Kremer et al.

Journal of Cardiothoracic Surgery (2019) 14:89

https://doi.org/10.1186/s13019-019-0910-2

RESEARCH ARTICLE Open Access

Differential lung ventilation for increased

oxygenation during one lung ventilation
for video assisted lung surgery
Ran Kremer1, Wisam Aboud2, Ori Haberfeld1, Maruan Armali2 and Michal Barak3*

Abstract
Background: One lung ventilation (OLV) is the technique used during lung resection surgery in order to facilitate
optimal surgical conditions. OLV may result in hypoxemia due to the shunt created. Several techniques are used to
overcome the hypoxemia, one of which is continuous positive airway pressure (CPAP) to the non-dependent lung.
Another technique is ventilating the non-dependent lung with a minimal volume, thus creating differential lung
ventilation (DLV). In this study we compared the efficacy of CPAP to DLV during video assisted thoracoscopic lung
resection.
Patients and method: This is a prospective study of 30 adult patients undergoing elective video assisted thoracoscopic
lung lobectomy. Each patient was ventilated in four modes: two lung ventilation, OLV, OLV + CPAP and OLV + DLV.
Fifteen patients were ventilated with CPAP first and DLV next, and the other 15 were ventilated with DLV first and then
CPAP. Five minutes separated each mode, during which the non-dependent lung was open to room air. We measured
the patient’s arterial blood gas during each mode of ventilation. The surgeons, who were blinded to the ventilation
technique, were asked to assess the surgical conditions at each stage.
Results: Oxygenation during OLV+ CPAP was significantly lower that OLV + DLV (p = 0.018). There were insignificant
alterations of pH, PCO2 and HCO3 during the different ventilating modes. The surgeons’ assessments of interference in
the field exposure between OLV + CPAP or OLV + DLV was found to be insignificant (p = 0.073).
Conclusions: During OLV, DLV is superior to CPAP in improving patient’s oxygenation, and may be used where
CPAP failed.
Trial registration: ClinicalTrials.gov NCT03563612. Registered 9 June 2018, retrospectively (due to clerical error).
Keywords: One lung ventilation, Continuous positive airway pressure, Differential lung ventilation

Introduction vasoconstriction, decrease the shunt [4, 5]. The hypox-

During lung resection surgery, optimal surgical access is emia is usually not severe; however, in some cases, life
attained when the operated lung is deflated and its threatening hypoxemia occurs that responds poorly to
movements are avoided. This is achieved by one lung corrective maneuvers [6]. Applying positive end expira-
ventilation (OLV) [1]. Ventilation of one lung creates a tory pressure (PEEP) to the dependent lung and venti-
trans-pulmonary shunt through the non-ventilated lung lating with 100% oxygen, are initial steps. In case there
and causes hypoxemia [2, 3]. Both mechanical factors, is no improvement in oxygenation, additional tech-
such as gravitation and pressure by the surgeon, and niques are used. One technique is insufflating oxygen
the physiological response, mainly hypoxic pulmonary with a constant pressure to the non-dependent lung,
called continuous positive airway pressure (CPAP).
Another option is to ventilate the non-dependent lung
* Correspondence: 10michal@gmail.com
3
Department of Anesthesiology, Rambam Health Care Campus and the
with a minimal volume and rate, creating differential
Rappaport Faculty of Medicine, Technion - Israel Institute of Technology, POB lung ventilation (DLV). Both techniques may impair
9602, 31096 Haifa, Israel exposure to the operated area to some extent.
Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Kremer et al. Journal of Cardiothoracic Surgery (2019) 14:89 Page 2 of 6

