j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 2 ( 2 0 1 4 ) 2 4 2 e2 4 9

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.JournalofSurgicalResearch.com

Association for Academic Surgery

“Open lung ventilation optimizes pulmonary

function during lung surgery”

John B. Downs, MD,a Lary A. Robinson, MD,b,*

Michael L. Steighner, CRNA,a David Thrush, MD,a Richard R. Reich, PhD,c,d
and Jukka O. Räsänen, MDa
a
Department of Anesthesiology, Moffitt Cancer Center, Tampa, Florida
b
Department of Thoracic Oncology (Surgery), Moffitt Cancer Center, Tampa, Florida
c
College of Arts and Sciences, University of South Florida Sarasota-Manatee, Florida
d
Department of Biostatistics, Moffitt Cancer Center, Tampa, Florida

article info abstract

Article history: Background: We evaluated an “open lung” ventilation (OV) strategy using low tidal volumes,
Received 27 February 2014 low respiratory rate, low FiO2, and high continuous positive airway pressure in patients
Received in revised form undergoing major lung resections.
7 June 2014 Materials and methods: In this phase I pilot study, twelve consecutive patients were anes-
Accepted 17 June 2014 thetized using conventional ventilator settings (CV) and then OV strategy during which
Available online 20 June 2014 oxygenation and lung compliance were noted. Subsequently, a lung resection was performed.
Data were collected during both modes of ventilation in each patient, with each patient acting
Keywords: as his own control. The postoperative course was monitored for complications.
Anesthesia ventilation Results: Twelve patients underwent open thoracotomies for seven lobectomies and five
Protective ventilation segmentectomies. The OV strategy provided consistent one-lung anesthesia and improved
Lung resection static compliance (40
7 versus 25
4 mL/cm H2O, P ¼ 0.002) with airway pressures similar
Thoracic surgery to CV. Postresection oxygenation (SpO2/FiO2) was better during OV (433
11 versus
Postoperative pulmonary 386
15, P ¼ 0.008). All postoperative chest x-rays were free of atelectasis or infiltrates. No
complications patient required supplemental oxygen at any time postoperatively or on discharge. The
Airway pressure release ventilation mean hospital stay was 4
1 d. There were no complications or mortality.
Pulmonary atelectasis Conclusions: The OV strategy, previously shown to have benefits during mechanical venti-
Lung surgery lation of patients with respiratory failure, proved safe and effective in lung resection pa-
tients. Because postoperative pulmonary complications may be directly attributable to the
anesthetic management, adopting an OV strategy that optimizes lung mechanics and gas
exchange may help reduce postoperative problems and improve overall surgical results. A
randomized trial is planned to ascertain whether this technique will reduce postoperative
pulmonary complications.
ª 2014 Elsevier Inc. All rights reserved.

