Objectives: To evaluate the usefulness of Gram staining and culture of skin lesions in patients with acute meningococcal infections.
Design: Retrospective study.
Setting: Community hospital and intensive care unit of a teaching hospital.
Subjects: 51 patients admitted from 1989 to 1993 with proved meningococcal infections and microbiological examination of specimens from skin lesions.
Interventions: Needle aspiration of a skin lesion before start of antibiotic treatment in 26 patients in the community hospital; punch biopsy of skin lesion after start of antibiotic treatment in 25 patients in the teaching hospital.
Main outcome measures: Detection of meningococci by Gram staining of specimens from skin lesions according to category of infection (meningococcaemia, meningitis, meningitis with shock, or septic shock without meningitis).
Results: Bacteria were detected in the specimen from haemorrhagic skin lesions by culture or Gram staining, or both in 32 (63%) patients. The sensitivity of the Gram stain was 51% and did not differ significantly from its sensitivity in detecting bacteria in cerebrospinal fluid. In meningococcal sepsis, however, a Gram stained skin lesion was significantly more sensitive (72%) than Gram stained cerebrospinal fluid (22%). In patients with meningitis skin lesions gave positive results on staining more often if shock was present. The results for punch biopsy specimens were not affected by antibiotics as Gram staining gave positive results up to 45 hours after the start of treatment and culture gave positive results up to 13 hours.
Conclusion: Microbiological examination of skin lesions is informative, especially in patients with sepsis and inconclusive results from cerebrospinal fluid, and may provide a diagnosis in such patients within 45 minutes. It differentiates well between meningitis with and without haemodynamic complications, and the result is not affected by previous antibiotic treatment.