Protective Mechanical Ventilation during General Anesthesia for Open Abdominal Surgery Improves Postoperative Pulmonary Function

POSTOPERATIVE pulmonary complications, especially postoperative respiratory failure, are important causes of perioperative morbidity and mortality.^1–4  Induction of general anesthesia promotes a reduction in lung volume and atelectasis formation associated with a deterioration of both gas exchange and respiratory mechanics.^5,6  There is unequivocal evidence from both experimental and clinical studies that mechanical ventilation in critically ill patients has the potential to aggravate or even initiate lung injury. Two retrospective studies^7,8  and one randomized controlled trial⁹  suggested that lower tidal volumes are beneficial in patients who need long-term mechanical ventilation but do not suffer from lung injury.

Mechanical ventilation is mandatory in patients undergoing general anesthesia. Higher tidal volumes may overdistend noninjured lungs, in particular, nondependent lung tissue. During surgical procedures, both phenomena may stress the noninjured lung, triggering local inflammation and local coagulation.^10,11  Retrospective and prospective studies have shown possible beneficial effects of lower tidal volumes in patients who are on short-term mechanical ventilation because of surgery.^12,13  However, the beneficial effects of short-term intraoperative mechanical ventilation with lower tidal volumes on pulmonary integrity remains undefined.^14,15  In addition, zero-positive end-expiratory pressure (ZEEP) or low levels of positive end-expiratory pressure (PEEP) may promote atelectasis, resulting in repetitive collapse/reopening of dependent lung tissue. The role of PEEP and low tidal volume (the so-called protective lung strategy) during the intraoperative period in preventing lung damage and postoperative pulmonary complications is not clearly defined.^16,17  Indeed, recently an experimental study suggested that such strategy might even lead to increased inflammation and lung injury in normal lungs.¹⁸  Despite these pathophysiological considerations, the use of PEEP in the operating room is controversial. Recently, an observational study conducted in 28 centers in France revealed that most patients undergoing general surgery were ventilated without PEEP.¹⁹  Notably, the use of recruitment maneuvers (RMs) to open the lungs has also been found to improve the effectiveness of PEEP with regard to gas exchange during general anesthesia.^20–22  However, to date, there is no clear evidence of an additional benefit of RMs for routine anesthesia.

The current study compared a lung-protective mechanical ventilation strategy combining the use of lower tidal volume (Vt), higher PEEP levels, and intraoperative RMs, with a conventional standard mechanical ventilation (higher tidal volume, ZEEP without intraoperative RMs) during abdominal nonlaparoscopic surgery lasting more than 2h.¹ 

We hypothesized that in patients with normal lungs scheduled for general anesthesia, a protective ventilation strategy might prevent lung function modifications and lung morphological alterations.

The aim was to determine in this patient population the effect of an intraoperative protective ventilation strategy on modifying chest x-ray images, oxygenation, and pulmonary functional tests.

This unfunded, prospective, randomized, open-label, clinical trial was registered at ClinicalTrials.gov (ID NCT00426790). This trial was designed to demonstrate a superiority of treatment in protective group compared to the standard group.

The medical ethics committee of the Ospedale di Circolo e Fondazione Macchi, Varese, Italy, approved the trial protocol, and informed consent was obtained from all patients before inclusion. The trial was not overseen by an independent safety board monitoring due to its design as a pilot physiological study.

Patients scheduled for elective nonlaparoscopic abdominal surgery under general anesthesia from May 2006 to May 2008 were selected through the clinical anesthesia service of our regional university hospital—Azienda Ospedaliera Ospedale di Circolo e Fondazione Macchi of Varese, Italy.

