Impact of different weaning strategies of high-frequency ventilation (HFV) on neonatal cerebral oxygen saturation and protocol for a prospective randomized controlled trial

Abstract

Background

High-frequency ventilation (HFV) is commonly used in neonatal intensive care units to provide respiratory support for critically ill neonates. Currently, there is no standardized procedure for weaning from HFV. Two commonly used strategies are transitioning from HFV to conventional mechanical ventilation (CMV) before extubation (HFV-CMV) and extubation after decreasing mean airway pressure during HFV (HFV-HFV). The impact of these strategies on neonatal cerebral oxygenation and hemodynamics remains incompletely understood.

Methods

We will conduct a prospective, single-center, randomized controlled trial to investigate the effects of two different HFV weaning strategies (HFV-CMV, HFV-HFV) on neonatal cerebral oxygenation and hemodynamics. The patients enrolled in the trial will be randomly allocated to either the HFV-CMV group or the HFV-HFV group in a 1 ratio. The primary outcome will be cerebral oxygen saturation (ScO2) before and after the intervention. Second outcomes are cerebral fractional tissue oxygen extraction, heart rate, blood pressure, and the incidence and severity of intraventricular hemorrhage and periventricular leukomalacia. We hypothesize that HFV-CMV results in positive impact on neonatal cerebral oxygenation compared to HFV-HFV. This study aims to identify a better weaning strategy for HFV and contribute evidence-based data to enhance its clinical application in newborns, potentially improving the care and outcomes for neonates receiving HFV.

Discussion

This study aims to assessing the impact of different HFV weaning strategies on neonatal cerebral oxygenation and hemodynamics, as well as the relationship between the duration of HFV under different strategies and neurological complications, to identify better weaning methods for HFV. We hope to contribute evidence-based data to enhance clinical application of HFV in newborns, potentially improving the care and outcomes for neonates receiving HFV.

Trial registration

Chinese Clinical Trial Registry: ChiCTR2400088628. Registered on August 22, 2024, https://www.chictr.org.cn/bin/project/edit?pid=235926.

Keywords: Neonates, Cerebral oxygen saturation, High-frequency ventilation, Conventional mechanical ventilation, Wean

Administrative information

Introduction

Background and rationale {6a}

High-frequency ventilation (HFV) is commonly employed in the treatment of critically ill neonates to provide respiratory support and improve ventilation and oxygenation. It uses a consistent mean airway pressure (Pmean), small tidal volumes, and high respiratory frequencies to facilitate effective gas exchange [1]. As an effective rescue ventilation mode, HFV has been widely used in various NICUs when other invasive mechanical ventilation (IMV) modes fail [2–5].

There is no standardized procedure for weaning from HFV, and currently, there are two commonly used transition from HFV to conventional mechanical ventilation (CMV) prior to extubation (HFV-CMV) and extubation after decreasing Pmean during HFV (HFV-HFV). Previous studies have primarily focused on respiratory outcomes and have shown benefits of HFV-HFV, such as shorter duration of IMV and reduced incidence of bronchopulmonary dysplasia (BPD) [6, 7]. And a cohort study reported a 90% success rate of extubation from HFV [8]. However, despite these potential benefits, 70% of NICUs in the UK still prefer transitioning from HFV to CMV before extubation [2].

The impact of HFV on the nervous system of neonates has long been a controversial issue. Continuous high Pmean in HFV may increase intrathoracic pressure, reducing venous return, cardiac output, and cerebral blood flow, which is closely related to the occurrence of intraventricular hemorrhage (IVH) [9–11]. Two multicenter RCTs, including 673 and 176 newborns respectively, concluded that HFV is associated with an increased risk of IVH [12, 13]. Changing modes during mechanical ventilation may also result in fluctuations in airway and intrathoracic pressures, affecting cerebral oxygenation and hemodynamics. An observational study on 70 neonates showed that transitioning from IMV to non-invasive ventilation increased cerebral oxygen saturation (ScO2) due to decreased peak inspiratory pressure (PIP) [14]. Another crossover RCT found that transitioning from CPAP to NIPPV may have an unfavorable effect on brain tissue perfusion in premature newborns, possibly due to intermittent elevation of pressure [15]. Therefore, we speculate that switching from HFV to CMV before extubation may lower airway pressure, reduce the duration of high Pmean, and potentially mitigate its impact on hemodynamics and nervous system.

