Serum cortisol and thyroid hormone concentrations and survival in foals born from mares with experimentally induced ascending placentitis.

Abstract Background There are few publications on occurrence of nonthyroidal illness syndrome in foals and on the prognostic value of cortisol and thyroid hormone (TH) concentrations in newborn foals. Objectives To determine serum cortisol and TH concentrations (total and free thyroxine: T4 and FT4; total and free triiodothyronine: T3 and FT3) in foals born from mares with placentitis, to determine their association with survival, and their use as prognostic markers. Animals A cohort of 29 newborn foals comprising 5 Control, 14 Low‐risk, and 10 Sick foals were evaluated over the first week of life. Methods In this prospective study foals born to mares with experimentally‐induced placentitis were assigned to Low‐risk or Sick groups while foals born to control mares were classified as Control based on clinical findings. Foals were also classified as Term (n = 13), Dysmature (n = 7), or Premature (n = 9), and survival rate was recorded. Serum cortisol and TH hormone concentrations were measured at 0, 12, 24, 48, and 168 hours of life. Results Sick non‐surviving foals had lower (P < .05) T3 : cortisol ratio at 12 (3.68 ± 1.06 versus 18.58 ± 2.78), 24 (5.47 ± 2.34 versus 23.40 ± 3.82), and 48 (10.47 ± 6.29 versus 26.6 ± 2.90) hours of life when compared to Sick surviving foals and lower (P < .05) T4 : cortisol ratio at 12 (75.12 ± 21.71 versus 414.47 ± 58.47) and 24 hours (127.83 ± 55.21 versus 430.87 ± 80.31) after birth than Sick surviving foals. Conclusion and Clinical Importance Placental infections can impair fetal thyroid function. Low T3 : cortisol and T4 : cortisol ratios seem to be good prognostic markers in newborn foals.


| INTRODUCTION
Placentitis is one of the most important causes of fetoplacental unit impairment and can affect equine fetal maturation and postnatal development. 1 Foals delivered by mares with placentitis can range from small sized foals with compromised organ function to full sized foals with no clinical and laboratory abnormalities, depending on both the time and the duration of the insult. 2 Immaturity and sepsis are the main complications of placentitis to newborn foals and can lead to dysfunction of different organ systems. 3,4 In human neonates, disorders such as sepsis suppress the hypothalamic-pituitary-thyroid axis (HPTA), reducing thyroxine (T 4 ), and triiodothyronine (T 3 ) concentrations, leading to a condition known as nonthyroidal illness syndrome (NTIS). 5,6 The reduction in thyroid hormone (TH) concentrations possibly represents a physiological adaption to reduce metabolic activity to allow energy reallocation to respond to disease. 5 Besides TH concentrations, T 3 : T 4 and F T 3 : F T 4 ratios have been used in human medicine to assess thyroid function and showed to be a useful aid in the diagnosis of NTIS as other thyroid dysfunctions. 7 In neonates of different species, TH are important for thermoregulation in the first hours of life. 8 In addition, these hormones act directly on cell metabolism and stimulation, which makes them essential for organs maturation and development. 8,9 Activation of the hypothalamic pituitary adrenal axis in the last weeks of pregnancy is essential for equine fetal tissue maturation and differentiation, but also for thyroid gland maturation and TH secretion. 10,11 The relation between adrenal and thyroid gland have been demonstrated in humans 12 and rats. 13 These studies suggested that excess of glucocorticoids can suppress the HPTA and that TH regulate adrenocortical function through changes in hepatic glucocorticoid metabolism. 12,13 Although the relation between adrenal and thyroid have not been deeply studied in foals, a relationship between cortisol and T 3 concentrations have been demonstrated in the perinatal period. 10 However, TH : cortisol ratios were not assessed before and could have clinical value.
There are few publications on NTIS in neonatal foals. [14][15][16][17] Furthermore, the information available is generally from hospital admissions, making it difficult to compare TH concentrations in sick and healthy age-matched foals. Two studies demonstrated that sick foals have lower serum TH concentrations than healthy foals, suggesting that NTIS is prevalent in critically ill equine neonates. These studies also found that serum TH concentrations were lower in nonsurviving foals. 14,17 The aim of our study was to determine serum cortisol and TH concentrations (total thyroxine, T 4 ; free thyroxine, F T 4 ; total triiodothyronine, T 3 ; and free triiodothyronine, F T 3 ) in foals born from mares with experimentally-induced ascending placentitis and to determine their association with survival. We hypothesized that (1) foals born to mares with experimentally induced ascending placentitis will have decreased serum cortisol and TH concentrations over the first week of life, (2) that these endocrine alterations will be associated with death, and (3) that cortisol and TH concentrations in the first hours of life will have prognostic value for survival.

