Continuous Veno-Venous Hemoltration (CVVH) Improves Iron Metabolism Disorders in Patients with Sepsis(cid:0)A Single-Center Prospective Cohort Study

Background: Iron metabolism disorder is commonly seen in patients with sepsis. This study aimed to evaluate whether continuous veno-venous hemoltration (CVVH) improved the iron metabolism disorders in sepsis. Methods: In a single-center, retrospective cohort study, totally 89 sepsis patients were prospectively enrolled and divided into the CVVH group (n=39) and the control group (n=50). Clinical and laboratory data were collected and compared between the groups on days 1, 3 and 7 of ICU admission. Plasma interleukin (IL)-6, hepcidin, erythropoietin (EPO), ferritin and soluble transferrin receptor (sTfR) were determined by enzyme linked immunosorbent assay (ELISA). Sequential organ failure scores (SOFA) on days 1 and 7, and 28-day survival between groups were compared. Results: Plasma IL-6, hepcidin, ferritin and RDW on days 3 and 7 were signicantly reduced in the CVVH group compared with those in the control group (all P<0.05). The CVVH group had a signicantly lower SOFA score on day 7 compared with the control group (P<0.05). Hemoglobin and EPO were gradually decreased within the rst week of ICU admission in both groups although no signicant differences between the groups were observed. There was no signicant difference in sTfR between the two groups along with the time (all P > 0.05). In addition, there were no signicant differences in 28-day survival rate and median survival time between the two groups. Conclusions: CVVH improves iron metabolism disorders and the disease severity in sepsis. However, it does not alleviate anemia and fails to improve the survival.


Background
Sepsis with an in-hospital mortality rate of 25.8% has so far been one of the leading causes of death in critically ill patients [1]. Anemia is commonly observed in patients with sepsis, which is mostly due to the involvement of in ammatory mediators [2,3]. The release of in ammatory mediators were associated with elevated plasma hepcidin, leading to iron metabolism disorders and subsequent anemia in severe sepsis [3]. Interleukin (IL)-6 is a key in ammatory cytokine which induces an abruptly increased synthesis of hepcidin and decreases plasma iron in patients with sepsis [4]. The survival of sepsis was closely related with the indicators re ecting iron metabolism disorders and in ammation-associated anemia. We previously observed that plasma iron, hemoglobin and soluble transferrin receptor (sTfR)/log ferritin were signi cantly decreased and the plasma erythropoietin (EPO), sTfR, hepcidin, ferritin and IL-6 were signi cantly increased in patients with sepsis within the rst week on intensive care unit (ICU) admission.
No effective therapies have been developed so far for the treatment of severe in ammation-associated anemia and related iron metabolism disorders in sepsis. Conventional blood transfusion and iron supplement are not routinely recommended in the current clinical practice. Blood transfusion was reported to be associated with higher mortality and also failed in anemia correction in critically ill patients [5,6]. Although enteral iron supplementation (ferrous sulfate 325 mg three times daily) reduced transfusion rate in critically ill patients with baseline iron de ciency, iron is a toxic compound which may induce oxidative stress and bacterial growth [7,8]. In addition, the optimal route of iron administration still need to be determined [7].
Continuous venous-venous hemo ltration (CVVH) as a commonly used mode of continuous blood puri cation is a pivotal treatment strategy for renal failure [9][10][11][12][13][14]. CVVH can also non-selectively remove endotoxin and in ammatory mediators, rebalance acid-base disorders, regulate immune stability, and maintain stable environment through the mechanisms of diffusion, convection and adsorption [15]. It is therefore applicable for eliminating in ammatory mediators in sepsis [16]. However, the impact of CVVH on iron metabolism disorders in patients with sepsis remains unclear. This prospective study aimed to determine the effect of CVVH on iron metabolism disorders in patients with sepsis.

Ethics statement
This study was performed in accordance with the Declaration of Helsinki adopted by the World Medical Association (2013 edition) [17]. The study protocol was approved by the Medical Ethics Committee of the First A liated Hospital of Dalian Medical University (PJ-KS-KY-2020-138), Dalian, China. A written informed consent was obtained from all participants or legal guardians at the time of initial admission.

