The Impact of Fluid Resuscitation on Hemodynamic of Hemorrhagic Shock: An Animal Model Experimental Study

Background: To investigate the effect of fluid resuscitation on glycocalyx shedding, and extravascular lung water index (ELWI), mean arterial pressure (MAP) and oxygen delivery (DO 2 ) changes. Methods: Male domestic piglets ( Sus scrofa ) 6-10 weeks old anesthetized and bled until mean arterial pressure drop to 20% of baseline and resuscitated with normal saline as much as blood drowned, followed with 40 mL/kg of normal saline after 30 minutes. Cardiac index (CI), ELWI, systemic vascular resistance index (SVRI), MAP, atrial natriuretic peptide (ANP) and syndecan-1 were measured before and after each fluid resuscitations. Results: Serum ANP was increased after normal volume fluid resuscitation (p= 0.043) and return its baseline value after hypervolemia fluid resuscitation. Serum Syndecan-1 levels did not increase. A small increase in ELWI only found 60 minutes after fluid resuscitation (p= 0.021). SVRI undergo a gradual decrease, until the lowest value at hypervolemia volume resuscitation. There was no difference between the MAP of the two groups (p= 0.105). Hemoglobin concentration significantly decreased from normal to hypervolemia volume resuscitation (p= 0.009). Oxygen delivery in hypervolemia resuscitation is higher than in normal volume resuscitation (p= 0.012), due to a significant increase in CI at hypervolemia volume resuscitation (p<0.001). Conclusions: Hypervolemia fluid resuscitation in the animal hemorrhage model is not induced glycocalyx shedding. Small increase ELWI was found in 60 minutes after fluid resuscitation. DO 2 is maintained by increasing CI in spite of decreasing hemoglobin level due to hemodilution. Increasing CI is balanced by reducing SVRI to sustain stable MAP.

(ANP) and syndecan-1 were measured before and after each fluid resuscitations.
Results: Serum ANP was increased after normal volume fluid resuscitation (p= 0.043) and return its baseline value after hypervolemia fluid resuscitation. Serum Syndecan-1 levels did not increase. A small increase in ELWI only found 60 minutes after fluid resuscitation (p= 0.021). SVRI undergo a gradual decrease, until the lowest value at hypervolemia volume resuscitation. There was no difference between the MAP of the two groups (p= 0.105). Hemoglobin concentration significantly decreased from normal to hypervolemia volume resuscitation (p= 0.009). Oxygen delivery in hypervolemia resuscitation is higher than in normal volume resuscitation (p= 0.012), due to a significant increase in CI at hypervolemia volume resuscitation (p<0.001).
Conclusions: Hypervolemia fluid resuscitation in the animal hemorrhage model is not induced glycocalyx shedding. Small increase ELWI was found in 60 minutes after fluid resuscitation. DO 2 is maintained by increasing CI in spite of decreasing hemoglobin level due to hemodilution. Increasing CI is balanced by reducing SVRI to sustain stable MAP.

Background
Most of the pediatric guidelines recommend liberal fluid resuscitation. However, excess fluid resuscitation may lead to complications including hemodilution, impairment of oxygen delivery, and hypothermia.
[i] , [ii] Fluid resuscitation increases the hydrostatic pressures of the pulmonary circulation and stimulates the release of atrial natriuretic peptides (ANP) due to acute stretching of the walls of the atria. [ The goal of fluid resuscitation is to increase the patient's preload and subsequently the stroke volume. According to the Frank-Starling law, increases in preload will produce an increase in stroke volume. However, this relationship only continues up until the Frank-Starling curve reaches a plateau, after which no further increase in stroke volume occurs. Understanding the mechanism of vascular leakage and hemodynamic changes during fluid resuscitation is essential to develop safe clinical guidelines for fluid resuscitation in children. In this experimental animal study, we investigated the effects of normal volume, i.e. euvolemic fluid resuscitation, and hypervolemic fluid resuscitation on serum ANP and syndecan-1 levels, ELWI, CI, SVRI, MAP, Hb, and DO 2 . days, in which they were given antibiotics oxytetracycline (10mg/kgBW) via intramuscular route, and anti-helminth (oxfendazole) bolus mixed with their food. The experiment was conducted in the Laboratory Animal Management Unit where the animals were fed commercially available food twice daily with free access to water, and were placed in a cage that was cleaned twice daily.

