A total of 122 patients was enrolled in the preliminary screening list, that included patients with initial and secondary infections by Enterococcus spp., and 38, 12, and six patients were excluded due to other bacterial co-infections, immunocompromised status, and receiving corticosteroid treatment during ICU admission, respectively. Finally, the study cohort included 66 patients, and these were divided into two groups: infected only with Enterococcus spp. (n=31; single-pathogen BSI group) and infected simultaneously with Enterococcus spp. and Candida spp. (n=35; mixed-pathogen BSI group) (Fig. 1).
Demographic characteristics of included patients
Table 1 summarizes the demographic characteristics of included patients. Single- and mixed-pathogen BSI groups had 31 (47.0%; 19 men and 12 women; mean age 77 years) and 35 (53.0%; men, 11 women; mean age 78 years) patients, respectively. Major diseases in patients of these two groups were cardiovascular, digestive, urinary system and neurological diseases, and diabetes mellitus. No significant differences were observed in age, sex, and main diseases between the two groups (all P>0.05).
Distribution of pathogens
Fig. 2 shows the distribution of pathogens in the included patients. In the single-pathogen BSI group, 27 and four patients were infected with E. faecalis and E. faecium,respectively, while in the mixed-pathogen BSI group, 29 and 6 patients were infected with E. faecalisand E. faecium, respectively. Among these, 22, 4, and 3 patients were co-infected with E. faecalis and C. albicans, E. faecalis and C. tropicalis,and E. faecalis and C. glabrata,respectively. The remaining 6 patients were all co-infected with E. faecium and C. albicans. No significant differences were observed in the distribution of Enterococcus spp. between the two groups.
Disease severity
Disease severity was defined by APACHE II scores, ICU stay duration, and 30-day mortality. APACHE II scores of all the patients were determined within 24 h of ICU admission. The mixed-pathogen BSI group demonstrated a significantly higher APACHE II scores than that of the single-pathogen BSI group (22.6 [3.7] vs. 20.3 [4.5], respectively; P < 0.05) (Table 2). Further, the mixed-pathogen BSI group also showed a significantly longer ICU stay duration than that of the single-pathogen BSI group (24 [7.2] vs. 19 [6.8], respectively; P < 0.01). The 30-day mortality in the mixed-pathogen BSI group (42.86%) was also higher than that of the single-pathogen BSI group (19.35%). Survival curves showed that in the mixed-pathogen BSI group, excessive mortality started on the 20th day of ICU admission (Fig. 3). Hence, mixed-pathogen BSIs seemed to be associated with a worse outcome, and the log-rank test demonstrated a significant difference in 30-day percent survival between the two groups (P = 0.0355).
Inflammatory indicators
Fig. 4 shows the comparisons of WBC counts, NEU%, and plasma PCT and CRP levels between the two study groups. The plasma CRP levels in the mixed-pathogen BSI group were significantly higher than that of the single-pathogen BSI group (P < 0.01). In contrast, no significant differences were observed in other marker levels (WBC counts, NEU%, and plasma PCT levels) between the two groups (all P>0.05).
Correlation between inflammatory markers and disease severity
Table 3 shows the correlations between inflammatory markers and disease severity in both study groups. In the single-pathogen BSI group, WBC counts and NEU% showed no significant correlation with APACHE II scores, ICU stay duration, and 30-day mortality. In contrast, plasma PCT levels showed a positive correlation with ICU stay duration (odds ratio [OR]=0.4051, 95% confidence interval [CI] 0.0438-0.6702, P = 0.0238), whereas plasma CRP levels were positively correlated with APACHE II scores (OR=0.4325, 95% CI 0.0814-0.6881, P = 0.0151) and ICU stay duration (OR=0.4362, 95% CI 0.0860-0.6905, P = 0.0142). However, PCT and CRP levels in the plasma were not associated with 30-day mortality. In the mixed-pathogen BSI group also, WBC counts and NEU% showed no significant correlation with APACHE II scores, ICU stay duration, and 30-day mortality. However, both PCT and CRP levels in plasma were positive correlated with APACHE II scores ([OR=0.3364, 95% CI -0.0067-0.6086, P = 0.0482] and [OR=0.4606, 95% CI 0.1404-0.6935, P = 0.0054] respectively) and ICU stay duration ([OR=0.3516, 95% CI 0.0105-0.6194, P = 0.0384] and [OR=0.5306, 95% CI 0.2300-0.7387, P = 0.0010] respectively). Further, only the plasma CRP levels showed positive correlation with 30-day mortality (OR=0.3836, 95% CI 0.0475-0.6417, P = 0.0229) in this group.