Dynamic Changes in Circulating Plasma B-type Natriuretic peptide levels and its Inuencing Factors in Patients with Hemorrhagic Fever with Renal Syndrome

Background: Typically, Hemorrhagic fever with renal syndrome (HFRS) occurs in ve sequential stages: Febrile, hypotensive, oliguric, polyuria, and convalescent stage. The pathophysiological process involves the release of inammatory mediators, the uctuation of blood volume and the injury of the Multiple organs, and all the factors mentioned above may cause the elevation of the level of plasma B-type natriuretic peptide (BNP). In addition, the physiological effects of BNP in regulating blood volume and promoting angiogenesis may in theory alleviate the pathological state of HFRS. So, this study is to explore the clinical application of BNP in HFRS by dynamically monitoring the BNP levels and analyzing the factors associated with BNP expression. Methods: Eighty-six in-patients with HFRS were divided into the mild, moderate and severe groups according to disease severity. Mixed-effects linear model was used to analyze the differences in BNP expression according to disease severity and disease stage. The factors affecting BNP expression were analyzed using a linear regression model for each disease stage. Results: BNP showed dynamic changes that corresponded with disease progression. The more severe the disease, the overall BNP level was higher. Disease severity, neutrophil count and pulse pressure were independent factors for BNP, while cardiac injury related indicators were not. Conclusions: In HFRS the main factors promoting BNP expression were inammation and blood volume, rather than heart disease. BNP can be regarded as an index for judging the severity of the disease and guiding body uid treatment. Basic clinical data recorded for all patients, including sex, weight, comorbidities, signs, clinical type, date onset, and blood collection date. Blood pressure (pulse pressure = systolic − diastolic pressure), heart rate, urine output, and body temperature were monitored daily. Neutrophil, creatine kinase-MB and creatinine clearance rate ([{140 – age} × body weight {kg}]/ [0.818 × serum creatinine {μ mol/L} × 0.85 for women]) were at various stages of the disease. All blood samples were collected early in the morning (06:00–07:00), with the patient in the horizontal position. All the patients fasted for at least 8 h before blood collection.

. Heilongjiang province ranked one of the highest levels of HFRS incidence in China, HFRS here was mainly caused by HTNV and SEOV (3). Hantavirus infection causes extensive damage in capillaries and small vessels via the release of vasoactive substances and in ammatory mediators (IL-1/6, TNF-α and so on) (1,4,5), additionally leading to uctuations in blood volume, coagulation disorders and injury to body tissues and organs. Typically, HFRS occurs in ve sequential stages: febrile, hypotensive, oliguric, polyuria, and convalescent. The polyuria stage is subdivided into a transitional stage, early stage, and late stage.
B-type natriuretic peptide (BNP) is a cardiac natriuretic peptide hormone that was identi ed in a recent study (6,7). Its concentration in systemic circulation among the healthy population is approximately 3.5 pg/ml, but its expression increases signi cantly in the presence of ventricular wall tension (8,9), volume expansion (10,11), in ammation cytokines (IL-1β\TNF-α) (12)(13)(14)(15) tissue hypoxia (16,17) and other stress factors. The physiological effects of BNP include diuresis, natriuresis, vasodilation, and inhibition of the renin-angiotensin-aldosterone system, which play crucial roles in the regulation of water and electrolyte homeostasis (18). In addition, BNP can stimulate endothelial progenitor cell proliferation and affect vascular growth and regeneration (19,20).
We hypothesized that in ammatory mediators (IL1/TNF-α), increased blood volume, and hypoxia induced by tissue injury in the course of HFRS may stimulate BNP synthesis, and the physiological effects of BNP in regulating volume homeostasis and promoting angiogenesis may reduce volume load and repair extensive vascular damage in HFRS.
Therefore, the study of plasma BNP has important clinical signi cance for HFRS. Elevated plasma BNP levels in PUUV infection, Crimean-Congo hemorrhagic fever, and dengue infection have been reported in recent years, but these studies tested BNP only once throughout the course of the disease (21)(22)(23)(24)(25) In this study all patients were by the clinical diagnostic criteria for HFRS. The exclusion criteria were: (1) less than 18 years of age; (2) chronic heart disease (coronary artery disease, cardiac failure, severe valvulopathy, and/or cardiomyopathy); (3) chronic renal failure; (4) known pulmonary hypertension, pulmonary embolism, or chronic obstructive pulmonary disease; and (5) diseases of the central nervous system (e.g., meningitis, brain abscess, and cerebral hemorrhage).
All the cases were discussed daily at multi-expert meetings, and clinical management decisions were made by a competent physician. The therapeutic schedule included rehydration, electrolyte regulation and, in some patients, abatement of fever and dialysis therapy. In some patients, sample collection was discontinued while they received treatment in the intensive care unit; they were returned to our department after their condition had stabilized and were included in the study again. Each patient's disease was classi ed as severe, medium, or mild according to the classi cation criteria for HFRS (26), Supplementary 1 (the severe and critical types described in the criteria were collectively termed "severe" in this study). was measured in the M mode, and the time-velocity integral method was used to measure peak blood ow velocity in the mitral valve in early diastole/late diastole (E/A). For 12-lead ECG recording, a Cardio Fax 3 system (Nihon Kohden, Tokyo, Japan) and a Wilson lead network were used, with the patient supine.

