2.1 Patient characteristics
A total number of 62 patients were recruited in our current study, with the baseline clinical characteristics listed in Table 1. Of the 62 recruited patients, 38 were male and 24 were female, with age ranging from 18 to 85 years old. In this cohort, 34 patients had progressive HAP, 28 had controlled HAP, 21 were K.p-positive infected HAP and 10 were complicated with BSI. In addition, we enrolled 10 healthy individuals from the volunteer team as a healthy control group.
Table 1
Subject characteristics
Characteristics
|
n=62
|
Age(y)
≧65
<65
Gender
|
18-85
25
37
|
male
female
Underlying diseases
Intracerebral hemorrhage
Cerebral embolism
Pelvic fracture
Renal failure
Pulmonary contusion
Brain contusion
Clinic status
|
38
24
19
10
6
6
10
11
|
Uncontrolled
Controlled
Pathogens
|
34
28
|
Kp
Non-Kp
|
21
41
|
HAP with BSI
Yes
No
|
10
52
|
2.2 The ratio of Tfh/Th is increased in uncontrolled HAP patients
Firstly, we analyzed if these cell subsets were different in controlled and uncontrolled HAP cases. As shown in Figure 2, Tfh cells were significantly increased in uncontrolled HAP (15.47%±5.49) compared to controlled HAP (11.83%±4.69, P=0.01) and heathy group (10.76%±4.76, P=0.02). Notably, the ratio of Tfh/Th performed a significant increase in progressive HAP (0.19±0.08) compared to disease-controlled HAP (0.13±0.06, P=0.01) and healthy group (0.12±0.05, P=0.02). These data indicated that the ratio of Tfh/Th is increased in progressive HAP patients and maybe useful for rapidly recognizing severity in HAP.
2.3 The ratio of Tfh/Th is elevated in K.p-infected HAP patients
The pathogenic bacteria were detected in 62 patients with HAP, of which twenty-one patients were Klebsiella pneumoniae (K.p)-infected and forty-one patients were non-K.p-infected including Acinetobacter baumannii (n=11), Escherichia coli (n=4), Proteus mirabilis (n=2), Pseudomonas maltophilia (n=6), Pseudomonas aeruginosa (n=8), Pseudomonas cepacia (n=2), Staphylococcus aureus (n=6), Aspergillus(n=2). Next, we further calculated Tfh cells to decide if these cells were relatively increased or decreased in special pathogenic bacteria-infected patients. Interestingly, we found that Tfh cells and the ratio of Tfh/Th were both significantly elevated in K.p-infected (16.20%±4.37, 0.20±0.06) compared to non-K.p-infected patients (12.44%±5.61, 0.15±0.08) with P=0.01 and P=0.02, respectively (Figure 3A). These data suggest that the immune response varied by different bacterial infections, which is related to the Tfh.
2.4 Blood stream infection in HAP induced the raise of the ratio of Tfh/Th
BSI is one of the most frequent lethal conditions that are managed in the ICU. We then investigated these different CD4+ cell subsets in HAP with BSIs and those without BSIs. As shown in Figure 3B, Tfh cells were significantly increased in HAP with BSIs compared with those without BSIs patients (16.20%±4.37 vs 12.98%± 4.82, P=0.01). Also, the ratio of Tfh/Th showed significantly increasing in patients with BSIs than those without BSIs (0.22±0.11 vs 0.15±0.06, P=0.01). These data showed that the immune response caused by bacteria entering the blood will further mediate a more serious Tfh/Th imbalance.
2.5 The correlation between PCT and the ratio of Tfh/Th
PCT, the precursor of calcitonin, as a marker for bacterial pneumonia, is usually low or untraceable in the circulation in healthy people, but can increase significantly in individuals with bacterial infection. PCT 0.25 μg/l is recommended as the threshold to evaluate whether antibiotics are used in patients with respiratory infection [5, 6]. We next determined whether the ratio of Tfh/Th was correlated to PCT in subjects of HAP with PCT>0.25 μg/L. As shown in Figure 3C, Spearman’s correlation test showed that a good correlation was found between the ratio of Tfh/Th with PCT level (r=0.370, P= 0.02).
2.6 The combination of Tfh/Th and PCT shows better value to predicting deteriorated HAP than PCT alone
We next hypothesized whether the ratio of Tfh/Th could predict the progression of HAP. PCT alone had an AUC of 0.703 (95% CI, 0.57-0.83), with the sensitivity of 88.89% and the specificity of 50%, respectively, when the critical value was 0.77. Tfh/Th alone had an AUC of 0.733 (95% CI, 0.61-0.86), with the sensitivity of 74.07% and the specificity of 70.59%, respectively, when the critical value was 0.15. As expected, the combination of Tfh/Th and PCT had an AUC of 0.807 (95% CI, 0.70-0.92), with the sensitivity of 74.07% and the specificity of 82.35% to predict the deterioration of HAP (Figure 4). These data indicated that the addition of Tfh/Th could increase the potential of PCT as a biomarker to predict the deterioration of HAP.
2.7 The ratio of Tfh/Th influenced the survival of HAP patients.
Lastly, we investigated the prognostic value of Tfh/Th in HAP. Given the critical value of Tfh/Th in predicting deteriorated HAP, we divided the individuals into two groups by the value of 0.15. By Kaplan-Meier survival analysis, the 28-day mortality of HAP patients with high (>0.15) and low (≤ 0.15) Tfh/Th ratios was assessed after the test day. As shown in Figure 5, HAP patients with high Tfh/Th ratio had a decreased rate of survival in 28 days, comparing to patients with low Tfh/Th ratio. These data suggest that the ratio of Tfh/Th could be used as a predictor of survival rate in HAP.