Early High-Dose Continuous Veno-Venous Hemoltration Alleviates the Alterations of Cd4+ T Lymphocyte Subsets in Septic Patients Combined With Acute Kidney Injury

Background This study aims to observe the changes in CD4 + T lymphocyte subsets in septic patients combined with acute kidney injury and the effect of early high-dose CVVH on CD4 + T lymphocyte subsets. Methods Enrolled septic patients combined with acute kidney injury were randomly divided into CVVH (n = 50) and conventional treatment (non-CVVH, n = 53) groups. Healthy volunteers (n = 21) were enrolled. CVVH was initiated within 12 h of intensive care unit (ICU) admission with the doses of 35 ~ 60 mL/kg/h and maintained for at least 72 h. Th1, Th2, Th17 and T reg were measured by ow cytometry on days 1, 3 and 7 of ICU admission. Sequential organ failure assessment (SOFA) scores were calculated.


Abstract Background
This study aims to observe the changes in CD4 + T lymphocyte subsets in septic patients combined with acute kidney injury and the effect of early high-dose CVVH on CD4 + T lymphocyte subsets.

Methods
Enrolled septic patients combined with acute kidney injury were randomly divided into CVVH (n = 50) and conventional treatment (non-CVVH, n = 53) groups. Healthy volunteers (n = 21) were enrolled. CVVH was initiated within 12 h of intensive care unit (ICU) admission with the doses of 35 ~ 60 mL/kg/h and maintained for at least 72 h. Th1, Th2, Th17 and T reg were measured by ow cytometry on days 1, 3 and 7 of ICU admission. Sequential organ failure assessment (SOFA) scores were calculated.

Results
Th1 percentages and Th1/Th2 ratios were lower, and Th2, Th17 and T reg percentages and Th17/T reg ratios were higher in patients with sepsis, when compared to healthy volunteers (all P < 0.05). Early highdose CVVH treatment signi cantly increased Th1 percentages and Th1/Th2 ratios, and signi cantly decreased Th2, Th17 and T reg percentages and Th17/T reg ratios, when compared to non-CVVH treatment (all P < 0.05). Th1 percentages and Th1/Th2 ratios were negatively correlated with SOFA scores, while Th2, Th17 and T reg percentages and Th17/T reg ratios were positively correlated with SOFA scores (all P < 0.05). Patients with sepsis in the CVVH group had signi cantly lower SOFA scores on day 7 of ICU admission and a shorter ICU stay when compared to those in the non-CVVH group (P < 0.05). However, the difference in 28-day mortality was not signi cant between the CVVH and non-CVVH groups (22.0% vs. 37.7%, P = 0.104).

