Lung collapse is common after anesthesia and surgery as well as in the intensive care unit (1,2). There is no consensus as to whether this condition should be actively treated, but a lung recruitment maneuver is probably indicated if severe hypoxemia occurs because of a large shunt. Although all common lung recruitment maneuvers increase intrathoracic pressure and reduce venous return, and also might increase right ventricular outflow impedance, the maneuvers are usually well tolerated hemodynamically (3,4). However, profound circulatory depression can occur, particularly during hypovolemia (5–9).
We recently proposed an alternative method to recruit collapsed lung tissue, the selective lung recruitment maneuver (S-LRM), in which recruitment is performed by selective inflation of the affected lung region via a bronchial balloon catheter (10). This method had no circulatory side effects in an experimental porcine model with isolated right lower lobar collapse, and was as effective as a global recruitment maneuver in improving oxygenation and lung volume (10). However, the animals were normovolemic and their circulation was stable. We hypothesized that the S-LRM has minimal circulatory side effects in unstable circulatory conditions as well. Therefore, the aim of the present experimental study was to examine the circulatory and respiratory effects of S-LRM during significant hypovolemia with compromised hemodynamics.
The study was approved by the national animal ethics committee and followed the National Institutes of Health principles of laboratory care. Eight pigs, 25–30 kg, were premedicated with apazerone 80 mg IM and midazolam 10 mg IM. Anesthesia was induced by fentanyl 5 μg/kg IV and ketamine 2 mg/kg IV. A tracheotomy was performed and the trachea was intubated with a Portex 9.0 ID tube (Smiths Medical, London, UK). The lungs were ventilated by a Servo 900 C ventilator (Siemens-Elema, Solna, Sweden) with volume control, tidal volume 8 mL/kg, positive end-expiratory pressure (PEEP) 10 cm H2O and fraction of inspired oxygen, Fio2 1.0. Inspiratory time was set to 35% and end-inspiratory pause time was 10%, with the rate adjusted to achieve an arterial pH of approximately 7.4. After a bolus dose of propofol (1 mg/kg), anesthesia was maintained with ketamine 10 mg · kg−1 · h−1, fentanyl 5 μg · kg−1 · h−1, propofol 2 mg · kg−1 · h−1, and pancuronium 0.25 mg · kg−1 · h−1. Ten milliliter per kilogram of lactated Ringer's solution was infused during the instrumentation phase of the experiment. In three animals, 2–5 mL/kg of a starch solution (Voluven, Fresenius Kabi, Sweden) was administered to stabilize the circulation before the main experiment was initiated.
Catheters were placed in the right and left carotid arteries and the right internal jugular vein for sampling of blood gases, monitoring of intravascular pressures, and obtaining pulse contour cardiac output (PCCO). A pulmonary artery catheter (Swan-Ganz CCO mbo CCO/Svo2 7.5F, Edwards Lifescience, Irvine, CA) was placed via the right external jugular vein to monitor pulmonary artery pressure and central venous pressure (CVP). A suprapubic urinary catheter was inserted for monitoring diuresis.
PCCO was obtained through a catheter (Pulsiocath, 4F, 16 cm Pulsion Medical Systems, Munich, Germany) placed in the left carotid artery and connected to the PiCCO monitor (Pulsion Medical Systems, Munich, Germany). The PCCO measurement was calibrated using the transpulmonary arterial thermodilution technique, using cold saline injectate 10 mL thrice immediately after induction of anesthesia and before each measurement sequence. In addition, values for intrathoracic blood volume, pulse pressure variation (PPV), and stroke volume variation (SVV) were obtained from the PiCCO device. During the entire study period, values for cardiac output, mean arterial blood pressure (MAP), and heart rate were collected at 1 s intervals using an interface and a computer with PiCCOwin Plus software (Pulsion Medical Systems, Munich, Germany). Quasistatic compliance of the respiratory system (Crs) was calculated as tidal volume/ (end-inspiratory plateau pressure − end-expiratory pressure). End-expiratory pressure was obtained after a 5-s end-expiratory hold.
Blood gases were sampled from the right carotid artery and the pulmonary artery and analyzed by an ABL 710 (Radiometer, Copenhagen, Denmark). Venous admixture (QS/QT) was calculated using the standard formula: QS/QT = [(Cc′O2 − CaO2)/ (Cc′O2 − CvO2)], where CcO2 is the end-capillary blood oxygen content estimated using the alveolar air equation, and CaO2 and CvO2 are the arterial and mixed venous oxygen content, respectively (10). Because a Fio2 of 1.0 was used, the venous admixture reported is a very close estimate of the intrapulmonary shunt (11).
