1249152

research-article20242024
TAR0010.1177/17534666241249152Therapeutic Advances in Respiratory DiseaseS Tongyoo, T Viarasilpa

Therapeutic Advances in
Respiratory Disease Original Research

Comparison of limited driving pressure

Ther Adv Respir Dis

2024, Vol. 18: 1–14

ventilation and low tidal volume strategies DOI: 10.1177/

https://doi.org/10.1177/17534666241249152
https://doi.org/10.1177/17534666241249152
17534666241249152

in adults with acute respiratory failure

© The Author(s), 2024.

Article reuse guidelines:

sagepub.com/journals-

on mechanical ventilation: a randomized permissions

controlled trial
Surat Tongyoo , Tanuwong Viarasilpa, Phitphiboon Deawtrakulchai,
Santi Subpinyo, Chaiyawat Suppasilp and Chairat Permpikul

Abstract
Background: Ventilator-induced lung injury (VILI) presents a grave risk to acute respiratory
failure patients undergoing mechanical ventilation. Low tidal volume (LTV) ventilation has
been advocated as a protective strategy against VILI. However, the effectiveness of limited
driving pressure (plateau pressure minus positive end-expiratory pressure) remains unclear. Correspondence to:
Surat Tongyoo
Objectives: This study evaluated the efficacy of LTV against limited driving pressure in Faculty of Medicine,
Siriraj Hospital, Mahidol
preventing VILI in adults with respiratory failure. University, 2, Prannok
Design: A single-centre, prospective, open-labelled, randomized controlled trial. Road, Bangkok Noi,
Bangkok 10700, Thailand
Methods: This study was executed in medical intensive care units at Siriraj Hospital, Mahidol surat.ton@mahidol.ac.th
University, Bangkok, Thailand. We enrolled acute respiratory failure patients undergoing Tanuwong Viarasilpa
Santi Subpinyo
intubation and mechanical ventilation. They were randomized in a 1:1 allocation to limited Chairat Permpikul
driving pressure (LDP; ⩽15 cmH2O) or LTV (⩽8 mL/kg of predicted body weight). The primary Division of Critical Care
Medicine, Department
outcome was the acute lung injury (ALI) score 7 days post-enrolment. of Medicine, Faculty of
Medicine, Siriraj Hospital,
Results: From July 2019 to December 2020, 126 patients participated, with 63 each in the LDP Mahidol University,
and LTV groups. The cohorts had the mean (standard deviation) ages of 60.5 (17.6) and 60.9 Bangkok, Thailand
Phitphiboon
(17.9) years, respectively, and they exhibited comparable baseline characteristics. The primary Deawtrakulchai
reasons for intubation were acute hypoxic respiratory failure (LDP 49.2%, LTV 63.5%) and Division of Critical Care
Medicine, Department
shock-related respiratory failure (LDP 39.7%, LTV 30.2%). No significant difference emerged of Medicine, Faculty of
in the primary outcome: the median (interquartile range) ALI scores for LDP and LTV were Medicine, Siriraj Hospital,
Mahidol University,
1.75 (1.00–2.67) and 1.75 (1.25–2.25), respectively (p = 0.713). Twenty-eight-day mortality Bangkok, Thailand

rates were comparable: LDP 34.9% (22/63), LTV 31.7% (20/63), relative risk (RR) 1.08, 95% Subdivision of Critical
Care, Division of Internal
confidence interval (CI) 0.74–1.57, p = 0.705. Incidences of newly developed acute respiratory Medicine, Faculty of
Medicine, Khon Kaen
distress syndrome also aligned: LDP 14.3% (9/63), LTV 20.6% (13/63), RR 0.81, 95% CI 0.55– University, Khon Kaen,
1.22, p = 0.348. Thailand

Conclusions: In adults with acute respiratory failure, the efficacy of LDP and LTV in averting Chaiyawat Suppasilp
Department of Clinical
lung injury 7 days post-mechanical ventilation was indistinguishable. Epidemiology and
Biostatistics, Faculty of
Medicine, Ramathibodi
Clinical trial registration: The study was registered with the ClinicalTrials.gov database Hospital, Mahidol
University, Bangkok,
(identification number NCT04035915). Thailand

journals.sagepub.com/home/tar 1

Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License
(https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission
provided the original work is attributed as specified on the Sage and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).
Therapeutic Advances in
Respiratory Disease Volume 18

Plain language summary

Limited breathing pressure or low amount of air given to the lung; which one is better
for adults who need breathing help by ventilator machine

We conducted this research at Siriraj Hospital in Bangkok, Thailand, aiming to compare

two ways of helping patients with breathing problems. We studied 126 patients who were
randomly put into two groups. One group received a method where the pressure during
breathing was limited (limited driving pressure: LDP), and the other group got a method
where the amount of air given to the lungs was kept low (low tidal volume: LTV). We
checked how bad the lung injury was at seven days later. The results showed that there
was no difference between the two methods. Both ways of helping patients breathe had
similar outcomes, and neither was significantly better than the other in preventing lung
problems. The study suggests that both approaches work about the same for patients who
need help with breathing using a machine.

Keywords: acute lung injury, acute respiratory failure, driving pressure, low tidal volume,
ventilator-induced lung injury

Received: 6 January 2024; revised manuscript accepted: 4 April 2024.

