Randomized clinical study of electrical impedance tomography-guided chest physiotherapy in difficult-to-wean patients: study protocol

Trial design and study setting

This single-center, single-blind, randomized pilot study employed a parallel-group design. This protocol adhered to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidelines (Nydahl et al., 2017) (see Table 1).

Inclusion criteria are mechanical ventilation patients who are ≥18 years old, received invasive mechanical ventilation more than 96 h before randomization, are satisfied the preconditions for the machine to be withdrawn, and at least have one failed attempt to withdraw the ventilator (re-need ventilator support within 48 h after extubation), and willing to participate in the study and sign the informed consent.

Exclusion criteria are as follows: malignant arrhythmia or acute myocardial ischemia; pneumothorax, pulmonary bulla and barotrauma and other lung diseases; hemorrhagic disease or abnormal coagulation mechanism with bleeding tendency; chest skin trauma; pulmonary hypertension and pulmonary embolism; with a permanent or temporary pacemaker; there is malignant tumor; present and previous history of neuromuscular diseases affecting respiratory muscle; participated in another clinical study related to mechanical ventilation withdrawal; can not cooperate with the study for any reason or the researcher thinks that it is not suitable to be included in this experiment.

Only when we obtain the written consent before ICU admission can we verity the eligibility of the patients and start our trial process.

Recruitment

The recruitment is mainly completed through the introduction of the ICU clinicians in our hospital and posters. For patients with severe or critical patients, we will work with their legal representatives to obtain permission.

Ethics

Approval was given from our ethics committee (Ethics Committee of Beijing Rehabilitation Hospital of Capital Medical University, Agreement number 2022bkky-139). If adjustments or changes to the protocol are necessary, a project amendment will be requested from the committee. Before enrollment, patients are required to sign the informed consent form and agree to participate in the study.

Interventions

Prior to the CPT plan being dispatched to centers, subjects in control group and EIT-guided group are given clinical treatment which is jointly formulated by the pulmonary rehabilitation team, and the course of treatment is determined according to the specific conditions of the patient. The medical and paramedical staff at each investigation center will be informed of the CPT plan and the use of EIT, in this way, make the teams can familiarize themselves with the technique. To limit operation bias, therapists should obtain training on standard operating procedures for EIT as well as a corresponding consistency assessment for EIT image evaluation results. Training on the standard operating procedure for EIT should include the correct use of EIT electrode strips, reduction of image artifacts during EIT, and correct understanding of the physiological significance of EIT ventilation area changes. The consistency assessment for EIT image evaluation should be conducted regularly (monthly), including experience sharing of clinical classic cases and detailed EIT use records of enrolled patients. All therapists participating in the study must have a minimum of three years of work experience.

Control group: Two sessions of CPT (morning and afternoon, 20 min each) are conducted. The CPT session consist of modified postural drainage (Freitas et al., 2018), assisted cough technique (Spinou, 2020), positive expiratory pressure (McIlwaine, Button & Dwan, 2019) and chest percussion, vibration (Wilson, Morrison & Robinson, 2019). The appointed therapist performed traditionally pulmonary auscultation and thoracic palpation to assess the status of pulmonary ventilation and secretion retention, and whether the patient’s cough ability can complete effective airway clearance. Subsequently, CPT will be implemented according to the evaluation results and the experience of the therapist.

EIT-guided group: Similar to the patients in the control group, operation frequency and duration of CPT are predefined for the patients according to the assessment prior to the randomization. In addition to the traditional pulmonary auscultation and thoracic palpation, for each CPT session, EIT measurement is conducted and the images are used to guide the CPT treatments. The uses of EIT to guide individual treatments are briefly described as follows (Fig. 1).

EIT-guided modified postural drainage combined with vibrations and chest percussion: tidal variation images in EIT reveals heterogeneously ventilated regions. Physiotherapist identified such regions at the bedside and instructed the patient to take the appropriate drainage position, so that the poorly ventilated regions became gravity non-dependent regions. Subsequently, the physiotherapist put her hands on the poorly ventilated area with a vibratory force. A compressive pressure is produced by the therapist’s arms, and the strength and therapeutic efficacy of CPT manipulation will be reported based on the EIT image results.

EIT as a tool for instant feedback and motivation: After each treatment session, the effect of the treatment is assessed immediately. The ventilation improvement is visible and explained using the EIT images. The patient visualized the improvement and is motivated for further CPT sessions.

