The impact of single-dose trazodone administration on plasma endogenous adrenocorticotropic hormone and serum cortisol concentrations in healthy dogs
Abstract
Background
Conditions affecting the hypothalamic-pituitary-adrenal (HPA) axis are common in dogs. Testing the function of the HPA axis includes measurement of endogenous adrenocorticotropic hormone (eACTH) and performance of an adrenocorticotropic hormone (ACTH) stimulation test. Trazodone is commonly administered to dogs to decrease stress. In humans, trazodone significantly decreases plasma cortisol concentration via alpha-1 adrenergic activity.
Objectives
Determine the influence of trazodone on eACTH and serum cortisol concentrations in healthy dogs.
Animals
Fourteen healthy, adult, companion dogs.
Methods
Prospective, randomized placebo-controlled study. Trazodone (8-10 mg/kg) or placebo was administered PO 1 hour before eACTH measurement and ACTH stimulation testing. After a ≥7-day wash-out period, dogs received the opposite treatment. Differences in eACTH, pre- and post-ACTH stimulation cortisol concentrations, and delta (difference between pre- and post-ACTH) cortisol concentrations were analyzed using a paired t or signed-rank test with a P < .05 significance level.
Results
The eACTH concentrations were not significantly different (P = .23) between treatments. Similarly, no significant differences were found in the pre-ACTH cortisol concentrations between treatments (P = .40). Post-ACTH cortisol concentrations (P = .05) and delta cortisol concentrations (P = .04) were significantly lower when the dogs were treated with trazodone.
Conclusions
Preliminary data suggest trazodone administration dampens the adrenocortical response to stimulation in healthy dogs. If similar effects are found in dogs with adrenal disease, the use of trazodone may affect diagnosis and clinical decision making in these populations.
Abbreviations
-
- °F
-
- degrees Fahrenheit
-
- ACTH
-
- adrenocorticotropic hormone
-
- BCS
-
- body condition score
-
- bpm
-
- beats per minute
-
- brpm
-
- breaths per minute
-
- CBC
-
- complete blood count
-
- CIRCI
-
- critical illness related cortisol insufficiency
-
- CRH
-
- corticotropin releasing hormone
-
- eACTH
-
- endogenous adrenocorticotropic hormone
-
- HAC
-
- hyperadrencorticism
-
- HPA
-
- hypothalamic-pituitary-adrenal
-
- HT
-
- hydroxytryptamine
-
- HypoA
-
- hypoadrenocorticism
-
- IACUC
-
- Institutional Animal Care and Use Committee
-
- IQR
-
- interquartile range
-
- Tmax
-
- time to maximum plasma concentrations
1 INTRODUCTION
Management of stress is relevant to both animal welfare and the ability to properly investigate and treat disease.1, 2 In the interest of promoting “fear-free” practices, the administration of anti-anxiety medications before veterinary appointments is becoming increasingly common. Trazodone is frequently administered to decrease stress-related behaviors in hospitalized dogs.3, 4 Oral trazodone administration appears to be generally well-tolerated in dogs but serious adverse effects have been reported.3-7 As such, trazodone is a popular drug used to mitigate stress and anxiety in dogs at home and in the hospital.1, 3, 4
Conditions affecting the hypothalamic-pituitary-adrenal (HPA) axis are common in dogs. These include both diseases of cortisol excess (hyperadrenocorticism [HAC]) and cortisol deficiency (hypoadrenocorticism [HypoA]). For both conditions, an accurate diagnosis is critical to allow for appropriate and timely intervention. Both are evaluated by tests related to the function of the HPA axis including measurement of endogenous adrenocorticotropic hormone (eACTH) and performance of an adrenocorticotropic hormone (ACTH) stimulation test.8-10 Interference with basal HPA function affects both eACTH and serum cortisol concentrations. Drugs capable of causing interference with the basal HPA axis include neurotransmitter-modulating drugs such as trazodone.11