| Animals
This observational cohort study was performed during the 2012-2014 breeding seasons in Southern Brazil. Twenty-nine foals were enrolled in our study from the time of birth to 7 days postpartum. These foals were also included in a parallel study, in which 46 foals were enrolled. 18 However, only foals which serum samples were available were included in the present study. Five dams carried healthy gestations to term, while 24 dams were submitted to experimental induction of ascending placentitis through intracervical inoculation of 10 7 colonyforming units of Streptococcus equi subsp. zooepidemicus at 300 days of gestation, as previously described. 18 Mares were assessed daily for the presence of vulvar discharge, early udder development and measurement of the combined thickness of uterus and placenta (CTUP), as described elsewhere. 19 Of these mares, 100% (24/24) showed vulvar discharge,

| Clinical assessment of the foals
All foalings were closely monitored. Foals had gestation duration recorded in days and were submitted to physical exams, evaluation of righting reflexes, total blood count and measurement of creatinine, urea, glucose and blood lactate immediately after birth. Foals born to mares with normal gestations showed normal adaptive behavior (sternal recumbency < 5 minutes, suckling reflex < 20 minutes, and stand < 1 hour), no evidence of immaturity, appropriate serum immunoglobulin G concentrations (IgG > 8.0 g/L), normal physical exams and total blood counts 20 during the first week of life and were assigned to the Control group (n = 5). Foals delivered by mares submitted to experimental induction of ascending placentitis were classified as Low-risk or Sick at 12 hours after birth, when the sepsis scores were calculated. 21 Foals with normal adaptive behavior, physical exams, serum IgG concentrations > 800 mg/dL, unremarkable CBCs, and sepsis score ≤ 10 were assigned to Low-risk group (n = 14), while foals that needed assistance to nurse and showed abnormal adaptive behavior, physical exams, CBCs, sepsis scores ≥ 11, or all were assigned to the Sick group (n = 10). In Sick group, 7 foals were septic while 3 foals had sepsis score ≤ 10 but were considered sick because of, abnormal adaptive behavior and alterations on clinical parameters, CBCs, and biochemical results.
Foals delivered with ≥320 days of gestation and showing no physical signs of immaturity were classified as Term (n = 13). Foals born with ≥320 days of gestation and showing physical signs consistent with immaturity (small body size, short and shiny hair coat, prominent rounded head, periarticular laxity, droopy ears and weakness) were classified as Dysmature (n = 7), while those delivered with <320 days of gestation and showing physical signs consistent with immaturity were classified as Premature (n = 9), according to the classification described elsewhere. 3 In the Low-risk group, 1 foal was considered premature and 6 were considered dysmature. All foals in the Sick group had physical findings consistent with immaturity. Eight were considered premature and 2 dysmature. Six foals from the sick group did not survive. Two of them had to be euthanized due the grave prognosis.

| Hormonal analysis
Foals had blood collected from the external jugular vein immediately after birth (0 hours), and at 12, 24, 48 hours and 7 days (168 hours) postpartum. Samples were allowed to clot at room temperature for 20 minutes, centrifuged at 600g for 10 minutes and immediately stored at −20 C until further analysis. Serum concentrations of total thyroxine (T 4 ), free thyroxine ( F T 4 ), total triiodothyronine (T 3 ), and free triiodothyronine ( F T 3 ) were determined by enzyme-linked immunosorbent assay (MP Biomedicals LLC, New York), whereas cortisol was determined by radioimmunoassay (MP Biomedicals LLC). The sensitivity and specificity of the kits used are described in Table 1.

