Renin-Angiotensin-Aldosterone System and Hypothalamic-Pituitary-Adrenal Axis in Hospitalized Newborn Foals
Results of this work were partially presented as an oral abstract at the 2012 American College of Veterinary Internal Medicine (ACVIM) Forum, New Orleans, LA.
Corresponding author: Dr R.E. Toribio, 601 Vernon L. Tharp Street, College of Veterinary Medicine, The Ohio State University, Columbus, OH 43210; e-mail: firstname.lastname@example.org.
The renin-angiotensin-aldosterone system (RAAS) and hypothalamic-pituitary-adrenal axis (HPAA) and their interactions during illness and hypoperfusion are important to maintain organ function. HPAA dysfunction and relative adrenal insufficiency (RAI) are common in septic foals. Information is lacking on the RAAS and mineralocorticoid response in the context of RAI in newborn sick foals.
To investigate the RAAS, as well as HPAA factors that interact with the RAAS, in hospitalized foals, and to determine their association with clinical findings. We hypothesized that critical illness in newborn foals results in RAAS activation, and that inappropriately low aldosterone concentrations are part of the RAI syndrome of critically ill foals.
A total of 167 foals ≤3 days of age: 133 hospitalized (74 septic, 59 sick nonseptic) and 34 healthy foals.
Prospective, multicenter, cross-sectional study. Blood samples were collected on admission. Plasma renin activity (PRA) and angiotensin-II (ANG-II), aldosterone, ACTH, and cortisol concentrations were measured in all foals.
ANG-II, aldosterone, ACTH, and cortisol concentrations as well as ACTH/aldosterone and ACTH/cortisol ratios were higher in septic foals compared with healthy foals (P < .05). No difference in PRA between groups was found. High serum potassium and low serum chloride concentrations were associated with hyperaldosteronemia in septic foals.
Conclusions and Clinical Importance
RAAS activation in critically ill foals is characterized by increased ANG-II and aldosterone concentrations. Inappropriately low cortisol and aldosterone concentrations defined as high ACTH/cortisol and ACTH/aldosterone ratios in septic foals suggest that RAI is not restricted to the zona fasciculata in critically ill newborn foals.
critical illness-related corticosteroid insufficiency
plasma renin activity
relative adrenal insufficiency
Septicemia, defined as a systemic inflammatory response to pathogenic microorganisms or their toxins, is the leading cause of foal mortality in the 1st week of life.[1-3] Critically ill foals often present to intensive care units with hypotension, tissue hypoperfusion, energy dysregulation, acid-base and electrolyte abnormalities, and organ failure.[1, 2, 4-7] The survival rate under these circumstances is dictated by the ability of various homeostatic systems to control and overcome these derangements as well as the host response to adequate medical therapy. Decreased fluid intake, fluid losses, inflammation, vasoplegia (low systemic vascular resistance), and myocardial impairment are the main factors leading to hypotension and hypoperfusion in septic foals.[2, 5, 6, 8, 9]
The hypothalamic-pituitary-adrenal axis (HPAA) and the renin-angiotensin-aldosterone system (RAAS) are both activated in response to sepsis-related stress in humans. HPAA activation leads to an increase in ACTH and cortisol concentrations in critically ill foals.[1, 10, 11] Recent studies have shown that activation of the HPAA is directly associated with the severity of sepsis and mortality in septic foals.[1, 10-12] Although usually presented as separated systems, in reality, the HPAA and RAAS are highly interactive, sharing a number of endocrine factors, including ACTH, vasopressin, angiotensin-II (ANG-II), and aldosterone.
Previous studies have shown that the aldosterone response to hypovolemia is different in neonatal foals compared with adult horses, likely because of differences in neonatal sodium and water regulation.[5, 8] Healthy human and equine neonates have higher aldosterone and renin concentrations than adults, which may reflect a relative insensitivity of the distal renal tubules to aldosterone.[5, 13, 14] RAAS dysfunction during sepsis has received extensive attention in critically ill humans.[15-18] Activation of the RAAS, characterized by increased plasma renin activity (PRA), ANG-II, and aldosterone concentrations, has been linked with multiple organ failure and high mortality in septic human patients.[15, 19] However, limited information is available in hospitalized foals.