Presented at the Academic Surgical Congress, San Diego, CA, February 4e6, 2014.
* Corresponding author. Department of Thoracic Oncology, Moffitt Cancer Center, 12902 Magnolia Drive, Tampa, FL 33612. Tel.: þ1 813
745 6895; fax: þ1 813 745 3027.
E-mail address: lary.robinson@moffitt.org (L.A. Robinson).
0022-4804/$ e see front matter ª 2014 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jss.2014.06.029
 j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 2 ( 2 0 1 4 ) 2 4 2 e2 4 9
1. Introduction
Postoperative pulmonary complications (PPC) are a major
cause of morbidity, mortality, and health-care costs [1,2]. As
many as 33% of patients undergoing upper abdominal surgery
and up to 56% of lung resection patients will develop PPC’s
[1,3]. Two of every three in-hospital postoperative deaths are
caused by pulmonary complications [4]. The economic impact
of PPC is estimated at $11.9 billion annually [1,4]. Cigarette
smoking, upper abdominal or thoracic surgery, age >70,
chronic obstructive pulmonary disease, obesity (body mass
index 30 kg/m2), anesthetic times >180 min, renal failure,
poor nutritional status, and significant blood loss are felt to be
preoperative risk factors [5]. However, early postoperative
atelectasis is considered to be the common pathway to PPC’s
[1,3,5].
Patients undergoing general anesthesia for lung resections
are particularly prone to PPC and additionally are often diffi-
cult to ventilate and oxygenate [6]. General anesthesia, muscle
relaxation, high levels of inspired oxygen, supine, lateral,
and lithotomy positions all have been shown to promote
atelectasis, which in some individuals may persist for >24 h
postoperatively [7]. Almost all patients receive elevated con-
centrations of oxygen intraoperatively and postoperatively,
sometimes for days, in an attempt to alleviate the arterial
hypoxemia resulting from atelectasis. Most investigators feel
that postoperative atelectasis likely leads to retained secre-
tions and subsequent pneumonia, which may progress under
some circumstances to acute respiratory failure and even
acute respiratory distress syndrome (ARDS) [1,6].
Conceivably, some of these complications may be attrib-
utable directly to atelectasis due to anesthetic management,
yet there is no consensus regarding the optimal intraoperative
ventilatory strategy during one-lung ventilation for noncar-
diac thoracic surgery [8] or abdominal surgery [2]. Most rec-
ommendations revolve around attempts to prevent arterial
hypoxemia, with little regard for optimization of lung function
[9]. In spite of the overwhelming evidence that anesthetic
management may play a significant role in the occurrence of
PPC, little or no attention has been given to intraoperative
maneuvers to optimize lung function [1]. Nevertheless, one
recently published clinical trial by Severgnini et al. [10] in 2013
focusing on using protective ventilation during anesthesia for
abdominal surgery demonstrated improved postoperative
lung function and decreased clinical signs of infection.
Table 1 e Anesthesia techniques.
Ventilator settings Conventional (CV)
Vt 6 mL/kg
RR Amount needed to keep exhaled CO2
(PetCO2) 40e45 mm Hg
I:E 1:2
Inspired oxygen (FiO2) Usually 0.21 or amount needed to
maintain oxygen saturation (SpO2) >88%
Positive end-expiratory 3 cm H2O
pressure (PEEP) or CPAP
RP None
243
Remarkably, in this randomized trial, postoperative lung
function abnormalities persisted as long as 5 d after surgery in
the control ventilation group.
We hypothesized that an open lung ventilation (OV)
strategy, using low (physiologic) tidal volume (Vt), low respi-
ratory rate (RR), low concentration of inspired oxygen (FiO2),
and a high level of continuous positive airway pressure (CPAP)
all with proven benefits during mechanical ventilation of pa-
tients with respiratory failure, would allow us to optimize
intraoperative lung mechanics and gas exchange. In essence,
the open lung ventilation strategy is designed to maintain
alveoli open and functional throughout the ventilatory cycle.
This pilot study was designed to evaluate the safety and effi-
cacy of an open lung ventilation strategy in the challenging
group of surgical patients undergoing an open thoracotomy
for major lung resection.
The open thoracotomy lung resection method was chosen
for this pilot study because it is the most common lung sur-
gery approach currently used in the United States, and this
method is perceived to be the most difficult postoperatively
for patients to ventilate and avoid postoperative complica-
tions. In fact, most of the lobectomies and segmentectomies
in the United States are still performed by open thoracotomy
(57% open, 40% Video-assisted thoracic surgery (VATS), and
3.4% robotic) [11], so this surgical approach is also quite
appropriate to study. Finally, because the conventional and
test anesthesia ventilation techniques were used and data
were obtained sequentially in all surgical procedures, patients
acted as their own controls for this phase I study.
2. Methods
After obtaining informed consent, adult patients electively
scheduled to undergo an open thoracotomy for lung resection
were enrolled. The investigation was approved by the Insti-
tutional Review Boards of the Moffitt Cancer Center (protocol
#MCC17297) and the University of South Florida College of
Medicine. Procedures followed were in accordance with the
ethical standards of The Committee on Human Experimen-
tation at the Moffitt Cancer Center. All patient information
was collected in compliance with Health Insurance Portability
and Accountability Act regulations.
Patients were excluded from the study if they had a prior
operative procedure in the ipsilateral pleural cavity, or if the
Open lung (OV)
6 mL/kg (maintained by adjusting CPAP and RP)
Amount needed to keep exhaled CO2 (PetCO2) 40e45 mm Hg
2:1, 3:1, or 4:1
Usually 0.21 or amount needed to maintain oxygen saturation (SpO2)
>88%
30 cm H2O initially, then the amount needed to maximize compliance
Level required for Vt of 6 mL/kg
244 j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 2 ( 2 0 1 4 ) 2 4 2 e2 4 9

Fig. 1 e Research protocol.

expected duration of the procedure was <2 h. The two venti- reduced in 2 cm H20 decrements and RP was adjusted at each
lation techniques used in the study are summarized in level to maintain a Vt 5e6 mL/kg. At each pressure level,
Table 1. The research protocol is outlined in Figure 1. Each respiratory system static compliance was calculated as Vt/
patient in this trial underwent both ventilation techniques (PaweRP). The Paw and RP were left at levels that produced
sequentially during each of the three phases of the operative the highest respiratory system compliance. The Paw was
procedure, and therefore, each patient acted as their own recorded at a “0” flow state, that is, it was a prolonged
control for comparison of the two techniques. For this phase I “plateau” pressure. After a 10-min stabilization period, data
pilot study, there was no separate control group or randomi- collection was repeated.
zation schedule. The patient was positioned in a lateral decubitus position
Before induction of anesthesia, standard monitors, for the operative procedure. One-lung ventilation was estab-
including noninvasive blood pressure, electrocardiogram lished using CV as described previously. Following the thora-
leads II and VI, Bispectral Index (Covidien, Mansfield, MA), and cotomy and equilibration of all values, data were collected and
a pulse oximeter (SpO2) were applied and the patient was OV was then instituted, as described previously. After a period
preoxygenated to an exhaled oxygen concentration of 80%. of stabilization, the data collection was repeated for the OV
General anesthesia was induced with intravenous propofol technique. Open lung ventilation was maintained for the
140e200 mg, and neuromuscular blockade was obtained using remainder of the operative procedure, until chest wall closure.
intravenous rocuronium bromide 50 mg. A left-sided, double- After lung resection and before chest closure, a continuous
lumen endotracheal tube was inserted and its proper position Paw of 30 cm H2O was applied momentarily to the operated
was verified by fiberoptic bronchoscopy. A 20 gauge radial lung with the chest cavity filled with saline to ensure there
artery catheter was inserted for continuous monitoring of was no bronchial stump air leak. A tunneled extrapleural
systemic blood pressure. General anesthesia was maintained catheter (On-Q, I-Flow, LLC, Lake Forest, CA) was inserted
with inhaled sevoflurane and intermittent doses of intrave- percutaneously for postoperative pain management, the
nous fentanyl. A Mindray A5 anesthesia machine and venti- operative lung was reexpanded, and two-lung ventilation was
lator (Mindray DS USA, Inc, Mahwah, NJ) were used to provide instituted using CV settings. After stabilization, data collec-
anesthesia and ventilation to all patients. Conventional tion was repeated in the lateral position using CV and OV in
ventilator settings (CV) consisted of a Vt of 5e6 mL/kg, a RR
required to maintain end-tidal exhaled carbon dioxide tension
(PetCO2) between 40 and 45 mm Hg, an inspiratory:expiratory
ratio (I:E) of 1:2, and a positive end-expiratory pressure (PEEP) Table 2 e Patient demographics.
of 3 cm H2O (Table 1). The inspired oxygen concentration was Value (n ¼ 12; M:F 6:6) Mean Standard Range
set to the lowest value possible to maintain SpO2 >88%. After deviation
10 min, plateau airway pressure (Paw), RR, PEEP, Vt, FiO2, Age (y) 67.9 7.8 55e80
PetCO2, and SpO2 were recorded. Respiratory system static Weight (kg) 82.1 6.7 48e140
compliance was calculated as Vt/ (Paw-PEEP). BMI (kg/m2) 29.2 7.3 17e43
To implement the research protocol for an “open lung” Smoking (pack-years) 43.1 3.0 30e60
ventilation strategy, the I:E ratio was changed to 2:1, 3:1, or 4:1 FEV1.0 (% of predicted) 80.0 4.7 50e106
DLCO (% of predicted) 74.8 3.9 52e95
so that each mechanical breath began just as expiratory flow
Preoperative PaO2 77.1 3.9 61e108
of the previous breath approached 0 L/min. Paw was initially Preoperative PaCO2 39.5 1.0 36e48
set to 30 cm H20, a value arbitrarily selected for lung recruit- Preoperative pH 7.40 0.01 7.38e7.45
ment, and expiratory pressure setting of the machine,
BMI ¼ body mass index; DLCO ¼ diffusing capacity for carbon
henceforth referred to as the release pressure (RP), was
monoxide; FEV1 ¼ forced expiratory volume at 1 second.
adjusted to maintain a Vt of 5e6 mL/kg. After 2 min, Paw was
 j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 2 ( 2 0 1 4 ) 2 4 2 e2 4 9 245

Statistical comparison of results obtained during OV and

Table 3 e Perioperative clinical characteristics.
CV was conducted using the Wilcoxon signed rank test. Dif-
Value (n ¼ 12) Mean Standard Range ferences with P <0.05 were considered significant.
deviation

Preinduction SpO2 (%, on 94.3 2.6 88e97

room air) 3. Results
American Society of 2.8 0.6 2e4
Anesthesiologists class
Twelve patients, six male and six female, were enrolled in the
Total blood loss (mL) 131.1 70.0 50e300
Intraoperative fluids (mL) 1850 576.6 1100e3200 study. Seven patients underwent lobectomies and five had
Total surgery time (min) 151.1 40.1 90e235 segmentectomies, of which seven were right-sided and five
Total anesthesia time (min) 197.3 40.3 139e277 were left-sided resections. Segmental resections were done to
PACU FiO2 0.21 0 0.21 conserve lung tissue because of a prior contralateral lobec-
PACU SpO2 (%) 93 1.8 91e96 tomy (2 patients), multiple primary cancers in different lobes
Hospital discharge SpO2 (%, on 97 1.3 94e97
(2 patients), and pulmonary metastasectomy (1 patient).
room air)
Eleven of the patients had primary lung cancer and one had a
Total hospital days 4.1 0.9 3e5
colon cancer metastasis to the lung. All patients were current
or former smokers and most were overweight or obese. The
patients’ preoperative lung function was reasonable and
succession. Thereafter, the patient was placed supine and preoperatively none required chronic oxygen therapy during
muscle relaxation was reversed. Following resumption of waking hours. Based on these mean values, this patient group
adequate spontaneous ventilation with 10 cm H2O CPAP, the does not appear particularly high risk although the mean-
FiO2 was increased until exhaled oxygen was 80%, and the American Society of Anesthesiologists (ASA) class was 2.8,
patient’s trachea was extubated. After extubation, the patient two patients had markedly decreased pulmonary functions
breathed room air for 5 min to ensure SpO2 >90% before and three were class II or class III morbidly obese. Patient
transport to the postanesthesia care unit (PACU). demographic data are shown in Table 2. Perioperative patient
In the PACU, if the SpO2 was <90%, the patient was clinical characteristics are shown in Table 3.
stimulated to encourage respiration, but supplemental O2 Results of measurements reflecting lung mechanics at the
was not applied without first notifying an investigator various phases of the operative procedure are shown in
(J.B.D.). A portable chest radiograph was taken before Table 4. Assuming a normal value of 100 mL/cm H2O, the
discharge from the PACU. Standard postoperative care was respiratory system static compliance was markedly decreased
provided by the thoracic surgeon (L.A.R.). Pain relief was in every patient following induction of general anesthesia
provided with continuous extrapleural bupivacaine 0.5%, (Table 4). During the two-lung ventilation in the supine posi-
scheduled IV acetaminophen for the first 24 h, IV ketorolac, tion, the Paw required to optimize compliance during OV was
and patient-controlled IV hydromorphone. All subsequent higher than the plateau pressure necessary to provide
postoperative chest radiographs were two-view PA and adequate ventilation with CV (P ¼ 0.04; Table 4). However,
lateral x-rays. Radiograph results were taken from the re- during OV the compliance was nearly twice that was calcu-
ports dictated by departmental faculty radiologists reading lated during CV (P ¼ 0.002).
x-rays using standard radiology criteria [12] who were not Compliance decreased significantly when one-lung venti-
involved in the study and de facto they were effectively lation was instituted (P ¼ 0.03) with both modes of ventilation
blinded because they did not know which patients were on (Fig. 2). During one-lung ventilation using CV, higher airway
study. pressure was required to provide adequate Vt and ventilation

Table 4 e Results of conventional (CV) and open lung (OV) ventilation during different phases of surgery. Mean (standard
deviation).
Parameter Two lung ventilation (supine) One lung ventilation (lateral) Two lung ventilation (lateral)

CV OV P CV OV P CV OV P

Vt (mL) 429 (77) 452 (80) NS 442 (75) 458 (87) 0.030 432 (61) 456 (97) NS
Vt/weight (mL/kg) 5 (1) 6 (1) NS 6 (1) 6 (1) 0.030 5 (1) 6 (1) NS
Paw (cm H2O) 17 (4) 19 (4) 0.040 20 (4) 20 (3) NS 15 (3) 19 (2) 0.003
Compliance (mL/cm H2O) 31 (10) 58 (18) 0.002 26 (4) 40 (7) 0.002 35 (7) 48 (14) 0.002
Ventilator rate (bpm) 10 (1) 9 (1) NS 10 (1) 9 (1) NS 10 (1) 10 (1) NS
PetCO2 (mm Hg) 44 (5) 44 (5) NS 42 (4) 41 (6) NS 47 (3) 45 (12) NS
PEEP (cm H2O) 3 (0) N/A N/A 3 (0) N/A N/A 3 (0) N/A N/A
RP (cm H2O) N/A 10 (4) N/A N/A 8 (2) N/A N/A 8 (2) N/A
FiO2 0.21 (0.03) 0.21 (0.01) NS 0.28 (0.23) 0.22 (0.02) NS 0.24 (0.04) 0.22 (0.02) NS
SpO2 (%) 95 (3) 96 (2) NS 94 (3) 93 (2) NS 91 (1) 94 (3) 0.010
SpO2/FiO2 430 (50) 448 (14) NS 398 (97) 422 (35) NS 386 (54) 433 (38) 0.008

N/A ¼ not applicable; NS ¼ not significant.

