Abstract
Objectives
The objective of this retrospective study was to describe the clinical signs and diagnostic findings in cats with histopathologically confirmed adrenal neoplasms, and to assess correlations with survival data.
Methods
Study data were acquired by reviewing medical records for all cats diagnosed with adrenal neoplasms at seven referral institutions between 2002 and 2013. Inclusion criteria required a histopathologic diagnosis of an adrenal neoplasm (ante-mortem or on necropsy).
Results
Thirty-three cats met the inclusion criteria for the study. The most common presenting complaints included weakness (n = 12), respiratory signs (n = 4), blindness (n = 4) or gastrointestinal signs (n = 3). Laboratory abnormalities included hypokalemia (n = 18), alkalemia (n = 12), elevated creatine kinase (>3000, n = 5) and azotemia (n = 4). In addition, hypertension was noted in 13 cats. Thirty cats were diagnosed with cortical tumors (17 carcinomas, 13 adenomas) and three cats were diagnosed with pheochromocytomas. Twenty-five cats underwent tests to evaluate the function of the adrenal tumors; 19/25 cats had functional tumors (hyperaldosteronism [n = 16], hypercortisolemia [n = 1], high estradiol [n = 1], and hypersecretion of aldosterone, estradiol and progesterone [n = 1]). Twenty-six cats underwent adrenalectomy, one cat was medically managed and six were euthanized without treatment. Long-term survival postoperatively ranged from 4–540 weeks, with 20 (77%) cats surviving the perioperative period of 2 weeks. The only variable that was found to be negatively associated with survival was female sex. The most common complications noted during the perioperative period were hemorrhage and progressive lethargy and anorexia.
Conclusions and relevance
Surgical treatment for feline adrenal tumors (regardless of tumor type) resulted in good long-term survival. Given that pre- and postoperative hypocortisolemia was identified in this study, and, in addition, hypersecretion of more than one adrenal hormone occurred in one cat, adrenal panels prior to surgery may be beneficial as part of the preoperative work-up.
Introduction
Adrenal neoplasia is an uncommon finding in cats, accounting for 0.2% of all feline neoplasms. 1 The type of adrenal tumor is classified based upon its origin as either a cortical tumor (adenoma or adrenocortical carcinoma) or a medullary tumor (pheochromocytoma). 1 Cortical tumors are further divided into functional (producing excess aldosterone, cortisol, sex hormones, alone or in combination) or non-functional tumors. Fifty cases are reported in the veterinary literature of cats with primary hyperaldosteronism.2,3 These were mostly isolated case reports; however, one case series of adrenal hyperplasia and two case series of adrenal neoplasia have been published.3–5 In addition, a review of the literature reveals four cases describing pheochromocytomas and six cases of cats with cortical tumors producing excess sex hormones.6–15 This information is in contrast to the human literature where adrenal tumors are frequently an incidental, non-functional and benign finding.16–18
Primary hyperaldosteronism is the most frequently described clinical syndrome associated with adrenal tumors in cats.1,2 Perioperative mortality rates of up to 40% have been reported previously, with hemorrhage and sepsis being the most common cause of death. 5 However, cats that survived surgery had a good long-term prognosis with survival for up to 5 years. 5 A recent retrospective study suggested that the median survival time was affected by anesthesia time; however, no other factors were identified to be associated with prognosis. 3 Compared with previous studies, this study also suggested a higher survival rate for cats that underwent unilateral adrenalectomy for hyperaldosteronism (80%). 3 Factors that affect long-term prognosis have been evaluated in dogs that underwent adrenalectomy, but a similar study has not yet been conducted in cats.19–23
Approximately 20% of cats with spontaneous hyperadrenocorticism have functional adrenal tumors. 24 Treatment options for hyperadrenocorticism include medical management or surgery, in which bilateral adrenalectomy is associated with a high complication rate; all 10 cats from one study that underwent surgery experienced complications ranging from minor electrolyte abnormalities to pancreatitis, thrombosis and pneumonia. 25