A major complication in critically ill children is relative adrenal insufficiency (RAI) or critical illness-related corticosteroid insufficiency (CIRCI), in which impaired secretion of a number of adrenal corticosteroids contributes to mortality.[19, 20] RAI, defined as an inappropriately low basal cortisol concentration or inadequate production of cortisol in relation to increased ACTH concentration, is common in septic and premature foals.[1, 10, 11] Although there is evidence that sepsis affects cortisol secretion in foals,[1, 10, 11, 21, 22] it is unlikely that CIRCI in critically ill foals is limited to glucocorticoid production. Thus, the aldosterone response to severe illness in foals remains to be evaluated.
The aim of this study was to investigate the RAAS response as well as factors of the HPAA that may interact with the RAAS (eg, ACTH, cortisol) in healthy and hospitalized (septic and sick nonseptic) newborn foals, and to determine their association with clinical findings. We hypothesized that RAAS activation, characterized by increased PRA, ANG-II, and aldosterone concentrations, is associated with severity of disease and higher mortality in septic foals. We also proposed that septic foals will have inappropriately low aldosterone concentrations (based on ACTH, ANG-II, hormone ratios, and electrolyte concentrations), indicating that mineralocorticoid deficiency is part of the RAI (CIRCI) that occurs in critically ill newborn foals.
Materials and Methods
Foals ≤3 days old of any breed and either sex admitted to The Ohio State University Galbreath Equine Center, Hagyard Equine Medicine Institute, and Rood & Riddle Equine Hospital from 1 foaling season were included. Hospitalized foals were classified into 2 groups: septic foals and sick nonseptic (SNS) foals. Foals in the septic group had a sepsis score of ≥12, a positive blood culture, or both. Foals in the SNS group were presented for illnesses other than sepsis (eg, hypoxic ischemic encephalopathy, meconium impaction, failure of transfer of passive immunity, orthopedic conditions) requiring hospitalization. These foals had negative blood cultures and a sepsis score of ≤11. The control group consisted of 24-hour-old foals, classified as healthy based on physical examination, normal CBC, serum biochemistry profile, serum immunoglobulin G (IgG) concentrations (>800 mg/dL), and sepsis score of 0–4. Foals with a history of receiving any medications (eg, glucocorticoids, IV crystalloid fluids, plasma) before admission were excluded from the study. Foals with uroabdomen also were excluded from the study.
Inappropriately low aldosterone concentration was defined as an increased ACTH/aldosterone ratio in septic compared with healthy foals and a negative correlation between aldosterone concentrations and markers of hypoperfusion (L-lactate concentration) in septic foals.
Survival was defined as having been discharged alive from the hospital. Foals that died or were euthanized because of a grave medical prognosis were defined as nonsurvivors. Foals euthanized for other reasons such as financial constraints were excluded from the study.
This study was approved by the OSU Veterinary Clinical Trials Office, the Institutional Animal Care and Use Committee and adhered to the principles of humane treatment of animals in veterinary clinical investigations, as stated by the American College of Veterinary Internal Medicine and National Institute of Health guidelines.
Clinical history obtained upon presentation included expected foaling date, duration of pregnancy, parity of the mare, maternal illnesses, premature lactation, observed or assisted parturition, dystocia, passing and appearance of fetal membranes, and medications (mare and foal). Clinical data collected from the foal included signalment (sex, age, breed), physical examination findings, CBC, biochemistry profile, plasma fibrinogen concentrations, IgG concentrations, and blood culture results. Endocrine measurements included PRA, ANG-II, ACTH, aldosterone, and cortisol concentrations. For consistency, the sepsis score was calculated by one of the investigators (KAD) for each foal, based on history, physical examination, and laboratory findings.