Trazodone is classified as a serotonin antagonist/reuptake inhibitor. It also antagonizes 5-hydroxytryptamine (HT)2A, 2B and 2C receptors as well as alpha 1 and alpha 2 adrenergic receptors. At high doses trazodone blocks serotonin reuptake at the serotonin transporter (SERT).5, 12 In rodents, 5-HT2 antagonism suppresses eACTH and corticosterone (the main glucocorticoid in rodents) secretion through the regulation of corticotropin-releasing hormone (CRH).13 In humans, trazodone significantly decreases plasma cortisol concentration via alpha-1 adrenergic activity.14 Despite its frequent use in small animal medicine, no dose determination studies and few pharmacokinetic studies have been published on trazodone in dogs. Although these studies provided important species-specific information on drug absorption,15 bioavailability,6 and adverse effects,6 the influence of trazodone on the HPA axis in dogs has yet to be examined. If trazodone suppresses the HPA axis in dogs, it could lead to false negative test results for HAC, incorrect dosing recommendations for dogs undergoing medical treatment for HAC, and false positive test results for HypoA. Decreases in eACTH and cortisol concentrations would complicate discrimination between adrenal- and pituitary-dependent HAC and between primary and secondary HypoA. Further investigation into the effects that 5-HT2A and 5-HT2B receptor antagonists and alpha-1 antagonists such as trazodone have on eACTH and serum cortisol concentrations is warranted to avoid misinterpretation of adrenal testing.
Our aims were to determine the influence of trazodone on eACTH and serum cortisol concentrations in healthy dogs. We hypothesized that the administration of a single dose of trazodone would result in decreased eACTH concentrations as well as decreased pre-ACTH, post-ACTH and delta (difference between pre- and post-)-ACTH serum cortisol concentrations in healthy dogs.
2 MATERIALS AND METHODS
2.1 Animals
Fourteen healthy, adult (1-8 years), companion dogs belonging to the faculty, staff, and students of the Auburn University College of Veterinary Medicine and the Wilford and Kate Bailey Small Animal Teaching Hospital (AU-VTH) were prospectively enrolled with informed owner consent and approval of the Institutional Animal Care and Use Committee (IACUC #4091). Dogs were enrolled if they were determined to be healthy based on patient history, physical examination, CBC, serum biochemical profile, heartworm antigen testing, and urinalysis. The enrolled dogs also had no recent history or clinical evidence of systemic illness within the last 30 days inclusive of clinical signs consistent with either hyper- or hypoadrenocorticism. Dogs with clinical or diagnostic findings consistent with hyper- or hypoadrenocorticism, diabetes mellitus or other chronic systemic illness were excluded. Dogs with a history of corticosteroid, azole antifungal drug, or etomidate administration within 8 weeks of sample collection, trazodone administration within 2 weeks of sample collection, a temperament that was not conducive to sample collection, or a body weight < 3 kg were excluded. At the start of each data collection period, dogs were assessed for subjective markers of stress: heart rate (beats per minute [bpm]), respiratory rate (breaths per minute [brpm]), and rectal temperature (degrees Fahrenheit [°F]).
2.2 Endocrine testing
Trazodone and placebo were supplied by the AU-VTH pharmacy and administered by the investigators (MB, MG). Dogs were presented for testing in the morning, after a 12-hour fast, and either trazodone (8-10 mg/kg) or placebo was administered with food. All dogs were presented the morning of testing. No dog was hospitalized overnight before sample collection. The volume of food was based on patient size and expected volume for a single meal. A dosing range was provided to accommodate different body weights given available tablet sizes while accounting for a minimum target dose based on previous studies. Dogs were assigned to receive either trazodone or placebo using a random number generator. The food was a commercially available maintenance dog food and held consistent across all dogs to minimize the effect of food on bioavailability.