Background pressure correlates with increased pulmonary

Invasive mechanical ventilation, frequently complications in both ARDS and non-ARDS
employed to assist critically ill patients, is increas- perioperative patients.10,11 By contrast, dimin-
ingly identified as a potential factor exacerbating ished driving pressure has been linked to improved
lung injury. Injuries may arise through various survival outcomes in ARDS patients.12 However,
mechanisms, encompassing barotrauma, volu- a definitive consensus on whether LDP can pre-
trauma and biophysical stress.1 A lung-protective vent VILI in mechanically ventilated patients
ventilation strategy termed ‘low tidal volume’ without ARDS remains elusive. Additionally,
(LTV) has been established as an effective information is limited on whether to choose the
approach to curtailing ventilator-induced lung LDP or the LTV strategy to prevent VILI in non-
injury (VILI). The strategy focuses on averting ARDS patients with acute respiratory failure who
alveolar overdistension and minimizing volu- need mechanical ventilation.
trauma. It has demonstrated improved outcomes
in patients needing mechanical ventilation, espe- This research aimed to compare the efficacy of
cially those diagnosed with acute respiratory dis- the LDP and LTV approaches in preventing VILI
tress syndrome (ARDS).2,3 among non-ARDS patients with acute respiratory
failure who require mechanical ventilation. By
Nevertheless, the efficacy of the LTV strategy in assessing these methodologies, we sought to gain
mechanically ventilated patients without ARDS insights into the optimal ventilation approach for
remains inconclusive. Data from small rand- preventing lung injury in this patient population.
omized controlled trials and meta-analyses have
suggested that the LTV approach correlates with
fewer pulmonary complications and reduced Methods
dependency on mechanical ventilation.4–8 By
contrast, a recent large randomized controlled Study design
trial found no significant benefits.9 This prospective, randomized, unblinded clinical
trial was conducted at the medical intensive care
A parallel strategy to mitigating VILI involves unit (ICU) of Siriraj Hospital, Mahidol University,
ventilation with limited driving pressure (LDP). Bangkok, Thailand. The study enrolled partici-
Emerging evidence posits that elevated driving pants between 25 July 2019 and 31 December

2 journals.sagepub.com/home/tar
 S Tongyoo, T Viarasilpa et al.

Figure 1. Flow diagram illustrating the screening, enrolment and randomization of the patients.

2020. The coinvestigators performed all partici- •• Were suspected of needing invasive
pant screening and enrolment procedures (Figure mechanical ventilation for <48 h. This cri-
1). Before enrolling patients, written informed terion is specific for patients who require
consent for participation was obtained from the endotracheal intubation for airway protec-
patients or, in cases where patients were unable to tion during an invasive procedure.
provide consent, their next of kin or legal guardi- •• Had a tracheostomy tube
ans. The principal investigator and a statistician •• Were pregnant or actively lactating
conducted the outcome evaluation data analysis, •• Had a do-not-resuscitate decision
with both blinded to the patient’s treatment •• Had extracorporeal membrane oxygenation
group. The Siriraj Critical Care Research Fund support before randomization
funded the trial. The funder had no role in the
study design, data analysis or outcome assess- After obtaining informed consent, we recorded
ment. The reporting of this study conforms to the the patients’ baseline characteristics, baseline
CONSORT (Consolidated Standards of mechanical ventilator settings and lung mechani-
Reporting Trials) statement.13 cal parameters. Based on the underlying patho-
physiology, we classified acute respiratory failure
requiring intubation into four categories. The cat-
Participants egories were type 1, acute hypoxic respiratory fail-
We screened all patients 18 years or older admitted ure; type 2, acute hypercapnic respiratory failure;
to the medical ICU with acute respiratory failure. type 3, acute respiratory failure caused by periop-
Patients were eligible if they had been intubated erative atelectasis; and type 4, acute respiratory
and placed on mechanical ventilation within 24 h failure during shock or hypoperfusion.15,16
before enrolment. We excluded patients who:
The acute lung injury (ALI) score was calculated
•• Were suspected of having a high risk of to assess the severity of lung injury. The compu-
death within 48 h after randomization, this tation was based on a patient’s oxygenation sta-
depends on the judgement of the attending tus, chest X-ray results, positive end-expiratory
physicians, which is mostly based on the pressure (PEEP) level and respiratory compliance
patient’s hemodynamic parameters during measurements. Each component was assigned a
the screening process. score ranging from 0 (no injury) to 4 (the most
•• Had ARDS as defined by the Berlin crite- severe injury), and their average was subsequently
ria14 before randomization used as the ALI score17 (Supplemental Table 1).

journals.sagepub.com/home/tar 3
Therapeutic Advances in
Respiratory Disease Volume 18

