Introduction

Pheochromocytomas are neuroendocrine tumors arising from the adrenal medulla that produce catecholamines.¹ Clinical signs associated with pheochromocytomas are most commonly those associated with increased catecholamine production and include tachycardia, cardiac arrhythmias, hypertension, and syncopal episodes.¹ Clinical signs can be transient due to the episodic secretion of epinephrine and norepinephrine from the tumor and are not specific for the disease process. The reported incidence of canine adrenal tumors is approximately 1% to 2% of all neoplasms and 0.01% to 0.1% of all neoplasms for pheochromocytomas specifically.² Unlike in human medicine where the vast majority of pheochromocytomas are benign, they are considered a malignant process in dogs with an incidence of anywhere from 39% to 50% for local invasion.³ Metastasis is reported to occur approximately 13% to 24% of the time, and most common locations are regional lymph nodes and the liver, lungs, spleen, and kidney.³

In human medicine, the current diagnostic recommendation regarding pheochromocytomas is quantifying metanephrine and normetanephrine (the metabolites of epinephrine and norepinephrine) in plasma or urine samples.^4–6 For the purpose of this manuscript, the reference to metanephrines (plural) refers to metanephrine and normetanephrine. Although evidence regarding diagnostic options is not as plentiful in veterinary literature, there are reports validating the use of measuring urinary metanephrines in pheochromocytoma diagnosis.^7–10 In addition, a more recent study¹¹ evaluated the feasibility of spot urine collection and concluded that free metanephrines in spot urine samples are a useful tool in discriminating pheochromocytomas from other adrenal disease; however, the study was limited by a small sample size. This collection of human and veterinary literature serves as the evidence behind spot urine sampling being the diagnostic choice in this study. On the contrary, there are populations of pheochromocytomas that are biochemically negative.¹² This is further elaborated on in the Discussion of this text; however, for the purposes of this study, biochemically negative pheochromocytomas are referred to as “silent.” Veterinary studies illustrating the specificity and sensitivity of spot urine collection in dogs with pheochromocytomas are lacking.

When further evaluating the use of urinary metanephrines, it has been discussed in human literature that the diameter of pheochromocytomas is directly related to urine metanephrine/normetanephrine levels, with larger tumors correlating to higher ratios.^7,13 In veterinary medicine, the presence of vascular invasion precludes a straightforward tumor diameter measurement (a 2-D measurement) as an accurate representation of tumor size, as the thrombus component has the potential to be larger than the tumor itself.^8,9 This potential serves as the basis behind obtaining a 3-D measurement of the tumor and thrombus component to quantify tumor size in dogs within this study. To our knowledge, there are currently no veterinary studies comparing tumor volume to urinary metanephrines.¹

The primary objectives of this investigation were to (1) evaluate the specificity/sensitivity and predictive values of metanephrine/normetanephrine-to-creatinine ratios in spot urine samples to diagnose canine pheochromocytomas and to differentiate from nonpheochromocytomas, (2) determine whether there is a difference in utility between urine metanephrine versus normetanephrine values in the diagnosis of pheochromocytoma, (3) determine whether there is a correlation between spot urine catecholamine levels and tumor volume, and (4) determine what percentage of pheochromocytomas have negative urine catecholamine levels and whether this correlates to tumor volume. Secondary objectives were to (1) assess whether higher urine catecholamine levels are a predictive value of vascular invasion, (2) establish whether there is an association between larger tumor volume and elevated urine catecholamine levels and decreased survival time, and (3) determine whether large tumor volume and elevated urine catecholamine levels are associated with increased intraoperative complications.

Our hypotheses were that (1) the specificity and sensitivity of urine spot samples to diagnose canine pheochromocytoma would be high and values needed to make a diagnosis of pheochromocytoma would fall below the previously reported values of 4 times the upper reference limit, (2) normetanephrine would be more significant in the diagnosis of pheochromocytomas, (3) tumor volume would correlate with urine values, and (4) silent pheochromocytomas would be rare and correlate with a smaller tumor volume. Hypotheses for the secondary objectives were that (1) higher urine catecholamine levels would be more likely to be associated with vascular invasion, (2) there would be no association between tumor volume and/or urine catecholamine levels and survival time, and (3) large tumor volume would be more likely to have intraoperative complications.

