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.

233 patients with erysipelas, admitted to the Department of Infectious Diseases, Danderyd Hospital, during a 2-year period were analysed for epidemiological, bacteriological and complicating features. Erysipelas was defined clinically as a febrile skin infection with a sudden onset of a red indurated expanding plaque with a distinct border. Common predisposing factors were alcohol abuse, diabetes mellitus and venous insufficiency, and complications were more common among such patients. No seasonal variation was found. 5% of patients with blood culture had streptococcemia (7/149). Erysipelas emerging from an infected ulcer was seen in 52% (122/233) and in 46% of these streptococci were isolated (57/122), 67% of which were of type A (38/57). Staphylococcus aureus was isolated from 59% of ulcerative cases (72/122) and in 3 of them staphylococci were found in the blood.

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.

After an average follow-up time of three years, recurrent erysipelas was observed in 29% of 143 patients admitted primarily with erysipelas. Nineteen patients (13%) had two or more recurrences during this period. The predisposing factor with the highest recurrence rate was venous insufficiency. Regular prophylaxis with phenoxymethylpenicillin (or erythromycin in penicillin allergics) after the second recurrence may be cost-effective. This antibiotic prophylaxis is only recommended in patients with predisposing factors who have suffered severe attacks.