SEARCH

SEARCH BY CITATION

Keywords:

  • Adrenal responses;
  • Equine;
  • Pituitary

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References

Background

Changes in both adrenocorticotropin (ACTH) and cortisol concentration in response to thyrotropin releasing hormone (TRH) administration have been used to diagnose equine pituitary pars intermedia dysfunction (PPID), but the use of the 2 hormones has not been compared.

Hypotheses

Measuring ACTH concentration is superior to measuring cortisol concentration after TRH administration in differentiating between normal horses and those with PPID, and the 2 hormone concentrations are disassociated in PPID horses.

Animals

Eleven horses and 2 ponies with PPID and 19 normal horses.

Methods

A study evaluating cortisol and ACTH concentrations before and at 14, 30, and 60 minutes after TRH administration.

Results

At 14 and 30 minutes after TRH administration, cortisol concentration increased in PPID horses, and ACTH increased in all groups; ACTH, but not cortisol concentration, was significantly higher in PPID horses compared with normal horses. A relationship between cortisol concentration and ACTH concentration was seen in normal horses, but not in horses with PPID. Compared with normal castrated males, normal female horses had a greater change in cortisol concentration per unit change of ACTH concentration.

Conclusions and Clinical Importance

ACTH and cortisol concentrations are disassociated in horses with PPID. Measuring ACTH concentration after TRH administration appears superior to measuring cortisol concentration as a diagnostic test for PPID.

Abbreviations
ACTH

adrenocorticotropin

AUC

area under the curve

BCS

body condition score

BMI

body mass index

CN

clinically normal horses

IL6

interleukin 6

PPID

pituitary pars intermedia dysfunction

TRH

thyrotropin releasing hormone

α-MSH

alpha-melanocyte stimulating hormone

Equine pituitary pars intermedia dysfunction (PPID) is a well-recognized condition of middle-aged or older horses and ponies. Various endocrinologic tests have been developed for diagnosing the condition. Among these tests is the evaluation of serum cortisol concentrations before and after thyrotropin releasing hormone (TRH) administration.[1-3] The combined dexamethasone suppression/TRH stimulation test was developed because of the potential influence of basal cortisol concentration on horses’ responses to TRH, and the hypothesis that TRH administration would override the induced cortisol suppression only in horses with PPID.[4, 5] A high sensitivity and specificity for adrenocorticotropin (ACTH) concentrations ≥35 pg/mL at 0 and at 30 minutes post-TRH administration have been reported for differentiating between normal horses and those with PPID,[6] and another study showed a greater increase in ACTH concentration, but a similar increase in cortisol concentration in PPID horses compared with normal horses 30 minutes after TRH administration.[7] The major objectives of this study were to compare ACTH and cortisol responses to TRH administration in both normal horses and in those with PPID, determine whether the hormone concentration changes correlated, and determine whether measurement of 1 hormone more reliably differentiated between normal horses and those with PPID.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References

Animals

Results of 36 TRH stimulation tests performed in 30 horses and 2 ponies were used in this study. All procedures were approved by an institutional animal care and use committee and owners of privately owned animals gave consent to the tests. Clinical signs used to classify horses as having PPID or mild PPID have been published previously.[5] There were 9 female horses and 2 castrated male horses, and 2 female ponies in the PPID group (2 Morgan horses, 1 Standardbred, 3 Arabians, 2 Quarter Horses, 2 Paints 1 Thoroughbred Cross, and 2 Welsh Ponies). Ages ranged from 13 to 34 years (mean, 22.2 ± 6.5 years; median, 22.5 years). A total of 8 horses and 2 ponies had at least two of the classical clinical signs of PPID, including long wavy haircoat, abnormal hair shedding, abnormal fat deposition, lethargy, loss of epaxial muscle mass and pendulous abdomen, recurrent infections, and laminitis. Five of these were necropsied; 3 had pituitary macroadenomas and 2 had microadenomas. Three horses had mild signs of PPID; 1 had a pituitary macroadenoma and 2 had microadenomas. Two PPID horses had 3 tests each; 1 horse was receiving pergolide treatment PO. All other horses had single tests performed. The 19 horses that were clinically normal (CN) included 7 that were necropsied and confirmed to have normal pituitary gland histology, 7 that were necropsied and had histologic evidence of pituitary hyperplasia, and 5 that were not necropsied. Within the CN group (10 Thoroughbreds, 4 Quarter Horses, 2 Warmbloods, 1 Standardbred, 1 Saddlebred, and 1 Thoroughbred), there were 11 castrated males, 7 females, and 1 intact male horse; ages ranged from 2 to 25 (mean, 10.5 ± 5.5 years; median, 9 years). In the CN group, 16 horses were tested between December and March, 1 was tested in April, 1 was tested in July, and 1 was tested in August. In the PPID group, 12 tests were performed between December and March, 2 in May, 1 in August, 1 in October, and 1 in November. ACTH data from 15 CN horses and 10 tests in 9 PPID horses were included in a previous report.[6]