Methods

Data collection

Computed tomography studies were performed with the use of a multidetector (128-slice) CT scanner (Aquilion Prime; Toshiba Medical Systems) with helical acquisition. Pre- and postcontrast medium and delayed postcontrast images were acquired. Nonionic, iodinated positive contrast medium (iohexol [Omnipaque], 300 mgI/mL) was administered as an IV bolus to patients using a power injector (Medrad Stellant CT injection system; Bayer AG) at injection rates of 2 mL/kg. Computed tomography equipment and study protocol were consistent across all 3 institutions. The data collected from the CT images included location of the adrenal mass and tumor volume. Tumor volume was measured by a board-certified radiation oncologist using Eclipse (version 15.6; Varian Medical Systems Inc). Differentiation between tumor thrombus and vascular thrombus was reported by board-certified radiologists using criteria of a 7-point grading system to evaluate vascular invasion.¹⁴ Serial area measurements of the tumor and associated thrombus were obtained and calculated by the system to provide a complete lesion volume (Figure 1).

View Full Size

View Full Size

View Full Size

View Full Size

Serial area measurements of pheochromocytoma in a 10-year-old male neutered Maltese using the Eclipse system. The red outlines highlight serial area measurements of the adrenal tumor and any associated thrombus. The red structure in panel B shows 3-D representation of the adrenal tumor. Transverse portal view (A), 3-D portal model (B), frontal portal view (C), and sagittal portal view (D).

Citation: Journal of the American Veterinary Medical Association 263, 3; 10.2460/javma.24.06.0409

• Download Figure
• Download Figure
• Download Figure
• Download Figure
• Download figure as PowerPoint slide

Urine (5 mL) was obtained from patients via cystocentesis or free catch and placed into a tube containing 50 mg of sulfamic acid. All samples were evaluated via tandem mass spectrometry through Marshfield Labs to provide a urine metanephrine/normetanephrine-to-creatinine ratio.

Data obtained from the medical record for nonpheochromocytoma patients included age, weight, breed, sex, surgery date, surgeon, urine collection method, metanephrine-to-creatinine and normetanephrine-to-creatinine ratios, and histopathology results. The same laboratory utilizing multiple pathologists was used to evaluate samples. Data obtained from the medical record for pheochromocytoma patients included the information listed previously in addition to preoperative/pretherapy diagnostics regarding hyperadrenocorticism testing, abdominal ultrasound and CT findings related to the diagnosed adrenal tumor including mass location and evidence of vascular invasion, institution performing the CT, whether preoperative a blockade was administered, intraoperative evidence of vascular invasion and location, whether venotomy was performed, whether concurrent nephrectomy was performed, intraoperative complications, postoperative complications, survival at last available follow-up date, and whether evidence was available to conclude that death was related to pheochromocytoma diagnosis. Intraoperative complications evaluated included tachycardia, hypertension, arrhythmia, and any additional complications provided in the record. Tachycardia was defined as > 160 beats/min.¹⁰ Systemic hypertension was defined as systolic measurement above 160 mm Hg.^15,16 Evidence of a light plane of anesthesia was evaluated for when tachycardia and/or hypertension was observed. The specific anesthetic protocol for each patient was beyond the scope of this study and will not be discussed. Postoperative complications that were evaluated included hypertension, hypotension, arrhythmias, suspected thromboembolic events, adrenocortical insufficiency, hypoglycemia, and/or renal failure prior to discharge.¹⁷