The outline of the experiment is found in Figure 1. A bronchial blocker (Cook C-AEBS-9.0-65-SPH, Cook Critical Care, Bloomington, IN) was inserted via the endotracheal tube in the right lower lobe bronchus (about 50 cm distal to the endotracheal tube opening) under fiberoptic guidance (Olympus BF-3C40, Tokyo, Japan). To standardize the lung volume history, a global lung recruitment maneuver was performed using pressure-controlled ventilation with peak pressure of 40 cm H2O, PEEP 10 cm H2O, inspiratory:expiratory ratio 1:1, and a respiratory rate of 6 breaths/min over 2 min. Arterial blood gases were measured and the respiratory rate adjusted. No further ventilator setting changes were made during the experiment. A new set of blood gases (mixed venous and arterial) were then sampled. Under fiberoptic visualization the balloon of the bronchial blocker was inflated and the gas in the isolated right lower lobe was exsufflated and measured (the exsufflated gas volume is recorded as “lobe volume”). The lobe was selectively lavaged (with a lobe volume of 37°C normal saline) using a syringe 15×. Frothing of the lavaged fluid was not seen in any of animals at the 15th lavage. The volume of injected and removed fluid was measured. The inner lumen of the bronchial blocker was connected to 10 cm H2O negative pressure and the inspiratory and expiratory tidal volumes were checked to confirm the seal between the bronchial blocker cuff and the bronchial wall. Blood gases were obtained to confirm lobar collapse and to establish baseline values.
The animals were then subjected to S-LRM. An airway pressure of 40 cm H2O was applied for 30 s through the inner lumen of the bronchial blocker. A calibrated water manometer connected to an oxygen gas source ensured correct pressure. During the maneuver, ventilation of the noncollapsed lung was uninterrupted. Three S-LRMs were performed: one in normovolemia, one after removal of 20% of the blood volume by venesection, and finally one after removal of 40% of the blood volume. Blood volume (mL) was estimated as 179 × body weight (kg)(0.73) (12). After each venesection an interval of 10 min was allowed before a new experimental sequence was started, and after each S-LRM lobar collapse was reestablished by exsufflation of the gas from the isolated lobe and application of 10 cm H2O negative pressure.
Cardiac output, MAP, heart rate, mean pulmonary artery pressure, CVP, PPV, and SVV were registered 3 min before the S-LRM, during the final 5 s of the S-LRM, and 3 min after the S-LRM. Blood gases were sampled, and tidal volume and airway pressures were obtained for calculation of Crs 3 min before and 3 min after the S-LRM. In addition, intrathoracic blood volume was registered 3 min before the S-LRM.
Results are presented as median and 25th and 75th percentiles, if not otherwise indicated. The Wilcoxon's ranked sum test (Statview, SAS Institute, Cary, NC) was used for all comparisons. A maximum of three analyses was performed using the same value. We present the exact P values so that the results can be more easily assessed, but we did not correct for multiple comparisons, because in this study we believed that it was more important to decrease the risk of Type II errors than that of Type I errors. We considered P < 0.05 as significant.
The lobar collapse induced a significant change in Pao2, venous admixture, and Crs (initial data not shown). These variables were normalized by S-LRM at all volemic levels (Table 1). Before S-LRM there were well-defined differences in Pao2 (P = 0.012) and venous admixture (P = 0.012), and a minor difference in Crs (P = 0.012) between the different volemic levels. However, 3 min after S-LRM no significant differences (P = 0.12–0.48) were found between the volemic levels in any of the variables.
Removal of 20% and 40% of the blood volume caused significant decreases in MAP, cardiac output, intrathoracic blood volume, and CVP, and increases in heart rate, PPV, and SVV (Table 2). There was a minor decrease in heart rate after the S-LRM in 20% hypovolemia and a minimal decrease in MAP in the end of the S-LRM at 40% hypovolemia. No other significant changes caused by the S-LRM in MAP, cardiac output, or heart rate were found at any of the volemic levels (Table 2). A minor decrease in mean pulmonary artery pressure was registered after the S-LRM at normovolemia and 20% hypovolemia (Table 2). Although CVP changed significantly with S-LRM, this change was <1 mm Hg (median and mean) (Table 2). PPV decreased by a median of 2% at the end of and after S-LRM at normovolemia, whereas SVV was unaffected by the S-LRMs (Table 2).
This study confirms that a S-LRM effectively reduces venous admixture and improves respiratory mechanics in an experimental porcine model of dense lobar collapse, and shows that this is achieved without markedly affecting circulation negatively, even with severe hypovolemia.
In normovolemic pigs with stable circulation and isolated lobar collapse, S-LRM improved oxygenation and lung volume as effectively as a general lung recruitment maneuver (G-LRM), but had no circulatory side effects (10). In the same animals, a G-LRM reduced cardiac output and MAP by approximately 50% (10). Furthermore, in a previous study examining the effect of G-LRM during normovolemia, hypovolemia and hypervolemia in lung-injured pigs, we found that a 30 s G-LRM using 40 cm H2O, after removal of 15% of the blood volume, decreased cardiac output from 2.6 to 0.2 L/min and MAP from 58 to 20 mm Hg (9). In contrast, in the present study using S-LRM after removal of 40% of the blood volume (reducing intrathoracic blood volume by 33%, cardiac output by 26%, and MAP by 44%), only a minor decrease in MAP and no change in cardiac output occurred, indicating that hemodynamically the animals tolerated this maneuver well.
It can be argued that the PiCCO device is inaccurate when there are quick changes in hemodynamics (13). However, in previous studies examining hemodynamics during PEEP ventilation and recruitment maneuvers, we found that PiCCO clearly indicated changes in cardiac output, even with brief interventions (7,9). Therefore, we are confident that cardiac output did not decrease because of the S-LRM in this study, but because of the small number of animals studied we cannot exclude that cardiac output could be affected by the maneuver. A prolonged S-LRM might theoretically over-distend the newly recruited lung regions, compressing the surrounding lung and thus compromising circulation.
The S-LRM effectively regained oxygenation, decreased venous admixture, and normalized compliance at all volemic levels, confirming our previous results. In addition, we found a small but significant decrease in mean pulmonary artery pressure after the S-LRM. This could have been because of the fact that improved oxygenation after the maneuver reduced a possible hypoxic vasoconstriction in the previously collapsed lung regions, or to the fact that vessels in a noncollapsed lung have lower resistance than in a collapsed lung (14). Our aim was not to evaluate the cardiac filling or arterial pressure variations during the S-LRM. These measures were obtained to verify that the removal of blood did change the intravascular volume. However, the changes in CVP, SVV, and PPV because of the S-LRM were nonexistent or marginal.
We induced lung collapse in the lower dorsal part of the lungs, which is the most common location both after anesthesia and surgery, as well as in intensive care patients (1,2). In addition, we placed the animals in supine position to mimic the clinical situation. As expected, Pao2 decreased with hypovolemia, because of the reduced cardiac output. If oxygen consumption and hemoglobin concentration are unchanged, mixed venous saturation (Svo2) is lower when cardiac output is reduced, and a lower Svo2 exaggerates the effect of a right-to-left intrapulmonary or cardiac shunt on arterial oxygenation (11). For example, if Svo2 is 80% and the fixed shunt is 50%, arterial oxygen saturation would be 90%, but if Svo2 is 50% with an equal shunt, arterial oxygen saturation would only be 75%. However, in our study, although the amount of collapsed lung tissue was similar at all volemic levels, not only did Pao2 decrease, but venous admixture (which is almost similar to the shunt with Fio2 1.0) also increased in the case of hypovolemia. This could be explained by the fact that in the low-pressure, gravity-dependent pulmonary circulation, when cardiac output was reduced, blood flow in the anterior, well-ventilated, and oxygenated lung regions became reduced. Thus, the collapsed lung, which was located in the posterior, caudal lung region, received a larger proportion of cardiac output. This would increase shunt fraction, despite the fact that anatomically the amount of collapsed lung was unchanged. In addition, we used 10 cm H2O of PEEP, which probably enhanced this effect by increasing the airway pressure and thus increasing the amount of “West zone I” condition in the noncollapsed lung (11).
Our results reinforce that in hypovolemia combined with hypoxemia because of dependent lung collapse, intravascular volume resuscitation is of extreme importance for improving arterial oxygenation. First, the increased Svo2 produced by the increased cardiac output will reduce the shunt effect. Second, the redistribution of pulmonary blood flow to well-ventilated lung regions will reduce the actual shunt fraction. In addition, our study suggests that S-LRM might be a safer alternative to a global recruitment maneuver in this situation. However, this has to be studied in patients.