To calculate the baseline score, we used the worst 0.91 × [height (cm) − 152.4] for men and
information for each component during the 24 h 45.5 + 0.91 × [height (cm) − 152.4] for women.2
before enrolment. The ventilator settings were periodically adjusted
by dropping the inspiratory pressure by 1 cmH2O
(PCV) or by reducing the tidal volume by 50 mL
Randomization (VCV) every 15–30 min. The goal was to consist-
Patients providing informed consent were ently maintain the tidal volume at ⩽8 mL/kg. As
enrolled in the study and allocated randomly to with the LDP group, the respiratory rate, PEEP
the LDP or LTV ventilation strategy group. and FiO2 were flexible, allowing maintenance of
Randomization was achieved using a computer- oxygen saturation at ⩾94% and pH within the
generated table sourced from www.randomiza- range of 7.35–7.45.
tion.com. The process adopted a 1:1 ratio,
employing blocks of four opaque, pre-numbered, In both cohorts, sedatives, analgesic agents or
sealed envelopes. These envelopes were dis- muscle relaxants were used to counteract patient-
patched to the participating ICUs, with the ICU ventilator asynchrony at the attending physician’s
allocations based on the computer-generated ran- discretion. All patients received standard care for
dom number table. The randomization process acute respiratory failure and treatments tailored
was overseen by an investigator (S.T.) who was to their individual ICU admission diagnoses for
excluded from patient enrolment and manage- acute critical illnesses.
ment. Although other investigators, the patients
and their families remained blinded to the study
allocation, the attending clinicians and nursing Outcome assessments
staff were cognizant of each patient’s assignment The primary outcome of this study was the ALI
to the LDP or LTV ventilation group. score on day 7. The score was derived from the
most severe measurements of each patient’s oxy-
genation, chest X-ray results, PEEP level and res-
Study intervention piratory compliance on day 7. For patients who
Post-randomization, patients in the LDP group died before day 7, an ALI score of 4 was assigned.
underwent ventilation targeted at a driving pres- The secondary outcomes were 28-day mortality
sure of ⩽15 cmH2O. When patients were venti- and the onset of ARDS within 28 days post-enrol-
lated in pressure-controlled ventilation (PCV) ment. The Berlin criteria were employed for
mode, the driving pressure corresponded to the ARDS diagnoses. We also recorded rates of new-
applied inspiratory pressure. When volume-con- onset pneumothorax, ventilator-associated pneu-
trolled ventilation (VCV) was utilized, an inspira- monia, lung atelectasis and hypoventilation. The
tory hold lasting between 0.5 and 1 s was instigated calculation of mechanical ventilator support-free
until the inspiratory pressure plateaued. The dif- days up to day 28 followed the formula suggested
ference between the plateau inspiratory pressure by Russell et al.18 The on-duty radiologist con-
and the PEEP equated to the measured driving ducted the chest X-ray interpretation, which
pressure. Ventilator settings were modulated to included the extent of abnormal pulmonary infil-
maintain the driving pressure at 15 cmH2O or tration as part of the ALI score calculation and
below. The inspiratory pressure was gradually the detection of pneumothorax and atelectasis.
decreased by 1 cmH2O (PCV mode), or the tidal The radiologist was blinded to patient enrolment
volume was reduced by 50 mL (VCV mode) at and group assignments.
15- to 30-min intervals. Adjustments to respira-
tory rate, PEEP and the fraction of inspired oxy-
gen (FiO2) were permitted to sustain an oxygen Statistical analyses
saturation level ⩾94% and a pH between 7.35 We postulated that a difference of 1.5 points in
and 7.45 during mechanical ventilation. the ALI scores on day 7 could influence patient
outcomes. Enrolling at least 63 participants per
For patients in the LTV group, ventilation tar- group would afford an 80% power to discern a
geted a tidal volume of ⩽8 mL/kg of predicted 1.5-point difference in the ALI score on day 7
body weight (PBW). To compute the PBW, the between groups, with a two-sided alpha error of
following equations were employed: 50 + 0.05.

4 journals.sagepub.com/home/tar
 S Tongyoo, T Viarasilpa et al.

Continuous variables underwent normality test- distribution of baseline ALI scores and the asso-
ing using the Kolmogorov–Smirnov test. ciation of baseline ALI scores and hospital mor-
Normally distributed variables are presented as tality is shown in Supplemental Table 2.
the means [standard deviation (SD)] and were
assessed using a t-test. Variables deviating from a Upon enrolment, most patients in both groups
normal distribution are shown as medians and were under PCV mode: LDP 93.7% (59/63),
interquartile ranges (IQRs) and were analysed LTV 88.9% (56/63), p = 0.53. As shown in Table
using the Wilcoxon rank-sum test. Categorical 2, the baseline respiratory parameters of the LDP
variables are depicted by frequency and percent- and LTV groups were comparable. Specifically,
age. Depending on suitability, the chi-square test there were no significant differences in tidal vol-
or Fisher’s exact test was employed. The primary ume per PBW [mean (SD) 8.6 (2.1) versus 8.3
outcome was subjected to a t-test. Secondary out- (2.3) mL/kg, p = 0.45], driving pressure [mean
comes were evaluated with the chi-square test (SD) 17.2 (4.1) versus 17.9 (4.2) cmH2O,
and are presented as relative risk (RR) with a 95% p = 0.31] or peak inspiratory pressure [median
confidence interval (CI). Kaplan–Meier curves (IQR) 23 (20–26) versus 24 (20–26) cmH2O,
were used to appraise mortality outcome days, p = 0.47]. Baseline arterial blood gas analyses for
followed by a log-rank test comparison at 28 days. arterial blood pH, partial pressure of oxygen in
For mortality assessment, the 28-day mortality arterial blood (PaO2), oxygen saturation, partial
was computed from the enrolment date. All pri- pressure of carbon dioxide in arterial blood
mary and secondary outcome analyses adhered to (PaCO2) and the PaO2:FiO2 ratio were consistent
the intention-to-treat principle, and p values across both groups (Table 2). An analogous num-
<0.05 denoted statistical significance. Data anal- ber of patients in each group were administered
yses were executed using PASW Statistics, ver- vasopressors, inotropes, sedatives and paralytic
sion 18 (SPSS Inc, Chicago, IL, USA). The study agents. Renal replacement therapy was needed
was registered with the ClinicalTrials.gov data- for 38.1% (24/63) of the LDP group and 25.4%
base (identification number NCT04035915). (16/63) of the LTV group (p = 0.13, Table 2).