Results

Thirty-four cases met the inclusion criteria: 14 nonpheochromocytoma patients (10 adrenocortical carcinomas, 4 adrenocortical adenomas) and 20 pheochromocytoma patients. One case from the nonpheochromocytoma group (adrenocortical carcinoma) was excluded due to the urine metanephrine/normetanephrine-to-creatinine results being obtained from a human laboratory. One case from the pheochromocytoma group was excluded, as the CT was unable to be downloaded to the Eclipse system to allow for volume measurements. Thirty-two cases were included for statistical analysis (Supplementary Table S1). Mean age for nonpheochromocytoma patients was 11.5 years. There were 4 male dogs (30.7%; 4 castrated, 0 intact) and 9 female dogs (69.3%; 9 spayed, 0 intact). Mean weight was 21.4 kg (range, 3.92 to 40.8 kg). The most common breed was mixed breed (5 [38.4%]), with 1 each of other breeds including Labrador Retriever, Beagle, Labradoodle, Shih Tzu, Affenpinscher, Great Pyrenees, American Bulldog, and Maltipoo. Mean age for pheochromocytoma patients was 11.33 years. There were 12 male dogs (63.1%; 11 castrated, 1 intact) and 7 female dogs (36.9%; 7 spayed, 0 intact). Mean weight was 9.1 kg (range, 3.62 to 37 kg). The most common breed was mixed breed (5 [26.3%]), followed by Shih Tzu (2 [10.5%]), French Bulldog (2 [10.5%]), and 1 each of Maltese, Coton de Tulear, Boston Terrier, Cockapoo, Australian Shepherd, Siberian Husky, Beagle, Labrador Retriever, Staffordshire Terrier, and Chihuahua. Age, sex, and body weight were not statistically different between pheochromocytoma and nonpheochromocytoma patients, and breed comparison was not possible due to population size. Although body weight tended to be higher for nonpheochromocytoma patients than pheochromocytoma patients, this was not statistically significant.

There was no statistical difference in metanephrine-to-creatinine values for pheochromocytoma and nonpheochromocytoma patients (P = .2498). The median metanephrine-to-creatinine value in the nonpheochromocytoma group was 95 with a range of 38 to 255. The median metanephrine-to-creatinine value in the pheochromocytoma group was 119 with a range of 26 to 998. The normetanephrine-to-creatinine ratio was significantly elevated in pheochromocytoma patients compared to nonpheochromocytoma patients (P = .0009). The median normetanephrine-to-creatinine value in the nonpheochromocytoma group was 259 with a range of 45 to 749. The median normetanephrine-to-creatinine value in the pheochromocytoma group was 523 with a range of 147 to 3,270.

When metanephrines were evaluated, 0 patients in the nonpheochromocytoma group were elevated and all 13 were within normal values. In the pheochromocytoma group, only 2 patients had elevated metanephrines and 17 were within normal values. Sensitivity, specificity, positive predictive value, and negative predictive value for the metanephrine test were not statistically significant (Table 1). When normetanephrine was evaluated, 3 patients in the nonpheochromocytoma group were elevated and 10 were within normal values. In the pheochromocytoma group, 15 patients had elevated normetanephrine and 4 were within normal values. Sensitivity, specificity, positive predictive value, and negative predictive value for normetanephrine results were statistically significant (P = .033).

Larger tumor volume did not correlate to higher urine levels of metanephrine or normetanephrine (r² = 0.066 or 0.021, respectively). Regarding silent pheochromocytomas, metanephrine values in 89.5% of dogs with pheochromocytomas were within the reference range and normetanephrine values in 21% of dogs with pheochromocytomas were within the reference range. Tumor volume did not predict whether a pheochromocytoma was silent (Table 2) on the basis of previous conclusions that larger tumor volume did not correlate to higher urine levels of metanephrine/normetanephrine. Higher urine metanephrine/normetanephrine levels were not predictive of vascular invasion. Fourteen of 19 pheochromocytoma patients had phrenicoabdominal vein invasion, 1 of 19 had renal vein invasion, and 12 of 19 had caudal vena cava invasion. For the purposes of statistical analysis, all venous invasions were combined together to total 14 of 19 for venous invasion. When the invasion group was compared with the noninvasion group to metanephrine/normetanephrine values, there was no statistical difference (P values of .467 and .53, respectively) (Supplementary Table S2).

Tumor volume and higher urine catecholamine values did not predict perioperative complications. These were not statistically significant, with P values of .101, .183, and > .999 for tumor volume (Supplementary Table S3), metanephrine, and normetanephrine, respectively. Perioperative complications were defined as tachycardia, hypertension, or cardiac arrhythmia on the basis of the criteria listed previously in the Methods section. Median tumor volume for patients that experienced no intraoperative complications was 5.835 cm³, with a range of 1.87 to 97.72 cm³. Median tumor volume for patients that experienced intraoperative complications was 3.78 cm³, with a range of 1.86 to 10.71 cm³. Table 3 shows metanephrine/normetanephrine levels compared to patients with or without intraoperative complications.

