Anesth Analg 2006;102:1201-1206
© 2006 International Anesthesia Research Society
doi: 10.1213/01.ane.0000198664.74731.d3
CRITICAL CARE AND TRAUMA
Regional Blood Flow in Respiratory Muscles During Partial Ventilatory Assistance in Rabbits
Akinori Uchiyama*,
Yuji Fujino*,
Kikumi Hosotsubo*,
Eriko Miyoshi*,
Takashi Mashimo*, and
Masaji Nishimura
*Osaka University Hospital Intensive Care Unit, Yamadaoka, Suita; and Department of Emergency and Critical Care Medicine, University of Tokushima Graduate School, Kuramotocho, Tokushima, Japan
Address correspondence and reprint requests to Akinori Uchiyama, Osaka University Hospital Intensive Care Unit, 2-15 Yamadaoka, Suita, Osaka Prefecture 565-0871, Japan. Address e-mail to auchiyama{at}hp-icu.med.osaka-u.ac.jp.
 |
Abstract
|
|---|
We tested the hypothesis that even partial ventilatory assistance would reduce respiratory muscle blood flow to levels similar to those found during control mechanical ventilation (CMV). Three levels of pressure support ventilation (PSV) and 2 CMV settings were compared in 10 rabbits. PSV 0, 6, and 12 cm H2O, under continuous positive airway pressure mode, were applied, and then pressure control ventilation (PCV) values of 6 (36 breaths/min) and 12 cm H2O (18 per breaths/min) were applied to each CMV setting with a muscle relaxant. Using colored microspheres, we measured regional tissue blood flow in respiratory muscles, lower extremities, kidney, and liver. Regional tissue blood flow in the diaphragm during PSV6, PCV6, and PCV12 were less than those during PSV0. During PSV12, blood flow in the crural diaphragm was more than that during PCV12 and similar to that during PSV0. Whereas the transdiaphragmatic pressure of PSV6 was –0.8 ± 1.6 cm H2O, that of PSV12 was –3.1 ± 2.4 cm H2O. Inspiratory asynchrony, arising from an ineffective triggering effort, was observed in PSV12. The ventilatory settings did not affect blood flow of the lower extremities, liver, and kidney. In conclusion, ventilatory settings affected blood flow in the diaphragm. At certain PSV settings, blood flow in the diaphragm was minimal.
|
Introduction
|
|---|
Respiratory muscle blood flow increases as the work of breathing increases (1–4). In critical care patients, increased respiratory muscle blood flow caused by the larger respiratory load reduces blood flow to other major organs (1–3). Mechanical ventilation is useful for maintaining blood flow to the other major organs in critically ill patients (5,6). Mechanical ventilation is a well-established method of reducing the patient’s work of breathing. Because mechanical ventilation alters both blood flow and the metabolic demands of respiratory muscles, it may also affect the balance between blood supply and metabolic demand, influence the development of respiratory muscle fatigue, and interfere with recovery from fatigue (7,8). Previous studies have indicated that completely controlled mechanical ventilation (CMV) with or without muscle relaxants, in which spontaneous breathing effort almost disappears, reduces respiratory muscle blood flow to a very low level (1,5,6).
Recently, partial ventilatory assistance modes that preserve spontaneous breathing effort have come into widespread use. Because these modes make it easy to control the amount of support the ventilator provides, physicians can change the patient’s work of breathing by adjusting the ventilator settings (9,10). However, the effects of ventilatory support on respiratory muscle blood flow have not yet been elucidated. In the present study, we tested the hypothesis that even partial ventilatory assistance, if high enough, would reduce respiratory muscle blood flow to levels similar to those found during totally CMV. In a rabbit model, using the multiple colored microspheres method, we measured regional tissue blood flow in respiratory muscles, lower extremities, liver, and kidney under different mechanical ventilatory settings, including totally CMV and partial ventilatory assistance.
