Effect of Stepwise Lung Recruitment Maneuver on Oxygenation, Lung Mechanics and Lung Injury Biomarkers During Lung Resection Surgery: A Prospective Randomized Controlled Single Blinded Study

ABSTRACT Background One lung ventilation for lung resection surgery may affect oxygenation, impair lung mechanics and increase the incidence of acute lung injury. Lung recruitment maneuver may improve these conditions. Settings and Design This study was a prospective randomized controlled single blinded clinical trial. Methods Forty patients scheduled for elective lung resection surgeries using one lung ventilation under general anaesthesia were randomly allocated into two groups. Control (C) group was subjected to conventional mechanical ventilation of tidal volume of 8 ml/kg and 5 ml/kg for total lung ventilation (TLV) and one lung ventilation (OLV), respectively, and PEEP of 5 cmH2O and (LR) group which was subjected to stepwise lung recruitment twice: The first lung recruitment (LR1) was performed after (OLV) to the dependent lung, while the second lung recruitment (LR2) was performed after resuming (TLV). Arterial blood gases, lung mechanics and serum tumor necrosis factor alpha were recorded at multiple time points during the study. The proposal and raw data were registered on PACTR as PACTR202001518687696. Results There was a significant decrease in PaO2/FIO2 during OLV baseline in comparison to TLV baseline (P = < 0.001*) in both groups. While, in LR group, there was a significant increase in PaO2/FIO2 and dynamic compliance during OLV-20 min after recruitment (P = <0.001*), during TLV-20 min after recruitment (P = < 0.001*) and during TLV-end (P = <0.001*). TNF-⍺ level was significantly higher in control group 1 h after surgery and 24 h postoperative. Conclusion Stepwise lung recruitment maneuver improved oxygenation parameters and lung mechanics during lung resection surgery and decreased the expression of lung injury biomarkers. Abbreviations: OLV: one lung ventilation; TLV: total lung ventilation; RM: recruitment maneuver; LR: lung recruitment; TNF-⍺: tumor necrosis factor alpha; VCV: volume controlled ventilation; PCV: pressure controlled ventilation; VT: tidal volume; FIO2: fraction of inspired oxygen; ALI: acute lung injury; ppoFEV1%: predicted postoperative forced expiratory volume in 1 s; BAL: Bronchoalveolar lavage.


Introduction
Lung resection surgery is one of the most commonly performed thoracic surgical procedures [1]. One-lung ventilation (OLV) is regularly used to facilitate surgical exposure and to isolate and protect the lungs throughout the surgery [2].
An inevitable intrapulmonary shunt during OLV arises as a result of non-dependent lung collapse and increased atelectatic regions in the dependent lung, resulting in hypoxemia and impairment of gas exchange and activation of hypoxic pulmonary vasoconstriction aiming to reduce the shunt [3].
During OLV, a variety of inflammatory cytokines are released in response to a variety of pathophysiological disturbances such as high airway pressures, elevated pulmonary capillary pressure and pulmonary alveolar resistance, compression of alveolar vessels, alveolar cell stretch, over-distension and cyclic alveolar collapse, ventilation/perfusion mismatch with shunt, all of which can trigger local or systemic inflammatory responses [4,5].
Lung protective ventilation and lung recruitment maneuvers are utilized during OLV to limit the severity of acute lung injury and postoperative pulmonary complications [6]. Although lung-protective ventilation can help to minimize acute lung injury, reduced tidal volume can lead to alveolar collapse, so lung recruitment maneuver is regarded the best method for increasing oxygenation, resolving ventilation-perfusion mismatch, improving lung mechanics and reversing atelectasis during one lung ventilation [7]. Therefore, the primary aim of the present study was to study the effect of stepwise lung recruitment maneuver in patients undergoing lung resection surgery on oxygenation parameters, while the secondary aim to evaluate the effect of recruitment on lung mechanics and lung injury biomarker represented by tumor necrosis factor alpha (TNF-⍺).