The median survival time was not reached in this group at the time of study completion. Of 19 patients diagnosed with pheochromocytoma, 2 died due to potential tumor-related causes at the time of study completion (pulmonary metastatic disease and a ruptured hepatic mass that could not definitely be ruled out as a metastatic lesion). Thirteen out of 19 dogs were alive at last follow-up, and 4 of 19 died due to nontumor-related causes (cardiopulmonary arrest while under chemotherapy for prostatic carcinoma, progressive pulmonary hypertension, suspected progression of pituitary macroadenoma, and progressive, preexisting renal disease). The median follow-up time from the date of surgery was 334 days, with a range of 9 to 1,127 days.

Discussion

In general, the patient population data within this study appeared to align with the patient population of previous studies evaluating canine pheochromocytomas.^3,18,19 There was no significant difference in regard to breed, sex, age, or weight predilection between pheochromocytoma and nonpheochromocytoma patients within the current study.¹⁹

Results of the present study in dogs with adrenal tumors supported the hypothesis that specificity and sensitivity of urine spot samples to diagnose canine pheochromocytoma would be high when using spot urine normetanephrine values. Human studies have recognized the burden associated with 24-hour urine sample collection and sought to evaluate the diagnostic accuracy of spot urine metanephrines. The resulting literature provided supporting evidence that spot urine sampling is an acceptable alternative to plasma and 24-hour urine collections in the diagnosis of pheochromocytomas.^20,21

On the basis of findings within this study, normetanephrine levels have the potential to differentiate a pheochromocytoma from a nonpheochromocytoma. A second study²² from 2010 reiterated the ability of normetanephrine to differentiate pheochromocytomas from nonpheochromocytomas; however, the sample size was considerably low, with results based on a single-digit population size. Diagnostic laboratories reference the Quante study findings when interpreting urinary metanephrines, and a cutoff of 4 times the upper reference limit is used to be diagnostic for a pheochromocytoma.²³ Figure 2 highlights the number of patients that fell below 4 times the upper reference limit, showing that only 5 patients would have been diagnosed with a pheochromocytoma by use of that metric. Arguably, this presents the most significant conclusion that the vast majority of patients would be diagnosed as not having a pheochromocytoma if the rule of 4 times the upper reference limit were followed.

View Full Size

Each patient is portrayed along the x-axis of the plot with their respective normetanephrine-to-creatinine value on the y-axis of the plot. The upper reference limit is portrayed by the solid horizontal line at value 380, and 4 times the upper reference limit is portrayed by the dashed line at value 1,520.

Citation: Journal of the American Veterinary Medical Association 263, 3; 10.2460/javma.24.06.0409

• Download Figure
• Download figure as PowerPoint slide

The aforementioned finding is supportive specifically for normetanephrine values as opposed to both metanephrine and normetanephrine. Within the current study population, metanephrine levels could not differentiate a pheochromocytoma from a nonpheochromocytoma. Figure 3 highlights that the majority of patients fell below the upper reference limit and no patients lay above the previously suggested rule of 4 times the upper reference limit. Overall, the sensitivity, specificity, and negative and positive predictive values for metanephrine testing were poor and likely not worth considering as part of evaluation to differentiate pheochromocytoma from nonpheochromocytoma. This is consistent with previous human and veterinary studies that have recommended evaluating normetanephrine as the primary value.^5,22,24–26 Although pheochromocytomas secrete epinephrine and norepinephrine, these catecholamines are metabolized within the tumor to their respective metabolites, metanephrine and normetanephrine. These metabolites are continuously secreted into circulation. The constant secretion of metabolites as opposed to the episodic secretion of epinephrine and norepinephrine is why plasma or urine metanephrines are superior to catecholamines when evaluating for a pheochromocytoma.²⁷ The proposed reasoning behind normetanephrine being superior to metanephrine has been shown to be due to the fact that pheochromocytomas produce higher levels of norepinephrine than epinephrine.^22,25

View Full Size

Each patient is portrayed along the x-axis of the plot with their respective metanephrine-to-creatinine value on the y-axis of the plot. The upper reference limit is portrayed by the solid horizontal line at value 359, and 4 times the upper reference limit is portrayed by the dashed line at value 1,436.

