Open Access
ACHIM
American College of Veterinary Internal Medicine
STANDARD ARTICLE OPEN ACCESS
Small Animal Internal Medicine Oncology
Cytologic Evaluation as a Diagnostic Tool to Differentiate Adrenocortical Tumors and Pheochromocytomas
Hunter J. Piegols1 0 | Rachel W. Williams2 | Nirali Pathak2 D | Laura E. Selmic1 | Giovanni Tremolada3 | Russell Sakacs1 Laurie Millward1 | Janis Lapsley1
1The Ohio State University Veterinary Medical Center, The Ohio State University, Columbus, Ohio, USA | 2UF Small Animal Hospital, University of Florida, Gainesville, Florida, USA | 3Flint Animal Cancer Center, Colorado State University, Fort Collins, Colorado, USA
Correspondence: Hunter J. Piegols (piegols.3@osu.edu)
Received: 26 July 2024 | Revised: 21 January 2025 | Accepted: 29 January 2025
Funding: The authors received no specific funding for this work.
Keywords: adenoma | adrenal | aspirate | carcinoma | cytology
ABSTRACT
Background: The utility of cytologic evaluation to distinguish adrenocortical tumors and pheochromocytomas in dogs has not been thoroughly investigated, partly because of the perceived risks of the procedure.
Objectives: Report test characteristics of fine needle aspiration (FNA) and cytologic evaluation for differentiation of adrenocor- tical tumors and pheochromocytomas in dogs. Complications associated with FNA also were recorded.
Animals: Thirty-eight dogs with 40 adrenal tumors that had FNA and cytologic evaluation performed before adrenalectomy were included in the study from three institutions. Tumors included 17 pheochromocytomas, 21 adrenocortical tumors, 1 con- current adrenocortical adenoma and pheochromocytoma, and 1 malignant neoplasm.
Results: Of the 40 FNA cytologic aspirations performed, 35 (87.5%) had a predominant cell type identified and therefore were considered of diagnostic quality. Of these, 30 (85.7%) correlated with the final histopathological diagnosis. When all samples were included, FNA and cytologic evaluation had a sensitivity of 77.3%, specificity of 76.5%, positive predictive value of 81.0%, negative predictive value of 72.2%, and accuracy of 76.9% for identifying adrenocortical tumors. For pheochromocytomas, these values were 72.2%, 95.2%, 92.8%, 80.0%, and 84.5%, respectively. Six (15.9%) dogs had self-limiting complications associated with the FNA procedure.
Conclusion and Clinical Importance: Fine needle aspiration and cytologic evaluation of adrenal tumors has a low compli- cation rate and can help differentiate adrenocortical tumors and pheochromocytomas. Thus, cytologic evaluation of adrenal tumors should be considered to help differentiate adrenal tumors and allow more individualized treatment of affected dogs.
1 Introduction
Advanced imaging has increased the detection of adrenal tu- mors with incidental tumors identified in up to 9.3% of dogs . The most common tumors reported in dogs undergoing adrenalectomy are pheochromocytomas and adrenocortical tumors [2]. Multiple differentiating tests have been proposed ☒ 1
to distinguish adrenocortical tumors and pheochromocyto- mas including advanced imaging, plasma and urine norme- tanephrine concentrations, serum inhibin concentration, endogenous adrenocorticotropic hormone concentration, urine cortisol-to-creatinine ratio, and low dose dexamethasone sup- pression testing [3-5]. Challenges include the potential for non-corticosteroid-producing adrenocortical tumors as well
Abbreviations: CI, confidence interval; cm, centimeter; FNA, fine needle aspiration; IQR, interquartile range; NPV, negative predictive value; PPV, positive predictive value.
@ 2025 The Author(s). Journal of Veterinary Internal Medicine published by Wiley Periodicals LLC on behalf of American College of Veterinary Internal Medicine.
as pituitary-dependent hyperadrenocorticism with concurrent adrenal tumors. Furthermore, pheochromocytomas can have episodic secretion leading to variation in normetanephrine con- centrations [6]. Other limitations include testing availability and prolonged turnaround time [7].
If an adrenalectomy is pursued, pre-operative knowledge of the tumor type may guide medical management and iden- tify potential risks. For example, pheochromocytomas have been associated with peri-operative instability secondary to excessive secretion of catecholamines [8, 9]. Pre-operative treatment with phenoxybenzamine has been proposed pre- viously to minimize peri-operative mortality in dogs with pheochromocytomas, but the benefit has been questioned [2, 8]. Pathologic changes to the myocardium also have been identified at necropsy in dogs with pheochromocytomas [10]. Carcinomas have been associated with hypercoagulability [11]. Experimental studies in rats found that corticosteroids affect wound healing [12]. Additionally, pre-operative corti- costeroid administration has been associated with increased risk of surgical site infection in dogs [13]. It is therefore possi- ble that dogs with increased corticosteroids secondary to hy- peradrenocorticism may have increased risk of surgical site infection. Knowledge of the tumor type may identify risks and enable appropriate risk-minimizing mitigations, which has been proposed to allow more individualized medicine in dogs with adrenal tumors [14].