Patients
Patients diagnosed with sepsis and admitted to the ICU at the First A liated Hospital of Dalian Medical University from October 1, 2015 to December 30, 2017, were prospectively collected. All included patients met the diagnostic criteria of the third international consensus de nitions for sepsis and septic shock (Sepsis-3), accompanied by sepsis-induced acute kidney injury with increase in creatinine or oliguria with uid overload, and hyperkalemia [10][11][12][18][19][20]. The exclusion criteria were as follows: less than 18 years old; pregnancy; with stomach cancer and acute bleeding; with blood system diseases [e.g., leukemia, myeloma and red blood cell diseases, including polycythemia vera, abnormal mean red blood cell volume (MCV < 80.8 or >99.2 μm 3 ), abnormal mean hemoglobin content (MCH < 26.4 or > 34.2 pg) or abnormal mean hemoglobin concentration (MCHC < 32.0 or > 35.4 g/dL)]; with acute or chronic renal insu ciency or renal failure requiring hemodialysis; with liver cirrhosis; prior blood transfusion within 1 week before admission or during hospitalization; with anemia treated with iron and erythropoietin within 3 months before admission.
The enrolled patients with sepsis were divided into the CVVH group that received CVVH treatment or the control group that did not receive CVVH treatment due to refusal either from the patient or from legal guardians because of nancial di culty. A written informed consent was obtained from all participant's legal guardians at the time of initial admission.

Treatment
The control group received routine treatment according to the Surviving Sepsis Campaign guidelines, including intensive monitoring, uid resuscitation, oxygen administration or mechanical ventilation, antimicrobial therapy (blood culture was conducted before the administration of antibiotics), vasopressor if necessary, glucose control, diuretics for oliguria with uid overload, etc [21].
Patients in the CVVH group underwent CVVH (Golden Prisma Flex, Sweden) within 24 h after diagnosis in addition to the routine treatment. A 11.5Fr double lumen hemofiltration catheter (Lily Technology, Guangdong, China) was percutaneously inserted into the femoral vein or the internal jugular vein. Hemofiltration was performed using the PrimaFlex apparatus (ALPRI, France) equipped with an acrylonitrile and sodium methallyl sulfonate copolymer hollow-fiber high-flux hemofilter (1.5 m 2 , Braun Diapact, Germany), and was delivered in CVVH mode. CVVH was performed at a dose of 35 mL/kg/h for 72 h [22], and thereafter for 12 h per day, depending on the condition of the patient. The average blood flow rates were 150-200 mL/min and the dehydration volume was 300-380 mL/h. The hemofilter was changed every 48 h. The bicarbonate replacement solutions were commercially prepared by Qingshan Likang, Pharmaceutical Co. (Chengdu, China), and was pre-hemofilter diluted. Unfractionated heparin as an anticoagulant or no heparinization was used. The dosage of the unfractionated heparin was adjusted according to the patient's blood coagulation state.

Data collection
The following clinical data were collected upon admission: gender, age, sites of infection, comorbidities, numbers of post-operative patients, patients using vasopressors and patients receiving mechanical ventilation, laboratory results, and survival time within 28 days. Hemoglobin, red blood cell distribution width (RDW), reticulocytes, mean corpuscular volume, mean corpuscular hemoglobin, and mean corpuscular hemoglobin concentration were measured by an automatic blood cell analyzer (XN-2000; Sysmex, Kobe, Japan). Sequential organ failure assessment (SOFA) scores were assessed on ICU admission based on age, medical history, vital signs, and laboratory results.

Statistical analyses
Statistical analyses were performed using SPSS 23.0 (IBM, Armonk, NY, USA). Continuous variables are expressed as mean ± standard deviation (SD) for normal distribution, or median and range for skewed distribution. Categorical variables were compared using Pearson's chi-squared or Fisher's exact test. Continuous variables were compared by Mann-Whitney U test or t test. Cumulative survival curves were depicted using the Kaplan-Meier method and compared by log-rank test. P < 0.05 was considered to be statistically signi cant.

