Characteristics of study subjects
Initially, 112 septic patients were admitted to the ICU of emergency department. However, five septic patients having a delayed measurement. Five patients were discharged automatically and losing follow-up. One patient with gastrointestinal bleeding was excluded. Thus, 101 septic patients were included in survival analysis. 32 critically ill patients without sepsis were enrolled as control. Flow diagram of selecting study subjects was shown in Figure 1. The demographic and clinical characteristics of patients were shown in Table S1. There were no differences in baseline characteristics between the septic patients and critically ill patients without sepsis on admission (Table S1).
Nosocomial Infection Characteristics
For NIs analysis, 8 septic patients who died within 48 hours after admission were excluded, and 93 patients were enrolled finally (Figure 1). Twenty-nine percent (27/93) of patients developed NIs. Of these, 23 patients had a NI at one site, 5 patients had NIs at two sites. The median time of the first diagnosis of NIs was 8 days [interquartile ranges (IQR):4-15]. Among 33 NIs, pulmonary infection (PI) was the most frequent NI (60.6%), followed by urinary tract infection (UTI) (21.2%), bloodstream infection (BSI) (9.1%) and catheter-related infections (CRI) (9.1%). Of all NIs, 39 microorganisms were isolated. PIs were mostly caused by Acinetobacter baumannii and Stenotrophomonas maltophilia, followed by Pseudomonas aeruginosa, Burkholderia cepacian, and corynebacterium. Candida was the most common pathogen responsible for UTIs and CRIs. Staphylococcus accounted for 66.7% of all isolates from BSIs. Details regarding the frequency of isolated microorganisms are given in Table S2.
The baseline characteristics of patients with and without NI were shown in Table 1. There were no significant differences in age, gender, and co-morbidities between patients that will develop NI or not. Septic patients who developed an NI were more severely ill on admission than those who did not, as indicated by higher APACHE II score and SOFA score. Moreover, these patients were characterized by greater exposure to intubation and central venous catheterization, and longer duration of mechanical ventilation, and urinary tract and central venous catheterization. 30-day mortality was significantly higher in patients with NI (P=0.034) (Table 1).
Table 1
Baseline characteristics of septic patients with and without NIs
| Septic patients | |
Variables | Total (n=93) | Without NI (n=66) | With NI (n=27) | P value |
Age (year) | 65.0(55.5, 70.0) | 65.0(54.0, 69.3) | 65.0(60.0, 70.0) | 0.744 |
Gender (male), n(%) | 50(53.8%) | 34(51.5%) | 16(59.3%) | 0.327 |
BMI, kg/m2 | 22.0(20.0, 25.0) | 22.0(21.0, 26.3) | 21.0(20.0, 24.0) | 0.125 |
Percentage of CD71+cells | 2.9(1.2, 6.4) | 2.5(1.1, 5.7) | 4.0(1.4, 12.6) | 0.113 |
Percentage of CD71+CD235a+cells | 0.9(0.2, 1.9) | 0.7(0.2, 1.3) | 1.2(0.5, 4.1) | 0.021 |
Percentage of CD45+CD71+CD235a+cells | 0.10(0.04,0.19) | 0.08(0.04,0.17) | 0.15(0.07,0.36) | 0.016 |
Circulating cytokines (pg/ml) | | | | |
IL-2 | 0.7(0.5, 1.2) | 0.7(0.5, 1.2) | 0.6(0.5, 1.4) | 0.879 |
IL-4 | 0.4(0.1, 0.6) | 0.4(0.1, 0.5) | 0.3(0.2, 0.7) | 0.549 |
IL-6 | 267.1(72.1, 2971.3) | 181.1(66.4, 2971.3) | 922.9(105.3, 3000.1) | 0.139 |
IL-10 | 44.1(12.2, 398.1) | 43.2(13.2, 407.7) | 50.7(7.6, 313.2) | 0.997 |
TNF-α | 1.8(0.7, 5.6) | 1.8(0.7, 5.6) | 1.5(0.5,5.6) | 0.836 |
IFN-γ | 1.6(0.5, 3.2) | 2(0.6, 3.2) | 1(0.1, 4.8) | 0.381 |
Co-morbidities, n(%) | | | | |
Diabetes | 20(21.5%) | 13(19.7%) | 7(25.9%) | 0.343 |
