| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2002;73:1204-1209
© 2002 The Society of Thoracic Surgeons
a Department of Anesthesiology Hospital Privado de Comunidad, Mar del Plata, Argentina
b Department of Surgery Hospital Privado de Comunidad, Mar del Plata, Argentina
c Intensive Care Medicine, Hospital Privado de Comunidad, Mar del Plata, Argentina
d Department of Anesthesiology, University Hospital Hamburg-Eppendorf, Hamburg, Germany
Accepted for publication December 12, 2001.
* Address reprint requests to Dr Tusman, Department of Anesthesiology, Hospital Privado de Comunidad, Cordoba 4545, 7600 Mar del Plata, Argentina
e-mail: gtusman{at}hotmail.com
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentReferences |
|---|
Methods. Ten patients undergoing open lobectomies were studied at three time points: (1) during two-lung ventilation (TLV), (2) during OLV before, and (3) after ARS. The ARS maneuver was done by increasing peak inspiratory pressure to 40 cm H2O, together with a positive end-expiratory pressure (PEEP) of 20 cm H2O for 10 respiratory cycles. After the maneuver, ventilation parameters were returned to the settings before intervention.
Results. During OLV, PaO2 was statistically lower before the recruitment (data as median, first, and third quartile, 217 [range 134 to 325] mm Hg) compared with OLV afterwards (470 [range 396 to 525] mm Hg) and with TLV (515 [range 442 to 532] mm Hg). After ARS, PaO2 values during OLV were similar to those during TLV. During OLV, the degree of pulmonary collapse in the nondependent lung, the hemodynamic status, and the ventilation parameters were similar before and after ARS.
Conclusions. Alveolar recruitment of the dependent lung augments PaO2 values during one-lung ventilation.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
Arterial oxygenation deteriorates during ventilation of one lung (OLV). It has been estimated that shunt values during this procedure can range from 20% to 40% [4], and it was postulated that intrapulmonary shunt originates from zones of total alveolar collapse and from zones of relative hypoventilation in both the dependent and the nondependent lung.
Many treatments have been proposed to overcome hypoxemia during OLV. Some of the proposed measures are the use of positive end-expiratory pressure (PEEP) and vasodilators in the dependent lung or treating the nondependent lung with continuous positive airway pressure (CPAP), vasoconstrictors, mechanical blockage of the pulmonary artery branch, or intermittent two lung ventilation (TLV) [5]. However, such therapeutic maneuvers are invasive, time-consuming, and may interfere with the work of the surgeon.
In patients undergoing general anesthesia, lung recruitment maneuvers proved to be easy to perform and effective in reverting alveolar collapse, hypoxemia, and decreased compliance [6, 7]. In previous publications, our group has demonstrated the beneficial effect of an alveolar recruitment strategy (ARS) on arterial oxygenation and respiratory compliance in anesthetized patients undergoing nonthoracic surgery in the supine position [7].
We hypothesized that recruiting atelectatic zones of the dependent lung and applying sufficient levels of PEEP should attenuate the decrease in PaO2 during OLV. The main goal of this study was to evaluate the impact of the alveolar recruitment strategy on gas exchange during OLV.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
Anesthesia
Upon arrival at the operating room, patients were premedicated with intravenous midazolam 0.03 to 0.04 mg/kg. Before induction of anesthesia, a thoracic epidural catheter was placed at the T2 to T4 intervertebral space. After induction of anesthesia, a total volume of 0.1 mL/kg of bupivacaine 0.5% without epinephrine plus phetidine 1 mg/kg was administered into the epidural catheter. A Teflon catheter (18 G) was inserted in the radial artery of the nondominant hand for invasive arterial pressure monitoring and blood gas analysis.
After 3 minutes of preoxygenation, anesthesia was induced with fentanyl 5 µg/kg, thiopental 3 mg/kg, and vecuronium 0.08 mg/kg. The trachea and the left bronchus were intubated with a left double-lumen tube (DLT) of the appropriate size (Broncho-Cath; Mallinckrodt Laboratories, Athlone, Ireland). Correct DLT position was corroborated using fiber optic bronchoscopy. To determine possible air leakage, the capnograph’s side stream sensor was placed consecutively into each lumen of the DLT while the other one was ventilated. The correct position of the DLT was confirmed bronchoscopically again after positioning the patient in lateral position. On-line dynamic pressure/volume loops were used to monitor the correct position of DLT during surgery. At OLV time, the lumen of the nonventilated side was left open to atmosphere.
Before the start of epidural anesthesia, infusing 7 mL/kg of colloid (Hemacell) expanded intravascular volume. Patients with intraoperative blood losses of more than 500 mL were excluded from the study. Losses of blood up to this amount were compensated by infusions of colloid solutions, with each volume of blood being replaced by an equal amount of these solutions. After the institutional protocol for thoracic surgery, all patients received 2 to 3 µg/kg/min of dopamine, and fluid balance was closely monitored to maintain the normovolemic state and avoid fluid overload.
