Trial of an Alternate Mode of Providing Artificial Breaths to Children With Very Severe Pneumonia

Study setting:

15-bedded pediatric intensive care unit (PICU) of a multi-specialty, tertiary referral and teaching hospital- the Advanced Pediatrics Centre (APC) in Post-Graduate Institute of Medical Education and Research (PGIMER), Chandigarh, India Study period Recruitment: January 2014 to December 2016 Data Analysis: Jan 2017 to June 2017 Study design An open-labelled, parallel-arm, efficacy/feasibility randomized controlled trial Ethics approval Ethics approval has been obtained from the Institute Ethics Committee. The study has been registered with Clinical Trials Registry - India (CTRI) (ctri.nic.in). Informed consent will be obtained from the parents/legal guardian and the conversation would be recorded using a camera (audio-visual documentation of evidence).

Sample size estimation Assuming alpha error of 5% and power of 80% with non-inferiority limit of 4 days (Standard deviation of VFDs being 8.2 days in the conventional low tidal volume ventilation group in pilot trial), sample size was calculated to be 52 per group. As this is a safety and feasibility trial, an interim analysis would be done at 50% enrolment.

Enrolment Parents or legal guardians of children who satisfy the above eligibility criteria will be invited by the investigator to participate in the study. Parents are free not to participate, or to withdraw from the study at any point of time. All children, irrespective of their enrollment in the study, will receive standard pediatric intensive care unit (PICU) care as per the unit's existing protocol. An information sheet (in Hindi/ English) furnishing details of the study will be provided to the parents. Given the fact that all these children are sedated and on mechanical ventilation, obtaining assent would not be feasible in this study.

Randomization Sequence generation A computer-generated, unstratified, block randomization with variable block sizes will be performed to determine group allocation. A person not involved in the study will perform the random number allocation and prepare opaque, sealed envelopes containing the allocation.

Concealment allocation Each pre-sealed opaque envelope would be opened only after obtaining a written consent and audio-visual record of the same. As the randomization is done using a variable block size, and prepared by a statistician not directly involved in the study, there would be no way of predicting the random allocation, thus minimizing the risk of allocation bias.

Randomization implementation After parents provide informed consent, randomization would be done within the next one hour and child initiated on appropriate mode of ventilation. The supportive care for both the groups, would be as per the attached supplementary protocols

Intervention protocol:

The Airway pressure release ventilation (APRV) intervention protocol has been designed based on the available APRV literature as of Dec 2013.

Start the child on APRV mode of ventilation with the following settings:

• P HIGH would determine the degree of baseline lung inflation. A rough estimate can be obtained based on the plateau pressure requirements on the conventional mode of ventilation. Perform an inspiratory hold to ascertain the plateau pressures:
• If P plateau > 30 cm water, set P HIGH at 30 cm water
• If P plateau < 30 cm water, set P HIGH at or 1-2 cm above the measured P plateau.
• Alternatively, if P plateau cannot be measured, P HIGH can be set according to the following guide:

PaO2/ FiO2 ratio P High < 250 15-20 < 200 20-25 < 150 25-28

After initiating a particular P HIGH, a clinical assessment of lung volume needs to be followed with a chest radiograph to determine the degree of lung inflation (similar to setting of Mean airway pressure in High frequency oscillatory mode of ventilation). The child's P HIGH is adjusted to maintain optimal lung volume, without clinical or radiological evidence of hyperinflation: no signs of decreased cardiac output/ hypotension and/or the level of the diaphragm visible greater than the ninth rib.

• Start at T High of 4 seconds; Titrate T High based on oxygenation status. At least 80 -95% of the total cycle time should be spent in T High.
• Set P Low at Zero cm H2O
• Set T Low so that expiratory flow decreases by 25 % of peak expiratory flow rate (PEFR); usually 0.1-0.8 seconds. The ratio of T-PEFR to PEFR should be targeted near 75%. This entity needs to be titrated every 2-4 hours and may have to be shortened as lung injury advances.
• Set Pressure Support at ZERO
• The number of breaths per minute or number of releases is a function of T High and T Low as depicted below:

Weaning from APRV would also be carried out in a structured, protocolized manner. The following strategies would be adopted:

1. As child's clinical condition and oxygenation index improves, and Fraction of inspired oxygen levels are brought down to 0.6, T HIGH is increased in steps of 0.5-2 seconds till it is 10-12 seconds.
2. P HIGH can be subsequently decreased in steps of 2-3 cm H2O till a value of 12-16 is reached. Decrease in P HIGH can be carried out earlier if features of hyperinflation (clinical/radiological) appear at any point.
3. The goal is to reach pressure levels of 12-16 cm H2O and then switch to Continuous positive airway pressure (CPAP) of 6-8 cm H2O, from which child can be extubated directly to nasal prong CPAP or gradually tapered off CPAP to Endotracheal-T piece and subsequently extubated depending on the overall clinical status.