Patients and methods
Agreement of the Ethical Committee of Alexandria Faculty of Medicine (IRB-NO: 00007555-FWA-NO: 00018699) and trial registry on PACTR as PACTR202001518687696 and a written informed consent from all the patients were obtained. The present study was carried out in Alexandria University hospitals on 40 adult patients, American Society of Anaesthesiologists (ASA) physical status class II-III with age ranged from 20-60 years scheduled for elective lung resection surgeries using one lung ventilation under general anaesthesia. The study was carried out between September 2018 till June 2021.
The patients were randomly allocated using computer generated sequence numbers, into two equal groups (20 patients in each group) using closed envelope technique in a single blinded manner, where patients did not know in which group, they were allocated. Group (C) control group in which conventional mechanical ventilation was performed without performing lung recruitment maneuver and Group (LR) lung recruitment group was subjected to stepwise lung recruitment maneuver.
Preoperative evaluation was carried, covering proper history taking clinical examination, routine laboratory investigations, arterial blood gases, ECG, echocardiography, chest x ray, CT chest, pulmonary function tests and serum tumor necrosis factor alpha (TNF-⍺).
On arrival to the operating theatre, a peripheral venous catheter was inserted, and all patients were monitored by standard monitoring using multichannel monitor (AVANCE CS 2 , GE Healthcare, Madison, USA) to display electrocardiogram (ECG) for heart rate (beats/ min) and rhythm, oxygen saturation (SpO2%), noninvasive arterial blood pressure (NIBP) (mmHg) and invasive arterial blood pressure (mmHg) through radial artery catheter after performing modified Allen's test [9]. End tidal CO 2 (ETCO 2 ) (mmHg) was connected after intubation. A nerve stimulator in AVANCE GE for neuromuscular blockade monitoring and entropy were additionally applied.
Thoracic epidural catheter was inserted at the thoracic interspaces (T5-T6) or (T6-T7) under midazolam sedation of 0.03-0.05 mg/kg. 0.1 ml/kg bolus of epidural infusion of 0.125% of plain bupivacaine in combination with fentanyl (2 microgram/ml) was delivered before surgical incision followed by 0.1-0.2 ml/kg/hour to keep arterial blood pressure and heart rate within 20% of the pre-induction values.
General anaesthesia was induced with fentanyl 2 µg/kg, propofol 1.5-2 mg/kg and cisatracurium 0.15-0.2 mg/kg intravenously, followed by tracheal intubation using left sided double-lumen endobronchial tube (DLT) size 37-39 French, in which its correct position in supine and lateral position was confirmed using a fiberoptic bronchoscope. Anaesthesia was maintained with isoflurane 1-2% to maintain entropy between 40 and 60 and cisatracurium increments of 0.015-0.02 mg/kg intravenously guided by train of four (TOF) ratio.
Initially, in both studied groups, two lung ventilation with volume controlled ventilation mode (VCV) was performed using tidal volume (8 ml/kg) and initial PEEP of 5 cmH 2 O, the inspiratory to expiratory ratio was 1:2 and the respiratory rate was adjusted to keep end tidal CO 2 between 35 and 45 mmHg with maximum airway pressure (P max ) of 35 cmH 2 O, Followed by One lung ventilation (OLV) to the dependent lung with (VCV), with tidal volume of (5 ml/kg) and PEEP of 5 cmH 2 O; with same the inspiratory to expiratory ratio of 1:2 and the same respiratory rate to keep end tidal CO 2 between 35 and 45 mmHg with maximum airway pressure (P max ) of 35 cmH 2 O in LR group and 30 cmH 2 O in control group and inspired fraction of oxygen (FIO 2 ) was adjusted to maintain oxygen saturation above 95%.
Group (LR) lung recruitment group was subjected to stepwise Lung recruitment twice: The first was performed after the start of one lung ventilation (OLV) in lateral position to the dependent lung, while the second was performed after resuming total lung ventilation (TLV).
Stepwise lung recruitment maneuver was performed with pressure controlled ventilation (PCV), maximum airway pressure (P max ) was increased to 38 cmH 2 O during recruitment with inspiratory to expiratory ratio of 1:2 and respiratory rate of 12 cycles/min. PEEP was increased in increments of 5 cm H 2 O for 3 respiratory cycles in a stepwise manner till reaching 20 cmH 2 O lasting for 6 respiratory cycles with a constant driving pressure of 15 cmH 2 O above the PEEP. The dynamic compliance, oxygen saturation (SpO 2 %) and blood pressure were monitored at each step. This was followed by progressive reduction of PEEP step by step with decrements of 2 cmH 2 O where each step lasted for 3 respiratory cycles, until the maximum dynamic compliance is obtained. If signs of overdistension appeared (dynamic compliance C dyn decreased or decrease in oxygen saturation or severe hypotension) before reaching 20 cmH 2 O PEEP, this patient was excluded from the study. During decremental decrease in PEEP, the point at which dynamic compliance C dyn decreased or oxygen saturation decreased by ≥ 1% from the maximum reading observed. This was defined as the derecruitment point and PEEP was returned to the previous step, which was defined as optimal PEEP which was kept all over the surgical period. After lung recruitment maneuver, VCV was resumed with setting the optimal PEEP. Tidal volume and the respiratory rate were modified to keep end tidal CO 2 within the target range. Maximum airway pressure (P max ) was adjusted to 35 cmH 2 O.
In both groups after resuming total lung ventilation and before the second lung recruitment maneuver in the study group, routine manual hyperventilation to both lungs will be done aiming to inflate the nondependent lung in all patients.
At the end of surgery, after reversal of neuromuscular blockade with neostigmine (0.04-0.08 mg/ kg) and atropine (0.01-0.02 mg/ kg), and after recovery from anaesthesia, all patients were extubated and transferred to post-anaesthesia care unit (PACU) with epidural infusion of dose 0.01 ml/kg/hour of 0.125% of plain bupivacaine in combination with Fentanyl 2 microgram/ml. The epidural catheter was removed on the second postoperative day.

