Isolation of Group a Streptococci From Children With Perianal Cellulitis and From Their Siblings

Perianal cellulitis caused by group A beta-hemolytic streptococci occurs mainly in children ages 6 months to 10 years. Rarely has the disease been described in adults. The incidence of perianal cellulitis ranges from 1 in 2000 to 1 in 218 pediatric outpatients visits.1 It is a characterized by sharply circumscribed perianal erythema accompanied by varying degrees of itching, anal fissures, painful defecation or constipation, rectal bleeding and mucopurulent anal discharge.2-8 The epidemiology of this infection is largely unclear. The disorder can be associated with guttate psoriasis7, 9 and with vulvovaginitis in prepubescent girls. In boys involvement of the penis can occur.9-11 The anal carrier rate in children with positive group A streptococcal pharyngeal cultures has been reported to be 6%.12 Two reports of intrafamily spread 1, 13 and an outbreak of perianal streptococcus infection in a day-care center have been published.14 Treatment of perianal streptococcal cellulitis has never been prospectively evaluated. Treatment of sporadic cases with oral penicillin,5 erythromycin8 or mupirocin cream alone4 has been effective, but recurrence rates as high as 40 to 50% have also been reported.15 The purposes of the present prospective study were to determine the frequency of concomitant throat and skin lesion of group A beta-hemolytic streptococci (GABHS) in children with perianal streptococcal cellulitis and their siblings, before and after antimicrobial treatment, and to evaluate the efficacy of oral amoxicillin 40 mg/kg daily divided into three equal doses for 10 days and topical application to the perianal area of mupirocin 2%, 3 times per day, for treatment. Methods. The study was conducted between January, 1992, and December 31, 1996. Eighteen children ages 7 months to 5.4 years were enrolled. Patients were entered into the study if they had erythema in the perianal region and positive anal swab culture for GABHS and after signed informed consent was obtained from their parents. All siblings living in the same house as the index patient were examined within 48 h of diagnosis. Medical history and signs of GABHS infection were recorded and pharyngeal cultures were obtained. In cases with skin lesions appearing to be pyoderma or impetigo, those sites were cultured as well. Rectal swab cultures were taken from all siblings younger then 10 years of age. Cultures from the perianal area, from throat swab specimens and from skin lesions were collected before treatment, 3 to 5 days into treatment and 2 to 3 weeks posttreatment. Patients who had received antimicrobial agents within 72 h of study enrollment or those treated with any other investigational drug within 30 days were excluded. Other reasons for exclusion included a serious underlying condition and history of hypersensitivity reaction to betalactam antibiotic. Children with perianal streptococcal cellulitis were treated with amoxicillin 40 mg/kg daily, divided into three equal doses for 10 days and with topical application of mupirocin 2%, 3 times per day. Children with recurrence were retreated for additional 7 days. The clinical and bacteriologic responses for each patient were assessed at 3 to 5 days into treatment and at 2 to 3 weeks posttreatment follow-up, and in a telephone interview 1 year later. Clinical responses were defined as cure (resolution of signs and symptoms of infection) and failure (persistence or worsening of signs and symptoms of infection). Bacteriologic response to therapy was evaluated by determining the presence or absence of GABHS in the perianal cultures obtained at follow-ups conducted 3 to 5 days into therapy and 2 to 3 weeks posttreatment. Bacteriologic success was defined as elimination of GABHS in the posttreatment follow-up perianal cultures, and failure was regarded as no improvement or return of signs and symptoms suggestive of perianal dermatitis with a positive perianal culture for GABHS. Results. Eighteen patients (10 male and 8 female), ages 7 months to 5.4 years, were enrolled into the study. Three patients had perianal erythema as the sole clinical sign (Table 1); the remaining 15 patients had other concomitant clinical manifestations: 9 (50%) perianal pruritus; 8 (44%) constipation or painful defecation; 7 (39%) fissura ani; 5 (27%) purulent secretion; and 3 (16%) rectal bleeding. Patients 6 (with guttate psoriasis) and 7 were brother and sister who presented with onset of disease 3 days apart. Patient 11 (brother of Patient 10) was diagnosed as having perianal cellulitis after GABHS was isolated from his perianal area as part of the requirements of the sibling's protocol, at physical examination he had perianal erythema, pruritus and purulent secretion that had been unnoticed by his parents.TABLE 1: Clinical features of patients and siblings of patients with perianal streptococcal dermatitis Thirteen (72%) patients (Table 1) had additional isolates of GABHS: 5 from the pharynx; 4 from skin lesions and 8 from cultures taken from the vulva or penis. Four of 13 had GABHS isolated from more than 1 place. In 4 siblings GABHS was isolated from the pharyngeal swab; none had clinical symptoms compatible with streptococcal disease. Sixteen patients were clinically cured after 10 days of the treatment protocol, and perianal cultures on Treatment Days 3 to 5 and Posttreatment Weeks 2 to 3 were negative for GABHS. Patients 3 and 16 had recurrence of the clinical signs and symptoms as well as positive perianal culture for GABHS; 1 week after they had completed the treatment protocol they were successfully retreated. Discussion. Streptococcal perianal cellulitis may be more frequent than previously thought because similar symptoms and signs are commonly seen with other dermatologic conditions. The mean duration of symptoms before correct diagnosis ranges from a few weeks to months.8 Fifteen (83%) of the children in the present study had several signs and symptoms including pruritus, constipation, painful defecation, fissura ani, purulent secretion and rectal bleeding. Only three patients had perianal erythema as the sole clinical symptoms. Asymptomatic anal carriers have been implicated in epidemics. Schaffner et al.16 described an outbreak of 21 postoperative wound infections traced to an asymptomatic anal carrier of GABHS. Two reports of intrafamily spread have been published.1, 13 Hirschfeld et al.13 suspected that fomites or shared baths are possible sources of infection but the exact epidemiology of these infections remains unclear. Kokx et al.1 could demonstrate intrafamily spread in 50% of possible cases. The route of transmission could not be traced, but common bathing habits at home were suspected. A small outbreak of 3 culture-confirmed, genetically identical, perianal streptococcal infections and 2 additional nonverified clinical perianal infections occurring within 9 days in a group of 20 day-care children has been reported.14 Despite molecular epidemiology investigation the route of transmission remained unclear. Our study shows that GABHS is commonly isolated from the pharynx, skin lesions, vulva or penis of children with streptococcal perianal cellulitis and from pharyngeal swabs of 4 of 48 (8%) asymptomatic siblings. We speculate that the intense itch accompanying perianal infection caused the children to scratch the anal area and thereby to transport the etiologic pathogens from the anal area to the throat and other part of the body. This hypothesis is strengthened by the observation that children with streptococcal pharyngitis had only a 6% anal carriage rate.13 Treatment of sporadic cases with oral penicillin,5 erythromycin8 or mupirocin cream alone4 or with oral penicillin5 has been effective, but recurrence rates as high as 40 to 50% have also been reported.15 In our study combined systemic and topical therapy was successful in most patients. Asher Barzilai, M.D. Herman-Avner Choen, M.D.