Cytology provides a readily available means to help differentiate adrenal tumors. Historically, perceived risk led to strong recom- mendations against adrenal gland aspiration, but such risk is not well supported by recent literature [15, 16]. Currently, the accu- racy of cytology to distinguish adrenocortical tumors and pheo- chromocytomas is poorly reported. Accuracy of distinguishing adrenocortical tumors from pheochromocytomas was reported to range from 90%-100% among different pathologists in a re- cent study, but limitations included having a combined study population of dogs and cats, having only 11 samples acquired by cytology, and having a final histologic diagnosis in only six cases [17]. Another study found 71% of adrenal cytologic samples were diagnostically conclusive, but only eight had a histologic diagnosis for comparison, limiting ability to interpret accuracy of the diagnostically conclusive samples [16].
We sought to assess the value of fine needle aspiration (FNA) to distinguish primary adrenal tumors in dogs before adrenalec- tomy. Our primary objective was to assess the diagnostic utility of cytology to distinguish adrenocortical tumors and pheochro- mocytomas before undergoing an adrenalectomy when com- pared with the final histopathology report. A secondary aim was to identify complications associated with adrenal gland aspiration.
2 | Methods
2.1 | Study Population
Ours was a multi-institutional, cross-sectional study involving retrospective review of medical records to assess the diagnos- tic utility of cytologic evaluation for pre-operative diagnosis of
adrenal tumors. Institutional Animal Care and Use Committee approval was not required given the retrospective nature of the study. Electronic medical records at three academic veterinary hospitals were searched for dogs undergoing an adrenalectomy that had pre-operative FNA and cytologic evaluation of the ad- renal gland being excised. Dogs were included if they underwent adrenalectomy between January 1, 2013 and March 1, 2024. For inclusion, the cytologic evaluation of the adrenal tumor had to be interpreted by a veterinary clinical pathologist. Exclusion criteria were if a pre-operative cytology report or post-operative histopathology report was not available for review. If bilateral adrenalectomy, either single-session or staged, was performed, both tumors were included as separate entities.
2.2 | Data Collection
Collected data included: age at the time of surgery, breed, sex and neuter status, and body weight. Continuous data were eval- uated for normality of distribution using the Shapiro-Wilk test and reported as mean +SD when normally distributed and me- dian (interquartile range [IQR]) when non-normally distributed. The affected side (left or right) was noted as well as maximal diameter of the tumor on pre-operative ultrasound examination or computed tomography. It was recorded if the tumor was sam- pled under general anesthesia or sedation. The cytologic descrip- tion was categorized as an epithelial or neuroendocrine cell type based on the predominant cell type identified in the cytology report during medical record evaluation. Alternatively, the sam- ple was categorized as non-diagnostic if a predominant cell type could not be identified other than cells consistent with blood contamination. Other neoplastic or inflammatory cytologic di- agnoses of adrenal tumors were not encountered and thus were not recorded in the study. Complications noted in the medical record in association with FNA of the adrenal gland were re- corded. Complications were graded using a scheme previously described for adverse events in investigational therapy. Briefly, grade 1 was considered mild, grade 2 moderate, grade 3 severe (clinically relevant but not immediately life threatening), grade 4 life-threatening, and grade 5 death [18]. The surgical proce- dure performed and histologic diagnosis were recorded.
2.3 | Statistical Analysis
All statistics were performed using Microsoft Excel (Version 2308 Build 16.0.16731.20542) with a StatPlus Add-In (Version 7.8.4). Diagnostic accuracy was calculated for all cases identi- fied as well as for samples having a cytologic description pre- dominated by an epithelial or neuroendocrine cell type, which was considered as diagnostic quality. The number of diagnostic quality and nondiagnostic samples collected under general anes- thesia and sedation was recorded with group differences tested using a Pearson Chi-square test. The 95% confidence intervals (CI) were calculated. Age at surgery, body weight, and maximal tumor diameter on imaging were reported for correctly catego- rized tumors and for those that were either non-diagnostic on cytology or incorrectly categorized with group differences tested using a t-test or Mann-Whitney U test, depending on if data were normally or non-normally distributed, respectively. Tumor laterality and sex also were reported for correctly categorized
tumors and for those that were either non-diagnostic cytologic samples or incorrectly diagnosed with group differences tested using a Pearson Chi-square test. Tumors were assigned as cor- rectly categorized when the cytologic description identified an epithelial or neuroendocrine cell type and were diagnosed as an adrenocortical tumor or pheochromocytoma on histopathology, respectively. Tumors were assigned as incorrectly categorized when the cytologic diagnosis did not correlate with the histo- pathologic diagnosis.