TRH Stimulation Test

The TRH stimulation test was performed by the same technician, with the horses in their familiar surroundings. The technique has been described previously.[6] Briefly, an 18 ga IV catheter1 was placed aseptically 30 minutes before the test. Baseline jugular blood samples were obtained before injecting 1 mg of synthetic TRH2 and flushing the catheter with 10 mL of heparinized saline (0.9% NaCl) solution. The TRH was reconstituted by sterile technique, under a biohazard hood, with sterile Dulbecco PBS3 to a concentration of 1 mg of TRH per mL. One milliliter of aliquots was filtered using a 0.22-μm syringe filter4 and stored in sterile Eppendorf tubes at −70°C until use. A volume of 6 mL was aspirated and discarded before obtaining each blood sample. For the current study, samples obtained at time 0 (baseline) and at 14, 30, and 60 minutes after TRH were used for measuring ACTH and cortisol concentrations. Samples were collected in plastic 6 mL potassium EDTA tubes that were centrifuged for separation of plasma within 45 minutes of collection. Plasma was stored at −20°C in plastic tubes, placed on ice packs, and sent by overnight mail to the Animal Health Diagnostic Center at Cornell University. For cortisol analyses, plasma samples were frozen at −80°C and shipped in 2 batches for analysis. Cortisol concentration was measured by a chemiluminescence immunoassay5 validated for horses.[8] A sequential immunometric assay5 that used chemiluminescence for signal generation was used to measure ACTH concentration. The test generally is specific for ACTH fragment 1–39, but has approximately 12–14% cross reactivity with fragment 18–39 and has been validated for use in horses.[9] The laboratory's normal reference range for cortisol concentration is 2–6 μg/dL, and for ACTH concentration is ≤35 pg/mL.

The cortisol and ACTH concentrations at time 0, 14, 30, and 60 minutes after TRH administration, increases in cortisol and ACTH concentrations and AUC for ACTH and cortisol concentrations at 14 and 30 minutes from baseline were compared for the groups. The magnitude of increase in cortisol concentration per unit increase in ACTH concentration was also evaluated.

Statistical Analysis

Tabulation and plotting were used to evaluate the data as suggested by Hosmer and Lemeshow,[10] and by Altman.[11] This exposed patterns in the data as well as possible outlying observations. To quantitate relationships between variables, we used regression analysis and logistic regression clustering on horses.

Scalar linear regression was used (in the fashion suggested by Campbell[12]) to explore the degree to which detectable ACTH and cortisol concentrations were quantitatively associated. To establish the association between specific ACTH and cortisol concentrations and the presence of PPID, logistic regression[10, 13] was used and the strength of association was quantified in terms of odds.

Area under the ACTH and cortisol curves (AUCs) was calculated using the trapezoidal method. The advantage of this approach in estimating areas is that it avoids dramatic deviations to which spline interpolation is susceptible when the surface shape is highly variable between consecutive time points. Nonparametric statistical tests were used in all AUC-based explorations of hormone responses.

To confirm the normality of the data, validating confidence interval estimates, and overall regression significance, we used the Shapiro–Wilks test and Tukey's ladder test (confirming the overall robustness of the untransformed versus transformed data6). The raw ACTH and cortisol data used in this investigation were normally distributed by the tests described, and no animals were included in this study if they did not have normally distributed data.

To confirm the robustness of our logistic regression results, we performed the following explorations: (1) the P value of the logistic regression results using Fisher's exact test6; (2) the logistic link by comparing our results with similar models fitted with alternate link functions (eg, cloglog and probit[3, 13]); and the sense, magnitude, and significance of outcome-exposure association with the aid of Somer's D statistic.[14],6

We used a P value of .05 to distinguish between significant and nonsignificant results in conjunction with hypothesis tests. The statistical software Stata 11.16 was used for all analyses reported.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References

ACTH Concentration

Mean (±SD), median (p50), minimum, and maximum values for ACTH concentrations and log values are shown in Table 1. In all groups, ACTH increased above baseline at 14 and 30 minutes, but baseline and 60-minute concentrations did not differ. At 14 minutes, ACTH concentrations were higher than at 60 minutes post-TRH in all groups, but the 30-minute ACTH concentration was higher than at 60 minutes only in the CN group. ACTH was higher in the PPID group than in the CN group at all times except 60 minutes. The AUC was significantly different in the PPID group compared with the CN group (Fig 1). The AUC median was 1,323 pg/mL in the CN group and 8,155 pg/mL in the PPID group, with the 9 most severely affected horses having a higher median AUC of 21,224 pg/mL. The minimum and maximum AUC values for the CN group were 557–7,362 pg/mL compared to 1,713–224,138 pg/mL for the PPID group. Within the PPID group, ACTH was >36 pg/mL in 10 of the 17 tests at baseline, 17/17 tests at 14 minutes, and 16/17 tests at 30 minutes. One test value was missing at 30 minutes. Within the CN group, ACTH concentration >36 pg/mL was seen in 1 horse at baseline, 5 horses at 14 minutes, and 3 horses at 30 minutes. There was no difference in ACTH concentrations between CN male and female horses. One CN horse with pituitary hyperplasia that was tested in August had ACTH concentrations above the 95% confidence interval for the rest of the CN group. One PPID horse tested in October had a baseline ACTH concentration above the 95% confidence interval for the rest of the group. Using an ACTH concentration >36 pg/mL to differentiate between PPID and CN horses had 94% specificity and 59% sensitivity at baseline, and 78% specificity and 94% sensitivity at 30 minutes post-TRH. The ACTH concentrations in the PPID horse receiving pergolide were not significantly different from those of the other PPID horses except that ACTH was higher at 30 minutes in the treated horse.

image

Figure 1. AUC for adrenocorticotropin (ACTH) concentrations image and cortisol concentrations image in normal (CN) horses and the pituitary pars intermedia dysfunction (PPID) horses. The ordinate is log scale. The AUC for ACTH but not cortisol differs between groups.

