Treating cellulitis: meta-analysis hampered by low-quality evidence

  • Brindle R & al.
  • JAMA Dermatol
  • 12 Jun 2019

  • International Clinical Digest
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.

Takeaway

  • A systematic review and meta-analysis failed to determine the most effective treatment for cellulitis, citing low-quality evidence.

Why this matters

  • The optimal antibiotic treatment for cellulitis is unclear.

Key results

  • Cellulitis was the primary diagnosis in 35% of included studies.
  • No evidence supported a specific antibiotic as most effective:
    • Penicillin vs cephalosporins: risk ratio (RR), 0.98; 95% CI, 0.68-1.42.
    • Cephalosporin A vs cephalosporin B: RR, 1.02; 95% CI, 0.96-1.09.
    • β-lactam vs macrolide, lincosamide, or streptogramin: RR, 0.94; 95% CI, 0.85-1.04.
    • Quinolone or vancomycin vs other antibiotic: RR, 1.04; 95% CI, 0.94-1.16.
    • Linezolid vs other antibiotic: RR, 1.00; 95% CI, 0.95-1.05.
    • Clindamycin vs trimethoprim sulfamethoxazole: RR, 1.05; 95% CI, 0.96-1.15.
  • Antibiotics with activity against methicillin-resistant Staphylococcus aureus did not add to effectiveness (RR, 0.99; 95% CI, 0.92-1.06).
  • Intravenous antibodies were inferior to oral antibiotics (RR, 0.83; P<.001 low-quality evidence>
  • Longer treatment duration did not improve effectiveness (RR, 0.99; 95% CI, 0.94-1.04).

Study design

  • 43 studies (5999 participants) were included in the meta-analysis.
  • The primary outcome was the proportion of patients cured, improved, symptom-free, or symptom-reduced after treatment.
  • Funding: None disclosed.

Limitations

  • Most included studies lacked clear endpoints.