Blood samples for hormone measurements from hospitalized foals were obtained on admission (within 1 hour of hospitalization) by sterile jugular venous catheterization. Samples from healthy foals were obtained during routine examination of newborn foals and stored at 4°C until processing. Blood was collected in serum clot and aprotinin-EDTA tubes. Aprotinin was added to inhibit protease-mediated degradation of hormones (500 kU/mL of blood). Samples were centrifuged at 2,000 × g for 10 minutes at 4°C. Serum and plasma were aliquoted and stored at −80°C within 12 hours after collection for up to 6 months until analyzed. Blood samples for CBC, serum biochemistry, fibrinogen, and IgG concentrations were processed within 2 hours after collection.
Serum aldosterone and cortisol concentrations were determined using human-specific radioimmunoassays1,2 validated for horses.[1, 23, 24] Plasma ACTH concentrations were determined with an immunochemiluminometric assay3 validated for horses. Plasma ANG-II concentrations were measured with a multi-species ANG-II radioimmunoassay,4 after peptide extraction using phenylsilylsilica solid-phase columns.5 Intra- and interassay coefficients of variation for this assay using equine samples were 8.3 and 10.7%, respectively. Parallelism for equine samples diluted 1 : 2–1 : 16 ranged from 84 to 103%.
Determination of PRA was based on generation of angiotensin-I (ANG-I) at 4°C and 37°C for 1 h, measured with a radioimmunoassay6 and expressed as ng/mL/h of generated ANG-I. Intra- and interassay coefficients of variation for this assay, using equine samples, were 7.6 and 9.4%, respectively.
Data sets were tested for normality by the Shapiro-Wilk statistic. ACTH and cortisol concentrations in healthy foals and aldosterone concentrations in all 3 groups of foals were normally distributed. The remainder of the data was not normally distributed. Medians and ranges were calculated for continuous variables. Comparisons between groups of foals were carried out with the Kruskal–Wallis statistic, and the Dunn's posttest was used to compare each group individually. The Mann–Whitney-U-test was used to compare survivors and nonsurvivors within each group. Significance was set at P < .05. Correlations were determined with the Spearman rank order (rho). Continuous variables were categorized by 2 cut-off values into tertiles based on distribution within a group, and analyzed using logistic regression for binomial distribution. Clinicopathologic data analyzed included sodium, potassium, chloride, L-lactate, IgG, and hormone concentrations. Crude odds ratios and 95% confidence intervals were calculated and based on categories. The dependent variables were survival and nonsurvival. All variables were screened and those with a P < .25 were tested by forward stepwise multivariate logistic regression to determine a final model. Variables that resulted in P value <.05 were retained in the model. Data were analyzed with statistical software.7,8
A total of 167 neonatal foals (133 hospitalized; 34 healthy foals) ≤3 days of age were included. The median age of all hospitalized foals at admission was 10 hours (septic foals, 12 hours; SNS foals, 10 hours). For healthy controls, the median age was 24 hours. Age was not significantly different between septic surviving and nonsurviving foals (P = .12) or between healthy and hospitalized foals (P = .207).
Fifty-six percent of hospitalized foals (74/133) were classified as septic and 44% (59/133) as SNS. The survival rate in septic foals was 55% (41/74) and in SNS was 88% (52/59). Forty-one percent (30/74) of septic foals had positive blood cultures. The median sepsis score for all hospitalized foals was 8, and for septic foals was 14.
Breeds represented by hospitalized foals included Thoroughbred (n = 92), Quarter Horse (10), Standardbred (12), Appaloosa (6), Saddlebred (2), Warmblood (3), Friesian (1), American Paint Horse (2), Arabian (2), Belgian (2), and Percheron (1). All healthy foals were Thoroughbreds (34).
Of the hospitalized foals, 54% (72/133) were colts and 46% (61/133) fillies. In the healthy foals, 47% (16/34) were colts and 53% (18/34) were fillies, whereas in the septic and SNS foals, 52% (39/74) and 56% (33/59) were colts and 48% (35/74) and 44% (26/59) were fillies, respectively.