Download figure to PowerPoint

Table 1. ACTH concentrations before and after TRH administration.
 014 min30 min60 min
  1. PPID, horses with pituitary pars intermedia dysfunction; CN, all clinically normal horses regardless of postmortem classification; p50, median; 1, log value; min–max, minimum and maximum numbers.

  2. For each parameter in columns, different letter superscripts indicate a significant difference between groups. Across columns, different superscript numbers indicate a difference from baseline. Only mean and log mean values were analyzed. Numbers of tests per group are in parentheses under mean ACTH, and when the number of tests per sampling time differs, the number is in parentheses next to the value.

ACTH mean ± SD
PPID (17)78.0 ± 78.4a,1902.5 ± 1374.5a,2429.3 ± 598.5a,2,3 (16)106.9 ± 142.51,3 (10)
CN (19)21±b,136.7 ± 27.1b,228.5 ± 14.8b,2 (17)24.3 ± 17.81 (13)
ACTH p50
PPID4828211445.6
CN19.22824.818.5
ACTH min–max
PPID17.3–29051.7–5,11731.2–1,877.122.8–408
CN14.1–39.98.5–10911.8–71.68.6–79.4
l ACTH mean ± SD
PPID4 ± 0.83a5.84 ± 1.43a5.19 ± 1.34a4.09 ± 1.02
CN3 ± 0.29b3.4 ± 0.64b3.25 ± 0.45b3.04 ± 0.53
l ACTH p50
PPID3.95.644.733.81
CN2.953.333.212.92
l ACTH min–max
PPID2.8–5.73.94–8.543.44–7.543.13–6.01
CN2.65–3.682.14–4.692.47–4.272.15–4.37

Cortisol Concentration

Mean (±SD), median (p50), minimum and maximum values, and log values for cortisol concentrations are presented in Table 2. Cortisol concentration exceeded the maximum normal range (6 μg/dL) in 2/17 tests in 13 PPID horses and 10/19 CN horses at baseline. One CN horse with pituitary hyperplasia that was tested in August had cortisol concentrations above the 95% confidence interval for the rest of the CN group. When data from this horse were included in the CN group, the cortisol concentration was lower in the PPID group than in the CN group at baseline, but not different between the groups at any other time. One PPID horse tested in August and 1 tested in October had baseline cortisol concentrations below the 95% confidence interval for the rest of the group. Cortisol concentrations at 14 and 30 minutes were higher than at baseline in the PPID group, but not in the CN group. The PPID horses had a significantly higher percentage increase and change in cortisol concentration than the CN horses at 14, but not at 30 minutes. The baseline cortisol concentration did not affect the percentage increase in either group at 14 or 30 minutes. There was no overall difference for the unit increase in cortisol concentration between the PPID and CN groups after TRH administration. There was no difference in cortisol concentration AUC between the PPID group and the CN group. The AUC for the PPID and CN groups is shown in Figure 1. Median AUC was 376 μg/dL in the CN group and 399 μg/dL in the PPID group, and the minimum and maximum values were 190–1,433 and 165–1,177 μg/dL, respectively.

Table 2. Cortisol concentrations before and after TRH administration.
 014 min30 min60 min
  1. For each parameter in columns, different letter superscripts indicate a significant difference between groups. Across columns, different numbers indicate a significant difference from baseline. Only mean and log mean values were analyzed.

  2. See Table 1 for abbreviations.

  3. a

    When the CNP female tested in August was eliminated, baseline concentration of the CN group (5.75 ± 1.6 μg/dL) did not differ from the PPID group and median was 5.98 μg/dL.

Cortisol mean ± SD
PPID (17)4.96 ± 1.291,a6.53 ± 1.2426.49 ± 1.3425.91 ± 2.341
CN (19)6.41 ± 2.13ba6.85 ± 2.436.92 ± 2.26.29 ± 2.78
Cortisol p50
PPID4.816.646.075.38
CN6.056.466.55.5
Cortisol min–max
PPID3.08–7.434.28–9.064.6–8.743.63–10.6
CN2.9–13.33.51–13.23.58–13.74.33–14.7
l Cortisol mean ± SD
PPID1.57 ± 0.261,a1.86 ± 0.1921.85–0.2021.71 ± 0.381
CN1.81–0.30b1.87 ± 0.321.89 ± 0.281.78 ± 0.34
l Cortisol p50
PPID1.571.891.801.68
CN1.801.871.871.70
l Cortisol min–max
PPID1.12–2.011.45–2.21.52–2.171.29–2.36
CN1.06–2.591.26–2.581.27–2.621.47–2.69

An increase in cortisol concentration >50% above baseline was seen in 2/19 CN tests at 14 minutes and in 2/18 CN tests at 30 minutes. In the PPID group, a >50% and a >66% increase in cortisol from baseline was seen in 2 of 17 tests at both 14 and 30 minutes, at 14 minutes in 1 horse that lacked a 30-minute sample result and only at 30 minutes in another horse. The cortisol concentrations in the PPID horse receiving pergolide were not significantly different from those of the other PPID horses.