|
Methods
|
|---|
This study was approved by the Osaka University Animal Use Committee. Ten New Zealand white rabbits (body weight, 3.4 ± 0.6 kg) were studied under general anesthesia. Anesthesia was induced by an IV bolus infusion of propofol 3 mg/kg and ketamine 3 mg/kg. After subcutaneous infiltration of a local anesthetic (1.0% lidocaine), the animals underwent tracheotomy with a 4-mm internal diameter endotracheal tube. To inject colored microspheres, a 19-gauge catheter was inserted into the left ventricle through the left carotid artery. The position of this catheter was determined by pressure tracing at the distal end. To draw reference blood samples and measure arterial blood pressure, the right femoral artery was cannulated with a 20-gauge catheter. To enable measurement of esophageal and gastric pressures, thin-walled latex balloons were inserted into the midesophagus and stomach through the cervical esophagus. Both balloons were inflated with 0.3 mL of air just before each measurement.
During the study, anesthesia was maintained by continuous infusion of ketamine (10 mg·kg–1·h–1) and propofol (20 mg·kg–1·h–1). If the animal exhibited voluntary movements of the extremities, the infusion rates of the drugs were increased. If the continuous infusion rates of the anesthetics were changed, the protocol was interrupted. The protocol was restarted after confirmation of a stable anesthetic state of 15-min duration.
Regional tissue blood flow was determined with a technique using 15-µm microspheres dyed with 1 of 5 different colors (Triton Technology Inc., San Diego, CA) during the 5 experimental ventilatory settings (11). This method allows five measurements of regional tissue blood flow in a single experimental preparation. For each measurement, about 3 million microspheres suspended in 1 mL were injected. The withdrawal of arterial reference blood samples was started 15 s before injection of the microspheres and continued for 150 s at a rate of 0.7 mL/min. Postmortem, tissue samples were taken from the crural and costal diaphragms, the rectus abdominis, the lower extremity muscles, the liver, and the kidney. Each organ was cut into samples of 1.0 g. Seven milliliters of a 4-M KOH solution containing 2% Tween 80 was added to each sample for digestion of the tissue. The glass tubes were placed in a water bath shaker for 4 h at 72°C. To extract the dye, the digested tissue solution was placed on a polyester filter with a burette while the filter was rinsed with 2% Tween 80 solution. The reference blood samples were processed in the same manner as the tissue samples. The color spectra of each organ sample and reference blood samples drawn from arteries were measured with a spectrophotometer (BioSpec-1600; Shimadzu, Kyoto, Japan). The regional tissue blood flow of each organ and cardiac output were calculated from the color spectra using the MISS program (Triton Technology, Inc.).
The MISS program can handle five different colors, many tissue samples, and reference blood samples per color. The program interacts with the photospectrometer and stores the spectra of each organ sample and each reference blood sample. The basis for the calculation of regional tissue blood flow is the matrix inversion technique (12). The absorption spectrum of each dye was measured separately and was used as a reference for the matrix inversion technique. The matrix inversion technique determines the contribution of each color to the measured composite spectra of the organ sample at five fixed wavelengths, one of which corresponds to each color. Thereby, a composite spectrum, including all colors, is separated into individual portions for each color in any given organ sample. The regional tissue blood flow and the cardiac output in each experimental setting were calculated from the individual data of each color in the MISS program.
For ventilation, we used a Servo 300 (Siemens-Elema, Solna, Sweden) and a standard pediatric ventilatory circuit without a humidifier. Every animal was examined under 5 settings: pressure support ventilation (PSV) 0, 6, and 12 cm H2O and pressure control ventilation (PCV) 6 and 12 cm H2O. After the preparatory procedures, followed by 30 min of stabilization, microspheres of one color per setting were injected. Randomly applied, the PSV settings were 0, 6, and 12 cm H2O under continuous positive airway pressure mode. Then, after muscle relaxant (pancuronium bromide 1 mg/kg IV) was administered, PCV 6 cm H2O (assist/control mode; respiratory rate = 36 breaths/min and inspiratory time = 0.6 s) and PCV 12 cm H2O (assist/control mode; respiratory rate = 18/min and inspiratory time = 0.6 s) were randomly applied. Inspiratory triggering sensitivity was set at the minimum level for detecting inspiratory effort in the flow-triggering mode. Fraction of inspire oxygen was set at 0.7 and positive end-expiratory pressure was set at 0 cm H2O throughout the experiment.