One hour after administration of either trazodone or placebo, blood was drawn for measurement of eACTH and an ACTH stimulation test was started using validated assays as previously described.16, 17 The ACTH stimulation test was performed using a 5 μg/kg IV dose of cosyntropin. The timing of trazodone administration was based on previous pharmacokinetic studies in dogs, which demonstrated that dogs reach targets for therapeutic plasma concentrations in humans (130 ng/mL-2 μg/mL) after 1 hour when trazodone is orally administered PO at 8 mg/kg.6 Furthermore, the therapeutic concentrations exceed plasma concentrations sufficient to suppress the HPA axis in people.14
Samples intended for evaluation of eACTH were placed into tubes containing EDTA and centrifuged within 15 minutes of collection. The plasma was separated, placed in plastic tubes, and immediately transported to the endocrinology laboratory at Auburn University for storage at −80°C until analysis using a previously validated chemiluminescent immunoassay (Immulite 1000, Siemens Healthcare Diagnostics, Tarrytown, New York).17 The reference range for eACTH for our laboratory was 10-80 pg/mL. Samples collected for cortisol measurement were placed into serum collection tubes and centrifuged within 1 hour after clotting. The serum was separated and stored at −80°C until analysis. Serum cortisol concentrations were measured using a previously validated chemiluminescent immunoassay (Immulite 1000).18 The reference ranges for serum cortisol concentrations for our laboratory were 20-160 and 220-560 nmol/L for pre-ACTH and post-ACTH stimulation cortisol, respectively. After a minimum 7-day wash-out period, dogs received the opposite treatment based on their initial random group assignment. Samples were analyzed in duplicate with sample results averaged for reporting. All samples were analyzed in a single batch for each hormone.
2.3 Statistical analysis
Statistical analysis was performed using commercial statistical analysis software (Sigma Plot 14.5). Descriptive statistics were performed where appropriate. Normality was evaluated by a Shapiro-Wilk test. Data for quantitative variables are presented as mean ± SD or median (interquartile range [IQR]) for normally and non-normally distributed data, respectively. Outliers were defined as values exceeding 1.5 × the IQR. Statistical comparisons between treatments for eACTH, pre-ACTH cortisol, post-ACTH cortisol, and delta cortisol concentrations were made using a paired t test (normally distributed data) or Wilcoxon signed-rank test (non-normally distributed data). Each dog served as its own control. Significance was set at P < .05. A targeted sample size of 12 was calculated using a power of 0.80 and an alpha of 0.05 assuming an approximately 50 ng/mL (137.9 nmol/L) difference in cortisol between group means, a difference extrapolated from the human medical literature.14
3 RESULTS
3.1 Animals
Fourteen healthy companion dogs were prospectively enrolled. Six breeds were represented: mixed breed (n = 7), Labrador retriever (n = 2), Brussels Griffon (n = 2), Boston terrier (n = 1), Great Dane (n = 1), and Great Pyrenees (n = 1). There were 11 castrated males and 3 spayed females. The mean ± SD age (years) was 4 ± 1.6 (range, 1-7). The mean ± SD weight (kg) at the start of the study was 28.8 ± 18.7 (range, 4.5-74). No significant differences were detected for weight between the 2 data collection times (P = .91). The median (IQR) body condition score (9-point scale) was 5 (5-6; range, 5-6). No significant differences in clinical markers of stress were found when comparing heart rate (P = .91), respiratory rate (P = .66) or rectal temperature (P = .63) between collection periods (Table 1). The median (IQR) trazodone dose (mg/kg) was 8.6 (8.3-8.7; range, 8-10.2).
| Collection 1 | Collection 2 | P value | |
|---|---|---|---|
| Heart rate (bpm) | 125.8 ± 16.4 | 126.9 ± 19.6 | .91 |
| Respiratory rate (brpm) | 36.8 ± 13.2 (n = 8) Pant (n = 6/14) |
30.8 ± 9.1 (n = 10) Pant (n = 4/14) |
.66 |
| Rectal temperature (°F) | 101.4 ± 0.8 | 101.6 ± 0.8 | .63 |
- Abbreviations: °F, degrees Fahrenheit; ACTH, adrenocorticotropic hormone; bpm, beats per minute; brpm, breaths per minute; eACTH, endogenous adrenocorticotropic hormone.