Hormone Concentrations, White Blood Count (WBC), Total Protein, IgG and Sepsis Score
Results of PRA, ANG-II, aldosterone, cortisol and ACTH concentrations, sepsis score, WBC, total protein and IgG for all foals are presented in Table 1. Septic foals had significantly higher ANG-II, aldosterone, cortisol, and ACTH concentrations compared with healthy foals (P < .05). ANG-II and ACTH concentrations were higher in septic than in SNS foals (P < .05). Aldosterone concentrations were not different between SNS and healthy foals. SNS foals had higher cortisol concentrations than healthy foals (P < .05). There were no statistical differences in PRA between foal groups. There were no statistical differences in hormone concentrations between fillies and colts among foal groups (P = .26).
Table 1. Blood hormone and electrolyte concentrations and hormone ratios in neonatal foals (values expressed as median and range)
|PRA (ng/mL/h)||6.82 (1–20)||6.52 (1–15)||5.89 (3–16)|
|ANG-II (pg/mL)||31.99 (3–863)*§||13.42 (2–291)||28.55 (10–83)|
|ALD (pg/mL)||368 (25–1515)*||254.67 (38–1343)||202.55 (55–678)|
|Cortisol (μg/dL)||13.68 (1–74)*||10.7 (1–48)*||4.7 (1–14)|
|ACTH (pg/mL)||176 (10–1250)*§||37 (11–647)||21.4 (10–40)|
|BUN (mg/dL)||22 (9–98)**||19 (4–69)||18 (12–31)|
|Creatinine (mg/dL)||3.1 (1–21)**§||2.3 (1–22)||1.85 (1–3)|
|Na+ (mEq/L)||138 (128–155)||137 (107–147)||137.5 (132–143)|
|K+ (mEq/L)||4.1 (2–6)||3.9 (2–7)||4.1 (3–5)|
|Cl− (mEq/L)||96 (76–107)**||97.9 (81–105)*||99 (92–105)|
|L-lactate (mmol/L)||5.7 (1.3–18.9)**§||4.4 (1.3–17.9)||2.1 (0.8–4.8)|
|ACTH/cortisol||11 (1–208.5)*§||5.94 (1–30.9)||5.08 (1.23–22.35)|
|ACTH/ALD||0.4 (0.02–3.7)*||0.12 (0.01–2.45)||0.09 (0.02–0.42)|
|ALD/K+||187.39 (29.34–326.57)**||161.64 (35.7–319)||60 (19.22–188)|
|ALD/Cl−||8.05 (0.8–17.97)**||6.2 (1.1–16.1)||2.27 (0.85–6.98)|
|ALD/Na+||5.43 (0.6–11.56)*||4.34 (0.8–10.92)||1.63 (0.63–4.97)|
|Total protein (g/dL)||4.4 (3.2–7.7)**||4.45 (3.5–6.8)*||5.05 (4.2–6.7)|
|IgG (mg/dL)||420 (0–1323)**§||760 (0–2420)||1200 (450–1200)|
|Albumin (g/dL)||2.9 (2.2–3.9)||2.85 (1.9–3.8)||2.9 (2.6–3.2)|
|Fibrinogen (mg/dL)||245 (100–800)||276 (100–600)||300 (200–700)|
|PCV (%)||38.9 (22–59)||39 (23–48)||39.2 (22–46)|
|WBC (×103/μL)||4.3 (0.6–22.6)**§||8.4 (0.9–20.6)||9.6 (0.8–15)|
|Segmented neutrophils (×103/μL)||2.3 (0.08–71)||6 (0.05–89)||7.1 (3.1–12.2)|
|Band neutrophils (×103/μL)||0.3 (0.01–32)||0.2 (0.01–7.1)||0.3 (0.2–9)|
|Lymphocytes (×103/μL)||1.1 (0.1–75)||1.3 (0.1–30)||1.8 (0.1–2.2)|
|Sepsis score||14 (11–21)**§||7 (4–11)**||2 (0–4)|
WBC and total protein concentrations were lower in septic compared with healthy foals (P < .01), and SNS foals had lower total protein concentrations compared with healthy foals (P < .05). Septic foals had lower IgG concentrations compared with healthy and SNS foals (P < .01). Septic foals that died had lower cortisol concentrations than survivors (P < .05), but aldosterone concentrations were not different between surviving and nonsurviving septic foals (P = .31) (Table 2).
