Experimental endocarditis in rabbits. 4. Experiments with Serratia marcescens: on the significance of serum susceptibility and proteolytic capacity of the strains and the influence of an indwelling catheter.

In order to investigate the course of Serratia marcescens endocarditis in groups of rabbits with and without an indwelling catheter, 130 rabbits were pretreated to produce left-sided endocarditis. Three clinical isolates of S. marcescens were used to infect the rabbits, i.e. CDC O13 (serum sensitive, proteolytic), SM 104 (serum resistant, proteolytic) and SM 55 (highly serum resistant, non-proteolytic). Ten rabbits with an indwelling catheter were challenged with CDC O13 and none of them died or showed evidence of endocarditis 28 days later. In groups of rabbits with indwelling catheters which were challenged with SM 104 or SM 55 there was a high incidence of endocarditis (19/20, 18/20, respectively), while groups without catheters inoculated with the same strains had a lower incidence (5/20, 15/20, respectively). In contrast to earlier observations with Streptococcus faecalis, the clinical and pathological data were not significantly influenced by the presence or absence of proteolytic capacity of the infecting strains. The results indicate that the ability of S. marcescens to establish endocarditis depends significantly on the degree of serum resistance of the strains. This difference was only demonstrable in experiments without an indwelling catheter during the infection period. The distrurbing influence of an indwelling catheter is discussed, and it is concluded that experimental models using indwelling catheters are inappropriate for studies on the pathophysiology of endocarditis.