To confirm stabilization of the respiratory status at each experimental setting, blood flow was measured 15 min after the ventilatory setting was changed. In the preliminary study, the respiratory rate decreased to about 20 breaths/min, far less than the normal respiratory rate of a healthy rabbit. In addition to respiratory depression caused by the anesthetics, other reasons for the decreased respiratory rate were thought to be a reduction of CO2 production because of the general anesthesia and a decrease in dead space as a result of the tracheostomy. Therefore, we mixed CO2 into the inspiratory gas to maintain the inspiratory concentration of CO2 at 0.75% throughout all the protocols.
Respiratory flow and airway pressure signals were obtained from the analog output of the ventilator. Pressure transducers (TP603T; ±50 cm H2O; Nihon Kohden, Tokyo, Japan) were used to monitor esophageal pressure (Pes) and gastric pressure (Pga). The data signals from these devices were amplified (AR601G; Nihon Kohden), digitized, and recorded at a 100-Hz sampling rate using data acquisition software (WINDAQ; Dataq Instruments, Akron, OH). A single minute’s worth of recorded data was later analyzed for each setting. The respiratory rate was calculated from the respiratory flow tracings. Using WINDAQ, transdiaphragmatic pressure (Pdi) was calculated from the differences between the Pes and Pga values. Negative values for Pdi indicated that the positive values for Pes deflection were larger than those for Pga deflection during mechanical ventilation. The rates of respiratory effort were derived from the Pes tracings. Negative deflection in Pes tracings was treated as evidence of inspiratory effort. Arterial blood gas data, mean arterial blood pressure, heart rate, tidal volume, respiratory rate, respiratory effort rate, mean airway pressure, and Pdi were measured at each setting.
All values are expressed as mean ± sd. Significance was set at P < 0.05. The data were analyzed using one-way analysis of variance for repeated measures. Post hoc analysis was performed with Tukey honestly significant difference test by using a software application (Statistica 5.1; Stat Soft, Tulsa, OK). The relationships between regional tissue blood flow and other factors were assessed by regression analysis. The regression analysis was performed by a software application (StatView 4.58; Abacus Concept, Inc., Berkeley, CA).
|
Results
|
|---|
The measurements at all the experimental settings were finished within 150 min of the beginning of the protocol. Circulatory and respiratory characteristics at each setting are shown in Table 1. Heart rate, mean arterial blood pressure, cardiac output, and arterial blood gas data did not differ significantly among the settings (Table 1). Mean airway pressure at PSV0 was significantly less than at other settings (P < 0.05). Mean airway pressures at PSV6 and PSV12 were significantly less than those at PCV6 and PCV12 (P < 0.05). Pdi values at PSV6 and PCV6 were nearly equal to zero, whereas those at PSV12 and PCV12 were negative. Respiratory rate and respiratory effort rate were significantly greater at PSV0 than at other settings (P < 0.05). Although respiratory rate at PSV12 was significantly less than that at PSV6 (P < 0.05), respiratory effort rates at PSV6 and PSV12 were not different. Figure 1 shows typical tracings of airway pressure and Pdi during PSV12. The difference between respiratory rate and respiratory effort rate shows ineffective triggering efforts, which indicate inspiratory asynchrony.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 1. Typical airway pressure tracings and transdiaphragmatic pressure during pressure support ventilation at 12 cm H2O (PSV12).
|
|
Regional tissue blood flow in the costal and crural diaphragms at PSV6, PCV6, and PCV12 was significantly less than that at PSV0 (P < 0.05). Regional tissue blood flow in the crural diaphragm was significantly more at PSV12 than at PCV12 (P < 0.05) (Table 2). The crural diaphragmatic blood flow at PSV12 was higher relative to PSV6; however, this difference did not reach significance (P = 0.19). The blood flow of other organs did not vary significantly among the five settings (Table 2).