Outcomes (primary and secondary)

Sample size

The sample size for this randomized clinical trial was determined utilizing G*Power software version 3.1, predicated on an effect size of −0.62 observed in change in right-side diaphragm thickening fraction, as detailed in previous research (Dres et al., 2022). To achieve a statistical power of 0.80 at an alpha level of 0.05, a total of 84 participants is requisite, allocating 42 individuals to each of the two comparative groups. The statistical analysis of the sample size calculation is conducted using the t-test for the difference between two independent means (two groups). Assuming a 20% drop-out rate, we ultimately decided to enroll a sample of 100 patients. However, after data is collected from the first 40 patients, an interim analysis will be performed to determine the size of the expected effect to be reasonable. Following that, a reassessment of the appropriate sample size will be performed based on this analysis.

Randomization

Eligible patients will be randomly assigned (1:1) to either the EIT-guided or control group using a computer-generated block design managed by an independent operator. Treatment allocation will be concealed through sequentially numbered, opaque, sealed envelopes, ensuring that neither researchers nor participants are aware of group assignments. An impartial researcher, uninvolved in data collection, will oversee the randomization process.

Blinding

An independent researcher, uninvolved in outcome evaluation, will assign group allocations. Blinding will extend to physiotherapists assessing outcomes. Participants and EIT-guided CPT providers will be instructed to avoid disclosing allocation to evaluators. The participants and researchers involved in the assessments will be blinded. Blinding will be maintained until data entry is completed, with any accidental unblinding will be documented. Outcome assessors will receive thorough training to ensure data quality.

Data collection and management

To protect confidentiality, all real-time data will be coded, and random checks will be performed to ensure consistency across original records, forms, and database entries, with any discrepancies promptly corrected. Tomographic data will be securely stored alongside participant codes for future reference if needed. A comprehensive checklist will ensure strict adherence to research protocols, minimizing data omission or bias. In cases where missing data cannot be retrieved, affected participants will be excluded from the analysis. Data and analysis codes can be accessed by contacting the corresponding author. If any withdraw, we will perform an intention-to-treat analysis or imputation data.

Statistical analysis

Normally distributed continuous data are described as means and standard deviation (SD) and non-normal distributed data are described as median (IQR). Normality of continuous data are assessed through the Shapiro–Wilk test. Comparisons between groups are performed by using the independent-samples t test or the Mann–Whitney U test, for the continuous variables normally or non-normally distributed, respectively. Categorical variables will be reported as numbers and percentages and compare with a Chi-square test or Fisher test. In addition, after the completion of the study, the sample can be divided into several subgroups for stratified analysis according to specific criteria (the type of primary disease, duration of mechanical ventilation, etc.). Primary and secondary endpoints will be evaluated using the modified intent-to-treat (mITT) population (the same as the ITT population for the control group but excluding patients for whom CPT plan is not achieved in the treatment group). We will use GraphPad Prism software (version 5.0) for the paper drawing, all statistical analyses is done with SPSS 22.0 (IBM, Armonk, NY, USA). Significant differences between groups or across time are reported at the alpha level of 0.05. All reported p values are two-sided.

Quality assurance

An external researcher uninvolved in measurements will perform random audits every 14 days to verify consistency between original equipment data, recorded formats, and the database. Detected errors will be corrected promptly. The study’s progress will be reviewed at monthly project meetings.

Adverse events

In this trial, adverse events are defined as any unexpected medical occurrences presenting symptoms not observed before the study. Predicted adverse events include severe pain, removal of endotracheal, arterial, or central lines, arrhythmia, bradycardia, systolic blood pressure > 200 mmHg or < 90 mmHg due to hemodynamic instability during exercise, and desaturation ≤88% due to respiratory instability during exercise, along with minor reactions such as dyspnea, dizziness, tachypnea, and sinus tachycardia (Schaller et al., 2016; Bailey et al., 2007). Predicted side effects are adverse events, with severity categorized as mild, moderate, or severe. Both device-related and unrelated adverse events are recorded. In cases of unexpected serious adverse events or other unintended interventions, reports are submitted to the institutional review board. Data and safety monitoring will occur after every five participants are enrolled. In case of any harm related to the intervention, the researchers will provide all the necessary support.