Recent evidence suggests that inhaled furosemide relieves dyspnoea in patients and in normal subjects made dyspnoeic by external resistive loads combined with added dead-space. Furosemide sensitizes lung inflation receptors in rats, and lung inflation reduces air hunger in humans. We therefore hypothesised that inhaled furosemide acts on the air hunger component of dyspnoea. Ten subjects inhaled aerosolized furosemide (40 mg) or placebo in randomised, double blind, crossover experiments. Air hunger was induced by hypercapnia (50+/-2 mmHg) during constrained ventilation (8+/-0.9 L/min) before and after treatment, and rated by subjects using a 100 mm visual analogue scale. Subjects described a sensation of air hunger with little or no work/effort of breathing. Hypercapnia generated less air hunger in the first trial at 23+/-3 min after start of furosemide treatment (58+/-11% to 39+/-14% full scale); the effect varied substantially among subjects. The mean treatment effect, accounting for placebo, was 13% of full scale (P=0.052). We conclude that 40 mg of inhaled furosemide partially relieves air hunger within 1h and is accompanied by substantial diuresis.