Pheochromocytomas are extremely rare tumors in cats, with only four reported cases in the literature. 6,13–15 Clinical signs are typically due to excess catecholamine secretion or can be due to presence of a space-occupying mass. 6 Hypertension may or may not be consistently demonstrated as catecholamine secretions may be paroxysmal.1,6
Several case reports of functional adrenal neoplasia with excess sex hormone production are described in the veterinary literature. The first case report was in 1999 and described a cat with hyperprogesteronism. 9 Subsequent reports often describe cats with hypersecretion of more than one hormone (ie, progesterone, aldosterone or cortisol). Of the seven case reports since 1999, 3/7 cats had elevations in sex hormones other than progesterone (estradiol, testosterone, androstenedione, 17-hydroxyprogesterone).7,8,26 These cats were treated either surgically or medically; surgery resulted in favorable long-term outcome in three cats, but one cat died three days postoperatively. Medications administered included aminoglutethimide or trilostane. Clinical improvement was noted with both drugs for several months before return of clinical signs.7,8,26
The objective of this retrospective study was to describe the clinical signs, and diagnostic findings in cats with histopathologically confirmed adrenal neoplasms and assess correlations with survival data.
Materials and methods
Medical records for all cats diagnosed with adrenal neoplasms at each of the seven referral institutions between 2002 and 2013 were reviewed. Search terms that were used included ‘adrenal tumor’, ‘adrenal neoplasia’, ‘hyperaldosteronism’, ‘Conn’s syndrome’, ‘hyperadrenocorticism’ and ‘pheochromocytoma’. The inclusion criteria required a histopathologic diagnosis of an adrenal neoplasm (ante-mortem or on necropsy).
Data retrieved from the medical records included signalment, presenting signs, physical examination findings, complete blood count and chemistry profile results, urinalysis and indirect systolic blood pressure (on admission). Cats were considered hypertensive if the admitting systolic blood pressure was >160 mmHg. 27 Blood pressures were not uniformly rechecked. Tests to evaluate for functional adrenal tumors were performed at the discretion of the primary clinician. Results from these tests were recorded, which included adrenocorticotropic hormone (ACTH) stimulation tests, low-dose dexamethasone suppression tests (LDDSTs), plasma aldosterone concentrations (Diagnostic Center for Population and Animal Health, Lansing, MI, USA) and adrenal panels (University of Tennessee Veterinary Medical Center Diagnostic Laboratory Services, Knoxville, TN, USA). Hormones assessed with adrenal panels included pre- and post-ACTH concentrations of cortisol, androstenedione, estradiol, progesterone, 17-OH progesterone, aldosterone and testosterone. The diagnosis of hyperadrenocorticism was determined by using either the ACTH stimulation test or the LDDST. When the LDDST was used, cortisol levels that were ⩾1.5 μg/dl 8 h post-dexamethasone administration were considered diagnostic for hyperadrenocorticism. 28 When the ACTH stimulation test was performed, post-ACTH cortisol levels >19 μg/dl were considered diagnostic for hyperadrenocorticism.28,29 Protocols for ACTH stimulation test varied by clinician and institution in that while all cats had a pre- and post-ACTH cortisol sample (attained at 60 mins), an additional third sample was also attained in many cases, either at 30 or 90 mins post-ACTH administration. The dose used for cosyntropin was either 5 μg/kg intravenously or 125 μg/cat intramuscularly.
Therapeutic interventions were recorded in all cases. For cats where surgery was performed, duration of medical management prior to surgery was noted. The medical record was also reviewed for complications associated with surgery, including both intra- and postoperative complications. Telephone interviews with owners or referring veterinarians were conducted to attain follow-up information.
