| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of 1Anesthesiology, 2Cardiovascular Surgery, and 3Critical Care Medicine, Hospital Privado de Comunidad, Mar del Plata, Argentina; Departments of 4Critical Care Medicine and 5Pneumonology, Fundación Jimenez Díaz, Madrid, Spain; and 6Respiratory Intensive Care Unit, Pulmonary Division, Hospital das Clínicas, University of Sao Paulo, Brazil
Submitted 5 October 2004 ; accepted in final form 28 March 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
cardiopulmonary bypass; carbon dioxide; capnography
The origin of this slope has been the subject of debate for many years. Convection-dependent inhomogeneities (CDI) and diffusion- and convection-dependent inhomogeneities (DCDI) within the lungs have been postulated as the main mechanisms of phase III sloping in single- and multiple-breath tests (3, 5, 28). CDI reflect large-scale inhomogeneities caused by the flow sequence between unequally ventilated units originating from the same (intraregional CDI) or different (interregional CDI) lung regions. DCDI are small-scale inhomogeneities that result from the interaction of both diffusion and convection transport observed at asymmetric branch points within lung acini.
Single- or multiple-breath tests using poorly soluble inert gases, such as N2, mainly describe the effect of the distribution of ventilation on SIII due to their minimal diffusion through the alveolar-capillary membrane. When a highly soluble gas like CO2 is used instead, the effect of pulmonary blood flow (PBF) on the shape of SIII could play an important role too. According to these concepts, the continuous transfer of CO2 molecules by PBF into the alveolar space has been postulated as an additional reason for the sloping of phase III (3, 8, 24).
Schwardt et al. (24) used a mathematical model to describe the effects of PBF on SIII. They found that changes in cardiac output mostly affected the absolute value of the SBT-CO2 curve with only minor effects on SIII. This theoretical finding suggests that continuous CO2 elimination through the alveolar-capillary membrane does not play an important role in the sloping of phase III.
Interest has recently shifted also to the study of the slope of phase II (SII). It seems to be a useful index for diagnosing emphysema, evaluating the severity of bronchospasm, and detecting lung recruitment (18, 25, 27, 30). The rapid increase in the CO2 concentration during expiration marks the appearance of alveolar gas at the airway opening, and its origin seems to have the same mechanism as the one described for SIII. The effect of PBF on its shape has never been studied before.
The aim of this study was to determine the role of PBF in the genesis of SII and SIII of the SBT-CO2. To address this question, we used an in vivo human model of patients who were submitted to cardiac surgery during the weaning from cardiopulmonary bypass (CPB). In this model, changes in PBF can be analyzed at extreme conditions while maintaining lung ventilation constant.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
On arrival at the operating room, a right radial artery indwelling catheter and a right internal jugular vein and large-bore intravenous catheters were placed under local anesthesia. Standard monitoring included ECG, rectal temperature, and invasive systemic and central venous pressure. A pulmonary artery catheter (PAC) (Baxter Healthcare, Irvine, CA) was inserted through an internal jugular vein access in patients whose ventricular ejection fraction was lower than 40%. Thus invasive pulmonary pressures and cardiac output could be measured in those patients.
Anesthesia was induced with 1.5–2 mg/kg propofol, 10 µg/kg fentanyl, and 0.1 mg/kg vecuronium and was maintained with infusions of 80–100 µg·kg–1·min–1 propofol and 0.5–1 µg·kg–1·min–1 remifentanyl. Intraoperative fluids and inotropic and vasoactive drugs were administered according to the hemodynamic status. After tracheal intubation, the lungs were mechanically ventilated with a Narkomat anesthesia machine (Heyer Medical, Germany) in a constant-flow, volume-controlled mode with the following settings: tidal volume of 8 ml/kg, respiratory rate of 15 breaths/min, inspiratory time of 33% without pause, positive end-expiratory pressure (PEEP) of 7 cmH2O, and a fraction of inspired oxygen of 50%. Alveolar ventilation was adjusted to obtain a normal arterial PCO2 by modifying the respiratory rate while maintaining tidal volume unchanged.