246 j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 2 ( 2 0 1 4 ) 2 4 2 e2 4 9
Fig. 2 e Static compliance by ventilation mode (CV [ white
bars; OV [ black bars) at each operative stage. Mean
(standard deviation).
compared with two-lung ventilation (P ¼ 0.03). This was not
the case using the OV technique. Paw required to optimize
compliance during one-lung ventilation with OV was not
significantly different from Paw during CV and it was no
higher than it was during two-lung ventilation. Although
compliance decreased during one-lung ventilation with OV, it
still was higher than during CV (P ¼ 0.002). The required Paw
was never >25 cm H2O in any patient during either ventilation
modality. During one-lung ventilation, Paw during OV did not
increase significantly. At the end of surgery, when both lungs
were ventilated in the lateral position, compliance during CV
was similar to that observed after induction in the supine
position (Fig. 2). However, during OV under these conditions,
compliance did not reach the level observed before thoracot-
omy (P ¼ 0.008). Still, compliance during OV was higher than
that calculated during CV (P ¼ 0.02). Postoperatively, all pa-
tients maintained spontaneous ventilation without difficulty.
One morbidly obese patient (body mass index 43) with
obstructive sleep apnea required face-mask BiPAP with room
air for 2 h until he was fully awake.
Intraoperatively, one patient had persistent SpO2 <90%
during one-lung ventilation with CV and received 90%e100%
Fig. 3 e SpO2/FiO2 ratio by ventilation mode (CV [ white
bars; OV [ black bars) at each operative stage. Mean
(standard deviation).
inspired oxygen to maintain SpO2 90%. However, during OV,
this same patient received 25% inspired oxygen and the
lowest SpO2 was 90%. In no patient was it necessary to
interrupt the operative procedure or to ventilate the nonde-
pendent (operative) lung during the lung resection. FiO2, SpO2,
and SpO2/FiO2 were not significantly different between
ventilator modalities during two-lung ventilation.
During one-lung ventilation, FiO2, SpO2, and SpO2/FiO2
were similar with both modes of ventilation when compared
with that observed during two-lung ventilation, before tho-
racotomy. After surgery, FiO2 and SpO2 were similar during OV
and CV for both one- and two-lung ventilation. However, SpO2
and SpO2/FiO2 were significantly greater with OV than CV
following chest closure (P ¼ 0.008) (Fig. 3). There was no dif-
ficulty maintaining ventilation in any patient, as assessed by
end-tidal carbon dioxide. We continuously monitored blood
pressure and heart rate throughout the investigation. At no
time, even during the recruitment maneuver stages of the
investigation, was there any indication of compromised cir-
culatory function or hypotension (systolic blood pressure
<100 mm Hg).
In the PACU, no patient required supplementation of
inspired oxygen to maintain SpO2 90%. Additionally, no pa-
tient had radiographic evidence of atelectasis in either lung in
the PACU or subsequently through to hospital discharge.
Although lobar atelectasis is readily discernible by chest
radiography, lesser degrees of atelectasis may not be as
apparent [12]. During the postoperative hospital stay, there
were no pulmonary or other complications, and no mortality.
The mean hospital stay was 4
1 d (range 3e5 d) and the mean
room air SpO2 on discharge was 97
1% (range 94%e99%). At
the postoperative outpatient clinic visit 2 wk after hospital
discharge, no long-term complications or pulmonary prob-
lems were observed.
4. Discussion
This pilot study was designed to evaluate the feasibility and
safety of an “open lung” ventilation strategy in lung surgery
patients who are known to be difficult to manage intra-
operatively and who have an elevated incidence of PPC as high
as 56% [1,3,6]. To maintain “open” alveoli and prevent atel-
ectasis, we used low Vts, low ventilation rates, low FiO2, and
high levels of CPAP comparing this to a conventional venti-
lation technique in the same patient. This investigational OV
technique was not only safe and feasible in this high-risk
patient population but it provided consistent one-lung anes-
thesia and improved compliance by 58% with similar airway
pressures, compared with CV (Fig. 2). The postresection
oxygenation (SpO2/FiO2) was significantly better during OV
(Fig. 3). Use of the OV technique did not impact the perfor-
mance of the lung resection by the surgeon because the
operative lung was fully collapsed for the entire resection,
there was no impairment of exposure or visualization, and the
test ventilation technique did not increase the operative time.
Likewise, the OV strategy should not hamper the surgeon in
any respect in performing a lung resection by minimally
invasive techniques (VATS or robotic-assisted), and it has
 j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 2 ( 2 0 1 4 ) 2 4 2 e2 4 9
been used previously by one of the authors (J.B.D.) without
problems in prior VATS resections.
General anesthesia alone is well known to produce
significant atelectasis that may persist for >24 h post-
operatively [1e3,6]. Assuming atelectasis precedes pulmonary