Patients were eligible for participation if they met the following criteria: nonlaparoscopic abdominal surgery under general anesthesia expected to last more than 2h and age more than 18 yr. Exclusion criteria were as follows: body mass index more than 40kg/m², laparoscopic surgery, need for surgery in emergency, previous lung surgery (any), persistent hemodynamic instability, intractable shock considered unsuitable for the study by the patient’s managing physician, history of chronic obstructive pulmonary disease, repeated systemic corticosteroid therapy for acute exacerbations of chronic obstructive pulmonary disease, asthma or sleep disorders, recent immunosuppressive medication defined as need of chemotherapy or radiation therapy, less than 2 months after chemotherapy or radiation therapy, severe cardiac disease defined as New York Heart Association class III or IV, or acute coronary syndrome, or persistent ventricular tachyarrhythmias, pregnancy (excluded by laboratory analysis), acute lung injury or acute respiratory distress syndrome, expecting to require prolonged postoperative mechanical ventilation, any neuromuscular disease, contraindications to position an epidural catheter because of major clotting disorders,^23–25  or sign of infection at the site of the procedure.²⁶ 

A central venous line was inserted in all patients, and a conservative fluids infusion (12–15 ml⋅kg⁻¹⋅h⁻¹) was administered during the study period to assure hemodynamic stability. Before they were given general anesthesia, patients underwent epidural anesthesia at the T8-T12 level whenever not contraindicated. After surgery the patients received a continuous infusion of ropivacaine 0.2% at 4–6ml/h and morphine 0.1–0.15mg for at least 48h (AmbIT PCA; Summit Medical Products, Inc., Sandy, UT) with the possibility of having patient-controlled bolus volume of 4–6mg/h. The catheter was scheduled to be removed on the fourth day after surgery. Removal of the epidural catheter was planned at least 12h after the last dose of low–molecular weight heparin (Enoxaparine sodium 4,000U/die in a single subcutaneous administration) and 4h before the next administration.

In other patients analgesia was provided by subcutaneous continuous infusion through an elastomeric infusion system with morphine (0.3–0.4mg/kg in 24h) and ketorolac (1.0–1.5mg/kg in 24h) according to creatinine serum level. In only one patient analgesia was performed by a continuous intravenous infusion of ketorolac (60mg in 24h) and tramadol (300mg in 24h); rescue dose was provided by subcutaneous morphine (5–10mg). All patients were preoxygenated with Fio₂ 0.8 before tracheal intubation, and maintained at 0.4 during the entire anesthesia procedure, irrespective of study group, and received a routine anesthesia according to protocol, including intravenous fentanyl (1–3μ/kg), propofol (2–3mg/kg) at induction; thereafter, anesthesia was maintained with propofol (5–10 mg⋅kg⁻¹⋅h⁻¹) or sevoflurane (inspiratory concentration between 1.5 and 2%); analgesia was provided with continuous remifentanil infusion (0.05–0.3 μ⋅kg⁻¹⋅min⁻¹) or fentanyl (1–3μ/kg) as required. Patients were intubated after they were administered rocuronium bromide (0.8mg/kg); rocuronium was administered every 40min, and the last administration was at least 1h before the end of surgical suture. Routine intraoperative monitoring was performed using a dedicated monitor (IntelliVue Mp70; Philips Electronics, Eindhoven, The Netherlands) and included noninvasive blood pressure, pulse oximetry, end-tidal fractions of carbon dioxide, and electrocardiogram.

According to the standard of care in our institution all patients underwent conventional physiotherapy²⁷  (early mobilization, stimulation of cough, and incentive spirometry), control of pain to achieve a Visual Analogue Scales (VAS) below 3, antibiotic prophylaxis, and antithrombotic treatment as required in the postoperative period.

Concealed randomization was conducted to ensure a fair comparison between groups: to select patients for treatment we generated a randomization list by Random Allocation Software (Windows software, version 1.0, May 2004, Saghaei, licensee BioMed Central Ltd.) (allocation ratio 1:1) and inserted the group-identification paper in envelopes, which were then sealed and clouded to not reveal allocations.