CPB was established by cannulation of the ascending aorta and the right atrium. The extracorporeal circuit was primed with 1,500 ml of a mixture of lactated Ringer solution, 4% Gelafundin (Braun Medical, Crissier, Switzerland) and mannitol. Once CPB was started, mild hypothermia (32–34°C) and hemodilution (28% hematocrit) were induced. Pump flow was maintained at 2.5 l·min–1·m–2, and mean arterial pressure was kept between 50 and 70 mmHg. During CPB, blood gases were managed according to the alpha-stat protocol (26).
Protocol. All measurements were performed during an open-chest condition with the sternum retractor in place. SBT-CO2 was analyzed at three periods (Fig. 1): 1) 15 min before CPB outset (baseline measurement), 2) during CPB weaning, and 3) 15 min after CPB weaning.
|
During the CPB weaning protocol, pump flow was decreased in steps of 20%, from 100% CPB (2.5 l·min–1·m–2) to 0% CPB. In this model, we assumed that PBF increased inversely but proportionally from 0 to 20, 40, 60, 80, and 100%, respectively. Every step change of 20% in the pump/pulmonary flow was maintained for al least 2 min (Fig. 1).
At each decrement in pump flow, a partial and progressive reduction in pump venous return was performed by the perfusionist to maintain a proportional augmentation in lung blood volume and the patient’s intravascular volume status. The decrease in venous return to the patient was accomplished by partially and progressively clamping the CPB venous line in parallel with the reduction of CPB pump flow.
Criteria to move from one step of the weaning protocol to the next included a 20% decrement in blood volume inside the blood oxygenator recipient, a mean arterial pressure of 
60 mmHg, a central venous pressure between 10 and 18 mmHg (or a wedge pressure between 10 and 18 mmHg when PAC was in place), and an increase in the end-tidal CO2 (ETCO2) of at least 4 Torr. We took ETCO2 as an online marker of lung perfusion during CPB weaning (7) as ventilation and metabolism were maintained constant during the study. In preliminary patients, ETCO2 normally increases from 0 to 5 Torr at 100% CPB to 25 to 30 Torr at 0% CPB during the CPB weaning phase. We therefore considered that ETCO2 increments of at least 4 Torr at each step represented an increase in PBF of 
20%.
A final set of data was obtained 15 min after CPB weaning, once lung volume normalized and normal cardiovascular circulation was restored.
Expired gas measurements. SBT-CO2 and respiratory mechanics were recorded continuously during the protocol with the COSMOplus capnograph (Novametrix, Wallinford, CT). With this device, CO2 is measured by a main-stream sensor using the nondispersive infrared absorption technique (accuracy ±2 mmHg). Airway flow is measured by a fixed orifice differential pressure flow sensor (accuracy >3%). Before the protocol was started, a routine calibration was performed using the reference cell according to the manufacturer’s instructions.
The SBT-CO2 was monitored and recorded online on a PC using the commercially available software Aplus for Windows (Novametrix, Wallingford, CT). CO2 and respiratory mechanics raw data were downloaded from the Aplus and exported into Excel 98 (Windows, Microsoft) for offline analysis. To minimize error and variability of the analysis, the mean value of the last 15 breaths of every step was calculated for each variable.
Figure 2 shows a typical SBT-CO2 curve and its derived variables as used in this study. SII and SIII were calculated automatically by a linear regression analysis between 30 and 70% of these slopes. SII and SIII were normalized by dividing them by the mean alveolar CO2 of the corresponding expiration. Normalized SII (SnII) and SIII (SnIII) allowed comparison of slopes from breaths with different CO2 excretion rates (22), as could be expected to occur during the entire CPB weaning process.
|
Dynamic compliance of the respiratory system was calculated as expired tidal volume divided by end-inspiratory airway pressure minus total PEEP. Dynamic expiratory resistance of the respiratory system was calculated as the quotient between delta pressure and flow. These two variables related to respiratory mechanics were recorded online and used as markers for a constant ventilatory condition.