Citation: Journal of the American Veterinary Medical Association 263, 3; 10.2460/javma.24.06.0409

• Download Figure
• Download figure as PowerPoint slide

Vascular invasion of pheochromocytomas in human literature is quoted to be “an extremely rare” phenomenon, with the majority of literature noted to be case reports.^28,29 This is contrary to veterinary cases, in which up to 50% of tumors have local vascular and/or soft tissue invasion.³⁰ Approximately 35% of pheochromocytomas are quoted to invade the caudal vena cava.³¹ This study aimed to evaluate 3-D measurements of the tumor and tumor thrombus to allow for a more accurate representation of tumor size in dogs. In the current study, larger tumor volume did not correlate to higher urine levels of metanephrine nor normetanephrine levels, and the null hypothesis that tumor volume would correlate with urinary metanephrines was rejected. In human medicine, it has been repeatedly noted that tumor size is associated with higher free metanephrines.^7,13,32 On the basis of the findings of this study, the same does not hold true in veterinary medicine. This means that tumor size alone likely cannot increase the diagnostic suspicion of a pheochromocytoma. The conclusion can also be made that small tumors are capable of producing large levels of catecholamines, and even a small pheochromocytoma can cause unexpected and catastrophic consequences.

As observed in Figure 2, there is a subset of patients with a histologically diagnosed pheochromocytoma that have normetanephrine values within the reference range, classically termed “silent” pheochromocytomas. Recent human literature established more specified nomenclature for the ways in which “silent” can be interpreted. The categories established were clinically silent, nonsecretory, biochemically negative, and nonfunctional.¹² In short summary, “clinically silent” refers to absence of signs/symptoms, “nonsecretory” to being without catecholamine secretory activity consistently over a 24-hour period, “biochemically negative” to free metanephrines values within the reference range, and “nonfunctional” to complete absence of catecholamine synthesis.¹² In the case of this study, the term “silent” refers to biochemically negative pheochromocytomas. The incidence of biochemically negative pheochromocytomas in human medicine varies between studies but overall appears to be uncommon, ranging from 0.2% to 9%.^33,34 Results of this study showed that normetanephrine levels were within the reference range in 4 of 19 patients. Considering that normetanephrine has been established as the value to interpret, this concludes that 21% of cases were biochemically silent pheochromocytomas. Table 2 highlights the findings comparing tumor volume in patients with normal (biochemically silent) and elevated normetanephrine values. In this case, we rejected the null hypothesis that silent pheochromocytomas would be rare. We also rejected the null hypothesis that silent pheochromocytomas would be predictive on the basis of tumor volume due to the previously discussed finding that tumor volume does not correlate with normetanephrine levels. Although published less frequently in human medicine, the findings from these data suggest that biochemically negative pheochromocytomas are not uncommon. Although a normal normetanephrine makes a pheochromocytoma less likely, it does not definitively rule it out. This possibility should be kept in mind when considering possible anesthetic complications and when handling the tumor intraoperatively in attempts to avoid massive catecholamine release.

Secondary objectives within this investigation included assessing whether higher urinary metanephrines are a predictive value of vascular invasion; establish whether there is an association between larger tumor volume, elevated metanephrines, and decreased survival time; and determine whether increasing tumor volume and elevated metanephrines is associated with more intraoperative complications. On the basis of the results, it was found that higher levels of urinary metanephrines are not predictive for vascular invasion. Therefore, the null hypothesis was rejected.

The comparison between increasing tumor volume and elevated metanephrines with overall survival time was unable to be made, as only 2 dogs died from potential tumor-related causes, so tumor size and levels could not be correlated. The null hypothesis that there would be no association between tumor volume and/or urinary metanephrines and survival time could not be rejected or proven on the basis of obtained data. A conclusion could potentially be extracted with a larger subset of patients followed over a longer period of time. More recent studies quote a tumor-related median survival of 3.96 years for patients with pheochromocytoma.³⁵ This was not compared specifically to tumor volume or metanephrines; however, those authors did note that vascular invasion did not affect survival to discharge.³⁵