Within the CN group, cortisol concentrations were higher in females (7.53 ± 0.52 μg/dL) compared with castrated males (5.76 ± 0.23 μg/dL). When individual time points were examined, values in female horses were higher (8.14 ± 1.09 versus 5.90 ± 0.39 μg/dL) at 14 minutes, but not significantly different (P = .056) at 30 minutes (8.14 ± 1.11 versus 6.14 ± 0.34 μg/dL). Deletion of the CN female with pituitary hyperplasia that was tested in August and had cortisol concentrations outside of the 95% confidence interval for the rest of the CN group eliminated the sex difference at 14 minutes, but cortisol concentration remained significantly higher overall in females compared with castrated males.

Relationship between ACTH and Cortisol Concentrations

The relationship between cortisol concentration and ACTH concentration that was seen in the CN horses was absent in the PPID horses. When time points including baseline were examined for the CN group, for each unit ACTH concentration, cortisol concentration increased by 0.058 ± 0.013 units. When time periods after TRH administration were examined for a relationship, the relationship remained significant for the CN group (0.052 ± .014 units cortisol/unit ACTH concentration). When cortisol concentration at 30 minutes was regressed on ACTH concentration at 14 minutes, there was a significant relationship only for the normal horses. When the relationship between ACTH and cortisol concentrations was compared in CN male and female horses, females had a greater change in cortisol concentration (0.088 ± 0.013 units versus 0.020 ± 0.014 units) per unit change of ACTH concentration (Fig 2). Robust regression showed that this sex difference persisted even with omission of data from the female with high cortisol concentrations that was tested in August. It was not possible to examine PPID horses for sex effect because most of the horses were females.

image

Figure 2. Predicted and observed mean values of cortisol associated with adrenocorticotropin (ACTH) concentration in 12 clinically normal (CN) males (—; image) and 7 CN females (- -; image) The data were collapsed by case and by time. The line represents the averaged means, not individual data points.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References

The premise of the TRH test is that TRH stimulates ACTH release from the pituitary gland, but cortisol originally was measured because of lack of a commercially available test for ACTH in horses at that time.[1] Measuring ACTH concentrations has been shown to have high sensitivity and specificity for diagnosing PPID,[6] and TRH has been shown to have a direct stimulatory effect on the pars distalis and the pars intermedia.[7] Both ACTH and α-MSH concentrations increase after TRH administration in normal horses and those with PPID, but the concentrations were higher in horses with PPID,[6, 7] similar to this study. To our knowledge, only 1 other study has examined both cortisol and ACTH concentration changes after TRH administration, but only baseline and 30-minute samples were obtained.[7] Unlike in this study where cortisol concentrations significantly increased only in the PPID horses, in that study, all horses had an increase in cortisol concentrations, but similar to the current study's findings, actual concentrations were not significantly different between normal horses and those with PPID.

Results of the current study are similar to earlier reports on cortisol concentrations, but also have some differences. Previous studies[1, 2, 4] measuring cortisol response used 15 rather than 14 minutes as the first sampling period, but we do not believe that this difference would influence results or conclusions from the current study. The 1st study suggesting cortisol response to TRH administration might differentiate between normal and PPID horses reported no change in cortisol concentrations in normal horses, whereas concentrations in PPID horses were significantly higher than baseline at 15, 30, 60, and 90 minutes after administration of TRH.[1] In the current study, horses with PPID had a significant increase from baseline at 14 and 30 minutes, but unlike the first report, the 60-minute concentration did not differ from baseline. Another study reported a significant increase in cortisol concentration at 15 and 30 minutes after TRH in 2 PPID and 3 CN horses, but maximum cortisol concentration was different from the 3 CN horses only at 15 minutes.[4] A descriptive study on 11 horses without clinical signs and 2 with signs of PPID reported PPID horses had greater percentage increases in cortisol concentrations at 15 minutes (127–245%), but unaffected horses’ concentrations also increased, varying from 17% to 105% above very variable basal concentrations.[2] In our study, PPID horses had a greater percentage increase in cortisol concentration at 14 minutes, but the mean cortisol concentrations and percentage increase in cortisol concentrations did not differ between the groups at 30 minutes after TRH administration, similar to another study that reported no significant difference in percentage increase or absolute cortisol concentration in horses with PPID compared with normal horses.[7] The initial study showing a significant increase in cortisol in PPID horses[1] did not compare the mean concentrations of the 2 groups at different times nor did it report a certain percentage increase as being diagnostic for PPID. Using a t test to compare the mean values of the 2 groups at different times in that study, there is no difference except that the PPID group's concentration is lower at baseline (33.5 ± 11.8 versus 48.4 ± 13.9 ng/mL).