Test characteristics including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated to assess the ability of cytology to correctly identify adrenal tumors as epithelial tumors with the histopathology results consistent with an adrenocortical tumor considered the gold standard; 95% CI were also calculated. The same values were calculated to assess the ability of cytology to correctly identify adrenal tumors as neuroendocrine tumors with the histopathology results consistent with a pheochromo- cytoma considered the gold standard.
The proportion of dogs developing complications was recorded with 95% CI calculated.
3 Results
3.1 Clinical Findings
Thirty-eight dogs with 40 adrenal tumors were identified that met the inclusion criteria at three academic institutions. The different institutions contributed 3, 13, and 22 dogs. The most common dog breeds were mixed breed dogs (12), dachshunds (4), bichon frise (2), shih tzu (2), and unspecified terriers (2). Dog breeds with only one dog identified included American Staffordshire terrier, Australian shepherd, Boston terrier, Cavalier King Charles spaniel, French bulldog, golden retriever, Jack Russell terrier, keeshond, Manchester terrier, miniature poodle, Pomeranian, pug, Rhodesian ridgeback, rottweiler, springer spaniel, and Yorkshire terrier. Median body weight was 15.1 kg (IQR, 6.3-27.1). Mean +/-SD age at the time of surgery was 11.3+/-1.9years old. There were 20 spayed female dogs and 18 neutered male dogs. There were 20 left-sided and 20 right-sided tumors. Maximal tumor diameter on imaging was 3.2+/-1.4cm (range, 0.8-7.3 cm). Thirty-six dogs underwent unilateral adrenalectomy, and two dogs underwent bilateral adrenalectomy.
3.2 | Diagnostic Findings
Twenty tumors were sampled with the dog under general anesthesia and 18 were sampled with the dog sedated; this information was not recorded for two tumors. Fine needle aspiration was performed under ultrasound guidance in all cases. Of the cytologic samples obtained, 87.5% (n =35; 95% CI, 73.2%-95.8%) had an identifiable predominant cell type and were considered of diagnostic quality. All 20 FNA sam- ples collected under general anesthesia were considered as diagnostic quality whereas 72% (n=13) of FNA samples col- lected under sedation were considered as diagnostic quality
(p=0.01). Of the 35 cytologic samples considered as diag- nostic quality, 60% (n=21) were consistent with an epithelial cell type (Figure 1) and 40% (n=14) were consistent with a neuroendocrine cell type (Figure 2). On histologic evaluation 42.5% (n =17) of tumors were consistent with a pheochromo- cytoma, 52.5% (n=21) were consistent with adrenocortical tumors, 1 was consistent with adrenocortical adenoma with concurrent pheochromocytoma, and 1 was consistent with a malignant neoplasm. Five pheochromocytomas were positive on Grimelius stain for argyrophilic granules; the tumor diag- nosed as an adrenocortical adenoma with concurrent pheo- chromocytoma also had regions positive on Grimelius stain for argyrophilic granules. One adrenocortical tumor was neg- ative on Grimelius stain, and 1 adrenocortical tumor was pos- itive for melan-A. The malignant neoplasm was analyzed by immunohistochemistry and was negative for inhibin-alpha, paired box 8, and factor VIII-related antigen with the cell of origin not determined. Of the adrenocortical tumors, 66.7% (n = 14) were adrenocortical adenomas, 28.6% (n =6) were ad- renocortical carcinomas, and 4.8% (n =1) were adrenocortical hyperplasia.
3.3 | Testing Characteristics of Cytology to Diagnose Adrenal Tumors
Overall, cytology of all the samples had a diagnostic accuracy of 75% (CI, 58.8%-87.3%). Accuracy was defined by identifying either an epithelial or neuroendocrine cell type on cytology with the respective cell types being identified on histopathology. The malignant neoplasm on histopathology was non-diagnostic on cytology and thus was included in the inaccurate category for the calculation of overall diagnostic accuracy; it was excluded from subsequent calculations because a more definitive diagno- sis was not achieved. The tumor consistent with an adrenocor- tical adenoma and pheochromocytoma on histopathology was consistent with a neuroendocrine cell type on cytology and was considered correctly categorized because a pheochromocytoma was present. When considering only cytologic samples of diag- nostic quality, cytology had an accuracy of 85.6% (CI, 69.7%- 95.2%; Table 1).
Baseline characteristics associated with each tumor were di- vided into correctly categorized tumors and those that were non-diagnostic cytologic samples or incorrectly categorized (Table 2). Of the variables assessed, age was the only factor that was significantly different between correctly categorized cyto- logic samples and non-diagnostic or incorrectly categorized cy- tologic samples.
