Plasma iron concentrations and systemic inflammatory response syndrome in neonatal foals.

Abstract Background Sparse information regarding plasma iron concentration in neonatal foals and its utility as an inflammatory marker in this population has been published. Objectives To determine the physiologic plasma iron concentration in neonatal foals. To assess its utility as an inflammatory marker to predict systemic inflammatory response syndrome (SIRS) and as a prognostic marker. Animals Forty‐seven ill neonatal foals admitted to a referral equine hospital were divided in 2 groups based on the SIRS criteria (24 SIRS and 23 non‐SIRS). Two control groups of 43 hospital and 135 stud farm healthy neonatal foals were also included. Methods Observational prospective study. Data were summarized by mean and its 95% confidence interval and absolute frequency and percentage for quantitative andqualitative variables. One‐way ANOVA, ANCOVA (group and age effects) and Dunnett as posthoc analysis were used to compare plasma iron concentration among groups. Results Neonatal foals with SIRS did not have had any statistically significant different plasma iron concentrations compared to non‐SIRS (P = .56) and stud farm control group (P = .99), 172.8 μg/dL (95% CI; 126.0‐219.6), 193.1 μg/dL (139.1‐247.2), and 181.8 μg/dL (171.3‐192.4), respectively. Plasma iron concentration had a large variability in healthy neonatal foals, and was negatively correlated with age in hospital controls (rho = −0.387) and sick neonatal foals (rho = −0.598) (P < .001). Conclusions and Clinical Importance Plasma iron was not a useful marker of SIRS in neonatal foals and was not associated with outcome.

of systemic inflammation is essential in order to provide appropriate treatment. 2 The original systemic inflammatory response syndrome (SIRS) criteria in neonatal foals have a significant association with prediction of sepsis and nonsurvival. 3 The SIRS term provides a reference for the complex findings that resulted from a systemic activation of the innate immune response, regardless of the cause. 4 Systemic inflammatory response syndrome describes the clinicopathologic effects of the inflammatory response to a variety of insults, including bacterial infection, endotoxemia, ischemia, hypoxia, trauma, and burns. 5 The original SIRS criteria used to describe this clinical syndrome in neonatal foals included presence of 2 or more of: (1) hyperor hypo-thermia (rectal temperature >39.2 C or <37.2 C); (2) leukocytosis or leukopenia (peripheral white blood cell count >12.5 × 10 3 /μL or <4 × 10 3 /μL) or > 10% immature ("band") neutrophils; (3) tachycardia (>120 beats/min); and (4) tachypnoea (> 30 breaths/min). 3,[5][6][7][8] The SIRS criteria is much simpler and faster to use than sepsis scores and might serve as a more rapid screening tool for sepsis in neonatal foals. A positive result to the original SIRS criteria has a sensitivity of 60% and specificity of 69% to predict sepsis when compared to other proposed SIRS criteria 3,9,10 and could be more useful for predicting nonsurvival associated with sepsis. 3 Based on the human pediatric literature, an updated SIRS score for foals was proposed but it does not provide an improved ability in predicting sepsis. 3,11 SIRS criteria for foals of the aforementioned study required the presence of at least 3 of: (1) fever or hypothermia, (2) tachycardia, (3) tachypnea, (4) leukocytosis, leukopenia, >5% band neutrophils, (5) venous blood lactate concentration, or (6) venous blood glucose concentration; at least one of which had to be abnormal temperature or leukocyte count. 9 In adult horses, plasma iron concentration acutely decreases in cases with systemic inflammation. 2 Low iron and high fibrinogen plasma concentrations are both sensitive indicators of systemic inflammation in horses, with sensitivity of 90 and 82%, respectively. 12 Rapid development of hypoferremia is particularly valuable during the earliest phases of infection, before other components of innate and adaptive immunity are mobilized. 2 Red blood cell (RBC) mass in human neonates is highly variable, because of changes in both the mass and the composition of RBCs occurring during the transition from the intra uterine to the extra uterine environment. 13 Sparse information is published about iron in neonatal foals. 14,15 The objectives of this study were to determine the physiologic values of plasma iron concentrations in neonatal foals (<14 days old), and to assess its utility as both an early inflammatory marker to predict SIRS and a prognostic marker in sick neonatal foals.