Statistical analysis. Statistical analysis was performed using the Instat software program (version 2.0, GraphPad, San Diego, CA). For comparison of variables, repeated-measurements analysis of variance was used. If the analysis of variance (F statistic) was significant, the Student-Newman-Keuls posttest was added for confirmation. PAC data (7 patients) was analyzed using the Wilcoxon test. Values are reported as means ± SD, and significance was accepted at the 5% level.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Breath tests performed with poorly soluble gases from an external source are used to describe the effects of ventilation maldistribution on the shape of SIII (3, 5, 19, 20). On the contrary, CO2 has some particular features such as its high solubility, its inverse blood-to-alveoli pathway, and its complex transport mechanism within the blood. In this regard, SBT-CO2 has a greater dependence on ventilation-perfusion relationship changes, making it almost impossible to differentiate the effects of ventilation from those of perfusion on the shape of SIII.
We describe an in vivo model in which PBF is systematically controlled, allowing a close analysis of lung perfusion during the breath tests, especially those performed with highly soluble gases such as CO2 that come from an internal source. Our results show that SII and SIII changed as lung blood flow progressively increased during the CPB weaning phase. However, when SII and SIII were normalized by dividing them by the mean expired alveolar CO2, these slopes remained unchanged. Normalizing the slopes in this way cancels out the effect of the magnitude of PBF on the SBT-CO2. Thus any change in SnII and SnIII can be interpreted as being caused by alterations in the distribution of ventilation and perfusion irrespective of the magnitude of PBF.
Phase III slope. The present study partially reproduces the findings of the model described by Schwardt et al. (24) in real patients. They used a single-path trumpet bell model to explain a number of aspects of normal expirograms. CO2 transport in the airways was simulated by a mass balance across the trumpet model to yield the airway convection-diffusion equation. The last term of this equation represents the CO2 evolution from the blood into the alveoli according to the local blood flow distribution. The model controls both the amount of PBF and its distribution, predicting two facts related to SIII: 1) SIII is minimally affected by the amount of PBF, i.e., cardiac output, when PBF is distributed to the whole lung; and 2) SIII decreases when PBF is distributed proximally until generation 17 and then increases when PBF is distributed to the lung periphery from generation 17 to 23.
In the model, stable ventilation conditions are assumed, and changes in cardiac output from 1.8 to 13.2 l/min had minimal effect on SIII. In the present study, we found a higher increment in SIII during extreme ranges of PBF but with a behavior similar to the one described in the model for a proximal-to-distal PBF distribution. Differences observed between these two findings could be explained by the characteristics of the trumpet model in which a single symmetric path cannot represent the true multibranched asymmetric structure of the airways and pulmonary vessels (6).
We believe that when PBF increases in our patients, it follows a proximal-to-peripheral distribution vector. This gravity-independent central-to-peripheral PBF gradient simulated in Schwardt’s model has been observed in humans (15) as well as in other mammals (1, 12, 13, 14, 29). This finding is consistent with the fractal branching pattern of pulmonary circulation where vascular resistance is increased due to branching and the existence of longer circuit pathways. Hakim et al. (16) showed in dogs that increments in cardiac output were associated with an increase in the absolute blood flow to both the central and the peripheral regions of the lung, while maintaining its central-to-peripheral gradient.
We speculate that, under constant ventilation conditions, as PBF increases and reaches more peripheral regions during CPB weaning, the asymmetric geometry of the vascular bed creates lung areas with different alveolar CO2 concentrations. Similar to the CDI mechanism, inter- and intraregional gradients of CO2 are produced in the alveoli by the regional pulmonary perfusion. This phenomenon causes a sloping of the CO2 tracer irrespective of any inhomogeneity in the distribution of ventilation. In this way, normalized slopes cancel the effect of the magnitude of PBF. In addition, as alveolar gas is exhaled, ventilation-dependent CDI and DCDI generate an additional and known cause of phase III sloping.
Although this theoretical assumption supports the concept that the amount and distribution of PBF are interrelated phenomena, whether normalization has any effect on the distribution of PBF will have to be confirmed in further studies in which PBF magnitude and distribution can be manipulated separately.