Statistical analysis
Median survival time of all cats undergoing adrenalectomy was calculated with Kaplan–Meier product limit estimates. The following factors were investigated to determine their influence on overall survival time using log rank statistics: sex, breed, tumor type (adenoma, adenocarcinoma, pheochromocytoma), tumor size (<2 cm or ⩾2 cm), weakness as a presenting sign, blood pressure (<160 or >160 mmHg), potassium concentration on presentation (<3 or ⩾3 mEq/l), aspartate aminotransferase (<70 or ⩾70 u/l), blood urea nitrogen (<30 or ⩾30 mg/dl), bicarbonate (<25 or ⩾25 mmol/l). Cats were censored in the survival statistics if they were still alive or died of causes unrelated to adrenal disease at the end of the study period. Descriptive statistics in Table 1 were calculated in the standard method. All statistical evaluations were performed with SAS version 9.2. Values of P <0.05 were considered significant.
Results of preoperative complete blood cell count, serum biochemistry profile and blood pressure measurement
CK = creatine kinase; AST = aspartate aminotransferase; BUN = blood urea nitrogen; BP = blood pressure; WBC = white blood cell
Results
Signalment and clinical findings
Thirty-three cats met the inclusion criteria for the study. Mean age at the time of initial evaluation was 12 years (range 5–17 years). Twenty-two cats were castrated males, one was an intact male and 10 were spayed females. Of the 33 cats in the study, 23 cats were domestic shorthairs, four were domestic longhairs, and there were three Siamese, one Persian, one Bombay and one Ocicat.
Weakness was the predominant presenting sign, characterized by hindlimb weakness, cervical ventroflexion, lethargy or plantigrade stance in 12/33 (36%) cats. Other clinical signs included blindness in 4/33 (12%) cats, respiratory signs in 4/33 (12%), gastrointestinal signs in 3/33 (9%), and polyuria and polydipsia in 1/33 (3%) cat (Table 2). The respiratory signs were dyspnea (three cats) and sneezing (one cat). Clinical signs associated with blindness included hyphema, retinal detachment or hemorrhage. In addition, nine cats were referred for the following: 5/33 (15%) cats were referred for a palpable abdominal mass noted on examination by the referring veterinarian, 2/33 (6%) for further work-up of diabetes mellitus, one cat (3%) for consultation of cardiac disease and one cat (3%) for kidney disease.
Percentage of cats included in the study with each presenting complaint
Other included four cats that were referred for further evaluation of diabetes mellitus (n = 2), cardiac disease (n = 1), kidney disease (n = 1) and polyuria/polydipsia (n = 1).
Clinicopathologic data
Results from complete blood counts and chemistry profiles were available for all cats. Mild-to-moderate leukocytosis (characterized by mature neutrophila) was noted in five cats. Two cats were anemic on admission (packed cell volumes of 24% and 25%). Abnormalities noted in biochemical laboratory data included hypokalemia (potassium ⩽3 mEq/l; n = 20), high creatine kinase (>3000 U/l; n = 5), alkalemia (bicarbonate >24 mmol/l; n = 12) (Table 1). Nineteen cats were considered to have chronic kidney disease International Renal Interest Society (IRIS) stages 2 (serum creatinine >1.6 mg/dl) or 3 (serum creatinine >2.9 mg/dl). 30 Urine specific gravities ranged from 1.009–1.070. Blood pressure was measured in 23 cats; 16 (70%) of these cats were hypertensive. Three of the four cats presenting with blindness were hypertensive.
Diagnostic imaging
All cats enrolled in the study had an abdominal ultrasound and thoracic radiographs performed to look for evidence of metastatic disease during initial work-up. To further characterize the adrenal mass noted on ultrasound, and to assist with surgical planning, two cats underwent CT exams. Caudal vena caval invasion was suspected based on ultrasound in one cat, which was confirmed with CT and later in surgery. Another cat had questionable lung nodules that did not progress over 2 years. No other cats had evidence of metastatic disease on initial evaluation.
