Key words : Lung compliance, Respiratory complications, Recruitment maneuver, Surgery

Patients with advanced hepatic disease may present before transplant surgery with peripheral atelectasis as a result of ascites or pleural effusion, which may be aggravated by prolonged surgical retraction.^1,2 The alveolar recruitment maneuver (ARM) can help reexpand these collapsed lung alveoli and prevent further atelectasis. The ARM is not a new technique, and its beneficial role for improvement of oxygenation is known at various clinical conditions and during surgery.^3-6 However, few data are available about the effect of lung recruitment during and after liver surgery⁷ and specifically liver transplant. Prolonged general anesthesia and subcostal chest retraction during liver resection surgery will reduce the functional residual capacity. The combined ARM and positive end-expiratory pressure (PEEP) during liver resection was found to reexpand the peripherally collapsed alveoli and increase the lung dynamic compliance (C_dyn) without an increase in blood loss.⁷ Recruitment of the lungs of hepatic patients with end-stage liver disease undergoing liver transplant is expected to be different from that required for critically ill patients with adult respiratory distress syndrome. The liver transplant procedures have specific operative considerations, and recipients’ clinical conditions may vary from one surgical phase to another. One of the contributing factors during surgery is the need to apply extensive abdominal surgical retraction for a long duration, which can affect the aeration of the peripheral alveoli. Recruitment maneuvers and high PEEP could be harmful to the liver blood inflow if it induces low cardiac output or obstructs hepatic outflow. Lung recruitment can induce venous congestion as a result of the high central venous pressure (CVP) associated with the recruitment process.

The primary goal of this study was to recruit the lungs of recipients in a stepwise mode during liver transplant surgery guided by C_dyn and report the effects on arterial oxygenation (Pao₂) and systemic and liver graft hemodynamics. The secondary goal was to study the effect of ARM on postoperative mechanical ventilation duration and postoperative pulmonary complications (PPC) in the intensive care unit (ICU).

This study was a randomized controlled trial. Fifty-eight recipients scheduled electively for living donor liver transplant during the year 2018 were enrolled (1 excluded). Institutional Review Board approval was obtained from the Anesthesia Departments of the National Liver Institute and the Faculty of Medicine, Menoufia University, Egypt (NLI IRB 00003413-00126/2017). This trial was registered with the Pan African Clinical Trial Registry (retrospective) (PACTR201811462663708) and started on October 15, 2017, following the National Liver Institute Institutional Review Board approval, and the trial ended on October 31, 2018.

After written informed consent was obtained from the patients, we used a random number generator with sealed envelopes to allocate the patients into 2 groups. One group was subjected to alveolar recruitment maneuver (ARM) with an optimal PEEP, and the other group (control) was given standard volume-controlled ventilation with a tidal volume of 6 mL/kg, inspiratory-to-expiratory time ratio of 1:2, and a fixed PEEP of 5 cm H₂O.

Inclusion criteria included consecutive recipients with hepatitis C between 18 and 60 years old (both sexes) with a Model for End-Stage Liver Disease score (MELD) of no more than 24 were enrolled. Exclusion criteria applied to patients with preexisting significant pulmonary disease and cardiac dysfunction such as cirrhotic cardiomyopathy and acute liver failure and those in need of retransplant. Recipients with pulmonary hypertension, body mass index (BMI, calculated as weight in kilograms divided by height in meters squared) >40 and refusal to participate were excluded.

Anesthesia monitoring included basic monitors, fraction of inspired oxygen concentration (Fio₂), anesthetic agent fractioned inspired and expired, urine output (mL/h), continuous invasive arterial pressure (IBP; mm Hg), and CVP (cm H₂O). Transesophageal Doppler (TED) probe (Cardio Q, Deltex Medical) was inserted for corrected flow time (FTc)-guided perioperative fluid optimization and cardiac output (L/min) monitoring. The TED probe was inserted orally and positioned approximately 35 to 40 cm from the incisor teeth.

Anesthesia depth was monitored with a SedLine brain function monitor (Masimo), with a processed electroencephalogram parameter (patient state index, PSi).

