The bacteriologic, serological, and clinical characteristics of 229 patients with erysipelas diagnosed during a 2-year period at a university hospital for infectious diseases in Sweden are presented. Beta-Hemolytic streptococci were detected in 34% of these patients. Group A was the dominant serogroup, but group G streptococci were found in about half as many cases. Bacteremia was present in 5%. A serological response with antistreptolysin O (ASO) and antideoxyribonuclease B (ADNase B) was seen primarily in patients harboring group A streptococci but also in those from whom no pathogen was isolated. ASO was also found in high titers in some patients with Group G streptococcal infection. The clinical course was usually benign, with few complications, but recurrences were common (occurring in 21% of the patients). No cases of streptococcal toxic shock were seen. Culture of skin biopsy specimens had low sensitivity; Beta-hemolytic were isolated from only two of 15 patients.

Recurrences of erysipelas are especially prevalent in patients suffering from local impairment of circulation and intervention might thus be of benefit. Therefore a prospective, randomized, open study was undertaken to evaluate whether daily antibiotic prophylaxis would reduce the risk of recurrence. Patients with venous insufficiency or lymphatic congestion who had suffered two or more episodes of erysipelas during the previous 3 years and were admitted to the Infectious Disease Department at Roslagstull Hospital, Stockholm, Sweden, between November 1988 and November 1991 were included. Fourty patients, 20 on prophylaxis and 20 controls were followed according to a life table analysis during a median time of 15 months. Phenoxymethylpenicillin was prescribed as daily prophylaxis (while erythromycin was given to patients allergic to penicillin). Recurrences of erysipelas appeared to be reduced by daily antibiotic prophylaxis but the effect was not dramatic (p = 0.06). Only in patients with a high recurrence rate continuous antibiotic prophylaxis against erysipelas is indicated.

Group A streptococcal isolates (n = 53) recovered from 38 erysipelas patients in 1988 and 1990 in Sweden were analysed with respect to serotype, erythrogenic toxin production and polymorphism in the emm gene region. Serotype determination showed a dominance of type T1M1 (28.6% of the strains), but T type 8 was also prevalent (14.3%). In the majority of the strains only a low production of erythrogenic toxin A was demonstrated, while both toxin B and C production were high. Polymorphism was detected in the emm gene region of T1M1 strains at a frequency of 64%. These erysipelas associated group A streptococci were more heterogenic with respect to serotype distribution and polymorphism in the emm gene region compared to previously studied group A streptococci isolated during an outbreak of serious streptococcal infections in Sweden in 1988/1989. The material included isolates from two cases of recurrence, and typing of the isolates indicated that the patients had been infected by the same serotype as in the primary infection.

In 45 patients hospitalized with febrile erysipelas that had been treated with oral penicillin, punch biopsies of infected skin were performed and analysed for phenoxymethylpenicillin (pcV) concentrations. The curves for the concurrent serum and tissue levels followed the same course, indicating a rapid diffusion of pcV from serum into tissue. Penicillin concentrations in infected skin exceeded the minimal inhibitory concentrations (MIC) of the streptococci isolated for the first 4 h after tablet ingestion. Venous insufficiency was highly prevalent and pcV concentrations in affected patients were slightly but not significantly higher. A theoretical basis for the successful treatment of erysipelas with oral penicillin is therefore at hand.

No abstract