The purpose of this study is to compare ventilation the dependent lung during OLV. All patients were
modalities during OLV, which may improve oxygenation placed in the lateral decubitus position for surgery.
with minimal impairment of the surgical field condi- When OLV was initiated, the non- dependent lung
tions. In this study, the non-dependent lung was venti- was open to the atmosphere. Ten minutes after the first
lated alternately, in crossover fashion, by CPAP and by a trocar was introduced, measurements were recorded and
portable ventilator with low rate and pressure in a DLV then the operated lung was either connected to a CPAP
technique [7–9]. We anticipated that a low ventilation system or to a small portable time-cycled ventilator
rate and pressure would produce the least interference paraPAC-2D (Transport Ventilator, SIMS pneuPAC Ltd.,
with the surgeon’s exposure and, at the same time, im- Luton, UK). The order of intervention was randomized
prove oxygenation. by one of the researchers (MB), using computerized soft-
ware (random.org). The CPAP pressure was set to 5 cm
H2O. Differential Lung Ventilation of the non-dependent
Patients and methods
lung was set at a rate of 8 breaths per minute, inspired gas
This is a prospective randomized controlled crossover
100% oxygen, peak pressure and tidal volume set to the
study of adult patients scheduled to have video-assisted
lowest available values, resulting in a peak pressure of 10
thoracoscopic surgery (VATS) of lung lobectomy under
cm H2O and a tidal volume around 50 ml. The treatment
general anesthesia. The study was approved by the institu-
of the operated lung was alternated with a 5-min interval
tion local Ethic Committee and registered (ClinicalTrials.
between modes of ventilation, without additional ventila-
gov NCT03563612). The patients had a detailed explan-
tory support or oxygen insufflations, to avoid the influence
ation of the study before the surgery by one of the anes-
of one modality upon the other. In that 5 min interval, the
thesiologists who took part in this study, and signed an
tube connection of the non-dependent lung was opened
inform consent if they agreed to participate.
to room air.
Exclusion criteria included: American Society of Anes-
thesiologists (ASA) grade ≥ 4, pregnancy, and difficult
Measurements
intubation. Decreased oxygen saturation below 85% at
Demographic data regarding the patients’ age, weight,
any time during the surgery was set as an end point, at
gender and ASA classification was recorded. Arterial
which the study would be stopped. The primary out-
blood gas, peak inspiratory pressure (PIP), and plateau
come variables are the PaO2 measurements, while the
pressure (Pplat) were measured during: two lung ventila-
secondary outcome variables are the spirometry mea-
tion, OLV (10 min after the first trocar was introduced),
surements and the surgeons’ evaluation of the surgical
OLV+ CPAP and OLV+ DLV. PIP and Pplat were
field.
recorded from the anesthesia machine spirometer. At
the same time, the chief surgeon was asked to comment
Study protocol on the surgical field conditions. The surgeon’s evaluation
On arrival in the operating room, an intravenous line was graded from 0 (no interference) to 3 (maximal inter-
was placed. Each patient was monitored with an electro- ference). The surgeons were blinded to the ventilation
cardiogram, pulse oximeter, invasive blood pressure, end mode used at that time since the patient’s sterile covers
tidal capnography, and eosophageal thermometer. Fol- were pulled up, concealing the anesthesia machine and
lowing the induction of general anesthesia using fentanyl the ventilator.
2–5 microgram/kg; propofol 1–3 mg/kg and rocuronium
0.6–0.8 mg/kg, the trachea was intubated with a left Statistical analysis
double lumen tracheal tube VivaSight (ETView Ltd. Sample size was calculated assuming difference in PaO2
Misgav Business Park, Israel) where verification of the between CPAP and mini-ventilation of 50 mmHg; a
tube position was monitored continuously with on-line safety level of 95%; standard deviation of 60; intensity of
video surveillance. The dependent lung was ventilated 80% and measurement ratio of 1:1. Comparison between
by anesthesia machine (Dräger Narkomed 2A) with
sevoflurane for maintenance of anesthesia.
Table 1 Patients’ demographics
Ventilation parameters were as follows: volume con-
Age, years 65 ± 10a
trolled mode with tidal volume was set to 8 ml/kg during
two lung ventilation (TLV) and reduced to 6 ml/kg dur- Male/Female 18/11
ing OLV; respiratory rate of 10–12 breaths per minute Weight, kg 82 ± 19a
during TLV, increased to 12–15 per minute during OLV, BMI 28 ± 5a
adjusted to keep PaCO2 below 50 mmHg; inspired oxy- ASA 2/3/4 13/15/1
gen of 100% at all times; positive end expiratory pressure Note
(PEEP) was 2 cm H2O, first in both lungs and then in a
Data presented in mean ± standard deviation
Kremer et al. Journal of Cardiothoracic Surgery (2019) 14:89 Page 3 of 6