Statistical analysis
The sample size was calculated using G.Power software. It was estimated that a total number of 40 patients were needed to detect at least 10% difference in partial arterial oxygen tension after lung recruitment using alpha error = 5% with 80% power [7].
Data were fed to the computer and analyzed using IBM SPSS software package version 20.0. (Armonk, NY: IBM Corp). The Kolmogorov-Smirnov was used to verify the normality of distribution of variables, and comparisons between groups for categorical variables were assessed using Chisquare test (Fisher Exact). Student t-test was used to compare two groups for normally distributed quantitative variables. While ANOVA with repeated measures using Post Hoc Test (adjusted Bonferroni). Significance of the obtained results was judged at the 5% level.

Results
Forty-nine subjects were assessed for eligibility to participate in this study, 43 patients were enrolled in the study, and 6 patients were excluded for not meeting the inclusion criteria. 3 patients discontinued the intervention in the LR group due to overdistension during the recruitment maneuver. Statistical analysis was done on 40 patients (20 patient in each group) as represented in Figure 1.
Regarding demographic data, duration of the surgery and duration of OLV, there was no statistically significant difference between the two groups as represented in Table 1.
Regarding lung mechanics as represented in Table 3, there was a significant increase in dynamic compliance in LR group during OLV-20 min after recruitment (P = <0.001*), TLV-20 min after recruitment (P = <0.001*) and TLV-end (P = <0.001*) on comparing both groups. Also, the mean airway pressure was increased significantly in LR group during OLV-20 min (P = <0.001*), TLV-20 min (P = <0.001*) and TLV-end (P = <0.001*), while there was no significant difference in peak pressure and plateau pressure between the two groups at the studied times.
Regarding, serum TNF-α, on comparing both groups, there was a significant increase in TNF ⍺ in the control group 1 h after surgery (P = <0.001*) and 24 h postoperative (P = <0.001*) as represented in Table 4.