Preoxygenation was with O2/air mixture (Fio₂ = 0.8) for 3 to 5 minutes. The PSi-guided general anesthesia was induced with propofol, fentanyl, and rocuronium. All patients were initially ventilated with a volume-control mode. Anesthesia was maintained with desflurane in O2/air mixture (Fio₂ = 0.4), and end-tidal desflurane concentration was adjusted to keep PSi between 25 and 50. Mechanical ventilation was performed via a semi-closed system (Macquet Flow-i anesthesia machine), adjusted to keep arterial oxygen saturation >95% and end-tidal CO₂ between 35 and 45 mm Hg. Boluses of rocuronium were guided by the train-of-four (TOF) nerve stimulation, and ARM was only started when the patient was fully relaxed.

Stepwise alveolar recruitment maneuver protocol

Alveolar recruitment maneuver was performed after induction and application of surgical retractors. Recruitment was repeated again on admission to the ICU after stabilization of the recipients and whenever disconnected.

The recruitment maneuver, only performed with pressure-controlled ventilation (inspiratory-to-expiratory time ratio, 2:1), was monitored by C_dyn, as follows. A step-by-step increase in inspiratory pressure (IP) is induced with an increase in PEEP every 2 respiratory cycles with a constant driving pressure (IP-PEEP) of 15 cm H₂O until a peak IP of 35 and a PEEP of 20 cm H₂O is reached. This is followed by sustained inflation with an increase in peak IP to 40 cm H₂O with a PEEP of 20 cm H₂O for 10 seconds.⁸ A progressive reduction of the pressures is followed until the optimal PEEP is attained, corresponding to 2 cm H₂O more than the collapse point.⁹ The collapse point corresponds to the point at which lung C_dyn decreases abruptly.

Finally, each recipient is switched to lung-protective volume-controlled ventilation with a minimum tidal volume of 6 to 7 mL/kg (ideal body weight) for an end-tidal CO₂ of 45 mm Hg with the measured optimal PEEP. The maximum airway pressure is set not to exceed 30 cm H₂O and driving pressure not to exceed 15 cm H₂O for both groups.

Crystalloid fluids were given as a background infusion of Ringer acetate of 6 mL/kg/h. Additional fluids, crystalloid and colloid (albumin, 5%), were infused guided by TED (Cardio Q), while blood products transfusion was guided by rotational thromboelastometry.^10,11 Recruitment could be aborted any time if hemodynamic instability occurred. Hypotension (mean IBP <60 mm Hg) during recruitment was initially treated with fluid load (Ringer acetate and/or 5% albumin) if Doppler FTc indicated hypovolemia (FTc < 350 ms); otherwise, initially 5 mg ephedrine and/or epinephrine (10 μg) was injected to restore hemodynamic stability during the postreperfusion hypotension. Noradrenaline was administered as a continuous catecholamine intra­venous infusion when the mean arterial pressure persisted to be <60 mm Hg despite adequate volume resuscitation, adequate cardiac output, and incre­mental boluses of ephedrine and/or epinephrine. All hemodynamic monitors were observed continuously during the trial, particularly cardiac output and systemic vascular resistance (SVR), and any dysfunction was treated accordingly.

Measurement study times were as follows: baseline; postinduction recruitment (20 minutes after general anesthesia induction); postretraction recruit­ment (20 minutes after surgical retractor application); anhepatic phase; mid-dissection phase; reperfusion (20 minutes after reperfusion of the new liver graft); and ICU recruitment (20 minutes after ICU admission and stabilization).

Demographics included age (years), sex, weight (kg), BMI, and MELD score. Ventilation related measurements were Pao₂ (mm Hg), Pao₂/Fio₂, optimal PEEP (cm H₂O), and C_dyn (mL/cm H₂O). Lung dynamic compliance was calculated automatically on a breath-by-breath basis using the anesthesia machine (Maquet) and the ICU ventilator (Maquet, Servo-I). Preoperative pulmonary function test data were checked, ie, forced expiratory volume at first second (FEV₁1, in L) and the ratio of FEV₁1 to forced vital capacity (FEV₁1/FVC). The duration of mechanical supportive ventilation was described as from admission to ICU until extubation. Extubation was only performed when recipients were hemodynamically stable and with no acid/base disturbances or graft vascular flow abnormalities.