Table 2. Blood hormone concentrations and hormone ratios in surviving and nonsurviving septic foals (values expressed as median and range)
|PRA (ng/mL/h)||10 (8.17–19.69)||7.2 (4.3–8.8)|
|ANG-II (pg/mL)||123.31 (18.15–862.55)||23.8 (9.81–44.06)|
|ALD (pg/mL)||338.3 (125–1515)||385 (144.3–1401)|
|Cortisol (μg/dL)||21.7 (6.9–36.7)||5.02 (1.97–6.63)*|
|ACTH (pg/mL)||97.8 (17.5–718)||147 (26.4–267.4)|
|ACTH/ALD||0.26 (0.02–0.59)||0.34 (0.08–1.95)|
|ACTH/cortisol||3.87 (2.23–19.52)||28.58 (3.98–53.18)*|
In the hospitalized population, foals with aldosterone concentrations in the range of 25–186.8 pg/mL were more likely to survive (OR = 2.7) than foals with aldosterone concentrations >428.4 pg/mL (Table 3). Foals with ACTH concentrations in the range of 10–26.6 pg/mL and 26.7–152 pg/mL were 5.4 and 2.7 times more likely to survive than those with ACTH concentrations >153 pg/mL, respectively (Table 3).
Table 3. Hormone univariate analysis for survival among neonatal foals
|ALD (pg/mL)||25 – 186.8||2.7*||1.01 – 7.42|
|186.9 – 428.3||1.5||0.61 – 3.84|
|428.4 – 1515||Referent|| |
|Cortisol (μg/dL)||1 – 5.34||0.68||0.24 – 1.9|
|5.35 – 14.03||0.21||0.21 – 1.63|
|14.03 – 74||Referent|| |
|ACTH (pg/mL)||10 – 26.6||5.4*||1.7 – 16.8|
|26.7 – 152||2.7*||1.03 – 7.13|
|153 – 1250||Referent|| |
|ANG-II (pg/mL)||2.1 – 12.5||0.5||0.18 – 1.57|
|12.6 – 38.4||1.12||0.31 – 3.19|
|38.5 – 863||Referent|| |
|PRA (ng/mL/h)||0.1 – 4.0||0.7||0.1 – 3.5|
|4.1 – 8.0||1.4||0.2 – 10.0|
|8.1 – 20||Referent|| |
|ACTH/cortisol||0.2 – 4.14||6.7*||1.9 – 23.14|
|4.15 – 11.0||5.2*||1.64 – 16.57|
|11.1 – 112.8||Referent|| |
|ACTH/ALD||0.01 – 0.16||3.06*||1.08 – 8.68|
|0.17 – 0.65||1.87||0.5 – 6.01|
|0.66 – 2.62||Referent|| |
|ALD/K+||8.1 – 31.94||2.8*||1.05 – 7.5|
|31.95 – 84.81||1.02||0.45 – 2.33|
|84.82 – 351.86||Referent|| |
|ALD/Cl−||0.25 – 1.94||3.02*||1.09 – 8.33|
|1.95 – 4.6||1.3||0.76 – 4.9|
|4.7 – 19.17||Referent|| |
|ALD/Na+||0.17 – 1.39||3.2*||1.32 – 9.48|
|1.4 – 3.16||1.55||0.61 – 3.9|
|3.17 – 11.56||Referent|| |
|Na+ (mEq/L)||107 – 135||0.68||0.28 – 1.66|
|136 – 139||0.95||0.37 – 2.44|
|139 – 155||Referent|| |
|Cl− (mEq/L)||76 – 95||0.9||0.4 – 2.23|
|96 – 99||1.01||0.4 – 2.49|
|100 – 107||Referent|| |
|K+ (mEq/L)||2.1 – 3.6||8.8*||2.5 – 30.9|
|3.6 – 4.5||2.4*||1.01 – 6.27|
|4.6 – 7||Referent|| |
|Sepsis score||0 – 6||81*||9.8 – 679.0|
|7 – 14||5.4*||2.2 – 13.3|
|15 – 21||Referent|| |
|L-lactate (mmol/L)||0.8 – 2.8||42.22**||5.24 – 339.7|
|2.9 – 4.9||4.76*||1.68 – 13.477|
|5.0 – 18.9||Referent|| |
|Creatinine (mg/dL)||1 – 1.9||3.16*||1.27 – 7.87|
|2 – 3.3||2.15||0.89 – 5.1|
|3.4 – 22||Referent|| |
|BUN (mg/dL)||4 – 18||2.5||0.92 – 7.11|
|19 – 24||0.85||0.36 – 2.01|
|25 – 98||Referent|| |
Serum sodium and potassium concentrations were not different between foal groups, but septic foals had lower chloride concentrations than healthy foals (Table 1). Of interest, hospitalized foals with potassium concentrations of 2.1–3.6 mEq/L and 3.7–4.5 mEq/L were more likely to survive (OR = 8.8 and OR = 2.4) than foals with potassium concentrations >4.6 mEq/L, respectively (Table 3). Foals with sepsis scores of 0–6 and 7–14 were 81 and 5.4 times more likely to survive than those with sepsis scores >15, respectively (Table 3).