The difference between two weaning strategies (HFV-CMV and HFV-HFV) on cerebral oxygenation and hemodynamics remains incompletely understood. We will conduct a RCT to compare the effects of these strategies. The primary outcome will be ScO2, as it accurately reflects cerebral tissue microcirculation oxygenation and perfusion and can be monitored in real-time using near-infrared spectroscopy [16, 17]. Additionally, we will record other important indicators including tissue oxygen extraction fraction (cFTOE), heart rate, and blood pressure. The occurrence of neurological complications including IVH and periventricular leukomalacia (PVL) will also be recorded. This study will provide insights into the optimal and safe weaning strategy of HFV in neonates, potentially guiding its future clinical practice.

Objectives {7}

This study aims to assessing the impact of different HFV weaning strategies on neonatal cerebral oxygenation and hemodynamics, as well as the relationship between the duration of HFV under different strategies and neurological complications, to identify better weaning methods for HFV and provide evidence to enhance its clinical application in newborns.

Trial design {8}

This is a prospective, single-center, parallel-group, superiority exploratory randomized controlled trial. Patients will be allocated into two groups, HFV-CMV and HFV-HFV, in a 1 ratio. We hypothesize that HFV-CMV results in less negative impact on neonatal cerebral oxygenation compared to HFV-HFV.

Methods: participants, interventions, and outcomes

Study setting {9}

This trial will be conducted at West China Second University Hospital, Sichuan University, a tertiary hospital boasts a team of highly skilled medical professionals with the requisite expertise and capabilities to treat and provide care for infants undergoing IMV.

Eligibility criteria {10}

Inclusion criteria.

Infants participating in the trial must meet the following inclusion

• Newborns using HFV[1] Patients with or at high risk of air leak syndrome requiring IMV will be directly ventilated with HFV[2] Patients primarily under CMV will be switched to HFV if the following criteria are PIP > 25 cmH2O, PEEP > 8 cmH2O, or FiO2 > 60% and the patient is unable to maintain SpO2 > 90% and pH > 7.25, PCO2 ≤ 60 mmHg
• The newborn is eligible for weaning from HFV:[1] Pmean ≤ 12 cmH2O, FiO2 ≤ 40%, pH ≥ 7.30, PCO2 ≤ 55 mmHg[2] Stable vital signs[3] Absence of severe metabolic acidosis and active pulmonary hemorrhage

Exclusion criteria.

• Patients with untreated air leak syndrome
• The guardian signed and refused to participate in the trial

Who will take informed consent? {26a}

The informed consent process will be obtained by personnel who possess qualifications in good clinical practice both verbally and in writing.

Additional consent provisions for collection and use of participant data and biological specimens {26b}

Not applicable. No biological samples will be collected in this study.

Interventions

Explanation for the choice of comparators {6b}

From the perspective of extubation success rate, both weaning strategies for HFV are feasible [6–8]. However, there is currently no research comparing the effects of the two strategies on neonatal cerebral oxygenation and hemodynamics. Therefore, this study will set up two groups of newborns using two different strategies (HFV-CMV, HFV-HFV) for comparison.

Intervention description {11a}

After confirming eligibility, randomization will be conducted immediately. Once the allocation is determined, infants will start the corresponding parameter adjustments according to the assigned intervention.

Instruments and equipment

• Ventilator: Dräger C500 (Drägerwerk Ag & Co., Lübeck, Germany) and Fabian HFO (ACUTRONIC Medical Systems AG, Hirzel, Switzerland) can provide both HFV and CMV
• Cerebral oximetry monitoring Enginmed EGOS-600 B
• Vital XPREZZON (Spacelabs Healthcare, USA) can continuously record and track essential parameters such as heart rate, blood pressure, and SpO2
• PCO2: TCM 400 monitoring (Radiometer, Denmark) will be used for continuous monitoring of transcutaneous PCO2

Ventilatory management

• HFV-CMV switch to CMV[1] Select SIMV+ pressure support ventilation (PSV)[2] FiO2 ≤ 40%[3] SIMV + volume-targeted ventilation (VTV): tidal volume 4–8 ml/kg, Pmax ≤ 25 cmH2O[4] PS: PIP*60–80%[5] RR: 30–55 b/min[6] PEEP: 6–8 cmH2O[7] Ti: 0.3–0.7 s to make sure inspiratory flow reaches zero