Episodic secretion of epinephrine and norepinephrine by pheochromocytomas can cause intraoperative complications not as commonly associated with other surgical procedures. Lang et al³⁶ found that 35% of elective and 100% of emergency adrenalectomies for diagnosed pheochromocytomas in canine patients experienced intraoperative complications (hypotension, hypertension, tachycardia, arrhythmias, and hemorrhage). Although intraoperative complications in veterinary patients is described and documented, primary anesthesia studies are lacking. Although perioperative complications is a secondary focus of this study, an in-depth analysis of anesthetic procedures and management was not covered and is beyond the scope of this manuscript. However, intraoperative complications due to pheochromocytomas are well documented in human literature. These tumors store and secrete excessive levels of epinephrine and norepinephrine, and release is further perpetuated by anesthetic induction and surgical manipulation of the mass. The most common intraoperative complications for patients undergoing pheochromocytoma resection are a hypertensive crisis and arrhythmias.^37,38 A study³⁸ looking at hemodynamics of human patients undergoing pheochromocytoma resection found that hemodynamic oscillations were more pronounced when preoperative catecholamine levels were greater than twice the upper reference limit. The findings of the current study concluded that tumor volume does not predict whether a patient is more likely to experience the aforementioned complications. In this case, the null hypothesis that increasing tumor volume would be more likely to have intraoperative complications was rejected. Catecholamine secretion in excess will cause activation of a- and b-receptors, producing the negative perioperative effects observed with pheochromocytomas.¹⁶ While activation of the a receptors leads to vasoconstriction and hypertension, activation of b-1 receptors will increase cardiac inotropic and chronotropic effects, leading to tachycardia.¹⁶ When considering the above mechanism of the development of clinical signs and the previous finding that larger tumor volume does not correlate with increased catecholamine production, it thus makes sense why larger tumor volume does not correlate to an increase in intraoperative complications related to epinephrine/norepinephrine production.

Limitations existed within the current study. Due to the retrospective nature of the study, only patients with complete medical records and accessible CT images were included. Study data were dependent on accuracy of medical records. In addition, records were selected on the basis of a system search of the procedure term “adrenalectomy.” It is possible that patients that underwent an adrenalectomy concurrently with another procedure were sorted under a different code and missed in the search process. Histopathology samples were reviewed by multiple pathologists from an external lab; it may be a limitation that not all samples were reviewed by a single pathologist. Although this study presents a case population larger than many previous reports evaluating pheochromocytomas and urinary metanephrines, the sample size is still not considered large and as such could present type II statistical error. For example, when the relationship between urine catecholamine values and perioperative complications was evaluated, although the normetanephrine value appeared higher in patients experiencing complications, this was not statistically significant. Potentially, a higher power of study would change the statistical significance.

In conclusion, arguably the most significant finding of this study was that the rule of 4 times the upper reference limit in spot urinary metanephrines used to diagnose pheochromocytomas will lead to a significant number of missed diagnoses. In the present study, only 5 of 19 cases would have been considered within the diagnostic range. When normetanephrine values were being evaluated, there was a statistical difference between pheochromocytomas and nonpheochromocytomas, proving that normetanephrine levels obtained from spot urine samples have the potential to differentiate a pheochromocytoma from a nonpheochromocytoma. This did not hold true for metanephrine values within our study population. With the above information in consideration, it is reasonable to conclude that any normetanephrine value above the upper reference limit should increase the diagnostic suspicion of a pheochromocytoma, and the case should be treated as such to best avoid pre- and intraoperative complications. Of the included cases, 21% were biochemically silent pheochromocytomas. Although a normal normetanephrine makes a pheochromocytoma less likely, it does not definitively rule it out. Larger tumor volume did not correlate to higher urine levels of metanephrines, and higher levels of metanephrines were not predictive of vascular invasion. Within our study population, tumor volume was also unable to predict whether a patient is more likely to experience intraoperative complications related to catecholamine release. Small tumors should be treated no differently in terms of anticipated anesthetic complications and tumor handling to avoid life-threatening complications.

Supplementary Materials

Supplementary materials are posted online at the journal website: avmajournals.avma.org.

Disclosures

The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.

Funding

The authors have nothing to disclose.

ORCID

D. H. Thamm https://orcid.org/0000-0002-8914-7767