The ventilation protocol consisted of volume-controlled mechanical ventilation (Datex Ohmeda S/5 Avance; GE Healthcare, Helsinki, Finland) at an inspired oxygen fraction of 0.40, inspiratory to expiratory ratio of 1:2, and a respiratory rate adjusted to normocapnia (end-tidal carbon dioxide partial pressure between 30 and 40 mmHg). The compliance value was calculated with the plateau pressure measured during the normal ventilation setting, with an inspiratory pause set at 40% of the inspiratory time. Patients were randomly assigned to mechanical ventilation with either a tidal volume of 9ml/kg ideal body weight (IBW) and ZEEP (the standard ventilation strategy) or 7ml/kg IBW and 10cm H₂O PEEP with RMs (the protective ventilation strategy). IBW was calculated according to a predefined formula: 50 + 0.91 (height [cm] −152.4) for men and 45.5 + 0.91 (height [cm] −152.4) for women.²⁸  RMs, as part of the protective strategy, were performed directly after induction of anesthesia, after any disconnection from the mechanical ventilator and directly before extubation, in hemodynamic stable situation as judged by the attending physician. Briefly, RMs were performed in volume-controlled ventilation as follows: the limit of peak inspiratory pressure was initially set at 45cm H₂O, the tidal volume at 7ml/kg IBW, and respiratory rate at 6 breaths/min, PEEP at 10cm H₂O, and the inspiratory to expiratory ratio at 3:1; then the tidal volume was increased in steps of 4ml/kg IBW until plateau pressure reached 30cm H₂O and three breaths were allowed. Finally, the respiratory rate, the inspiratory to expiratory ratio, inspiratory pause, and tidal volume were set back at values preceding the RM, whereas the PEEP was maintained at 10cm H₂O.

We defined a remarkable reduction in systolic arterial pressure when less than 90 mmHG and ensured that a mean arterial pressure less than 60 mmHg was not accepted.

Anesthesiologists were allowed to change the ventilation protocol at any point on the surgeon’s request, or if there was any concern about patient safety.

During mechanical ventilation, airway pressures, tidal volume, and the respiratory rate were measured by means of the facilities of the ventilator. The compliance of the respiratory system was calculated as V_T/(plateau pressure of the respiratory system − PEEP).

Before and after surgery, pulmonary functional tests were performed at the bedside by using a spirometer (FERRARIS PiKo-6 FEV1/FEV6 METER; Pulmonary Data Services, Inc., Louisville, KY) while the patient was in a seated, comfortable position. A clip was placed over the nose and the patient breathed through the mouth into a tube connected to the spirometer. First the patient breathed in deeply, and then exhaled as quickly and forcefully as possible into the tube. The patient repeated this test three times and the best of the three results was considered to be the measure of lung function. The forced vital capacity and the forced expiratory vital capacity in 1 s were measured whereas the ratio between the forced vital capacity and the forced expiratory vital capacity in 1 s (forced expiratory vital capacity in 1 s/forced vital capacity) was calculated by the internal algorithm of the spirometer. We also calculated the predicted values of pulmonary functional tests according to Quanjer et al.²⁹ 

Arterial blood gas analysis (Copenhagen abl 700 series; Radiometer, Brønshøj, Denmark) and peripheral oxygen saturation were measured in sitting position in room air, after 10min of adaptation. After surgery, if the patient was using a Venturi oxygen mask (Breathing Solutions, Castelbolognese, Italy). the mask was removed. If peripheral oxygen saturation dropped below 88% during the 10min of adaptation, the maneuver was stopped and arterial blood gas analysis immediately obtained.

Pre- and postoperative dyspnea, cough, presence of secretions, abdominal and thoracic pain were measured by means of specifically targeted VAS. VAS was obtained by an attending physician not involved in the study. Patients were asked to report their level of comfort by pointing to a horizontal line, 100mm in length, anchored by word descriptor at each end, after answering one of the following questions: “How is your sensation of dyspnea?”, “How severe was your cough today?”, “How is your level of pain?”^30,31  The VAS (in millimeters) was determined by measuring from the left-hand end of the line to the point that the patient or the physician marked.

Pre- and postoperative (day 1 and day 3) chest x-ray, performed at the bedside, was examined in a blinded way by an independent specialist in radiology, who was not involved in the study. Pathological chest x-ray was defined as the presence of at least one of the following: atelectasis, pleural effusions, or other chest radiological alterations.

