A prospective, randomised, single-blind comparative trial was carried out to determine whether double beta-lactam treatment with pivampicillin plus pivmecillinam is more effective than pivampicillin alone in the treatment of recurrent and chronic lung infections with Haemophilus influenzae in patients with chronic obstructive pulmonary disease (COPD) or cystic fibrosis (CF). Fifty-six children and young adults with COPD or CF were randomised to the clinical study. The patients were allocated at random to receive perorally either pivmecillinam, 40 mg/kg/day, combined with pivampicillin, 50 mg/kg/day, or pivampicillin 50 mg/kg/day alone for 14 days. A cross-over pharmacokinetic study using the same drugs was carried out in 10 CF patients to determine the antibiotic concentrations in serum and sputum after a single dose of each drug. The clinical study showed no significant differences in clinical scoring, lung function tests or adverse events after treatment with pivampicillin plus pivmecillinam or pivampicillin alone. Follow-up microbiological evaluation 2 and 6 weeks after the end of treatment showed that the offending pathogen was eradicated in 68% of the patients treated with pivampicillin plus pivmecillinam and in 67% of the patients treated with pivampicillin alone. Reinfection with another biotype was more common in the combination group (50% vs. 21%) than in the pivampicillin group. In the pharmacokinetic study the median peak serum concentration occurred two hours after intake of tablets. The efficacy of double beta lactam treatment in lung infections with H. influenzae appears to be equivalent to that of ampicillin on clinical lung symptoms, lung function tests, adverse effects and bacteriology.