Chronic heart disease | 9(9.7%) | 5(7.6%) | 4(14.8%) | 0.240 |
Hypertension | 50(53.8%) | 37(56.1%) | 13(48.1%) | 0.320 |
Chronic liver disease | 8(8.6%) | 4(6.1%) | 4(14.8%) | 0.167 |
Chronic kidney disease | 12(12.9%) | 9(13.6%) | 3(11.1%) | 0.519 |
Cancer | 17(18.3%) | 11(16.7%) | 6(22.2%) | 0.361 |
Infection sites, n(%) | | | | 0.262 |
Digestive tract | 30(32.3%) | 19(28.8%) | 11(40.7%) | |
Urinary tract | 25(26.9%) | 21(31.8%) | 4(14.8%) | |
Skin and soft tissue | 15(16.1%) | 9(13.6%) | 6(22.2%) | |
Respiratory tract | 7(7.5%) | 4(6.1%) | 3(11.1%) | |
Others | 16(17.2%) | 13(19.7%) | 3(11.1%) | |
Pathogen isolates, n(%) | | | | 0.036 |
Gram-negative bacteria | 19(20.4%) | 15(22.7%) | 4(14.8%) | |
Gram-positive bacteria | 9(9.7%) | 5(7.6%) | 4(14.8%) | |
Fungus | 2(2.2%) | 1(1.5%) | 1(3.7%) | |
Virus | 1(1.1%) | 0(0%) | 1(3.7%) | |
aOther pathogens | 3(3.2%) | 3(4.5%) | 0(0%) | |
Mixed | 3(3.2%) | 0(0%) | 3(11.1%) | |
No | 56(60.2%) | 42(63.6%) | 14(51.9%) | |
bSOFA score, median (IQR) | 7(4, 9) | 6(4, 9) | 9(7, 12) | 0.002 |
bAPACHE II score | | | | |
Overall, median (IQR) | 13(9, 17) | 12(8, 15) | 17(13, 20) | 0.001 |
<8 | 21(22.6%) | 17(25.8%) | 4(14.8%) | 0.005 |
9-13 | 26(28%) | 23(34.8%) | 3(11.1%) | |
14-17 | 25(26.9%) | 17(25.8%) | 8(29.6%) | |
>17 | 21(22.6%) | 9(13.6%) | 12(44.4%) | |
Hemoglobin (g/L) | 103.0(86.5, 116.0) | 105.0(90.8, 118.0) | 97.0(71.0, 112.0) | 0.069 |
Red blood cells (×10^12/L) | 3.5(2.9, 3.9) | 3.5(3, 3.9) | 3.5(2.6, 3.8) | 0.345 |
White blood cells (×10^9/L) | 13.9(8.2, 20.3) | 14.5(8.3, 19.9) | 12.4(7.6, 21.2) | 0.651 |
Lymphocytes (×10^9/L) | 0.7(0.5, 1.0) | 0.7(0.5, 1) | 0.7(0.5, 1) | 0.343 |
Serum creatinine (µmol/L) | 136.0(83.5, 266.5) | 111.5(76.8, 219.0) | 214.0(102.0, 296.0) | 0.048 |
Interventions, n(%) | | | | |
Intubation | 22(23%) | 6(9.1%) | 16(59.3%) | <0.001 |
Duration of Mechanical ventilation | 0(0, 0) | 0(0, 0) | 3(0, 11) | <0.001 |
Central venous catheterization | 65(69.9%) | 43(65.2%) | 22(81.5%) | 0.093 |
Duration of Central venous catheterization | 4.0(0, 8.0) | 3(0, 6) | 9(3, 15) | <0.001 |
Urinary tract catheterization | 87(93.5%) | 61(92.4%) | 26(96.3%) | 0.436 |
Duration of urinary tract catheterization | 6.0(4.0, 11.5) | 5(3, 7) | 12(7, 26) | <0.001 |
30-day mortality, n(%) | 13(14.0%) | 6(9.1%) | 7(25.9%) | 0.034 |
Abbreviations: NI, nosocomial infection; IQR: interquartile ranges; APACHE, Acute Physiology and Chronic Health Evaluation; SOFA: Sequential Organ Failure Assessment. |
a Other pathogens are rickettsia, leptospira and plasmodium falciparum. |
b Scores were calculated within the first 24 h after the ICU admission using the value associated with the greatest severity of illness. |
CD71+cells and CECs in different groups
Representative flow dot plots of gating strategies of CD71+cells, CD71+CD235a+ cells and CD45+ subset of CD71+CD235a+cells were shown in Figure 2. Their frequency varies greatly among adult critically ill patients; The proportion of CD71+CD235+ cells in PBMC ranged from under detectable level to 38.80%. Compared with critically ill controls, the percentages of CD71+cells, CD71+CD235a+cells as well as CD71+CD235a+CD45+cells in PMBCs were significantly higher in septic patients (all P<0.01) (Figure 3 A-C). Septic patients who developed an NI had significantly higher percentages of CD71+CD235a+cells (P=0.021) and CD71+CD235a+CD45+cells (P=0.016), but not CD71+cells (P=0.113), than those who did no (Figure 3 D-F). The percentage of CD71+CD235a+ cells was higher in non-survivors when compared with survivors [1.3(0.4,6.6) vs 0.7(0.2, 1.3); P=0.037] (Figure 3 G-I).
