Inhaled furosemide for relief of air hunger versus sense of breathing effort: a randomized controlled trial

Abstract

Background: Inhaled furosemide offers a potentially novel treatment for dyspnoea, which may reflect modulation of pulmonary stretch receptor feedback to the brain. Specificity of relief is unclear because different neural pathways may account for different components of clinical dyspnoea. Our objective was to evaluate if inhaled furosemide relieves the air hunger component (uncomfortable urge to breathe) but not the sense of breathing work/effort of dyspnoea.

Methods: A randomised, double blind, placebo-controlled crossover trial in 16 healthy volunteers studied in a university research laboratory. Each participant received 3 mist inhalations (either 40 mg furosemide or 4 ml saline) separated by 30-60 min on 2 test days. Each participant was randomised to mist order 'furosemide-saline-furosemide' (n- = 8) or 'saline-furosemide-saline' (n = 8) on both days. One day involved hypercapnic air hunger tests (mean ± SD PCO2 = 50 ± 3.7 mmHg; constrained ventilation = 9 ± 1.5 L/min), the other involved work/effort tests with targeted ventilation (17 ± 3.1 L/min) and external resistive load (20cmH2O/L/s). Primary outcome was ratings of air hunger or work/effort every 15 s on a visual analogue scale. During saline inhalations, 1.5 mg furosemide was infused intravenously to match the expected systemic absorption from the lungs when furosemide is inhaled. Corresponding infusions of saline during furosemide inhalations maintained procedural blinding. Average visual analogue scale ratings (%full scale) during the last minute of air hunger or work/effort stimuli were analysed using Linear Mixed Methods.

Results: Data from all 16 participants were analysed. Inhaled furosemide relative to inhaled saline significantly improved visual analogues scale ratings of air hunger (Least Squares Mean ± SE - 9.7 ± 2%; p = 0.0015) but not work/effort (+ 1.6 ± 2%; p = 0.903). There were no significant adverse events.

Conclusions: Inhaled furosemide was effective at relieving laboratory induced air hunger but not work/effort in healthy adults; this is consistent with the notion that modulation of pulmonary stretch receptor feedback by inhaled furosemide leads to dyspnoea relief that is specific to air hunger, the most unpleasant quality of dyspnoea.

Funding: Oxford Brookes University Central Research Fund.

Trial registration: ClinicalTrials.gov Identifier: NCT02881866 . Retrospectively registered on 29th August 2018.

Keywords: Aerosolized; Breathlessness; Dyspnoea; Hypercapnia; Loop diuretics; Nebuliser; Pulmonary stretch receptors; Resistive load.

Conflict of interest statement

Ethics approval and consent to participate

Oxford Brookes University Research Ethics committee approved the protocol (UREC Registration No: 150939) and all participants provided written informed consent.

Consent for publication

Consent to publish was obtained from all participants.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures

Fig. 1

Breathing circuit. The Breathing Circuit was identical for air hunger (AH) test and work/effort (WE) test, except that the external resistance was removed in the AH test. To elicit AH, CO2 was added to the flow of fresh gas into the bag and this flow was fixed at baseline alveolar ventilation. To elicit WE, individuals were instructed to empty the bag with each breath while the flow of fresh bag into the bag was increased and CO2 was added to maintain normocapnia.  PETCO2 = end tidal PCO2

Fig. 2

Standard tests of air hunger and work/effort. Left: Typical raw data set for the air hunger (AH) test during which two levels of end-tidal PCO2 were imposed and ventilation was constrained. The vertical dashed lines indicate the steady state level of AH associated with the test level of CO2 chosen to elicit 50%VAS ratings in pre-mist trials. Right: Typical raw data set for the work effort (WE) test in which two levels of targeted VT were imposed and normocapnia was maintained. The vertical dashed lines indicate the steady state level of WE associated with the test level of VT chosen to elicit 50% VAS ratings in pre-mist trials. During both tests ventilatory constraint or targeting was suspended briefly and participants were instructed to take a sigh. VAS ratings were provided every 15 s in response to a LED cue. VT Tidal volume, PAW continuous airway pressure measured at the mouth

Fig. 3

Effect of mist inhalations on steady state air hunger (AH) and work/effort (WE). This dataset is from an individual who received the mists in the order of furosemide-saline-furosemide (FSF) with the corresponding saline-furosemide-saline (SFS) intravenous infusions on both days. Panels a to f show the last minute of each test level of end-tidal CO2 (for AH) or of VT (for WE) –these regions of interest are shown by the vertical dashed lines in Fig. 2. AH test day: Air hunger ratings were reduced after furosemide inhalation (a to b and e to f) but not after saline inhalation (c to d). WE test day: No obvious differences in ratings were evident before and after any mist inhalations

Fig. 4

Patient Flow Diagram

Fig. 5

Overall changes in air hunger (AH) and work/effort (WE) associated with mist inhalations. Panel a. Mean ± SEM AH (left panels) and WE (right panels) before and after furosemide inhalations (black bars) and before and after saline inhalations (grey bars) in the 8 individuals who were allocated to the saline-furosemide-saline order of mist inhalations (top panels) and in 8 individuals who were allocated to the furosemide-saline-furosemide order of mist inhalations (bottom panels). VAS ratings improved to a greater extent after furosemide compared to saline mist inhalations for AH, but this pattern was not evident for WE. ETCO2 = end tidal CO2 (mean ± SD mmHg). Panel b. Least Squares Mean change in VAS ratings before and after inhaled furosemide relative to the change before and after inhaled saline for AH and WE

Fig. 6

Individual data for change in visual analogue scale for air hunger (AH) and work/effort (WE). Individual change in visual analogue scale ratings (VAS, % full scale) of AH at fixed test levels of PETCO2 (left panel), and of the sense of breathing WE at fixed test levels of tidal volume (right panel) following inhaled furosemide (dark bars) and inhaled saline (grey bars). Closed bars indicate the average change in VAS for two furosemide inhalations in half the participants (S1, 2, 4, 7, 10, 11, 14, 17) or the change in VAS for one furosemide inhalation in the other half. Open bars indicate the average change in VAS for two saline inhalations in half the participants (S3, 5, 6, 8, 9, 12, 13, 15) or the change in VAS for one saline inhalation in the other half. Inhalation of furosemide tends to produce a reduction in VAS after furosemide more often than after saline for the AH test. For WE test reductions were evident for both inhaled furosemide and inhaled saline. Participants are arranged in order of response to furosemide for the AH test

Fig. 7

Second dose effect. Left panels (a and c): Individual (n = 8) changes in VAS ratings of AH in response to first and second doses of inhaled furosemide in the furosemide-saline-furosemide (FSF) group (a). Corresponding changes for the first and second doses of inhaled saline in the saline-furosemide-saline (SFS) group (c). For the FSF group the second dose of furosemide had a greater reduction in AH relief than the first dose in all but one participant. This was not true for the second dose of saline in the SFS group. Participants are arranged in order of response to first dose of furosemide for the FSF group or first dose of saline for SFS group. Right panels (b and d): The mean reduction in AH for the first and second dose of furosemide (b) and saline (d)

Fig. 8

Dyspnoea Descriptors. Frequency with which each descriptive phrase was selected by participants to describe their experience during air hunger tests (AH; left panel) and WE tests (WE; right panel). AH cluster of descriptors dominated the participants’ choice of the respiratory sensations felt during the AH tests while the WE cluster of descriptors dominated the participants’ choice of the respiratory sensations felt during the WE tests