Invasive group A streptococcal infection : host and pathogen interactions

Abstract

Invasive group A streptococcal (GAS) infections were studied in 229 patients with a clinical diagnosis of erysipelas and 151 patients with GAS bacteremia, respectively.

Beta-hemolytic streptococci were identified in about a third of the erysipelas patients, mostly group A, whereas group G streptococci were identified in about half as many episodes. Serological findings supported GAS and GGS as dominant etiology. Erysipelas episodes usually followed a benign course with few severe complications, but relapses were common (20%), and the streptococcal reservoir for many episodes of primary as well as relapsing erysipelas remained unknown. GAS isolates from erysipelas patients were shown to be genetically diverse as judged by serotype distribution and diversity of emm gene hybridization patterns, as well as profiles of pyrogenic exotoxin production. The emm hybridization patterns among T1M1 isolates indicated that the same subclones were prevalent among erysipelas patients and contemporary T1M1 bacteremia episodes.

The incidence of GAS bacteremia in the study area peaked during 1988, coincident with a national outbreak of T1M1 GAS infections in Sweden, and returned thereafter to a lower level. Incidence increased with age, and was about three times as high among residents over 65 years of age as in younger age groups. Thirteen percent of the bacteremic patients developed streptococcal toxic shock syndrome (STSS) with a mortality of 47%, whereas overall mortality was 11%. T1M1 infections appeared epidemically 1986 to 1990 and again 1993 to 1995. This serotype as well as alcohol abuse were associated with a significantly increased risk for the development of STSS. Mortality was independently associated with increasing age, severe underlying disease, and the development of STSS.

Genetic subtyping, using restriction fragment length polymorphism on whole cell DNA obtained from bacteremia isolates during a 13 years period, showed great genetic diversity. Superimposed epidemics of genetically homogenous T1M1 isolates were found to be an increased risk for STSS. The speA gene, noticed primarily among T1M1 and T3 isolates, was associated with a greatly increased risk for STSS and statistical analysis indicated that the speA gene, encoding streptococcal pyrogenic exotoxin A (SpeA), was a factor significantly associated with STSS. In addition, patients with STSS but not erysipelas totally lacked SpeA neutralizing ability in their serum and had low ability to neutralize SpeB.

Opsonizing activity in patient sera, studied by a chemiluminescence method, was low in acute sera from patients with tonsillitis, erysipelas as well bacteremia. A substantial increase of opsonizing activity appeared after five to six weeks. Opsonization by a single immune serum showed great variability in susceptibility between wild T1M1 isolates collected during 1980 to 1998. These data indicated that opsonizing immunity may be strain specific rather than serotype specific. However, isolates from the years 1987 to 1990 were generally better opsonized than isolates from before or after this period, possibly indicating ongoing antigenic shifts in surface components affecting resistance to opsonization of T1M1 isolates.