We conducted the study of 71 patients admitted to Siriraj Hospital due to S. suis septicemia. The overall prevalence of S. suis IE in patients with S. suis septicemia was 26.8%, which was higher than those reported in the previous studies (6.4–25.6%).(1,2,6) However, those studies included all patients with S. suis infection, unlike our study, which included only patients with S. suis septicemia. Furthermore, those studies did not mention the frequency of the patients that had echocardiography performed. Despite a higher prevalence of S. suis IE in our study than previous literature, we expected the underestimation of S. suis IE in our study since TTE was done in less than half (42.3%) of the patients and even less for TEE (9.8%). This may be due to lack of awareness of the organism’s invasiveness and our facility’s echocardiography availability.
Regarding the aggressiveness of S. suis IE, we found that this pathogen can highly destroy cardiac valves and perivalvular structures. Our present study has shown that 80.0% and 52.6% of S. suis IE patients had significant valvular regurgitation and perivalvular complications of IE, respectively. A total of 13 out of 19 patients underwent valvular surgery according to standard guideline.(8) A study of S. suis IE from Trirattanapa et al.(4) had a similar rate of significant valvular regurgitation (81.4%) and perivalvular complications (35%) to our present study, even though a higher rate of cardiac surgery was reported (81%). We believed that this may be due to the higher rate of systemic embolism (40.0%) in the latter study. When compared to the previous S. aureus report(9), we found the prevalence of S. aureus IE in S. aureus septicemia was 22%, perivalvular complications and cardiac intervention were occurred in 23.0% and 64.2%, respectively. These outcomes were similar to our report. One would argue that in the S. aureus literature, IE was included only as “definite IE” regarding the modified Duke criteria, compared to our reports that definite IE by ESC clinical criteria was only diagnosed in 12 patients (63.2%). However, this is the different case since S .aureus was already classified as a “typical organism” (8, 10), but S. suis was not. This posed a difference in the daily routine clinical practice and vigilantness of primary physicians between these two organisms; therefore, the “IE blood drawing” protocol was far less utilized in S. suis IE. In our point of view, with S. suis septicemia and signs of IE/perivalvular complication by echocardiography would be appropriate enough to diagnose those patients as IE, whether definite or possible, and included them in the present study. Nonetheless, considering the recent update of 2023 Duke-International Society of Cardiovascular Infectious Diseases (ISCVID) criteria(10), all streptococcus spp. (except streptococcus pneumoniae and streptococcus pyogenase) were added as typical organisms and the requirement for timing or separate venipuncture was removed. If this 2023 updated Duke-ISCVID criteria was applied to our report, it will reach definite clinical criteria for IE in 19 patients (100%). However, if we included both pathological and clinical criteria, our cohort would reach a definite IE in 94.7% (18 out of 19 patients) by the ESC criteria and 100% (19 out of 19 patients) by the 2023 updated Duke-ISCVID criteria. Given that, generally, not all suspected IE patients will undergo cardiac surgery and the pathologic criteria will never be sought out. Therefore, the ESC clinical criteria can only detect 63.2% of all IE patients. On the contrary, 2023 updated Duke-ISCVID clinical criteria can detect up to 100% of patients. This information highlighted the need to re-classify S. suis as a “typical” organism.
As for the predictors of S. suis IE in S. suis septicemia, after univariate analyses, we found that dyspnea, immunologic phenomena, new murmurs, and heart failure were all predictors of IE. Nevertheless, heart failure was the only independent predictor for S. suis IE with adjusted OR 35.61 (95%CI 3.24-390.44). This highlighted the group of patients who would benefit most if echocardiography were performed.
Other echocardiographic findings, including LVEDV, LVESV, LVDd, and LVDs, were significantly higher in those with IE. We believe that this points out to subacute nature of S. suis IE, which allowed the left ventricle to remodel, even though there was high valvular destruction. From a treatment perspective, patients with IE had twice the length of stay compared to the non-IE group. This finding may lead to more nosocomial complications, physical and psychological deterioration, and even economic issues for the patients. Fortunately, there were no deaths in our IE group when compared to a 6-month mortality rate of 26% in S. aureus IE(9). This may be due to different antibiotic susceptibilities between organisms, advances in treatment, and the difference in prevalence of prosthetic valve IE.
In summary, the prevalence and aggressive nature of S. suis IE are comparable to those of S. aureus IE(9), but there is still no recommendation for routine TTE surveillance for S. suis IE in patients with S. suis septicemia. We believe that this is due to the scarcity of S. suis IE studies compared to S. aureus, leading to underrecognized the invasiveness of this pathogen. We hope that our study might be one of the studies that may affect the change in clinical practice, and we further encourage reclassifying this pathogen as a “typical” pathogen in all future IE guidelines.
Limitations of the study
There are a few limitations of this study that needed to be mentioned. First, the study design was a retrospective cohort study. This inevitably led to confounding factors that could not be adjusted and might affect the outcomes of the study. Second, the sample size was small due to the low prevalence of S. suis septicemia, even though this is almost a 16-year cohort. Third, to find the true prevalence of S. suis IE in patients with S. suis septicemia, all patients should have echocardiography performed. However, this might not be possible for a single-centered study, as the incidence of S. suis septicemia is only 4 patients per year.