Mechanical Insufflation-Exsufflation in Critically Ill Patients

Overview

Mechanical insufflation-exsufflation (MI-E) is an established airway clearance technique for patients with chronic conditions like neuromuscular diseases. However, its use in critically ill ICU patients remains inconsistent and lacks standardized guidelines. Despite growing research, current practices vary widely in patient selection, treatment protocols, and safety management, with limited high-quality evidence to support clear recommendations. To address this gap, an international, multidisciplinary Delphi consensus study is needed to establish expert-based best practices and feasible guidelines for the safe and effective implementation of MI-E in the intensive care setting.

Background: Mechanical insufflation-exsufflation (MI-E) functions as a noninvasive method which clears airways through its ability to enhance cough performance by delivering positive pressure to initiate airflow and then rapidly applies negative pressure for exhalation. MI-E enables bronchial secretions to be moved and eliminated through its ability to produce high expiratory flows while decreasing the chances of secretions being retained in the lungs. MI-E is well established in patients with neuromuscular diseases and chronic respiratory disorders, in which weak cough and airway clearance dysfunction are the main contributors to respiratory morbidity. Nevertheless, the use of this technique in critically ill patients, particularly those admitted to the intensive care unit (ICU), is still diverse and insufficiently standardized. Invasive and non invasive ventilation, sedation, impaired consciousness, artificial airways, and respiratory muscle dysfunction substantially alter cough physiology in ICU patients, increasing risk of secretion retention, atelectasis, extubation failure, and reintubation. Research on MI-E applications for ICU patients has become more popular but current clinical practices show different standards for selecting patients and determining treatment times and establishing device settings and conducting safety checks and managing restricted treatment conditions and integrating MI-E use with other ventilation modalities. A few studies available are restricted by small sample sizes, mixed populations, and varying outcome definitions, thus no clear recommendations can be made. Concerns about safety, such as the risk of barotrauma and damage to the airway, changes in clinical practice, and the requirement for staff skills and training are some of the factors that cause uncertainty and inconsistency in the use of MI-E in critically ill patients. In situations where there are few high-quality randomized trials giving the evidence, expert consensus is now generally considered to be a proper and effective methodology to determine clinical best practice. The Delphi method with the addition of fuzzy logic is a way of systematically collecting and combining the experts' opinions, and it also counteracts the influence of the strongest participants and allows the identification of the recommendations that should be given priority, both from the point of view of their clinical relevance and of their feasibility. Therefore, it is necessary to carry out a global, multidisciplinary Delphi consensus study in order to identify best practices and agree on consensus, based guidelines for implementing MI-E therapy in critical care patients. Study Objectives: To achieve international expert consensus on the clinical use of MI-E in critically ill patients Fuzzy Delphi methodology, and to develop consensus-based recommendations applicable to intensive care practice. * To identify and prioritize the clinical indications, contraindications, and patient selection criteria for MI-E use in critically ill patients. * To define consensus-based recommendations regarding timing of application, device settings, titration strategies, and integration with ongoing mechanical ventilation. * To establish expert agreement on safety considerations, including monitoring requirements and potential risks associated with MI-E use in the ICU. * To explore organizational, educational, and practical barriers affecting the adoption of MI-E in critical care settings. * To assess the perceived feasibility and clinical relevance of MI-E practices across different ICU contexts. * To identify knowledge gaps and future research priorities related to MI-E use in critically ill patients. Sample Size: As this study is an expert consensus study using Delphi and Fuzzy Delphi methodology, the sample size refers to the number of participating experts rather than patient enrollment. An international, multidisciplinary panel of 20-25 experts with recognized experience in critical care and/or mechanical insufflation-exsufflation (MI-E) will be included in the study. To account for potential attrition across Delphi rounds, approximately 30 experts will be invited initially, with the aim of retaining at least 20 experts through the final consensus round. This panel size is considered sufficient to achieve stable and reliable consensus while maintaining feasibility and diversity of expert perspectives. Ethical Approval and Informed Consent: Ethical approval for this study has been obtained from the appropriate Ethics Committee (The Campania 2 Territorial Ethics Committee) Approval Number: AOC/0018028/2026) prior to study initiation. All