Anesthesia was maintained with Isofluorane 0.5 to 0.6 MAC, with vecuronium boluses of 0.015 mg/kg and epidural lignocaine 1% boluses of 5 mL as clinically necessary.
A servo 900 C (Siemens Elema, Solna, Sweden) was used for mechanical ventilation, initially set as follows: volume control mode, VT of 8 mL/kg of ideal body weight, square flow waveform; respiratory rate of 10 to 14 breaths per minute; inspiratory to expiratory ratio of 1:2 with no inspiratory pause, and PEEP of 5 cm H2O. Before the start of the protocol, no volume expansive maneuver to reaerate the lung was done in any patient. Inspired oxygen fraction (FIO2) was kept 100% throughout the study. During OLV, tidal volumes were reduced to 6 mL/kg of ideal body weight and respiratory rates were increased to values between 14 and 18 breaths per minute to avoid peak inspiratory pressure higher than 30 cm H2O. Before obtaining the results of the blood gas analysis, ventilation was adjusted according to the end-tidal of CO2 (etCO2); later on, ventilation was adjusted to normalize PaCO2.
Monitoring and variables
Invasive arterial pressure, heart rate, esophageal temperature, central venous pressure (CVP), pulse oxymetry (SpO2) and electrocardiogram were monitored with the Cardiocap II monitor (Datex Instrument, Corp, Helsinki, Finland). A Capnomac Ultima monitor (Datex Instrument, Corp) was used to measure the following ventilation parameters and gas concentrations: peak inspiratory pressure (Pip), PEEP, tidal volume (TV), respiratory rate, expired minute volume, and etCO2.
During two-lung ventilation (TLV) intrinsic PEEP was measured in the lateral position by pressing the expiratory pause button for 5 seconds and reading the airway pressure value off the ventilator’s display.
Protocol
All measurements were carried out with the patient in lateral position. Arterial oxygenation, respiratory, and hemodynamic parameters were recorded at three points. (1) During TLV, 15 minutes after placing the patient in the lateral position with the chest still closed. (2) During OLV before applying the recruitment strategy. The arterial blood sample was obtained 15 minutes after clipping the lobar artery. (3) During OLV, 15 minutes after the recruitment maneuver.
The recruitment strategy was carried out in both lungs simultaneously by temporarily releasing the clamp from the nondependent lumen of the DLT. The aim of the increase in airway pressure was twofold. (1) Following surgery’s protocol for lobectomies in our hospital, a volume expansive ventilatory maneuver was done after lobectomy to evaluate the bronchial sutures of the nondependent lung. (2) This maneuver was used to expand the dependent lung too, restoring its FRC. Thereafter, OLV of the now expanded dependent lung was reestablished, possible bleeding was controlled, and pleural drainage was in place. A last blood sample was obtained 15 minutes later with the chest still open.
Recruitment maneuver
Figure 1
shows the alveolar recruitment strategy as used in this study. Before performing the strategy, the ventilator was switched to pressure control ventilation, adjusting the level of pressure to obtain the same tidal volume that the patient was receiving during volume control ventilation. With these parameters, ventilation was allowed to equilibrate for 3 minutes. Then, the ARS was performed as described previously [7]. The critical alveolar opening pressure is assumed to be around 40 cm H2O in the healthy lung [6, 7]: (1) Inspiratory time was increased to 50%. (2) Respiratory frequency was set at 12 breaths per minute. (3) The critical alveolar opening pressure was reached by adjusting the peak inspiratory pressure. PIP and PEEP were increased stepwise from 30/10 to 35/15, and finally 40/20 cm H2O. Every level of pressure was maintained during 1 minute. PIP-PEEP difference was limited to 20 cm H2O to avoid high tidal volumes during the maneuver. (4) Finally, airway pressure was gradually decreased returning to the previous settings. PEEP was set above 5 cm H2O.
|
Because elevated intrathoracic pressures can diminish venous return and thus cause decreases in blood pressure, hemodynamic and ventilatory variables were monitored closely while performing the opening maneuver. If mean arterial pressure decreased by more than 15% from its initial value, the ARS was discontinued and 500 mL of crystalloids was administered. Once hemodynamic stability was reached, the ARS was tried again.
During surgery, hemoglobin saturation was kept above 90% at all times. If during OLV SpO2 fell below 90%, surgery was temporarily interrupted to allow ventilation of both lungs until oxygen saturation reached at least 97% (intermittent ventilation).
Blood specimens were processed within 5 minutes of extraction by a blood gas analyzer (ABL 510 Radiometer Medical, Copenhagen, Denmark). Values were corrected for body temperature.