After randomization, patients in the LDP group

Results exhibited a higher tidal volume per PBW than
We screened a total of 230 patients requiring those in the LTV group, although the difference
invasive mechanical ventilator support. Of these, was not statistically significant [Figure 2(a)].
126 met the inclusion criteria and were rand- There was no difference in driving pressure
omized to the LDP or LTV treatment group. between the LDP and LTV groups [Figure 2(b)].
Every randomized patient was monitored until There were 19 (30.2%) patients in LDP and 33
the study’s conclusion and included in the final (52.4%) patients in LTV who had tidal volume
intention-to-treat analysis (Figure 1). Patients’ per PBW lower than 8 mL/kg at all times after
baseline characteristics – age, sex, PBW, underly- study enrolment (p = 0.01). The percentage of
ing conditions and disease severity – were compa- timing that patients had tidal volume per PBW
rable between the two groups (Table 1). The higher than 8 mL/kg within 7 days after enrolment
primary reason for intubation was hypoxemic res- was significantly higher among LDV than LTV
piratory failure: LDP 49.2% (31/63), LTV 63.5% group (7.14 + 7.30 versus 4.05 + 5.88, p = 0.01).
(40/63), p = 0.11. Sepsis or septic shock was the The proportion of patients who had driving pres-
predominant ICU admission diagnosis for both sure lower than 15 cmH2O [21 (33.3%) versus 19
groups: 87.3% (55/63) in each group (p = 1.00). (30.2%), p = 0.70] and the percentage of timing
Pneumonia was the leading infection cause: LDP that patients had driving pressure higher than
44.4% (28/63), LTV 47.6% (30/63), p = 0.72. 15 cmH2O (6.67 + 6.87 versus 7.57 + 7.10,
The median duration from intubation to enrol- p = 0.47) were not different between groups
ment was 8 h 30 min (4:00–12:00) for the LDP (Table 2).
group and 7 h 30 min (3:25–10:45) for the LTV
group. The mean (SD) baseline ALI scores were The primary outcome, the ALI score on day 7,
1.83 (0.68) for the LDP group and 1.75 (0.72) showed no significant difference between the LDP
for the LTV group, with p = 0.51 (Table 1). The and LTV groups [mean (SD) 1.90 (1.14) versus

journals.sagepub.com/home/tar 5
Therapeutic Advances in
Respiratory Disease Volume 18

Table 1. Baseline characteristics of the patients.

Baseline characteristics Limited driving Limited tidal volume p

pressure group (n = 63) group (n = 63)

Age, mean (SD), years 60.5 (17.6) 60.9 (17.9) 0.90

Male sex, no. (%) 32 (50.8) 34 (54.0) 0.72

Predicted body weight,a mean (SD), kg 57.3 (11.1) 57.0 (9.8) 0.86

APACHE II score,b median (IQR) 26 (17–30) 24 (17–30) 0.476

SOFA score,c median (IQR) 11 (8–14) 9 (7–12) 0.21

Acute lung injury score, mean (SD) 1.83 (0.68) 1.75 (0.72) 0.51

Time from intubation to enrolment, 8:30 (4:00–12:00) 7:30 (3:25–10:45) 0.48

median (IQR), h:min

Comorbidities, no. (%)

Hypertension 37 (58.7) 32 (50.8) 0.37

Diabetes mellitus 23 (36.5) 24 (38.1) 0.85

Chronic kidney disease 19 (30.2) 19 (30.2) 1.00

Malignancy 20 (31.7) 13 (20.6) 0.16

Coronary artery disease 5 (7.9) 5 (7.9) 1.00

ICU admission diagnosis, no. (%)

Sepsis/septic shock 55 (87.3) 55 (87.3) 1.00

Site of infection

Pneumonia 28 (44.4) 30 (47.6) 0.72

Intra-abdominal infection 7 (11.1) 8 (12.7) 0.78

Urinary tract infection 5 (7.9) 3 (4.8) 0.72

Skin and soft tissue infection 5 (7.9) 7 (11.1) 0.76

Bacteraemia 14 (22.2) 16 (25.4) 0.68

   Congestive heart failure 8 (12.7) 8 (12.7) 1.00

Causes of respiratory failure, no. (%)

Hypoxemic respiratory failure 31 (49.2) 40 (63.5) 0.11

Shock-related respiratory failure 25 (39.7) 19 (30.2) 0.26

Hypercapnic respiratory failure 6 (9.5) 4 (6.3) 0.75
aPredicted body weight was calculated using these equations: 50 + 0.91 × [height (cm) − 152.4] for men and
45.5 + 0.91 × [height (cm) − 152.4] for women.
bAPACHE II scores, severity-determining scores, range from 0 to 71. Higher scores represent more severe disease.
cSOFA scores range from 0 to 24. Higher scores represent more organ failure.

APACHE, acute physiology and chronic health evaluation; ICU, intensive care unit; IQR, interquartile range; kg, kilogram;
SD, standard deviation; SOFA, sequential organ failure assessment.

6 journals.sagepub.com/home/tar
 S Tongyoo, T Viarasilpa et al.

Table 2. Baseline respiratory parameters of the patients at the time of enrolment and other treatments.

Baseline respiratory parameters Limited driving pressure Limited tidal volume p

group (n = 63) group (n = 63)

Mechanical ventilator mode, no. (%)

Pressure control ventilation 59 (93.7) 56 (88.9) 0.53

Volume control ventilation 4 (6.3) 7 (11.1) 0.53

Respiratory parameters, mean (SD) or median (IQR)

Tidal volume, mL 486.1 (135.3) 463.5 (124.5) 0.33

 Tidal volume per predicted body weight, 8.6 (2.1) 8.3 (2.3) 0.45
mL/kg

Peak inspiratory pressure, cmH2O 23 (20–26) 24 (20–26) 0.47

Positive end-expiratory pressure, cmH2O 5 (5–5) 5 (5–6) 0.31

Driving pressure, cmH2O 17.2 (4.1) 17.9 (4.2) 0.31

Respiratory rate, time per minute 18 (16–23) 20 (16–24) 0.22

FiO2 0.4 (0.4–0.6) 0.4 (0.4–0.6) 0.59

Arterial blood gas analysis, mean (SD)

pH 7.36 (0.10) 7.37 (0.11) 0.90

PaO2 115.8 (52.3) 128.4 (51.2) 0.18

O2 saturation 96.6 (2.8) 97.1 (3.2) 0.36

PaCO2 29.1 (8.7) 31.2 (10.1) 0.22

HCO3 17.4 (6.6) 18.9 (6.7) 0.20

PaO2:FiO2 ratio 248.4 (135.4) 270.1 (136.4) 0.38

Respiratory parameters after enrolment

 Tidal volume per predicted body weight 19 (30.2) 33 (52.4) 0.01

<8 mL/kg,a no. (%)

 Percent duration of tidal volume per 7.14 + 7.30 4.05 + 5.88 0.01
predicted body weight over 8 mL/kg,b
mean (SD)

Driving pressure <15 cmH2O,c no. (%) 21 (33.3) 19 (30.2) 0.70

 Percent duration of driving pressure over 6.67 + 6.87 7.57 + 7.10 0.47
15 cmH2O,d mean (SD)

Other treatments received

Vasopressor 48 (76.2) 42 (66.7) 0.24

Inotrope 12 (19.0) 10 (15.9) 0.64

Renal replacement therapy 24 (38.1) 16 (25.4) 0.13

(Continued)

journals.sagepub.com/home/tar 7
Therapeutic Advances in
Respiratory Disease Volume 18

Table 2. (Continued)

Baseline respiratory parameters Limited driving pressure Limited tidal volume p

group (n = 63) group (n = 63)

Sedative and paralytic agents

Midazolam 12 (19.0) 12 (19.0) 1.00

Fentanyl 36 (57.1) 45 (71.4) 0.14

Cis-atraculium 3 (4.8) 5 (7.9) 0.72
aProportion of patients who had tidal volume per predicted body weight <8 mL/kg during 7 days (study period) after
enrolment.
bDuration of tidal volume per predicted body weight over 8 mL/kg is calculated by the duration (h) that the patient had tidal

volume per predicted body weight over 8 mL/kg divided by 168 h.

cProportion of patients who had driving pressure <15 cmH O during 7 days (study period) after enrolment.
2
dDuration of driving pressure over 15 cmH O is calculated by the duration (h) that the patient had driving pressure over
2
15 cmH2O divided by 168 h.
cmH2O, centimetres of water; FiO2, fraction of inspired oxygen; HCO3, bicarbonate; IQR, interquartile range; mL/kg,
millilitres per kilogram; PaCO2, partial pressure of carbon dioxide in arterial blood; PaO2, partial pressure of oxygen in
arterial blood; SD, standard deviation.

1.89 (0.99), p = 0.95]. The result remained no sig- LDP 8.71 (10.44) days, LTV 10.22 (11.21) days,
nificant different after a sensitivity analysis, exclud- p = 0.44, Table 3. A subgroup analysis for hypox-
ing 15 patients who died before day 7, was emic respiratory failure was performed. The
performed [1.51 (0.74) versus 1.71 (0.80), p = 0.17]. results of the subgroup analysis are shown in
An increase in the ALI score was observed in 38.1% Supplemental Table 3. There was no significant
(24/63) of LDP patients and 34.9% (22/63) of difference in primary and secondary outcomes
LTV patients (p = 0.71). Figure 2(c) illustrates the among hypoxemic respiratory failure patients in
progression of the ALI score for both groups. The the LDP and the LTV groups.
rate of successful extubation at 14 days was compa-
rable between the groups: LDP 39.7% (25/63),
LTV 41.3% (26/63), p = 0.86. New-onset ARDS Discussion
appeared in 14.3% (9/63) of the LDP group com- Our randomized controlled trial studied adult
pared to 20.6% (13/63) in the LTV group (RR patients with acute respiratory failure who lacked
0.81, 95% CI 0.55–1.22, p = 0.35). Other adverse ARDS and required mechanical ventilation sup-
events linked with mechanical ventilator support, port. We found no significant difference in the
such as pneumothorax, ventilator-associated pneu- lung injury scores of the LDP and LTV strategy
monia, atelectasis and carbon dioxide (CO2) reten- groups on day 7. Similarly, the rates of pulmonary
tion, showed no significant differences between the complications – new-onset ARDS, pneumotho-
groups (Table 3). rax, ventilator-associated pneumonia and atelec-
tasis – exhibited no significant variations between
The 28-day mortality rate was 34.9% (22/63) for the groups. The 28-day and in-hospital mortality
the LDP group and 31.7% (20/63) for the LTV rates also showed no notable variance between
group (RR 1.08, 95% CI 0.74–1.57, p = 0.71, the two groups.
Table 3). The Kaplan–Meier curves for 28-day
mortality are depicted in Figure 3. The ICU This study used a driving pressure threshold of
mortality rates showed no significant difference: ⩽15 cmH2O for the LDP group and a tidal volume
LDP 36.5% (23/63), LTV 38.1% (24/63), RR threshold of ⩽8 mL/kg for the LTV group. These
0.97, 95% CI 0.68–1.38, p = 0.85. In-hospital values were selected because they aligned with pre-
mortality rates were also consistent: LDP 50.8% vious research findings.9,10,19 Past observations
(32/63), LTV 52.4% (33/63), RR 0.97, 95% have shown that a driving pressure increase exceed-
CI 0.68–1.37, p = 0.86, Table 3. The mean (SD) ing 15 cmH2O is linked to higher hospital mortality
number of days alive and free from mechanical among ARDS patients. The earlier studies also
ventilation up to day 28 showed no difference: found that non-ARDS patients ventilated with an

8 journals.sagepub.com/home/tar
 S Tongyoo, T Viarasilpa et al.