participating experts will be provided with detailed written information regarding the study objectives, methodology, data management procedures, and their rights as participants. Electronic informed consent will be obtained from all experts before participating in the first Delphi round. Participation will be voluntary, and experts may withdraw from the study at any stage without providing justification. Expert responses will be collected anonymously using secure electronic survey platforms. Personal identifiers (such as name and email address) will be stored separately from survey data and accessed only by the study coordination team. All data will be handled confidentially and in accordance with applicable data protection regulations. Primary Consensus Output: Development of a core set of international consensus statements on the clinical use of MI-E in critically ill patients, finalized through predefined Fuzzy Delphi consensus criteria (agreement level, threshold value, and defuzzified score). Secondary Outcomes: Consensus-based identification and prioritization of key clinical domains related to MI-E use in critical care, including: * Indications and contraindications * Patient selection criteria * Timing of application * Device settings and titration strategies * Safety monitoring and risk considerations * Integration with ongoing mechanical ventilation and ICU workflows Ranking of consensus statements according to their clinical importance and practical feasibility using Fuzzy Delphi weighting methods. Identification of areas of strong agreement and persistent disagreement among experts regarding MI-E practices in critically ill patients. Consensus-derived description of organizational, educational, and practical barriers to MI-E implementation in intensive care units. Definition of priority areas for future research based on items that fail to reach consensus or are considered clinically important but insufficiently supported by existing evidence. Duration of Study: The total duration of the study is planned to be 12 months, including expert recruitment, execution of all Delphi and Fuzzy Delphi rounds, data analysis, and finalization of the consensus statements. Data Collection and Anonymization: Data will be collected using secure, web-based electronic survey platforms specifically designed for Delphi studies. Each participating expert will be assigned a unique anonymous identification code to ensure confidentiality and to allow tracking of responses across Delphi rounds without revealing personal identity. All Delphi rounds will be conducted electronically. Experts will complete structured questionnaires, including open-ended questions in the initial round and structured rating scales in subsequent Fuzzy Delphi rounds. No patient-level data or sensitive personal information will be collected. Personal identifiers (such as name, institutional affiliation, and email address) will be stored separately from survey responses in a password-protected file accessible only to the study coordination team. Survey data will be anonymized prior to analysis and will be used exclusively for research purposes. Anonymized datasets will be exported for statistical and Fuzzy Delphi analysis. Results will be reported in aggregate form, ensuring that individual experts cannot be identified. Data storage, handling, and sharing procedures will comply with applicable data protection regulations and institutional policies. Qualitative Analysis (Delphi Round 1): Responses from the first open-ended Delphi round will be analyzed using thematic content analysis. Two members of the research team will independently review and code the responses to identify recurring themes, concepts, and recommendations related to the clinical use of MI-E in critically ill patients. Discrepancies in coding will be resolved through discussion and consensus. The resulting themes will be used to generate a refined list of statements for subsequent Delphi rounds. Fuzzy Delphi Analysis (Rounds 2 and 3): In the structured Delphi rounds, experts will rate each consensus statement using a 5-point Likert scale based on predefined dimensions (e.g., clinical importance and feasibility). Likert-scale responses will be transformed into Triangular Fuzzy Numbers (TFNs). For each statement: * Individual TFNs will be aggregated to obtain group fuzzy values. * A threshold value (d) will be calculated to assess the degree of agreement among experts. * Defuzzification will be performed using the centroid method to obtain a crisp score for each statement. Consensus will be defined according to prespecified criteria: * Threshold value (d) ≤ 0.20 * Level of agreement ≥ 75% * Defuzzified score ≥ 0.50 Statements meeting all criteria will be accepted as consensus items. Statements not meeting consensus criteria will be revised and re-evaluated in a subsequent Delphi round or excluded based on expert feedback. Statistical Analysis: Descriptive statistics will be used to summarize expert characteristics and response distributions. Categorical variables will be reported as frequencies and percentages. Continuous variables derived from defuzzified scores will be presented as means and standard deviations or medians with interquartile ranges, as appropriate. Analysis will be performed using IBM SPSS Statistics 26 (IBM Corp., Armonk, NY, USA) and jamovi v.2.6 (The jamovi project, Sydney, Australia).