| Statistical analysis
All variables were tested for normality with the Shapiro-Wilk test.
Hormone concentrations and T 3 : T 4 , F T 3 : F T 4 , T 3 : cortisol and T 4 : cortisol ratios were compared among age-matched foals (at birth and

Serum T 3 and F T 3 concentrations were lower in Sick compared to
Control and Low-risk foals at 0, 12, 24, and 48 hours after birth Serum T 3 , F T 3 , T 4 , F T 4 , and cortisol concentrations were not different between surviving and nonsurviving foals ( Note: Results expressed as mean ± SE. Statistical differences were not detected among groups (P > .05). Abbreviations: T 3 , total triiodothyronine; F T 3 , free triiodothyronine; T 4 , total thyroxine; F T 4 , free thyroxine.
F I G U R E 4 T 3 : cortisol, A, and T 4 : cortisol, B, ratios in surviving (n = 4) and nonsurviving (n = 6) Sick foals. Statistical difference (P < .05) among groups is represented by different letters at each time point. T 3 , total triiodothyronine; T 4 , total thyroxine Sick nonsurviving foals had lower T 3 : cortisol ratio from 12 to 48 hours of life and lower T 4 : cortisol ratio at 12 and 24 hours after birth than Sick surviving foals (Figure 4; P = .031). The AUC for TH was not different between Sick surviving and Sick nonsurviving foals (P > .05).

| DISCUSSION
In our study, sick foals born from mares with experimentally induced placentitis had lower concentrations of T 3 , F T 3 , and F T 4 than healthy control and Low-risk foals at birth and during the first week of life, suggesting that these foals developed NTIS in response to intrauterine disease. Similar results were found in that sick foals up to 6 hours of life had lower concentrations of T 4 and T 3 than healthy foals. 15 The mechanism by which sick foals delivered by mares with pla-

tion. A correlation between T 3 and cortisol concentrations has been
documented in equine fetuses, suggesting that increased adrenocortical activity and subsequent cortisol release few days before birth promotes thyroid gland maturation and TH secretion. 10 In the same study, premature foals had lower concentrations of cortisol and T 3 than term foals, whereas dysmature foals had lower concentrations of T 3 than term foals, but higher than premature foals. 10 These findings were similar to the ones from the study reported here.
Sick foals from the present study had lower TH serum concentrations at birth and likely in late gestation, supporting impaired TH synthesis from placental and fetal disease, which is consistent with NTIS.  24 There is evidence that the tumor necrosis factor alpha (TNF-α), IL-1, IL-6, and IL-10 have negative effects on thyroid-stimulating hormone (TSH) release, on the expression of deiodinases that convert T 4 to T 3, and decrease TH binding to serum proteins. 25 In addition, inflammatory cytokines inhibit genes involved in the metabolism of TH. 26,27 Although inflammatory cytokines were not assessed in mares and foals from our study, they might have contributed to the reduction of TH in sick foals. The fact that some foals have experienced an inflammatory process in uterus, but were considered Low-risk foals and had normal TH levels lead us to believe that some mares had a mild inflammatory response to infection or responded more efficiently to treatment.
In the present study, we found no association between serum hormone concentrations and foal survival. Similar results were found in another study with foals affected with perinatal asphyxia syndrome in which TH concentrations were not associated with survival. 16 Other foal studies have found association between low TH concentrations and death. 14,17 In a study in which a single blood sample was collected to assess TH concentration in foals that were between 1 and 192 hours old, septic nonsurviving foals had lower levels of T 3 , F T 3 , and F T 4 than septic surviving foals. 14   TH serum concentrations were not associated with neonatal death in our study. However, T 3 : cortisol and T 4 : cortisol ratios seem to be good prognostic markers of disease severity in the first hours of life.

ACKNOWLEDGMENTS
The authors thank the Federal University of Pelotas and The Ohio State University for supporting our study.

CONFLICT OF INTEREST DECLARATION
Authors declare no conflict of interest.

OFF-LABEL ANTIMICROBIAL USE DECLARATION
Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) DECLARATION
All procedures carried out in the present study were approved by the Ethical Committee on Animal Experimentation of the Federal University of Pelotas under protocol #4750.