Figure 2. Comparison of the tidal volumes (a), driving pressures (b) and acute lung injury scores (c) of the
limited driving pressure and low tidal volume groups.

LTV strategy (6–8 mL/kg) had similar outcomes to Our results support the proposition of targeting
those using an intermediate tidal volume strategy lower driving pressure to prevent VILI10,18 and
(8–10 mL/kg). Specifically, there was no significant suggest that it is an alternative to the lower tidal
difference in the number of days they were alive volume strategy. Considering the heightened het-
without mechanical ventilation at 28 days.9,10,19 erogeneity of lung parenchyma pathology in

journals.sagepub.com/home/tar 9
Therapeutic Advances in
Respiratory Disease Volume 18

Table 3. Primary and secondary outcomes.

Outcomes Limited driving Limited tidal Relative risk p

pressure group volume group (95% CI)
(n = 63) (n = 63)

Primary outcome, no. (%)

Acute lung injury score on day 7, mean (SD) 1.90 (1.14) 1.89 (0.99) 0.95

Secondary outcome, no. (%)

ICU mortality 23 (36.5) 24 (38.1) 0.97 (0.68–1.38) 0.85

28-Day mortality 22 (34.9) 20 (31.7) 1.08 (0.74–1.57) 0.71

In-hospital mortality 32 (50.8) 33 (52.4) 0.97 (0.68–1.37) 0.859

ICU length of stay, median (IQR), days 11 (6–20) 13 (6–22) 0.96

Hospital length of stay, median (IQR), days 23 (12–46) 19 (11–39) 0.56

 Day alive and free of mechanical ventilator to 8.71 (10.44) 10.22 (11.21) 0.44
day 28, mean (SD), days

Extubation at 14 days 25 (39.7) 26 (41.3) 0.97 (0.68–1.47) 0.86

 Acute lung injury score on day 7 changes from 0.07 (1.20) 0.14 (1.07) 0.73
baseline, mean (SD)

 Acute lung injury score on day 7 increased from 24 (38.1) 22 (34.9) 1.07 (0.74–1.55) 0.71
the baseline

Adverse events, no. (%)

New-onset ARDS 9 (14.3) 13 (20.6) 0.81 (0.55–1.22) 0.35

Pneumothorax 2 (3.2) 1 (1.6) 1.51 (0.30–7.56) 1.00

Ventilator-associated pneumonia 7 (11.1) 11 (17.5) 0.79 (0.52–1.20) 0.31

Atelectasis 3 (4.8) 1 (1.6) 2.03 (0.37–11.20) 0.62

CO2 retention 3 (4.8) 7 (11.1) 0.69 (0.44–1.08) 0.33

Cumulative adverse events 19 (30.2) 24 (38.1) 0.84 (0.59–1.20) 0.35

ARDS, acute respiratory distress syndrome; CI, confidence interval; CO2, carbon dioxide; ICU, intensive care unit; IQR, interquartile range; SD,
standard deviation.

non-ARDS patients relative to ARDS patients,20 predominantly on PBW-based calculations, the

constraining the driving pressure could mitigate LDP approach offers a more physiologically con-
the mechanical stress on preserved lung paren- gruent method to guard against VILI.
chyma. Reducing this stress is particularly impor-
tant as the inconsistency in lung expansion in Another potential advantage of the LDP over the
non-ARDS patients results in regional strains. LTV strategy is its allowance for variations and
The strains are notably more severe than the potentially higher tidal volumes. This attribute is
overall strains seen in the healthy and inflamed particularly beneficial in response to an escalating
lungs of such patients. By adopting a lower driv- respiratory drive, especially in patients with meta-
ing pressure, we can preclude overdistension of bolic acidosis or shock-related respiratory failure.
preserved alveoli, thereby reducing the risk of When patients are ventilated using an LTV strat-
VILI. Conversely, while the LTV strategy relies egy, their inability to augment tidal volume might

10 journals.sagepub.com/home/tar
 S Tongyoo, T Viarasilpa et al.

utility and pertinence of the ALI score for both

non-ARDS and ARDS patients with acute res-
piratory failure.

Several limitations of this study merit attention.

