Exercise in patients with a tracheostomy and speaking valve: a randomised crossover-controlled trial

Dear Editor,

Critically ill patients expected to receive prolonged invasive mechanical ventilation commonly receive a tracheostomy. Whilst a tracheostomy reduces sedation requirements and improves comfort, mobility, and communication, it is often associated with prolonged intensive care unit (ICU) and hospital length of stay [1]. These patients require exercise and rehabilitation which have well documented benefits in critical illness [2]. A speaking valve (SV) may be used to facilitate verbal communication and improve lung recruitment; however, is often removed during exercise due to potential (but unsubstantiated) safety concerns such as lung de-recruitment [3]. While the safety and benefits of SVs have been demonstrated with patients resting in bed, no studies have examined their impact during exercise. The aim of this study was to investigate the effect of performing exercise with a SV on lung function.

A randomised crossover-controlled study was performed in a large metropolitan, tertiary referral ICU. Ethical approval was obtained and written informed consent gained from all study participants.

Participants completed two 10-minute, in-bed cycle ergometry sessions in random order, either with a SV (intervention) or without a SV (control) in situ (Methods, online supplement). The primary outcome was safety, determined by changes in end-expiratory lung impedance (EELI) and tidal variation impedance (TVI) via electrical impedance tomography (EIT), and via monitoring of participants’ vital signs. Secondary outcomes included exercise performance measures and participant reported experiences of exercising with/without a SV. Adverse events were defined as reaching criteria that necessitated termination of the cycling session (Table E2).

Patients with a tracheostomy were screened according to inclusion/exclusion criteria (Table E1). Twenty eligible participants were recruited between May 2021 and June 2024 (median age 68 (54–71) years, 80% male) (Table E3). All participants completed both exercise sessions, and no adverse events occurred. Exercise performance measures were similar between control and intervention sessions with no statistically significant differences (Table E4). Lung aeration (EELI) reduced during exercise in both groups (Table E5), with significantly less lung aeration loss at mid-exercise and significantly larger increases in aeration during the recovery period when exercising with a SV (Table E7 and Fig. 1A). Across control and intervention sessions, TVI increased during exercise (Table E5), with significantly larger increases in the intervention group (Table E7 and Fig. 1B).

Fig. 1.

Magnitude of change in lung aeration (EELI) and tidal variation impedance (TVI) from baseline. Data are presented as estimate (95% CI) from linear mixed effects. A: No speaking valve vs. speaking valve (end-expiratory lung impedance (EELI)), B: No speaking valve vs. speaking valve (tidal variation). T1 = baseline, T2 = mid-exercise, T3 = end-exercise, and T4 = recovery (30 min post-exercise). *Values are statistically significant between control and intervention sessions. regression.

At baseline and exercise timepoints, peripheral oxygen saturation (SpO₂) was significantly lower in the intervention group, although these differences were small and not clinically significant (Table E6). Mean oxygen saturations were maintained above 96% at all timepoints. Statistically significant differences in heart rate (HR) were observed at baseline and recovery, although these were also small and not clinically significant, with mean HR maintained below 100 bpm at all timepoints (Table E6). There were no significant differences in mean respiratory rate (RR) which was maintained below 26 across all timepoints (Table E6). Patient experience surveys (Appendix E2) completed by 18 participants demonstrated statistically significant differences in the ability to ask questions and make needs known when exercising with SVs (Figure E1 and Table E8), and 83% (n = 15) of participants preferred this intervention. Shortness of breath and satisfaction ratings were similar when exercising with or without a SV (Table E9).

The findings of this study indicate that exercise with a SV is safe and potentially beneficial for patients with a tracheostomy as highlighted by improvements in EELI and TVI. While there is currently no established minimally important difference in EELI or TVI (with impedance values expressed in arbitrary units) [4], these findings align with previous data demonstrating improved lung aeration associated with SV use for patients at rest [3], and would support a recommendation to keep a SV in situ for exercise to maximise lung recruitment and communication.

While some statistically significant differences were observed in vital signs, none were clinically significant or an indication of adverse events. Key clinical safety criteria for in-bed exercises with mechanically ventilated patients outline SpO₂ levels ≥ 90%, RR ≤ 30, and ventricular rates < 120 bpm for any tachyarrhythmias [5], all of which were maintained in this study across both sessions at all timepoints. Exercise was as effective with or without a SV, with the added benefit of SVs allowing communication. While outside the scope of this study, the ability to communicate during exercise therapy in ICU may lead to improved engagement in rehabilitation and reduced feelings of anxiety and frustration.

The main study limitation was the choice of in-bed stationary cycling to allow the use of EIT. Early mobility and rehabilitation involve a range of interventions that may combine position changes and exercise (e.g. out-of-bed mobilisation). Therefore, the results may not be comparable to different types of upright exercise in patients with a tracheostomy, and this warrants further investigation.

In conclusion, our study indicates that exercise with a SV is safe and may be beneficial for improving lung aeration after exercise. It is also preferred by most patients compared to exercise without a SV. The findings of this study support patients with a tracheostomy and SV being able to safely engage in physical rehabilitation while maintaining autonomy in communication.

Abbreviations

Bpm

Beats per minute

EELI

End-expiratory lung impedance

EIT

Electrical impedance tomography

HR

Heart rate

ICU

Intensive care unit

RR

Respiratory rate

SpO₂

Peripheral oxygen saturation

SV

Speaking valve

TVI

Tidal variation impedance

Author contributions

L.Ch., L.C., J.F.F., A.M., J.P., A.S., P.T., S.VW., and O.T. conceived the study and contributed substantially to the study design and writing of the manuscript. L.Ch., L.C., and N.M. had full access to all of the data in the study, contributed substantially to formal data analysis and interpretation of the results, and take full responsibility for the integrity of the data and accuracy of the data analysis. N.M. also contributed substantially to the writing of the manuscript. All authors have read and approved the final version of this manuscript.

Funding

This trial is supported by a New Investigator Grant from The Prince Charles Hospital Common Good Foundation.

Data availability

The datasets are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by The Prince Charles Hospital Human Research Ethics Committee on 14/08/2018 (HREC/18/QPCH/183) and conducted in accordance with the ethical standards of the Declaration of Helsinki. Informed consent to participate was gained from all participants.

Clinical Trial Number

The study was registered prior to commencement through the Australian and New Zealand Clinical Trials Registry (ANZCTR), Clinical Trial Number: 12619000148178.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.