Hormone and Electrolyte (K+, Na+, Cl−) Ratios and Correlations
Hormone ratios were determined for each foal group (Tables 1–2). Aldosterone/electrolyte ratios were higher in septic foals compared with healthy and SNS foals (P < .05) (Table 1). Septic foals had significantly higher ACTH/aldosterone and ACTH/cortisol ratios than healthy foals (P < .05). The ACTH/cortisol ratio also was higher in septic than SNS foals (P < .05), and higher in nonsurviving than in surviving septic foals (P < .05) (Table 2). In septic foals, aldosterone concentrations were positively correlated with ANG-II, ACTH, and cortisol concentrations (Table 4).
Table 4. Spearman correlations (rho) between hormones, serum electrolytes, IgG, creatinine and BUN concentrations, and sepsis score in septic foals
|ALD|| || ||0.43**||0.32*||0.07||0.41**||−0.10||−0.33*||−0.41*||0.15||0.33*||0.30*|
|Cortisol|| || || ||0.28*||0.49**||−0.09||0.27||−0.17||−0.03||−0.08||0.01||0.15|
|ACTH|| || || || ||0.29*||0.27||0.26||0.26*||−0.39*||0.20||0.12||0.11|
Association of Hormones with Electrolytes, Creatinine, L-Lactate, BUN, IgG, and Sepsis Score in Septic Foals
In septic foals, aldosterone was positively correlated with potassium, creatinine, and BUN concentrations (eg, an increase in aldosterone concentration was associated with an increase in creatinine concentration), and negatively correlated with IgG and L-lactate concentrations (Table 4). ACTH and cortisol concentrations were positively correlated with sodium concentration in septic foals. ANG-II and ACTH concentrations were negatively correlated with IgG concentrations in septic foals. ACTH was positively correlated with L-lactate concentrations. ANG-II concentrations were positively correlated with creatinine concentrations in septic foals (Table 4). No correlations between PRA and electrolytes, L-lactate, sepsis score and IgG were found. We found no association between hormones and electrolyte, creatinine and BUN concentrations in healthy foals.
The final model is presented in Table 5 (P = .001). The Hosmer and Lemeshow Goodness-of-Fit test indicated that the data fitted the model (P = .58). The model includes 2 variables: sepsis score and potassium concentrations. Foal survival was associated with low sepsis score (<12) and lower serum potassium concentrations (<4 mEq/L).
