Surgical Treatment of Adrenocortical Tumors: 21 Cases (1990-1996)
Twenty-four adrenocortical tumors were surgically removed from 21 dogs. Histopathological exami- nation confirmed 18 carcinomas and six adenomas. Four dogs died in the perioperative period. Fifteen of the 17 dogs that survived the perioperative period had long-term resolution of their clinical signs. Two dogs with incompletely resected tumors were treated with mitotane to control their clinical signs. Overall median Kaplan-Meier life-table survival for dogs with carcinomas was 778 days (range, one to 1,593 days). Median survival for dogs with adenomas was not reached (range, 11 to 730 days). Histopathological diagnosis, histopathological cellular features, age of the dog, and tumor size were not prognostic of outcome. J Am Anim Hosp Assoc 2001;37:93-97.
Christine R. Anderson, DVM, MS Stephen J. Birchard, DVM, MS, Diplomate ACVS
Barbara E. Powers, DVM, PhD, Diplomate ACVP Gerardo A. Belandria, DVM Charles A. Kuntz, DVM, MS, Diplomate ACVS
Stephen J. Withrow, DVM, Diplomate ACVS, Diplomate ACVIM
RS
From the Departments of Clinical Sciences (Anderson, Kuntz, Withrow) and Pathology (Powers), College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Ft. Collins, Colorado 80523; the Department of Veterinary Clinical Sciences (Birchard), College of Veterinary Medicine, The Ohio State University, Columbus, Ohio 43210; and the School of Veterinary Medicine (Belandria), Tufts University, North Grafton, Massachusetts 01536.
Introduction
Adrenocortical tumors are the cause of 15% to 20% of cases of canine hyperadrenocorticism.1 Medical therapy with o,p’-DDD (mitotane) has been shown to be less effective in dogs with adrenal-dependent hyper- adrenocorticism (ADH) than in dogs with pituitary-dependent hyper- adrenocorticism (PDH).2 Adverse side effects of mitotane, including anorexia, vomiting, diarrhea, and adrenocortical insufficiency, have been reported in 59% of dogs, and relapses of hyperadrenocorticism were seen in 63%.3 Treatment with ketoconazole is effective but may be cost prohib- itive in large dogs.4 Adrenalectomy is generally considered to be the treat- ment of choice for ADH. However, survival times, postoperative mortality, and metastatic rate following adrenalectomy are conflicting in previously published reports.5-7 The purposes of this study were to evalu- ate response to adrenalectomy in a series of 21 dogs with adrenocortical tumors and to identify factors that affect outcome. Factors evaluated included histopathological diagnosis, histopathological cellular features, tumor size, microscopic vascular invasion, surgeon’s ability to completely resect the tumor, and age of patient.
Materials and Methods
Records of all dogs that underwent adrenalectomy for ADH between 1990 and 1996 at Colorado State University (n=7), Tufts University (n=6), and The Ohio State University (n=8) were reviewed for signal- ment, clinical signs, diagnostic tests performed, surgical procedure, tumor size and location, histopathological diagnosis, postoperative management, and outcome. Dogs were excluded if follow-up information was not avail- able. Histopathology slides from 18 cases were reviewed by a single pathologist (Powers). The main criterion used for determining malignancy was histopathological evidence of invasive behavior. Tumors were assessed for degree of nuclear pleomorphism, percentage of tumor necro- sis, mitotic rate in 10 high-powered (400X) fields, and presence or absence of vascular invasion. Follow-up was not performed at regular intervals in all cases. Outcome was determined by telephone calls to own- ers or referring veterinarians. Necropsy was not routinely performed. All
dogs included had histopathologically confirmed adrenocor- tical tumors. Hyperadrenocorticism was diagnosed by a vari- ety of methods, including adrenocorticotropic hormone (ACTH) stimulation test and low-dose dexamethasone sup- pression (LDDS) test. Adrenal-dependent hyperadrenocorti- cism was differentiated from PDH by using a high-dose dexamethasone suppression (HDDS) test, endogenous ACTH concentrations, abdominal ultrasound, or a combina- tion of these methods.
Statistical Analysis
Disease-free interval was defined as the interval between surgery and evidence of local recurrence or metastasis. Sur- vival time was defined as the interval between surgery and death. Cause of death was classified as tumor related or non- tumor related. Two dogs were lost to follow-up at 11 and 322 days. Survival times were determined by use of the Kaplan- Meier life-table analysis.
Cox proportional hazards univariate analysis was per- formed for histopathological diagnosis, tumor size, and age of patient. Tumors were classified histopathologically as adenomas or carcinomas based on absence or presence of microscopic invasion, other features of malignancy, or both.
