Presumed pituitary apoplexy in 26 dogs: Clinical findings, treatments, and outcomes

Abstract Background Pituitary apoplexy refers to hemorrhage or infarction within the pituitary gland resulting in acute neurological abnormalities. This condition is poorly described in dogs. Objectives To document presenting complaints, examination findings, endocrinopathies, magnetic resonance imaging (MRI), treatments, and outcomes of dogs with pituitary apoplexy. Animals Twenty‐six client‐owned dogs with acute onset of neurological dysfunction. Methods Retrospective case series. Dogs were diagnosed with pituitary apoplexy if MRI or histopathology documented an intrasellar or suprasellar mass with evidence of hemorrhage or infarction in conjunction with acute neurological dysfunction. Clinical information was obtained from medical records and imaging reports. Results Common presenting complaints included altered mentation (16/26, 62%) and gastrointestinal dysfunction (14/26, 54%). Gait or posture changes (22/26, 85%), mentation changes (18/26, 69%), cranial neuropathies (17/26, 65%), cervical or head hyperpathia (12/26, 46%), and hyperthermia (8/26, 31%) were the most frequent exam findings. Ten dogs (38%) lacked evidence of an endocrinopathy before presentation. Common MRI findings included T1‐weighted hypo‐ to isointensity of the hemorrhagic lesion (21/25, 84%), peripheral enhancement of the pituitary mass lesion (15/25, 60%), brain herniation (14/25, 56%), and obstructive hydrocephalus (13/25, 52%). Fifteen dogs (58%) survived to hospital discharge. Seven of these dogs received medical management alone (median survival 143 days; range, 7‐641 days) and 8 received medications and radiation therapy (median survival 973 days; range, 41‐1719 days). Conclusions and Clinical Importance Dogs with pituitary apoplexy present with a variety of acute signs of neurological disease and inconsistent endocrine dysfunction. Dogs that survive to discharge can have a favorable outcome.

Conclusions and Clinical Importance: Dogs with pituitary apoplexy present with a variety of acute signs of neurological disease and inconsistent endocrine dysfunction.
Dogs that survive to discharge can have a favorable outcome.

K E Y W O R D S
adenoma, carcinoma, endocrionopathy, hemorrhage, magnetic resonance imaging, suprasellar, survival

| INTRODUCTION
Pituitary apoplexy is a well-described clinical syndrome in humans, which results from acute hemorrhage or infarction within the pituitary gland. 1 Typically, this condition occurs with a preexisting insidious pituitary adenoma [2][3][4] and affects an estimated 2.0%-14.1% of patients with such tumors. 2,5,6 Human patients experience a constellation of acute signs of neurological and endocrine disease, including severe headache, nausea, vomiting, visual impairment, altered levels of consciousness, cranial nerve palsies, hormonal disturbances, and sudden death. 2,7 Several predisposing factors have been identified in people, including pregnancy, anticoagulant therapy, thrombocytopenia, and gonadotropinreleasing hormone treatment. 5 The etiology of pituitary apoplexy, however, remains elusive, with explanatory theories including tumor growth outstripping its vascular supply and tumor-induced compromise and compression of the thin intrasellar vascular network. 2,3,8 Recommended treatment is controversial and there are varied opinions on transsphenoidal surgical decompression vs conservative management based on the severity of clinical signs. 9 The case fatality rate of pituitary apoplexy in humans ranges from 1.6% to 1.9%, with most patients needing long-term hormonal replacement therapy to manage secondary adrenal insufficiency or central diabetes insipidus. 4,5,10,11 By comparison, there is a paucity of literature on pituitary apoplexy in dogs and cats. Four case reports with a combined total of 6 dogs described acute onsets of clinical signs, including vomiting, altered consciousness, behavioral changes, vision loss, seizures, internal ophthalmoplegia and collapse. [12][13][14] A recent case series of 19 dogs (including 2 of the previously reported 6 dogs) described mainly altered mentation, vestibular dysfunction and seizures. 15 While computed tomography, histopathology, and postmortem findings are described in this literature, in addition to magnetic resonance imaging (MRI) findings of pituitary apoplexy in a cat, 16 published MRI findings in dogs with pituitary apoplexy appear to be limited to 8 cases. 14,15 Treatment options and outcome data are also limited in the veterinary literature, as many dogs are euthanized at the time of diagnosis or fail to survive to hospital discharge. 13,15 The objective of our study was to augment preexisting literature on pituitary apoplexy in dogs by describing a larger cohort of dogs, with a focus on historical comorbidities and endocrinopathies, presenting complaints from owners, physical as well as neurological exam abnormalities, and clinicopathologic abnormalities. Additional goals were to document MRI findings and to detail attempted therapies and outcomes for pituitary apoplexy in dogs.

| Clinical information
Dog signalments, histories, presence of preexisting endocrinopathies or other diseases, physical and neurological exam abnormalities, diagnostic imaging findings, postmortem pathology, treatments initiated, and outcomes were obtained from the medical record. A modified Glasgow coma score (MGCS) was determined retrospectively for each dog after review of the medical record. 17,18 Additionally, the results of clinical pathology testing, including complete blood counts, serum chemistries, urinalyses, coagulation testing, endocrine evaluation, and other diagnostic tests were recorded. Methods of follow-up included recheck examinations at our hospital or with referring veterinarians, and telephone interviews with owners. Survival was determined as the time from diagnosis until death from any cause.