Ethics
As there was no need to sacrifice the animals, following the experiment and fluid resuscitation, the animals were returned to the institution and treated by veterinarian.
The animals were stabilized and given analgesia and antibiotics for one week course. No animals died during the course of the study.

5
This was a two-phase fluid resuscitation animal study performed at the Experimental Surgery and Radiology Laboratory, Faculty of Veterinary Medicine, Bogor Agricultural University. Eleven healthy male domestic piglets (Sus scrofa), age 6-10 weeks old were anesthetized with Ketamine and Xylazine and supported by volume control mechanical ventilation, adjusted to blood gas. We infused 3 mL/kg/hour of 0.9% normal saline as maintenance fluid. Environmental temperature was maintained with a thermal blanket.
After one hour of stabilization, hemodynamic parameters were measured, and a blood sample taken for laboratory assay of plasma levels. Following baseline data collection, we induced pressure targeted shock via venous blood drawing until the MAP was reduced by 20% to 80% of the initial MAP. Hemodynamic parameters were recorded. Following 30 minutes of shock, we performed fluid resuscitation in two phases. [i] In phase 1, a bolus of normal saline equal in volume to the blood loss needed to induce shock was administered. Phase 2 was performed 30 minutes later, with 40 mL/kg of saline given as a bolus, to stimulate hypervolemic resusictation. Hemodynamic parameters were measured three times, at 3 minute intervals, for each stage of fluid resuscitation, and at 30 minutes and 1 hour after the last fluid administration. The numerical means of each data set were used for the statistical analysis.

Statistical Analysis
Sample size calculation was performed using Federer's formula.
[iii] As we used a twophase model, euvolemic vs hypervolemic fluid resuscitation in one single study group, the minimum sample size required was 9 animals. To allow for errors in data collection and subject dropout, we studied 11 piglets.
Normally distributed data are presented as means and standard deviation values, whilst nonparametric data are presented as medians and ranges. ANP, syndecan-1, CI, SVRI, MAP, Hb, and DO2 were normally distributed, but ELWI at baseline, after hypervolemic fluid resuscitation, and at 30 and 60 minutes after hypervolemic resuscitation were not normally distributed. A statistical hypothesis test with paired t-test was used for normally distributed data and Wilcoxon signed-rank test for nonparametric data. Pearson correlation coefficient was used to measure the strength of linear association between ANP and SVRI. The statistical analysis was performed using SPSS Version 20.0.

Results
The baseline data and characteristics of all subjects are shown in Table 1. Mean blood drawn to produce a 20% fall in MAP was 101 ± 56 mL. The average time required for euvolemic fluid resuscitation was 2 ± 1 minutes, and for hypervolemic resuscitation, 9 (7-24) minutes. Changes in hemodynamic profiles during the study are shown in Table 2.
The highest CI, MAP, and DO 2 were recorded following hypervolemic resuscitation, whilst SVRI was lowest at this time. ELWI increased gradually until 1 hour after hypervolemic resuscitation. Figure 1 shows that ANP increased significantly from 85.20 ± 40.86 ng/L at baseline to 106.42 ± 33.71 ng/L following euvolemic resuscitation (p = 0.043). Serum ANP decreased to 82.60 ± 41.21 ng/L and 83.55 ± 46.09 ng/L respectively immediately following hypervolemic resuscitation and 30 minutes after hypervolemic resuscitation. Serum syndecan-1 levels did not increase during this study. Figure 2 shows the data for ELWI at baseline. There was no significant difference between ELWI at euvolemia and baseline value (p= 0.722), nor following hypervolemic resuscitation and baseline (p= 0.398), but there was a significant difference between baseline value and 60 minutes following hypervolemic resuscitation (p = 0.021). SVRI shows a gradual decrease, with nadir values after hypervolemic resuscitation. There was no correlation between serum ANP changes and SVRI (r= 0.106; p= 0.281). There was no difference between the MAP between euvolemic and hypervolemic resuscitation at any point in the study (p = 0.105). Hemoglobin concentration decreased significantly from 9.32 ± 0.86 g/dL after euvolemic resuscitation to 8.39 ± 0.79 g/dL following hypervolemic resuscitation (p = 0.009). Oxygen delivery following hypervolemic resuscitation at 2281 ± 525 ml/minute was significantly higher than after normovolemic resuscitation, 2028 ± 409 ml/minute (p = 0.012), as seen in Figure 4.

