Role of Endoscopic Ultrasound-guided Fine-needle Aspiration in Adrenal Lesions: Analysis of 32 Patients

Abstract

Objective:

Endoscopic ultrasound-guided fine-needle aspiration cytology (EUS-FNAC) is a precise and safe technique that provides both radiological and pathological diagnosis with a better diagnostic yield and minimal adverse events. EUS-FNAC led to the remarkable increase in the detection rate of incidentaloma found during radiologic staging or follow-up in various malignancy or unrelated conditions.

Aims:

We did this preliminary study with an aim to evaluate the role of EUS-FNA in diagnosing and classifying adrenal lesions, clinical impact, and compare the outcome with the previously published literature.

Materials and Methods:

We included 32 consecutive cases (both retrospective and prospective) of EUS-guided adrenal aspirate performed over a period of 3.3 years. The indications for the aspirate in decreasing order were metastasis (most common carcinoma gall bladder) > primary adrenal mass > disseminated tuberculosis > pyrexia of unknown origin. On EUS, 28 cases revealed space occupying lesion or mass (two cases bilateral) and four cases revealed diffuse enlargement (two cases bilateral) with a mean size of 21 mm.

Results:

The cytology reports were benign adrenal aspirate (43.8%), metastatic adenocarcinoma (15.6%), histoplasmosis (9.4%), tuberculosis (9.4%), round cell tumor (6.2%), adrenocortical carcinoma (3.1%), and descriptive (3.1%). Three cases (9.4%) yielded inadequate sample. The TNM staging was altered in 22.23% of the cases by result of adrenal aspirate.

Conclusions:

EUS-FNA of the adrenal gland is a safe, quick, and sensitive and real-time diagnostic technique, which requires an integrated approach of clinician, endoscopist, and cytopathologist for high precision in diagnosis. Although the role of EUS-FNA for right adrenal is not much described, we found adequate sample yield in all the four patients that underwent the procedure.

INTRODUCTION

Endoscopic ultrasound-guided fine-needle aspiration cytology (EUS-FNAC) has emerged as a precise and safe technique that enables real-time visualization and onsite cytological examination to characterize not only the luminal gastrointestinal tract lesions but also the periluminal and abluminal organs such as pancreas, liver, kidney, adrenal glands, and both mediastinal and intra-abdominal lymph nodes for quick and more accurate diagnosis.[1,2] It has advantage over other simple imaging techniques as it provides both radiological and pathological diagnosis of the lesions. It is a simple, very sensitive, cost effective and safer procedure in comparison to previously used techniques with per-cutaneous approach (non-diagnostic yield about 14%), which enables timely optimal management of the disease.[3,4] The adverse event rate in EUS-FNA is 0.01%, which is much lesser in comparison to earlier used conventional computer tomography (CT) or trans-abdominal ultrasound guided per-cutaneous approaches (adverse events 0.4–12%).[5,6,7]

Since the advent of EUS technique, incidentally detected adrenal masses (incidentaloma) found during radiologic staging or follow-up in various malignancy and or unrelated conditions is more frequently diagnosed and characterized.[8] However, no well-defined criteria have been formulated to differentiate between benign and malignant adrenal lesions.

The aim of this preliminary study was to evaluate the role of EUS-FNA in diagnosing and classifying adrenal lesions, clinical impact, and compare the outcome with the previously published literature [Table 1].

Table 1.

Previous studies and their outcomes evaluating the role of endoscopic ultrasound-guided fine-needle aspiration of the adrenal gland

MATERIALS AND METHODS

This study was done in a single tertiary care centre over a period of 3.3 years including both retrospective and prospective cases between January 2013 and April 2016. A written informed consent was obtained from the relative of each patient after explaining the procedure and related complications. A combination of anaesthetic agents including intravenous midazolam (0.1 mg/kg) and Propofol (1 mg/kg) was given as per standard recommendations based on body weight and age of the patients for conscious sedation under cardio-respiratory monitoring.

Linear EUS was performed using (Olympus GF-UCT180, Tokyo, Japan) with a longitudinal convex ultrasound transducer and an ultrasound scanner in all patients. The left adrenal gland was identified by tracing descending aorta till celiac axis then rotating the scope clockwise with slight withdrawal. A trans-duodenal approach was followed to visualize right adrenal gland with endoscope in the long position along the greater curvature of the stomach. The right adrenal gland was identified between the upper pole of the right kidney and the inferior vena cava after visualizing these landmarks.

EUS-FNA was performed using a 22 gauge, 8 cm needle (Cook-Medical, Winston-Salem, NC; United States). Minimum two passes were done often avoiding suction. The aspirated material was smeared immediately and kept in 95% alcohol as well as air dried by a trained pathologist with rapid on-site evaluation (ROSE) for adequacy of the sample. ROSE was performed by keeping slide in Toluidine blue stain in a coplin jar for 30 seconds followed by washing in running tap water and examination under microscope. In selected cases with enough material, cell block preparation were also done. Post-procedure, patients were monitored in the recovery room for at least 60 min before discharge.

