Synergy with double and triple antibiotic combinations compared.

Combinations of two sets of antibiotics (gentamicin-carbenicillin-rifampicin and trimethoprim/sulphamethoxazole-carbenicillin-rifampicin) were tested against isolates of Pseudomonas maltophilia. An abbreviated checker-board method permitted tests with combinations of all three antibiotics and of each of the possible three pairs in each set. Triple combinations were usually synergistic, whereas 25% of combinations of pairs were antagonistic or additive. The modal degree of synergy was greater with combinations of three antibiotics than with combinations of any two. For any isolate, greater synergy was generally obtained with a triple than with a double combination, and the mean degree of synergy was always greater with triple than with double combinations. It was advantageous to add to double combinations even an antibiotic to which the organism was resistant. Organisms multiply resistant to three antibiotics were also resistant to 25% of combinations of pairs, but never resistant to combinations of all three. Synergy was more likely to be present at all antibiotic ratios with triple than with double combinations. Since ratios at the site of action may differ greatly from that of the administered combination, there would be substantial clinical benefit in identifying sets of antibiotics of which combinations at all ratios showed synergy. The search for these may entail exploring multicomponent combinations and, with conventional methods, this raises a considerable logistic problem. A strategy is therefore proposed, designed to find antibiotic sets showing synergy at all ratios.