Elevations in Sex Hormones in Dogs With Sudden Acquired Retinal Degeneration Syndrome (SARDS)

Introduction

Canine sudden acquired retinal degeneration syndrome (SARDS) is an idiopathic disorder resulting in rapid and irreversible vision loss in affected dogs. To diagnose the disorder, a triad of findings is required: a history of sudden vision loss, normal funduscopic examination, and extinguished retinal activity as determined by electroretinography (ERG).^1–3 Affected animals are typically middle-aged to older dogs, with females being overrepresented.^4,5 Any breed may be affected; however, a breed predilection for dachshunds, miniature schnauzers, and Brittany spaniels has been previously reported.^4,5 Despite being a well-recognized clinical disorder, little is known regarding the pathogenesis of SARDS. An association between heightened adrenocortical activity and SARDS has been documented on several occasions, although it is not clear whether the association is one of cause or effect.^2–7

Histopathological examination of retinas from animals with SARDS demonstrates loss of photoreceptor outer segments, with sparing of the inner segments in the early stages of the disease.^1,3,8 With progression of the disorder, the retina gradually atrophies, and widespread degeneration of the photoreceptors becomes evident.^1,3 This is characterized ophthamoscopically as retinal vascular attenuation, increased tapetal hyperreflectivity, and pallor of the optic nerve head.

Various studies have investigated the potential roles of antiretinal immunoglobulins, excitotoxic amino acids, and photoreceptor cell apoptosis in the pathogenesis of SARDS; however, the mechanism of this disease remains elusive.^5,9,10 It has been postulated that while apoptosis may be the final pathway resulting in photoreceptor death, the apoptotic event may be triggered by a steroid hormone or excitotoxin.¹⁰

Although variable in their occurrence and severity in many dogs, concurrent clinical and historical findings of polyuria (PU), polydipsia (PD), polyphagia, excessive panting, lethargy, obesity (often with dramatic weight gain preceding or around the time of blindness), and hepatomegaly are similar to findings in dogs with hyperadrenocorticism (HAC).^1–4,11 Although these clinical signs have been postulated to perhaps be stress induced and to resolve with time, clinical signs of PU/PD were found to not resolve over time in all dogs diagnosed with SARDS.¹² In one retrospective study, these concurrent clinical signs were often perceived by the owner as representing a poor quality of life and sub-sequently led to euthanasia in some cases.¹³

Serum biochemical abnormalities identified in many SARDS-affected dogs include hypercholesterolemia and elevations in serum alkaline phosphatase (ALP), aspartate aminotransferase (AST), and alanine aminotransferase (ALT) activities.^1,3,4 Isoenzyme analysis of ALP in six dogs with SARDS confirmed that the enzyme elevation was steroid induced, although none of these dogs had reportedly received exogenous corticosteroids.² In many cases, affected dogs also have a low urine specific gravity.⁴ These clinical findings, combined with abnormal adrenocorticotropic hormone (ACTH) stimulation and low-dose dexamethasone suppression test results reported in some dogs with SARDS, are suggestive of HAC.^4,6 Canine HAC recently has been recognized to be associated with elevations of adrenal steroid hormones not normally measured in clinical practice. ^14,15 These sex hormones also exhibit glucocorticoidlike activity, but they are not traditionally evaluated. Evaluation of serum concentrations of progesterone, 17- hydroxyprogesterone, androstenedione, testosterone, and estradiol—along with traditional cortisol profiles—may allow identification of HAC in a larger population of dogs with SARDS. In addition, this may help to elucidate whether or not adrenal steroid hormones are involved in the pathogenesis of retinal degeneration.

Because of the frequency of clinical signs of HAC in dogs with SARDS and inconsistent findings on traditional cortisol testing, we elected in this study to evaluate the elevation of sex hormones in dogs with SARDS. The objectives of this study were to identify and characterize steroid hormonal abnormalities and the clinicopathological profiles of dogs affected by SARDS. Additionally, we also planned to determine if clinical signs of HAC were associated with elevated sex-hormone values.

Materials and Methods

Inclusion criteria for dogs were a history of sudden, complete vision loss; an ophthalmic examination that was deemed as normal or, if subtle abnormalities were present, they were insufficient to explain the degree of vision loss; and extinguished bright-flash ERG performed at the time of diagnosis. Owners of dogs with a known previous diagnosis of SARDS were retrospectively contacted to participate in the study. Additionally, dogs newly diagnosed with SARDS according to the above criteria were enrolled in the study. All dogs were enrolled with the informed consent of their owners.