Statistical analysis
Continuous variables were expressed as mean standard deviation, and categorical variables were expressed as case numbers and percentages. Differences between groups were compared by analysis of variance for normally distributed variables, Wilcoxon rank-sum test for variables with a skewed distribution, and the chi-square test or Fisher's exact test for categorical variables. Mixed-effects linear model was used to analyze the differences in BNP expression according to disease severity and disease stage. The effect of clinical examination and laboratory indicators on BNP expression were analyzed using a linear regression model for each disease stage. Variables identi ed as signi cant (P < 0.05) in the univariate analyses were further screened in multivariate analyses. All statistical analyses were conducted using the SAS software (version 9.2), and a two-sided P value of <0.05 was established as the level of statistical signi cance for all the tests.

Patient characteristics
The clinical features of the patients are provided in Table 1. 241 samples were collected from 86 patients with HFRS. The reasons are as follows: This is due to two or three natural course overlaps in some severe patients and the lack of transitional stage in some mild patients; Some patients had missed the fever stage when they were hospitalized; In addition, 1 person abandoned treatment during the oliguria phase, and 5 people interrupted BNP testing due to treatment in ICU. 61.1% of severe received dialysis, while only 6.1% of moderate; Pulmonary edema occurred in 2 patients with severe disease.
The number of neutrophils, pulse pressure and creatinine clearance showed dynamic changes, and the change range was severe > medium >mild. Pulse pressure was the highest and creatinine clearance was the lowest in the oliguria stage of severe HFRS and in the transition stage of moderate HFRS. During the febrile stage of severe and moderate patients, the creatinine clearance rate decreased signi cantly (Table 2).
Plasma BNP levels at different stages in patients with mild, medium and severe HFRS The dynamic changes observed in the plasma BNP levels are shown in Figure 1, As the disease progresses, BNP levels gradually increase to reach a peak, and then decline as the disease improves. The peak of BNP levels was observed in the oliguric stage of patients with severe HRFS, in the transition stage of patients with moderate HFRS, and in the early polyuria stage of patients with mild HFRS. Comparison BNP levels between consecutive stages in HFRS are shown in Table 4. In addition, plasma BNP levels increased with increasing disease severity at the same stage (except during the febrile phase), with statistically signi cant differences, Table 3. After adjusting for age, sex, weight and dialysis, the severity and stage as well as their interactions had a signi cant effect on BNP levels, Table 5.