Discussion
The improvement in oxygenation parameters after LR in the present study may be related to a decrease in intrapulmonary shunt by application of constant end expiratory pressure which cause less V/Q mismatch due to homogeneous distribution of inspired gas during pressure-controlled recruitment maneuver, reversal of atelectasis.
In agreement with our study, Ferrando et al. [7] found that stepwise recruitment applied during one lung ventilation had improved the oxygenation parameters, as when compared to 5 cmH 2   was significantly greater with individualized PEEP at the end of one-lung ventilation. Also, Unzueta et al. [3] found that recruitment of both lungs before instituting OLV, improved arterial oxygenation in which there was a significant increase in PaO 2 after lung recruitment.
In contrast to our results, Roze et al. [11] found that lowering VT and raising PEEP with the same low plateau pressure (VT at 5 ml/kg and a PEEP level that kept a constant plateau pressure recorded before), decreased oxygenation when compared with higher VT and lower PEEP (VT at 8 ml/ kg and a PEEP of 5 cm H 2 O), during OLV and their finding might be explained that the increased mean alveolar pressure in the dependent lung causes redistribution of pulmonary blood flow from overdistended lung units in the dependent lung to the non-dependent lung, in contrast to the present study in which the individualized optimal PEEP was used.
As regards lung mechanics, in the present study, there was a significant increase in dynamic compliance in LR group after applying stepwise lung recruitment during OLV and TLV till the end of surgery. This could be explained that stepwise lung recruitment with pressurecontrolled ventilation used a decelerating flow pattern  that resulted in a more homogeneous distribution of the tidal volume and improvement of the lung compliance due to recruitment of lung areas that were poorly ventilated and maintenance of individualized optimal PEEP which could keep the alveoli open and prevent alveolar recollapse and also minimizes overdistension.
In agreement with our results, Rauseo et al. [12] found that the application of an open lung ventilation strategy consisting of cycling recruitment followed by a decremental PEEP titration during OLV significantly improved lung compliance and compliance of the respiratory system. Also Ferrando et al. [7] found that recruitment applied during one lung ventilation had increased the static compliance significantly suggesting a constant end-expiratory lung volume. Moreover, Miura et al. [13] found that respiratory compliance was significantly increased after recruitment during OLV.
In contrast to our results, Park et al. [14] found no differences in static or dynamic compliance between groups with and without preemptive alveolar recruitment strategy before OLV. And this could be attributed to stoppage of lung recruitment before OLV and application of standardized fixed PEEP of 5 cmH 2 O during OLV with partial loss of end expiratory lung volume.
In the present study, there was a significant increase in TNF ⍺ in the control group 1 h after surgery and 24 h postoperative in relation to LR group. Lung recruitment in this study showed   [15]. Moreover, the use of stepwise recruitment maneuvers with gradual increase in peak pressure and PEEP was less detrimental to the alveolar capillary membrane, because the pressures were built up slowly [16], in contrast to the sustained inflation, which appeared to be more harmful [17].
In agreement with our results, Kim et al. [18] found that the level of TNF-α in the bronchoalveolar lavage (BAL) fluid of dependent lungs was significantly greater in the protective ventilation group than in the protective ventilation recruitment maneuver group when recruitment was applied 10 minutes after initiation of OLV. They demonstrated that recruitment and protective ventilation reduced inflammatory responses in the ventilated lung and serum compared to protective ventilation alone. Also, Schilling et al. [19] demonstrated that repetitive RMs during either OLV or two lung ventilation did not induce any significant pro-inflammatory responses. They concluded that it is OLV not lung recruitment that is detrimental and that repetitive alveolar recruitment maneuvers and bronchoscopic procedures did not increase the alveolar cytokine release.
In contrast to our results, Leite et al. [20] found that lung re-expansion of 30 cmH 2 O during 30 seconds after OLV, caused cytokine release, protein extravasation and neutrophil infiltration into the alveolus. As elevated levels of interleukin (IL)-6, IL-1 and tumor necrosis factor-alpha in BAL fluid were detected, and elevated levels of IL-6 and IL-10 in serum were found. And this might be related to the use of sustained inflation maneuver not the stepwise maneuver.

Conclusion
Stepwise lung recruitment maneuver improved oxygenation parameters and lung mechanics during lung resection surgery and decreased the expression of lung injury biomarkers.

Limitations
The main limitation in the present study was small sample size and lack of patient follow-up to assess long-term effects of lung recruitment and its impact on hospital stay.