Dogs were subjected to a complete medical history and physical and ophthalmic examinations. Complete ophthalmic examination included vision assessment by menace and tracking responses, evaluation of pupillary light and dazzle reflexes, tonometry, slit-lamp biomicroscopy, and indirect ophthalmoscopy. Following pupillary dilation with tropicamide,^a dogs were dark-adapted for 30 minutes for ERG. A gold-foil contact lens^b was used as the positive electrode, and a reference electrode was placed under the skin proximal to the ipsilateral zygomatic arch. For the ground, a needle was inserted into the skin over the occiput. Full-field, bright-flash electroretinograms^c were performed without general anesthesia, and, on average, five readings per eye were obtained.

Laboratory assessment consisted of one or more of the following diagnostic tests: biochemical profile, complete blood count (CBC), urinalysis, ACTH stimulation test, and endogenous ACTH levels. For ACTH stimulation, baseline values for cortisol and sex-hormone analysis were obtained using prestimulation serum samples. Additionally, prestimulation plasma samples for endogenous ACTH were obtained. One hour after administration of 5 μg/kg of ACTH (Cosyntropin)^d intravenously, poststimulation serum samples were obtained. Previous studies have found that this protocol results in effective stimulation for maximal production of cortisol, sex hormones, and adrenocortical steroid intermediates at 1 hour.¹⁶ Samples were submitted on dry ice for analysis using previously validated assays.^17–19 Values were compared to laboratory reference ranges that have been established to account for gender and reproductive status.

Additionally, when possible, systemic blood pressure was also obtained by Doppler ultrasound techniques as previously described.²⁰ Systolic pressures were recorded, and five readings at a minimum were obtained and averaged. A mean systolic blood pressure >160 mm Hg was considered hypertensive.

The proportion of dogs with increases in one or more sex hormones, cortisol, or sex hormones only was calculated. The 95% confidence interval (CI) of the binomial proportion was calculated. For the analysis, PROC FREQ^e was used.

Results

Thirteen dogs were available for study. Nine dogs were spayed females, two were neutered males, and two were intact males. The ages of the dogs at the time of diagnosis ranged from 6 years to 11.5 years, with a median age of 9 years at the time of diagnosis. Breeds represented included Brittany spaniel (n=2), miniature schnauzer (n=2), and one each of West Highland white terrier, mixed-breed dog, dachshund, chow chow, German shepherd dog, Cairn terrier, Lhasa apso, Pomeranian, and Welsh springer spaniel. Duration of vision loss prior to the time of diagnosis ranged from <1 week to 4 months. No dogs in the study had a history of receiving exogenous corticosteroids.

All owners were questioned as to the presence of clinical signs consistent with a diagnosis of HAC. Of the 13 dogs, 11 had polyuria, 10 had polydipsia, nine had weight gain, seven had polyphagia, seven had panting, six had lethargy, and two had increased salivation. Physical examination findings were unremarkable for four dogs. The most common clinical findings were mild to moderate obesity (n=8). Two dogs demonstrated cardiac abnormalities on examination. One dog had a grade III/VI apical systolic murmur, and the other had a sinus arrhythmia with a first-degree atrioventricular block. Neither animal showed clinical signs related to these findings.

Discussion

Findings of this study support the previously reported predisposition for female, middle-aged dogs to develop SARDS.^4,5 The most common systemic clinical signs reported for dogs included PU, PD, polyphagia, and weight gain. All dogs were presented with two or more clinical signs consistent with HAC. A clinical sign not previously reported, but noted in this subset of SARDS cases was a history of increased salivation. The most common abnormality on physical examination was obesity. Ophthalmic examination demonstrated a greater percentage of fundic abnormalities than previously reported.⁴ Ten of 13 dogs were found to have clinical signs consistent with mild retinal degeneration, likely because of the advanced age (>10 years) of several of the dogs and the duration of time (up to 4 months) from vision loss to diagnosis in several cases.