3.2 Endogenous ACTH
Endogenous ACTH samples were successfully collected for dogs receiving trazodone (n = 12) and placebo (n = 14). Two dogs in the trazodone group could not have samples analyzed in duplicate because of inadequate sample volume and therefore were not included in the statistical comparison. Median (IQR) eACTH concentrations (pg/mL) were 27.4 (18.2-37.1; range, 13.1-117) and 18.8 (12.5-34.2; range, 10.4-50.1) for trazodone and placebo, respectively. No significant difference in eACTH concentrations were found between the trazodone and placebo groups (P = .23). One dog in the trazodone group had an eACTH concentration above the reference range of our laboratory (117 pg/mL).
3.3 Cortisol
An ACTH stimulation test was successfully completed in all dogs (n = 14) after receiving both trazodone and placebo (Table 2). Dogs receiving trazodone had significantly lower post-ACTH (P = .05) and delta cortisol concentrations compared to placebo (P = .04). The median (IQR) percentage decreases in serum cortisol concentrations were 3.16% (0.3-13.4) and 25.4% (12-38.7) for post-ACTH and delta serum cortisol concentrations, respectively. The same 2 dogs had post-ACTH cortisol concentrations slightly below the reference range for our laboratory after receiving both trazodone (217 and 207.5 nmol/L [7.87 and 7.52 μg/dL]) and placebo (197 and 214.5 nmol/L [7.14 and 7.78 μg/dL]).
| Trazodone | Placebo | P value | |
|---|---|---|---|
| eACTH (pg/mL) | 27.4 (18.2-37.1) | 18.8 (12.5-34.2) | .23 |
| Pre-ACTH cortisol (nmol/L) | 45.8 (25.0-72.4) | 40.2 (32.9-52.0) | .40 |
| Pre-ACTH cortisol (μg/dL) | 1.67 (0.91-2.62) | 1.46 (1.19-1.89) | |
| Post-ACTH cortisol (nmol/L) | 313.0 (257-328) | 320.0 (294-336) | .05* |
| Post-ACTH cortisol (μg/dL) | 11.3 (9.31-11.9) | 11.6 (10.7-12.2) | |
| Delta cortisol (nmol/L) | 230.7 (180-275) | 273.0 (224-305) | .04* |
| Delta cortisol (μg/dL) | 8.36 (6.54-9.98) | 9.89 (8.13-11.0) |
- Note: Data are displayed as median (interquartile range [IQR]).
- Abbreviations: ACTH, adrenocorticotropic hormone; eACTH, endogenous adrenocorticotropic hormone.
- * Significant differences between treatments (P < .05).
4 DISCUSSION
Trazodone administration significantly decreased post-ACTH and delta serum cortisol concentrations in healthy dogs suggesting that trazodone might affect diagnostic results during ACTH stimulation testing. The lack of significant differences in serum eACTH concentrations also might suggest that the decreases in post-ACTH and delta ACTH cortisol concentrations occurred by an eACTH-independent mechanism. Although additional studies are needed, caution should be exercised when administering trazodone to dogs before ACTH stimulation testing.