• 2. HFV-HFV decrease Pmean by 2–4 cmH2O (Fig. 1)

Fig. 1 Flow diagram of the study

Failure criteria

• After 5 min of parameter adjustment, the target SpO2 (preterm 90–94%, term infant ≥ 95% [18]) or PCO2 (35–60 mmHg) cannot be maintained
• If the ScO2 of the patient decreases by 20% from the baseline, it will be considered a failure [19]
• Emergency situations such as pulmonary hemorrhage, pneumothorax, and heart failure

Monitor procedure

• Once routine care is completed and the patient becomes calm, monitoring will begin
• Monitoring of PCO2 will be conducted during HFV mode using TCM 400 until the values stabilize. Afterward, the monitoring of ScO2 will be initiated. The probe will be placed 1 cm below the left frontal hairline and will be secured using a cap or adhesive tape. The monitoring will be continuous for 30 min during HFV
• The switch to CMV or a decrease in Pmean will be accompanied by a 60-min monitoring. The first 30 min will be considered the transitional phase from HFV or higher Pmean, while the subsequent 30 min will be the stable phase
• In addition to ScO2, the following parameters will be simultaneously monitored, and their mean values recorded every heart rate, blood pressure, PCO2, and Pmean. cFTOE will be calculated during the statistical data analysis

Criteria for discontinuing or modifying allocated interventions {11b}

After intervention, if the infant experiences an increase in PCO2, a decrease in ScO2 or SpO2, parameters can be adjusted within an acceptable range. If any of the failure criteria are met, the intervention will be discontinued. Mode will be switched back to HFV in HFV-CMV group, and Pmean will be adjusted in HFV-HFV group. If the guardian of an enrolled infant requests to terminate the study, their decision will be fully respected.

Strategies to improve adherence to interventions {11c}

Prior to the study, the research team will receive protocol-based training to ensure clear understanding of their roles and responsibilities. A pediatrician will assess patient eligibility, while a respiratory therapist will perform the interventions. Trained nurses will be responsible for placing monitoring probes. Data recording personnel will also receive training to ensure consistent data collection. Meetings will be promptly convened to address any difficulties encountered during the trial.

Relevant concomitant care permitted or prohibited during the trial {11d}

All enrolled neonates will receive routine care, and no necessary procedures will be prohibited.

Provisions for post-trial care {30}

Not applicable. After the trial, the subjects will be provided with standard care in accordance with local policies.

Outcomes {12}

Primary

The primary outcome is ScO2, which will be recorded every 2 s.

Secondary

• [1] cFTOE (cFTOE = (SpO2 − ScO2)/SpO2)
• [2] Heart rate
• [3] Blood pressure
• [4] The incidence and severity of IVH and PVL

Other

Other outcomes include the the incidence and severity of BPD.

Participant timeline {13}

The participant timeline is shown in Fig. 2.

Fig. 2 The schedule of recruitment, interventions, and assessments

Sample size {14}

This study is the first trial to compare the effects of different HFV weaning strategies on neonatal ScO2, and there is no identical study data available. To calculate the sample size, we refer to the results of two previous studies. One study found that the difference in ScO2 after switching from HFV to CMV was 2.3 ± 5.7% [20]. In another study, newborns received HFV for lung recruitment. The difference in ScO2 before and after the change in Pmean was found to be 0.08 ± 0.23% [21]. With α = 0.05% and β = 0.9%, a paired t-test was applied using the PASS15.0.5 software, which calculated a required sample size of 116 cases. Considering a dropout rate of 10%, the total sample size needed is 128 cases.

Recruitment {15}

Whenever a neonate is placed on HFV mode, eligibility assessment will be conducted by physicians. Neonates who meet the criteria of the study will be recruited by personnel with good clinical practice qualifications. Historical data from our medical center suggests that we can recruit the target sample size within 3 years.

Assignment of allocation

Sequence generation {16a}

Eligible infants will be randomly assigned to either the HFV-CMV group or the HFV-HFV group in a 1 ratio using computer-generated random numbers and variable block sizes.

Concealment mechanism {16b}

The allocated interventions will be placed in sealed envelopes labeled with participant numbers, in accordance with the order of randomization.