Three stains Giemsa, hematoxylin and eosin, and papanicolaou were routinely used as per the standard guidelines in all the cases along with special stains in case of infective etiologies such as Ziehl–Neelsen (ZN), periodic acid-Schiff (PAS) and silver methenamine (SM). Also, in selected cases Immunocytochemistry (ICC) was performed on smear/cell block sections as per the requirement. For ICC on smears, slides were kept in cold acetone overnight. Next day, slides were washed thrice in Tris buffer followed by peroxidise and protein blocking. Primary antibody was applied without performing antigen retrieval. Later, slides were washed in Tris buffer solution than HRP conjugated secondary antibody was applied and subsequently 3,3′-Diaminobenzidine (DAB), counterstaining with hematoxylin and mounting was done. However, standard procedure was followed in cell block immunohistochemistry. ICC was contributory in two cases of poorly differentiated metastatic carcinoma (Pan CK, Thermo-Fisher; RTU) and a case of adrenocortical carcinoma (synaptophysin and Inhibin-A, Thermo-Fisher; RTU).

RESULTS

Over a period of 3.3 years between January 2013 and April 2016, about 1500 EUS-FNA were performed in the department out of which adrenal aspirate comprised 2.13% (32 cases) [Table 2]. Two cases were managed by trans-abdominal ultrasound guided FNAC. The main indication for adrenal aspirate was metastasis 50% (16/32) followed by primary adrenal mass 28.13% (9/32), 12.5% (4/32) disseminated tuberculosis and 9.37% (3/32) cases of pyrexia of unknown origin. The clinical diagnosis in decreasing order were carcinoma gall bladder (10/32), carcinoma pancreas (4/32), adrenal mass/retroperitoneal mass (6/32), disseminated tuberculosis (4/32), unknown primary (2/32) and one case each of carcinoma lung, cholangiocarcinoma, carcinoma stomach, periampullary carcinoma, lymphoma and chronic liver disease.

Table 2.

Summarizing clinical details, indications, endoscopic ultrasound findings, radiological, and cytological diagnosis

The mean age was 53.8 years with a slight male predominance and male:female ratio of 1.2:1. On EUS, 28 cases revealed space occupying lesion (SOL) or mass lesion (two cases bilateral), and four cases showed diffuse enlargement (two cases bilateral) [Figure 1a–i]. The overall mean size of the adrenal lesion was 21 mm, while based on nature of pathology the mean size of adrenal was 10.64 mm in benign adrenal aspirate, 31.3 mm in infective etiology, and 32.4 mm in malignant lesions. On EUS, 27/30 cases were hypoechoic while in 3 cases they were heteroechoic. Three cases (9.37%) yielded inadequate sample. The cytology was reported as; benign adrenal aspirate (14/32), metastatic adenocarcinoma (5/32), histoplasmosis (3/32), tuberculosis (3/32), round cell tumor (2/32), adrenocortical carcinoma (1/32), and descriptive (1/32).

Figure 1.

Endoscopic ultrasound images showing spectrum of diseases; Primary adrenal tumors (a) A heteroechoic lesion in retroperitoneum, (b) A lobulated heteroechoic mass in right suprarenal region, (c) A homogenous hypoechoic right adrenal mass; histoplasmosis (d) A hypoechoic mass in left adrenal, (e) A hypoechoic circumscribed lesion in left adrenal, (f) Enlarged left adrenal along with right adrenal enlargement (not in the plane); metastasis (g) Enlarged left adrenal with multiple large lymph nodes, (h) A hypoechoic mass in left adrenal with multiple periportal lymph nodes, (i) A hypoechoic SOL in left adrenal

Cytomorphological spectrum

In tubercular adrenalitis, smears were largely necrotic with occasional collection of epithelioid histiocytes [Figure 2a]. The ZN staining showed characteristic acid-fast bacilli [Figure 2b].

Figure 2.

FNAC images infectious lesions adrenal; tuberculosis (a) Smear showing necrotic material (Giemsa stain ×200), (b) ZN stain showing acid fast bacilli (arrow) (ZN stain ×1000, oil immersion); histoplasmosis (c) Smear showing yeast form of histoplasma with narrow-based budding (Giemsa ×1000, oil immersion), (d) Clusters of epithelioid histiocytes with histoplasma (arrow) forming granuloma (PAP stain ×1000, oil immersion), (e) Silver methenamine stain highlighting the fungal yeast (SM stain ×1000, oil immersion), (f) ZN stain showing negative shadows of histoplasma (arrow) (ZN stain ×1000 oil immersion) with PAS positivity (inset)

In histoplasmosis, the aspirates revealed both intracellular (both within cortical cells and histiocytes) and extracellular clusters and scattered small yeast form of histoplasmawith narrow-based budding, and occasional epithelioid cell granuloma. These yeasts were highlighted by SM and PAS stain and showed negative shadow in ZN stain [Figure 2c–f].