The antagonism of serotonergic and alpha-1 adrenergic receptors by trazodone suppresses eACTH and serum cortisol and corticosterone concentrations in people and rodents. Similarly, we found that healthy dogs receiving trazodone had significantly lower post-ACTH and delta serum cortisol concentrations compared to those that received placebo. A suppression of cortisol secretion in response to ACTH may affect clinical decision-making, particularly diagnosis of either HAC or HypoA and dose adjustments for dogs undergoing treatment for HAC. Interestingly, the suppression of cortisol in our study was not accompanied by a decrease in eACTH. In a study in rodents, 5HT antagonism suppressed corticosterone secretion by impairing secretion of CRH and eACTH.13 Although our findings might reflect type 2 error, they also might be explained by an eACTH independent mechanism. Paracrine 5-HT-ergic control of adrenal steroidogenesis occurs in rodents and humans and is considered an adaptive response to stress.19 Furthermore, upregulation of the 5-HT signaling pathway has been demonstrated in people with adrenocortical hyperplasia and adenomas suggesting that this paracrine pathway may play a role in primary adrenal disease.19 Additional studies are needed to determine if a similar pathway exists for steroidogenesis in dogs.
No dog in our study showed a sufficient decrease in serum cortisol concentrations to be misclassified as having HypoA, but our study was performed in healthy dogs. We did, however, demonstrate suppression of delta cortisol concentrations of up to 50% in individual dogs. Although additional studies are needed, it is possible that dogs with existing adrenal insufficiency because of developing HypoA20 or critical illness-related cortisol insufficiency (CIRCI), which relies on delta cortisol concentrations as a means of diagnosis,21, 22 may be incorrectly classified as having HypoA because of suppression of the response to ACTH. Our findings regarding the effect of trazodone post-ACTH cortisol concentrations could have ramifications for the diagnosis of HAC in individual dogs. In our laboratory, for example, a post-ACTH cortisol concentration > 20.3 ug/dL (560 nmol/L) is consistent with HAC. With a potential decrease of 13.4% in the post-ACTH cortisol concentration as seen in some dogs in our study, any dog that would have had a post-ACTH concentration between 20.4 and 23.1 (561-6 nmol/L) could have a normal test result if receiving 8 mg/kg trazodone. Future studies examining whether the detected differences in post-ACTH and delta serum cortisol concentrations could result in a misdiagnosis in dogs suspected of having HypoA, CIRCI and HAC are warranted.
Our study had some limitations. The most notable limitation was a small sample size. Although a power calculation was performed, the projected difference in cortisol concentration was extrapolated from the human medical literature.14 As such, it is possible some of our findings reflect either type 1 or type 2 error, and larger studies are needed. Additionally, although a single 8-10 mg/kg dose of trazodone in dogs resulted in plasma concentrations that have been documented to cause HPA suppression in people after 1 hour, maximum plasma concentrations in dogs may occur later than 1 hour,6 and the dose we administered was at the upper end of the dosing range. The effects observed in our study may be dose-dependent because of variable receptor affinity for trazodone.23 Follow-up studies evaluating eACTH and cortisol at later timepoints reflecting maximum plasma concentrations are warranted in addition to studies evaluating both lower and higher doses of trazodone. Finally, our study investigated the influence of trazodone administration on the HPA axis in healthy dogs. However, evaluation of eACTH and serum cortisol concentrations are rarely performed on dogs without suspected pituitary or adrenal illness. Additional studies investigating the effect of trazodone on populations with HAC or actual or suspected adrenal insufficiency will be required to fully understand the effect of this drug on the diagnosis and management of these diseases.
In conclusion, single-dose trazodone administration in healthy dogs resulted in a suppressed response to ACTH stimulation compared to placebo. Because trazodone is commonly administered before veterinary clinical evaluation to minimize stress, this interference with the HPA axis has the potential to confound the evaluation of adrenal gland disease in dogs. Additional large-scale investigations in dogs with HAC or adrenal insufficiency are needed to define the full clinical effect of trazodone administration.
ACKNOWLEDGMENT
Funding provided by a research grant from the American Veterinary Medical Foundation (AVMF) and Veterinary Pharmacology Research Foundation (VPRF).
CONFLICT OF INTEREST DECLARATION
Authors declare no conflict of interest.
OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.
INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION
Approved by Auburn University IACUC, number 4091.
HUMAN ETHICS APPROVAL DECLARATION
Authors declare human ethics approval was not needed for this study.