| Blood sampling and measured variables
Blood samples from ill neonatal foals were collected at admission in ethylenediaminetetraacetic acid (EDTA) and lithium heparin tubes.
Similarly, blood samples from healthy hospital neonatal foals (hospital control group) were collected at different times during hospitalization, usually within the first 48 hours of admission. Finally, blood samples from stud farm control group were collected from 24 to 48 hours after birth.
Blood collected in EDTA from sick foals was used to determine manual hematocrit, total protein by refractometry, manual fibrinogen by heat precipitation method, 16 and also to perform Diff-Quick stained blood smears and complete cell blood count using automated hematology analyzers (Advia 120, Siemens Health Care Diagnostics SL, Barcelona, Spain; LaserCyte, Idexx laboratories, Inc, Netherlands). In the hospital control group, in addition to a normal physical examination, hematocrit, total protein and fibrinogen concentrations were determined to support that foals were healthy to be included in this group (data not shown). In the stud farm control group, serum IgG and plasma iron concentration were the only measured variables. Serum IgG was

| Data analysis
The main objective of this study was to establish the usefulness of iron concentration for diagnosis and prognosis in neonatal foals and obtain an estimate of limits of abnormality (ie, reference range) in neonatal foals. These limits were estimated by calculation of individual 95% confidence intervals (95% CI) (ie, from the SD as a measure of variability), and absolute range (minimum and maximum) for plasma iron concentration for each group, defined as sick, hospital control and stud farm control groups. One-way ANOVA and Dunnett as posthoc analysis of group comparisons detected differences between groups. To evaluate the effect of age on plasma iron concentration, ANCOVA analysis of group as independent factor and age as covariate was performed. Additionally, using t test for independent groups, differences between SIRS, non-SIRS, and outcome were explored from 95% CI of mean (ie, using SE as measure of variability). Cutoffs of less than 59, 79, and 105 μg/dL were tested in order to evaluate the utility of external limits of plasma iron concentration and a Fisher's exact test was performed. Other variables were described by mean and their 95% CI and absolute frequency and percentage for quantitative and qualitative variables, respectively. All statistical analyses were performed using a statistical software package (SPSS version 25, SPSS Inc, Chicago, Illinois) and all analyses were performed with a 2-sided Type I error of 5%.

| RESULTS
A total of 225 neonatal foals (<14 days old) were included in the study.
Forty-three were healthy foals admitted to the hospital, with mean age 4.5 days (95% CI; 3.4-5.6 days) and the 135 healthy neonatal foals left were from a stud farm with mean age 1.5 days (95% CI; 1.5-1.5 days).
The 95% CI limits of plasma iron concentration of the sick group were similar to the stud control group (P = .99) ( Table 1; Figure 1). The hospital control group had significantly lower 95% CI limits of plasma iron concentrations and lower mean plasma iron concentration than sick group (P = .002) (Table 1; Figure 1). With ANCOVA analysis, there was no significant difference between the hospital control and sick foals (P = .06) but age effect was present (P = .001; Figure 2). A negative moderate correlation with age was seen in the hospital control group (rho = −0.387) and in the sick group (rho = −0.598), so the higher the age the lower the iron concentration. Greater variability and higher iron plasma concentrations were specially seen in the first   External cutoff values (less than 59, 79, and 105 μg/dL) were applied for plasma iron concentration with no statistically significant differences between those ( Table 2).