The relationship between Pdi and regional tissue blood flow of the crural diaphragm is shown in Figure 2. Regional tissue blood flow of the crural diaphragm correlated significantly with Pdi (Y = 0.003 x X2 + 0.018 x X + 0.107, Y; regional tissue blood flow (mL·g–1·min–1), X; Pdi (cm H2O); r = 0.466; P < 0.05). The tissue blood flow of the crural diaphragm increased in high Pdi. When Pdi was close to zero, regional tissue blood flow of the crural diaphragm was minimal. In contrast, the tissue blood flow also increased as Pdi became more negative. At PSV12, all the animals had large negative values of Pdi and high levels of regional tissue blood flow. The relationship between regional tissue blood flow and the frequency of ineffective triggering efforts, which consisted of the difference between respiratory rate and respiratory effort rate during PSV, is shown in Figure 3. There was no significant correlation between regional tissue blood flow and frequency of ineffective triggering efforts. The relationship between regional blood flow and the product of ineffective triggering frequency and mean Pdi differences during ineffective triggering, which represented diaphragmatic work for ineffective triggering, was also shown in Figure 4. There was no significant correlation between them.
View larger version (19K):
[in this window]
[in a new window]
|
Figure 2. Relationship between transdiaphragmatic pressure (Pdi) and regional tissue blood flow in crural diaphragm. There was a significant correlation between Pdi and regional tissue blood flow of the crural diaphragm.
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Figure 3. Relationship between regional tissue blood flow of the crural diaphragm and frequency of ineffective triggering (respiratory effort rate – respiratory rate).
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Figure 4. Relationship between regional tissue blood flow of the crural diaphragm and product of ineffective triggering effort frequency and mean transdiaphragmatic pressure changes during ineffective triggering efforts.
|
|
|
Discussion
|
|---|
This study had three major findings. First, ventilatory settings affected regional tissue blood flows in the diaphragm. Regional tissue blood flow in the crural and costal diaphragms during PSV6 were minimal and similar to those during PCV6 and PCV12. Second, during PSV12, which provided greater partial ventilatory assistance than PSV6, regional tissue blood flow in the crural diaphragm was more than that in PCV12 and not different from that in PSV0. Increasing PSV level did not necessarily lead to a reduction of regional tissue blood flow in the diaphragm. Third, the ventilatory settings tested in this experiment did not affect regional tissue blood flows of the lower extremities, liver, and kidney.
Regional tissue blood flows in the diaphragm during PSV6, PCV6, and PCV12 were similar to previously reported blood flow data (1,5,6) and were similar to the flows in the lower extremities, which were almost inactive. We found that diaphragmatic blood flow during PSV12 was similar to that during PSV0.
Many factors influence respiratory muscle blood flow during mechanical ventilation. First, muscle activity determines the muscle’s metabolic demand and affects blood flow (13). Second, the positive airway pressure has some effect on circulatory status. However, the five settings in the present study produced no significant differences in cardiac output. Third, positive airway pressure might influence diaphragmatic tension. High abdominal cavity pressure decreases diaphragmatic blood flow during tetanic contraction (14). It was possible that the efficacy of the muscle’s pumping action within the diaphragm may differ between spontaneous breathing and mechanical ventilation (15,16).
Blood flow of the crural diaphragm was correlated with Pdi (Fig. 2). The increase in blood flow of the crural diaphragm as Pdi increased was attributed to the increase in muscular activities. When Pdi was close to zero, the regional tissue blood flow of the crural diaphragm was minimal. In that case, the diaphragmatic activity was thought to be minimal. In contrast, when Pdi was negative, as seen in PSV12, in which the positive pressure in the pleural cavity pushed out the component of the abdominal cavity on contact with the diaphragm, the blood flow of the crural diaphragm increased. Thoraco-abdominal pressure swings during nonphysiological passive diaphragmatic movement, and asynchrony between spontaneous breathing effort and mechanical ventilation might affect the blood flow of the diaphragm at PSV12.
The difference between respiratory rate and respiratory effort rate is a result of ineffective triggering efforts and indicates inspiratory asynchrony (Fig. 1). At high levels of PSV, associated with a tidal volume of 10 mL/kg or more, the incidence of ineffective effort increases as the PSV level increases (17). Meanwhile, increases in the amount of ventilator assistance cause more frequent rates of ineffective triggering (18). In the present study, ineffective triggering efforts occurred more often at PSV12 than that at PSV6. Thus, even if more ventilatory assistance is applied, the asynchrony between spontaneous breathing effort and the assistance provided by the ventilator might not reduce the respiratory muscle activity. Whereas ineffective triggering efforts were seen during PSV12 in all the animals, neither frequency of ineffective triggering nor work for ineffective triggering efforts was related to the diaphragmatic tissue blood flow in the present study. The ventilatory settings, which induced ineffective triggering, were assumed to lead to increase of the diaphragmatic blood flow.