Systemic hemodynamics were heart rate (beats/min), IBP (mm Hg), CVP (cm H₂O), and the TED-derived variables of cardiac output (L/min), FTc (ms), and systemic vascular resistance (SVR, dyn/s.cm^-5). For liver graft monitoring during ARM, hepatic vein flow and portal vein Doppler study parameters (cm/s) were measured on admission to ICU and after weaning from ventilation. Postoperative blood levels of aspartate aminotransferase (AST, IU/L), alanine aminotransferase (ALT, IU/L), lactate (mmoL/L), and international normalized ratio were measured on postoperative day 1 (POD1) and POD3. Operative data included the need for catecholamine and blood transfusion, graft/body weight ratio, and warm and cold ischemia time. Postoperative pulmonary complication was defined as the number of patients with oxygen desaturation (<94%), chest infection, respiratory failure, and pleural effusion. The pleural effusion volume was diagnosed by ultrasonography for volumes more than 20 mL.12 Durations of mechanical ventilation and ICU stay were reported, and the 3-month mortality rate was noted. The primary endpoint was defined by changes in C_dyn, Pao₂, and hemodynamics represented in cardiac output, IBP and CVP, and graft portal vein and hepatic vein flows. The secondary endpoints were changes in postoperative AST (IU/L), ALT (IU/L), lactate (mmol/L), international normalized ratio, and PPC.

The minimal sample size was calculated based on the study by Futier and colleagues (2013).⁶ Compared with a practice of nonprotective mechanical ventilation, they concluded that the use of a lung-protective ventilation strategy in intermediate-risk and high-risk patients undergoing major abdominal surgery was associated with improved clinical outcomes and reduced health care utilization. Based on their finding that respiratory system compliance (mL/cm H₂O) at the end of surgery was 45.1 ± 12.9 in the nonprotective ventilation group versus 55.2 ± 26.7 in lung-protective ventilation group, we concluded that a sample size of 28 patients per group (number of groups = 2) (total sample size needed = 56) is the sufficient required sample to detect a standardized effect size of 0.6769 (minimum difference in mean respiratory system compliance [mL/cm H₂O] divided by pooled variance) change in the primary outcome, as statistically significant with 80% power and at a significance level of 95% (alpha error probability = 0.05). The sample size per group does not need to be increased to control for attrition bias. The sample size was calculated using G Power version 3.1.9.2. The allocation sequence was generated using the permuted block randomization technique, and the block size was variable. Allocation sequence/code was concealed from the person allocating the participants to the intervention arms using sealed opaque envelopes. Masking/blinding was employed with the participants and with outcome assessors who were blinded to the group allocation of patients; however, the anesthesiologist and surgeons were not blinded. Surgeons were informed to report any surgical difficulty during recruitment.

Data were collected and entered into the computer with SPSS software (version 21) for statistical analyses.^13,14 Data were entered as numerical or categorical, as appropriate. The Kolmogorov-Smirnov test of normality revealed significance in the distribution of some of the variables, so the nonparametric statistics were adopted. An alpha level was set to 5% with a significance level of 95%, and a beta error was accepted up to 20% with a power of the study of 80%.

Fifty-eight recipients were enrolled and randomized into ARM (n = 28) and control (n = 29) groups. The Consolidated Standards of Reporting Trials (CONSORT) flow chart is presented in Figure 1. Data in text, figures, and tables are presented as median values with interquartile ranges (IQR). The preope­rative recipients’ demographics were comparable, including sex (Table 1). The following data are presented as median values (IQR) for ARM versus control groups, respectively. Preoperative lung function tests: FEV₁1 (in L) and FEV₁1/FVC ratio (%) were 2.90 (2.68-3.20) and 79.00 (75.9-82.80) versus 3.10 (2.70-3.50) and 81.00 (79.00-87.00) in the control group (P = .508 and P = .052), respectively. Cold ischemia duration (minutes) was 60.0 (50.0-67.5) versus 60.0 (45.0-70.0) (P = .61), warm ischemia duration (minutes) was 50.0 (40.0-60.0) versus 40.0 (35.0-50.0) (P = .03), and median graft-to-body weight ratio was 0.9 (0.8-1.0) versus 1.0 (0.9-1.06) (P = .03), respectively. Comparable noradrenaline support (P = .134) and infused crystalloids (Ringer acetate) were observed (P > .05). The PEEP and C_dyn increased significantly in ARM versus control (P < .001) as presented in Table 2 and Figure 2, respectively. The arterial blood Pao₂ (mm Hg) as a result increased with the ARM group versus control (Figure 3). The number of recipients with PPC complications was reduced in the ARM group versus the control group (0/28 vs 8/29; 27.59%; P = .001).

