Caracterização do uso da insuflação-exsuflação mecânica em Unidades de Terapia Intensiva brasileiras e o impacto de seu uso em pacientes em ventilação mecânica

Abstract

Introduction: Preliminary studies have suggested potential benefits of adding mechanical insufflation-exsufflation (MI-E) to routine physiotherapy in critically ill patients receiving invasive mechanical ventilation (MV). However, data on how this technique is applied in Brazilian intensive care units (ICUs) remain limited. Likewise, evidence on the impact of MI-E on clinically relevant outcomes—such as ventilator weaning—remains limited. Objectives: This thesis comprised three studies with the following aims: (1) to assess the availability of MI-E devices in Brazilian ICUs, evaluate clinicians’ knowledge, current practices, and barriers to its routine implementation; (2) to evaluate the shortterm effects of MI-E compared with standard physiotherapy in mechanically ventilated patients with respiratory hypersecretion; and (3) to systematically review and analyze the findings of studies investigating the impact of adjunctive MI-E on weaning from mechanical ventilation. Methods: The objectives were addressed through three distinct studies. The first, a cross-sectional investigation, involved an electronic survey distributed to physicians, physiotherapists, and nurses registered with the Brazilian Intensive Care Medicine Association (Associação de Medicina Intensiva do Brasil—AMIB). The questionnaire assessed the availability of MI-E devices, participants’ knowledge regarding their use, current clinical practices, and perceived barriers to implementation. The second study was a randomized, crossover clinical trial conducted in the Intensive Care Unit of the University Hospital at the Federal University of Juiz de Fora (ICU-HUUFJF), Brazil. The study included mechanically ventilated patients with respiratory hypersecretion, defined as requiring three or more tracheal suctioning procedures within an 8-hour period. Participants were randomized to one of two groups: the MI-E group or the usual physiotherapy group. In the MI-E group, patients received two sessions on day 1, each consisting of five cycles of insufflation at +40 cmH₂O followed by exsufflation at –40 cmH₂O, with a 0.5-second pause between phases. The usual physiotherapy group received the routine physiotherapeutic care provided in the ICU. On day 2, after an 18-hour washout period, the groups were crossed over. The primary outcome was the number of tracheal suctioning procedures required within 12 hours following the initial intervention. Secondary outcomes included changes in arterial blood gas parameters and respiratory mechanics after each treatment session. The third study was a systematic review and meta-analysis of randomized clinical trials and quasi-experimental studies that evaluated the use of MI-E during weaning from MV and/or after extubation. Eligible studies included adult patients receiving mechanical ventilation in ICUs and assessed at least one of the following outcomes: weaning success, ICU length of stay, or mortality. A comprehensive search was conducted in Medline, Cochrane Library, Scopus, Web of Science, Embase, and LILACS/BVS databases, without language or publication restrictions, including studies published up to July 31, 2025. Titles and abstracts were independently screened by two reviewers to identify potentially eligible studies. The full texts of selected articles were then retrieved and independently assessed for eligibility by two reviewers. The study protocol was registered on the PROSPERO platform (Prospective Register of Systematic Reviews – CRD42021274417). The risk of bias of randomized trials was evaluated using the Cochrane Risk-of-Bias Tool for Randomized Trials (RoB 2), and the certainty of evidence for each outcome was appraised according to the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach. Results: Study 1 obtained responses from 232 healthcare professionals — 125 physicians, 92 physiotherapists, and 15 nurses — representing 148 hospitals, most of which were in the southeastern region of Brazil (54.7%), with limited representation from other regions. The availability of mechanical insufflation–exsufflation (MI-E) in Brazilian ICUs was low (20.9%). Among respondents, 36 professionals reported using MI-E, with 80.6% applying it in intubated patients and 64.5% in extubated patients. In intubated patients, the main reported indications were excessive airway secretions (100%) and use during weaning from mechanical ventilation (86.2%). For extubated patients, indications included secretion overload (100%) and prevention of extubation failure (30.4%). The most frequently reported barriers to MI-E implementation were lack of knowledge (52.1%), limited availability of equipment (36.4%), and insufficient scientific evidence (11.5%). In Study 2, a total of 21 patients were included, with a mean age of 67.4 ± 13 years, 15 (71.4%) male, a mean SAPS 3 score of 67.6 ± 11.4, and a mean SOFA score of 6.1 ± 2.7. At the time of randomization, patients were clinically stable, with a mean PaO₂/FiO₂ ratio of 256 ± 84.7 mmHg, driving pressure of 9.5 cmH₂O (interquartile range [IQR], 7.75–12.40 cmH₂O), and PEEP of 5 cmH₂O (IQR, 5–6 cmH₂O). During the period in which patients received MI-E, the median number of tracheal suctioning procedures within 12 hours was 2 (IQR, 2–3) compared with 3 (IQR, 3–4) during conventional physiotherapy (p = 0.06). No statistically significant differences were observed between groups regarding arterial blood gas parameters (PaO₂, PaCO₂, and PaO₂/FiO₂) 8 hours after the intervention. Two-way ANOVA (group × time) yielded p values of 0.58, 0.78, and 0.16, respectively, for these variables. On the day MI-E was applied, there was a reduction in respiratory system resistance (Rsr) and an increase in respiratory system compliance (Csr) after the second session. These variables were analyzed using a linear mixed-effects model comparing intergroup differences, longitudinal changes (five time points), and the group × time interaction. The p value for this interaction was 0.006 for Rsr and 0.018 for Csr, indicating statistically significant improvements in respiratory mechanics following MI-E. In Study 3, the systematic review identified four randomized clinical trials, comprising a total of 292 patients (148 in the MI-E group and 144 in control groups). Two studies were judged to have an overall high risk of bias, and the remaining two presented some concerns regarding bias. No statistically significant differences were found between groups regarding extubation failure rates, with an odds ratio (OR) of 0.70 (95% CI, 0.39–1.25; p = 0.23; I² = 62%). Similarly, there were no significant differences in mortality (OR, 1.38; 95% CI, 0.64–2.98; p = 0.41; I² = 0%) or ICU length of stay (mean difference, –3.26 days; 95% CI, –8.63 to 2.12; p = 0.23; I² = 92%). Conclusions: In Brazilian intensive care units, the availability of devices for mechanical insufflation–exsufflation (MI-E) is limited. The main barriers to its implementation include resource constraints, lack of knowledge, and insufficient scientific evidence. Among mechanically ventilated patients with respiratory hypersecretion, the use of MIE did not produce a statistically significant reduction in the need for tracheal suctioning. However, it was associated with improved respiratory system mechanics, characterized by an increase in respiratory system compliance (Csr) and a reduction in respiratory system resistance (Rsr). The systematic review and meta-analysis of randomized clinical trials showed that incorporating mechanical insufflation–exsufflation (MI-E) into weaning protocols was not associated with reductions in weaning failure rates, ICU length of stay, or mortality. However, the included studies involved a limited number of participants and exhibited high risk of bias, underscoring the need for further well-designed randomized clinical trials to clarify the role of MI-E in the weaning process.