Results
Signalment, History, Clinical Signs
The median age of the dogs was 11 years (range, seven to 15 yrs). Sixteen dogs (76%) were female and five (24%) were male. One of the female dogs was intact; all other dogs were neutered. The median weight was 20 kg (range, four to 51 kg). Breeds represented were mixed-breed (n=5), Labrador retriever (n=2), standard schnauzer, English springer spaniel, toy poodle, Finnish spitz, Rhodesian ridgeback, beagle, Newfoundland, Doberman pinscher, basset hound, Shetland sheepdog, bichon frise, cocker spaniel, golden retriever, dachshund, and Cairn terrier (n=1 each).
Clinical signs were consistent with hyperadrenocorticism in 20 dogs and had been present for a duration of two weeks to 19 months (median, 2.5 mos). Signs included polyuria and polydipsia (67%), polyphagia (33%), dermatological changes (33%), weakness/lethargy (19%), abdominal enlargement (14%), recurrent urinary tract infections (14%), and panting (5%). Two dogs had documented systemic hypertension. The remaining dog had a concurrent renal tubular adenocarcinoma and presented for hematuria.
Laboratory Findings
Complete blood count (CBC) changes included lymphopenia (67%), mature neutrophilia (19%), eosinopenia (14%), and monocytosis (5%). Serum biochemical analysis revealed ele- vated serum alkaline phosphatase (ALP) in 90% of the dogs, elevated cholesterol (76%), elevated triglycerides (19%), ele- vated alanine transferase (10%), and decreased blood urea nitrogen (10%).
Endocrine Screening Tests
An ACTH stimulation test was performed in 12 dogs and was consistent with hyperadrenocorticism in four (33%) dogs. A LDDS test was performed in 11 dogs and was con- sistent with hyperadrenocorticism in all 11.
Endocrine Differentiation Tests
A HDDS test was performed in nine dogs. Suppression was not seen in any of the nine dogs. Suppression was defined as a plasma cortisol concentration less than 50% of baseline concentration or an absolute concentration of less than 1.4 ug/dl. Endogenous ACTH concentrations were measured in five dogs and were undetectable (i.e., less than 20 pg/ml) in four (80%). The remaining dog had an endogenous ACTH concentration of 28 pg/ml (reference range, 20 to 40 pg/ml).
Imaging Studies
Abdominal ultrasonography was performed in 18 dogs, and adrenal masses were identified in all 18. Tumor invasion of the vena cava was diagnosed by ultrasonography in the two dogs in which invasion was identified at surgery. Abdominal radiography was performed in seven dogs. Two tumors (one adenoma and one carcinoma) were identified as soft-tissue densities (28%). None of the tumors were calcified. Hepatomegaly was seen in three (43%) dogs. Thoracic radio- graphs were taken in six dogs, and evidence of metastasis was not seen in any of these dogs.
Surgery and Postoperative Management
Eighteen (86%) dogs underwent unilateral adrenalectomy, and three (14%) dogs underwent bilateral adrenalectomy. A ventral midline abdominal approach was used in 20 dogs, and one dog required a ventral midline approach and a right paracostal approach for adequate exposure. Four dogs also underwent a unilateral nephrectomy. Nephrectomy was per- formed due to tumor invasion of the renal vein in two dogs, intraoperative laceration of the renal vein in one dog, and presence of a concurrent primary renal adenocarcinoma in one dog.
Adrenal tumors were identified at surgery in all 21 dogs. Nine (43%) tumors were in the right adrenal gland, nine (43%) were in the left adrenal gland, and three (14%) dogs had bilateral tumors. Tumor size ranged from approximately 2 cm3 to 6 cm3. Complete resection of the tumor was achieved in 19 dogs. In the other two dogs, the tumor invaded the caudal vena cava and was partially resected.
Postoperative medication varied. Fifteen dogs were treated with varying doses of prednisone on a decreasing schedule. Two dogs were treated with dexamethasone. One dog that had bilateral adrenalectomies was treated with pred- nisone and fludrocortisone acetate until death of unrelated causes 25 months after surgery. Fludrocortisone acetate was used on a short-term basis in four dogs that had electrolyte abnormalities in the immediate postoperative period.
Four of the 21 (19%) dogs died or were euthanized between days one and 14 postoperatively. One dog devel- oped ischemic necrosis of the jejunum for unknown reasons
and died on day three. Discoloration of the intestine had been noted during surgery. One dog had a hypoadrenocorti- cal (i.e., Addisonian) crisis 14 days postoperatively and was also diagnosed with dilated cardiomyopathy and was eutha- nized. Another dog died of multiple organ failure one day after surgery. Necropsy revealed multifocal renal infarction, glomerulonephritis, pulmonary thromboemboli, suppurative and necrotizing pancreatitis, and adrenocortical carcinoma pulmonary metastases. The remaining dog developed abdominal pain, vomiting, and azotemia and died four days after surgery at the referring veterinarian’s clinic. Necropsy was not permitted.