Table 5. Multivariate logistic regression analysis for survival among neonatal foals (final model)
|>12||Referent|| || |
|4.1–7||Referent|| || |
In the study reported here, we documented that during critical illness in newborn foals, the RAAS and HPAA are activated. Septic foals had higher ANG-II, aldosterone, cortisol, and ACTH concentrations compared with SNS and healthy foals. We also propose that RAI, in addition to a high ACTH/cortisol ratio, is characterized by inappropriately low aldosterone concentrations (based on a high ACTH/aldosterone ratio). This poor glucocorticoid and mineralocorticoid response is consistent with CIRCI, as documented in critically ill children.[19, 26]
Sepsis is a stressful event that activates the HPAA and RAAS to maintain organ function by modulating a multitude of homeostatic processes, including tissue perfusion, immune defense mechanisms, inflammation, and cellular metabolism.[16, 27-29] Activation of the HPAA, characterized by increased ACTH and cortisol concentrations, is well documented in sick foals.[1, 10, 22, 30] RAI has been associated with mortality in septic human patients.[31-33] Similarly, RAI seems to be common in septic and premature foals.[1, 10, 11, 30, 33] The reported prevalence of RAI is up to 50% in septic foals and 60% in septic human patients.[33, 34]
In the study reported here, basal ACTH, aldosterone, cortisol, ANG-II concentrations, and ACTH/cortisol and ACTH/aldosterone ratios were increased in septic foals. Similar to recent studies in foals,[1, 11] a high ACTH/cortisol ratio was associated with mortality in the foals of this study. Of interest, septic foals also had high ACTH/aldosterone ratios, indicating that RAI is a complex process that affects more than one layer of the adrenal cortex. Other explanations for the high ACTH/aldosterone ratio in the septic foals of this study include excessive ACTH secretion from stress (ie, hypotension, hypovolemia, endotoxemia, proinflammatory cytokines),[1, 12, 22] adrenocortical resistance to ACTH, and pituitary refractoriness to cortisol. RAI is likely to enhance ACTH release because of a lack of cortisol-mediated suppression of the HPAA. In a recent study, ACTH concentrations were positively correlated with IL-6 mRNA expression in septic foals.
There is no reason to believe that adrenal gland insufficiency is limited to the zona fasciculata because the adrenocortical layers share circulation and innervation, and ACTH stimulates both the zona glomerulosa and the zona fasciculata.[15, 16] Therefore, systemic inflammation, cytokines, poor perfusion, and hypotension may affect both glucocorticoid and mineralocorticoid production.[15, 16, 34, 36] ACTH deficiency leading to severe hyponatremia and hypoaldosteronism in people further supports the importance of ACTH in mineralocorticoid secretion. Cortisol and aldosterone are released in response to ACTH stimulation in cats and dogs.[38, 39] Of interest, exogenous ACTH failed to increase aldosterone secretion in healthy adult horses. Aldosterone secretion is controlled by ANG-II, ACTH, sodium, and potassium concentrations. ANG-II and potassium are considered the main regulators of aldosterone secretion.[36, 41] Although ACTH has little effect in controlling the rate of aldosterone secretion, it is necessary for aldosterone synthesis. Thus, adrenal resistance to ACTH can lead to mineralocorticoid deficiency. In fact, mineralocorticoid secretory activity in humans is assessed by using the ACTH (cosyntropin) stimulation test. Some work also indicates that in humans aldosterone secretion may be more sensitive than cortisol or dehydroepiandrosterone to ACTH stimulation. To our knowledge, the ACTH/aldosterone ratio and aldosterone response to ACTH stimulation test have not been evaluated in foals. Preliminary work from our group indicates that the administration of synthetic ACTH (cosyntropin) to healthy foals causes a modest release of aldosterone (R.E. Toribio, unpublished data, 2012).
Dissociation of aldosterone and renin (hyperreninemic hypoaldosteronism), characterized by an inappropriately low aldosterone/renin ratio (mineralocorticoid deficiency), has been associated with mortality in critically ill humans (adults and children).[18, 26, 44] However, we found no differences in PRA and PRA/hormone ratios between foal groups. The lack of differences in PRA between groups could be explained with a poor renin response to sepsis and hypotension in neonatal foals, or it could be analytical. Unlike other endocrine factors, it is recommended that PRA be determined using fresh samples (impractical for the type of study reported here), because it is a dynamic test and factors such as sample collection, storage, temperature changes, and processing can alter enzymatic activity. Although not designed to assess plasma renin activity, a direct renin assay that measures enzyme concentrations may have been a better method to determine circulating renin in stored equine plasma samples.[45, 46]
Low cortisol and aldosterone concentrations during illness are consistent with CIRCI and represent the basis for hydrocortisone and fludrocortisone replacement therapy in children diagnosed with this condition.[20, 29, 47] Adding fludrocortisone to hydrocortisone therapy in septic patients has been shown to decrease the need for vasopressor use. Furthermore, an inappropriate aldosterone response to sepsis could have contributed to some of the electrolyte abnormalities (eg, hyponatremia, hypochloremia, hyperkalemia) in critically ill foals.
In the present study, low serum chloride and high potassium, but not low sodium, concentrations were associated with increased aldosterone concentrations. It is possible that in critically ill foals renal tubular chloride concentrations may be more important than sodium to activate the RAAS. Low chloride concentrations in the renal distal tubules induce renin release in other species. Because we found no differences in PRA, it is also possible that direct adrenal activation by potassium at this stage of life may be important in foals. We assume that multiple factors, including hypotension, ANG-II, ACTH, chloride, and potassium concentrations, led to RAAS activation in hospitalized foals.
Multivariate analysis was performed using a number of clinical and laboratory factors related to sepsis and HPAA and RAAS regulation in hospitalized foals. Only two of these factors were retained in the final model: sepsis score and potassium concentrations. Low sepsis score and low potassium concentrations were associated with higher odds for survival, which is similar to what has been observed in other studies.[1, 48] This finding suggests that serum potassium concentrations should be taken into consideration in establishing a prognosis in critically ill foals. Causes of sepsis-associated hyperkalemia include metabolic and tissue acidosis, cell necrosis, renal hypoperfusion, insulin resistance, and hypoaldosteronemia.
Limitations of this study include differences in the median age between hospitalized (10 hours) and healthy foals (24 hours) (P = .207), and lack of blood pressure measurements in hospitalized foals. Using age- and sex-matched controls may be way to further evaluate the HPAA and RAAS in the future studies. Although we did not determine blood pressure in the foals of this study, we demonstrated that hyperlactatemia (an indirect indicator of hypoperfusion and hypotension[2, 6, 50-53]) was negatively correlated with aldosterone concentration in septic foals. The negative correlation between aldosterone concentration and clinical evidence of hypoperfusion supports relative hypoaldosteronism as a part of RAI, because increased aldosterone concentrations would have been expected. High variability in ACTH, cortisol, and aldosterone concentrations associated with an early or late stage of sepsis exists in neonatal foals.[35, 54] Therefore, our definition of relative hypoaldosteronism based solely on high ACTH/aldosterone should be interpreted with caution.
This study provides evidence that the RAAS and HPAA are activated in sick foals. Our data also suggest that RAI in critically ill foals is characterized by impaired cortisol and aldosterone secretion. Functional studies evaluating the adrenal response to various stimuli (ACTH, ANG-II, potassium) will enhance our understanding of the endocrine response to sepsis in foals.
The authors thank Krista Hernon, Alicia Griffin, and Eason Hildreth for their assistance with laboratory techniques. We are grateful to Dr Rhonda Rathgeber at Hagyard Equine Institute for collecting samples from healthy foals. The authors also thank Mary Kinee of Rood and Riddle Equine Hospital and Justine Elam of Hagyard Equine Institute for sample processing and Steven Naber of The Ohio State University for statistical assistance.
Grant Support: This study was funded by the United States Department of Agriculture Formula grants and the OSU College of Veterinary Medicine Equine Funds.
Conflict of Interest: Authors disclose no conflict of interest.
Coat-A-Count Aldosterone Radioimmunoassay, Siemens Healthcare Diagnostics, Los Angeles, CA
Coat-A-Count Cortisol Radioimmunoassay, Siemens Healthcare Diagnostics
Immulite ACTH assay, Siemens Healthcare Diagnostics
Angiotensin II Radioimmunoassay, ALPCO Diagnostics, Salem, NH
Phenylsilylsilica Extraction Columns, ALPCO Diagnostics
GammaCoat Plasma Renin Activity Radioimmunoassay, DiaSorin, Stillwater, MN
Prism, version 4.0a, GraphPad Software Inc, San Diego, CA
SAS version 9.1, SAS Institute Inc, Cary, NC