Endocrine testing
Ancillary diagnostics to test for function of adrenal tumors were not always performed consistently. However, 25 cats had one of the following tests performed: plasma aldosterone (n = 15), ACTH stimulation test (n = 4), LDDST (n = 3) or an adrenal panel (n = 3). The results of these tests showed that 19/25 cats had a functional tumor, which included hyperaldosteronism (n = 16), hypercortisolemia (n = 1), high estradiol concentration (n = 1) and hypersecretion of more than one hormone (aldosterone, estradiol and progesterone; n = 1). Interestingly, five cats also had low post-ACTH cortisol levels (1.11, 2.75, 1.23, 1.0, 1.7 μg/ml; reference interval 5–15 μg/ml) prior to surgery. Of the remaining two cats that had an ACTH stimulation test or an adrenal panel performed, both had normal post-ACTH cortisol levels (12.5 and 4.2 μg/ml, respectively).
Clinical outcome
Tumor distribution was noted to be right sided in 17 cats, left sided in 15 cats and bilateral in one cat. Tumor size varied from 0.5–4.0 cm in diameter. All cats had confirmed adrenal neoplasia based on histopathology results. Thirty (91%) had cortical tumors (17 carcinomas, 13 adenomas) and three (9%) had pheochromocytomas.
An adrenalectomy was pursued in 26/33 cats. In one cat, caudal caval venotomy in addition to adrenalectomy was performed. Surgery was scheduled after initial consultation for all cats, with the exception of one cat, where emergency surgery was needed owing to acute adrenal hemorrhage. The average time that medical management was instituted prior to surgery was approximately 2 weeks (1.97 weeks, range 0–48.0 weeks). Of the 26 cats that underwent surgery, 20/26 (77%) cats survived (defined as survival 2 weeks postoperatively). Causes of death included postoperative hemorrhage and refractory hypotension (n = 1), acute kidney injury (n = 1) and euthanasia (n = 4). The reasons for euthanasia were continued weakness, lethargy and anorexia postoperatively. Perioperative complications included pancreatitis (n = 1), progressive lethargy and anorexia (n = 7), and significant hemorrhage requiring blood transfusion (n = 7). Hypoadrenocorticism occurred in three cats during the postoperative period: one of these cats had low cortisol (pre- and post-ACTH stimulation) levels prior to surgery, and the other two cats were not tested prior to surgery. Cats that developed hypoadrenocorticism were treated with prednisone (or prednisolone) for an average of 1–3 months. Four cats had follow-up adrenal function panels performed (aldosterone or sex hormone panels), which demonstrated resolution of high hormone levels (aldosterone in three cats; progesterone and aldosterone in one cat).
The median survival time for cats that underwent surgery was 50 weeks (mean 61 weeks, range 0–149 weeks). Six of 26 cats that underwent surgery are still alive at the end of the study period (ranging from 52–530 weeks since surgery). For cats that died beyond the perioperative period, cause of death was not determined in three cases, death was due to potential metastatic disease in two cats and to unrelated disease in two cats (chronic cholangiohepatitis and chronic kidney disease). Whether or not cause of death was due to adrenal disease was not determined, as necropsies were not performed in these cases. Of the three cats lost to follow-up, one was transferred to the referring veterinarian’s practice the day following surgery, one was seen only for suture removal and one was lost to follow-up 8 weeks postadrenalectomy.
The results of the statistical analysis are summarized in Table 3, and in Figures 1 and 2. Cats that were hypokalemic on presentation (potassium <3 mEq/l) had a better survival rate (P = 0.0186). Female cats had a higher overall mortality rate that was statistically significant (P = 0.0005). No other variables showed statistical significance associated with survival.
