Background: Atelectasis, an important cause of impaired gas exchange during general anesthesia, may be eliminated by a vital capacity maneuver. However, it is not clear whether such a maneuver will have a sustained effect. The aim of this study was to determine the impact of gas composition on reappearance of atelectasis and impairment of gas exchange after a vital capacity maneuver.

Methods: A consecutive sample of 12 adults with healthy lungs who were scheduled for elective surgery were studied. Thirty minutes after induction of anesthesia with fentanyl and propofol, the lungs were hyperinflated manually up to an airway pressure of 40 cmH₂ O. FI sub O₂ was either kept at 0.4 (group 1, n = 6) or changed to 1.0 (group 2, n = 6) during the recruitment maneuver. Atelectasis was assessed by computed tomography. The amount of dense areas was measured at end-expiration in a transverse plane at the base of the lungs. The ventilation-perfusion distributions (V with dot A/Q with dot) were estimated with the multiple inert gas elimination technique. The static compliance of the total respiratory system (C_rs) was measured with the flow interruption technique.

Results: In group 1 (FIO₂ = 0.4), the recruitment maneuver virtually eliminated atelectasis for at least 40 min, reduced shunt (V with dot A/Q with dot < 0.005), and increased at the same time the relative perfusion to poorly ventilated lung units (0.005 < V with dot A/Q with dot < 0.1; mean values are given). The arterial oxygen tension (Pa_O2) increased from 137 mmHg (18.3 kPa) to 163 mmHg (21.7 kPa; before and 40 min after recruitment, respectively; P = 0.028). In contrast to these findings, atelectasis recurred within 5 min after recruitment in group 2 (FI_O2 = 1.0). Comparing the values before and 40 min after recruitment, all parameters of V with dot A/Q with dot were unchanged. In both groups, C_rs increased from 57.1/55.0 ml *symbol* cmH₂ O sup -1 (group 1/group 2) before to 70.1/67.4 ml *symbol* cmH₂ O sup -1 after the recruitment maneuver. C_rs showed as low decrease thereafter (40 min after recruitment: 61.4/60.0 ml *symbol* cmH₂ O sup -1), with no difference between the two groups.

Conclusions: The composition of inspiratory gas plays an important role in the recurrence of collapse of previously reexpanded atelectatic lung tissue during general anesthesia in patients with healthy lungs. The reason for the instability of these lung units remains to be established. The change in the amount of atelectasis and shunt appears to be independent of the change in the compliance of the respiratory system.

Key words: Anesthesia, general. Lung: atelectasis; compliance; gas exchange. Measurement techniques: computed tomography; multiple inert gas elimination. Ventilation, mechanical: ventilation perfusion ratio.

GENERAL anesthesia regularly impairs gas exchange, and despite preventive measures such as an increase of the inspired oxygen fraction, this often results in a decreased oxygenation of blood. [1,2] The formation of atelectasis with subsequent pulmonary shunt has been shown to be an important factor for the derangement of gas exchange. [3-5] However, atelectasis may be reexpanded and virtually eliminated by hyperinflation of the lungs. [6-8] We demonstrated that such a recruitment maneuver may result in a sustained effect on atelectasis, a decrease of pulmonary shunt, and a slightly improved oxygenation for at least 40 min. [9] In that investigation, air was used for the recruitment maneuver, the lungs of all patients were ventilated with 40% Oxygen₂ in nitrogen, and there was an increase in the perfusion to units with low ventilation-perfusion ratios (V with dot A/Q with dot) after the recruitment.

It has been shown that lung units with low V with dot A/Q with dot may be prone to collapse, particularly if high inspiratory concentrations of oxygen or other gases with a high solubility in blood are used. [10] On the other hand, a single deep breath may result in release of surfactant, [11,12] thus contributing to an improved alveolar stability and preventing lung collapse. We therefore tested the effect of low or high inspired oxygen concentration (FI_O2 0.4 or 1.0, balance nitrogen) on the reappearance of atelectasis, gas exchange, and compliance of the respiratory system after a vital capacity maneuver during general anesthesia.

Twelve consecutive patients scheduled for neurosurgical procedures or general surgery were included in the study (Table 1). Another four patients refused to participate in the study, i.e., the refusal rate was 25%. [14] As judged by patient history and clinical examination, no patient had a cardiac or pulmonary disease. There were two smokers.

The study was approved by the Ethics Committee of the University Hospital of Uppsala, and informed consent was obtained from each patient. The estimation of the sample size (n = 12) was based on previous studies, [5,8] detect a difference of 50% in the amount of atelectasis, at a P value of 0.05 and a power of 90%, between the two groups 40 min after reexpansion of atelectasis. [15].

On request, the patients received 0.04-0.08 mg *symbol* kg sup -1 cetobemidon (Ketogan, an opioid) intramuscularly as premedication. Before anesthesia, 0.5 mg atropine was given intravenously. Anesthesia was induced with 1-2 micro gram *symbol* kg sup -1 fentanyl and 2 mg *symbol* kg sup -1 propofol intravenously, followed by a continuous infusion of 4-8 mg *symbol* kg sup -1 *symbol* h sup -1 propofol. During the induction, the lungs were manually ventilated via a face mask, using 100% Oxygen₂. To facilitate orotracheal intubation, the patients received 0.1 mg *symbol* kg sup -1 pancuronium, and additional doses of 1-2 mg were given when needed. The lungs were mechanically ventilated (Servo Ventilator 900C, Siemens) at a rate of 10 breaths/min with 40% Oxygen₂ in nitrogen during the following phase of the study (Figure 1). No positive end-expiratory pressure was used. The minute ventilation was adjusted to maintain an end-tidal carbon dioxide concentration of approximately 4% (carbon dioxide analyzer Eliza, Engstrom).

The patients were assigned randomly (sealed envelopes) to two different groups. In group I, FI_O2 was kept at 0.4, whereas in group 2, FI_O2 was switched to 1.0 during the recruitment maneuver (see below).

Three sets of measurements were performed (details are given below and in Figure 1): an estimation of atelectasis by computed tomography (CT) of the lungs, an estimation of the ventilation perfusion relationship (V with dot A/Q with dot) with the multiple inert gas elimination technique (MIGET), and an estimation of the compliance of the total respiratory system (C_rs) with the flow interruption technique. In the awake state, the measurements included CT of the lungs and the estimation of V with dot A/Q with dot. After 20 min of stable anesthesia, the CT scans and the V with dot A/Q with dot measurements were repeated, and the C_rs was measured (= "anesthesia, baseline"). This was followed by the recruitment maneuver (see below). The CT, V with dot A/Q with dot and C_rs measurements were repeated until 40 min after the recruitment (Figure 1). Finally, the patient was moved to the operating theater.

To reexpand atelectasis, a recruitment maneuver was performed as described previously. [8] Using a super syringe, the lungs were inflated with air three times up to an airway pressure (P_aw) of 30 cmH₂ O (Figure 1). For the final inflation to a P_aw of 40 cmH₂ O, either air (group 1) or 100% Oxygen₂ (group 2) was used, and in group 2, FI_O2 was changed to 1.0 on the ventilator for the succeeding study period. Each inflation was held for 15 s, and between each inflation, the lungs were ventilated for 1-2 min with the baseline settings of the ventilator. During the recruitment maneuver, the airway pressure was measured with a manometer (BOC, Ohmeda) attached to the endotracheal tube.

Atelectasis was studied by CT (Somatom plus, Siemens). The subjects were in the supine position with arms above the head. A frontal scout view covering the chest was obtained at end-expiration, awake, and after induction of anesthesia. For each subsequent analysis, a CT scan in the transverse plane was performed at end-expiration, 1 cm above the top of the right diaphragm. An additional scan at the level of the hilum and at the apex of the lungs was taken after induction of anesthesia, immediately after the recruitment maneuver and 40 min thereafter. The scan time was 1 s at 255 mAs and 137 kV. The slice thickness was 8 mm, and a matrix of 512 x 512 was used, resulting in a pixel (picture element) of approximately 1.5 x 1.5 min. The total x-ray exposure for each patient was 3mSv (Millisievert; total average exposition in Sweden 5 mSv per year).

To identify atelectasis, a magnified image (approximately 3x) was made of the dorsal portion of the CT scan of both the right and the left lungs. The dorsal border between the thoracic wall and the dense areas was drawn manually, whereas the ventral border between inflated lung tissue and atelectasis was identified by the region-of-interest (ROI) program. The atelectatic area, including all pixels with density values between -100 and +100 Hounsfield units (HU), was calculated by the computer. As an example, lung tissue with a density value of -100 HU represents a lung unit with 10% gas and 90% tissue. The amount of atelectasis was expressed in absolute values (cm²) and as percent of the total intrathoracic area (including the mediastinum). For details, see references 3 and 16.

To evaluate possible changes in regional lung volume in nondependent parts of the lungs, vessels that could be seen on consecutive scans were identified, and the angles between these vessels were measured. For this purpose, the CT scans were transferred to an image analysis software package (BioScan OPTIMAS, Edmonds, WA). This program allows the simultaneous analysis of several CT scans and the measurement of angles between lines (an example is given in Figure 2).

The ventilation-perfusion ratios (V with dot A/Q with dot) were assessed with the multiple inert gas elimination technique. [17] Isotonic saline with a mixture of six inert gases (sulphur hexafluoride, ethane, cyclopropane, enflurane, diethyl ether, and acetone) was infused continuously into a peripheral vein. Under steady-state conditions, arterial blood and mixed expired gas samples were collected in duplicate for subsequent analysis by gas chromatography (Hewlet Packard Gas Chromatograph 5880A and 5890). Oxygen uptake (V_O2) and carbon dioxide elimination (V_CO2) were estimated with the Douglas-bag technique, thus collecting expired gas and measuring the concentrations of mixed expired oxygen and carbon dioxide (Beckman OM-14 and Leybold-Heraeus). During anesthesia, only V_CO2 was measured directly, whereas V_O2 was calculated assuming the same respiratory quotient as measured awake. The cardiac output was estimated as 20*V_O2, assuming an arteriovenous oxygen difference of 50 ml *symbol* l sup -1 blood and expressing V_O2 in ml *symbol* min sup -1. The mixed venous inert gas concentrations were computed from the arterial and mixed expired values using mass balance principles. [18] By mathematical analysis of the inert gas data, each V with dot A/Q with dot distribution was recovered in a 50-compartment model, and the result with the best fit of data (smallest remaining sum of squares, RSS) of each pair of duplicate samples was used for further statistical analysis. Intrapulmonary shunt (Q_S /Q_T) was defined as the fraction of total blood flow perfusing lung units with V with dot A/Q with dot < 0.005, and low V with dot A/Q with dot was defined as the fraction of total blood perfusion to lung units with 0.005 < V with dot A/Q with dot < 0.1. logSDQ, the standard deviation of the logarithmic distribution of perfusion, was calculated as a measure of the dispersion of blood flow distribution. logSDV is the standard deviation of the logarithmic distribution of ventilation. RSS for all measurements was on the average 2.0 and exceeded 6.0 in only two measurements, thus fulfilling the criteria as established by Wagner and West. [19].

In two patients (one from each group), a thorough analysis of V with dot A/Q with dot was performed by so-called Monte Carlo simulation and by linear programming, as described previously. [20] Thus, the inert gas data were analyzed taking into account both the experimental error of inert gas measurement and the uncertainty, inherent to the determination of V with dot A/Q with dot distributions by the multiple inert gas elimination technique.

Blood gas measurements were performed by standard technique (ABL-2, Radiometer).

Estimations of static compliance of the respiratory system (C sub rs) were obtained using the technique of flow interruption. [21] Pressure and flow were measured in the ventilator on the inspiratory side and fed into a computer for on-line signal processing (Macintosh IIFx, with LabView 2, National Instruments). The mean value of two "inspiratory hold" maneuvers (2 s hold) was used for each point. C_rs was calculated as tidal volume (V_T) divided by end-inspiratory pressure minus end-expiratory pressure. Thus, C_rs reflects the elastic behaviour of the lungs and the chest wall over the tidal volume range.

Where not stated otherwise, mean values and standard deviations are presented. In addition, for major variables, 95% confidence intervals (95% CI) are given. [22] To compare variables at different times within groups, we used the Friedman two-way ANOVA and the Wilcoxon's signed-rank test. To compare data between the two group, the Mann-Whitney U test was used. Spearman's rank correlation coefficient was used to analyze relationships between variables. For all calculations, the SYSTAT computer software package (SYSTAT, Evanston, IL) was used.

The CT scans in the awake patients displayed no abnormalities. All patients experienced atelectasis after induction of anesthesia, with an overall mean area of 8.0 plus/minus 8.2 cm² (group 1 10.0 plus/minus 7.1 cm², group 2 6.1 plus/minus 9.3 cm²). Figure 3 shows an example in two patients, Table 1 shows details of the data, and Figure 4 gives a summary.

The ventilation-perfusion measurements are summarized below and in Table 2. In the awake condition, a shunt (V with dot A/Q with dot < 0.005) was found in only one patient (2.5% of cardiac output [CO]), the same patient also had a considerable amount of low V with dot A/Q with dot (14.2% CO). On the average, low V with dot A/Q with dot (0.005 < V with dot A/Q with dot < 0.1) was 3.1 plus/minus 4.2% CO. Pa sub O₂, measured while breathing air, exceeded 75 mmHg (10.0 kPa) in 11 of 12 patients (mean 98 plus/minus 22 mmHg [13.0 plus/minus 2.9 kPa]), and mean Pa_CO2 was 39.0 plus/minus 5.3 mmHg (5.2 plus/minus 0.7 kPa). The patient with the rather marked derangement of V with dot A/Q with dot and the low Pa_O2 was well premeditated and fell asleep during the measurement of V with dot A/Q with dot "awake." Furthermore, he was obese and had the second highest body mass index (patient 8 in Table 1).

After induction of anesthesia the pulmonary shunt was increased to 6.5 plus/minus 5.2%, low V with dot A/Q with dot increased to 5.3 plus/minus 3.3%, Pa_O2 was 150 plus/minus 55 mmHg (20.0 plus/minus 7.4 kPa) at an FI_O2 of 0.4, and Pa_CO2 was 33.7 plus/minus 2.3 mmHg (4.5 plus/minus 0.3 kPa). The estimated mean cardiac output, used to calculate the V with dot A/Q with dot distributions, was 4.4 plus/minus 1.0 l *symbol* min sup -1 in the awake subjects and 3.5 plus/minus 0.81 l *symbol* min sup -1 after induction of anesthesia.

The measurement of C_rs yielded values in the normal range for patients whose trachea is intubated and whose lungs are mechanically ventilated [23] (group 1 57.1 plus/minus 19.5 ml *symbol* cmH₂ O sup -1, group 2 55.0 plus/minus 19.1 ml *symbol* cmH₂ O sup -1; Figure 4).

The recruitment maneuvers caused no clinically important adverse effects.

Immediately after recruitment, atelectasis was virtually eliminated in both groups (0.0 plus/minus 0.1 and 0.1 plus/minus 0.2 cm sup 2 in groups 1 and 2, respectively). In group 1 (FI_O2 = 0.4), there was a slow increase in the amount of atelectasis, which reached about one-sixth of the initial area of atelectasis 40 min after the recruitment (after 5 min 0.2 plus/minus 0.2 cm²; after 40 min 1.6 plus/minus 2.1 cm², 95% CI 0 to 3.8 cm², P = 0.028 vs. anesthesia baseline). In group 2 (FI_O2 = 1.0), the amount of atelectasis was increased to a value close to prerecruitment 5 min after the recruitment maneuver (5.3 plus/minus 8.7 cm², 95% CI 0 to 16.5 cm², no significant difference vs. anesthesia baseline), and there was a further slow increase during the rest of the study period (after 40 min 7.9 plus/minus 9.7 cm², no significant difference vs. anesthesia baseline). See Figure 3 for an example in two patients and Figure 4 for a summary of the data. The mean intrathoracic area, expressed as percentage of the intrathoracic area in the awake subject, showed only small, insignificant changes from anesthesia baseline to 0 and 40 min after recruitment both in group 1 (91.3 plus/minus 5.6%/94.5 plus/minus 4.9%/92.0 plus/minus 5.2%) and in group 2. (90.5 plus/minus 10.4%/92.0 plus/minus 9.3%/90.8 plus/minus 9.7%).

Pulmonary vessels, suitable for measurements of angles, could be identified on subsequent CT scans of four patients in group 1 and of all six patients in group 2. Between anesthesia baseline and the first measurement immediately after the recruitment, there was a decrease of the intervascular angle in all but two patients (P = 0.013), the mean decrease being 12 plus/minus 7%/11 plus/minus 10% (group 1/2). As compared to the value before recruitment, the mean angles remained decreased in group 1 by 9 plus/minus 6% and 11 plus/minus 10% at 5 and 40 min after the recruitment, respectively. In group 2, the respective values were 5 plus/minus 7% and 3 plus/minus 10% (no significant difference vs. prerecruitment); thus there was a gradual increase of the intravascular angle toward the prerecruitment value.

In group 1, shunt decreased from 7.2 plus/minus 5.2% CO at anesthesia baseline to 1.3 plus/minus 2.2 and 2.3 plus/minus 2.6% CO 20 and 40 min after the recruitment maneuver (P = 0.028 and 0.028, if compared to anesthesia baseline), respectively. Low V with dot A/Q with dot increased from 6.7 plus/minus 2.1% CO to 11.0 plus/minus 3.5 and 8.0 plus/minus 3.7% CO (P = 0.028 and 0.3, if compared to anesthesia baseline). In group 2, there was no significant change in any of the measured parameters of the ventilation-perfusion distribution. At 40 min after the recruitment, shunt was 6.4 plus/minus 6.6%, and low VA/Q was 3.3 plus/minus 2.5% CO. Further details of the V with dot A/Q with dot measurements are given in Table 2. There was a correlation between the amount of atelectasis and shunt before the recruitment (group 1 r = 0.91, group 2 r = 0.99) and 40 min after recruitment (group 1 r = 0.69, group 2 r = 0.83), In group 1, Pa_O2 measured at an FI_O sub 2 of 0.4, increased from 137 mmHg (18.3 kPa) after induction of anesthesia to 163 mmHg (21.7 kPa) 40 min after the recruitment maneuver (Table 2). In group 2, the change in FI_O2 from 0.4 to 1.0 per se has an influence on Pa_O2.

The V with dot A/Q with dot data at 20 min after the recruitment were analyzed in more detail in two arbitrarily chosen patients, one from each group. In both, low V with dot A/Q with dot was significantly different from zero. Shunt, however, was not different from zero in patient 5 (group 1, with no or minor atelectasis 20 min after the recruitment: largest possible shunt of 2,500 calculations 4% CO), whereas it was different from zero in patient 9 (group 2, with reappearance of atelectasis 20 min after the recruitment; smallest possible shunt of 2,500 calculations 6% CO). Furthermore, there was a significant difference between shunt and low V with dot A/Q with dot in patient 5 and a significant difference between the shunt of patient 5 and that of patient 9. An increase or a decrease of the estimated CO by plus/minus 25% did not alter the statistical outcome. Thus, in these patients, it was possible to detect shunt and low V with dot A/Q with dot and to separate them from each other.

The time course of C_rs showed no difference between the two groups (Figure 4). In both groups, C_rs was increased by the recruitment maneuver (70.1 plus/minus 27.5/ 67.4 plus/minus 21.2 ml *symbol* cmH₂ O sup -1 in group 1/2, P = 0.028 vs. anesthesia baseline in both groups). Forty minutes after the recruitment, C_rs was still higher than after induction of anesthesia (group 1 61.4 plus/minus 19.6 ml *symbol* cmH₂ O sup -1, group 2 60.0 plus/minus 20.2 ml *symbol* cmH₂ O sup -1, P = 0.028 vs. anesthesia baseline in both groups). The change in C_rs by the recruitment maneuver was not related to the amount of atelectasis immediately before the recruitment (Figure 5). There also was no correlation between compliance and shunt or low V with dot A/Q with dot awake or after induction of anesthesia nor between compliance and atelectasis, shunt or low V with dot A/Q with dot.

This study confirms that reexpansion of atelectasis during general anesthesia in patients with healthy lungs may have a sustained effect with respect to atelectasis, shunt, and Pa_O2, if the lungs are ventilated with 40% Oxygen₂ in nitrogen. If 100% Oxygen sub 2 is used, however, lung collapse reappears within a few minutes, and as compared to prerecruitment, the ventilation-perfusion relationship is essentially unchanged after the recruitment. This time course of atelectasis suggests that gas resorption plays an important role in the recurrence of collapse in previously reexpanded atelectatic lung tissue. Such lung regions appear to be unstable, and the mechanisms that keep a lung unit open possibly are overcome by the forces caused by gas absorption.

The amount of atelectasis was estimated by CT. Methodologic details concerning the analysis of dense regions in dependent parts of the lungs have been discussed in detail elsewhere. [3,16] Atelectasis that appears after induction of general anesthesia is located mostly in dependent, basal parts of the lung. [3] To avoid excessive exposure to radiation, we decided to perform only one CT scan per condition under investigation. At three occasions, an additional scan at the level of the hilum and the apex of the lungs was taken. As in previous investigations, [3,8] the amount of atelectasis at these additional levels was always smaller as compared to the CT scan close to the dome of the diaphragm. With respect to the time course of atelectasis, the analysis of these additional scans provided no further information and therefore is not presented in this paper.

The ventilation-perfusion distribution in the lungs was analyzed with the multiple inert gas elimination technique. This method is based on the elimination and retention of a number (usually six) of "inert" gases. To calculate these parameters, the measurement of cardiac output and the concentrations of the "inert" gases in mixed expired air, mixed venous blood, and arterial blood is necessary. Some of these parameters thus necessitate the use of a pulmonary artery catheter. However, a simplified methodology can be applied, requiring measurements in arterial (or venous) blood and in mixed expired air only. [24] We decided to use this latter technique as it was not considered justifiable to apply the more invasive procedure of pulmonary artery catheterization to the subjects eligible for this study. Thus, mixed venous "inert" gas concentrations were computed using mass balance principles, and cardiac output was estimated from the oxygen consumption. This approximation of cardiac output is imprecise, but it has been shown that indexes of V with dot A/Q with dot mismatch (e.g., logSDQ and logSDV) remain essentially unaffected by such uncertainties. [18] Errors in cardiac output measurements, however, may be transmitted to other parameters of V with dot A/Q with dot, including shunt, low V with dot A/Q with dot, and mean V with dot A/Q with dot. Thus, a quantitative estimation of these parameters is less reliable. However, if cardiac output is constant during the study period, relative changes in shunt and low V with dot A/Q with dot can be estimated. This may be seen from the formula used to calculate mixed venous partial pressures (Pv) of an "inert" gas from the arterial partial pressure (Pa) and the partial pressure in mixed expired gas (PE) [18]: Pv = Pa + ((PE * VE)/lambda * QT)), where lambda blood gas partition coefficient, VE: minute ventilation, and QT: cardiac output.

With VE, QT, and lambda unchanged, changes in Pa and PE will transmit directly to changes in PV. Variations in shunt and low V with dot A/Q with dot during unchanged anesthesia therefore will not merely reflect errors in measurement but may be interpreted as changes of shunt or low V with dot A/Q with dot. Because oxygen uptake did not vary during anesthesia in those cases in which repeated measurements were made, we assume that CO was more or less constant during the study. Moreover, an analysis of the data of our investigation showed that a variation of cardiac output by 25% will result in a rather modest variation of shunt and low V with dot A/Q with dot by about 20% of the measured values.

Finally, Monte Carlo simulation and linear programming showed that low V with dot A/Q with dot could be separated from shunt and that the V with dot A/Q with dot distribution in patient 5 of group 1 (no or minor atelectasis 20 min after the recruitment) was significantly different from that in patient 9 of group 2 (with atelectasis 20 min after the recruitment). This lends support to our conclusion that the reappearance of atelectasis after a recruitment maneuver affects the V with dot A/Q with dot distribution.

The findings of this study are in accordance with previous investigations with respect to atelectasis and ventilation-perfusion relationship (V with dot A/Q with dot) before and after induction of general anesthesia with mechanical ventilation. [4,5] Furthermore, the changes in the amount of atelectasis and V with dot A/Q with dot after the recruitment maneuver in the patients ventilated with a FI_O2 of 0.4, yielded results similar to a recently performed study. [9] The very fast reappearance of densities in the group ventilated with 100% Oxygen₂ after the recruitment maneuver, however, deserves further discussion.

That resorption of gas may play a role in the formation of atelectasis during anesthesia has been discussed on theoretical grounds [10,25,26] and in a number of experimental and clinical studies, [25,27,28] although it was not possible to demonstrate lung collapse on conventional x-ray. [29] The role of this mechanism for the formation of atelectasis during general anesthesia remained unclear until now. In the current investigation, the fast reappearance of atelectasis in the group ventilated with 100% Oxygen₂ suggests that gas resorption is a major factor for a renewed collapse of previously reopened lung tissue.

Collapse by gas resorption may be present in a lung unit if there is a total stop of gas flow to this unit (thus resulting in trapped gas) or if the expired ventilation of this lung unit falls to zero (critical V with dot A₁ /Q with dot¹⁰). In the current study, the amount of atelectasis and shunt was reduced by the recruitment maneuver, the amount of lung units with low V with dot A/Q with dot increased at the same time. Therefore, the amount of units with a critical V with dot A₁ /Q with dot probably increased, too. It may be assumed that units with a low V with dot A/Q with dot are the ones that were collapsed and caused shunt before the recruitment. Therefore, they are located in dependent parts of the lungs. As the preinspiratory lung volume in supine subjects is markedly smaller in dependent than in nondependent parts of the lungs, [30] the gas used for an inflation up to vital capacity will be distributed predominantly to dependent lung units, too. Thus, if 100% Oxygen₂ is used for this maneuver, the concentration of oxygen will increase mostly in such lung units.

To summarize, a recruitment maneuver with oxygen results in all increase of lung units with low V with dot A/Q with dot, prone to fast collapse because of increased oxygen content within such units. In a computer model of absorption atelectasis, the time to collapse in a lung unit filled with 100% Oxygen₂ and excluded from ventilation has been estimated to be about 8 min. [26] This collapse may be faster if a mixture of nitrous oxide and oxygen is used. [25,26] According to the same study, this time will be about 3 h if a lung unit was filled with 30% Oxygen₂ in nitrogen before exclusion from ventilation. Whether a decreased function of surfactant, as suggested during general anesthesia, [31] or other, yet unknown factors play an additional role for instability of lung units remains to be shown.

Further evidence that this atelectasis is caused by resorption of gas is suggested by the results of our measurements of initravascular angles in nondependent parts of the lungs. These angles were decreased by the recruitment maneuver, and they increased again thereafter. This indicates that competition for space may cause a compression of nondependent parts of the lungs when dependent parts are expanded and an expansion of nondependent parts when dependent parts of the lungs collapse. In two other studies, it was found that lung units in the immediate vicinity of atelectatic tissue were well aerated or even hyperinflated, [32,33] thus fitting with an interdependence between adjacent lung units.

The compliance of the respiratory system (C_rs) did not show the same course as atelectasis. In both groups, C_rs was increased and atelectasis was decreased by the recruitment maneuver, but thereafter C_rs decreased similarly in both groups, whereas the behavior of atelectasis was different between groups 1 and 2. Furthermore, there appears to be no close relationship between the amount of atelectasis and the relative change in C_rs before and immediately after the recruitment maneuver (Figure 5). For example, there were subjects showing an increase in C_rs by one-third and having almost no atelectasis. On the other hand, the subject with the second largest amount of atelectasis had only a small increase in C_rs with the recruitment maneuver, despite the fact that the atelectasis was eliminated.

These findings indicate that atelectasis is not the only cause of decreased compliance. [34-36] Other mechanisms that may be involved are changes in thoracic configuration [36] (although we did not see any significant change in the transverse thoracic area in the present study), changes in the micromechanics of acinus or alveolar walls, [37] and an altered function of surfactant. It has been shown that a single stretch of alveolar type II cells may be followed by a stimulation of surfactant secretion that is sustained for up to 30 min. [38] This time scale is fairly similar to that of the behavior of compliance in the current investigation. The compliance measurement was made over the tidal volume range, which makes it sensitive to the presence of poorly ventilated lung regions. Regions with low V with dot A/Q with dot were more common in group 1 (with little atelectasis) than in group 2, and they may have balanced the effect of atelectasis on compliance.

In conclusion, atelectasis as found during general anesthesia in patients with healthy lungs may be reexpanded by deep inflations of the lungs, but resorption of gas plays an important role in the recurrence of collapse in such lung units. Therefore, gas exchange may be improved only by a recruitment maneuver, if the lungs are ventilated with a gas mixture containing a poorly resorbed gas such as nitrogen. The change in the amount of atelectasis and shunt appears to be independent of the change in total compliance of the respiratory system. The reason for the instability of lung units, leading to atelectasis with induction of anesthesia, remains to be established.

The authors appreciate the assistance of Malin Lundin, laboratory technician, and Marianne Almgren and Lena Haglund, x-ray technicians. The authors thank Peter D. Wagner, Department of Medicine, University of California, San Diego, San Diego, California, for his help in the detailed analysis of V with dot A/Q with dot, and Johan Bring, Department of Statistics, Uppsala University, for statistical advice.

1. Cote CJ, Golstein EA, Cote MA, Hoaglin DC, Ryan JF: A single blind study of pulse oximetry in children. ANESTHESIOLOGY 68:184-188, 1988. Buy Now Bibliographic Links [Context Link]

2. Moller JT, Johannessen NW, Berg H, Espersen K, Larsen LE: Hypoxemia during anaesthesia: An Observer study. Br J Anaesth 66:437-444,437-444, 1991. [Context Link]

3. Brismar B, Hedenstierna G, Lundquist H, Strandberg A, Swenson L, Tokies L: Pulmonary densities during anesthesia with muscular relaxation: A proposal of atelectasis. ANESTHESIOLOGY 52:422-428, 1985. [Context Link]

4. Tokics L, Hedenstierna G, Strandberg A, Brismar B, Lundquist H: Lung collapse and gas exchange during general anesthesia: Effect of spontaneous breathing, muscle paralysis, and positive end-expiratory pressure. ANESTHESIOLOGY 66:157-167, 1987. [Context Link]

5. Gunnarsson L, Tokics L, Gustavsson H, Eledenstierna G: Influence of age on atelectasis formation and gas exchange impairment during general anaesthesia. Br J Anaesth 66:123-432, 1991. [Context Link]

6. Shambaugh GE, Harrison WG, Farrell JL: Treatment of the respiratory paralysis of poliomyelitis in a respiratory chamber. JAMA 9:1371-1373, 1930. [Context Link]

7. Mead J and Collier C: Relation of volume history of lungs to respiratory mechanics in anesthetized dogs. J Appl Physiol 14:669-678, 1959. [Context Link]

8. Rothen HU, Sporre B, Engberg G, Wegenius G, Hedenstierna G: Re-expansion of atelectasis during general anaesthesia: A computed tomography study. Br J Anaesth 71:788-795, 1993. [Context Link]

9. Rothen HU, Sporre B, Engberg G, Wegenius G, Hedenstierna G: Reexpansion of atelectasis during general anaesthesia: How long does it last? Acta Anaesthesiol Scand 39:110-125, 1995. [Context Link]

10. Dantzker DR, Wagner PD, West JB: Instability of lung units with low V with dot A/Q with dot ratios during Oxygen sub 2 breathing. J Appl Physiol 38:886-895, 1975. Bibliographic Links [Context Link]

11. Nicholas TE, Power JHT, Barr HA: The pulmonary consequences of a deep breath. Respir Physiol 49:315-324, 1982. [Context Link]

12. Mason RJ: Surfactant secretion, Pulmonary Surfactant, from Molecular Biology to Clinical Practice. Edited by Roberston B, Golde LMG, Batenburg JJ, Amsterdam, Elsevier, 1992, pp 295-312. [Context Link]

13. Cronk CE, Roche AF: Race- and sex-specific reference data for triceps and subscapular skinfolds and weight/stature. Am J Clin Nutr 35:347-354, 1982. Bibliographic Links

14. Anonymous: Ethics and clinical research in anaesthesia (editorial). Lancet 339:337-338, 1992. [Context Link]

15. Dawson-Saunders B, Trapp RG: Basic and Clinical Biostatistics East Norwalk, Appleton and Lange, 1990, pp 118-119. [Context Link]

16. Lunquist H, Hedenstierna G, Strandbereg A, Tokics L, Brismar B: CT-assessment of dependent lung densities in man during general anesthesia. Acta Radiol Scand (in press). [Context Link]

17. Wagner PD, Saltzman HA, West JB: Measurement of continuous distribution of ventilation perfusion ratios. Theory J Appl Physiol 36:588-599, 1974. [Context Link]

18. Wagner PD, Smith CM, Davies NJH, McEvoy RD, Gale GE: Estimation of ventilation-perfusion inequality by inert gas elimination without arterial sampling. J Appl Physiol 59:376-383, 1985. Bibliographic Links [Context Link]

19. Wagner PD, West JB: Information content of the multiple inert gas elimination method, Pulmonary Gas Exchange. Volume I. Edited by Wagner PD, West JB, New York, Academic. 1980, pp 233-235. [Context Link]

20. Ratner ER, Wagner PD: Resolution or the multiple inert gas method for estimating V with dot A/Q with dot maldistributions. Respir Physiol 49:293-313, 1982. Bibliographic Links [Context Link]

21. Gottfried SB, Rossi A, Higgs BD, Calverley PMA, Zocchi L, Bozic C, Milic-Emili J: Noninvasive determination of respiratory system mechanics during mechanical ventilation for acute respiratory failure. Am Rev Respir Dis 131:414-420, 1985. Bibliographic Links [Context Link]

22. Gardner MJ, Altman DG: Confidence intervals rather than P values: Estimation rather than hypothesis testing. Br Med J 292:746-750, 1986. [Context Link]

23. Rehder K, Sessler AD, Marsh HM: General anesthesia and the lung. Am Rev Respir Dis 126:541-563, 1975. Bibliographic Links [Context Link]

24. Wagner PD, Hedenstierna G, Bylin G, Lagerstrand L: Reproductibility of the multiple inert gas elimination technique. J Appl Physiol 62:1740-1746, 1987. [Context Link]

25. Webb SJS, Nunn JF: A comparison between the effect of nitrous oxide and nitrogen on arterial PO sub 2. Anaesthesia 22:69-81, 1967. [Context Link]

26. Joyce CJ, Baker AB, Kennedy RR: Gas uptake from an unventilated area of lung: Computer model of absorption atelectasis. J Appl Physiol 74:1107-1116, 1993. Bibliographic Links [Context Link]

27. Dale WA, Rahn H: Rate of gas absorption during atelectasis. Am J Physiol 170:606-615, 1952. [Context Link]

28. Dery R, Pelletier J, Jacques A, Clavet M, Houde J: Alveolar collapse induced by denitrogenation. Can Anaesth Soc J 12:531-544, 1965. Bibliographic Links [Context Link]

29. Prys-Roberts C, Nunn JF, Dobson RH, Robinson RH, Greenbaum R, Harris RS: Radiologically undetectable pulmonary collapse in the supine position. Lancet 2:399-401, 1967. [Context Link]

30. Rehder K, Sesler AD, Rodarte JR: Regional intrapulmonary gas distribution in awake and anesthetized-paralyzed man. J Appl Physiol 42:391-402, 1977. Bibliographic Links [Context Link]

31. Wollmer P, Schairer W, Bos JAH, Bakker W, Krenning EP, Lachmann B: Pulmonary clearance of 99mTc-DTPA during halothane anaesthesia. Acta Anaesthesiol Scand 34:572-575, 1990. Bibliographic Links [Context Link]

32. Hedenstierna G, Lundquist H, Lundh B, Tokics L, Strandberg A, Brismar B, Frostell C: Pulmonary densities during anaesthesia: An experimental study on lung morphology and gas exchange. Eur Respir J 2:528-535, 1989. Bibliographic Links [Context Link]

33. Hachenberg T, Lundquist H, Tokics L, Brismar B, Hedenstierna G: Analysis of lung density by computed tomography before and during general anaesthesia. Acta Anaesthesiol Scand 37:549-555, 1993. Bibliographic Links [Context Link]

34. Rahn H, Farhi LE: Gaseous environment and atelectasis. Fed Proc 22:1035-1041, 1963. Bibliographic Links [Context Link]

35. Egbert LD, Laver MB, Bendixen HH: Intermittent deep breaths and compliance during anaesthesia in man. ANESTHESIOLOGY 24:57-60, 1963. Buy Now [Context Link]

36. Klineberg PL, Rehder K and Hyatt RE: Pulmonary mechanics and gas exchange in seated normal men with chest restriction. J Appl Physiol 51:26-32, 1981. Bibliographic Links [Context Link]

37. Bachofen H, Wilson TA: Micromechanics of the acinus and alveolar walls. The Lung: Scientific Foundation. Edited by Crystal RG, West JB, New York, Raven, 1991, pp 809-820. [Context Link]

38. Wirtz HRW, Dobbs LG: Calcium mobilization and exocytosis after one mechanical stretch of lung epithelial cells. Science 250:1266-1269, 1990. Bibliographic Links [Context Link]