The modified Clinical Pulmonary Infection Score (mCPIS) was calculated by a modified original score as described by Pelosi et al.³²  Patients were also compared before and after surgery for the following parameters: the Glasgow Coma Scale for nervous central system, the mean arterial pressure or administration of vasopressors required for cardiovascular system, the serum aspartate transaminase, alanine transaminase, and bilirubin for liver, the prothrombin time and platelets for coagulation, and the serum creatinine for renal system.³³ 

During the intraoperative period (after induction of anesthesia, during the surgery, and before extubation) the following data were collected: airway pressures, arterial pressure, compliance of the respiratory system, peripheral oxygen saturation, and end-tidal fractions of carbon dioxide. Intraoperative fluids requirement and erythrocytes administration were recorded, as well as blood losses and urine output. Intraoperative complications were recorded during anesthesia, and defined as follows: peripheral oxygen saturation less than 90% and/or end-tidal fractions of carbon dioxide more than 45 mmHg for more than 1min, need to change the ventilation setting (tidal volume and/or respiratory rate), heart rate more than 100 beats/min or less than 60 beats/min, systolic arterial pressure more than 150 mmHg or less than 90 mmHg, need for vasoactive drugs. During RMs, noninvasive blood pressure measurements were performed by setting the monitor in continuous mode.

Preoperatively the following measurements were obtained: peripheral oxygen saturation and arterial blood gas analysis in air, pulmonary functional tests, tympanic temperature, VAS for abdominal and thoracic pain, dyspnea, cough, presence of secretions, laboratory tests for organ function, chest x-ray, and the mCPIS.

The same measurements were performed on postoperative days 1, 3, and 5 whereas the chest x-ray and the mCPIS were calculated only on postoperative days 1 and 3. Pulmonary complications were defined with Celli score³⁴  calculated postoperatively on days 1, 3, and 5.

Pulmonary complications were defined as the development of three or more of six new findings: cough, increased secretions, dyspnea, chest pain, temperature greater than 38°C, and pulse rate more than 100 beats/min.³⁴  Surgical complications were recorded and patients were followed up until hospital discharge or death. Wound infection was defined according to Horan et al.³⁵  criteria: infection within 30 days after the operative procedure with at least purulent drainage from the superficial incision, organisms isolated from an aseptically obtained culture of fluid or tissue from the superficial incision, with pain or tenderness, localized swelling, redness or heat.

Our hypothesis was that intraoperative lung-protective ventilation could protect against postoperative pulmonary and extrapulmonary complications. The primary endpoints were the changes in mCPIS in the postoperative period. The secondary endpoints were the changes in arterial oxygenation and peripheral oxygen saturation in air, the pulmonary functional tests, and the rate of complications during recruitment, anesthesia, as well as in the postoperative period.

The sample size was based on data previously published in the literature on oxygenation and chest x-ray alterations postoperatively, according to Hedenstierna and Edmark⁵  and Choi et al.,¹⁰  but a formal a priori sample size calculation was not conducted.

The normality of the distribution was tested with the D’Agostino-Pearson test. Data are given as mean ± SD or median and interquartile range (25–75%), as appropriate. Comparisons of normally distributed variables were performed with paired or unpaired t tests as appropriate, whereas the Mann–Whitney and the Wilcoxon tests were used for other variables. Comparisons of two or more proportions were conducted with the chi-square test; the Fisher exact test was used for small frequencies. The major outcome variables were tested with two-way repeated measures ANOVA (group effects), and pair-wise comparisons of each time point with baseline, adjusted according to Bonferroni correction, were conducted.

The Kaplan–Meier curve was used to analyze the length of hospital stay in groups; the log-rank test was used for the reported P value.

All tests were two-tailed and statistical significance was accepted at P value less than 0.05.

All statistical analyses were performed with MedCalc^® Version 9.3.7.0 (MedCalc Software bvba, Mariakerke, Belgium). The statistical analysis was performed independently of the team of the clinicians involved in the study, by a specialist in physics and biomedical statistics (Dr. Novario).