Frequent laboratory findings included elevated total protein, increased ALP, and proteinuria. Four of 10 dogs were also found to be hypertensive, making this an important diagnostic test in evaluating future dogs diagnosed with SARDS. Hypertension is frequently associated with an underlying systemic disease.²¹ Hypertension in dogs is most commonly associated with chronic renal disease (especially protein-losing renal disease), hyperadrenocorticism, and diabetes.²¹ The prevalence of hypertension in dogs with HAC has been reported to be as high as 80%.^21,22 Because of the retrospective nature of this study, not all dogs underwent blood pressure screening or urinalysis, and quantification of urine protein was not available. Therefore, it remains unclear whether the hypertension in these dogs was associated with HAC, protein-losing nephropathy, a combination of these abnormalities, or another concurrent condition.

The majority (11 of 13) of SARDS cases were found to have elevations in one or more sex hormones. Of these elevations, the most common were in 17-OH progesterone and progesterone. Additionally, cortisol was elevated in nine of 13 dogs. All of the dogs with increased cortisol had increased poststimulation values. Importantly, stimulated values (basal versus feedback) give the most information in regard to adrenal activity.²³ The majority (nine of 11) of dogs demonstrated an increase in both cortisol and sex hormones. A minority (three of 13) of dogs exhibited only an increase in adrenal sex hormones, and these dogs would not be identified on routine HAC screening. Clinical signs of HAC were associated with abnormal stimulation values in 92% of cases. Only one dog was found to have clinical signs of HAC with normal ACTH stimulation results.

Previous studies have shown that ACTH stimulation can be expected to detect 83% of dogs with HAC.²⁴ Although it has been suggested that many dogs with SARDS have concurrent HAC, results have been variable.^6,7 The ACTH stimulation results and correlation to clinical signs in this study show consistent evidence for HAC in SARDS. The small percentage of dogs demonstrating only increases in sex hormone values likely represents those with an atypical form of HAC. Dogs with atypical HAC are difficult to diagnose, because results of routine testing by ACTH stimulation or low-dose dexamethasone suppression are normal. However, clinical and hematological evidence of glucocorticoid excess is often present in these dogs.¹⁵ The finding of atypical HAC in a small population of SARDS dogs may account for the previously reported variability in HAC diagnosis.

Studies of atypical HAC have identified a pattern of hormone abnormalities characterized by normal to subnormal cortisol with increased adrenal sex hormones.^15,23 Progesterone and 17-OH progesterone are major precursors in the steroid pathway to the production of cortisol and androgens.^17,23 The majority of dogs (as in this study) with atypical HAC demonstrate increased concentrations of these two precursors. Fewer dogs demonstrated increases in aldosterone and estradiol, and, rarely, elevations in testosterone are identified.^14,23 Increases in adrenal sex hormone production could be the result of an enzyme deficiency (such as 11β-hydroxylase or 21β-hydroxylase) or a disruption of the enzymatic pathway necessary for cortisol production.^17,23,25 Either scenario would lead to the accumulation of cortisol intermediates and shunting to androgen synthesis.^17,23,25,26 Alternatively, elevated progestins or estradiol could result in a subnormal cortisol response.^23,27

Adrenal sex hormones have been reported to be elevated in dogs with pituitary-dependent hyperadrenocorticism.^15,26 Atypical HAC has been reported in association with congenital adrenal hyperplasia of humans.²⁸ In this disorder, cortisol synthesis is deficient, resulting in a lack of negative feedback on ACTH.^23,28 Excessive steroid precursors accumulate and lead to increased adrenal androgen production. ^23,28 A similar pattern of hormone release was described in dogs with adrenal tumors or Alopecia-X syndrome and in ferrets with adrenocortical disease.^23,29 Because the majority of HAC cases exhibit increased 17- OH progesterone, previous studies have attempted to determine if detection of this adrenal intermediate could aid in the diagnosis of HAC. Studies have found that measurement of 17-OH progesterone is a useful aid to the confirmation and treatment of HAC in both typical and atypical cases.^15,30 The authors also advocated running the entire adrenal sex hormone panel, as one particular hormone was not consistently increased after ACTH stimulation.^17,26

The systemic clinical signs in dogs with atypical HAC can be attributed to multiple factors. Sex hormones have been reported to be associated with glucocorticoid-like activity.^31,32 Sex hormones such as progesterone have an intrinsic glucocorticoid action, are capable of suppressing the hypothalamic-pituitary-adrenal axis, and have an affinity for the glucocorticoid receptor.^23,31,32 Also, 17-OH progesterone may increase the bioavailability of cortisol by displacing it from cortisol-binding protein.^23,27,30,33

Normal to high ACTH values in nine dogs were considered inappropriate given the degree of adrenal hyperactivity. These data support a potential abnormality at the level of the pituitary. Further evaluation of adrenal morphology by abdominal ultrasound and possibly MRI of the brain could aid in localizing the abnormality in the hypothalamic-pituitary- adrenal axis.