Conclusion
Septic patients combined with acute kidney injury exhibit the different alterations of CD4 + T lymphocyte subsets. Early high-dose CVVH treatment could alleviate the alterations, which may be one of factors associated with the improvement of sepsis severity, but do not signi cantly decrease the 28-day mortality.
Background Sepsis is de ned as a life-threatening organ dysfunction caused by a dysregulated host response to infection [1]. It is one of the leading causes of death in the intensive care unit (ICU). In spite of many therapeutic principles recommended by the international guidelines for the management of sepsis and septic shock, sepsis remains as a major health problem worldwide, with an ICU mortality rate of 25.8% in 84 countries in 2012 [2,3].
In sepsis, the immune response initiated by an invading pathogen fails to return to homeostasis, which culminates in a pathological syndrome characterized by sustained excessive in ammation and immune suppression [4]. Sepsis has been considered as a result of a complex network of events that involve immune-in ammatory and anti-in ammatory processes, which contribute to the clearance of infection and tissue recovery, or organ injury [4,5]. Recent studies have indicated that immune dysregulation (immune excess or suppression) might play a key role in the development of sepsis, and this dysregulation is closely associated with CD4 + T lymphocytes during both the early and late stage of sepsis [5]. Upon different stimulation in the microenvironment, naïve CD4 + T lymphocytes can be induced to differentiate into speci c subtypes, such as T helper (Th) 1, Th2, Th17 and regulatory T (T reg ) cells.
The effector functions of these cells are mediated by cytokines secreted by differentiated cells [6]. Th1 cells mainly secrete IFN-γ, TNF-β and IL-2, supporting cell-mediated immunity [6]. Th2 cells secrete the key effector-cytokines, such as IL-4, IL-5 and IL-13, and Th2 cells, mounting immune response to extracellular parasites and playing major roles in the induction and persistence of certain allergic diseases [6]. Th17 cells secrete key effector cytokines, including IL-17A, IL-17F, IL-21 and IL-22, and have a crucial role on the in ammatory response against parasites, and extracellular and fungal pathogens [7]. T reg cells balance T lymphocyte homeostasis, activation and function via numerous different mechanisms, including the secretion of IL-10, TGF-β and IL-35 [4].
Clearly, CD4 + T lymphocytes represent a unique branch of the adaptive immune system, which is crucial for achieving a regulated effective immune response to pathogens, and their proper functioning is vital for the survival of patients infected with invading pathogens [6]. Higher Th1 and Th17 differentiation have been found to be associated with higher survival rate in severe sepsis [8]. However, proin ammatory Th1 and Th17 cells were found to be suppressed during human experimental endotoxemia [9], and the inversed proportion of Th17/T reg may be associated with the aggravation of patients with sepsis [10]. Therefore, the adjustment of CD4 + T subtype may be one of promising options for the treatment of sepsis.
The direct removal of in ammatory mediators from circulation appears to be intuitively attractive, since this could e ciently interrupt the in ammatory cascade, affect the differentiation of CD4 + T subtypes, and attenuate in ammation. Continuous blood puri cation is a pivotal treatment strategy for renal failure in the intensive care setting [3,[11][12][13]. However, recent studies revealed that in addition to the removal of waste products, other molecules, such as in ammatory mediators/cytokines, were also the targets of removal in continuous blood puri cation, especially high-dose (namely high-volume) hemo ltration, for patients with sepsis [11,12,[14][15][16][17][18][19][20][21]. Continuous blood puri cation has been generally applied for the treatment of sepsis as a strategy for the continuous removal of in ammatory mediators [3,13]. However, to date, the effect of continuous venous-venous hemo ltration (CVVH), as a commonly used continuous blood puri cation mode, on the alterations of CD4 + T lymphocyte subsets in patients with sepsis, remains unclear. The present prospective study observed the changes in CD4 + T lymphocyte subsets in septic patients combined with acute kidney injury, and the effect of early high-dose CVVH on CD4 + T lymphocyte subsets alterations, disease severity and mortality.

Study design and population
The present study was conducted in accordance with the Declaration of Helsinki (2013 edition) adopted by the World Medical Association [22]. The study protocol was approved by the Ethics Committee of the First A liated Hospital of Dalian Medical University (PJ-KY-2018-12), Dalian, China. A written informed consent was obtained from all participants or legal guardians.
A total of 135 consecutive septic patients combined with acute kidney injury, who were admitted to the emergency ICU of our hospital from September 2016 to December 2018 and had the indication of CVVH treatment, were selected as subjects. The enrolled patients with sepsis were randomly divided into two groups: CVVH group, patients who received CVVH treatment; conventional treatment (non-CVVH) group. A total of 21 healthy volunteers were enrolled.

Inclusion and exclusion criteria
Patients with sepsis or septic shock were included in the present study if they met the requirement of the third international consensus de nitions for sepsis and septic shock (Sepsis-3), accompanied by sepsisinduced acute kidney injury (diagnosed according to the KDIGO clinical practice guidelines) with increase in serum creatinine of more than 300 μmol/L, oliguria (<100 mL/6 h) unresponsive to uid resuscitation, severe acidemia of less than pH of 7.2, hyperkalemia of more than 6.5 mmol/L, or clinically signi cant organ edema (e.g., pulmonary edema) [1,11,[23][24][25].
Patients were excluded from the present study based on the following criteria: patients < 18 years old; patients with immunode ciency virus infection, neutropenia (de ned as an absolute neutrophil count of < 1,000 neutrophils/μL), acute bleeding, malignant tumor, pregnancy, or chronic kidney disease; patients who used corticosteroids or nephrotoxic drugs; patients who received any kind of renal replacement therapy prior to admission to the ICU; patients who refused to take part in the present study.