The influence of baseline cortisol concentrations on the response to TRH has been questioned, and several studies have attributed the greater increase in cortisol concentration in the PPID horses to the group's lower basal cortisol concentrations.[2, 4] One study reported that 6 PPID horses with basal hypercortisolemia had a lower percentage increase in cortisol concentration 30 minutes after TRH administration than 5 horses with normal basal concentrations (40 ± 18 versus 116 ± 36%).[3] The initial study had not attributed the PPID horses’ greater increase to low baseline concentrations because the magnitude of increase was reported to be independent of the baseline cortisol concentration.[1] In the current study, there was no difference in basal cortisol concentrations between the CN and PPID groups when the high cortisol concentration from the CN horse with pituitary hyperplasia was deleted from the analysis, but the difference between the group's responses persisted. Basal cortisol concentration did not influence percentage increase at either 14 or 30 minutes after TRH administration. However, until larger numbers of horses are evaluated, the potential influence of basal cortisol concentrations on the response remains uncertain. A minimal increase of 30–50% in cortisol concentration has been proposed to be diagnostic for PPID, although no supporting data were presented,[15, 16] and the use of the 30% increase appears to have been based on the initial study,[1] which reported mean values and did not evaluate percent changes. One study of 11 horses with PPID reported a 74 ± 47% increase,[3] and another study reported a 50% increase in cortisol concentration 30 minutes after TRH in 5/7 PPID horses and 7/16 normal horses.[7]

In the combined dexamethasone/TRH test, a 66% increase in cortisol concentration 30 minutes after TRH is considered positive.[4, 5] In the current study, using the criterion of a 50–66% increase was not sensitive for identifying affected horses. The reasons for these differences among tests are speculative. Although cortisol was measured using different assays in different laboratories in the studies on the TRH stimulation test, this seems unlikely to explain the different results. None of the earlier studies on the TRH stimulation test specified in which seasons the tests were performed, but season seems unlikely to be a major factor. Studies on basal cortisol concentrations have shown no increase during August through October,[17-21] although concentrations were reported to be higher in May than in April in mares.[17] Reports on the effect of season on cortisol concentration response to dexamethasone suppression have been inconsistent.[18, 20] Cortisol concentrations were the same in PPID horses and control horses after oral domperidone, regardless of season.[22] Investigations of a seasonal effect on cortisol concentration changes in response to TRH stimulation have not, to our knowledge, been reported. However, there is no evidence to suggest an effect or indicate that the seasonal effect on ACTH concentration in response to TRH administration explains the different results among tests evaluating cortisol concentration response. The effect on ACTH concentration is modest,[6, 23] and ACTH and cortisol concentrations appear disassociated in horses with PPID. Also, in the current study, exclusion of the 3 horses tested between August and October did not alter results of the test. Numbers of CN horses with pituitary changes, and ages of horses could have varied among the studies on the TRH stimulation test and potentially influenced results. As the sex of horses was not always specified or only 1 sex was studied, it is unknown whether this could contribute to differences among reports. Age ranges appear similar among the studies, although median ages are unknown. Age-related increase in pars intermedia lesions has been reported,[6, 24] but the largest study[24] did not have accompanying diagnostic tests for PPID, and the impact of the pituitary changes on diagnostic tests remains unclear. Differences in the populations are likely to have contributed to different test results, and differences in the secretory state of the pituitary gland of the horses in the different studies also could have contributed. The stage of PPID could have varied among the studies and is likely to have been more advanced in at least the first study. Also, the CN groups are probably not homogeneous because some CN horses have pituitary hyperplasia that potentially could have functional effects even if the horse appears CN.[6]

There are several explanations for the disassociation between ACTH and cortisol concentrations, including a plateau effect of ACTH on the adrenal gland. Dose–response studies evaluating cortisol response to different doses of ACTH have shown a plateau effect,[25-27] but in humans, a plateau was not observed for endogenous concentrations[28] and the relationship with endogenous ACTH concentrations remains unknown in horses. Endogenous ACTH concentration changes after different transport distances have been shown not to be concomitant with those of cortisol concentrations in normal male horses, but the authors did not discuss the divergence.[29] As we measured total cortisol concentrations, it is unknown whether measuring the free cortisol concentration would have differentiated between normal horses and those with PPID, or revealed a different correlation with plasma ACTH concentration. The relationship between free and total cortisol concentrations after stimuli such as administration of TRH has not been reported in horses. Although measuring free cortisol concentrations may be preferable to measuring total cortisol concentration based on other studies,[26, 30, 31] we do not believe that it would alter this study's conclusions, and most veterinarians do not use this methodology.