CD71+cells and CECs for predicting NIs in septic patients
Cox proportional hazard model was used to analyze the relationship between CECs and NIs (Table 2). The results show that the percentage of CD71+cells (HR 1.06; 95% CI 1.02-1.11; P=0.006), CD71+CD235a+cells (HR 1.08; 95% CI 1.02-1.14; P=0.006) and CD71+CD235a+CD45+cells (HR 9.63; 95% CI 2.31-40.27; P =0.002) were associated with increased incidence of NIs after adjusting for BMI, APACHE II score, pathogen isolates, IL-6, hemoglobin, Serum creatinine, intubation, central venous catheterization, and durations of mechanical ventilation, central venous catheterization and urinary tract catheterization. From ROC curve analysis, 2.96 was the best cutoff value of CD71+cells to predict NI development [area under the curve (AUC) 0.605, p = 0.067, sensitivity 66.7%, specificity 57.6%]. The best cutoff values of CD71+CD235a+cells and CD71+CD235a+CD45+cells to predict NI development were 2.08 (AUC 0.653, p=0.021, sensitivity 40.7%, specificity 86.4%) and 0.25 (AUC 0.660, p=0.016, sensitivity 37.0%, specificity 90.9%).
In competing-risk regression analyses, the percentages of CD71+cells (HR 1.06; 95% CI 1.02-1.09; P<0.001), CD71+CD235a+cells (HR 1.08; 95% CI 1.04-1.12; P=0.001) and CD71+CD235a+CD45+cells (HR 6.77; 95% CI 1.39-32.93; P=0.018) were associated with NI development after adjusting for possible confounders (Table 2). According to cumulative incidence curves, for patients with high levels of CD71+cells (>2.96%), CD71+CD235a+cells (>2.08%) and CD71+CD235a+CD45+cells (>0.25%), the probabilities of developing an NI within the next 30 days were 52.8%, 66.6% and 71.3%, whereas for patients with low percentage of them (CD71+cells≤2.96%;CD71+CD235a+cells≤2.08%; CD71+CD235a+CD45+cells≤0.25%), the probabilities were 33.8%, 32.4% and 32.9%, respectively (all P<0.05) (Figure 4 A-C).
CD71+cells and CECs for predicting mortality in septic patients
In Cox regression analysis, the percentages of CD71+CD235a+CD45+cells were not significantly associated with 30-day mortality after adjusting for confounders (HR 2.20; 95% CI 0.51-9.45; P=0.288), while the percentage of CD71+cells (HR 1.05; 95% CI 1.00-1.10; P=0.041) and CD71+CD235a+cells (HR 1.06; 95% CI 1.00-1.12; P=0.031) can predict 30-day mortality outcome (Table 2, Table S3). From ROC curve analysis, 3.48% was the best cutoff value of CD71+CD235a+cells to predict 30-day mortality (AUC 0.640, p=0.046, sensitivity 42.9%, specificity 90.0%), while 6.05% was the best cutoff value of CD71+cells (AUC 0.62, p=0.068, sensitivity 47.6%, specificity 77.5%). In competing-risk regression analysis, the percentage of CD71+cells (HR 1.04; 95% CI 1.01-1.07; P=0.015) and CD71+CD235a+cells (HR 1.06; 95% CI 1.01–1.10; P=0.011) were associated with 30-day mortality after adjusting for possible confounding factors (Table 2). According to cumulative incidence curves, for patients with the percentage of CD71+CD235a+cells more than 3.48%, the probability of mortality in 30 days was 14.3%, whereas for patients with low percentage of the cells (≤3.48%), the probability was 52.9% (P<0.001).