To assess diagnostic utility, test characteristics consisting of sensitivity, specificity, PPV, NPV, and accuracy were deter- mined to assess the ability of cytology to identify a mass as consistent with an adrenocortical tumor (Table 3) and subse- quently pheochromocytoma (Table 4). When performing these calculations, the tumor that had neuroendocrine cell type on cytologic evaluation and a concurrent pheochromocytoma and adrenocortical adenoma on histopathology was consid- ered a false negative for the adrenocortical tumor calculations and a true positive for the pheochromocytoma calculations because the adrenocortical adenoma component failed to be
detected on cytology whereas the pheochromocytoma was de- tected on cytology.
3.4 | Complications Associated With FNA Procedure
Of the dogs undergoing FNA procedures, 15.9% (n=6; CI, 6.0%-31.3%) had complications identified in the medical re- cords in association with the FNA procedure. The following complications were identified: grade 1 bradycardia (n=2), grade 2 intermittent second-degree atrioventricular block (n=1), grade 2 hypercapnia (n=1), grade 2 hypothermia (n=1), and grade 3 apnea (n=1). Only 18 dogs had blood pressure monitoring during the FNA procedure, but no hy- pertensive episodes were recorded. No other complications were identified, and no care beyond monitoring recovery from sedation or general anesthesia was required for treatment. Of the six dogs with complications, four (66.7%) had pheochro- mocytomas and two (33.3%) had adrenocortical adenomas. One dog with bilateral adrenocortical tumors was receiving prazosin and one dog with a pheochromocytoma was receiv- ing amlodipine at the time of the FNA procedure; no other anti-hypertensive medications were being administered at the time of sampling.
4 | Discussion
Although multiple diagnostic tests to distinguish adrenocorti- cal tumors and pheochromocytomas are reported, the utility of
cytologic evaluation has received limited evaluation. Of the FNA performed, 87.5% were considered of diagnostic quality, based on having a predominant cell type identified. A previous study with 58 adrenal gland FNA across 15 institutions found 71% of cyto- logic samples to be diagnostic [16]. One potential explanation for this difference is that regular performance of the procedure may increase competency and subsequent obtainment of diagnostic samples with two of the centers in our study performing 37 of the 40 FNA. Sampling performed under general anesthesia was associated with more samples being of diagnostic quality, which is suspected to be associated with general anesthesia potentially facilitating ultrasound-guided FNA as well as the potential to collect multiple samples. Given an overall accuracy of 75% for all cytologic samples and 85.6% when confined to diagnostic cyto- logic samples, when using histopathology as a gold standard, the results of our study suggest that cytology could be considered as a diagnostic test for distinguishing adrenocortical tumors and pheochromocytomas. The accuracy found is similar but slightly less than previously reported accuracies of 87.5% and 90%-100% for comparing a cytologic diagnosis to histopathology. The spec- ificity of 95.2% and PPV of 92.9% of cytology for identification of pheochromocytomas is of particular interest because a positive result highly supports a dog truly having a pheochromocytoma. It is important to note that PPV and NPV are dependent on dis- ease prevalence with PPV increasing and NPV decreasing as the prevalence increases, respectively. Thus, the PPV found would be lower in a population of dogs with a lower prevalence of adre- nal neoplasia, such as a general population of dogs. There was a complication rate of 15.9% with overall low severity of complica- tions documented in the included cases, further suggesting that the risk of the procedure is relatively low while providing useful
| Histopathology | ||||
|---|---|---|---|---|
| Adrenocortical tumor | Pheochromocytoma | Total | ||
| Cytology | Epithelial cell type | 17 | 4 | 21 |
| Neuroendocrine cell type | 1 | 13 | 14 | |
| Total | 18 | 17 | 35 | |
Note: Overall accuracy of diagnostic cytologic samples =85.6% (95% CI, 69.7%, 95.2%). Abbreviation: CI, confidence interval.
information. Only a single grade 3 adverse event was noted with all others being mild or moderate [18].
Using cytology as a diagnostic test for distinguishing adrenal tumors has several potential benefits. Most adrenal tumors are diagnosed using imaging, and ultrasound-guided FNA of the tumor can be readily performed. The cytologic findings also can be interpreted in combination with advanced imaging find- ings. For example, pheochromocytomas tend to be more hyper- attenuating, are associated with a larger contralateral adrenal gland, and have more prominent tumor thrombi on computed tomography than do carcinomas [3]. Cytologic evaluation by a veterinary clinical pathologist is rapid, relatively inexpensive, and widely available, further facilitating this diagnostic option. Cytologic features to help distinguish adrenocortical tumors and pheochromocytomas have been described including adre- nocortical tumors being more likely to have intact cells, cells that either occur singly or in cohesive clusters, low nuclear to cy- toplasmic ratio, basophilic and variably vacuolated cytoplasm,
and coarse or condensed chromatin [17]. Other diagnostic tests such as evaluation of plasma and urine metanephrine concen- trations require high-pressure liquid chromatography with electrochemical detection or liquid chromatography-tandem mass spectrometry [4]. The need for specialized equipment for determination of metanephrine concentrations results in lim- ited availability and increased costs as compared with cytologic evaluation. Limitations for the two previous studies on adrenal cytology include that eight or fewer cases in each study had his- topathology results [16, 17]. Additionally, one of the studies in- cluded imprint cytology, masses collected during necropsy, and cats. Thus, this study may not represent the use of cytology for distinction of adrenocortical tumors and pheochromocytomas in dogs when used in a clinical setting.
Cytology had an overall accuracy of 84.6% for identifying pheo- chromocytomas with a PPV of 92.9% and specificity of 95.2%. Although measurement of metanephrines in plasma, urine, and saliva is documented to diagnose pheochromocytomas in dogs,
| Characteristic | Correctly categorized cytologic samples (n=30) | Non-diagnostic or incorrectly categorized cytologic samples (n=10) | d |
|---|---|---|---|
| Weight (kg; IQR) | 15.8 (7.7-24) | 7.4 (6.0-27.9) | p=0.39A |
| Age (years) +/-SD | 11.0+/-1.6 | 12.8+/-2.01 | T-test: p<0.01B |
| Sex (n) | p=0.36℃ | ||
| Female spayed | 17 | 4 | |
| Male neutered | 13 | 6 | |
| Tumor laterality | p=0.14℃ | ||
| Left | 17 | 3 | |
| Right | 13 | 7 | |
| Maximal tumor diameter (cm)+/-SD | 3.0+/-1.1 | 3.9+/-1.8 | T-test: p=0.09B |
Note: Comparison tests used were Mann-Whitney U test (A), T-test (B), and Pearson’s chi-square (C). Abbreviations: cm, centimeter; IQR, interquartile range; kg, kilogram; SD, standard deviation.
| Histologic diagnosis consistent with adrenocortical tumor | ||||
|---|---|---|---|---|
| Present | Absent | |||
| Cytologic diagnosis consistent with adrenocortical tumor | Present | 17 | 4 | PPV: 81.0% (95% CI 58.1%, 94.6%) |
| Absent | 5 | 13 | NPV: 72.2% (95% CI 46.5%, 90.3%) | |
Sensitivity: 77.3%
Specificity: 76.5%
(95% CI 54.6%, 92.2%)
(95% CI 50.1%, 93.2%)
Note: Accuracy in identifying adrenocortical tumors: 76.9% (95% CI, 60.7%, 88.9%).
Abbreviations: CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value.
urinary normetanephrine is perceived to be the most used and described in dogs [4, 19-21]. A cut-off ratio of four-times the highest urine normetanephrine to creatinine ratio measured in control samples, a ratio of > 364, was previously proposed, but it had a low sensitivity of 57.1% with three of the seven dogs with pheochromocytomas failing to be detected using this cut-off [19]. This contrasts with the use of cytology which has a sensitivity of 72.2%. Furthermore, a confounding factor is that dogs with hy- peradrenocorticism also have increased normetanephrine, sus- pected to be from increased catecholamine production secondary to increased glucocorticoids, which results in the need for a high threshold for urine normetanephrine to creatinine ratios to have a diagnostically useful PPV [19]. In contrast, cytologic evaluation has a sensitivity of 72.2% while still maintaining a PPV of 92.9%. Plasma free normetanephrine concentration using a cut-off of 5.52 nmol/L has a sensitivity and specificity of 100% and 97.6%, but the study only had eight dogs, all of which had clinical signs or a history of hypertension and a median tumor size of 5.0 cm
[7]. It is unclear if the reported sensitivity and specificity would ☒ remain the same in a larger population that contained dogs with incidentally found pheochromocytomas because such dogs may have lower normetanephrine concentrations or more episodic
secretion than dogs with reported clinical signs. Additionally, urine normetanephrine to creatinine ratio may be influenced by the location of sample collection, with collection in a clinic associated with higher ratios than when samples are collected at home [21]. Overall, cytologic evaluation of pheochromocytomas is not influenced by many of the confounding factors associated with measurements of urine normetanephrine concentrations.
Cytology had an accuracy of 76.9% for diagnosing adrenocor- tical tumors. Many diagnostic tests to support an adrenocorti- cal tumor, such as a low-dose dexamethasone suppression test and endogenous adrenocorticotropin releasing hormone, are dependent on the tumor secreting cortisol. Serum inhibin con- centrations are increased in dogs with hyperadrenocorticism because of production of inhibin by tumor cells, and limited evaluation of inhibin in dogs with non-cortisol secreting adre- nocortical tumors is available. In four dogs however the serum concentration was highly variable [5]. A more systematic char- acterization of adrenal corticosteroid production beyond cor- tisol in dogs with primary adrenal tumors is warranted. Dogs also can have concurrent pituitary and adrenal lesions, which is reported in 5% of dogs with hypercortisolism and in 10% of
| Histologic diagnosis consistent with pheochromocytoma | ||||
|---|---|---|---|---|
| Present | Absent | |||
| Cytologic diagnosis consistent with pheochromo-cytoma | Present | 13 | 1 | PPV: 92.9% (95% CI 66.1%, 99.8%) |
| Absent | 5 | 20 | NPV: 80.0% (95% CI 59.3%, 93.2%) | |
| Sensitivity: 72.2% (95% CI, 46.5%, 90.3%) | Specificity: 95.2% (95% CI, 76.2%, 99.9%) | |||
Note: Accuracy in identifying pheochromocytomas: 84.6% (95% CI, 69.5%, 94.1%).