complications in some patients, an intraoperative open lung
ventilation strategy might have a beneficial effect in pre-
venting such complications.
Protective ventilation strategy is considered by most to
consist only of a low Vt, defined as 6 mL/kg [13]. Actually, it has
been shown that such a low Vt in the absence of positive end-
expiratory pressure (PEEP) will not maintain an adequate
resting expiratory lung volume, will not prevent atelectasis,
and will not prevent arterial hypoxemia without oxygen
supplementation [14]. If a Vt of 6 mL/kg were to be applied
with conventional ventilation with PEEP levels necessary to
optimize respiratory system compliance, it is likely that the
resulting airway pressure would compromise cardiopulmo-
nary function.
In addition to low Vt, protective lung ventilation also may
include a low ventilator rate, a low FiO2, and high levels of
CPAP. Although well established, these factors are not widely
publicized and are relatively overlooked by the ARDS Network
and numerous pulmonary publications.
Anesthesiologists commonly believe that during general
anesthesia, a Vt >6 mL/kg is necessary to prevent atelectasis,
hypoxemia, and hypoventilation [15]. For instance, many an-
esthesiologists use a Vt of 10 mL/kg and a RR of 10 per minute
initially; then, during one-lung ventilation they decrease Vt,
increase RR by 50%, and increase inspired oxygen to 100%.
Because of the atelectasis that generally ensues, FiO2 is
frequently maintained at high levels, which may exacerbate the
atelectasis [16]. Our results do not support the need for
increased Vt and indeed reveal that a Vt of 6 mL/kg and a low RR
are capable of providing adequate oxygenation even with room
air, and adequate ventilation, as long as the Paw is sufficiently
high to prevent atelectasis and maintain the alveoli “open.”
This patient population was chosen for investigation
because one-lung ventilation in the lateral decubitus position
during lung resection provides the greatest challenge to anes-
thesiologists to prevent intraoperative arterial hypoxemia and
adequate ventilation [6]. We used static compliance as an in-
direct reflection of relative lung volume with regard to lung
collapse, atelectasis, and hyperinflation. The reduction in
compliance that almost always occurs during lung resections is
due in a large part to microatelectasis developing in the
dependent ventilated lung. However, during surgery, atelec-
tasis in the nonoperative lung is never directly visualized so
compliance provides an indirect but reliable indication of
atelectasis. Also at the end of the surgical case, the operative
lung is reexpanded under direct vision, but still enough time
passes before the patient is extubated to allow the occurrence
of significant atelectasis in the operative lung, especially if the
FiO2 is elevated with the usual accompanying absorption atel-
ectasis. The increase in compliance we observed with the OV
technique was most likely due to less atelectasis. If atelectasis
is reduced intraoperatively, it is reasonable to anticipate fewer
PPC, and this is what we plan to explore in a future much larger
randomized trial. The fact that no patient required any sup-
plemental O2 in the recovery room is a reflection that
247
pulmonary gas exchange approximated preoperative levels.
The ratio of SpO2/FiO2 is regarded to be a reflection of the gas
exchange efficiency of the lung [17]. If hypoxemia occurs during
one-lung ventilation with OV strategy, it most likely will be due
to right-to-left shunting of blood or mismatching of ventilation
and pulmonary perfusion. An increase in FiO2 will have mini-
mal effect on hypoxemia secondary to right-to-left shunting of
blood. Therefore, if an increase in inspired oxygen does in-
crease SpO2 significantly, it indicates a mismatching of V/Q and
the possibility that Paw is not sufficiently elevated, inspite of
optimal compliance [18,19]. In the single patient in our study
who developed right-to-left intrapulmonary shunting of blood,
increasing FiO2 was minimally effective in increasing SpO2. It is
likely that “shunted” blood flow from the operated, non-
dependant lung contributed to the relative hypoxemia. We
were able to maintain adequate SpO2 with increased FiO2
indicating that some V/Q mismatch was present as well.
Application of CPAP to the operative lung might have improved
oxygenation but was not used because it would have reex-
panded the lung and interrupted the operative procedure.
The “open lung” strategy applied in our investigation is
similar to that described by Bratzke et al. [20]. However, that
study used an anesthesia ventilator (bird Corp, Palm Springs,
CA) designed to provide airway pressure release ventilation
(APRV), which would allow spontaneous breathing with CPAP,
as well as mechanical ventilation with APRV. A recent animal
study by Roy et al. [21,22] using APRV in an anesthetized pig
model of acute lung injury demonstrated striking protection
against ARDS by using this ventilation technique immediately
after the lung injury compared with conventional low Vt
conventional ventilation. Their results suggested that APRV
protected the lungs by attenuating lung permeability,