Statistical analysis
Descriptive statistical analysis was performed for each variable during TLV and during OLV before and after the recruitment maneuver using INSTAT 2.0 (GraphPad Software, San Diego, CA). Due to the small sample size, comparison of variables between time points was carried out using one-way analysis of variance with a Kruskal-Wallis nonparametric test. Results are presented as median, first (25%), and third (75%) quartiles. A p value less than 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
|
|
The time period between the beginning of OLV and the vascular clipping ranged from 45 to 75 minutes. During OLV before clamping of the lobar artery (during the period of highest intrapulmonary shunt), patients 2 and 4 required six and four cycles of intermittent two-lung ventilation respectively because of oxipletismography values below 90%. After vascular clipping, there were no more episodes of hemoglobin desaturation and therefore no need for intermittent ventilation.
Intrinsic PEEP was less than 2 cm H2O in only 4 patients. No patient presented hemodynamic instability during the recruitment maneuver nor cardiorespiratory complications in the perioperative period.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
Arterial oxygenation during OLV before the recruitment maneuver
As explained in the Material and Methods section (point 2), arterial blood gases were not assessed before clipping the lobar artery. However, 2 patients required more than one cycle of intermittent ventilation for episodes of arterial desaturation. The effects of intermittent TLV on oxygen saturation were brief, lasting no longer than 5 minutes.
After vascular clipping, no more episodes of desaturation occurred, although all patients presented with a lower PaO2 during OLV compared with TLV. Not all patients responded with the same fall in PaO2 (See Fig 2). Using Benatar’s iso-shunt normogram [8], it can be determined that the shunt at this time of the surgery ranged from 15% to 30%. This variability of intrapulmonary shunt can be explained by differences in the degree of hypoxic pulmonary vasoconstriction (HPV) among individuals of the same species [9]. An individual’s HPV response correlates inversely with the degree of shunt; thus, the stronger the HPV, the lower the shunt.
One other possible explanation for the marked differences in PaO2 values may be the difference in the extent of the surgical resection. It can be expected that the larger the piece of lung parenchyma excised, the larger the decline in shunt after vascular clipping. However, no such correlation was found. Right lobectomies cause a more extensive compromise of oxygenation due to the higher amount of nonventilated lung tissue. In the present study, six lobectomies of the right and four of the left were performed (Table 1). Again, no relationship was found between arterial oxygenation and the lobe excised.
Arterial oxygenation during OLV after the recruitment maneuver
Arterial oxygenation during OLV was not statistically different from TLV after the recruitment maneuver. We argue that the alveolar recruitment also caused a recruitment of the pulmonary capillaries and thus caused a redistribution of blood flow to the dependent lung with a more equal matching of perfusion to ventilation [10]. Both an alveolar and a capillary recruitment are the preconditions for an effective exchange of the respiratory gases. Due to the optimized matching of ventilation to perfusion after a successful recruitment, the single dependent lung becomes as efficient with respect to gas exchange as the two nonrecruited lungs together.
According to Benatar’s diagram, the shunt fraction of our study population was around 10% during TLV and 12.5% during OLV after ARS. Following classic physiologic concepts, the nondependent lung is responsible for two-thirds of the total shunt (approximately 20%), whereas the dependent lung contributes an additional one-third or an absolute shunt fraction of about 10%. If we consider that the recruitment maneuver actively reduced the shunt within the dependent lung to a minimum, the total shunt would still remain around 20% due to the shunting within the nondependent lung. In our study, however, the total shunt fraction of only 12.5% during OLV after the recruitment contrasts with published data. A possible explanation for this increase in PaO2 beyond the expected values could be a systematic underestimation of the shunt in the dependent lung. This implies that nearly the whole impairment of oxygenation observed during OLV before ARS could be attributed to atelectasis of the dependent lung, and not to a shunting of blood within the upper lung in our patients. Thus, recruitment of the dependent lung should effectively revert to total shunt. We are aware that this concept clearly contradicts the classic description of shunt distribution during OLV. Following these classic thoughts, it would be difficult to explain why the ARS reduced the shunt fraction to approximately 10%, even if only one lung was ventilated.
A more pronounced response to HPV in our patient population could be another theoretical explanation. Several authors have confirmed the hypothesis that HPV is potentiated by repeated periods of hypoxia [11, 12], but other authors could not confirm these results [13, 14]. In addition, time has to be considered as an important factor influencing HPV. Carlsson and associates [14] demonstrated that HPV reaches its maximal affect within the first 15 minutes of a hypoxemic challenge, and that this effect does not increase any more after 60 minutes. On the other hand, the influence of lung inflation on the HPV response remains controversial. An increase in lung inflation is associated with a decrease in the magnitude of the reflex [9]. It seems unlikely that an increase in the response to HPV is the main reason for the increased oxygenation observed in our patients. Dopamine and Isofluorane are capable of diminishing the lung’s response to hypoxemia. However, in our study, these influencing factors were maintained constant during the entire study period.