Treatments of the patients
The conventional therapy was based on standard sepsis treatments and according to Surviving Sepsis Campaign guidelines, which included intensive monitoring, uid resuscitation, oxygen administration or mechanical ventilation, antimicrobial therapy (blood culture was conducted before the administration of antibiotics), vasopressor administration, glucose control, diuretics for oliguria with uid overload, etc [3].
For the CVVH treatment, a 11.5Fr double lumen hemofiltration catheter (Lily Technology, Guangdong, China) was percutaneously inserted into the femoral vein. Hemofiltration was performed using the PrimaFlex apparatus (ALPRI, France) equipped with an acrylonitrile and sodium methallyl sulfonate copolymer + polyethylene (AN69ST) hollow-fiber high-flux hemofilter (1.5 m 2 , Braun Diapact, Germany), and was delivered in CVVH mode. The extracorporeal circuit and hemofilter were pre-ushed with heparinized normal saline (12500 U of unfractionated heparin added in 1000 mL of normal saline) to prevent clotting. After circulation, the heparinized normal saline was ushed out of the hemo lter and circuit tubes prior to starting CVVH. Continuous intravenous pumping of unfractionated heparin (4 to 8 U/kg/h) was used to adjust the systemic activated partial thromboplastin time (APTT) between 35 and 45 s [26]. However, heparin-free CVVH was implemented if a patient had a bleeding tendency (e.g. low platelet count, prolonged activated partial thromboplastin time). CVVH was initiated within 12 h of ICU admission and maintained for at least 72 h with the doses of 35 ~ 60 mL/kg/h, and a blood ow rate of 150 ~ 200 mL/min. CVVH were stopped on basis of renal recovery and physician decision [24] The hemofilter was used till 72 h, or immediately changed if it did not work effectively because of clotting.
The bicarbonate replacement solutions were commercially prepared by Qingshan Likang, Pharmaceutical Co. (Chengdu, China), and was pre-hemofilter diluted.

Variables
The following variables were collected for each patient on admission: age, sex, infection sites, results of blood culture, laboratory data (plasma creatinine, lactate, procalcitonin, leukocyte count, hemoglobin, platelet count and C-reactive protein), mean arterial pressure, number of patients using vasopressors, time from sepsis onset to ICU admission and number of patients using mechanical ventilation. On days 1, 3 and 7 of ICU admission, CD4 + T lymphocyte subsets (Th1, Th2, Th17 and T reg ) were measured, and Th1/Th2 and Th17/T reg ratios were calculated; meantime, sequential organ failure assessment (SOFA) scores were evaluated.

Flow cytometry analysis
Two mL of peripheral venous blood was taken into EDTA-K3 coated tubes. The blood sample was rst stimulated for 5 h with a cell stimulation cocktail (Sigma, St. Louis, MO, USA), containing 50 ng/mL of protein transport inhibitors, 1 mg/mL of ionomycin and 1.7 mg/mL of monensin, according to manufacturer's instructions.

Statistical analysis
Data were analyzed using SPSS 22.0 (IBM, Armonk, NY, USA). Data were expressed as mean ± standard deviation (SD) for normal distribution, or median and interquartile range for skewed distribution.
Pearson's chi-squared or Fisher's exact test was used, as appropriate, to compare the categorical variables. A Mann-Whitney U test was used to compare the variables between the CVVH and non-CVVH groups. Variables among time points and groups were compared using repeated-measures analysis of variance (ANOVA), followed by Bonferroni test for multiple comparisons. The differences were considered statistically signi cant when P < 0.05.

Results
Baseline characteristics of the enrolled subjects A total of 135 eligible patients with sepsis or septic shock met the inclusion criteria (Fig.2). There were no signi cant differences between the CVVH (n = 50) and non-CVVH groups (n = 53) with regard to age, sex, infection sites, positive blood cultures, plasma creatinine, lactate and procalcitonin, leukocyte count, hemoglobin, platelet count, C-reactive protein, mean arterial pressure, the number of patients using vasopressors, time from sepsis onset to ICU admission and the number of patients receiving mechanical ventilation (all P > 0.05; Table 1). CVVH started at 5.5 (2.0, 8.5) hours after ICU admission. The parameters of CVVH were provided in Table 2.
The changes of T lymphocyte subsets in patients with sepsis on ICU admission Patients with sepsis (both non-CVVH and CVVH groups) showed signi cantly lower Th1 percentages and Th1/Th2 ratios, and signi cantly higher Th2, Th17 and T reg percentages and Th17/T reg ratios on admission, when compared to healthy volunteers (all P < 0.05; Fig.3). However, the differences in the above variables on admission were not signi cant between the non-CVVH and CVVH groups (all P > 0.05).
The effect of high-dose CVVH treatment on Th1, Th2 and the Th1/Th2 ratios in patients with sepsis After the one-week conventional therapy in the non-CVVH group, the low Th1 percentages and Th1/Th2 ratios were gradually improved, while the high Th2 percentages were gradually decreased (all P < 0.05; Fig. 3A, 3B and 3C). On day 7 of ICU admission, high-dose CVVH treatment exhibited more signi cant improvement in Th1 percentages and Th1/Th2 ratios, and more signi cant decrease in Th2 percentages compared with non-CVVH treatment (all P < 0.05). However, on day 3 of ICU admission, the differences in the above variables were not signi cant between the non-CVVH and CVVH groups (all P > 0.05).