Baseline characteristics
Totally 127 patients with sepsis were initially enrolled and 89 patients were nally included in the study ( Figure 1). The CVVH group included 39 patients while 50 patients were included in the control group. No signi cant differences of baseline characteristics including age, male, sites of infection, comorbidities, the numbers of postoperative patients, patients using vasopressors and patients receiving mechanical ventilation, PaO 2 /FiO 2 , procalcitonin, lactate, creatinine, reticulocytes, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, and SOFA scores were detected between the groups ( Table 1). The parameters of CVVH were summarized in Table 2.
The effect of CVVH on plasma IL-6 On admission, plasma IL-6 level did not differ between the two groups (P > 0.05, Figure 2). Signi cantly decreased IL-6 level from day 1 to days 3 and 7 was observed in both groups (all P < 0.05). However, plasma IL-6 on days 3 and 7 in the CVVH group were signi cantly lower than that in the control group (both P < 0.05).
The effect of CVVH on hemoglobin, plasma EPO and RDW There were no signi cant differences in hemoglobin, plasma EPO and RDW between groups upon admission (all P > 0.05). However, both groups presented with gradually decreased hemoglobin and EPO within the rst week of ICU admission without signi cant differences between the groups (all P < 0.05, Figures 3A and 3C). RDW on days 3 and 7 in both groups were higher than that on day 1 without signi cant difference between days 3 and 7 (both P < 0.05, Figure 3 B). However, the CVVH group had a signi cantly lower RDW on days 3 and 7 than the control group (both P < 0.05).
The effect of CVVH on iron metabolism disorder Plasma hepcidin and ferritin did not differ between the groups upon admission (all P > 0.05, Figure 3). However, plasma hepcidin on days 3 and 7 in both groups were lower than that on day 1 (all P < 0.05, Figure 3D). Plasma hepcidin on days 3 and 7 in the CVVH group was signi cantly reduced compared with that in the control group (both P < 0.05). Plasma ferritin on days 3 and 7 in both groups were higher than that on day 1 (all P < 0.05, Figure 3E) while plasma ferritin on days 3 and 7 in the CVVH group was signi cantly lower than that in the control group (both P < 0.05). There were no signi cant differences in plasma sTfR between the two groups, or among different time points (all P > 0.05, Figure 3F).
The effect of CVVH on the severity and outcome SOFA scores signi cantly decreased from day 1 to day 7 in both groups (all P < 0.05). However, SOFA score on day 7 in the CVVH group was signi cantly lower than that in the control group (P < 0.05).

Discussion
The present study evaluated the impact of CVVH on dynamic changes of indicators re ecting the iron metabolism disorder in sepsis patients. We found that plasma IL-6, hepcidin, ferritin and RDW was signi cantly increased in sepsis but reduced after CVVH treatment. However, there was no signi cant difference in sTfR between the two groups along with the time. In addition, CVVH did not seem to improve the 28-day survival for patients with sepsis.
Plasma hepcidin, IL-6, ferritin and sTfR are in ammatory anemia-associated parameters for patients with sepsis. Our previous study on sepsis observed that these parameters increased remarkably within the rst week of ICU admission [4]. Furthermore, plasma IL-6 was positively correlated with plasma hepcidin and ferritin in patients with sepsis [4]. Hepcidin as a liver-producing peptide hormone plays a central role in regulating plasma iron [23]. It inhibits iron absorption from duodenum, release from hepatocytes and circulating macrophages, resulting in a low level of plasma iron [24]. The production of hepcidin is affected by in ammation. IL-6 can induce up-regulation of hepcidin expression through the STAT-3 signaling pathway, thus causes a series of disorders in iron metabolism, which were signi cantly associated with anemia and 28-day mortality [2-4, 7, 25-28]. The current study revealed that CVVH treatment signi cantly reduced plasma IL-6, hepcidin and ferritin, supporting previous ndings that in ammatory mediators such as plasma IL-6 and TNF-α can be cleared by hemo ltration in animal and human experiments [10,13,16,29]. The mechanism that CVVH clears plasma IL-6 may be adsorption rather than convection because IL-6 with a molecular weight of 26 kDa is di cult to be ltered through the pores of the membrane in CVVH [29] . In addition, polymyxin B in the ltration membrane is a cyclic lipophilic peptide antibiotic which has high a nity for the lipid A moiety in the endotoxin, and thus CVVH can effectively adsorb lipopolysaccharide. Rapid removal of lipopolysaccharide as an activator of cytokines production by CVVH can also lead to a decreased production of IL-6 [29]. Moreover, hepcidin with a molecular weight of approximately 10 kDa can be effectively removed through the membrane in CVVH [30]. The reduction of plasma hepcidin and ferritin in sepsis patients treated by CVVH may also be associated with the reduction of plasma IL-6 because the clearance of IL-6 by CVVH can relieve the upregulation of hepcidin through IL-6-hepcidin axis and block the induction of pro-in ammatory cytokines such as TNF-α, IL-1β and IL-6 to the production of ferritin [25][26][27].
sTfR level seemed to be unaffected by CVVH in patients with sepsis in this study. sTfR has a large molecular weight of 85 kDa with a very low CVVH clearance rate. Moreover, sTfR re ects the degree of iron availability for cells, but is not affected by in ammation. Therefore, the clearance of IL-6 by CVVH does not affect the production of sTfR.
Plasma EPO was reported to be elevated with a tendency of gradual decrease within the rst week of ICU admission in sepsis patients [4]. However, we found in the present study that there was no signi cant change in EPO concentration in sepsis patients treated by CVVH. This may be mainly because EPO with a molecular weight of 34 kDa is di cult to be ltered through the membrane pores in CVVH [31]. In addition, CVVH and EPO production is mutually affected. Plasma EPO is usually increased in response to anemia and hypoxemia, whereas the production of EPO is compromised by impaired renal function (which decreases EPO production) and pro-in ammatory cytokines (such as TNF-α and IL-6, which inhibits EPO production) in patients with sepsis [4,32]. On the one hand, CVVH may ameliorate hypoxemia and acute lung injury, thus decreasing the production of EPO [33]. However, clearance of proin ammatory cytokines by CVVH may increase the production of EPO. Considering the complicated effects of CVVH on EPO production, it is not surprising to see in this study that CVVH failed to affect plasma EPO concentration in sepsis patients.
RDW is typically used as part of a complete blood count to quantify changes in the size of circulating red blood cells [34]. For patients with sepsis, signi cantly increased RDW is observed on admission and can serve as a useful predictor of mortality, as evidenced in our previous study [4,35]. In the present study, we found that sepsis patient treated with CVVH had a lower RDW. The mechanism that CVVH affects RDW remains unknown. However, previous research displayed a strong correlation between RDW and IL-6, and IL-6 was one of independent predictors of RDW [4,34]. Pro-in ammation cytokines such as IL-6 can also impair the maturation of RBC and thus accelerate the entrance of immature RBCs into circulation.
Therefore, we speculate that the decreased RDW by CVVH may be related to the clearance of IL-6.
Hemoglobin level in sepsis patients treated with CVVH did not change greatly within the rst week. The incidence of sepsis-related anemia is as high as 90% within 3 days. It is caused by multiple mechanisms including infection, iatrogenic blood loss, dilution during uid resuscitation, decreased plasma iron, inhibition of erythropoietin production, shortened red blood cell life, and malnutrition with infection to be the major reason [3,[36][37][38]. In ammation up-regulated hepcidin which inhibited plasma iron levels and had a measurable effect on hemoglobin levels from 7 days after diagnosis of sepsis [23]. Therefore, within the rst week, the effect of decreased hepcidin by CVVH on hemoglobin production may be insigni cant.
The 28-day survival of sepsis failed to be improved by CVVH, which is inconsistent with the metaanalysis published by Putzu et al. [14]. There are several explanations for this result. First, the relatively small sample size of this study may not be adequate to reach a statistically signi cant result. A multicenter randomized trial with a large sample size would be necessary in the future trials. Second, iron metabolism disorder partly contributes to the severity of sepsis and may be insu cient to affect the outcome. Many other complications accompanied by sepsis may also in uence the mortality.
There are several other limitations in the present study. First, this is a single-center study, which inevitably introduces selection bias. Second, CVVH is not a standardized technique for the treatment of sepsis. Therefore, the CVVH regimen and system used could affect the results. Third, we did not assess the differences in subgroups of sex and septic shock. However, the sex distribution and the number of patients using vasopressors on admission was balanced between the groups. Finally, the implement of the CVVH received by patients with sepsis was dependent on the choice of these patients or their legal guardians except the decision of their doctor-in-charge. Hence, this might produce a selection bias.

Consent for publication
Not applicable.

Availability of data and materials
The datasets analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.    Comparison of plasma IL-6 between the control and CVVH groups. CVVH, continuous veno-venous hemo ltration.