Does the Antisecretory Peptide AF-16 modulate fluid balance and inflammation in experimental peritonitis induced sepsis?

Sepsis is a life-threatening condition due to a dysregulated immunological response to infection. Apart from source control and broad-spectrum antibiotics, management is based on fluid resuscitation and vasoactive drugs. Fluid resuscitation implicates the risk of volume overload, which in turn is associated with longer stay in intensive care, prolonged use of mechanical ventilation and increased mortality. Antisecretory factor (AF), an endogenous protein, is detectable in most tissues and in plasma. The biologically active site of the protein is located in an 8-peptide sequence, contained in a synthetic 16-peptide fragment, named AF-16. The protein as well as the peptide AF-16 has multiple modulatory effects on abnormal fluid transport and edema formation/resolution as well as in a variety of inflammatory conditions. Apart from its’ anti-secretory and anti-inflammatory characteristics, AF is an inhibitor of capillary leakage in intestine. It is not known whether the protein AF or the peptide AF-16 can ameliorate symptoms in sepsis. We hypothesized that AF-16 decreases the degree of hemodynamic instability, the need of fluid resuscitation, vasopressor dose and tissue edema in fecal peritonitis. To test the hypothesis, we induced peritonitis and sepsis by injecting autologous fecal solution into abdominal cavity of anesthetized pigs, and randomized (in a blind manner) the animals to intervention (AF-16, n=8) or control (saline, n=8) group. After onset of hemodynamic instability (defined as mean arterial pressure < 60 mmHg maintained for > 5 minutes), resuscitation was initiated by an infusion of AF-16 or saline. We recorded respiratory and hemodynamic parameters hourly for twenty hours and collected post mortem tissue samples at the end of the experiment. No differences between the groups were observed regarding hemodynamics, fluid balance, lung mechanics, gas exchange or histology. This experimental study suggests that AF-16 does not modulate sepsis symptoms in peritonitis induced sepsis.


48
Introduction 49 Sepsis is defined as "life-threatening organ dysfunction caused by a dysregulated host response to 50 infection" [1]. In septic shock profound circulatory and metabolic abnormalities contribute to an increase 51 in mortality, with up to 40 % in-hospital mortality [1][2][3]. Requirement of vasopressor therapy to sustain 52 a mean arterial pressure (MAP) > 65 mmHg in combination with persistent serum lactate level > 2 53 mmol/L after fluid resuscitation are the clinical hallmarks of septic shock [1]. Sepsis and septic shock

153
The animals were euthanized with 100 mmol KCl i.v. at the end of the experiment under deep anesthesia.

154
Thereafter the chest wall was opened. Lung tissue samples were collected from both lungs from the 155 following regions: apical-medial, medial-medial, caudal-dorsal, caudal-medial and caudal-ventral.

156
Samples were also taken from heart, liver, kidney, intestine and skin. The samples were immediately 157 immersed in 10% buffered formalin. A veterinary pathologist who was blinded for the group allocation 158 evaluated the samples histologically. Wet-to-dry ratio was measured in the above mentioned lung 159 regions from the right lung. Samples were weighed, and dried in an oven, at 50° C, until the weight did

173
The two groups were comparable at baseline regarding hemodynamics and respiratory parameters 174 (Table 1). Mean time from peritonitis induction to onset of hemodynamic instability was 4.5 +/-2.2 and 175 4.9 +/-1.2 hours in treatment and control groups, respectively.

189
After established hemodynamic instability both groups presented with a decline in oxygenation 190 (PaO 2 /F I O 2 ratio). The AF-16 group went from 60 +/-3.0 kPa at baseline, to 33 +/-13.9 kPa at the end 191 of the protocol, the control group went from 61 +/-4.7 kPa, to 27 +/-15.7 kPa at the same time points

192
( Table 2). There was no statistically significant difference in oxygenation between the intervention and 193 control groups (two way ANOVA F (2, 54) = 0.093, p = 1) as a function of time. Respiratory rate was 194 adjusted to keep PaCO 2 under 6.5 kPa (Table 2). 195

Extravascular lung water (EVLW) and Stroke Volume Variation (SVV)
207 There was no statistically significant difference in EVLW evolution between intervention and control 208 groups as a function of time (two-way ANOVA, F (7,87) = 0.77, p = 0.614). EVLW increased from 360 209 +/-90 ml at baseline to 550 +/-370 ml at the end of the observation period, and from 290 +/-40 ml to 210 450 +/-300 ml in the intervention and control groups, respectively (Table 3). Neither was there any 211 statistically significant difference between groups as a function of time regarding SVV (Table 3).
212  The onset of hemodynamic instability was defined as mean arterial pressure under 60 mmHg, both 220 intervention and control groups presented with increases in heart rate at this stage of the experiment 221 (Table 3). There was no statistically significant difference between groups regarding heart rate 222 throughout the observation period (Two way ANOVA, F(21,252) = 0.765, p =0.761). Onset of 223 hemodynamic instability was also accompanied by an increase in hemoglobin concentration in both 224 groups, while no statistically significant difference between groups was detected (Two way ANOVA,

225
F(21,253) = 0.214, p = 1.00). The two groups did not differ in a statistically significant way in either 226 lactate, pH or oxygen extraction ratio (Table 3).

Fluid balance 228
There was no statistically significant difference between groups in fluid requirements, urinary output, 229 fluid balance (these parameters described as ml/kg/h of sepsis duration), norepinephrine consumption 230 (µg/kg/h of sepsis duration) nor percentage body weight gain (kg body weight before and after 231 experiment) (Table 4). Wet-to-dry ratio 238 liver were analyzed. Lung samples were analyzed separately and pooled together. Skin had the lowest 239 water content, kidney the highest (Table 5).

252
It could be an early stage in the process of leukocyte migration through the vessel wall, but leukocytes seem to 253 remain in the intima which is suggestive of endoarteriolitis, which could be predisposing for thrombosis (control).

255
Vessels often displayed prominent endothelial cells and leucocytes were found in the process of

294
The definition of septic shock in humans include an increased serum lactate > 2 mmol/L despite adequate 295 fluid resuscitation. In this study we did not observe any statistically significant hyperlactatemia. Oxygen

303
There was no statistically significant difference in EVLW evolution between the groups during the 304 experiment, all except three animals (one from intervention group and two control animals) did manifest study a statistically significant reduction of EVLW in the intervention group was found, as an isolated with an increase in EVLW, leaving out the "EVLW non responders" in post-hoc analysis did not yield 311 a different outcome.

326
In this study AF-16 was given in repeated doses, the initial dose being three times higher than in our 327 previous ventilator induced lung injury (VILI) model [23]. We cannot rule out that the intervention with 328 AF-16 would be more effective at an even higher or continuous dose, as AF in plasma has a rapid 329 turnover rate [12], or that an effect could have been observed in a less severe shock state. Moreover, the 330 number of animals studied was limited, so minor changes between the groups might not have been 336 experiment. One must also accept the possibility of interspecies variability in intestinal flora and host 337 response to both infection and intervention.

338
We conclude that, contrary to our hypothesis, in this pilot study in a porcine experimental model of fecal 339 peritonitis and sepsis we could not detect any differences between intervention and control groups