Secondary Logo

Journal Logo


Wong, J.J.M.1; Liu, S.2; Qian, S.3; Ong, J.S.M.4; Gan, C.S.5; Anantasit, N.6; Chor, Y.K.7; Samransamruajkit, R.8; Phan, H.P.9; Phumeetham, S.10; Feng, X.11; Feng, M.2; Lee, J.H.1

Pediatric Critical Care Medicine: June 2018 - Volume 19 - Issue 6S - p 18
doi: 10.1097/01.pcc.0000537382.07580.e4
Oral Abstracts

1KK Women’s and Children’s Hospital, Pediatric Medicine, Singapore, Singapore

2National University of Singapore, Saw Swee Hock School of Public Health, Singapore, Singapore

3Beijing Children’s Hospital, Pediatric Intensive Care Unit, Beijing, China

4National University of Singapore, Pediatric Intensive Care Unit, Singapore, Singapore

5University Malaya Medical Centre, Department of Pediatrics, Kuala Lumpur, Malaysia

6Ramathibodi Hospital, Pediatric Department, Bangkok, Thailand

7Sarawak General Hospital, Department of Pediatrics, Kuching, Malaysia

8King Chulalongkorn Memorial Hospital, Department of Pediatrics, Bangkok, Thailand

9National Children’s Hospital, Pediatric Intensive Care Unit, Hanoi, Vietnam

10Siriraj Hospital, Department of Pediatrics, Bangkok, Thailand

11Children’s Hospital of Chongqing Medical University, Pediatric Intensive Care Unit, Chongqing, China

Back to Top | Article Outline

Aims & Objectives:

High frequency ventilation (HFV) is frequently used as rescue therapy to avoid excessively high plateau pressures. However, evidence for its use in pediatric acute respiratory distress syndrome (PARDS) is weak. We aimed to investigate the association between the usage of HFV and mortality in children with PARDS.

Back to Top | Article Outline


Patients with PARDS from 10 pediatric intensive care units across Asia from 2009–2015 were identified. Data on epidemiology and clinical outcomes were collected. Primary outcome was pediatric intensive care unit (PICU) mortality. HFV and non-HFV patients were matched based on a propensity score model. ICU mortality between the matched HFV and non-HFV groups were then statistically compared using Fisher’s exact test.

Back to Top | Article Outline


There was a total of 336 PARDS patients included in this analysis, 130/336 (39%) were treated with HFV during their course of PARDS. The propensity score matching yielded a balanced cohort of 176 patients (88 in the HFV and non-HFV groups respectively). The matched groups had comparable demographics, severity of illness scores and risk factors for PARDS (Table 1). The characteristics of the propensity score model can be found in Table 2.

ICU mortality for the matched HFV group and non-HFV group were 31% and 36% respectively. HFV use was not associated with ICU mortality in PARDS [odds ratio (OR): 0.78, 95% confidence interval (CI) 0.41, 1.46; p=0.52].

Back to Top | Article Outline


HFV use was common (39%) in PARDS. The use of HFV was not associated with increased mortality in PARDS.

©2018The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies