The case described by Awayda and colleagues raises several relevant issues about recurrent adult group B streptococcal infection. The short intervals between their patient's episodes and the serotyping data suggest relapsing infection [1]. However, given the high proportion of adult group B streptococcal infection caused by serotype V [2], molecular subtyping is needed to determine whether the subsequent episodes were due to relapse or to infection with different strains. In one of our patients with recurrent group B streptococcal infection caused by serotype V organisms, restriction endonuclease analysis of chromosomal DNA indicated that the patient had infection with two different strains of streptococci [1].

We agree with the aggressive diagnostic evaluation that Awayda and colleagues used to determine whether their patient had deep-site infection. Of the 12 patients evaluated in our series, 6 were found to have endocarditis, osteomyelitis, or septic arthritis [1]. Recommendations have been made for the treatment of deep-site infection caused by group B streptococci. Although aggressive antibiotic therapy is also warranted for patients with recurrent streptococcal disease who do not have an identifiable deep-site infection, the optimal antibiotic regimen is not known.

This case also raises the issue of the role of humoral immunity in susceptibility to group B streptococcal infection. Patients with isolated humoral immunodeficiency, such as those with X-linked agammaglobulinemia, rarely develop streptococcal bacteremia [4], indicating that the absence of antibodies alone is insufficient to predispose patients to invasive group B streptococcal infection. We agree that the immune defects associated with diabetes mellitus, in addition to this patient's common variable immunodeficiency, may have acted together to make her highly susceptible to streptococcal infection.

It would be of interest to know what proportion of adults with either a single or multiple episodes of invasive group B streptococcal infection have hypogammaglobulinemia in addition to other chronic medical conditions [5]. We are unaware of data that address this issue. Until this question is resolved, we agree that it is reasonable to assess patients with group B streptococcal infection for immunoglobulin deficiency, given that it remains unknown whether relapsing infection is an isolated or frequent phenomenon. This issue is of clinical importance because patients with this group of disorders can benefit from the administration of intravenous immunoglobulin. See also [3]