In critical care patients, one of the purposes of mechanical ventilation is to transfer blood flow from the respiratory muscles to the major organs by reducing the work of breathing (5,6). According to the results of the present study, if an appropriate level of partial ventilatory assistance is given to critically ill patients, this assistance can reduce respiratory muscle blood flow and maintain blood flow to other major organs at the same levels as with CMV. However, in the present study, no significant difference was found in the hepatic and renal blood flows between each ventilatory setting. There could be several reasons for this: because the rabbits were under general anesthesia and the inhaled CO2 concentration was kept small, the animals were not hyperventilating; because our rabbits had normal lungs with high-lung compliance and low airway resistance, the amount of increased blood flow required for the respiratory muscles was small; or because the animals did not have circulatory failure, the effects of the mechanical ventilation were small. If blood flow in major organs were examined while the respiratory muscles’ metabolic demands were either further increased, placed under conditions of circulatory failure, or both, the partial ventilatory assistance settings might have affected hepatic and renal blood flows in ways similar to those in previous reports (1,3). Further study is required to examine the effects of partial ventilatory assistance on blood flow in major organs in such situations.
Clinically, it is important to determine a level of PSV specific to each patient. Our data show that a higher assistance level, such as PSV12, which provided greater partial ventilatory assistance than PSV6, does not reduce inspiratory muscle blood flow. There is a possibility that excessive PSV levels have negative effects, such as muscular atrophy and ventilator-induced diaphragmatic injury (19). Although Pdi might be a key factor in determining PSV level, it would be difficult to measure Pdi continuously. As seen in PSV12, an irregular respiratory rhythm accompanied by ineffective triggering efforts might exhibit a negative Pdi value and an excessive level of PSV.
One limitation of our experimental method is that the regional tissue flow was measured at a single point, marked by the injection of microspheres. The timing of the injection may have influenced the data of the regional blood flow. Throughout the experiment, we mixed CO2 into the inspiratory gas to achieve a normal respiratory rate for the rabbits. In the present study, the Paco2 values at all the experimental settings were within the normal range. In our animals, the minute ventilation ranged from 1100 mL/min (PSV0) to 830 mL/min (PCV12). The minute ventilation of healthy awake rabbits has been reported to be 1240 mL/min (20). The respiratory rate of a healthy rabbit is 56–157 breaths/min (21). Our rabbits were not in a hyperventilation state. Inhaled CO2 was thought to have little effect on the results of the present study. Even with mechanical ventilation, the subjects in some studies exhibited muscular activity (22). We used paralytic drugs in the PCV groups to confirm the resting state of the diaphragm in CMV.
In conclusion, during mechanical ventilation the ventilatory settings affected regional tissue blood flows in the diaphragm. In the diaphragm, it was possible to achieve minimal blood flow at certain levels of PSV that are similar to those during full-CMV with muscle relaxants. In mechanically ventilated rabbits, blood flow of the crural diaphragm may increase with increasing PSV level, especially when the PSV level increase promotes ventilator-animal synchrony.
|
Footnotes
|
|---|
Supported, in part, by Grant-in-Aid for Scientific Research No. 13770830 from the Japan Society for the Promotion of Science, Tokyo, Japan.
Accepted for publication November 18, 2005.
