Neonatal Toxic Shock Syndrome

ABSTRACT & COMMENTARY

Synopsis: Twenty neonates with "neonatal toxic-shock-syndrome-like exanthematous disease" are described.

Source: Takahashi N, et al. Exanthematous disease induced by toxic shock syndrome toxin 1 in the early neonatal period. Lancet 1998;351:1614-1619.

Takahashi and colleagues in tokyo describe 20 neonates with a diffuse generalized macular erythematous rash, and one or more of the following findings in the absence of another explanatory disease process: serum CRP greater than 10 mg/L, platelet count less than 150x109/L, and rectal temperature greater than 38° C. Twelve infants were full-term and eight preterm. All but one of the former group were febrile while all but one of the latter were afebrile. When present, fever was usually first observed on the second or third day of life and lasted only one day. It was followed by the onset of the rash, which spread centrifugally from the trunk, including the palms and soles, and tended to confluency. The rash did not desquamate.

The illness was generally mild and self-limited, although two preterm infants died, one at 66 days of a tracheo-esophageal fistula, and one at 31 days of necrotising enterocolitis. While all but one preterm infant received antibiotics, only four of the 12 term infants did so. Nasopharyngeal and umbilical swab cultures were obtained from each infant; all 40 cultures yielded methicillin-resistant Staphylococcus aureus (MRSA). MRSA grew from the blood of one preterm infant. While MRSA colonization was present in approximately one-third of all neonates, fewer than 10% of those carrying MRSA developed this disorder.

Seven of the seven MRSA isolates tested produced TSST-1 and four also produced enterotoxin C. Enterotoxins A and B were not detected. Peripheral blood lymphocytes were obtained from four patients and their T cell Vb repertoire was examined by flow cytometry. Each had significant expansion of Vb2 relative to control infants (P < 0.0001). Examination of T cell receptor b-chain genes of two affected infants demonstrated that the Vb2 T cell expansion represented polyclonal activation, consistent with a superantigenic, rather than an antigenic response.

COMMENT BY STAN DERESINSKI, MD, FACP

As pointed out by Takahashi et al, the findings in these infants do not fulfill the epidemiological definition of toxic shock syndrome. They did not develop hypotension, their rash did not undergo desquamation, and their illness was of limited severity. Nonetheless, the laboratory studies are consistent with the possibility that the rash and other clinical abnormalities were caused by the staphylococcal superantigen, TSST-1.

TSST-1 is the classical staphylococcal superantigen originally associated with menstruation-related toxic shock syndrome (TSS). TSS can also be caused by several staphylococcal enterotoxins and by toxins produced by other organisms, such as Streptococcus pyogenes. Neonatal TSS has been previously reported in association with both S. pyogenes and S. aureus, the latter as part of a mother-infant pair with this syndrome.1,2

Superantigens interact with specific Vb types of which there are approximately two dozen families.3 Each superantigen has affinity for specific Vb types. Superantigens are not processed by the cell, but instead act as soluble ligands, which, in contrast to conventional peptide antigens, do not fit into the central groove formed within the class II MHC molecule, but rather attach to the side of the MHC-TCR complex. The superantigen thereby forms a bridge between the MHC molecule and virtually all T lymphocytes expressing the particular Vb determinants which recognize that superantigen. This results in the stimulation and proliferation of all those cells, resulting in an expansion of the affected Vb type, which in the case of stimulation by TSST-1 is Vb2. Thus, in contrast to the very small proportion of T cells stimulated by conventional antigens, a very large proportion of T cells may be stimulated by a superantigen. The result of the immense immunologic response may have severe consequences, including death, for the host.

TSST-1 stimulation of T cells (both CD4+ and CD8+) bearing Vb2, leads to production of interleukin-2, which in turn results in the expansion of T cells bearing this Vb specificity. The stimulated T cells also produce lymphotoxin (TNF-b) and interferon gamma. The latter leads to monocyte/macrophage activation and the production of inflammatory cytokines, including interleukin-1 and TNF-a. There is some evidence that superantigens can also directly activate these phagocytic cells to produce these molecules, which appear to be largely responsible for the clinical characteristics of toxic shock syndrome, as they are for septic shock.

The findings in the patients described by Takahashi et al raise a number of questions. There were apparently no cases of this syndrome in patients colonized with methicillin-susceptible S. aureus. Unless these cases represented an outbreak related to a clone of MRSA, it is difficult to account for this observation. What accounts for the low incidence of disease among those colonized? Takahashi et al suggest that susceptibility to the disease was not uniform but instead depended upon the absence of specific maternal antibody. What accounts for the relative lack of severity of the illness? Finally, can these data be reproduced in other centers?

References

    1. Mahieu LM, et al. Congenital streptococcal toxic shock syndrome with absence of antibodies against streptococcal pyrogenic exotoxins. J Pediatr 1995;127: 987-989.

    2. Chow AW, et al. Variant postpartum toxic shock syndrome with probable intrapartum transmission to the neonate. Am J Obstet Gynecol 1984;148:1074-1079.

    3. Schafer R, Sheil JM. Superantigens and their role in infectious disease. Adv Pediatr Infect Dis 1995;10: 369-390.