Firstly, we could not blind attending physicians
and nurses to the intervention, introducing poten-
tial study bias. To address this, we designated an
investigator who was blinded to the intervention
allocations to evaluate patient outcomes.
Figure 3. Kaplan–Meier analysis of 28-day mortality. Secondly, our decision to use an ALI difference
There was no difference between the limited driving of 1.5 on day 7 for sample size calculation may
pressure and low tidal volume groups (log-rank have resulted in the study being underpowered to
p = 0.71). detect more subtle differences with the given
study population size. In a multicenter trial with
larger population size, control of the rate is cer-
lead to an amplified respiratory drive, which, in tainly required to assess the potential benefit of
turn, can potentially result in patients’ self- low driving pressure in the prevention of VILI.
inflicted lung injury.21 To mitigate this, sedative Thirdly, as the investigation focused on non-
and analgesic agents or muscle relaxants might be ARDS patients with good respiratory compliance,
employed.22,23 However, while these drugs can its findings might not directly apply to patients
enhance patient–ventilator synchrony, their over- with poor respiratory compliance. Finally, we
use can lead to delirium, protracted dependency on employed the inspiratory pressure of patients on
mechanical ventilation, weaning difficulties and PCV as a surrogate for driving pressure. This
extubation failure.24–26 Our findings revealed a approach might have underestimated the driving
marginally elevated use of the sedative and analge- pressure, especially in patients exerting high
sic agent fentanyl in the LTV group compared with inspiratory efforts. Notwithstanding this potential
the LDP group, although the difference was non- shortcoming, using inspiratory pressure to titrate
significant. Crucially, this variance in medication the driving pressure remains a practicable method
administration did not significantly influence the in bedside clinical practice.
extubation rate at 14 days or the number of days
alive without mechanical ventilation up to day 28.
Additionally, no significant differences were noted Conclusion
in the use of muscle relaxants by the two groups. In adults with acute respiratory failure but with-
out ARDS, there was no difference in the efficacy
Our study has several noteworthy strengths. of LDP and LTV in averting lung injuries 7 days
Firstly, it is a prospective randomized controlled after mechanical ventilation. Specifically, no vari-
trial, which is considered a robust design for eval- ation was observed in the prevention of injuries
uating interventions. Moreover, a blinded investi- when comparing an LDP of ⩽15 cmH2O to an
gator conducted the outcome evaluations, lending LTV of <8 mL/kg of PBW.
greater objectivity to the results. Another merit is
the adoption of the ALI score as the primary out-
come. This metric integrates four clinical param- Declarations
eters that are ubiquitously employed globally in
clinical practice. The use of the ALI score is Ethics approval and consent to participate
therefore feasible even in resource-constrained The study adhered to the ethical principles of the
settings, as it draws on easily accessible clinical Declaration of Helsinki, and its protocol was
data. Although introduced several decades ago, authorized by the Siriraj Institutional Review
the sustained employment of the ALI score in Board (approval number Si 828/2016). Before
acute respiratory failure, ARDS and emergent enrolling patients, written informed consent for
conditions such as COVID-19 attests to its worth participation was obtained from the patients or,
in severity evaluation and prognostic predic- in cases where patients were unable to provide
tions.27,28 This continued use accentuates the consent, their next of kin or legal guardians.

journals.sagepub.com/home/tar 11
Therapeutic Advances in
Respiratory Disease Volume 18

Consent for publication ac.th. After a proposed analysis protocol gains

Not applicable. approval, anonymized participant data will be
accessible 3 months post-publication.
Author contributions
Surat Tongyoo: Conceptualization; Data cura- ORCID iD
tion; Formal analysis; Funding acquisition; Surat Tongyoo https://orcid.org/0000-
Investigation; Methodology; Project administra- 0003-3772-2990
tion; Validation; Visualization; Writing – original
draft; Writing – review & editing. Supplemental material
Supplemental material for this article is available
Tanuwong Viarasilpa: Conceptualization;
online.
Formal analysis; Investigation; Methodology;
Supervision; Writing – original draft; Writing –
review & editing.
Phitphiboon Deawtrakulchai: Conceptuali­ References
zation; Data curation; Formal analysis; Funding 1. Slutsky AS and Ranieri VM. Ventilator-induced
lung injury. N Engl J Med 2013; 369: 2126–
acquisition; Methodology; Project administra-
2136.
tion; Validation; Visualization; Writing – original
draft; Writing – review & editing. 2. Acute Respiratory Distress Syndrome Network;
Brower RG, Matthay MA, et al. Ventilation with
Santi Subpinyo: Data curation; Formal analy- lower tidal volumes as compared with traditional
sis; Project administration; Visualization; Writing tidal volumes for acute lung injury and the acute
– original draft; Writing – review & editing. respiratory distress syndrome. N Engl J Med
2000; 342: 1301–1308.
Chaiyawat Suppasilp: Formal analysis;
Investigation; Supervision; Validation; Writing – 3. Tasaka S, Ohshimo S, Takeuchi M, et al. ARDS
original draft; Writing – review & editing. clinical practice guideline 2021. J Intensive Care
2022; 10: 32.
Chairat Permpikul: Conceptualization; Formal
analysis; Methodology; Supervision; Writing – 4. Pinheiro de Oliveira R, Hetzel MP, dos Anjos
Silva M, et al. Mechanical ventilation with high
original draft; Writing – review & editing.
tidal volume induces inflammation in patients
without lung disease. Crit Care 2010; 14: R39.
Acknowledgements 5. Determann RM, Royakkers A, Wolthuis EK,
Our gratitude goes to Mr. David Park for linguis- et al. Ventilation with lower tidal volumes as
tic editing. This research was conducted at the compared with conventional tidal volumes for
Division of Critical Care Medicine, Department patients without acute lung injury: a preventive
of Medicine, Faculty of Medicine, Siriraj Hospital, randomized controlled trial. Crit Care 2010; 14:
Mahidol University, Bangkok, Thailand. R1–R14.
6. Serpa Neto A, Simonis FD, Barbas CS, et al.
Funding Association between tidal volume size, duration
The authors disclosed receipt of the following of ventilation, and sedation needs in patients
financial support for the research, authorship, without acute respiratory distress syndrome: an
and/or publication of this article: The Siriraj individual patient data meta-analysis. Intensive
Critical Care Research Fund provided financial Care Med 2014; 40: 950–957.
support for the trial. The funder remained unin- 7. Serpa Neto A, Simonis FD, Barbas CS, et al.
volved in the study design, data analyses and out- Lung-protective ventilation with low tidal
come assessments. volumes and the occurrence of pulmonary
complications in patients without acute
Competing interests respiratory distress syndrome: a systematic review
The authors declare that there is no conflict of and individual patient data analysis. Crit Care
interest. Med 2015; 43: 2155–2163.
8. Guay J, Ochroch EA and Kopp S. Intraoperative
Availability of data and materials use of low volume ventilation to decrease
For data-sharing requests, researchers can reach postoperative mortality, mechanical ventilation,
the corresponding author at surat.ton@mahidol. lengths of stay and lung injury in adults without