Phase II slope. The model described by Schwardt et al. (24) did not study the effects of PBF on SII, and to our knowledge the hemodynamic-related effect on SII has not been documented yet. On the contrary, the effect of ventilation distribution on SII is now a better understood phenomenon. Asthmatic and emphysematous patients (18, 25, 30) show a decrease in SII when compared with normal subjects. In mechanically ventilated patients with anesthesia-induced atelectasis, we have shown that, after a lung-recruitment maneuver in which the lung is reexpanded, SII increases (27). The fact that lung recruitment increases SII indicates that the reopening of collapsed small airways results in a more homogeneous distribution of ventilation. Therefore, an inhomogeneous distribution of ventilation seems to influence SII.
In a recent study, Schulz et al. (23) describe the ventilation-induced changes on phase II independent from the influences of phase III by using labeled CO2 (C18O2). This technique might be very helpful in studying the determinants of PBF in more detail.
Despite the known influence of the distribution of ventilation on SII, our results have shown that PBF has an additional effect on this slope. SII showed its highest values before and after CPB. The lowest values were found during the highest pump flow (lowest PBF), observing a progressive increment as PBF was increased.
VTCO2,br. VTCO2,br depends directly on PBF. VTCO2,br progressively increased during CPB weaning because more CO2 molecules were transported into the lungs to be eliminated by ventilation. Our findings with respect to VTCO2,br behavior agree with those observed in the theoretical simulation performed by Schwardt et al. (24). These authors demonstrated that cardiac output and venous CO2 tension have a direct relationship with VTCO2,br values because average alveolar CO2 concentration mainly depends on the rate at which CO2 molecules are delivered by PBF.
Pitfalls. We assumed that PBF increases proportionally and inversely to the step decreases in CPB pump flow. However, this proportionality does not necessarily exist in the beating heart because the cardiac output is dependent on many factors such as contractility, cardiac rate, preload, and afterload. These factors undergo important changes in response to the ischemia-reperfusion phenomena responsible for the post-CPB stunning heart and the effects of the proinflammatory state that is associated with CPB circulation.
In addition, during CPB weaning, preload depends mainly on the control exerted by the perfusionist over the venous return from the pump. In this subjective maneuver, the perfusionist clamps the venous line returning to the patient to achieve a gradual and proportional increase in preload. It is common to reach the normovolemic state during CPB weaning within several minutes after all residual volume of blood inside the pump is reinfused to the patient through the arterial cannula. Thus, strictly speaking, the step changes in PBF could only yield a rough approximation of the protocol-targeted 20%. In our view, however, the progressive change in PBF and not the exact rate of change was important to study the effects of PBF on SBT-CO2.
The SBT-CO2 variables immediately after CPB weaning did not reach baseline and post-CPB values. The "stunning" heart (i.e., a myocardium failure observed immediately after CPB weaning caused mainly by ischemia reperfusion, electrolyte dysbalance, and the inflammatory response) and the transient hypovolemic state observed at this moment of the protocol were responsible for this limitation, a situation that cannot be overcome in clinical cardiac surgery under CPB.
In conclusion, this in vivo human model showed that PBF plays an important role in the genesis and morphology of the SBT-CO2 slopes. Nonnormalized SII and SIII were affected by the different degrees of PBF, whereas normalized slopes were not. Normalized slopes describe ventilatory and perfusion inhomogeinities in mechanically ventilated patients independent of changes in the magnitude of PBF. The role of slope normalization on PBF distribution remains to be determined.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. H. Bohm, S. Maisch, A. von Sandersleben, O. Thamm, I. Passoni, J. M. Arca, and G. Tusman The Effects of Lung Recruitment on the Phase III Slope of Volumetric Capnography in Morbidly Obese Patients Anesth. Analg., July 1, 2009; 109(1): 151 - 159. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
Post-CPB (after CPB) compared with 100, 80, 60, 40, and 20% of PBF (P < 0.05). Differences in SII were statistically significant between all % of PBF values.