The effect of high-dose CVVH treatment on Th17, T reg and Th17/T reg ratios in patients with sepsis
After the one-week conventional therapy in the non-CVVH group, the increased Th17 and T reg lymphocyte percentages and Th17/T reg ratios could be e ciently reversed in patients with sepsis. On day 7 of ICU admission, high-dose CVVH treatment exhibited a more e cient effect on the reversion of these T lymphocyte subsets when compared to non-CVVH treatment (all P < 0.05; Fig. 3D, 3E and 3F). However, on day 3 of ICU admission, the differences in the above variables were not signi cant between the non-CVVH and CVVH groups (all P > 0.05).

The correlation between CD4 + T lymphocyte subsets and SOFA scores in patients with sepsis
The Spearman's correlation analysis indicated that the Th1 percentages and Th1/Th2 ratios were negatively correlated with the SOFA scores, while the Th2, Th17 and T reg percentages, and Th17/T reg ratios were positively correlated with the SOFA scores (all P < 0.05; Table 3).

The effect of high-dose CVVH on the severity and outcome of patients with sepsis
Patients with sepsis in the CVVH group had signi cantly lower SOFA scores on day 7 of ICU admission, and a shorter ICU stay, when compared to those in the non-CVVH group (both P < 0.05; Table 4). Patients with sepsis in the CVVH group had a tendency to a lower 28-day mortality when compared to those in the non-CVVH group (22.0% vs. 37.7%; Table 4), but the difference was not signi cant (P = 0.104).

Discussion
In the present study, we observed that septic patients combined with acute kidney injury exhibited higher Th2 percentages, lower Th1 percentages and Th1/Th2 ratios in the rst week of ICU admission, when compared to healthy volunteers. This observation is similar to previous works, showing a shift from Th1 to Th2 cytokine pro les following infection [27,28]. Higher Th17, T reg and Th17/T reg ratios were also observed in patients with sepsis in the present study when compared to healthy volunteers, which is consistent with the reports of other studies [29,30]. These alterations of CD4 + T lymphocyte subsets indicate that T reg and Th17 cells are different differentiations from Th1 and Th2 cells [28,31,32].
Although the underlying mechanisms of CD4 + T lymphocyte subset differentiation remain under investigation, it was speculated that these different differentiations may play crucial roles in disease progression in patients with sepsis [6, 28].
In addition, we observed that Th1 percentages and Th1/Th2 ratios were negatively correlated with the SOFA scores, while Th2, Th17, T reg percentages and Th17/T reg ratios were positively correlated with the SOFA scores. These observations further indicated that the differentiation and different alterations of CD4 + T lymphocyte subsets may be associated with the severity of sepsis, which induced an immunoparalysis state [6,28,33]. T lymphocyte subsets, especially CD4 + T helper cells, have been demonstrated to contribute to the advancement of in ammatory diseases and microbial challenges [4, 6, 29-31, 34, 35]. The lack of the shift from Th1 to Th2 has been found to increase the survival among patients with sepsis [4]. Inhibiting a pro-in ammatory Th1 to an anti-in ammatory Th2 shift and attenuating the T reg population can improve survival in secondary pneumonia infections [28]. In the present study, it was found that after one week of ICU admission, the proportions of T lymphocyte subsets in patients were getting close to those in healthy volunteers, which indicates the reversion of the CD4 + T lymphocyte subsets after conventional therapy.
Surprisingly, the early high-dose CVVH treatment exhibited a more e cient effect on this T lymphocyte subset reversion when compared to non-CVVH treatment, which further illustrates the correlation between CVVH treatment and the signi cantly decreased SOFA scores of patients. The mechanisms underlying the more e cient reversion of early high-dose CVVH treatment to the alterations of CD4 + T lymphocyte subsets have not yet been completely clari ed. We speculated that this effect of early high-dose CVVH may be associated with the removal of high-dose CVVH to in ammatory mediators/cytokines, endotoxin, antigens and other solutes, which affected the differentiation, apoptosis and functions of CD4 + T lymphocyte subsets [5,12,14,15,17,27,33,36]. AN69ST, utilized in the present study, is an effective cytokine-adsorbing hemo lter, and can also remove endotoxin [14,37]. Ionic interactions have been demonstrated to play a role in cytokines (e.g. TNF-α, IL-6, and IL-8) adsorption by the AN69ST membrane [37]. Previous studies have shown that high-dose CVVH effectively removed serum in ammatory cytokines also via the mechanism of convection [19,21]. Furthermore, cutting the peaks of soluble mediators through continuous hemo ltration, namely, a concept for the peak concentration hypothesis, might partly explain for the more e cient reversion of CVVH treatment to the alterations of CD4 + T lymphocyte subsets [17]. Thus, CVVH has been considered as an e cient immunomodulatory therapy for patients with sepsis [36]; the effect of the non-speci c removal of unbound mediators and cytokines from the blood compartment is to reset to a lower level of immune dysregulation, both in the prevalently proin ammatory and counter-in ammatory phases of sepsis [17].
It was not surprising that the 28-day mortality of patients with sepsis failed to be signi cantly decreased by high-dose CVVH treatment in spite of the signi cantly decreased SOFA scores, which is inconsistent with the meta-analysis published by Putzu et al., stating that low-quality evidence indicates the correlation between CVVH and the signi cant reduction in mortality in patients with severe sepsis [3,13]. There are several potential explanations for this phenomenon. First, the single-center design and relatively small sample size in this study may not be adequate to reach a statistically signi cant reduction of mortality after CVVH treatment. Hence, a multi-center randomized trial with a large sample size would be needed in future trials. Second, the differentiations and functions of T lymphocyte subsets are complicate and dynamic, which might partly contribute to the severity of patients with sepsis and be insu cient to affect the outcome. Finally, the potential adverse effects of hemo lter during CVVH, such as being an in ammatory stimulus itself and activating leukocytes during hemo lter passage, excessive loss of electro lytes, micronutrients, vitamins or drugs (e.g., antibiotics), maybe weaken the bene cial effects [18,38]. In fact, to date the effects of CVVH treatment on mortality of patients with sepsis remains controversial. A recent meta-analysis also showed that high-volume hemo ltration did not change the mortality of patients with sepsis [16].
The present study had six main limitations. First, we failed to detect the levels of cytokines in plasma and ltered solution such as IL-4, IL-6, IL-1ra, IL-8, IL-10, IL-12 and tumor necrosis factor-α that may affect the differentiation and functions of these CD4 + T lymphocyte subsets. Second, CVVH is not a standardized technique. Hence, the settings (e.g. CVVH doses) and system used in the present study could result in different results from other studies [13,39]. Third, the differences in subgroups were not assessed on the basis of sex and septic shock. However, the sex distribution and the number of patients using vasopressors on admission in the CVVH and non-CVVH groups was even. Fourth, although 35 ~ 60 mL/kg/h (> 35 mL/kg/h generally accepted as the de nition of high dose in CVVH therapy) of CVVH doses used in the present study is not enough high, several meta-analyses showed that higher doses had no clinical advantage for blood puri cation purposes in the treatment of patients with sepsis [11,16,20,40]. Finally, the association between CVVH and lower SOFA scores might be biased by the fact that hemo ltration eliminates creatinine and bilirubin, which itself are parameters of the SOFA scores.

Conclusions
Overall, the present study indicates that septic patients combined with acute kidney injury exhibit the different alterations of CD4 + T lymphocyte subsets in the rst week of ICU admission. Early high-dose CVVH treatment could signi cantly alleviate the alterations of CD4 + T lymphocyte subsets, which may be one of factors associated with the improvement of sepsis severity, but do not signi cantly decrease the 28-day mortality.

Declarations
Ethics approval and consent to participate The present study was conducted in accordance with the Declaration of Helsinki (2013 edition) adopted by the World Medical Association. The study protocol was approved by the Ethics Committee of the First A liated Hospital of Dalian Medical University (PJ-KY-2018-12), Dalian, China. A written informed consent was obtained from all participants or legal guardians.

Consent for publication
All the authors agree for publication.

Availability of data and materials
The datasets generated and analyzed during this study are available from the corresponding author upon reasonable request.   ICU, intensive unit; NS, no signi cance; SOFA, sequential organ failure assessment. Figure 1 The ow cytometry of CD4+ T lymphocyte subsets from healthy volunteers and patients with sepsis. Th1 cells were identi ed with positive CD3 and IFN-γ, and negative CD8. Th2 cells were identi ed with positive Page 17/18 CD3 and IL-4, and negative CD8. Th17 cells were identi ed with positive CD3, CD4 and IL-17. Regulatory T cells were identi ed with positive CD4 and high CD25, and Foxp3.