Table 2 Data regarding the operation Regarding the surgeons’ assessment of interference in
CPAP first DLV first P value surgical field exposure:
(n = 15) (n = 14) None of the patients was graded as 3, where interfer-
Lobe resected: ence is maximal. One patient was graded as 1 during
RUL/RLL/LUL/LLL 5/4/5/1 4/4/3/3 0.53 CPAP and 2 during DLV. One patient was graded as 2
Length of surgery (min) 130 ± 34 * 137 ± 48 * 0.74 for both techniques. The rest of the patients received 0
Note
interference for CPAP while six had 1 grade interference
* Data presented in mean ± standard deviation with DLV (p = 0.073).

the groups of patients was performed using the Mann

Whitney non-parametric test. Data of the arterial blood Discussion
gas and spirometry variables during the different ventilation In this study we found that DLV is superior to CPAP as
techniques were compared with Wilcoxon non-parametric mean for improving patient’s oxygenation during OLV.
test. Surgeons’ estimations of interference with the surgical The theoretical explanation for DLV superiority relies on
field were compared with the chi-square test. Differences the physiologic phenomenon of heterogeneity in differ-
were considered statistically significant at p < 0.05. ent areas in the lung [10, 11]. Ventilating both lungs
with the same pressure results in fresh gas flow ventilat-
Results ing the lower resistance parts of the lung, while the areas
Thirty patients were recruited as participants. One with high resistance, such as atalectatic areas, remain
patient in the group that had DLV first was excluded, unventilated. Heterogeneous lung aeration may result in
due to difficulty in tracheal intubation. No significant lung inflammation and injury, which deteriorates gas ex-
difference was found between the two groups, CPAP change furthermore [12, 13]. Ventilating independently
first or DLV first, in all studied variables. Participant’s different parts of the lung with different pressures may
demographics are shown in Table 1 and data regarding help force the air into atalectatic parts, reduce V/Q mis-
the surgery in Table 2. match and improve oxygenation [14, 15]. This theory is
Regarding arterial blood gas: oxygenation reduced supported by experimental models [7, 12]. Clinical trials
significantly when changing from two lung ventilation to showed a similar beneficial effect of DLV in patients in
OLV (Fig. 1). Oxygenation during OLV+ CPAP was sig- intensive care units [16–18] and during open thoracic
nificantly lower that OLV + DLV (p = 0.018). Alterations surgery [9, 19]. However, we found no data regarding
of PCO2, pH and HCO3 during the different ventilating DLV during thoracoscopic surgery.
modes were not significant (Figs. 2, 3, 4 respectively). Theoretically, changing ventilation may improve oxy-
No significant difference was noted between peak and genation indirectly by reducing CO2 levels, according to
plateau pressures during OLV, OLV + CPAP or OLV + the formula: PaO2 = FiO2 (Pbr-PH2O)-PCO2/K. How-
DLV. ever, we found no significant change in CO2 in this

Fig. 1 Arterial PaO2

Kremer et al. Journal of Cardiothoracic Surgery (2019) 14:89 Page 4 of 6

Fig. 2 Arterial PaCO2

study. Thus, improved oxygenation was not the outcome that this time interval was long enough to allow patient’s
of CO2 levels. oxygenation return to its baseline. Another problem was
The main disadvantage of DLV, as with CPAP, is that the non-significant results of the surgical exposure
both may interfere with surgical field exposure. Spirometry assessment (p = 0.073). This marginally non-significant
of the dependent lung was recorded in order to find whether result may be caused by Type II error, and may have
the different ventilation modes of the non-dependent lung been different in a larger scale study. The surgeons were
influence the dependent one. No significant difference was blinded to the mode of ventilation, yet the anesthesiologist
found. was not. This may have set additional error to the study.
An important weakness of this study relies on its In the past decades, when cardio-thoracic surgery be-
design as a cross-over study. There is a possibility of came minimally invasive, good lung deflation became a
carry-over effects and its influence on interpretation of necessity. Video- and robot- assisted thoracoscopic oper-
the findings. In order to overcome it we set a period of ations require high quality OLV [20, 21]. Moreover,
5-min interval between modes of ventilation, without many of these procedures are performed with the patient
additional ventilatory support or oxygen insufflations. In in the supine position, with a large shunt and low
that 5 min interval, the tube connection of the non- oxygenation. At the same time, the patient often has
dependent lung was opened to room air. We assume poor basic lung functions, and hypoxemia occurs

Fig. 3 Arterial pH
Kremer et al. Journal of Cardiothoracic Surgery (2019) 14:89 Page 5 of 6

Fig. 4 Arterial HCO3. B = two lung ventilation. A = one lung ventilation. C = one lung ventilation + CPAP. D = one lung ventilation + DLV

rapidly. Patients with ischemic heart disease who Consent for publication
undergo coronary artery bypass graft are especially sus- Not applicable.

ceptible to injury during hypoxemia, and aggressive

treatment of arterial desaturation is mandatory to ensure Competing interests
The authors declare that they have no competing interests.
the patient’s safety [22, 23]. The challenge of the
anesthesiologist is to overcome hypoxemia without dis-
turbing exposure of the surgical field. We believe the Publisher’s Note
DLV is an additional tool for improving oxygenation Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
during OLV.
Author details
1
Department of Thoracic Surgery, Rambam Health Care Campus, Haifa, Israel.
Conclusions 2
Department of Anesthesiology, the Baruch Padeh Medical Center, Poriya,
The use of DLV while ventilating one lung may improve Tiberius, Israel. 3Department of Anesthesiology, Rambam Health Care
patient’s oxygenation, and was found to be better than Campus and the Rappaport Faculty of Medicine, Technion - Israel Institute of
Technology, POB 9602, 31096 Haifa, Israel.
CPAP. Differential lung ventilation may be used where
CPAP failed. Received: 6 February 2019 Accepted: 22 April 2019

Abbreviations
ASA: American Society of Anesthesiologists; CPAP: Continuous positive
airway pressure; DLV: Differential lung ventilation; OLV: One lung ventilation; References
PEEP: Positive end expiratory pressure; PIP: Peak inspiratory pressure; 1. Kirschner PA. The surgery-anesthesia relationship: a surgeon’s view. In:
Pplat: Plateau pressure; VATS: Video-assisted thoracoscopic surgery Cohen E, editor. The Practice of Thoracic Anesthesia. Philadelphia: JB
Lippincott Company; 1995. p. 163–80.
2. Karzai W, Schwarzkopf K. Hypoxemia during one-lung ventilation:
Acknowledgments
prediction, prevention, and treatment. Anesthesiology. 2009;110:1402–11.
None
3. Cohen E. Management of one-lung ventilation. Anesthesiol Clin N Am.
2001;19:475–95.
Funding
4. Nomoto Y. Perioperative pulmonary blood flow and one lung anesthesia.
This research did not receive any specific grant from funding agencies in the
Can J Anaesth. 1987;34:447.
public, commercial, or not-for-profit sectors.
5. Benumof JL. One lung ventilation and hypoxic pulmonary vasoconstriction:
implications for anesthetic management. Anesth Analg. 1985;64:821.
Availability of data and materials 6. Mierdl S, Meininger D, Dogan S, et al. Does poor oxygenation during one-
The datasets used and/or analyzed during the current study are available lung ventilation impair aerobic myocardial metabolism in patients with
from the corresponding author on reasonable request. symptomatic coronary artery disease? Interact Cardiovasc Thorac Surg. 2007;
6:209–13.
Authors’ contributions 7. Borges JB, Senturk M, Ahlgren O, Hedenstierna G, Larsson A. Open lung in
RK: study design. WA: study design and data collection. OH: patient lateral decubitus with differential selective positive end-expiratory pressure
recruitment and data analyzing. MA: manuscript preparation. MB: study in an experimental model of early acute respiratory distress syndrome. Crit
design, data collection and manuscript preparation. Review of manuscript: all Care Med. 2015;43:e404–11.
authors. All authors read and approved the final manuscript. 8. Skjeflo GW, Dybwik K. A new method of securing the airway for differential
lung ventilation in intensive care. Acta Anaesthesiol Scand. 2014;58:463–7.
Ethics approval and consent to participate 9. Shechtman MY, Ziser A, Barak M, Ben-Nun A. Mini-ventilation for improved
The study was approved by the institution local Ethic Committee and all the oxygenation during lung resection surgery. Anaesth Intensive Care. 2011;39:
patients included signed an inform consent. 456–9.
Kremer et al. Journal of Cardiothoracic Surgery (2019) 14:89 Page 6 of 6