Five hundred twenty-seven consecutive patients, who were scheduled to undergo an elective surgical procedure of 2h or more, were screened (fig. 1). Four hundred sixty-nine patients had one or more exclusion criteria, leaving 58 patients for randomization. Two patients were excluded because of change of surgical plan to videolaparoscopic surgery. Fifty-six were randomized. One patient assigned to standard ventilation was excluded due to a severe intraoperative surgical complication leading to a modification in the ventilator settings. Fifty-five patients entered the final analysis. There were no major differences between the two groups with regard to baseline characteristics (table 1).

During anesthesia, aside from the mechanical ventilator settings, there were no statistically significant differences between groups in intraoperative peripheral oxygen saturation, arterial pressure, and fluid balance (table 2). In the majority of cases anesthesiologists used sevoflurane to maintain anesthesia (26 of 28 cases for the protective ventilation group and 26 of 27 cases for the standard ventilation group; P = 1.0). Tidal volume, PEEP, plateau pressure, and mean airway pressure of the respiratory system were higher in the protective group compared with the standard ventilation strategy group. The respiratory rate was slightly lower in the standard ventilation strategy group, whereas the end-tidal carbon dioxide partial pressure was slightly higher in the protective ventilation strategy group.

Overall intraoperative complications, including those during RMs, were comparable between the two study groups (table 3). During RMs in eight patients systolic arterial pressure decreased less than 90 mmHg for more than 3min, whereas two patients showed a heart rate less than 60 beats/min. No other complications were observed during RMs. End-tidal carbon dioxide partial pressure was never less than 25 mmHg during RMs.

On postoperative days 1 and 3, the mCPIS was lower in the protective as compared with standard ventilation group (fig. 2 and table 4). Pulmonary complications³⁴  were higher in the standard group compared with the protective group on postoperative day 1 (7 of 26 vs. 1 of 27 respectively, P = 0.024), whereas no differences were found on day 2 (4 of 26 vs. 2 of 27 respectively, P = 0.42) and day 3 (3 of 26 vs. 1 of 25 respectively, P = 0.61).

Peripheral oxygen saturation and arterial oxygenation decreased on postoperative days 1 and 3 compared with preoperative period in standard while not in the protective ventilation strategy group (figs. 3 and 4). No differences were found in Paco₂ and pH between groups up to postoperative day 5.

Pulmonary functional tests were comparable preoperatively in both groups (table 5). Forced vital capacity and forced expiratory vital capacity in 1 s were higher in protective ventilation strategy group as compared with standard ventilation strategy group on postoperative days 1, 3, and 5. As shown in table 6, in protective compared with standard ventilation group dyspnea, secretions, and cough scores were not different between groups on postoperative days 1, 3, and 5. Chest x-ray changes were fewer on postoperative days 1 and 3 in protective compared with standard ventilation group (fig. 5 and table 7). Finally, pulmonary complications, as evaluated with the Celli score³⁴  (table 8), were lower in the protective ventilation group compared with standard ventilation group on postoperative day 1.

No differences were observed in extrapulmonary organ function pre- and postoperatively between groups (table 9). Also there were no differences in fluids administration and hydric balance during postoperative period (table 10).

The Kaplan–Meier curve of hospital length of stay did not show statistically significant differences between groups (fig. 6; P = 0.96; 95% CI, 0.5775–1.7729). On postoperative day 28, 7% of the patients in the protective ventilation strategy group as compared with 15% in the standard ventilation strategy group were still recovering in hospital (2 of 28 patients for protective ventilation group and 4 of 27 patients for the standard ventilation group, P = 0.42).

None of the patients died and all were discharged home.

In this randomized controlled trial, it was observed that in comparison to standard ventilation strategy with higher tidal volumes without PEEP and RMs, a lung-protective ventilation strategy with lower tidal volumes, 10cm H₂O PEEP, and RMs (1) improved mCPIS on postoperative days 1 and 3; (2) improved postoperative arterial oxygenation and pulmonary functional tests; and (3) showed no association with an increased incidence of intraoperative complications or nonpulmonary organ failures.