|
References
|
|---|
- Robertson CH Jr., Pagel MA, Johnson RL. The distribution of blood flow, oxygen consumption, and work output among the respiratory muscles during unobstructed hyperventilation. J Clin Invest 1977;59:43–50.[ISI][Medline]
- Magder S, Lockhat D, Luo BJ, et al. Respiratory muscle and organ blood flow with inspiratory elastic loading and shock. J Appl Physiol 1985;58:1148–56.[Abstract/Free Full Text]
- Robertson CH Jr., Foster GH, Johnson RL. The relationship of respiratory failure to the oxygen consumption of, lactate production by, and distribution of blood flow among respiratory muscles during increasing inspiratory resistance. J Clin Invest 1977;59:31–42.[ISI][Medline]
- Pang LM, Kim YJ, Bazzy AR. Blood flow to respiratory muscles and major organs during inspiratory flow resistive loads. J Appl Physiol 1993;74:428–34.[Abstract/Free Full Text]
- Viires N, Aubier SM, Roussos C. Regional blood flow distribution in dog during induced hypotension and low cardiac output: spontaneous breathing versus artificial ventilation. J Clin Invest 1983;72:935–47.[Medline]
- Hussain SNA, Roussos C. Distribution of respiratory muscle and organ blood flow during endotoxic shock in dogs. J Appl Physiol 1985;59:1802–8.[Abstract/Free Full Text]
- Supinski G, DiMarco A, Ketai L, et al. Reversibility of diaphragm fatigue by mechanical hyperperfusion. Am Rev Respir Dis 1988;138:604–9.[ISI][Medline]
- Supinski GS, DiMarco AF, Gonzalez J, et al. Effect of norepinephrine on diaphragm contractility and blood flow. J Appl Physiol 1990;69:2019–28.[Abstract/Free Full Text]
- Brochard L, Harf A, Lorino H, et al. Inspiratory pressure support prevents diaphragmatic fatigue during weaning from mechanical ventilation. Am Rev Respir Dis 1989;139:513–21.[ISI][Medline]
- Sassoon CS, Light RW, Lodia R, et al. Pressure–time product during continuous positive airway pressure, pressure support ventilation, and T-piece during weaning from mechanical ventilation. Am Rev Respir Dis 1991;143:469–75.[ISI][Medline]
- Kowallik P, Schulz R, Guth BD, et al. Measurement of regional myocardial blood flow with multiple colored microspheres. Circulation 1991;83:974–82.[Abstract/Free Full Text]
- Schosser R, Arfors K-E, Messmer K. MIC-II-A program for the determination of cardiac output, arterio-venous shunt and regional blood flow using the radioactive microsphere method. Comput Programs Biomed 1979;9:19–38.[ISI][Medline]
- Bark H, Supinski GS, Lamanna JC, et al. Relationship of changes in diaphragmatic muscle blood flow to muscle contractile activity. J Appl Physiol 1987;62:291–9.[Abstract/Free Full Text]
- Buchler B, Magder S, Katsardis H, et al. Effects of pleural pressure and abdominal pressure on diaphragmatic blood flow. J Appl Physiol 1985;58:691–7.[Abstract/Free Full Text]
- Sheriff DD, Bibber RV. Flow-generating capability of the isolated skeletal muscle pump. Am J Physiol 1998;274:H1502–8.
- Laughlin MH, Joyner M. Closer to the edge: contractions, pressures, waterfalls and blood flow to contracting skeletal muscle. J Appl Physiol 2002;94:3–5.
- Giannouli E, Webster K, Roberts D, et al. Response of ventilator-dependent patients to different levels of pressure support and proportional assist. Am J Respir Crit Care Med 1999;159:1716–25.[Abstract/Free Full Text]
- Leung P, Jubran A, Tobin MJ. Comparison of assisted ventilator modes on triggering, patient effort, and dyspnea. Am J Respir Crit Care Med 1997;155:1940–8.[Abstract]
- Vassilakopoulos T, Petrof BJ. Ventilator-induced diaphragmatic dysfunction. Am J Respir Crit Care Med 2004;169:336–41.[Free Full Text]
- Manderly JL, Tesarek JE, Sifford LJ, et al. Respiratory measurements of unsedated small laboratory mammals using nonbreathing valves. Lab Anim Sci 1979;29:323–9.[ISI][Medline]
- Frank DW. Physiologic data of laboratory animals. In: Melby EC Jr., Altman NH, eds. Handbook of laboratory animal science volume III. Boca Raton: CRC Press, Inc., 1976:23–64.
- Kondo T, Miyairi A, Kobayashi I, et al. Diaphragmatic electrical activity during controlled and assisted mechanical ventilation in conscious human subjects. Intern Med 1994;33:667–72.[Medline]
Top
Abstract
Introduction