Implementation {16c}

Once the consent is obtained, a designated research personnel will open the envelope and reveal the assigned intervention. Subsequently, a respiratory therapist involved in the trial will perform the intervention.

Assignment of blinding

Who will be blinded {17a}

Blinding of healthcare providers is not feasible in this trial. However, to minimize potential biases in the analysis of study outcomes, the data analysts will be blinded to the specific group assignments.

Procedure for unblinding if needed {17b}

The trial does not involve unblinding and follows an open-label design. Only the data analysts will be blinded.

Data collection and management

Plans for assessment and collection of outcomes {18a}

The primary outcome of this trial is ScO2, and the complete monitoring data for the 90-min period will be exported using the Enginmed EGOS-600 B device. Secondary outcomes will be measured using bedside monitors, with average values recorded per minute. The cFTOE value per minute will be calculated using the formula cFTOE = (SpO2 − ScO2)/SpO2. The baseline data and neurological and respiratory outcomes will be obtained from the patient’s medical records.

Plans to promote participant retention and complete follow-up {18b}

Detailed information about the purpose, procedures, duration, outcomes, potential benefits, and risks of the study will be provided to the guardians to ensure their understanding. Guardians can withdraw the patient from the study at any stage without affecting their treatment. Reasons and timing of withdrawals will be documented, and relevant measures will be recorded with guardian consent. If a guardian declines, the patient will be replaced.

Data management {19}

Bedside data will be entered into an electronic spreadsheet for storage and management. Maternal information during pregnancy will be retrieved from the hospital medical records system. The electronic spreadsheet will only be accessible to the research team and will be periodically reviewed by the principal investigator to identify and correct any erroneous data promptly.

Confidentiality {27}

To protect participant privacy, each participant will be assigned a unique research identification code to replace their real identity information in the electronic spreadsheet. After the study is completed, the identification codes will no longer be used, and participants’ identity information will not be disclosed in any public or shared data.

Plans for collection, laboratory evaluation, and storage of biological specimens for genetic or molecular analysis in this trial/future use {33}

Not applicable. No biological samples will be collected in this study.

Statistical methods

Statistical methods for primary and secondary outcomes {20a}

Statistical analysis will be conducted using both intention-to-treat (including all randomized participants) and per-protocol (including only those who completed the intervention) analysis. The primary and secondary outcomes will be analyzed using a one-way repeated measures ANOVA to test for differences across different time periods. These periods include the 30 min before the intervention, the first 30 min, and the second 30 min after the intervention. Multivariable mixed effect liner regression will also be conducted to evaluate the mean differences among different time periods, with considering the impact of potential confounding factors, such as PCO2 and Pmean. For patient demographics and maternal information, categorical variables will be analyzed using the chi-square test or Fisher’s exact test, while continuous variables will be analyzed using the t-test. For variables that follow an abnormal distribution, the Mann–Whitney U test will be used. A significance level of p < 0.05 will be considered statistically significant.

Interim analyses {21b}

Not applicable. No interim analysis will be conducted in this trial.

Methods for additional analyses (e.g., subgroup analyses) {20b}

Subgroup analysis will be conducted based on the gestational age (< 32 weeks versus ≥ 32 weeks) of the enrolled infants.

Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data {20c}

If any missing data occurs in analysis, multiple imputation will be used to impute the missing values. Sensitivity analysis will be conducted to compare the results before and after multiple imputation of missing values, in order to observe the stability of statistical results.

Plans to give access to the full protocol, participant-level data, and statistical code {31c}

After the completion of the study, the public can contact the research team to obtain the relevant experimental data.

Oversight and monitoring

Composition of the coordinating center and trial steering committee {5d}

The coordinating center includes a principal investigator, data manager, and other research personnel. The principal investigator oversees the experiment, while the data manager records and organizes the data. Other research personnel are involved in day-to-day operations. Monthly meetings are held to track study progress, with emergency meetings convened as needed. No trial steering committee or stakeholder/public involvement group is present.

Composition of the data monitoring committee, its role and reporting structure {21a}

The data monitoring committee (DMC) comprises two independent pediatricians and one statistician. They regularly review trial data and offer recommendations to enhance integrity and reliability. In the event of adverse events, they document reasons and provide suggestions to ensure participant safety. The DMC operates independently and has no conflicts of interest.