inflammation, edema, and surfactant degradation.
Current anesthesia ventilators are not capable of providing
APRV, or intermittent CPAP, as described by Bratzke et al. [20].
Therefore, we attempted to closely mimic Bratzke’s method-
ology using pressure-controlled ventilation with an I:E ratio of
2:1e4:1. The “release” pressure was obtained by using the
anesthesia machine “PEEP” setting. The difference between
the “pressure control” setting and the “PEEP” setting deter-
mined the ventilating pressure or “release” pressure. The
“Paw” was the pressure control setting.
Regardless of the technique used, we used consistently
lower FiO2 levels than commonly recommended or necessary
for these procedures, both in the operating room and PACU.
Yet, significant arterial hypoxemia was never a problem in
any patient at any time. This finding has several implications.
Even during induction of general anesthesia and just before
tracheal extubation by maintaining exhaled oxygen at only
80%, we avoided total denitrogenation of the lungs to mini-
mize absorption atelectasis [23]. The use of low FiO2 (room air)
allowed us to use the pulse oximeter as a monitor of adequacy
of both ventilation and gas exchange [24]. It reassured us that
intrapulmonary shunting of blood was minimal, even with
one-lung completely unventilated. Interestingly, this may
mean that optimal inflation of the dependent lung causes less
perfusion of the collapsed, operated lung.
We selected 30 cm H2O as the Paw level to recruit
the atelectatic lung, although a recruitment maneuver
usually entails a higher-suggested airway pressure. In every
248 j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 2 ( 2 0 1 4 ) 2 4 2 e2 4 9
patient, we observed an increase in respiratory system
compliance as we decreased the Paw level, indicating that at
Paw ¼ 30 cm H2O we not only recruited all possible atelectatic
lung but hyperinflated some lung regions, making them less
compliant. However, as we decreased the Paw, at a critical
point, compliance again decreased, indicating that we had
passed the inflection point on the pressure:volume curve. We
set the Paw level to 4 cm H2O above that point and determined
that to be the level to produce optimal compliance.
The level of Paw required to maximize compliance was
higher than the airway pressure used to ventilate patients
with CV. Therefore, the airway pressure applied during CV
was inadequate to open some atelectatic lung areas in the
dependent lung, even transiently. Such areas likely would
remain collapsed throughout the entire operative procedure.
Furthermore, it is likely that many areas of the lung collapse
during exhalation, open during the positive pressure breath,
and collapse again during exhalation. This repeated alveolar
collapse and expansion (RACE) has been shown to be injurious
in experimental lung injury models [14].
The first study designed to test intraoperative protective
anesthesia ventilation techniques was a small randomized trial
in open abdominal surgery patients published by Severgnini
et al. [10] in 2013. Their protective ventilation strategy used
lower Vts (7 mL/kg), 10 cm H20 PEEP, and various recruitment
maneuvers. They documented improved postoperative pul-
monary function compared with control ventilation tech-
niques but the study was underpowered to determine whether
protective ventilation may decrease pulmonary complications.
Of particular note, they did document that there were
depressed pulmonary function and oxygenation post-
operatively as well as increasing atelectasis in the control pa-
tients (low Vt and no PEEP) for long as 5 d but not in the
protective ventilation group. Recent studies in open-chest
rabbits ventilated with Vts of 8e12 mL/kg and no PEEP
revealed histologic damage to peripheral airways, and inflam-
mation likely related to repeated alveolar collapse and reex-
pansion (RACE) [25]. This phenomenon may be an
unrecognized complication in patients undergoing general
anesthesia [26].
Our phase I pilot study in lung resection patients was
designed to test the feasibility and safety of the use of a pro-
tective open lung ventilation strategy and consisted of only 12
patients. We now plan to follow this study with a much larger
randomized trial of lobectomies by minimally invasive (VATS
and robotic-assisted) and open thoracotomy techniques that
is powered to provide more definitive data about the potential
for reduction of PPC using OV.
5. Conclusions
Patients undergoing one-lung ventilation for elective lung
resection with currently recommended ventilation techniques
develop significant intraoperative atelectasis. However, using
an open lung ventilation strategy during elective lung resection
surgery appears to be safe and effective, will correct and pre-
vent intraoperative atelectasis, and potentially may have a
beneficial effect on the postoperative course of these patients
who are at high risk for pulmonary complications.
Acknowledgment
Authors’ contributions: J.B.D., L.A.R., M.L.S., D.T., R.R.R., and
J.O.R. were involved in the conception and design of the study.
J.B.D., L.A.R., M.L.S., and D.T. did collection of data. J.B.D.,