Statistical analysis of the median survival time for different prognostic factors
BUN = blood urea nitrogen; AST = aspartate aminotransferase
Kaplan–Meier survival curve of cats comparing castrated males (CM; solid line) and spayed females (FS; dotted line)
Kaplan–Meier survival curve of cats with potassium concentration ⩽3 mEq/l (dotted line) or >3 (solid line) on presentation
Discussion
The most common presenting sign for cats in this study was weakness, followed by blindness, respiratory signs and gastrointestinal signs. Signs of weakness have been reported to likely occur when potassium concentrations are ⩽2.5 mmol/l.2 However, 8/12 cats in this study that presented for weakness had a serum potassium concentration of >2.5 mmol/l. Therefore, although the serum potassium concentration may contribute to weakness, it is likely not the only factor that caused weakness in these cats. Other factors such as anemia (noted in two cats on presentation) and subclinical hypoadrenocorticism (noted in five cats on presentation, prior to surgery) are likely contributors. Interestingly, neither hypokalemia or weakness on presentation affected survival in this population. In fact, cats with lower potassium levels (<3 mEq/l) actually had a slightly higher survival rate. Therefore, hypokalemia on presentation is not necessarily a negative prognostic indicator. Hypokalemia was most commonly associated with hyperaldosteronism in this study (15/20 cats). Of the remaining five hypokalemic cats, four were euthanized without endocrine function tests and one tested negative for hyperadrenocorticism (plasma aldosterone levels were not assessed).
Nineteen of 33 cats had evidence of chronic kidney disease, with four cats considered to have IRIS stage 3 chronic kidney disease, while 15 cats were considered to have IRIS stage 2 chronic kidney disease. Interestingly, only two cats were hyperphosphatemic, and both of these cats had IRIS stage 3 renal disease. The tendency for normal-to-low phosphorus concentration may be a frequent finding in azotemic cats with adrenal neoplasia. 31 This tendency is perhaps owing to escape from chronic mineralocorticoid-induced sodium retention. The volume expansion induced by sodium retention resets the proximal tubules to increase fractional clearance of calcium and phosphorus. In turn, the negative calcium balance promotes parathyroid hormone secretion and further hypophosphatemia occurs. 32 However, only three cats demonstrated hypernatremia in this study, which is similar to previous reports, where hypernatremia is not a consistent finding.2,5,33 Of the three hypernatremic cats, the phosphorus concentrations were normal or low (1.2, 4.1 and 4.7 mg/dl [reference interval 3.0–6.3 mg/dl]).
As discussed in previous reports, cats with adrenal tumors may hypersecrete more than one adrenal hormone.7–12 However, there was only one cat from this study that showed hypersecretion of more than one hormone (estradiol, progesterone and aldosterone). The hypersecretion of the sex hormones likely was significant in this case as endogenous ACTH levels were assessed in this cat prior to surgery (2.0 pmol/l), which improved postoperatively (85.7 pmol/l). Adrenal panels were performed in two additional cats, which showed hypersecretion of only one hormone (aldosterone in one cat, estradiol in the other). Of the seven cats that were tested for hyperadrenocorticism (LDDST or ACTH stimulation test), only one had test results supportive of hyperadrenocorticism, and this cat was also treated for diabetes mellitus. Of the remaining five cats that were diabetic in this study, all tested negative for hyperadrenocorticism. Although seven cats were tested for hyperadrenocorticism with either an ACTH stimulation test or an adrenal panel, only three LDDSTs were performed. Hyperadrenocorticism may have been under-diagnosed in this study as the majority of cats were tested using a less sensitive test. 28 In addition, cats with adrenal carcinomas may exhibit aberrant ACTH receptors and demonstrate a blunted response. 12
