We read with interest the article by Harrison and colleagues [1] describing relapsing group B streptococcal infection. We report a case of recurrent bacteremia in a patient with humoral immunodeficiency.

A 73-year-old woman with diabetes, chronic obstructive pulmonary disease, hypertension, and penicillin allergy was hospitalized three times in 4 months with recurrent group B streptococcal bacteremia. She initially presented with fever and bronchitis, and blood cultures done at that time grew group B streptococci. She received cephalosporins for 10 days and 5 weeks later presented with fever and cellulitis of the left leg. A blood culture grew group B streptococci, and the patient then received erythromycin for 11 days. Six weeks later, fever and cellulitis returned. The blood culture again grew group B streptococci. Cultures from the patient's pharynx, interdigitis, stool, urine, and vagina were negative for group B streptococci. Colonoscopy, esophago-gastroduodenoscopy, and transesophageal echocardiography were unrevealing. The patient received intravenous vancomycin for 10 days, clindamycin for 10 days, then rifampin for 5 days [2]. Four months later, she remains well.

Streptococcal isolates showed serotype V (Baker CJ. Personal communication). Evaluation of the patient's humoral immune system showed quantitative and functional deficiencies [3]. Levels of IgG and IgA were 370 mg/dL (normal, 680 to 1445 mg/dL) and 42 mg/dL (normal, 71 to 407 mg/dL), respectively. Subclasses of IgG, except IgG (₃), were low: IgG₁, 307 mg/dL (normal, 450 to 900 mg/dL); IgG₂, 117 mg/dL (normal, 180 to 530 mg/dL); IgG₃, 59 mg/dL (normal, 13 to 80 mg/dL); and IgG₄, less than 7 mg/dL (normal, 8 to 100 mg/dL). The patient did not respond to tetanus vaccination and only partially responded to pneumococcal vaccination (Table 1).

This patient was at risk for group B streptococcal infections because she had diabetes [4] and humoral immunodeficiency. We suggest that, in addition to identifying risk factors outlined in recent reports [1, 4], the humoral immune status of patients be evaluated. Although no consensus on optimal treatment exists, aggressive antibiotic therapy [3] may prevent subsequent recurrence.