Another dog developed pancreatitis postoperatively but responded to medical management and was discharged from the hospital.
Pathology
Twenty-four adrenocortical tumors from 21 dogs were sub- mitted for histopathological evaluation. Twenty (83.3%) of the tumors were carcinomas, and four (16.7%) were adeno- mas. Of the three dogs with bilateral tumors, two dogs had bilateral carcinomas, and one dog had a carcinoma in one adrenal gland and an adenoma in the contralateral gland. Nuclear pleomorphism was absent or mild in 15 dogs and moderate or marked in three dogs. Percentage of necrosis ranged from zero to 90%, with a median of 15%. The mitotic rate varied from zero to three per 10 high-powered (400X) fields, with a median of one. Vascular invasion was present in 10 dogs, all of which had tumors that were carcinomas. All adenomas were well demarcated with no vascular inva- sion, and all carcinomas were poorly demarcated and inva- sive. Nuclear pleomorphism, percentage of necrosis, and mitotic rate had no prognostic significance.
Metastasis
The metastatic rate for dogs with carcinomas was one (5%) dog at time of surgery and two (9%) dogs long-term. Overall metastatic rate was 14%. Metastasis was not identified pre- operatively or intraoperatively in any of the dogs. One dog, which died in the postoperative period, had metastasis to the lung and thyroid vessels confirmed at necropsy. Another dog had recurrence of clinical signs 22 months after surgery, and hepatic metastasis was confirmed histopathologically follow- ing hepatic biopsy. The third dog had had chronic hyperten- sion and chronic renal failure and presented to the hospital 33 months after adrenalectomy for vomiting, dehydration, and anemia. This dog was euthanized after three days of hos- pitalization. Necropsy confirmed metastatic adrenocortical carcinoma on the serosal surfaces of the liver and pylorus.
Long-Term Outcome
Of the 17 dogs that survived the perioperative period, 15 (89%) had long-term resolution of clinical signs. The two dogs that had residual disease were treated with mitotane therapy to control clinical signs. The first dog was treated with an initial dose of 62.5 mg/kg body weight per day of mitotane divided into two doses daily. After three weeks, the
dose was decreased to 31.25 mg/kg body weight once every three days. Adrenocorticotropic hormone stimulation tests, performed at three- to six-month intervals, continued to be normal until 22 months after surgery. At this time, the ACTH stimulation test was consistent with hypoadrenocorticism, and the mitotane was discontinued permanently. Prednisone (0.25 mg/kg body weight daily) was administered from the time of surgery until the dog’s death of unrelated causes 31 months after surgery. The second dog with residual disease was treated with 50 mg/kg body weight per day of mitotane divided into two doses daily for two weeks after surgery. The dose was then decreased to 50 mg/kg body weight twice weekly. The owners discontinued treatment after two months. Additional testing was not performed. This dog was alive and asymptomatic for hyperadrenocorticism 11 months after surgery. The dog that had hepatic metastasis diagnosed 22 months after surgery was also treated with mitotane. This dog was euthanized three months later following an Addis- onian crisis.
Overall median Kaplan-Meier life-table survival time for dogs with carcinomas was 778 days (mean, 814 days; range, one to 1,593 days) [Figure 1]. Overall median survival time for dogs with adenomas was not reached (range, 11 to 703 days), as no dog died of disease or disease-related complica- tions. When dogs that died in the perioperative period, defined as days one to 14, were excluded from analysis, the median survival time for dogs with carcinomas was 992 days (mean, 1,083 days; range, 72 to 1,593 days). The difference between adenomas and carcinomas was not statistically sig- nificant (p equals 0.22). Neither histopathology, tumor size, nor age of patient were prognostic of outcome.
Discussion
Signalment, history, and serum biochemical and CBC results were consistent with previous reports.1,5-8 Sensitivity of 33% for the ACTH stimulation test was slightly lower than what has been reported for adrenocortical tumors.1,8 The LDDS test was more sensitive (100%) than the ACTH stimu- lation test in the dogs in this study and in others, but it is known to be less specific.1,8 The HDDS test was consistent with ADH in all dogs tested, but this test is inconclusive for definitively diagnosing ADH, because 20% to 30% of dogs with PDH will also fail to suppress following this test.8 Endogenous ACTH concentrations were nondetectable (less than 20 pg/ml) in four of five dogs tested, and the remaining dog had a nondiagnostic test result. Ultrasonography was beneficial in these patients for diagnosing adrenal masses and vena caval invasion if present. It is difficult to draw con- clusions regarding the sensitivity and specificity of these diagnostic tests, because the number of dogs in this study was low and not all tests were performed in all dogs. Other sources have evaluated these tests more thoroughly.1,8,9
Complete surgical resection was achieved in all dogs that did not have vena caval invasion of the tumor. The right adrenal gland is closer in proximity to the vena cava, and in both dogs with vena caval invasion, the tumor was in the right adrenal gland.
