Treatment of Pseudomonas aeruginosa lower respiratory tract infections with beta-lactam antibiotics alone (beta-lactam monotherapy) has been thought to result in a high incidence of therapeutic failure due to the emergence of multiply-resistant strains on the basis of induction of a chromosomal beta-lactamase. Review of published experience in patients without granulocytopenia or cystic fibrosis suggests that favorable clinical responses can be obtained in 80-90% of cases, and bacteriological cures in 45-55%, using any of the newer beta-lactam antipseudomonal agents alone (data from cefsulodin, cefoperazone, azlocillin and piperacillin). Resistance develops in 30-40% of the infecting organisms, and is associated with treatment failure in 10-20% of cases (data from cefsulodin, ticarcillin and carbenicillin). Cross-resistance to other beta-lactams and to aminoglycosides can occur but seems unlikely to be on the basis of induction of a chromosomal beta-lactamase (data from cefsulodin). The addition of an aminoglycoside antibiotic (combination therapy) has been recommended to prevent these outcomes. Retrospective comparison with results obtained using combination therapy in patients without granulocytopenia or cystic fibrosis suggests that the addition of an aminoglycoside cannot be expected to prevent either the development of resistance or therapeutic failure (which are frequently unassociated). Treating every patient with a Pseudomonas aeruginosa lower respiratory tract infection with combination therapy will expose all of them to the toxicity of an aminoglycoside but will rarely repay this risk with the prevention of a multiply resistant strain or the salvage of a patient destined to fail beta-lactam monotherapy.