Abbreviations: CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value.
dexamethasone-resistant dogs [22]. Of dogs with pituitary- dependent hyperadrenocorticism, 39% and 25% fail to suppress using a low- and high-dose dexamethasone tests, respectively [23]. Thus, it is possible to have dogs with pituitary-dependent hyperadrenocorticism and a concurrent non-functional adre- nocortical tumor or pheochromocytoma, potentially compli- cating the interpretation using a suppression test. Diagnostic tests based on functionality of the adrenocortical tumor present a challenge with diagnosing adrenocortical tumors. Because cytology is not dependent on tumor function for its diagnostic utility, it provides an opportunity to identify non-functional ad- renocortical tumors that may be otherwise difficult to identify. Another limitation for adrenal tumor differentiation is that en- docrine testing for adrenocortical tumors is only indicated when clinical signs of hyperadrenocorticism are present [24]. The util- ity of such diagnostic tests in dogs with adrenal tumors that do not produce signs of hyperadrenocorticism is unclear. Cytology may allow diagnosis and differentiation of non-functional ad- renocortical tumors that may fail to be diagnosed using other methods.
Of the dogs in our study that underwent FNA of the adrenal gland, 15.9% had complications related to the procedure. Most of the complications noted may have been secondary to general anesthesia or sedation required for sample procurement rather than the FNA procedure itself. In a similar safety study, only 5% (n=1) of dogs developed a complication, ventricular tachycar- dia, at the time of the procedure [15]. Another study estimated a death rate of approximately 1% associated with the procedure, based primarily on radiologist recollection, which is at risk for availability bias [16]. No dogs in our study died, had hemorrhage noted in the record, or required post-procedural supportive care beyond recovery from general anesthesia and sedation. Thus, the risk associated with performing a FNA was not perceived to be excessive.
Fine needle aspiration of the adrenal gland is performed in hu- mans using computed tomography guidance or endoscopic ul- trasound guidance with adequate samples retrieved in 90.4% of cases [25]. This result is similar to the 87.5% of samples having a predominant cell type identified that were considered of di- agnostic quality in our study. Complications associated with computed tomography-guided needle biopsy in humans include pneumothorax, pain, hemorrhage, pancreatitis, and needle tract metastasis [26]. The complication rate reported in humans varies
from 0% to 12% [25, 26]. Although we found a complication rate of 15.9%, it did not include any of these complications noted in humans. Confirmation of metastatic disease is reported in 53% of adrenal FNA in humans [25]. This finding contrasts with dogs in which only 2% of adrenal FNA are consistent with metastatic disease, and most are consistent with primary adrenal tumors [16]. Adrenal FNA in humans is regarded as a diagnostic that is useful for differentiating primary adrenal cortical and medul- lary tumors with complications considered uncommon, further suggesting utility in dogs [25, 26].
Limitations of our study are mostly related to its retrospec- tive nature. Despite being a multi-institutional study, a lim- ited number of dogs had pre-operative cytology of an adrenal tumor that ultimately was removed by adrenalectomy. Type II error is possible with failed identification of further group dif- ferences. Concurrent differentiating tests were not routinely performed, and direct comparisons regarding accuracy com- pared with endocrine testing such as low-dose dexamethasone suppression tests and urine normetanephrine concentrations could not be made. The clinical decision to perform FNA of the adrenal gland was determined by the clinician, which is a potential source of bias. One case identified had an undefined malignant neoplasm on histopathology and a non-diagnostic cytology. The role of cytology in such cases is unclear given the limited number of cases in our study. Another limitation is that differentiation between adrenocortical hyperplasia, adenomas, and adenocarcinomas was not performed because of the lim- ited number of cases. In a larger data set, ability of cytology to further differentiate adrenocortical tumors would be of inter- est. Variation in pathologist certainty of diagnosis also was not assessed. Immunohistochemistry to confirm the diagnosis of a pheochromocytoma or adrenocortical tumor was not routinely performed. Additionally, protocols for sedation and general an- esthesia, monitoring during the procedure, and performing the FNA were not standardized. Complications associated with the FNA procedure may have been secondary to the sedation or gen- eral anesthesia rather than FNA itself. Finally, there was a lack of long-term follow-up for development of complications such as needle-tract metastasis.