Discussion
This study showed that ANP levels increase, immediately after normovolemic resuscitation. The ANP levels then decrease to near baseline levels before shock was induced. This result is similar that found by Ozer et al.
[i] They showed that the factor limiting increased ANP secretion was the intact pericardium. Maximal increase in ANP secretion was significantly higher when the pericardium was removed. Increasing ANP was not found in the study of Chappell et al. where fluid infusion to maintain euvolemia was carried out simultaneously with blood removal resulting in euvolemic hemodilution. 9 We believe that the rapid volume expansion in this study induces ANP secretion. It should also 8 be noted that Chappell et al. expressed their serum ANP levels in ng/L of albumin which corrected for the hemodilution factor.
The limited degree and short duration of increased ANP did not cause any glycocalyx shedding. However extravascular lung water was increased by 0.93 mL/kg. A study in acute lung injury induced in pigs showed that the ELWI was increased from 6.3 mL/kg at baseline, to 9.4 mL/kg 3 hours after fluid resuscitation.
[ii] Zhao and coworkers showed in adult ARDS patients, ELWI in their survival group was 13.0 ± 3.6 mL/kg and in non- Although there was no correlation between ANP changes and SVRI, we found SVRI decreases gradually after fluid resuscitation. A decrease in arterial elastance and SVRI after fluid resuscitation has been reported in septic patients by Monge Garcia and colleagues.
[ix] SVRI reduction can be induced by parasympathetic activity[x] to compensate for the increase in cardiac output. Figure 3 shows the increased CI was balanced by the reduction of SVRI, such that the MAP remains stable. Our results show that the MAP at euvolemia and after hypervolemic resuscitation remain the same. Since increasing CI will be balanced by decreasing SVRI, a decision to use fluid for increasing MAP must be considered with the risk of increasing ELWI, especially when glycocalyx shedding is occurring. In humans, Glassford  Guyton and Lindsey showed that partial constriction of the aorta increased the left atrial pressure in dogs and lead to accumulation of lung edema.
[xii] In contrast, decreasing SVRI which results in "forward flow" with increasing CI could be a mechanism which opposes the development of excessive ELWI by reducing pulmonary hydrostatic pressure.
Fluid resuscitation increases stroke volume which helps maintain blood pressure.
However, fluid resuscitation also dilutes the hemoglobin concentration in the blood, potentially leading to a decrease in DO 2 . Adequate DO 2 is essential for ensuring patient wellbeing. In this study, hemoglobin concentration was significantly lower after hypervolemic resuscitation, but DO 2 was higher than after euvolemic resuscitation. Figure   4 shows the role of CI in increasing DO 2 after hypervolemic resuscitation. In those situations where CI cannot increase further, hypervolemic resuscitation should be carefully considered. Despite afterload reduction by decreasing SVRI, there may not be 10 sufficient cardiac reserve to generate an increased CI.

Conclusions
Hypervolemic fluid resuscitation in our animal hemorrhage model does not induce glycocalyx shedding. DO 2 is maintained by increasing CO in spite of decreasing hemoglobin level due to hemodilution. Increasing CO is balanced by decreasing SVR to sustain stable MAP. SVRI reduction could be a protective mechanism for a patient with limited cardiac reserve that may support the cardiac output and restrain extravascular lung water. To achieve optimal results with fluid resuscitation, all of these parameters should be monitored. Further studies looking at the interrelationship between fluid resuscitation and DO 2 in critically ill children are needed to identify those patients who may benefit from fluid resuscitation and those who could be harmed.

Consent for publication
Not applicable

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

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

Authors' contributions
17 Figure 2 Median Extravascular Lung Water. Serial measurement of extravascular lung water at baseline, after induction of shock, after euvolemic resuscitation, following hypervolemic resuscitation, 30 minutes after fluid resuscitation, and 60 minutes after fluid resuscitation. * = Nonparametric data.