HONOLULU — For community-acquired methicillin-resistant Staphylococcus aureus infections of the face, cephalexin should rarely be the antibiotic of choice, according to a decision analysis presented by researchers at the University of California, San Francisco, and at the annual meeting of the Pediatric Academic Societies.
And a second study, from Cardinal Glennon Children's Medical Center in St. Louis, questioned whether an antibiotic is needed at all to treat a primary skin infection if abscesses are incised and drained.
In the San Francisco study, Dr. Adam Hersh and associates weighed the trade-offs associated with three antibiotic choices—cephalexin, trimethoprim/sulfamethoxazole, or clindamycin—for empiric treatment of a purulent skin infection in a child younger than 18.
Cephalexin does not treat CA-MRSA; trimethoprim/sulfamethoxazole does not treat group A Streptococcus; and although clindamycin treats both, CA-MRSA is becoming increasingly resistant to it in some communities, said Dr. Hersh.
When community physicians consistently culture skin infections and prevalence and resistance rates can be developed, a decision tree analysis can be a method of “exquisitely weighing of the tradeoffs between treatment choices,” he said.
In communities that have a prevalence of CA-MRSA of greater than 10%, cephalexin is least likely to be effective, although it is the most widely prescribed antibiotic for this indication, the researchers noted.
Choosing between trimethoprim/sulfamethoxazole and clindamycin remains a delicate decision, depending on CA-MRSA resistance and prevalence of group A strep in an individual community.
In San Francisco, the overall prevalence of S. aureus in cultures of purulent pediatric skin infections is 90% (80% caused by CA-MRSA; 20% by methicillin-sensitive S. aureus). Of the remaining 10%, cultures show group A strep more than 99% of the time, said Dr. Hersh in an oral, late-breaking presentation.
Using these figures for a “base case” analysis, he concluded that probability rates for each drug having activity against an empiric skin infection were 95% for clindamycin, 89% for trimethoprim/sulfamethoxazole, and 28% for cephalexin.
Higher group A strep prevalence or high-rate clindamycin resistance in a community would tilt the decision tree model to trimethoprim/sulfamethoxazole or cephalexin, but it would be the rare community in which CA-MRSA prevalence was so low it would favor cephalexin for skin infections or other CA-MRSA affected conditions such as osteomyelitis or septic arthritis, said Dr. Hersh.
Dr. Myto Duong and associates in St. Louis selected tri-methoprim/sulfamethoxazole for a randomized, controlled, double-blind trial comparing antibiotic treatment of skin and soft tissue infections with treatment with incision and drainage in 161 immunocompetent children who presented to the emergency department at Cardinal Glennon Children's Medical Center. About half of the subjects were less than 5 years old.
Wound cultures showed CA-MRSA in 129 children (80%)—with 18% also clindamycin resistant—and methicillin-sensitive S. aureus in 14 (9%). Other bacteria were responsible for infections in remaining cases, including group A strep in 1%. Twelve patients were lost to follow-up.
In patients with complete data, there was complete resolution of lesions in 95% receiving a placebo after incision and drainage; and 96% of those receiving incision, drainage, and a 10-day antibiotic prescription before discharge. Compliance (taking at least half of the medication prescribed), was poor, at 66%.
Development of a new purulent skin lesion after treatment was equally likely in compliant patients on antibiotic or placebo. In the noncompliant subset, receipt of an antibiotic cut the risk of developing a new purulent skin lesion. In this group, 23% of those on placebo developed a new lesion versus 4% who received an antibiotic prescription.
Dr. Duong and Dr. Hersh disclosed no financial conflicts.