L.A.R., R.R.R., and J.O.R. did the analysis and interpretation of
data. J.B.D., L.A.R., and R.R.R. did the writing of the article.
J.B.D., L.A.R., M.L.S., D.T., R.R.R., and J.O.R. did the revision of
the article with approval of the final version of the article.
Financial support for study: Department of Anesthesiology
unrestricted research funds and the Hoenle Foundation, Sar-
asota, FL thoracic surgery research funds.
Disclosure
The authors reported no proprietary or commercial interest in
any product mentioned or concept discussed in this article.
None of the authors have received any personal or financial
support, nor are they involved with an organization that has
financial interest in the subject matter.
references
[1] Shander A, Fleisher L, Barie P, Bigatello L, Sladen R,
Watson C. Clinical and economic burden of postoperative
pulmonary complications: patient safety summit on
definition, risk-reducing interventions and preventative
strategies. Crit Care Med 2011;39:2163.
[2] Futier E, Constantin J, Paugam-Burtz C, et al. A trial of
intraoperative low-tidal-volume ventilation in abdominal
surgery. N Engl J Med 2013;369:428.
[3] Smetana GW. Preoperative pulmonary evaluation. N Engl J
Med 1999;340:937.
[4] Linde-Zwirble WL, Bloom JD, Mecca RS, Hansell DM.
Postoperative pulmonary complications in adult elective
surgery patients in the US: severity, outcomes, and resource
use. Crit Care 2010;14(Suppl 1):P210.
[5] Rock P, Rich PR. Postoperative pulmonary complications.
Curr Opin Anesthesiol 2003;16:123.
[6] Agostini P, Cieslik H, Rathinam S, et al. Postoperative
pulmonary complications following thoracic surgery: are
there any modifiable factors? Thorax 2010;65:815.
[7] Strandberg A, Tokics L, Brismar B. Atelectasis during
anaesthesia and in the postoperative period. Acta
Anesthesiol Scand 1986;30:154.
[8] Maslow A, Stafford T, Davignon K, Ng T. A randomized
comparison of different ventilator strategies during
thoracotomy for pulmonary resection. J Thorac Cardiovasc
Surg 2013;146:38.
[9] Roze H, Lafargue M, Outtara A. Case scenario: management
of intraoperative hypoxemia during one-lung ventilation.
Anesthesiology 2011;114:167.
[10] Severgnini P, Selmo G, Lanza C, et al. Protective mechanical
ventilation during general anesthesia for open abdominal
surgery improves postoperative pulmonary function.
Anesthesiology 2013;118:1307.
[11] Kent M, Wang T, Whyte R, Curran T, Flores R,
Gangadharan S. Open, video-assisted thoracic surgery and
robotic lobectomy: review of a national database. Ann
Thorac Surg 2014;97:236.
 j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 2 ( 2 0 1 4 ) 2 4 2 e2 4 9
[12] Ashizawa K, Hayashi K, Aso N, Minami K. Lobar atalectasis:
diagnostic pitfalls on chest radiography. Br J Radiol 2001;74:
89.
[13] Ventilation with lower tidal volumes as compared with
traditional tidal volumes for acute lung injury and the acute
respiratory distress syndrome. The Acute Respiratory
Distress Syndrome Network. N Engl J Med 2000;342:1301.
[14] Schiller H, McCann UG 2nd, Carney D, Gatto L, Steinberg J,
Nieman G. Altered alveolar mechanics in the acutely injured
lung. Crit Care Med 2001;29:1049.
[15] Bendixen HH, Hedley-Whyte J, Laver MB. Impaired
oxygenation in surgical patients during general anesthesia
with controlled ventilationda concept of atelectasis. N Engl J
Med 1963;269:991.
[16] Magnusson L, Spahn DR. New concepts of atelectasis during
general anaesthesia. Br J Anaesth 2003;91:61.
[17] Rice TW, Wheeler AP, Bernard GR, Hayden DL, Schoenfeld DA,
Ware LB. National Institutes of Health, National Heart, Lung,
and Blood Institute ARDS Network. Comparison of the SpO2/
FIO2 ratio and the PaO2/FIO2 ratio in patients with acute lung
injury or ARDS. Chest 2007;132:410.
[18] Douglas ME, Downs JB, Dannemiller FJ, Hodges MR,
Munson ES. Change in pulmonary venous
admixture with varying inspired oxygen. Anesth Analg 1976;
55:688.
249
[19] Downs J. Has oxygen administration delayed appropriate
respiratory care? fallacies regarding oxygen therapy. Respir
Care 2003;48:611.
[20] Bratzke E, Downs JB, Smith RA. Intermittent CPAP: a new
mode of ventilation during general anesthesia.
Anesthesiology 1998;89:334.
[21] Roy S, Sadowitz B, Andrews P, et al. Early stabilizing alveolar
ventilation prevents acute respiratory distress syndrome: a
novel timing-based ventilatory intervention to avert lung
injury. J Trauma Acute Care Surg 2012;73:391.
[22] Roy S, Habashi N, Sadowitz B, et al. Early airway pressure
release ventilation prevents ARDSda novel preventive
approach to lung injury. Shock 2013;39:28.
[23] Benoit Z, Wicky S, Fischer JF, et al. Effects of increased FiO2
before tracheal extubation on postoperative atelectasis.
Anesth Analg 2002;95:1777.
[24] Fu ES, Downs JB, Schweiger JW, Miguel RV, Smith RA.
Supplemental oxygen impairs detection of hypoventilation
by pulse oximetry. Chest 2004;126:1552.
[25] D’Angelo E, Peccharia M, Gentile G. Dependence of lung
injury on surface tension during low-volume ventilation in
normal open-chest rabbits. J Appl Physiol 2007;102:174.
[26] Hedenstierna G. Alveolar collapse and closure of airways:
regular effects of anesthesia. Clin Physiol Funct Imaging
2003;23:123.