10. Aliverti A, Pennati F, Salito C, Woods JC. Regional lung function and
heterogeneity of specific gas volume in healthy and emphysematous
subjects. Eur Respir J. 2013;41:1179–88.
11. Kaminsky DA, Irvin CG, Lundblad LK, et al. Heterogeneity of
bronchoconstriction does not distinguish mild asthmatic subjects from
healthy controls when supine. J Appl Physiol (1985). 2008;104:10–9.
12. Tojo K, Nagamine Y, Yazawa T, et al. Atelectasis causes alveolar hypoxia-
induced inflammation during uneven mechanical ventilation in rats.
Intensive Care Med Exp. 2015;3:56.
13. Bowser JL, Lee JW, Yuan X, Eltzschig HK. The hypoxia-adenosine link during
inflammation. J Appl Physiol (1985). 2017;123:1303–20.
14. Klingstedt C, Baehrendtz S, Bindslev L, Hedenstierna G. Lung and chest wall
mechanics during differential ventilation with selective PEEP. Acta
Anaesthesiol Scand. 1985;29:716–21.
15. Siegel JH, Stoklosa JC, Borg U, et al. Quantification of asymmetric lung
pathophysiology as a guide to the use of simultaneous independent lung
ventilation in posttraumatic and septic adult respiratory distress syndrome.
Ann Surg. 1985;202:425–39.
16. Anantham D, Jagadesan R, Tiew PEC. Clinical review: independent lung
ventilation in critical care. Crit Care Lond Engl. 2005;9:594–600.
17. Yamakawa K, Nakamori Y, Fujimi S, Ogura H, Kuwagata Y, Shimazu T. A
novel technique of differential lung ventilation in the critical care setting.
BMC Res Notes. 2011;4:134.
18. Wickerts CJ, Blomqvist H, Baehrendtz S, Klingstedt C, Hedenstierna G,
Frostell C. Clinical application of differential ventilation with selective
positive end-expiratory pressure in adult respiratory distress syndrome.
Acta Anaesthesiol Scand. 1995;39:307–11.
19. Baraka A, Muallem M, Baroody MA, Sibai AN, Usta N, Sibai AM. Differential
lung ventilation during thoracotomy. Middle East J Anaesthesiol. 1988;9:
357–62.
20. Umari M, Falini S, Segat M, et al. Anesthesia and fast-track in video-assisted
thoracic surgery (VATS): from evidence to practice. J Thorac Dis. 2018;
10(Suppl 4):S542–S54.
21. Seco M, Edelman JJ, Yan TD, Wilson MK, Bannon PG, Vallely MP. Systematic
review of robotic-assisted, totally endoscopic coronary artery bypass
grafting. Ann Cardiothorac Surg. 2013;2:408–18.
22. Liu TJ, Shih MS, Lee WL, et al. Hypoxemia during one-lung ventilation for
robot-assisted coronary artery bypass graft surgery. Ann Thorac Surg.
2013;96:127–32.
23. Deshpande SP, Lehr E, Odonkor P, et al. Anesthetic management of
robotically assisted totally endoscopic coronary artery bypass surgery
(TECAB). J Cardiothorac Vasc Anesth. 2013;27:586–99.