Introduction Dyspnoea is prevalent and reduces quality of life in patients with chronic disease. Inhaled furosemide offers a potential complementary novel treatment for these patients. The mechanism of action is unclear but current theory suggests sensitisation of slowly adapting pulmonary stretch receptors (saPSR) altering neural feedback that informs the brain of the level of breathing. Clinical dyspnoea comprises several components including air hunger (AH; an uncomfortable urge to breathe) and a sense of breathing work/effort (WE) which are thought to arise from different neural pathways. We therefore hypothesised that inhaled furosemide would relieve AH but not WE.
Methods A double-blind, placebo-controlled trial was conducted on healthy volunteers (n = 16; 9 males). Test sessions involved 3 inhalations of furosemide (40 mg) or saline (4 ml) separated by 30–60 mins. Order of inhalations was furosemide-saline-furosemide in half the subjects and saline-furosemide-saline in the other half. Before and after each inhalation, AH was induced with hypercapnia (mean ± SD PCO2 = 49.8 ± 3.7 mmHg) and constrained ventilation (mean ± SD 9.2 ± 1.5 l/min) on one test-day while WE was induced with targeted ventilation (mean ± SD 16.6 ± 3.1 l/min) and external resistive load (20cmH2O/L/s) on the other test-day. During saline inhalations 1.5 mg furosemide in 15ml saline was infused to match the expected systemic absorption of furosemide from the lungs over 15 mins of inhalation. Corresponding infusions of saline during furosemide inhalations maintained blinding from noticeable diuresis. Subjects rated AH or WE every 20s on a visual analogue scale (VAS). Hypercapnia (AH) or targeted ventilation (WE) were imposed for 4 mins and the ratings in the last minute were analysed using Linear Mixed Model procedure (SAS 9.4).
Results The final model produced a main effect of mist (furosemide or saline; p = 0.016), time (pre or post inhalation; p = 0.047) and a significant condition (AH or WE)*mist interaction (p = 0.004). Mean ± SE AH was significantly lowered by inhaled furosemide relative to inhaled saline (-9.7 ± 2.1 mm VAS; p = 0.0015) but mean ± SE WE was not (+ 1.6 mm ± 2.4; p = 0.903).
Conclusion Inhaled Furosemide is effective at relieving AH, but not WE. This is consistent with a mechanism involving modulation of parenchymal lung mechanoreceptor activity leading to dyspnoea relief that specifically applies to the AH component. The treatment may therefore benefit patients with the most unpleasant form of dyspnoea.