Directions for future study include determining if a link exists between photoreceptor apoptosis and increased sex hormones in SARDS. Steroid hormones have been shown to induce apoptosis in many cell types. Systemic increases in steroid hormones possibly trigger an apoptotic cascade.¹⁰ Glucocorticoid receptors in the chick retina have been shown to bind progesterone with a high affinity.³⁴ Pregnenolone sulfate is the most abundant neurosteroid found in the brain and retina, and estrogen receptors have been demonstrated on the rods, cones, and neuronal cells of the retina.^35–37 Potential mechanisms for steroid-induced apoptosis include induction of increased glutamate levels, which are excitotoxic and could trigger apoptosis.^5,10,38 Glutamate has been shown to be increased in SARDS affected eyes compared to eyes affected with progressive retinal atrophy.⁵ Additionally, pregnenolone sulfate has been shown to modulate N-methyl-D-aspartic acid (NMDA) receptors in the retina, leading to increased intracellular calcium and inducing cellular apoptosis.^36,37

Limitations of this study include a small number of dogs and inconsistent clinical testing (CBC, urinalysis, and blood pressure screening) because of its retrospective nature. Additionally, the study is limited by the possible influence of exogenous factors, such as stress, on the results of hormone testing. As with cortisol, 17-OH progesterone and other sex hormone levels likely can be increased secondary to stress.¹⁵ Elevated adrenocortical activity from physiological stress has been documented in several nonadrenal disease states in dogs.^39,40

Approximately 15% of dogs with nonadrenal illness have ACTH response tests consistent with HAC, while >50% may have an abnormal low-dose dexamethasone suppression test.^39,40 Additionally, the specificity for 17-OH progesterone as a diagnostic test for HAC may be as low as 71%, and chronic illness may lead to false-positive test results.^29,41

A study by Behrend et al evaluated 17-OH progesterone in dogs with nonadrenal neoplasia compared to dogs suspected of having HAC.⁴² In that study, 31.4% of dogs with neoplasia were found to have an increase in 17-OH progesterone compared to normal dogs.⁴² However, unlike the present study, none of the dogs with neoplasia were suspected to have HAC because of the lack of supportive historical, clinical, and laboratory findings.⁴² Additionally, the authors of the previous study could not rule out the possibility of exogenous steroid administration or the concurrent presence of HAC in studied dogs. Although one dog with SARDS subsequently developed lymphoma, no dogs enrolled in the present study had clinical, physical, or historical findings suggestive of chronic illness.

To address the above limitations, future studies should be performed to assess HAC at the time of SARDS diagnosis and several months after diagnosis, using a control group of systemically ill dogs. Lastly, the finding of systemic hypertension in four of the 10 dogs in which it was measured cannot be fully interpreted, because concurrent testing (other than HAC testing) to rule out causes of hypertension was not performed.

Therapeutic options for dogs with persistent clinical signs and hormone and laboratory abnormalities include melatonin, which has been shown to decrease steroidogenesis, trilostane, or mitotane. Melatonin is a neuro-hormone that is released by the pineal gland. High levels have been shown to decrease steroidogenesis.⁴³ One theory for melatonin’s effect on steroid hormones is that it inhibits gonadotropin-releasing hormone (GnRH) by the hypothalamus. Therefore, this is a safe alternative that may alter serum sex-hormone concentrations.⁴³ Melatonin also has been shown to inhibit aromatase and 21-hydroxylase enzymes, which can result in suppression of estradiol and cortisol, respectively.⁴⁴ Careful case selection is necessary before considering the use of potent therapies, such as mitotane, for HAC.

Conclusion

Our findings strongly suggest that further testing is indicated for dogs with SARDS in order to rule out the presence of concurrent, possibly atypical, HAC. Clinical signs of HAC are common in these dogs. Although routine HAC screening should identify a majority of affected cases, the dogs that have atypical HAC would not be identified despite clinical signs and hematological abnormalities. Dogs with SARDS may benefit from ACTH stimulation testing to evaluate cortisol and adrenal sex hormones, continued monitoring of clinical signs, abdominal ultrasonography, and possibly MRI of the brain. Based on the findings of this study, screening tests for systemic hypertension and proteinuria may also be warranted.