Overall, diagnostic cytologic samples have an accuracy of 85.6% in the population studied. The high PPV for identifying a pheo- chromocytoma within our study population suggests that a cyto- logic diagnosis of a neuroendocrine tumor should be viewed as
having a high likelihood of the dog having a pheochromocytoma diagnosed on histopathology. Information on differentiating ad- renal tumor type may allow for a patient-focused approach in medical management of such dogs, especially in the pre-operative period, as well as may play a role in the decision to pursue surgi- cal management. In a recent retrospective study of 302 dogs, 87% survived to discharge after adrenalectomy, with tumor-related survival of 3.96 years [2]. Pre-operative treatment has been best described for pheochromocytomas. Pre-treatment with phe- noxybenzamine previously was reported to decrease the risk of mortality in dogs with pheochromocytomas, but this benefit has not been demonstrated in more recent studies [2, 6, 8, 27]. Many of the medications used for pre-operative management would best be used with knowledge of the specific adrenal tumor type and the associated risks of that tumor. Performance of a FNA with subsequent cytologic evaluation appears to be a relatively low risk procedure that may supplement information from other diagnostic tests and advanced imaging to help the clinician dif- ferentiate adrenocortical tumors and pheochromocytomas. This information may allow more individualized management of dogs with adrenal tumors and disease-specific risk assessment, which may benefit clinical management.
Acknowledgments
The authors have nothing to report.
Disclosure
The authors declare no off-label use of antimicrobials.
Ethics Statement
The authors declare no Institutional Animal Care and Use Committee or other approval was needed. The authors declare human ethics ap- proval was not needed.
Conflicts of Interest
The authors declare no conflicts of interest.
References
1. J. I. Baum, S. E. Boston, and J. B. Case, “Prevalence of Adrenal Gland Masses as Incidental Findings During Abdominal Computed Tomogra- phy in Dogs: 270 Cases (2013-2014),” Journal of the American Veterinary Medical Association 249, no. 10 (2016): 1165-1169.
2. H. J. Piegols, B. E. Abrams, J. M. Lapsley, et al., “Risk Factors In- fluencing Death Prior to Discharge in 302 Dogs Undergoing Unilateral Adrenalectomy for Treatment of Primary Adrenal Gland Tumours,” Veterinary and Comparative Oncology 21, no. 4 (2023): 673-684.
3. P. Pey, S. Specchi, F. Rossi, et al., “Prediction of Vascular Invasion Using a 7-Point Scale Computed Tomography Grading System in Adre- nal Tumors in Dogs,” Journal of Veterinary Internal Medicine 36, no. 2 (2022): 713-725.
4. M. F. van den Berg, H. S. Kooistra, G. C. M. Grinwis, et al., “Refer- ence Intervals for Plasma, Urinary, and Salivary Concentrations of Free Metanephrines in Dogs: Relevance to the Diagnosis of Pheochromocy- toma,” Journal of Veterinary Internal Medicine 37, no. 1 (2023): 173-183.
5. C. Brömel, R. W. Nelson, E. C. Feldman, et al., “Serum Inhibin Con- centration in Dogs With Adrenal Gland Disease and in Healthy Dogs,” Journal of Veterinary Internal Medicine 27, no. 1 (2013): 76-82.
6. D. Enright, V. M. Dickerson, J. A. Grimes, S. Townsend, and K. M. Thieman Mankin, “Short- and Long-Term Survival After Adrenalec- tomy in 53 Dogs With Pheochromocytomas With or Without Alpha- Blocker Therapy,” Veterinary Surgery 51, no. 3 (2022): 438-446.
7. R. Gostelow, N. Bridger, and H. M. Syme, “Plasma-Free Metaneph- rine and Free Normetanephrine Measurement for the Diagnosis of Pheochromocytoma in Dogs,” Journal of Veterinary Internal Medicine 27, no. 1 (2013): 83-90.
8. M. A. Herrera, M. L. Mehl, P. H. Kass, P. J. Pascoe, E. C. Feldman, and R. W. Nelson, “Predictive Factors and the Effect of Phenoxybenzamine on Outcome in Dogs Undergoing Adrenalectomy for Pheochromocytoma,” Journal of Veterinary Internal Medicine 22, no. 6 (2008): 1333-1339.
9. F. Zublena, C. De Gennaro, and F. Corletto, “Retrospective Evalua- tion of Labetalol as Antihypertensive Agent in Dogs,” BMC Veterinary Research 16, no. 1 (2020): 256.
10. F. F. Argenta, M. Slaviero, L. S. de Mello, et al., “Pathological Aspects and Immunohistochemical Evaluation of Troponin C in the Cardiovas- cular System of Dogs With Pheochromocytoma,” Topics in Companion Animal Medicine 53-54 (2023): 100777.
11. P. Vilar Saavedra, A. L. Garcia, S. Z. Lopez, et al., “Hemostatic Ab- normalities in Dogs With Carcinoma: A Thromboelastographic Charac- terization of Hypercoagulability,” Veterinary Journal 190, no. 2 (2011): 378-383.