Previous randomized controlled trials showed conflicting results regarding the influence of ventilator settings on surrogate endpoints of pulmonary and systemic inflammation.^{10,11,16,17,36–40}  These studies were performed in cardiothoracic surgery,^{16,37–39,41}  in esophagectomy,³⁶  or in a nonhomogeneous groups of patients undergoing elective surgery.^10,11,40  Most of them investigated only the effects on the inflammatory response^11,40  or alveolar coaugulopathy¹⁰  during mechanical ventilation. RMs were seldom applied in most of these studies, and the level of PEEP differed among trials. The current trial differs from previous investigations with regard to (1) the combination of lower tidal volumes, PEEP and RMs—none of the previous investigations evaluated potential complications of higher PEEP levels and RMs during general anesthesia; (2) the selected population of patients undergoing elective abdominal surgical procedures, the majority lasting more than 2h with general anesthesia; notably, both abdominal surgery and longer duration of anesthesia have been reported as potential risk factors for higher incidence of postoperative pulmonary complications¹ ; (3) standardization of fluid management during surgery and in the postoperative period, as well as the physiotherapy and analgesic treatments; (4) the chosen endpoints in the postoperative period. From a recent survey in France¹⁹  the majority (>80% of patients) were managed without PEEP, so the use of ZEEP in the control group might not be considered unsafe for patients. Furthermore, in our study, 20% of patients underwent hepatic or biliary tract surgery, and some concerns might be also raised with regard to the use of higher PEEP in this specific set of patients during surgery. Thus we believe that the application of ZEEP and 9ml/kg tidal volume could have been considered as clinical practice at least in our unit before the study.

Use of higher PEEP levels is potentially associated with an increase in mean airway pressure within the respiratory system, likely promoting higher incidence of hemodynamic complications, higher fluids’ requirement, and blood losses.

We used tidal volumes of 9ml/kg in the control arm for the following reasons: first, this size of tidal volumes was used as a standard in our institution. Second, previous studies suggested that larger tidal volumes of, e.g., 12ml/kg could cause additional lung injury.^9,42 

We found that the use of higher PEEP levels was associated neither with major hemodynamic impairment nor with higher intraoperative requirement of fluids or blood losses. Nevertheless, use of RMs was associated with no life-threatening reductions in systolic arterial pressure and heart rate, but no other complications were observed during RMs.

In the current study we did not use a sustained inflation, as commonly suggested. In fact, sustained inflation might be associated with more deleterious hemodynamic effects. We used a modified RMs by allowing a progressive increase in tidal volumes, which may have promoted less negative hemodynamic impairment.

In the current trial, our aim was not to investigate major postoperative pulmonary complications, but the effects of intraoperative ventilation strategies on relevant clinical parameters associated with alterations in the pulmonary function. We evaluated (1) arterial oxygenation and peripheral oxygen saturation in air, (2) pulmonary functional tests, (3) changes in dyspnea, cough, and secretions, (4) chest x-ray, abnormalities, including atelectasis and pleural effusions, and (5) the mCPIS.

Oxygenation was studied while the patients were breathing in air, in seated position, after 10min of adaptation. This allows avoiding any possible influence of different inspiratory oxygen fractions on the arterial oxygenation. The pulmonary functional tests were studied by using a spirometer while the patient was in seated position, allowing reproducible measurements. In our study, we standardized the methods of performing the chest x-ray at the bedside, and the chest x-ray was evaluated by a specialist in radiology, blinded to the study. We used a modified CPIS score, as previously proposed,³²  as an indicator of pulmonary complications.