1
0.9
0.8
Percentage surviving
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
200
400
600
800
1000
1200
1400
1600
1800
Time (days)
Median survival time for dogs with adenomas was not reached. Median survival times for dogs with carcinomas was 778 days. In the literature, most studies have excluded dogs that died or were euthanized perioperatively from sur- vival analysis. If these dogs are censored in this study, the median survival time for dogs with carcinomas is 992 days. Emms et al. reported an average survival time of 18 months in 11 dogs that underwent unilateral adrenalectomy.6 Only eight of these dogs had adrenocortical tumors; the other three had adrenocortical hyperplasia. Three dogs that had died were excluded from survival analysis.6 Survival analy- sis was not performed by Scavelli et al. because of the large number of dogs that were euthanized or died during surgery (nine of 14 dogs).5 A median survival time of 17.5 months for 26 dogs discharged from the hospital was reported by van Sluijs et al.7 Eight dogs died in the perioperative period and were excluded from survival analysis.
In this study, the metastatic rate for dogs with carcinomas was 5% at the time of surgery and 14% long-term. In previ- ously published reports, Emms et al. found no metastases, but only five dogs in this series had carcinomas.6 Scavelli et al. found 50% hepatic metastasis at the time of surgery.5 The low rate of metastasis in this study may be due to better pre- operative diagnostics, resulting in dogs with metastasis not undergoing surgery and thus being excluded from this paper; or the low rate of metastasis may be due to the actual num- ber of eventual metastases being underestimated, because
necropsy was not performed in all patients.
Although the long-term outcome in these patients is often good, adrenalectomy is a technically difficult surgery, and perioperative complications are not uncommon. Periopera- tive complications that developed in the patients in this series included thromboembolism, pancreatitis, adrenal insuffi- ciency, and cardiac arrest.
In previously published reports, perioperative mortality ranged from 9% to 60%.5-7 In Scavelli et al.’s report, seven of 25 dogs were euthanized at the time of surgery due to metastasis, and eight additional dogs died or were eutha- nized due to postoperative complications. In the authors’ series, perioperative mortality was 19%, with none of these dogs euthanized intraoperatively. The two dogs with residual disease at surgery both had long-term survival with medical management (one died of disease at 922 days, and one was still alive at 343 days). Based on the responses of these dogs, nonresectable primary tumors or evidence of metastasis may not necessarily be indications for euthanasia.
Conclusion
The prognosis for dogs with adrenocortical neoplasia is gen- erally good if they survive the immediate postoperative period. Local tumor recurrence and metastasis appear to be low. The surgery, although technically demanding, appears to offer as good or better a prognosis as chronic medical management.
References
1. Feldman EC, Nelson RW. Hyperadrenocorticism (Cushing’s syn- drome). In: Canine and feline endocrinology and reproduction. 2nd ed. Philadelphia: WB Saunders, 1996:188-265.
2. Feldman EC, Nelson RW, Feldman MS, Farver TB. Comparison of mitotane treatment for adrenal tumor versus pituitary-dependent hyperadrenocorticism in dogs. J Am Vet Med Assoc 1992;200:1642-7.
3. Kintzer PP, Peterson ME. Mitotane treatment of 32 dogs with cortisol- secreting adrenocortical neoplasms. J Am Vet Med Assoc 1994;205: 54-61.
4. . Feldman EC, Bruyette DS, Nelson RW, Farver TB. Plasma cortisol response to ketoconazole administration in dogs with hyperadrenocor- ticism. J Am Vet Med Assoc 1990;197:71-8.
5. Scavelli TD, Peterson ME, Matthiesen DT. Results of surgical treat- ment for hyperadrenocorticism caused by adrenocortical neoplasia in the dog: 25 cases (1980-1984). J Am Vet Med Assoc 1986;189: 1360-4.
6. Emms SG, Johnston DE, Eigenmann JE, Goldschmidt MH. Adrenal- ectomy in the management of canine hyperadrenocorticism. J Am Anim Hosp Assoc 1987;23:557-64.
7. Van Sluijs FJ, Sjollema BE, Voorhout G, van den Ingh TSGAM, Rijn- berk A. Results of adrenalectomy in 36 dogs with hyperadrenocorti- cism caused by adrenocortical tumour. Vet Q 1995;17:113-6.
8. Reusch CE, Feldman EC. Canine hyperadrenocorticism due to adreno- cortical neoplasia: pretreatment evaluation of 41 dogs. J Vet Int Med 1991;5:3-10.
9. Feldman EC. Distinguishing dogs with functioning adrenocortical tumors from dogs with pituitary-dependent hyperadrenocorticism. J Am Vet Med Assoc 1983;183:195-200.