In humans, body mass index (BMI), sex, age, and IL6 concentrations can affect the ACTH-cortisol drive.[31-33] The sex difference for the relationship between ACTH and cortisol concentrations seen in the normal horses, to our knowledge, has not been reported previously, although a sex difference has been reported in dogs, with housing influencing whether a sex difference was detected.[34, 35] A study in normal horses reported that normal mares had higher endogenous cortisol concentrations than normal castrated males, but response to dexamethasone did not differ.[36] A study in normal horses and those with PPID showed no difference between the sexes in cortisol concentrations at baseline or 2 hours after administration of ACTH, but the numbers of CN horses were small, and any sex effect could be lost or obscured with PPID.[37] In the current study, age of CN females (12 ± 7.5 y) and castrated males (9.7 ± 4.1 y) was similar and therefore unlikely to explain sex differences, and BMI was not determined. Our study was not designed to compare responses in horses with high versus low BCS, but there was no significant difference in BCS between the groups or between the CN castrated males (BCS, 5.33 ± 1.75) and females (BCS, 5.5 ± 1.5). A seasonal effect did not appear to explain the apparent sex difference in the CN group; 6/7 females had tests between December and March with 1 tested in early August, and 11/12 males were tested between December and April with 1 tested in July. The sex difference remained after deleting data from the CN female tested in August. Overrepresentation of males or females in 1 CN subgroup did not appear to explain the sex difference because there was no statistical difference between numbers of values from males and females in the different CN categories. Also, the likelihood of subclinical pituitary hyperplasia causing the sex difference in the CN group is not supported by a study, showing no sex difference in PPID horses’ cortisol concentrations after ACTH administration.[37] Although estrus cycle status was not determined because mares were not palpated per rectum to determine ovarian activity at the time of the test, none was showing estrus behavior. The most definitive way to evaluate a sex effect would be to test ovariectomized mares and stallions.

It is unknown whether a difference in the ratio of immunoreactive to bioactive ACTH or events at the level of the adrenal gland could explain the disassociation between ACTH and cortisol concentrations. The apparent lack of a strong temporal relationship between plasma immunoreactive ACTH and cortisol concentrations has been reported in normal humans,[38-40] in dogs undergoing acute blood loss,[41, 42] and in stressed late gestation ovine fetuses.[43] Several bioactive and immunoreactive forms of ACTH have been reported in humans, and the proportions vary with the secretory state of the pituitary gland.[38, 39] The relationship between biologically active and immunoreactive ACTH appears to vary with the stimulus.[38, 44] A poor correlation between plasma immunoreactive ACTH concentrations and immunoreactive cortisol concentrations in response to a dopamine agonist has been reported previously in 2 horses with PPID, leading the authors to hypothesize that total immunoreactive ACTH concentration did not accurately reflect the concentration of bioactive ACTH.[45] The immunoreactive ACTH to bioactive ACTH ratio also was found to be higher in PPID horses’ adenomas compared with normal horses’ pars intermedia tissue.[46] Two studies utilizing in vitro ACTH bioassays indicated that the ratio of biological to immunoreactive ACTH concentration is lower in plasma from horses with PPID compared with normal horses.[47, 48] The reason for variable pathologic changes in the adrenal cortex reported in PPID horses with high plasma ACTH concentrations could be explained if the latter has variable bioactivity. To our knowledge, the relationship between the 2 forms of ACTH has not been examined after TRH administration. Whether TRH stimulates release of other hormones that affect the proportion of ACTH that is biologically active, or whether it stimulates different populations of corticotropes or melanotropes that process or release different forms of ACTH or ACTH-like peptides remains unknown. This seems unlikely to entirely explain the disassociation between ACTH and cortisol concentrations in the horses because the disassociation also was found for basal endogenous concentrations before TRH administration.

In this study, measuring ACTH concentration changes in response to TRH appeared superior to measuring cortisol concentrations in diagnosing PPID, but the dissociation between ACTH and cortisol concentrations, and the sex effect pose more questions about ACTH bioactivity and the ACTH-cortisol drive in horses.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References

The authors acknowledge Dr Barbara Schanbacher and Steve Lamb for endocrinologic assays, and Dr Perry Habecker for pathology evaluations. Private funding by M Gardiner accounted for partial support.