Hypoadrenocorticism was confirmed (based on ACTH stimulation test) in only three cats postoperatively; however, this condition may have been underappreciated as an additional 11 cats had clinical signs of progressive lethargy and anorexia, which may have been due to hypoadrenocorticism postoperatively (four of which were euthanized owing to these clinical signs). Additionally, low post-ACTH cortisol concentrations were identified in five cats prior to surgery. Three of these cats had hyperaldosteronism, one cat had hypersecretion of estradiol and one cat had hypersecretion of estradiol, progesterone and aldosterone. Potential mechanisms for hypoadrenocorticism include lack of enzymes within the tumor necessary to convert progesterone or estradiol to cortisol, excessive unbound cortisol (due to displacement from cortisol-binding protein) resulting in suppression of the hypothalamic–pituitary–adrenal axis, or lack of ACTH receptors within the neoplastic cells.12,34 Another possibility is negative feedback on ACTH secretion from an unmeasured adrenal hormone precursor such as corticosterone. 34 Given the preoperative low cortisol levels and possible hypersecretion of more than one hormone, adrenal panels and measurement of ACTH concentrations may be helpful preoperative tests in cats with adrenal neoplasia. A recent study showed that the DiaSorin immunoradiometric assay to measure ACTH is valid for use in cats. 35
Tumors >4–5 cm in size and acute adrenal hemorrhage have been shown to be negative prognostic indicators in dogs.20,21 In contrast, tumor size from this study was not found to be associated with a negative prognosis. Possible reasons for this discrepancy may be owing to the small sample size or species differences in natural progression of the disease. Only one cat presented with acute adrenal hemorrhage, which has not been previously reported in cats with adrenal tumors. Tumor thrombus is another factor that has been associated with negative prognosis in dogs, especially extensive tumor thrombi. 22 In general, tumor thrombi appear to be more commonly associated with pheochromocytomas in dogs.21–23 In a recent study by Barrera et al, 22 13/14 dogs with caval invasion had pheochromocytomas. In contrast, cats, in general, may have a lower incidence of tumor thrombi occurrence naturally, potentially owing to lower incidence of pheochromocytomas. Only one cat had evidence of thrombi associated with an adrenal tumor in this study. This cat had evidence of vascular invasion associated with adrenocortical carcinoma on histopathology.
The most common postoperative complication in this study was hemorrhage associated with surgery, which occurred in seven cats. Causes of hemorrhage postadrenalectomy are multifactorial. These tumors are usually highly vascular, can infiltrate the surrounding vasculature and are difficult to dissect owing to close proximity of major blood vessels. Right-sided tumors in particular are thought to be more difficult to remove owing to proximity and higher likelihood of invading the vena cava.21,36 However, there were no significant differences in survival with right- or left-sided tumors both from this study and previous studies. 3 Other complications included acute kidney injury (n = 1), pancreatitis (n = 1) and suspected thromboembolism (n = 1). Given that both hemorrhage and thromboembolism are complications encountered postoperatively, coagulation tests performed preoperatively may be helpful in anticipating these types of complications.
The main limitation of this study was the retrospective design. A single pathologist was not utilized for assessing histopathology. It is possible that malignancies were under- or over-represented in this population owing to various criteria used for assessing malignancy. In humans, characterization of adrenal cortical tumors can be challenging, and is preferably carried out by a pathologist with experience using the microscopic Weiss criteria.37,38
However, our conclusions on survival data based on tumor type were the same as two previous studies on feline adrenal tumors. When follow-up information was not included in the record, it was obtained (when possible) by contacting the referring veterinarians or the owners. Tests for adrenal tumor function were not consistently performed in all cases, and the functional test also varied when performed. Cats that were euthanized did not always have necropsies performed. While recruitment of cases from multiple institutions was beneficial to represent a varied population, it also lacks standardization of treatment.
Conclusions
The most common tumor type noted from this study was adrenocortical carcinoma. The most common presenting sign for cats with an adrenal tumor was weakness. Frequent laboratory abnormalities include hypokalemia, hypertension and alkalosis. The postoperative survival rate was 77% in this study, with favorable long-term survival. Pre- and postoperative hypocortisolemia was identified, and, in addition, hypersecretion of more than one adrenal hormone occurred in one case. The only variable found to be negatively associated with survival was female sex. Although statistically significant, it is unknown if this is a clinically significant finding, and further investigation is warranted.
Footnotes
Conflict of interest
The authors do not have any potential conflicts of interest to declare.
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.