12. H. P. Ehrlich and T. K. Hunt, “Effects of Cortisone and Vitamin A on Wound Healing,” Annals of Surgery 167 (1968): 324-328.
13. J. Espinel-Rupérez, M. D. Marín-Ríos, V. Salazar, et al., “Incidence of Surgical Site Infection in Dogs Undergoing Soft Tissue Surgery: Risk Factors and Economic Impact,” Veterinary Record. Open 6, no. 1 (2019): e000233.
14. G. Hayes, “Update of Adrenalectomy,” Veterinary Clinics of North America. Small Animal Practice 52, no. 2 (2022): 473-487.
15. J. A. Sumner, L. Lacorcia, A. M. Rose, A. P. Woodward, and J. E. Car- ter, “Clinical Safety of Percutaneous Ultrasound-Guided Fine-Needle Aspiration of Adrenal Gland Lesions in 19 Dogs,” Journal of Small Ani- mal Practice 59, no. 6 (2018): 357-363.
16. P. Pey, A. Diana, F. Rossi, et al., “Safety of Percutaneous Ultrasound- Guided Fine-Needle Aspiration of Adrenal Lesions in Dogs: Perception of the Procedure by Radiologists and Presentation of 50 Cases,” Journal of Veterinary Internal Medicine 34, no. 2 (2020): 626-635.
17. W. Bertazzolo, M. Didier, M. E. Gelain, et al., “Accuracy of Cytology in Distinguishing Adrenocortical Tumors From Pheochromocytoma in Companion Animals,” Veterinary Clinical Pathology 43, no. 3 (2014): 453-459.
18. A. K. LeBlanc, M. Atherton, R. T. Bentley, et al., “Veterinary Co- operative Oncology Group-Common Terminology Criteria for Adverse Events (VCOG-CTCAE v2) Following Investigational Therapy in Dogs and Cats,” Veterinary and Comparative Oncology 19, no. 2 (2021): 311-352.
19. S. Quante, F. S. Boretti, P. H. Kook, et al., “Urinary Catecholamine and Metanephrine to Creatinine Ratios in Dogs With Hyperadrenocor- ticism or Pheochromocytoma, and in Healthy Dogs,” Journal of Veteri- nary Internal Medicine 24, no. 5 (2010): 1093-1097.
20. N. Sasaki, Y. Ikenaka, Y. Inoue, et al., “Urinary Free Metanephrines Measurement in Dogs With Adrenal Gland Diseases Using a New Sim- ple Liquid Chromatography Tandem Mass Spectrometry Method,” Jour- nal of Veterinary Medical Science 83, no. 4 (2021): 648-655.
21. K. Höglund, H. Palmqvist, S. Ringmark, and A. Svensson, “Quanti- fication of Normetanephrine in Canine Urine Using ELISA: Evaluation of Factors Affecting Results,” Journal of Veterinary Diagnostic Investiga- tion 34, no. 1 (2022): 28-35.
22. K. L. van Bokhorst, H. S. Kooistra, S. A. E. B. Boroffka, and S. Galac, “Concurrent Pituitary and Adrenocortical Lesions on Computed
Tomography Imaging in Dogs With Spontaneous Hypercortisolism,” Journal of Veterinary Internal Medicine 33, no. 1 (2019): 72-78.
23. E. C. Feldman, R. W. Nelson, and M. S. Feldman, “Use of Low- and High-Dose Dexamethasone Tests for Distinguishing Pituitary- Dependent From Adrenal Tumor Hyperadrenocorticism in Dogs,” Journal of the American Veterinary Medical Association 209, no. 4 (1996): 772-775.
24. E. N. Behrend, H. S. Kooistra, R. Nelson, C. E. Reusch, and J. C. Scott-Moncrieff, “Diagnosis of Spontaneous Canine Hyperadrenocorti- cism: 2012 ACVIM Consensus Statement (Small Animal),” Journal of Veterinary Internal Medicine 27, no. 6 (2013): 1292-1304.
25. A. G. Novotny, J. P. Reynolds, A. A. Shah, et al., “Fine-Needle Aspi- ration of Adrenal Lesions: A 20-Year Single Institution Experience With Comparison of Percutaneous and Endoscopic Ultrasound Guided Ap- proaches,” Diagnostic Cytopathology 47, no. 10 (2019): 986-992.
26. M. K. Mody, E. A. Kazerooni, and M. Korobkin, “Percutaneous CT-Guided Biopsy of Adrenal Masses: Immediate and Delayed Com- plications,” Journal of Computer Assisted Tomography 19, no. 3 (1995): 434-439.
27. C. Appelgrein, G. Hosgood, E. Drynan, and A. Nesbitt, “Short-Term Outcome of Adrenalectomy in Dogs With Adrenal Gland Tumours That Did Not Receive Pre-Operative Medical Management,” Australian Vet- erinary Journal 98, no. 9 (2020): 449-454.