Table 2. Cox regression analysis and competing-risk regression analysis of CD71+cells and CECs for predicting nosocomial infection and mortality in septic patients
Variables
|
Unadjusted model
|
Adjusted model
|
HR
|
CI95
|
P value
|
HR
|
CI95
|
P value
|
a Nosocomial infection
|
|
|
|
|
|
|
Cox regression analysis
|
|
|
|
|
|
|
Percentage of CD71+cells
|
1.05
|
1.01~1.09
|
0.010
|
1.06
|
1.02~1.11
|
0.006
|
Percentage of CD71+CD235a+cells
|
1.09
|
1.05~1.15
|
<0.001
|
1.08
|
1.02~1.14
|
0.006
|
Percentage of CD45+CD71+CD235a+cells
|
6.34
|
2.15~18.69
|
0.002
|
9.63
|
2.31~40.27
|
0.002
|
Competing-risk regression analysis
|
|
|
|
|
|
|
Percentage of CD71+cells
|
1.05
|
1.01~1.09
|
0.011
|
1.06
|
1.02~1.09
|
<0.001
|
Percentage of CD71+CD235a+cells
|
1.09
|
1.06~1.13
|
0.001
|
1.08
|
1.04~1.12
|
0.001
|
Percentage of CD45+CD71+CD235a+cells
|
5.81
|
1.87~18.70
|
0.003
|
6.77
|
1.39~32.93
|
0.018
|
b30-day mortality
|
|
|
|
|
|
|
Cox regression analysis
|
|
|
|
|
|
|
Percentage of CD71+cells
|
1.04
|
1.00~1.08
|
0.035
|
1.05
|
1.00~1.10
|
0.041
|
Percentage of CD71+CD235a+cells
|
1.07
|
1.02~1.11
|
0.002
|
1.06
|
1.00~1.12
|
0.031
|
Percentage of CD45+CD71+CD235a+cells
|
3.14
|
0.93~10.63
|
0.065
|
2.20
|
0.51~9.45
|
0.288
|
Competing-risk regression analysis
|
|
|
|
|
|
|
Percentage of CD71+cells
|
1.04
|
1.01~1.07
|
0.009
|
1.04
|
1.01~1.07
|
0.015
|
Percentage of CD71+CD235a+cells
|
1.07
|
1.05~1.09
|
<0.001
|
1.06
|
1.01~1.10
|
0.011
|
Percentage of CD45+CD71+CD235a+cells
|
2.91
|
1.10~7.74
|
0.032
|
2.06
|
0.90~4.75
|
0.089
|
Abbreviations: CECs, CD71+ erythroid cells; HR, HR hazard ratio; CI, confidence interval.
a Admission variables included in this model were BMI, APACHE II score, SOFA score, pathogen isolates, IL-6, hemoglobin, serum creatinine, intubation, central venous catheterization, and durations of mechanical ventilation, central venous catheterization and urinary tract catheterization.
b Admission variables included in this model were gender, chronic heart disease, APACHE II score, SOFA score, IL-6, hemoglobin, serum creatinine, intubation, central venous catheterization, and durations of mechanical ventilation.
Factors associated with the expansion of CD71+cells and CECs in sepsis
A multiple linear regression was performed to detect independent factors associated with the expansion of CD71+cells and CECs (Table S4). Potential variables that included in the regression analysis were age, gender, co-morbidities, the severity scores (SOFA and APACHE II),WBC, RBC, Hb, Scr and cytokines (IL-2, IL-4, IL-10, TNF-αand IFN-γ). The results showed that the levels of IL-6 and IFN-γwere positively associated with percentage of CD71+cells, CD71+CD235a+cells and CD45+ subset of CD71+CD235a+cells in PBMCs, while IL-10 was negatively associated with them. Additionally, the levels of RBC were negatively associated with the percentage of CD71+CD235a+CD45+cells (Table S4).