When the lobe was collapsed, Crs was somewhat lower in hypovolemia than in normovolemia. The reason for this is unclear. We suggest that this may be related to differences in expiratory lung volume and consequent ventilation at different parts of the pressure–volume curve.
Our study has several limitations. First, it was performed in healthy young animals. Second, the study only examined an isolated lung collapse, and in conditions where the lung collapse is diffuse the method is, if not modified, not suitable. Third, because we did not find any significant hemodynamic side effects caused by the maneuver, other than a slight decrease in MAP, it could be that the number of animals studied was inadequate. However, we believe that the results indicate that the probability of serious hemodynamic side effects being caused by the S-LRM is small.
In conclusion, this study shows that in severely hypovolemic pigs with isolated lobar collapse, S-LRM effectively reduces venous admixture, increases oxygenation, and improves lung mechanics, without markedly affecting circulation negatively.
1. Brismar B, Hedenstierna G, Lundquist H, Strandberg A, Svensson L, Tokics L. Pulmonary densities during anesthesia with muscular relaxation—a proposal of atelectasis. Anesthesiology 1985;62:422–8
2. Gattinoni L, Pelosi P, Vitale G, Pesenti A, D'Andrea L, Mascheroni D. Body position changes redistribute lung computed-tomographic density in patients with acute respiratory failure. Anesthesiology 1991;74:15–23
3. Dyhr T, Nygard E, Laursen N, Larsson A. Both lung recruitment maneuver and PEEP are needed to increase oxygenation and lung volume after cardiac surgery. Acta Anaesthesiol Scand 2004;48:187–97
4. Borges JB, Okamoto VN, Matos GF, Caramez MP, Arantes PR, Barros F, Souza CE, Victorino JA, Kacmarek RM, Barbas CS, Carvalho CR, Amato MB. Reversibility of lung collapse and hypoxemia in early acute respiratory distress syndrome. Am J Respir Crit Care Med 2006;174:268–78
5. Lim SC, Adams AB, Simonson DA, Dries DJ, Broccard AF, Hotchkiss JR, Marini JJ. Transient hemodynamic effects of recruitment maneuvers in three experimental models of acute lung injury. Crit Care Med 2004;32:2378–84
6. Nunes S, Rothen HU, Brander L, Takala J, Jakob SM. Changes in splanchnic circulation during an alveolar recruitment maneuver in healthy porcine lungs. Anesth Analg 2004;98:1432–8
7. Nielsen J, Østergaard M, Kjærgaard J, Tingleff J, Berthelsen PG, Nygård E, Larsson A. Lung recruitment maneuver depresses central hemodynamics in patients following cardiac surgery. Intensive Care Med 2005;31:1189–94
8. Odenstedt H, Åneman A, Karason S, Stenquist O, Lundin S. Acute hemodynamic changes during lung recruitment in lavage and endotoxin-induced ALI. Intensive Care Med 2005;31:112–20
9. Nielsen J, Nilsson M, Fredén F, Hultman J, Alström U, Kjærgaard J, Hedenstierna G, Larsson A. Central hemodynamics during lung recruitment maneuvers at hypovolemia, normovolemia and hypervolemia. A study by echocardiography and continuous pulmonary artery flow measurements in lung-injured pigs. Intensive Care Med 2006;32:585–94
10. Hansen LK, Sloth E, Nielsen J, Koefoed-Nielsen J, Lambert P, Lunde S, Larsson A. Selective recruitment maneuvers for lobar atelectasis: effects on lung function and central hemodynamics: an experimental study in pigs. Anesth Analg 2006;102:1504–10
11. Lumb AB. Nunn's applied respiratory physiology. 6th ed. Philadelphia: Elsevier, 2005
12. von Englehardt W. Swine cardiovascular physiology: a review. In: Bustad LK, Mclellan RO, eds. Swine in biomedical research. Seattle: Frayn, 1966:307–29
13. Gunn SR, Kim HK, Harrigan PW, Pinsky MR. Ability of pulse contour and esophageal Doppler to estimate rapid changes in stroke volume. Intensive Care Med 2006;32:1537–46
© 2007 International Anesthesia Research Society
14. Wittenberger JL, Mcgregor M, Berglund E, Borst HG. Influence of state of the lung on pulmonary vascular resistance. J Appl Physiol 1960;15:878–82