Two recipients in the control group developed chest infection, and 1 required reintubation to support respiratory functions, whereas 5/8 recipients with PPC in the control group experienced mild to moderate pleural effusion.

There was no reduction in the ICU ventilation period in the ARM versus the control group (7 [4-8] vs 7 [6-9] hours; P = .08) or in ICU stay (5 [5-5] vs 6 [5-7] days; P = .89). Central venous pressure measurements were affected by the high ARM PEEP (Figure 4), but Doppler FTc readings were not (Figure 5). A negligible correlation existed between CVP and FTc (n = 401 paired readings, Kendall tau correlation = 0.053; P = .135). Invasive arterial pressure, cardiac output, and SVR variables were minimally affected by the recruitment process and the presence of high PEEP (Figures 6, 7, and 8). The liver graft portal vein and hepatic vein outflows were not affected by the high PEEP and CVP associated with the ARM (Table 3). The number of patients in need of blood products was comparable between the groups (P > .05) (Table 4). On POD1 and POD3, the international normalized ratio and blood levels of AST, ALT, and lactate were not different between the groups (P > .05) (Table 5). A significant reduction in the above laboratory parameters was observed between POD1 and POD 3 (P < .05), as a result of the functions of the newly transplanted liver graft. There were no reported surgical difficulties during the transplant procedure, except for the observation of a bulge of the lower right diaphragm from the distended lower lobe of the right lung. This was clear during the anhepatic phase when the right diaphragm becomes visible.

This trial is one of the few to address the effects of intraoperative prophylactic lung recruitment among liver transplant recipients during transplant surgery. The present trial results demonstrated the ability of liver transplant recipients to hemodynamically tolerate the stepwise mode of lung recruitment.

Yi and colleagues were among those few teams to report as early as 2006 that lung recruitment can be safely practiced among recipients.¹⁵ They successfully recovered the lungs of recipients diagnosed with postoperative acute lung injury and enhanced their weaning process, but their work was only in the ICU and not in the operating rooms.¹⁵

The improvement in oxygenation as a result of recruitment is not new and was previously investigated at different fields of surgery by several researc­hers including Pang and colleagues¹⁶ and Weingarten and colleagues.⁴ The ARM during liver transplant surgery represents a challenge because of the signi­ficant variations in operative circumstances from one phase to another during the transplant surgical procedure itself with the expected blood loss and hemodynamic changes. In addition, recipients with end-stage liver disease can present prior to surgery at different grades of liver dysfunction and with a variable ability to tolerate hemodynamic changes particularly during bleeding or with increased airway pressures.

Liver transplant recipients with end-stage liver disease are known to have a low SVR and are peripherally vasodilated. This makes them prone to hypotension particularly at the different intra­operative phases of transplant, such as reperfusion phase and following prolonged anesthesia when recruitment must be repeated during admission to ICU. Arnal and colleagues treated patients with adult respiratory disease syndrome and observed that most of the alveolar recruitment occurs during the first 10 seconds of the sustained inflation and that the hemodynamic impairments follow these initial 10 seconds.⁸

This observation by Arnal and colleagues⁸ was taken in consideration during the design of this present trial to increase the ability of the recipients to tolerate lung recruitment. Reduction of the sustained inflation during recruitment to 10 seconds may be a contributing factor toward the hemodynamic stability noted in the results of this present trial.

One of the important lessons learned from this present trial is to respect the individual variations in the ventilation parameters among the recipients as the optimal PEEP and the C_dyn values. It is advisable to monitor these changes from one phase to another during surgery as these can change. Monitoring pressure volume loops and specifically the C_dyn changes were preferred by the authors as a tool to guide the process of recruitment and to detect each recipient’s individual optimal PEEP. Pressure volume loops are available in most anesthesia machines but these require general anesthesia and full muscle relaxation when used to guide alveolar recruitment. The status of muscle relaxation was continuously monitored during this trial with a TOF muscle relaxant monitor to maximize the benefit of recruitment.

The ability of the lungs to be recruited (recruitability) and the degree of improvement in lung functions with ARM varies from individual to individual. One of the limitations of our study was application of ARM randomly to all recipients enrolled in the study group without a prior assessment of their lungs’ recruitability. The only method used in this study to assess lung functions before surgery was the set of pulmonary function tests.