Footnotes
  1. 1

    Terumo Surflo®; Terumo Medical Corp, Elkton, MD

  2. 2

    TRH; Sigma-Aldrich Co, St. Louis, MO

  3. 3

    Gibco® – Invitrogen Corp, Carlsbad, CA

  4. 4

    GE Healthcare, Minnetonka, MN

  5. 5

    Diagnostic Products, Corp, Los Angeles, CA

  6. 6

    Stata version 11.1; Stata 2010 Corporation, College Station, TX

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  • 1
    Beech J, Garcia M. Hormonal response to thyrotropin-releasing hormone in healthy horses and in horses with pituitary adenoma. Am J Vet Res 1985;46:19411943.
  • 2
    Thompson JC, Ellison R, Gillert R. Problems in the diagnosis of pituitary adenoma (Cushing's syndrome) in horses. NZ Vet J 1995;43:7982.
  • 3
    Hillyer MH, Taylor FGR, Mair TS, et al. Diagnosis of hyperadrenocorticism in the horse. Equine Vet Educ 1992;4:131134.
  • 4
    Eiler H, Oliver JW, Andrews FM, et al. Results of a combined dexamethasone suppression/thyrotropin-releasing hormone stimulation test in healthy horses and horses suspected to have a pars intermedia pituitary adenoma. J Am Vet Med Assoc 1997;211:7981.
  • 5
    Frank N, Andrews FM, Sommardahl CS, et al. Evaluation of the combined dexamethasone suppression/thyrotropin-releasing hormone stimulation test for detection of pars intermedia pituitary adenomas in horses. J Vet Intern Med 2006;20:987993.
  • 6
    Beech J, Boston R, Lindborg S, et al. Adrenocorticotropin concentration following administration of thyrotropin releasing hormone in healthy horses and those with pituitary pars intermedia dysfunction and pituitary hyperplasia. J Am Vet Med Assoc 2007;231:417425.
  • 7
    McFarlane D, Beech J, Cribb A. Alpha-melanocyte stimulating hormone release in response to thyrotopin releasing hormone in healthy horses, horses with pituitary pars intermedia dysfunction and equine pars intermedia explants. Domest Anim Endocrinol 2006;30:276288.
  • 8
    Singh AK, Jiang Y, White T, et al. Validation of nonradioactive chemiluminescent immunoassay methods for the analysis of thyroxine and cortisol blood samples obtained from dogs, cats, and horses. J Vet Diagn Invest 1997;9:261268.
  • 9
    Perkins GA, Lamb S, Erb HN, et al. Plasma adrenocorticotropin (ACTH) concentrations and clinical response in horses treated for equine Cushing's disease with cyproheptadine or pergolide. Equine Vet J 2002;34:679885.
  • 10
    Hosmer DW, Lemeshow S. Applied Logistic Regression, 2nd ed. Chapters 4 and 5. New York: Wiley; 2000.
  • 11
    Altman DG. Practical Statistics for Medical Research, 1st ed. Chapter 7. New York: Chapman and Hall; 1995.
  • 12
    Campbell MJ. Statistics at Square Two, 2nd ed. Chapters 2 and 3. London: BMJ Books, Blackwell Publ; 2006.
  • 13
    Collett D. Modelling Binary Data. 1st ed. Chapters 5 and 6. New York: Chapman and Hall; 1996.
  • 14
    Newson R. Confidence intervals for rank statistics: percentile slopes, differences, and ratios. Stata J 2006;6:497520.
  • 15
    Schott H. Pituitary pars intermedia dysfunction equine Cushing's disease. Vet Clin Equine 2002;18:237270.
  • 16
    McFarlane D, Cribb A. Alpha melanocyte stimulating hormone release in response to thyrotropin releasing hormone administration in healthy horses. J Vet Intern Med 2004;18:413.
  • 17
    Fazio E, Medica P, Ferlazzo A. Seasonal patterns of circulating β endorphin, adrenocorticotrophic hormone and cortisol levels in pregnant and barren mares. Bulg J Vet Med 2009;12:125135.
  • 18
    Schreiber CM, Stewart AJ, Behrend EN, et al. Seasonal variation in diagnostic tests for pituitary pars intermedia dysfunction in normal aged geldings. J Am Vet Med Assoc 2011; in press.
  • 19
    Place NJ, McGowan CM, Lamb SV, et al. Seasonal variation in serum concentrations of selected metabolic hormones in horses. J Vet Intern Med 2010;24:650654.
  • 20
    Donaldson MT, McDonnell SM, Schanbacher BJ, et al. Variation in plasma adrenocorticotrophic hormone concentration and dexamethasone suppression test results with season, age and sex in healthy ponies and horses. J Vet Intern Med 2005;19:217222.
  • 21
    Haritou SJA, Zylstra R, Ralli C, et al. Seasonal changes in circadian peripheral plasma concentrations of melatonin serotonin, dopamine and cortisol in aged horses with Cushing's disease under natural photoperiod. J Neuroendocrinol 2008;20:988996.
  • 22
    Sojka JE, Jackson PE, Moore GE, et al. Effect of season and dose on domperidone stimulated change in adrenocorticotrophic hormone concentrations in horses with pituitary pars intermedia dysfunction. Cambridge, MA: Dorothy Russell Havemeyer Foundation, Equine Geriatric Workshop; October 2010: p.41.
  • 23
    Funk RA, Stewart AJ, Woodridge AA, et al. Seasonal changes in plasma adrenocorticotrophic hormone and alpha-melanocyte stimulating hormone in response to thyrotropin releasing hormone in normal aged horses. J Vet Intern Med 2011;25:579585.
  • 24
    van der Kolk JH, Heinrichs M, Amerongen JD, et al. Evaluation of pituitary gland anatomy and histopathologic findings in clinically normal horses and horses and ponies with pituitary pars intermedia adenoma. Am J Vet Res 2004;65:17011707.