Adverse event reporting and harms {22}

During the trial, all adverse events will be recorded and reported to the DMC. The research team will take appropriate actions to ensure participant safety. Documentation of adverse events will include time of occurrence, duration, relationship to the intervention, and corresponding measures taken. The team will follow-up on each event until it is resolved. Potential adverse events include ventilator-associated lung injury, pneumothorax, burns from probes, and pressure injuries. Any serious adverse events related to the trial that pose a threat to life will be reported to the ethics committee and local health authorities.

Frequency and plans for auditing trial conduct {23}

The DMC is responsible for safeguarding the scientific integrity and participant safety of the trial and conducts independent reviews. They will conduct audits of the trial every 2 months to ensure safety and efficacy. In the event of serious adverse events or safety concerns, additional meetings will be convened for further auditing.

Plans for communicating important protocol amendments to relevant parties (e.g., trial participants, ethical committees) {25}

Modifications to the research protocol will be voted on and documented during meetings. If these modifications impact interventions or participant safety, ethical committees will be notified. The revised protocol will be communicated to the entire research team, with necessary training provided if needed.

Dissemination plans {31a}

The trial results, including both positive and negative findings, will be shared through conferences and peer-reviewed research publications at both domestic and international levels.

Discussion

This study will be the first to explore the effects of different weaning strategies of HFV on cerebral oxygenation and hemodynamics in neonates, which will provide more evidence for the application and weaning of HFV. The inclusion of subjects in this study will not restrict gestational age and reason for intubation. We will conduct a stratified analysis based on the subjects’ gestational age (< 32 weeks versus ≥ 32 weeks) to explore whether there are differences in cerebral oxygenation and hemodynamics during two weaning methods. In our study, every measurement of ScO2 taken during the 90-min period will be retained (every 2 s) to avoid any missing data. Due to the different ventilation mechanisms, there will be changes in ventilation volume after mode switching, and we will utilize continuous transcutaneous monitoring to observe the trend of PCO2.

If the patient exhibits strong spontaneous activity or requires airway suctioning during the monitoring process, it may affect the results. Increased secretions, spontaneous activities, and bowel movements can potentially affect Pmean. We will make annotations for these situations during recording and exclude these portions of data to minimize any potential bias during analysis.

We aim to identify a better weaning strategy for HFV and contribute evidence-based data to enhance its clinical application in newborns, potentially improving the care and outcomes for neonates receiving HFV.

Trial status

Protocol Version 2, May 26, 2024.

Recruitment start January 1, 2025.

Recruitment completion January 31, 2028 (estimated).

Acknowledgements

We would like to express our gratitude to Professor Xiaohong Li for providing statistical consultation.

Abbreviations

Authors’ contributions {31b}

Wenli Liu, Jun Tang, and Tao Xiong: literature review, protocol development, data analysis, and manuscript writing. Rong Zhou, Zhu Yuan, Chao Chen, and Yi Huang: protocol development, design of the work, design of data tables and sample size calculation. Ke Tian, Aoyu Wang, and Jun Zhu: trial coordination, data acquisition, and optimization of experimental protocols. Tao Xiong, Jun Tang, and Jing Shi: project initiation, modification of protocols and manuscripts. All authors have read and approved the final manuscript.

Funding {4}

This work was supported by National Natural Science Foundation of China (82,371,723), the National Key R&D Program of China (2021YFC2701700, 2021YFC2701704), the Science and Technology Bureau of Sichuan Province (24ZDYF0866), and Innovative Research Project from 0 to 1 of Sichuan University (2023SCUH0021).

Data availability {29}

The dataset generated in this study will be provided upon reasonable request after contacting the corresponding author.

Declarations

Ethics approval and consent to participate {24}

This study has been approved by the Ethics Committee of West China Second University Hospital of Sichuan University, with the approval number of 2024 (130). Written, informed consent to participate will be obtained from the parent or legal guardian of the neonates participating in the trial.

Consent for publication {32}

Not applicable. This manuscript does not contain individual personal data from patients. The participant information materials and informed consent form are available from the corresponding author on request.

Competing interests {28}

The authors declare that they have no competing interests.

Footnotes

References

Associated Data

Data Availability Statement

The dataset generated in this study will be provided upon reasonable request after contacting the corresponding author.