Another lesson learned from this present trial is to assess the fluid volume status prior to the cessation of lung recruitment during episodes of hypotension and before depriving the recipient from the ARM benefits. Three of the recipients in the present trial were subjected to a temporary hold of ARM in order to improve the venous return. They were found to be fluid deficit, and when replaced with fluids their lungs could be successfully re-recruited without hemodynamic disturbances. The fear that graft blood supply could be affected by the associated increase in CVP during the ARM process is a theoretical risk but was not detected at any stage following transplant. In this present trial, no effect from recruitment could be reported among the newly transplanted liver graft liver enzymes or the denervated graft portal vein inflow and hepatic vein outflow. Ukere and colleagues (2017) from the University Medical Center in Hamburg similarly found that the increase in PEEP and CVP during ARM minimally affects the hepatic hemodynamics, but their study was during liver resection and not transplant.¹⁷

The effects of stepwise ARM mode on the blood supply and drainage of the newly transplanted graft in the present trial was found to be negligible and could be performed safely. However, the option to decline or reduce PEEP should always be available whenever the systemic hemodynamics are depressed and not responding to fluids or catecholamines. Avoidance of extreme elevations in CVP during liver transplant is recommended to reduce hepatic congestion, particularly during the dissection of the diseased liver. The stepwise technique of recruitment was able to determine the individualized optimal PEEP for each recipient, in contrast to the rapid single-step recruitment maneuver and without sharp hemodynamic changes as reflected in the results section. Meade and colleagues¹⁸ and Marini¹⁹ investigated the various ARM techniques and came to the conclusion that the manual sustained hyperinflations can increase the risk for barotrauma, but the stepwise technique preserves the hemody­namics with reduced incidence of barotrauma.^18,19 The Italian single-center trial by Severgnini and colleagues²⁰ investigated the effectiveness of protec­tive ventilation with a tidal volume of 7 mL/kg and 10 cm H₂O PEEP versus nonprotective ventilation and found that the clinical pulmonary infection score and the chest radiograph abnormalities were reduced. In contrast, the PROVILO study^21,22 included 900 patients undergoing open abdominal surgery and found no difference between protective and nonprotective ventilation. In these 2 studies, a fixed PEEP was applied during recruitment for every patient, in contrast to the present trial, which allowed for individual variations in the PEEP requirements. A fixed PEEP for the patients would not be able to adapt to the changes during surgery and the variation in the clinical and operative status between patients. Development of scores or methods to identify or predict patients at risk for PPC is needed. Canet and colleagues (ARISCAT Group) in a prospective multicenter study developed a risk index based on 7 objectives to assess and identify the individuals prone to PPC.²³

The biggest difference in PPC in this present trial was in the incidence of pleural effusions, which was higher among the controls. It is difficult to relate the ARM process to this decrease in pleural effusion, particularly with liver transplant surgery, and the manipulations next to the diaphragm in hepatic patients known to have oncotic pressure disturbances, but this significant decrease in pleural fluid formation should be considered and investigated in future trials.

We also noticed that the FTc readings by Doppler were not affected by high PEEP values as the traditional CVP. This could be a promising tool to monitor fluid administration during surgery when high PEEP values are used.

Surgeons were kept aware and informed to report any related surgical difficulty during the transplant procedure, but no surgical technical difficulty was reported. The bulge of the right diaphragm with mechanical ventilation during anhepatic phase was noted among the recruitment group only. This was only clearly visible when the recipient’s own cirrhotic liver was dissected out prior to the insertion of the new liver graft. The reason behind this was that the abdominal side of the diaphragm became clearly visible to the surgeons with no organs covering it. This bulge of the diaphragm with recruitment was the result of the distention of the right lower lung region when the peripherally collapsed alveoli were reopened with recruitment. In contrast, the left side lung recruitment of lower collapsed alveoli was not evident to the surgeons’ eyes because it was covered by other abdominal organs such as stomach and spleen.

Unfortunately, it is believed that PEEP could increase CVP and hepatic venous pressures during liver surgery, particularly during resection, and this can lead to an increase in blood loss, transfusion requirement, morbidity, and mortality.^24,25

Neuschwander and colleagues in a secondary analysis of data from a randomized control trial named IMPROVE disagree with the above-mentioned assertions and state that mechanical ventilation using PEEP during liver surgery with lung-protective strategy was not associated with an increase in bleeding, and the same conclusion was reported by Yassen and colleagues (in 2018) in a randomized controlled trial with and without ARM during liver resection among a group of hepatic patients with liver cirrhosis as a result of hepatitis C.^7,26 In this present trial the blood transfusion requirements were comparable between both groups of recipients with and without recruitment.

A transfusion-free surgical transplant was possible in a considerable number of the recipients in both groups as demonstrated in Table 4, but at periods of hypotension and during reperfusion it is possible that recruitment of the lungs can be aborted temporarily when fluid replacement and catechola­mine boluses do not stabilize the hemodynamics as demonstrated in the flow chart of this present trial.

In conclusion, the stepwise approach during ARM significantly improved the pulmonary functions of the recipients both intraoperative and postoperative with minimal effects on systemic hemodynamics and on the newly transplanted liver graft blood supply and functions. A significant reduction in PPC with ARM was observed, but with no reduction in ICU ventilation or stay. This present study was able to prove that lung recruitment with a stepwise approach and with an individual and optimal PEEP can be tolerated by the recipients during liver transplant surgery.

References:

1. 1. Lin YH, Cai ZS, Jiang Y, Lu LZ, Zhang XJ, Cai QC. Perioperative risk factors for pulmonary complications after liver transplantation. J Int Med Res. 2010;38(5):1845-1855. doi:10.1177/147323001003800532
   CrossRef - PubMed
   
2. Feltracco P, Carollo C, Barbieri S, Pettenuzzo T, Ori C. Early respiratory complications after liver transplantation. World J Gastroenterol. 2013;19(48):9271-9281. doi:10.3748/wjg.v19.i48.9271
   CrossRef - PubMed
   
3. García-Fernández J, Romero A, Blanco A, Gonzalez P, Abad-Gurumeta A, Bergese SD. Recruitment manoeuvres in anaesthesia: How many more excuses are there not to use them? Rev Esp Anestesiol Reanim. 2018;65(4):209-217. English, Spanish. doi:10.1016/j.redar.2017.12.006
   CrossRef - PubMed
   
4. Weingarten TN, Whalen FX, Warner DO, et al. Comparison of two ventilatory strategies in elderly patients undergoing major abdominal surgery. Br J Anaesth. 2010;104(1):16-22. doi:10.1093/bja/aep319
   CrossRef - PubMed
   
5. Almarakbi WA, Fawzi HM, Alhashemi JA. Effects of four intraoperative ventilatory strategies on respiratory compliance and gas exchange during laparoscopic gastric banding in obese patients. Br J Anaesth. 2009;102(6):862-868. doi:10.1093/bja/aep084
   CrossRef - PubMed
   
6. Futier E, Constantin JM, Paugam-Burtz C, et al. A trial of intraoperative low-tidal-volume ventilation in abdominal surgery. N Engl J Med. 2013;369(5):428-437. doi:10.1056/NEJMoa1301082
   CrossRef - PubMed
   
7. Yassen K, Abuzaid A, Hussain N, Ibrahim E, Khater Y. The respiratory and haemodynamic effects of alveolar recruitment in cirrhotic patient undergoing liver resection surgery. a randomized controlled trial. Am Soc Anesthesiol. 2018:A2167.
   CrossRef - PubMed
   
8. Arnal JM, Paquet J, Wysocki M, et al. Optimal duration of a sustained inflation recruitment maneuver in ARDS patients. Intensive Care Med. 2011;37(10):1588-1594. doi:10.1007/s00134-011-2323-0
   CrossRef - PubMed
   
9. Garcia-Fernandez J, Canfran S, de Segura IA, Suarez-Sipmann F, Aguado D, Hedenstierna G. Pressure safety range of barotrauma with lung recruitment manoeuvres: a randomised experimental study in a healthy animal model. Eur J Anaesthesiol. 2013;30(9):567-574. doi:10.1097/EJA.0b013e3283607875
   CrossRef - PubMed
   
10. Sinclair S, James S, Singer M. Intraoperative intravascular volume optimisation and length of hospital stay after repair of proximal femoral fracture: randomised controlled trial. BMJ. 1997;315(7113):909-912. doi:10.1136/bmj.315.7113.909
    CrossRef - PubMed
    
11. Saner FH, Gieseler RK, Akiz H, Canbay A, Gorlinger K. Delicate balance of bleeding and thrombosis in end-stage liver disease and liver transplantation. Digestion. 2013;88(3):135-144. doi:10.1159/000354400
    CrossRef - PubMed
    