In our trial we found that the protective ventilation group was associated with a statistically significant reduction in chest x-ray alterations and mCPIS compared with the standard ventilation group. Especially, chest x-ray showed a clinically significant increase in postoperative atelectasis on both days 1 and 3 in the standard ventilation group. This suggests that patients ventilated with lower tidal volume and no PEEP in our trial could have gross atelectasis and potential peripheral airway injury, caused by tidal airway closure, which was maintained in the postoperative period. Indeed, a recent experimental study in open-chest rabbits, demonstrated that mechanical ventilation with tidal volumes of 8–12ml/kg and no PEEP causes permanent mechanical alterations and histologic damage to peripheral airways and inflammation in noninjured lungs.⁴³  During general anesthesia, atelectasis is potentiated by anesthesia and muscle relaxants altering diaphragmatic position.⁵  Also, tidal airway closure can occur and cause peripheral airway injury. This may be a common but unrecognized complication in patients undergoing general anesthesia.⁴⁴  Cyclic opening and closing from ZEEP leads to greater increases in bronchoalveolar lavage cytokines than atelectasis.⁴⁵ 

Furthermore, these morphological alterations were associated with a marked improvement in arterial oxygenation in air as well as better pulmonary functional tests in the protective ventilation strategy group. Interestingly, in the protective ventilation group compared with the standard ventilation group, on postoperative day 1, we observed a lower percentage of patients with an arterial oxygenation less than 60 mmHg (3.6 vs. 18.5% respectively). Similarly, we observed a lower percentage of patients with peripheral oxygen saturation levels less than 90% in air (3.6 vs. 19.2% respectively, P = 0.1). Our results suggest therefore that intraoperative protective ventilation strategy may play a relevant role to minimize potential oxygen desaturation in the postoperative period.

Although the study was intentionally not powered for outcome, the improvement in these clinical variables was not associated with a statistically significant reduction in the hospital length of stay in the protective ventilation group. However, in our study, on postoperative day 14, 20% of the patients in the protective ventilation strategy group as compared with 40% in the standard ventilation strategy group were in hospital.

The current trial suffers of some potential limitations, which need to be addressed. First, our study does not allow to differentiate the effects of lower tidal volumes from those of higher PEEP levels. We deliberately chose to combine lower tidal volumes with higher PEEP levels as well as RMs to identify a ventilation strategy aimed at keeping the lung open during general anesthesia for surgery, a strategy that might have potential benefits in the postoperative period. Second, we performed the RMs until plateau pressure reached 30cm H₂O for three breaths, but we had not systematically recorded the peak pressures reached during the RM. Third, we did not focus on the effects of ventilation strategies on major postoperative pulmonary complications. The mCPIS includes the evaluation of the chest x-ray. It has been reported that chest x-ray may underestimate the presence of atelectasis and lung morphology alterations as compared with computed tomography.⁴⁶  However, computed tomography is not easy to obtain in this group of patients for technical and ethical reasons. Fourth, our data cannot be directly translated to other categories of patients and types of surgery. Fifth, with regard to anesthesia conduction, several studies suggest that volatile anesthetics could modify the inflammatory process in general and we cannot exclude this as having affected our outcomes. Moreover, a recent study shows an induced inflammatory response from halogenated anesthetics versus propofol, which seems to preserve antiinflammatory and antioxidant defences during mechanical ventilation in pigs model, preventing the emergence of apoptosis.⁴⁷  In contrast, several other studies show that halogenated anesthesia attenuates the inflammatory response.^48,49  Another recent study, concerning the inflammatory response in major abdominal surgery, reveals that there are no statistically significant differences between total intravenous anesthesia and inhalational anesthesia.⁵⁰  Finally, the baseline pulmonary functional test results were a bit lower than predicted, likely due to the method of measurement. However, we believe that this did not affect the interpretation of our data, because we investigated the evolution of pulmonary functional tests with time in the two study groups.

In conclusion, we showed that a protective ventilation strategy with lower tidal volumes, PEEP, and RMs during anesthesia improved the respiratory function in the postoperative period after abdominal nonlaparoscopic surgery and reduced the clinical signs of pulmonary infection during the first 5 days after open abdominal surgery. Larger trials are warranted to determine whether intraoperative protective mechanical ventilation improves major outcome parameters.