Factors in uencing the expression of BNP in different stages of HFRS
Univariate linear regression analysis was used to examine the clinical parameters that may in uence BNP expression in each stage of HFRS. Disease severity, neutrophil count, creatinine clearance rate, pulse differential pressure, E/A, and age were found to be signi cantly correlated with serial BNP measurements by univariate linear regression analysis (P< 0.05) ( Table 6). Sex, weight, heart rate, LVEF, abnormal ECG and Creatine kinase-MB ndings were not correlated with BNP levels (P > 0.05) (supplementary 2). Multivariate linear regression analysis showed that the signi cant independent in uencing factors affecting BNP were distinct in different stages, for example, the neutrophils count was signi cant in the febrile stage, while pulse pressure and disease severity were in the oliguric stage, disease severity was in the transitional stage, neutrophil count and disease severity were in the early polyuria stage, and the age was signi cant in the late polyuria stage (Table 6).

Discussion
These were the main ndings of our study: (1) In HFRS patients, the more severe the disease, the overall BNP level was higher. (2) The BNP levels at different stages of HFRS showed dynamic changes that corresponded with disease progression.
(3) Disease severity, neutrophil count, and changes in pulse pressure were the signi cant independent factors associated with the plasma BNP level, while Indicators related to cardiac injury was not.
This study showed that the BNP levels were signi cantly different between the different stages of HFRS, and the stage at which the peak BNP level was observed varied according to disease severity .Therefore, to discuss the clinical application of BNP in HFRS, plasma BNP level testing in every stage was more scienti c value. This study is the rst to investigate HFRS patients with Hantan and Seoul virus infections, and comprehensively describes signi cant in uencing factors and the dynamic changes in BNP levels according to stage and severity.
The effect of renal function on BNP is widely concerned by doctors. BNP clearance is generally achieved by two routes, the most common one is binding to neutrophilic peptide chain endonuclease (NEP) on the membrane surface. In addition, BNP enters lysosomes through intracellular phagocytosis mediated by the receptor NPRC and then is degraded (27). Previous studies have suggested that BNP concentrations are strongly correlated with residual renal function (28,29), but several subsequent studies have reported that BNP only has a 21-22% or lower dependence on renal function for its clearance (30)(31)(32). In this study, creatinine clearance was found to be associated with BNP, but it was not an independent in uential factor. In the fever stage of severe and medium HFRS, the creatinine clearance rate was signi cantly reduced, but the level of BNP only slightly increased, which also indicated that the effect of creatinine clearance rate on BNP was very weak.
We focused on the correlation between heart injury and BNP in patients of HFRS (33). Heart rate, electrocardiogram, creatine kinase isoenzyme, LVEF, and E/A were used to assess cardiac status. Only E/F affects BNP level, but it is not an independent factor of BNP. The study suggests that heart damage is not the main cause of BNP elevation in HFRS. A report related to sepsis also showed that the main factor affecting BNP expression was not heart disease (33).
In this study, neutrophils were used as markers of in ammation. Proin ammatory cytokines IL-1β and TNF-α selectively increased the activity of BNP promoters, thereby regulating the expression of the myocardial BNP gene (12, 13). We did not further examine the cytokines (IL-1β and TNF-α) in patients, because previous studies on in ammatory mediators of HFRS have been clear (1,4,5). Pulse pressure were used as an indicator of blood volume changes. In the setting of volume expansion, the resulting wall stress can initiate the synthesis of BNP in the ventricular myocardium and, thus, cause an increase in the plasma BNP levels (10,11). BNP, with a half-life of only 22 min, can be used for real-time monitoring of blood volume changes. It has been widely used in the volume management of patients with heart failure in clinical practice. At present, no speci c approved therapy is available for HFRS, the treatment is primarily supportive.
Maintaining the uid and electrolyte balance together with circulatory volume is very important to avoid dangerous over hydration. The results suggest that BNP can be used to monitor blood volume changes in HFRS, and doctors can adjust the uid treatment regimen according to the change of BNP level, so as to reduce complications such as pulmonary edema and hypervolemia syndrome.
The main characteristic of the febrile stage is in ammatory reaction, while the tissue damage is relatively light and blood volume maintain normal or reduced. In this stage, the slight increase in plasma BNP levels observed in this study may be attributable to the in ammatory stimuli. As the disease progresses into the oliguric stage, patients with moderate and severe HFRS exhibit increased kidney damage, a signi cantly reduced urine output, signi cantly increased blood volume, and even pulmonary edema. These factors may result in signi cant increase of the observed BNP levels. The transitional stage is characterized by the transition from oliguria to polyuria. In this stage, although urine volume starts to increase, renal function is further decreased and blood volume continues to increase, leading to a further increase in the plasma BNP levels. In the present study, in patients with moderate HFRS, the plasma BNP levels peaked in this period. However, the peak value of plasma BNP levels in patients with severe HFRS was observed in the oliguric stage. This is probably because 61.1% of the patients with severe HFRS underwent hemodialysis. As a result of the hemodialysis treatment, their blood volume decreased rapidly and the concentration of some in ammatory mediators decreased. Therefore, their condition was alleviated in a short time and a rapid decline in their BNP levels was observed. During the early to late polyuria stage, organ function improved signi cantly and gradually returned to normal, with a decrease in blood volume, and a reduction in the in ammatory response. When the factors causing the elevation of BNP decreased, the BNP level would gradually reduce subsequently. The study reveals important information about the pathophysiological processes of HFRS associated with BNP. It can be seen that the pathophysiological changes of HFRS in each course of disease make BNP level signi cantly different. This suggests that clinicians must analyze BNP results in combination with the course of HFRS. In addition, BNP was involved in the maintenance of electrolyte homeostasis and promoted the regeneration of vascular cells, which may be an important part of the self-limiting mechanism in HFRS.
The higher the disease severity of HFRS is, the higher the BNP level will follow, and the severity independently affected the BNP level, which suggested that the plasma BNP level could be used as an indicator to evaluate the HFRS severity.
Papanikolaou also found that disease severity, rather than infective cardiomyopathy, might be the main factor associated with elevated BNP levels in patients with severe sepsis (33). The study will be continued on the thresholds for determining disease severity at the early stages of HFRS. Vascular endothelial dysfunction is the basic pathological change, which determines the severity of HFRS. In clinical trials, intravenous type A natriuretic peptide (ANP) can improve pulmonary edema and reduce pulmonary vascular permeability in patients with acute lung injury (34, 35). An increase in the plasma BNP levels was found to accelerate vascular regeneration in a mouse of model of limb ischemia induced by femoral artery ligation (20). In addition, Recombinant human BNP (nesiritide) has produced satisfactory results in the treatment of heart failure (36). Therefore, natriuretic peptides may be an effective method to treat HFRS in the future.