  • 25
    Alam MGS, Dobson H, Fitzpatrick RJ. Endocrine response to different doses of ACTH in cows. Br Vet J 1986;142:239245.
  • 26
    Bousquet-Melou A, Formentini E, Picard-Hagan N, et al. The adrenocorticotropin stimulation test: contributions of a physiologically based model developed in horse for its interpretation in different pathophysiological situations encountered in man. Endocrinology 2006;147:42814291.
  • 27
    Azziz R, Bradley E, Huth J, et al. Acute adrenocorticotropin (1-24) (ACTH) adrenal stimulation in eumonorrheic women: reproducibility and effect of ACTH dose, subject weight and sampling time. J Clin Endocrinol Metab 1990;70:12731279.
  • 28
    Krishnan KRR, Ritchie JC, Saunders W, et al. Nocturnal and early morning secretion of ACTH and cortisol in humans. Biol Psychiatry 1990;28:4757.
  • 29
    Fazio E, Medica P, Aronica V, et al. Circulating β-endorphin, adrenocorticotropic hormone and cortisol levels of stallions before and after short road transport: stress effect of different distances. Acta Vet Scand 2008;50:6.
  • 30
    Alexander SL, Irvine CHG. The effect of social stress on adrenal axis activity in horses: the importance of monitoring corticosteroid-binding globulin capacity. J Endocrinol 1998;157:425432.
  • 31
    Roelfsema F, Kok P, Frolich M, et al. Disordered and increased adrenocorticotropin secretion with diminished adrenocorticotropin potency in obese premenopausal women. J Clin Endocrinol Metab 2009;94:29912997.
  • 32
    Keenan DM, Roelfsema F, Carroll BJ, et al. Sex defines the age dependence of endogenous ACTH-cortisol dose responsiveness. Am J Physiol Regul Integr Comp Physiol 2009;297:R515R523.
  • 33
    Žarković M, Ignjatović S, Dajak M, et al. Cortisol response to ACTH stimulation correlates with blood interleukin 6 concentration in healthy humans. Eur J Endocrinol 2008;159:649652.
  • 34
    Pessina P, Fernández-Foren A, Cueto E, et al. Cortisol secretion after adrenocorticotropin (ACTH) and dexamethasone tests in healthy female and male dogs. Acta Vet Scand 2009;51:33.
  • 35
    Garnier F, Benoit E, Virat M, et al. Adrenal cortical response in clinically normal dogs before and after adaptation to housing. Lab Anim 1990;24:4043.
  • 36
    Cartmill JA, Thompson DL, Storer WA, et al. Endocrine responses in mares and geldings with high body condition scores grouped by high vs low resting leptin concentrations. J Anim Sci 2003;81:23112321.
  • 37
    van der Kolk JH, Wensing T, Kalsbeek HC, et al. Laboratory diagnosis of equine pituitary pars in intermedia adenoma. Domest Anim Endocrinol 1995;12:3539.
  • 38
    Goverde HJM, Pesman GJ, Smals AGH. The bioactivity of immunoreactive adrenocorticotrophin in human blood is dependent on the secretory state of the pituitary gland. Clin Endocrinol (Oxf) 1989;31:255265.
  • 39
    Poland RE, Hanada K, Rubin RT. Relationship of nocturnal plasma bioactive and immunoactive ACTH concentrations to cortisol secretion in normal men. Acta Endocrinol 1989;121:857865.
  • 40
    Krieger DT, Allen W. Relationship of bioassayable and immunoassayable plasma ACTH and cortisol concentrations in normal subjects and in patients with Cushing's disease. J Clin Endocrinol Metab 1975;40:675687.
  • 41
    Engeland WC, Miller P, Gann DS. Dissociation between changes in plasma bioactive and immunoreactive adreno-corticotrophin after hemorrhage in awake dogs. Endocrinology 1989;124:29782985.
  • 42
    Dempsher DP, Gann DS. Increased cortisol secretion after small hemorrhage is not attributable to changes in adrenocorticotropin. Endocrinology 1983;113:8693.
  • 43
    Castro MI, Valego NK, Zehnder TJ, et al. Bioactive-to-immunoreactive ACTH activity changes with severity of stress in late gestation ovine fetus. Am J Physiol 1993;265:E68E73.
  • 44
    Goverde HJM, Martens GJM, Pesman GJ, et al. Multiple forms of bioactive and immunoreactive adreno-corticotrophin in human pituitary and blood of patients with Nelson's syndrome. J Clin Endocrinol Metab 1993;77:443447.
  • 45
    Orth DN, Holscher MA, Wilson MG, et al. Equine Cushing's disease: Plasma immunoreactive proopiolipomelanocortin peptide and cortisol levels basally and in response to diagnostic tests. Endocrinology 1982;110:1430.
  • 46
    Orth DN, Nicholson WE. Bioactive and immunoreactive adrenocorticotropin in normal equine pituitary and in pituitary tumors of horses with Cushing's disease. Endocrinology 1982;111:559563.
  • 47
    Sommer K. Das Equine Cushing-Sydrom: Entwicklung cines ACTH-bioassays fur die Ermittlung des biologisch-immunreaktiven Verkaltnisses von endogenem ACTH in equinen Blutproben (The equine Cushing syndrome development of an ACTH-bioassay for determination of the biological-immunoreactive-ratio of endogenous ACTH in equine blood samples). Stiftung Tierarztliche Hockschule Hannover Dissertation. Hannover, Germany: School of Veterinary Medicine, 2003.
  • 48
    Shrauner B, Corder M, McFarlane D. Bioactivity of plasma ACTH in PPID-affected horses compared to normal horses. Cambridge, MA: Dorothy Russell Havemeyer Foundation, Equine Geriatric Workshop, October 2010; 53.