postapnea, PAPm was higher than at baseline but not significantly.
Recovery:
The values of all of the hemodynamic parameters in recovery were not different from those at baseline.
| Discussion |
|---|
|
|
|---|
Blood Gases
In this study, hypoxemia and hypercapnia occurred during the last part of apnea. This is not consistent with previous studies in patients with apnea,9 where the changes in arterial blood gas levels were observed after the termination of apnea. A possible reason for this difference could be that, in our study, the continuous arterial blood gas monitoring was done through an intravascular sensor, which is characterized by a fast dynamic response to changes in blood gas tension.10 The ear or finger oximeters that have been used in previous studies had a relatively limited rate of dynamic response to abrupt changes in arterial blood gas levels, which are recorded after a brief circulation delay, estimated at 10 to 35 s.11 We think that our measurements reflect more precisely the changes in arterial blood gas levels during the apnea-ventilation cycle. Another possible reason for this difference could be the longer duration of apnea in our model.
One could argue that the modest changes in blood gas values could be attributed to a reduced metabolic rate. However, the values of calculated
O2 before, during, and after apnea did not shown significant variations. Therefore, we suppose that the metabolic rate did not change during apnea.
HR and Right Arterial Pressure
HR remained unchanged throughout the apnea-ventilation cycle. Additionally, no change in HR was noticed during nonobstructive apneas in vagotomized and chemically sympathectomized pigs.5 This finding is also consistent with data from previous studies in brain-dead patients.121314
RAPm remained unchanged throughout the apnea-ventilation cycle, possibly because no respiratory movements occurred. Furthermore, RAPm was not significantly affected by the withdrawal and reinstitution of mechanical ventilation. It could be suggested that the intrathoracic pressure did not significantly change throughout the apnea-ventilation cycles. It is probable that the relatively low tidal volume used, in the absence of positive end-expiratory pressure (and/or auto-positive end-expiratory pressure), hardly affected the intrathoracic pressure in hemodynamically stable patients with normal respiratory systems. Therefore, it is quite possible that mechanical ventilation did not influence heart function, in terms of venous return, preload and afterload.
Systemic Arterial Pressure and SVR
In response to apnea, SAPm fell in parallel with a fall in SVR. These findings are consistent with data from animal studies,1516 in which hypoxemia and hypercapnia caused local vasodilatation in many vascular beds. After apnea termination, mild and nonhypoxic hypercapnia was associated with a rise in SVR and SAPm, suggesting that nonhypoxemic mild hypercapnia was not associated with vasodilatation. This finding is not consistent with those in previous studies in brain-dead patients,1417 in which hypercapnia was associated with systemic hypotension and a fall in SVR. Furthermore, studies in chemically sympathectomized and vagotomized animals5 have shown that vasodilatation occurs during brief hypercapnic, nonobstructive apneas, whether they were hypoxic or hyperoxic. Differences in the PaCO2 level and/or the duration of apnea could be responsible for these different findings. It should be noted that the apneas in those studies were induced in sedated pigs.
CO and SV
Compared with preapnea values, CO and SV values increased only slightly during apnea, but without reaching statistical significance. One could hypothesize that the limited increase in SV and CO in response to peripheral vasodilation could be attributed to diastolic dysfunction of the left ventricle. This hypothesis is supported by the concomitant rise in PCWP, which is an index of left ventricular end-diastolic pressure. At the same time, the increase in PAPm and PVR suggests that pulmonary hypertension-induced right ventricle-septal dysfunction causes diminished left diastolic ventricular function. This hypothesis is consistent with those of previous studies, in which pulmonary hypertension during apnea led to an increase in right ventricle end-diastolic pressure. The latter produces a marked leftward septal shift causing a decrease in left ventricle compliance and left ventricle end-diastolic volume.1819 On the other hand, the rise in PCWP could be attributed to decreased left ventricular contractility due to hypoxia unopposed by sympathetic stimulation.2021 However, in other studies in brain-dead patients,13141518 CO levels were significantly increased in response to a fall in SVR during long hypercapnic apneas. This difference could be attributed to the absence of hypoxemia during apnea in those studies, as well as to different methods of evaluation of heart function. In our model, hypoxemia could affect the function of both heart ventricles. However, in chemically sympathectomized and vagotomized sedated pigs, CO and SV were increased during brief nonobstructive hypoxemic and hypercapnic apneas.5 Differences in the neural state (sedation vs brain death) or in species could be responsible for these different findings. However, we should note that brain death could affect the cardiovascular and epithelial function altering the release of mediators, such as catecholamines. Hypoxia and hypercapnia, furthermore, might enhance these alterations induced by brain death. Therefore, the behavior of the cardiovascular system could be different in patients without brain death.
After the termination of apnea, CO and SV levels were decreased in comparison with the late apnea period. Because HR remained constant, the fall in CO and SV could be attributed to increased SVR and a short-term increase in left ventricular afterload. Also, it is possible that the effect of hypoxemia on the heart was persistent.
Pulmonary Circulation
PAPm and PVR were increased during apnea and fell again in the recovery. These findings indicate that hypoxemia and hypercapnia, when acting locally, cause pulmonary vasoconstriction even in the absence of central autonomic control. This finding is consistent with data from animal studies,22 in which it was observed that hypoxemia acts directly on the smooth muscles of the pulmonary vasculature to produce vasoconstriction.
Although PVR values declined after late apnea termination, they remained higher than those at baseline. Mild hypercapnia without hypoxemia was associated with pulmonary vasoconstriction. These findings indicate that the local effect of carbon dioxide seems to play a significant role in the reaction of the pulmonary vasculature to apnea. Pulmonary circulation is probably more sensitive than the systemic to the local hypercapnic stimulus.
Limitations
A limitation of this study is that our findings could not be extrapolated to real central apneas where central autonomic activity exists. However, it could be suggested that autonomically impaired individuals could be at a significant risk during central sleep apneas because of the local effects of hypoxemia and hypercapnia on the cardiovascular system. Another limitation is the lack of monitoring of heart function during apnea, for example, by echocardiography, hence the direct evaluation of the influences of hypoxemia on the heart.
| Conclusion |
|---|
|
|
|---|
| Footnotes |
|---|
O2 = mixed venous oxygen saturation; SVR = systemic vascular resistance;
O2 = oxygen consumption. Received for publication May 21, 2004. Accepted for publication December 22, 2004.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |