Materials and Methods

Animals—Ten healthy female Bergamasca sheep from a research herd at the University of Bari (mean ± SD weight, 56 ± 9 kg [range, 45 to 68 kg]; mean ± SD age, 17 ± 2 months [range, 14 to 22 months]) were included in the study. The sheep were housed and all experiments were conducted at the School of Veterinary Medicine at Bari. During the experimental period, the sheep were housed together in 1 indoor stall (size, approx 40 m²). The diet consisted of grass and hay, and the sheep had free access to water; they were acclimated to these conditions for 1 month prior to the start of the study. Food was withheld for 24 hours before anesthetic induction. The sheep were considered healthy on the basis of the results of physical examination, which included hematologic, arterial blood gas, and serum biochemical analyses. The study protocol was approved by the Italian Ethical Committee of the Ministry of Health, Bari.

Anesthetic protocol and monitoring—All anesthetic inductions were performed in the CT suite. Skin at the site of venipuncture was surgically prepared and infiltrated with 1 mL of a 2% solution of lidocaine hydrochloride,^a and a 14-gauge catheter was placed in a jugular vein of each sheep. Midazolam hydrochloride^b (0.4 mg/kg, IV) was administered for sedation. After 5 minutes, anesthesia was induced with propofol^c (5 mg/kg, IV), and tracheal intubation was accomplished by use of a cuffed endotracheal tube (inner diameter, 12 mm). Sheep were positioned in sternal recumbency for intubation and subsequently moved into right lateral recumbency for the remainder of the study. Anesthesia was maintained by use of a constant rate infusion of propofol (0.4 mg/kg/min), and lactated Ringer's solution was infused at a rate of 10 mL/kg/h during the entire anesthetic procedure. Muscular paralysis was induced by administration of vecuronium^d (25 μg/kg, bolus IV) to facilitate mechanical ventilation and to avoid any interference of chest wall muscle tone on lung mechanics measurements.

The median auricular artery of the left pinna was percutaneously catheterized by use of a 20-gauge catheter to record arterial blood pressures (systolic, diastolic, and MAP)^e and to facilitate collection of arterial blood samples. Hemoglobin oxygen saturation (via pulse oximetry), Petco₂ (via a capnometer), heart rate (via ECG), and rectal temperature (via a digital thermometer) were continuously recorded.

Neuromuscular function was monitored with a peripheral nerve stimulator^f operating in the train-of-4 stimulation mode. Complete neuromuscular blockade (absence of any twitch response during train-of-4 nerve stimulation) was confirmed before the respiratory maneuvers were performed; on completion of the experiment after edrophonium and atropine administration^23,g, 4 responses (twitches) to train-of-4 nerve stimulations were required before the constant rate infusion of propofol was discontinued and sheep were allowed to recover from anesthesia with appropriate monitoring and assistance.

Mechanical ventilation and experimental protocol—Immediately after intubation, sheep were connected to a nonrebreathing circuit and mechanically ventilated^h by use of a respirator operated in a volume-controlled, time-cycled mode with a constant fresh gas flow, Fio₂ of 1 (ie, 100% oxygen), V^T of 12 mL/kg, and inspiratory-to-expiratory ratio of 1:3 with ZEEP. The respiratory rate was adjusted to maintain Petco₂ at 35 to 45 mm Hg. The Fio₂ (maintained at 1) was continuously monitored during the experiment by use of an oxygen sensor located inside the ventilator.

Sheep were positioned in right lateral recumbency; this was considered time 0 (ie, T₀). Spiral CT scan of the thorax was performed at 15, 30, and 60 minutes with ZEEP (ie, T₁₅, T₃₀, and T₆₀, respectively), and arterial blood gas analysis was performed from samples obtained at each time point. Sixty minutes after positioning the sheep in right lateral recumbency, 10 cm H₂O of PEEP was applied for 20 minutes, without changing the ventilatory pattern settings. The CT and blood gas analysis were repeated after PEEP was applied for 20 minutes (ie, T_PEEP).

Evaluation of respiratory system mechanics— Measurement of respiratory system mechanics was performed at T₁₅, T₃₀, T₆₀, and T_PEEP according to the following methods. Flow was measured with a heated pneumotachograph,ⁱ which was connected to a differential pressure transducer^j inserted between the Y-piece of the ventilator circuit and the endotracheal tube. The response to the pneumotachograph was linear over the experimental range of gas flows. The transducer was calibrated before each experiment via a 1-point calibration procedure. Volume was determined via numerical integration of the flow signal. The Pao was measured proximal to the endotracheal tube by use of a pressure transducer.^j All variables for respiratory system mechanics were displayed and collected on a personal computer through a 12-bit analogue-to-digital converter board^k at a sampling rate of 200 Hz.^l The difference between the PEEP set on the ventilator (ie, the Pao value at the end of a regular breath [PEEP_external]) and the pressure in Pao during a 3- to 5-second end-expiratory pause (ie, PEEP_total) was measured, and this value was regarded as the static intrinsic PEEP (ie, PEEP_intrinsic).²⁴ The end-expiratory pause was performed by use of the expiratory hold on the ventilator. The CSTAT_rs was calculated at T₁₅, T₃₀, T₆₀, and T_PEEP as: CSTAT_rs = V^T/Pao_plat − PEEP_total, where Pao_plat is the value of Pao measured at the end of an end-inspiratory pause of 4 seconds by use of the inspiratory hold of the ventilator. After confirming complete neuromuscular blockade, the end-inspiratory and end-expiratory pauses were performed sequentially at each experimental time point, with 1 or 2 regular mechanical tidal breaths delivered between the 2 pauses (Figure 1).

View Full Size

Concurrent tracings of oxygen flow (top) and airway opening pressure (bottom) obtained from a representative healthy adult sheep mechanically ventilated with pure oxygen during anesthesia induced and maintained with propofol (5 mg/ kg and 0.4 mg/kg/min, respectively, IV); muscular paralysis was induced via bolus administration of vecuronium (25 μg/kg, IV). Tracings indicate the end-inspiratory and end-expiratory pauses separated by a regular breath (ie, without an occlusive pause). Pao_plat = Airway plateau pressure corresponding to the value of Pao after an end-inspiratory pause of 3 to 4 seconds. PEEP_external = PEEP (applied by the ventilator) corresponding to the value of the Pao at the end of a regular breath. PEEP_total = PEEP value that takes into account the possible intrinsic PEEP and corresponds to the value of the Pao at the end of an end-expiratory pause of 3 to 4 seconds.

Citation: American Journal of Veterinary Research 71, 8; 10.2460/ajvr.71.8.867

• Download Figure
• Download figure as PowerPoint slide

Collection and analysis of CT images—Frontal topograms and helical CT images of the thorax were obtained by use of a third-generation spiral CT scanner.^m The CT images were acquired at a setting of 120 kVp and 160 mA (1 second • time) by use of a lung algorithm; matrix size was 512 × 512, field of view was 50, and pitch was 10 mm. All CT images were obtained during an end-expiration apnea, which was induced by briefly disconnecting the endotracheal tube from the ventilator circuit at T₁₅, T₃₀, and T₆₀; at T_PEEP, the ventilator was configured in a continuous positive-pressure mode at 10 cm H₂O. Mechanical ventilation was resumed immediately after the CT image was obtained (60 to 70 seconds after disconnection from the ventilator). All CT images were analyzed for lung abnormalities; pathological changes were considered cause to exclude sheep from the study.

The CT images were analyzed by use of a commercially available computer program.ⁿ The right and left lungs were chosen as ROIs for analysis; portions of the pulmonary hila containing the trachea, main bronchi, and hilar blood vessels were excluded from the ROIs. Distributions of radiographic attenuations (expressed in HUs) among the selected ROIs were plotted by use of the computer software. In accordance with a previous study⁴ in dogs, the following regions or compartments were identified within the lungs: hyperinflated (ie, composed of pixels with CT numbers of −1,000 to −901 HUs), normally aerated (ie, composed of pixels −900 to −501 HUs), poorly aerated (ie, composed of pixels with CT numbers of −500 to −101 HUs), and nonaerated (ie, atelectatic; composed of pixels with CT numbers of −100 to 100 HUs).

The gas volume of each ROI (ie, ROI_V) was computed by including pixels with density values of −1,000 to 100 HUs according to the following formula: gas volume (mL) = CT number/(–1,000) • voxel volume,²¹ where each voxel is a pixel (ie, parallelogram with a square base [0.59 mm/side]) and a height corresponding to the CT slice thickness (10 mm); this was calculated as the area of the pixel times the CT slice thickness. The volume of each compartment (ie, hyperinflated, normally aerated, poorly aerated, and atelectatic) within the ROI was calculated by use of the same formula.

The EELV for each sheep was calculated as the sum of all ROI_V for all CT images obtained, and the total volume of each lung compartment was calculated as the sum of all volumes for a specified compartment in all CT slices. The total volume of each compartment was expressed as a percentage of the EELV.

The PEEP-induced alveolar recruitment was calculated²⁵ as the difference in the volume of the atelectatic lung compartment in the entire lung between T₆₀ and T_PEEP and expressed as a percentage of the value at T₆₀. The right and left lungs were also considered separately as ROIs, in accordance with the same method. To determine the craniocaudal distribution of atelectasis within both lungs, 26 slices (equally distributed throughout each lung in each sheep) were selected for evaluation at T₆₀ and T_PEEP and the number of voxels with an HU value between −100 and 100 was calculated for each slice.

Blood gas measurements—Blood gas measurements and related analyses were performed by use of an automated arterial blood gas analyzer.^o The analyzer was calibrated each day and prior to each experiment. Arterial blood samples were collected anaerobically at T₁₅, T₃₀, T₆₀, and T_PEEP before CT images were obtained; samples were analyzed immediately. The pH of arterial blood, Pao₂, and Paco₂ were measured. All arterial blood gas values were corrected within the analyzer for body temperature of the sheep (measured per rectum at the time of sampling). The Pao₂-Pao₂ was calculated for each sheep, according to the alveolar gas equation: Pao₂-Pao₂ = ([P_B − P_H2O] × Fio₂ − Paco₂) − Pao₂, where P_B is the barometric pressure and P_H2O is the water vapor pressure. The P_B was recorded by the arterial blood gas analyzer during each analysis, and the P_H2O was corrected for the rectal temperature of the sheep recorded at the time of arterial blood collection. The Vds_alv was estimated as a percentage of V_t by use of the following equation: Vds_alv/V^T = ([Paco₂ − Petco₂]/Paco₂) × 100, where Petco₂ is the value measured at the moment of arterial blood collection.

Statistical analysis—The mean ± SD was determined for each numeric variable. Normal distribution of data was verified by use of a Shapiro-Wilk test. Variables for lung function, lung aeration, and cardiovascular function obtained at T₁₅, T₃₀, T₆₀, and T_PEEP were compared among time points. Statistical analysis included a 1-way ANOVA for repeated measures followed by Student-Newman-Keuls test. Values of P < 0.05 were considered significant.

Results

Experimental procedures were completed in all sheep without complications. Mean ± SD scanning time per helical CT was 65 ± 5 seconds. No pathologically altered lung parenchyma was detected in any sheep during preliminary examination of CT images; thus, no sheep were excluded from the study.

Heart rate, MAP, respiratory rate, Petco₂, and hemoglobin oxygen saturation measured during mechanical ventilation at T₁₅, T₃₀, T₆₀, and T_PEEP did not differ significantly among time points (Table 1). The EELV was significantly greater at T_PEEP (3.0 ± 0.4 L), compared with values recorded at T₁₅ (2.1 ± 0.3 L), T₃₀ (2.2 ± 0.5 L), or T₆₀ (2.2 ± 0.6 L). The relative sizes of lung compartments (ie, hyperinflated, normally aerated, poorly aerated, and atelectatic) were determined for the various experimental conditions (Figure 2). The volume of the normally aerated compartment represented a significantly greater percentage and the poorly aerated and atelectatic compartments represented a significantly lesser percentage of the EELV at T_PEEP, compared with the volumes at ZEEP (ie, T₁₅, T₃₀, and T₆₀). Notably, the volume of the hyperinflated compartment did not change among the different experimental conditions.

View Full Size

Relative sizes of 4 functional lung compartments identified via analysis of CT images obtained from 10 propofol-anesthetized sheep positioned in right lateral recumbency (time of positioning = time 0). Bars represent mean ± SD percentage of the EELV. Data were recorded at predetermined intervals (15 minutes [T₁₅, white bars], 30 minutes [T₃₀, gray bars], and 60 minutes [T₆₀, vertically dotted line bars]) with ZEEP and after application of 10 cm H₂O of PEEP for 20 minutes (T_PEEP; black bars). ^a,bWithin each anatomic compartment, values with different letters are significantly (P < 0.05) different among time points on the basis of a 1-way ANOVA for repeated measures and a Student-Newman-Keuls test. See Figure 1 for remainder of key.

Citation: American Journal of Veterinary Research 71, 8; 10.2460/ajvr.71.8.867

• Download Figure
• Download figure as PowerPoint slide

When both lungs were considered together as an ROI, the volume of the atelectatic compartment (ie, atelectatic lung) was decreased by 5.8 ± 3.2%, compared with the value at T₆₀. When each individual lung was considered as an ROI, the volume of the atelectatic compartment was significantly (P = 0.01) decreased by 33.8 ± 6.3% in the right (ie, dependent) lung (from 0.143 ± 0.002 L at T₆₀ to 0.095 ± 0.009 L at T_PEEP) and by 6.62 ± 2.32% in the left lung (from 0.045 ± 0.003 L at T₆₀ to 0.042 ± 0.004 L at T_PEEP). The distribution of atelectatic lung tissue (detected as voxels with an HU value between −100 and 100) across the entire length of both lungs in a craniocaudal direction revealed increased atelectasis formation in the caudal (slices 12 through 26) lung fields (Figure 3).

View Full Size

Distribution of atelectasis throughout the lungs in 10 propofol-anesthetized sheep. One voxel is a pixel (ie, parallelogram with a square base [0.59 mm/side]) and a height corresponding to the CT slice thickness (10 mm); bars represent the mean number of voxels with an HU value between −100 and 100 (ie, the HU range indicative of atelectasis). Transverse CT slices were equally distributed across the entire lung; CT slice numbers represent images obtained serially as the scan proceeded in a cranial to caudal direction (slices with low numbers were obtained from the most cranial lung fields, and slices with high numbers were obtained from the most caudal lung fields). Data were recorded at T₆₀ with ZEEP (white bars) and at T_PEEP (black bars). See Figures 1 and 2 for remainder of key.

Citation: American Journal of Veterinary Research 71, 8; 10.2460/ajvr.71.8.867

• Download Figure
• Download figure as PowerPoint slide

Transverse CT images of the thorax (at the level of the body of T8) were obtained (Figure 4). Atelectasis of the parenchyma in the dependent lung was detectable in the CT images obtained at T₁₅, T₃₀, and T₆₀; at T_PEEP, the amount of atelectasis was reduced.

View Full Size

Transverse CT images of the thorax obtained at the level of the body of T8 from the same propofol-anesthetized representative sheep used to obtain the tracing in Figure 1. Images were obtained at T₁₅ (A), T₃₀ (B), and T₆₀ (C) with ZEEP and at T_PEEP (D). Atelectatic lung parenchyma (black arrows) is apparent in the dependent (right) lung at T₁₅, T₃₀, and T₆₀; at T_PEEP, the amount of atelectasis is reduced (white arrow). See Figure 2 for remainder of key.

Citation: American Journal of Veterinary Research 71, 8; 10.2460/ajvr.71.8.867

• Download Figure
• Download figure as PowerPoint slide

The Pao₂ was significantly increased and Pao₂-Pao₂ was significantly decreased at T_PEEP, compared with the respective values at T₁₅, T₃₀, and T₆₀ (Table 2). Because pure oxygen was used for ventilation throughout the experiment (ie, the Fio₂ remained constant at 1), the effects of PEEP on the Pao₂-to-Fio₂ ratio were identical to those on Pao₂. The CSTAT_rs was significantly increased at T_PEEP, compared with values calculated at T₁₅, T₃₀, and T₆₀ (Table 3).

Statistical analysis was performed for the remaining data (rectal temperature, pH, PaCO₂, and Vds_alv) obtained during the experiments. Results of this analysis did not indicate any significant differences among time points.

Discussion

Results of the study reported here supported our hypothesis that mechanical ventilation with pure oxygen in propofol-anesthetized sheep would result in lung atelectasis, impaired pulmonary mechanics, and reduced gas exchange and that application of 10 cm H₂O of PEEP would substantially improve these functions by reopening small airways and recruiting previously collapsed alveoli. Even in healthy sheep, poorly aerated and atelectatic lung compartments begin to form rapidly during anesthesia when animals are mechanically ventilated and inspire pure oxygen.²⁶ Our data indicate that the change in lung aeration in sheep is not necessarily the result of a slowly progressive process but develops quickly and may then remain fairly static throughout the anesthetic procedure (Figures 2 and 4). This observation is consistent with the results of a study² in anesthetized humans ventilated with pure oxygen, in which pulmonary atelectasis was detected within the first 5 to 10 minutes of anesthesia but did not increase with advancing time. The results of the present study revealed that the percentages of poorly aerated and atelectatic lung parenchyma detected during the initial anesthetic period (ie, T₁₅) were approximately 30% and 7%, respectively, indicating that nearly 40% of the lung tissue was not properly aerated. Atelectases were predominant in the more caudal lung fields (ie, slices 12 through 26), close to the diaphragm, where the main abdominal compression forces are localized.

In sheep and in ruminants in general, the compression exerted by the rumen can substantially impair lung function during anesthesia, compared with the lung function of anesthetized monogastric animals.²⁷ Indeed, the rumen frequently remains filled with content, even after feed is withheld for an extended period prior to anesthesia. Fermentation also continues during general anesthesia and eructation usually ceases, which promotes tympanic expansion of the rumen and further impairment of lung inflation during inspiration.^27,28 We determined that actelectases were more pronounced in the dependent lung of sheep in lateral recumbency (Figure 4). This finding was similar to the results of a study²⁹ in anesthetized, mechanically ventilated human patients in lateral recumbency, which indicated that the dependent lung, although better perfused, is subject to more widespread small-airway closure and atelectasis formation and thus to increased V/Q mismatching. Furthermore, the V_t distributes preferentially in the nondependent lung; this increases the risks of alveolar hyperinflation and further impairment of gas exchange (caused by increased dead space) and may result in lung injury.²⁹

The principal consequence of poorly aerated and atelectatic lung tissue is impaired gas exchange with intrapulmonary shunt formation and a low V/Q.⁹ In the present study, the high Pao₂-Pao₂ gradient (Table 2) detected during the first 60 minutes of anesthesia with mechanical ventilation was compatible with impaired lung aeration; this possibility was considered even more likely because the MAP and heart rate were within respective reference ranges¹⁷ reported for sheep (Table 1). The fact that Paco₂ and Vds_alv/V^T remained within the physiologic range possibly indicates that minute ventilation was adequate and alveolar dead space remained constant throughout the experiments.

Prevention of alveolar collapse and recruitment of alveoli in atelectatic and poorly aerated tissue are important objectives that may prevent pulmonary complications during and after general anesthesia.^30–32 Administration of a low Fio₂ (ie, 0.3 to 0.6; 30% to 60% oxygen) significantly reduces atelectasis formation in humans and nonhuman animals.^2–4 However, ventilation with a low Fio₂ is not always applicable in a clinical setting because of technical (eg, lack of a second gas source, lack of control over the oxygen mixture, or inability to monitor Fio₂) or clinical reasons (eg, compromise of lung function prior to surgery). In those circumstances, PEEP and recruitment maneuvers are the only ventilatory techniques available for treatment of anesthesia-related airway closure and atelectasis during surgery.^30–36 The term PEEP describes an increased pressure in the airways relative to ambient pressure at the end of expiration. The desired effect of PEEP on the pulmonary parenchyma is to increase FRC as a result of the opening of closed bronchioli and collapsed alveolar units.^10,36,37

In the study reported here, the application of 10 cm H₂O of PEEP for 20 minutes significantly reduced anesthesia-induced atelectasis (Figures 2 to 4). As expected, alveolar recruitment was primarily in the more caudal lung fields that were most affected by airway closure and atelectasis (Figure 3). Given the lack of any significant time-dependent change in lung aeration, gas exchange, and lung compliance from T₁₅ through T₆₀, it is extremely unlikely that lung function would have improved without the application of PEEP during the 20 minutes after T₆₀. Interestingly, alveolar recruitment was predominant in the dependent (right) lung. Our finding confirms the results of a study³⁸ in anesthetized horses, in which addition of PEEP to positive-pressure ventilation significantly improved aeration of the dependent lung; in that study, the improvement was attributed to a reversible shift of inspired gas from the nondependent toward the dependent lung. Similar results were reported in a study³⁹ in anesthetized humans.

It is important to distinguish between PEEP-induced recruitment and V_t-induced alveolar recruitment; these are different events that have different impacts on lung function. Positive end-expiratory pressure is an end-expiratory treatment that is persistent throughout the respiratory cycle. This ensures that alveolar units recruited by PEEP participate fully in gas exchange (provided that the lung parenchyma is physiologically normal). In contrast, V_t recruitment is a temporary (ie, inspiratory phase only) event. Although V_T recruitment may contribute to the improvement of gas exchange, the previously opened lung units collapse again (referred to as derecruitment) during the expiratory phase. The application of PEEP can influence V_T recruitment as a result of the increase in airway pressure; thus, it is likely that, in the sheep in the present study, there was further improvement of lung aeration during inspiration at T_PEEP. To accomplish the objective of this study, however, the authors included an evaluation of PEEP-induced alveolar recruitment. For this reason, CT was performed at end expiration.

The improved gas exchange observed at T_PEEP (Table 2) in sheep in the study reported here emphasizes the positive effect PEEP ventilation has on lung function. In studies^10,15 in humans, however, it is cautioned that PEEP-induced improvement of lung aeration may not always be associated with better gas exchange because of the potentially negative impact of PEEP on the pulmonary (ie, redistribution of pulmonary blood flow) and systemic (ie, reduction of cardiac output) circulation. Regarding redistribution of the pulmonary blood flow, even if the V/Q ratio had not been specifically determined in the present study, the significant improvement in oxygenation and in CSTAT_rs at T_PEEP supported the premise that 10 cm H₂O of PEEP improved the V/Q. Regarding systemic hemodynamic function, a possible limitation of the study reported here is that no data were obtained to indicate whether cardiac output was impaired in the sheep. Hemodynamic impairment is an important potential limitation for the application of PEEP during mechanical ventilation in anesthetized patients. In particular, the persistent increase in intrathoracic pressure causes a reduction in venous return; thus, blood flow in the pulmonary and systemic circulation could potentially be reduced.¹⁶ However, clinical studies¹⁰ in humans and nonhuman animals revealed that in healthy subjects, the cardiovascular compromise imposed by relatively low PEEP (≤ 10 cm H₂O) can usually be counteracted with administration of adequate fluids, inotropic drugs, or both. Because MAPs and heart rates remained constant in sheep in the study reported here, we speculate that application of a low PEEP had only minor effects on cardiac output, but we recognize that this aspect deserves further investigation. Moreover, we cannot exclude that the reduction of the Pao₂-Pao₂ detected in the present study may have been partly attributed to a reduction in the shunt fraction as a consequence of a small reduction in cardiac output induced by PEEP.

Collapsed airways and atelectatic alveoli are usually characterized by higher opening pressures than those required for physiologically normal alveoli; to be recruited (ie, reopened), they require a PEEP high enough to increase the inspiratory airway pressure above the required opening pressure and to prevent recollapse of those units at end of expiration.¹⁰ If the PEEP applied is insufficient to accomplish this and alveoli in the atelectatic lung regions cannot be opened, PEEP ventilation fails to recruit lung parenchyma and can potentially result in pulmonary hyperinflation, with PEEP and V_t disproportionally redistributed to normally aerated lung parenchyma, thereby reducing lung compliance further and impairing gas exchange.¹⁵ On the basis of clinical studies^32–35 in human patients, the most rational and successful approach during those circumstances (ie, failure to recruit the atelectatic alveolar units) is to perform a recruitment maneuver via inflation of the lungs with high inspiratory pressures (eg, 30 to 40 cm H₂O) for a brief period to recruit as many previously collapsed alveolar units as possible, then to apply a PEEP titrated to a value that will keep alveoli and small airways open. In the group of sheep in the present study, it appeared that the use of PEEP did not induce pulmonary hyperinflation (Figure 2). The opening pressures of alveolar units in atelectatic lung parenchyma are heterogeneous; these vary with the factors described as contributing to atelectasis formation as well as with pathological changes in the lung tissue.¹² It is important to stress the notion that anesthesia-induced atelectasis usually consists of collapsed alveolar units in healthy lung parenchyma that can easily be recruited and returned to physiologic function.^8,40 The situation is different for pathological pulmonary atelectasis (eg, that which results from acute lung injury, acute respiratory distress syndrome, pneumonia, or tumors), in which inflammatory cell infiltrates, degenerative processes, or both decrease compliance and impair alveolar recruitment.⁴⁰

The application of PEEP should be considered a temporary treatment that is limited to the time that a patient is receiving mechanical ventilation, acting primarily during the end-expiratory phase to address temporary changes of the respiratory system (eg, decreased FRC, airway closure, and pulmonary atelectasis) induced by the anesthetic drugs and techniques. The results of the present study indicated that PEEP improved oxygenation during anesthesia induced and maintained with propofol in healthy sheep ventilated with pure oxygen; application of PEEP also increased FRC via recruitment of atelectatic lung regions. The possible clinical impact of these results in veterinary medicine deserves further investigation, but human data clearly suggest that limiting atelectasis formation during anesthesia is an important step in limiting pulmonary complications during and after surgery.⁵