| Statistical analyses
Categorical data were reported as frequency of occurrence. Ordinal data (MGCS) were reported as medians with ranges. Continuous data were listed as means with standard deviations and ranges, after D'Agostino and Pearson testing showed these distributions to be normal. The MGCS between groups were compared with a Mann-Whitney test. Associations between clinical signs, brain herniation, hyperosmolar treatment and survival to hospital discharge were analyzed using a Fisher's exact test. Overall survival time was described using the Kaplan-Meier method and comparisons between groups were made using a log-rank test. All analyses were performed with Prism (version 9.0.0, Graphpad Software, La Jolla, California) and a P value <.05 was considered significant.    were no significant associations detected between vomiting, ptyalism or vestibular dysfunction and brain herniation (Table 3). Endocrine testing included ACTH stimulation (4 dogs), baseline cortisol (4), total thyroxine (T 4 ) and thyroid stimulating hormone (4), total T 4 (3), lowdose dexamethasone suppression (1), ACTH concentration (1), total and free T 4 (1), and a thyroid panel with total and free T 4 , thyroid stimulating hormone, triiodothyronine (T 3 ) and autoantibodies to T 4 and T 3 (1). Analysis of cerebrospinal fluid collected from the cerebellomedullary cistern was performed in 5 dogs. All analyses were abnormal and showed pleocytosis in 4 dogs, albuminocytologic dissociation in 1 dog and evidence of erythrophagocytosis in 2 dogs (Table S1).

| RESULTS
The median survival time for the entire cohort was 11.5 days (range, 1-1719 days; Figure 2A). Eleven dogs (42%) were hospitalized for supportive care, including hyperosmotic agents, glucocorticoids, intravascular fluid therapy, and anticonvulsant medications before or after the MRI, but did not have a favorable response (eg, deterioration despite supportive care or inadequate improvement suitable for discharge as determined by the clinician or owner) and humane euthanasia during hospitalization was elected. The median MGCS of these dogs was 15 (9)(10)(11)(12)(13)(14)(15)(16)(17). One of these dogs had an MRI and a subsequent postmortem examination, that revealed a pituitary adenoma with intralesional hemorrhage. Another dog was euthanized without an T A B L E 2 Clinicopathologic and magnetic resonance imaging findings in dogs with pituitary apoplexy. Defined as contrast enhancement that is more pronounced for the periphery of the mass lesion as compared to the centrally. The remaining 15 dogs (58%) survived to hospital discharge. The median MGCS of these dogs was 16 (10)(11)(12)(13)(14)(15)(16)(17)(18), which was significantly different from the MGCS of dogs not surviving to discharge (P = .014). We found no statistically significant associations between clinical signs or radiographic evidence of brain herniation and outcome (  24 and it is possible that compression of the thalamus from suprasellar tumors can also cause vestibular dysfunction. A recent case series of dogs with pituitary apoplexy also described altered mentation and vestibular dysfunction as common clinical signs. 15 However, in contrast to our study, seizures were a prominent clinical sign while postural or gait abnormalities and hyperpathia were uncommon and signs of gastrointestinal dysfunction and pyrexia were not mentioned. 15 Almost half (46%) of the dogs in our study had an increased rectal temperature. In humans, 30% of patients with intracerebral hemorrhage will have a fever. 25,26 Theories for neurogenic fever include prostaglandin production, hemotoxic damage to the hypothalamus, and inhibition of blood supply to the midbrain, which might disinhibit thermogenesis. 27 Fever has been shown to be negatively associated with a return to function and survival after intracerebral hemorrhage in humans. 28 Although the mean systolic blood pressure of the dogs in our study was similar to reference values reported for clinically normal F I G U R E 2 Survival curves for (A) the entire cohort of dogs with pituitary apoplexy and (B) those dogs that survived to hospital discharge. Dogs that received radiation therapy in addition to medical therapy (solid line) survived longer than those receiving medical therapy alone (dashed line, P = .008). Ticks represent dogs that were censored during analysis.
dogs in the literature, there was considerable variability in our cohort and 6 dogs had values that would be considered hypertensive or severely hypertensive according to consensus guidelines (>160 mm Hg). 29 It is possible that pre-existing hypertension precipitated or contributed to the hemorrhage associated with apoplexy in some of these cases. Systemic hypertension has been suggested to be a predisposing factor for pituitary apoplexy in humans, [30][31][32] although several other reports, including a large, case-control study found no association and has called this relationship into question. 33,34 The causative role of blood pressure in dogs is difficult to disentangle from a potential compensatory response to an acute intracranial increase in volume, situational hypertension because of anxiety, and other contributing comorbidities.
The severity of the signs of neurological dysfunction varied in the dogs in our study. This parallels reports of pituitary apoplexy in humans, in which presenting complaints range from headache to coma, and there are described grades of severity. 35    However, because of the study's retrospective nature, there was no standardization of dosing or administration and we suspect this association is primarily related to the use of these agents in cases with more severe clinical signs and advanced disease.
In humans, the decision to pursue decompressive surgery vs medical management, and the timeframe in which surgery should be performed, are controversial. In general, visual impairment and neurological deterioration are indications for surgery, as there is more likely to be visual improvement if the surgery is performed within 7 days of symptom onset. 67 Transsphenoidal surgery has been well described in dogs as an effective treatment for pituitary-dependent hyperadrenocorticism. [68][69][70][71][72][73] As in humans with pituitary apoplexy, this approach could be considered in dogs with this condition that are stable enough for general anesthesia.
This study has a number of limitations, many associated with its retrospective nature. Exclusion of dogs with clinical signs that had been present for longer than 1 week may have eliminated some cases that had an acute apoplectic event, but were not presented to our hospital within this initial time period. There was no standardization of the diagnostic tests performed, including coagulation testing, CSF