12 journals.sagepub.com/home/tar
 S Tongyoo, T Viarasilpa et al.

acute lung injury. Cochrane Database Syst Rev among ventilated patients with acute respiratory
2018; 7: CD011151. distress syndrome: a systematic review and
meta-analysis. Crit Care Med 2018; 46: 300–306.
9. Writing Group for the PReVENT Investigators;
Simonis FD, Serpa Neto A, et al. Effect of a 20. Kaczka DW, Lutchen KR and Hantos Z.
low vs. intermediate tidal volume strategy on Emergent behavior of regional heterogeneity in
ventilator-free days in intensive care unit patients the lung and its effects on respiratory impedance.
without ARDS: a randomized clinical trial. J Appl Physiol (1985) 2011; 110: 1473–1481.
JAMA 2018; 320: 1872–1880.
21. Carteaux G, Parfait M, Combet M, et al. Patient-
10. Amato MB, Meade MO, Slutsky AS, et al. self inflicted lung injury: a practical review. J Clin
Driving pressure and survival in the acute Med 2021; 10: 2738.
respiratory distress syndrome. N Engl J Med
2015; 372: 747–755. 22. Kassis EB, Beitler JR and Talmor D. Lung-
protective sedation: moving toward a new
11. Neto AS, Hemmes SN, Barbas CS, et al. paradigm of precision sedation. Intensive Care
Association between driving pressure and Med 2023; 49: 91–94.
development of postoperative pulmonary
complications in patients undergoing mechanical 23. Sklienka P, Frelich M and Burša F. Patient
ventilation for general anaesthesia: a meta- self-inflicted lung injury – a narrative review
analysis of individual patient data. Lancet Respir of pathophysiology, early recognition, and
Med 2016; 4: 272–280. Erratum in: Lancet Respir management options. J Pers Med 2023; 13: 593.
Med 2016; 4: e34. 24. Luz M, Brandão Barreto B, de Castro REV, et al.
12. Toufen Junior C, De Santis Santiago RR, Practices in sedation, analgesia, mobilization,
Hirota AS, et al. Driving pressure and long-term delirium, and sleep deprivation in adult intensive
outcomes in moderate/severe acute respiratory care units (SAMDS-ICU): an international
distress syndrome. Ann Intensive Care 2018; 8: survey before and during the COVID-19
119. pandemic. Ann Intensive Care 2022; 12: 9.

13. Schulz KF, Altman DG, Moher D, et al. 25. Shehabi Y, Bellomo R, Reade MC, et al. Early
CONSORT 2010 statement: updated guidelines intensive care sedation predicts long-term
for reporting parallel group randomized trials. mortality in ventilated critically ill patients. Am J
Ann Intern Med 2010; 152: 726–732. Respir Crit Care Med 2012; 186: 724–731.

14. ARDS Definition Task Force; Ranieri VM, 26. Aragón RE, Proaño A, Mongilardi N, et al.
Rubenfeld GD, et al. Acute respiratory distress Sedation practices and clinical outcomes in
syndrome: the Berlin definition. JAMA 2012; mechanically ventilated patients in a prospective
307: 2526–2533. multicenter cohort. Crit Care 2019; 23: 130.

15. Naureckas ET and Wood LDH. The 27. Kangelaris KN, Calfee CS, May AK, et al. Is
pathophysiology and differential diagnosis of there still a role for the lung injury score in the
acute respiratory failure. In: Hall JB, Schmidt GA era of the Berlin definition ARDS? Ann Intensive
and Kress JP (eds.) Principles of critical care. 4th Care 2014; 4: 4.
ed. New York: McGraw-Hill, 2015, pp.370–376. 28. Cai S, Zhu F, Hu H, et al. Assessment of
16. Tongyoo S, Tantibundit P, Daorattanachai respiratory support decision and the outcome of
K, et al. High-flow nasal oxygen cannula vs. invasive mechanical ventilation in severe COVID-
noninvasive mechanical ventilation to prevent 19 with ARDS. J Intensive Med 2022; 2: 92–102.
reintubation in sepsis: a randomized controlled
trial. Ann Intensive Care 2021; 11: 135.
17. Murray JF, Matthay MA, Luce JM, et al. An Appendix
expanded definition of the adult respiratory
distress syndrome. Am Rev Respir Dis 1988; 138: Abbreviations
720–723.
ALI    acute lung injury
18. Russell JA, Lee T, Singer J, et al. Days alive and APACHE acute physiology and chronic health
free as an alternative to a mortality outcome in evaluation
pivotal vasopressor and septic shock trials. J Crit ARDS    acute respiratory distress syndrome
Care 2018; 47: 333–337. CI      confidence interval
19. Aoyama H, Pettenuzzo T, Aoyama K, et al. CO2    carbon dioxide
Association of driving pressure with mortality FiO2      fraction of inspired oxygen

journals.sagepub.com/home/tar 13
Therapeutic Advances in
Respiratory Disease Volume 18

HCO3   bicarbonate PBW   predicted body weight

ICU    intensive care unit PCV     pressure-controlled ventilation
LDP      limited driving pressure PEEP    positive end-expiratory pressure
LTV      low tidal volume RR    relative risk
Visit Sage journals online
PaCO2    partial pressure of carbon dioxide in SD    standard deviation
journals.sagepub.com/ arterial blood SOFA    sequential organ failure assessment
home/tar
PaO2    partial pressure of oxygen in arterial VCV     volume-controlled ventilation
Sage journals blood VILI     ventilator-induced lung injury

14 journals.sagepub.com/home/tar