12. Ibitoye BO, Idowu BM, Ogunrombi AB, Afolabi BI. Ultrasonographic quantification of pleural effusion: comparison of four formulae. Ultrasonography. 2018;37(3):254-260. doi:10.14366/usg.17050
    CrossRef - PubMed
    
13. Daniel WW. A Foundation for Analysis in the Health Sciences. Biostatistics Toronto, John Wiley; 1991:209-215.
    CrossRef - PubMed
    
14. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;39(2):175-191. doi:10.3758/bf03193146
    CrossRef - PubMed
    
15. Yi HM, Cai CJ, Lu MQ, et al. [The treatment strategy of early ALI after liver transplantation]. Chinese J Surg. 2006;44(13):889-893.
    CrossRef - PubMed
    
16. Pang CK, Yap J, Chen PP. The effect of an alveolar recruitment strategy on oxygenation during laparascopic cholecystectomy. Anaesth Intensive Care. 2003;31(2):176-180. doi:10.1177/0310057X0303100206
    CrossRef - PubMed
    
17. Ukere A, Meisner S, Greiwe G, et al. The influence of PEEP and positioning on central venous pressure and venous hepatic hemodynamics in patients undergoing liver resection. J Clin Monit Comput. 2017;31(6):1221-1228. doi:10.1007/s10877-016-9970-1
    CrossRef - PubMed
    
18. Meade MO, Cook DJ, Guyatt GH, et al. Ventilation strategy using low tidal volumes, recruitment maneuvers, and high positive end-expiratory pressure for acute lung injury and acute respiratory distress syndrome: a randomized controlled trial. JAMA. 2008;299(6):637-645. doi:10.1001/jama.299.6.637
    CrossRef - PubMed
    
19. Marini JJ. Recruitment by sustained inflation: time for a change. Intensive Care Med. 2011;37(10):1572-1574. doi:10.1007/s00134-011-2329-7
    CrossRef - PubMed
    
20. Severgnini P, Selmo G, Lanza C, et al. Protective mechanical ventilation during general anesthesia for open abdominal surgery improves postoperative pulmonary function. Anesthesiology. 2013;118(6):1307-1321. doi:10.1097/ALN.0b013e31829102de
    CrossRef - PubMed
    
21. Hemmes SN, Severgnini P, Jaber S, et al. Rationale and study design of PROVHILO - a worldwide multicenter randomized controlled trial on protective ventilation during general anesthesia for open abdominal surgery. Trials. 2011;12:111. doi:10.1186/1745-6215-12-111
    CrossRef - PubMed
    
22. PROVE Network Investigators for the Clinical Trial Network of the European Society of Anaesthesiology, Hemmes SN, Gama de Abreu M, Pelosi P, Schultz MJ. High versus low positive end-expiratory pressure during general anaesthesia for open abdominal surgery (PROVHILO trial): a multicentre randomised controlled trial. Lancet. 2014;384(9942):495-503. doi:10.1016/S0140-6736(14)60416-5
    CrossRef - PubMed
    
23. Canet J, Gallart L, Gomar C, et al. Prediction of postoperative pulmonary complications in a population-based surgical cohort. Anesthesiology. 2010;113(6):1338-1350. doi:10.1097/ALN.0b013e3181fc6e0a
    CrossRef - PubMed
    
24. McNally SJ, Revie EJ, Massie LJ, et al. Factors in perioperative care that determine blood loss in liver surgery. HPB (Oxford). 2012;14(4):236-241. doi:10.1111/j.1477-2574.2011.00433.x
    CrossRef - PubMed
    
25. Li Z, Sun YM, Wu FX, Yang LQ, Lu ZJ, Yu WF. Controlled low central venous pressure reduces blood loss and transfusion requirements in hepatectomy. World J Gastroenterol. 2014;20(1):303-309. doi:10.3748/wjg.v20.i1.303
    CrossRef - PubMed
    
26. Neuschwander A, Futier E, Jaber S, et al. The effects of intraoperative lung protective ventilation with positive end-expiratory pressure on blood loss during hepatic resection surgery: A secondary analysis of data was from a published randomized control trial (IMPROVE). Eur J Anaesthesiol. 2016;33(4):292-298. doi:10.1097/EJA.0000000000000390
    CrossRef - PubMed