Conclusions
In summary, this study provides comprehensive information about the dynamic changes in BNP levels that occur in HFRS according to clinical stage and severity of the disease, and the factors that in uence plasma BNP levels. Based on our observations, BNP can be regarded as an index for judging the severity of the disease and guiding body uid treatment; The main factors affecting BNP level in HFRS were not heart injury and creatinine clearance rate, while were blood volume and in ammatory mediators; In addition, BNP may be an important part of the self-limiting mechanism in HFRS. BNP plays a crucial role in the regulation of water and electrolyte homeostasis and the stimulation of vascular growth and regeneration, and future research should focus on whether it can be used to treat HFRS as well.

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated and/or analysed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This study was funded by China Postdoctoral Science Foundation (20100471095) and Natural Science Foundation of China (30700700).
Authors' contributions NM, YL and YL designed the study. NM, YL, YH, MX, YZ, SS and LJ collected, analyzed and interpreted the patient data.
NM was a major contributor in writing the manuscript. All authors read and approved the nal manuscript.    Notes: P < 0.05 was considered to indicate statistical signi cance. "Transitional" refers to the transitional stage of the polyuria stage. "Early" refers to the early stage of the polyuria stage. "Late" refers to the late stage of the polyuria stage.  Notes: β is the parameter estimate: a positive value indicates a positive correlation, and a negative value indicates a negative correlation. P < 0.05 was considered to indicate statistical signi cance. "Ref." indicates the reference level.

Figures
Circles and bars represent the mean standard deviation. P < 0.05 indicates that the BNP levels were signi cantly different between two adjacent stages. Mild HFRS, solid line; moderate HFRS, dashed line; and severe HFRS, dotted line.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.