One more explanation for the high PaO2 seen in our study could be that the oxygen remaining within the upper lung after its reaeration could have contributed to arterial oxygenation also during OLV. We consider this effect to be highly unlikely, because the degree of collapse observed in the nondependent lung during OLV was macroscopically similar before and after the maneuver. Also, in pure oxygen ventilation, absorption is completed within 5 to 7 minutes. Rothen and associates [15] have demonstrated that the rate of formation of reabsorption atelectasis after a recruitment maneuver depends primarily on the fraction of inspired oxygen (FiO2) used. If pure oxygen was used, atelectasis reappeared on CT scans within 5 minutes after lung recruitment and to the same extent as before.
Model limitation
Before the ligation, the lobar vessels at the moment of highest shunt oxygenation were not evaluated due to reasons inherent to the protocol (see Material and Methods, point 2). In view of the overall study design, there was no time to evaluate the effect of an alveolar recruitment on arterial oxygenation before vascular clipping. We considered it nonethical to prolong the period of OLV with its high risk of hypoxemia. For this reason, the effect of the recruitment strategy on gas exchange during periods of highest intrapulmonary shunting cannot be appreciated. Further investigations should address this interesting question.
To keep anesthesia time within acceptable limits, 15 minutes of equilibrium was used to evaluate oxygenation at each time point. Torda and associates [4] showed that intrapulmonary shunt reaches its maximum 15 to 20 minutes after the beginning of OLV. HPV reaches its maximum effect after 15 minutes of OLV. Carlsson and associates considered 15 minutes of steady-state ventilation enough to evaluate the body’s HPV response to hypoxemic stimuli [14].
Compared with the above mechanisms, changes in the alveolo-capillary surface area due to the collapse or aeration of lung tissue occur very quickly. Brismar and associates [1] used computer tomography to demonstrate that anesthesia-induced atelectasis appears within the first 5 minutes after induction of anesthesia and does not progress in magnitude over time. Neumann and associates [16], in a pig model of acute lung injury, studied the timing of lung collapse and recruitment by dynamic CT. The time constant for the formation of atelectasis was 0.86 ± 0.67 and 0.69 ± 0.54 seconds for recruitment. Böhm and associates [17] studied the pulmonary opening and collapsing process by means of on-line arterial blood gases. They observed that the changes in PaO2 are seen within 1 minute. Thus, the results of the above-mentioned studies support the concept that 15 minutes of equilibrium should be enough to evaluate the effect of changes in alveolar-capillary surface area on arterial oxygenation.
Conclusions
In patients undergoing lobectomies, the alveolar recruitment strategy proved successful in augmenting arterial oxygenation during OLV. The recruitment maneuver could optimize the matching of ventilation and perfusion within the dependent lung. Further research is needed to clarify the precise mechanisms involved in alveolar and capillary recruitment and their potential benefit for patients who desaturate severely during OLV at the moment of highest shunt.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. T. Lytle and D. R. Brown Appropriate Ventilatory Settings for Thoracic Surgery: Intraoperative and Postoperative Seminars in Cardiothoracic and Vascular Anesthesia, June 1, 2008; 12(2): 97 - 108. [Abstract] [PDF]
|
|
|
|
 |
|
 S. Maisch, H. Reissmann, B. Fuellekrug, D. Weismann, T. Rutkowski, G. Tusman, and S. H. Bohm Compliance and Dead Space Fraction Indicate an Optimal Level of Positive End-Expiratory Pressure After Recruitment in Anesthetized Patients Anesth. Analg., January 1, 2008; 106(1): 175 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Tusman, S. H. Bohm, F. S. Sipmann, and S. Maisch Lung Recruitment Improves the Efficiency of Ventilation and Gas Exchange During One-Lung Ventilation Anesthesia Anesth. Analg., June 1, 2004; 98(6): 1604 - 1609. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Magnusson and D. R. Spahn New concepts of atelectasis during general anaesthesia Br. J. Anaesth., July 1, 2003; 91(1): 61 - 72. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Senturk and M. Tugrul Alveolar recruitment during one-lung ventilation--really "one" lung? Ann. Thorac. Surg., February 1, 2003; 75(2): 635 - 635. [Full Text] [PDF]
|
|
|
|
|
|
G. Tusman, F. Melkun, D. Staltari, C. Quinzio, C. Nador, E. Turchetto, and S. H. Bohm Alveolar recruitment during one-lung ventilation--really "one" lung?: Reply Ann. Thorac. Surg., February 1, 2003; 75(2): 635 - 636. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
