Salbutamol solution is usually administered by nebuliser in a dose of 5 mg. Little evidence exists that this is the optimal dose for bronchodilatation or that this dose is without sideeffects. Twelve patients with asthma were given increasing doses of salbutamol, 1 5 mg, 3-0 mg, 7-5 mg, and placebo. Treatments were administered twice daily for four days in a double-blind manner. Measurements of ventilatory capacity, pulse rate, and tremor were recorded before and for three hours after treatment. There was a significant dose-related response for FEVI and peak flow rate. There was also a significant dose-related response in pulse rate and tremor. The incidence of palpitations was similarly related to dose. Plasma levels of salbutamol were measured before and after treatment with salbutamol and showed a dose related increase in salbutamol absorption which begins to be evident after the 3 0 mg dose. Three milligrams of salbutamol nebuliser solution may be an optimal dose, producing satisfactory bronchodilatation but fewer side-effects related to systemic absorption.Adequate bronchodilatation with minimal systemic side-effects can be achieved with 100 jug of aerosolised salbutamol in patients with mild or moderate airways obstruction caused by asthma.' For cases with severe asthma, salbutamol respirator solution is commonly prescribed using various delivery techiques.Previous studies of salbutamol respirator solution have shown adequate bronchodilatation using 10mg2 and 5 mg.3 These doses may be associated with systemic side effects of tremor and palpitations. To clarify the optimal dose of salbutamol respirator solution without associated side-effects, three doses were compared and correlated to the plasma levels of salbutamol. MethodsTwelve patients (seven men, five women; mean age 56 years, range 24 to 67 yr) were studied. All had been admitted to hospital with persistent uncontrolled wheezing and a peak expiratory flow rate (PEFR) of less than 50 % of predicted normal. Most were receiving concurrent oral or inhaled corticosteroids and this dose was not altered during the study. No bronchodilator drugs were administered for 12 hours before study. Four treatment regimens were compared using a modified Latin Square distribution. The regimens were (1) placebo, (2) salbutamol 1-5 mg in 3 ml solution, (3) salbutamol 3-0 mg in 3 ml solution, and (4) salbutamol 7-5 mg in 3 ml solution.These were administered double-blind over a study period of four days, two treatments being given each day at 0900 and 1400 hours respectively. Each treatment was given once in the morning and once in the afternoon. Drugs