| DISCUSSION
The main findings of this study were: (1) the large variability of plasma iron concentration in healthy neonatal foals, (2) the lack of statistically significant differences in plasma iron concentration in sick and SIRS affected foals relative to those in healthy foals, and (3)  but presented a wider distribution (58.8-305 μg/dL) compared to previously published plasma iron concentration in adults (105-277 μg/dL). 2,12 In healthy foals at birth, plasma iron and ferritin concentrations are lower than in adult horses, followed by a rapid increase in the first 24 hours as a result of absorption of colostrum iron. 14  healthy foals we expected fluctuations depending on foals' age; however, it was unexpected to observe a large variability of plasma iron concentrations in neonatal foals of a given age (ie, 24-48 h). In addition, significant differences between healthy and SIRS foals were expected in this study but were not observed. Perhaps, grouping foals according to age (ie, 1-3 days, 3-7 days, and <14 days) would have helped us to better assess plasma iron dynamics.
In the studied population of neonatal foals, there was no significant difference in mean plasma iron concentrations between healthy foals and sick foals (P = .99). In equine medicine, previous reports suggest that plasma iron concentrations in adult horses are an acute and sensitive indicator of systemic inflammation. 2,12 In adult horses plasma iron concentration is considered decreased when values fall below the reference interval (105-277 μg/dL). 2 We expected to find lower iron plasma concentrations in sick foals compared to healthy foals, as occurs in the equine adult population because of the iron withholding mechanism. In the setting of infectious, inflammatory, and neoplastic diseases, a primitive defense mechanism of the organism is to withhold iron from microorganisms. 20 The iron withholding defense system includes constitutive iron-binding components such as transferrin, lactoferrin, and ovotransferrin, as well as the suppression of iron efflux from macrophages, reduction in plasmatic iron, and increased synthesis of ferritin by macrophages to accommodate iron from phagocytised lactoferrin iron. 21 At baseline and in response to nutritional iron deficiency, infection, bleeding and pregnancy, hepcidin regulates iron metabolism. 22 Hepcidin is potently induced by inflammation, predominantly by the cytokine IL-6, and has shown to be essential for innate immunity to gram-negative bacteria. 22 This hormone controls iron flows into plasma through inhibition of the only known mammalian cellular iron exporter ferroportin. 23 Hepcidin is feedback-regulated by iron status and strongly modulated by inflammation and erythropoietic demand. 23 Studies of plasma iron concentrations in neonatal foals with SIRS are sparse or nonexistent. The SIRS criteria have been defined in veterinary medicine by different authors. 3,7,24,25  respectively) could at least partly explain the differences observed in plasma iron concentrations (Figure 2). To rule out the age effect in future studies, we suggest classifying foals according to age (ie, 1-3 days, 3-7 days, and <14 days), in order to differentiate plasma iron concentrations in SIRS and non-SIRS neonates as some other authors have done measuring age-dependent plasma biochemical variables in neonatal foals. 31 Another of the limitations of this study could be the absence of information regarding iron metabolism in the animals included, because, except for plasma iron concentration, no other variables related to iron metabolism were analyzed. Iron metabolism in equine medicine can be directly evaluated by measuring the amount of iron in the blood, including: plasma iron concentration, the capacity of the blood to transport iron by measuring transferrin or total iron binding capacity (TIBC) and the amount of iron storage measuring ferritin. 18 Another limitation of this study could be the lack of a full hematology data from the stud control group. All samples were taken from healthy foals based on: normal foaling, normal physical exam, and serum immunoglobulin G concentration above 800 mg/dL within 24-48 hours postpartum. Despite this limitation, we considered that a large homogeneous healthy group, with a daily close veterinary follow-up was still of a great value for the study.

| CONCLUSIONS AND CLINICAL RELEVANCE
In summary, plasma iron concentration has a negative moderate correlation with age in neonatal foals. Unlike what is reported in adult horses' plasma iron concentration is not a useful early inflammatory marker to predict SIRS in sick neonatal foals. Based on results of this study plasma iron concentration in healthy neonatal foals has a larger variability compared to adults. Finally, plasma iron concentration is not a useful prognostic marker in this population.

ACKNOWLEDGMENT
This study was partially presented at the 12th ECEIM Congress, Valencia, Spain, November 2019.