The round cell tumor aspirates were richly cellular comprising of monomorphic round to oval cells displaying mainly hyperchromatic nuclei, occasional nuclear moulding and overlapping, scant to moderate amount of cytoplasm, and indistinct cell borders and marked vacuolization artefacts in the background [Figure 3a–c].

Figure 3.

FNAC images of benign and malignant adrenal lesions; round cell tumor (a) and (b) a cellular aspirate showing monomorphic cells with hyperchromatic nuclei, nuclear moulding and scant cytoplasm (a-b: Giemsa stain ×400), and (c) nuclear pseudoinclusion (PAP stain ×400); metastatic adenocarcinoma (three cases) (d) (H and E stain ×400), (e) (PAP ×200), (f) (Giemsa stain ×400) A cellular aspirate showing tight cluster of atypical cells; benign aspirate (g) showing scant cellularity and vacuolization artefacts (Giemsa stain ×200), (h) scattered adrenal cortical cells (Giemsa stain ×400) and (i) A benign duodenal epithelial cell cluster (Giemsa stain ×400), Immunocytochemistry images of adrenocortical carcinoma showing (j) synaptophysin (×400), and (k) Inhibin-A (×400) positivity and metastatic poorly differentiate carcinoma showing (l) pan-CK positivity (×400)

Metastatic adenocarcinomas aspirates were variably cellular and showed tight to loosely cohesive cell clusters of atypical cells with moderate nuclear pleomorphism, hyperchromatic to vesicular nuclei, high N:C ratio, occasional prominent nucleolei, and moderate amount of eosinophilic cytoplasm with/without focal mucinous changes [Figure 3d–f].

In most of the benign adrenal aspirates, smears were scant to moderately cellular containing cortical cells with maintained N:C ratio in a hemorrhagic background with abundant vacuolization artefacts [Figure 3g and h]. In addition, many aspirates also showed benign intestinal epithelial cells [Figure 3i].

ICC showed synaptophysin and inhibin-A positivity in adrenocortical carcinoma [Figure 3j and k] and pan-cytokeratin [Figure 3l] positivity in two cases of poorly differentiated metastatic carcinoma.

DISCUSSION

EUS is a real time high resolution imaging modality that has significantly increased the detection of adrenal incidentaloma especially in patients with known primary malignancy.[16] Although EUS-FNA introduced about 2 decades [17] ago, but in India this technique came lately and more so available only in tertiary care centres. Adrenal gland is one of the common sites of metastasis in lung and gastro-intestinal cancers. It is very important to differentiate between primary benign/malignant lesions and metastasis for the proper management and staging of a case. Although conventional radiological techniques such as CT scan and MRI are useful modalities to rule out involvement of adrenal in different malignancies, but false positive and negative results were observed in about 10% of the cases.[11] EUS-FNA is a highly accurate technique with a good adequacy in comparison to earlier per-cutaneous methods. It has an advantage of providing real-time diagnosis in some cases due to incorporation of ROSE by expert pathologist. Also, the rate of adverse events associated with the per-cutaneous approach are much higher; 0.4–12% in comparison to 0.01% in EUS-FNA procedure.[5,7] In our study, none of the cases showed any procedure-related complications. In this series, adequacy of yield was 90.63%, which is comparable to previously published literature.[12,14,15]

Of the 18 patients including two with unknown primary evaluated for metastasis, five (27.78%) showed malignant lesion which is comparable to 29% positivity in series by Jhala et al.[9] and TNM staging altered in 22.23% of the cases. One case of carcinoma gall bladder revealed a second primary of the adrenal (adrenal cortical carcinoma) instead of metastasis. Incorporation of ancillary technique like ICC has proven very helpful in cytopathology. In this series, three cases (two cases of poorly differentiated metastatic carcinoma and one case of adrenocortical carcinoma) were conclusively diagnosed based on the results of ICC.

Size of the adrenal gland alone is not a very sensitive and specific criterion. In our study, the mean size of adrenal was 10.64 mm in benign adrenal aspirate, 31.3 mm in infective etiology, and 32.4 mm in malignant lesions. Matinez et al. reported a higher median adrenal diameter in benign lesions in comparison to malignant aspirates, while Eloubeidi et al. found an opposite result in their series.[12,18]

Adrenal incidentalomas are mostly non-functional (>90%) with a low (<10%) risk of being malignant, and the cumulative risk of malignant transformation is less than 1:1000.[19] About 2% of the incidentalomas shows a metastatic tumor.[10] In our series, 14/32 (43.75%) cases showed a benign aspirate and 8/32 (25%) malignant lesions. In 2/3^rd of the patients with a known malignancy undergoing evaluation for staging, if an adrenal lesion is detected show metastatic tumor commonly from carcinomas of lung, gall bladder, stomach, kidney, breast, and lymphomas.[20]

As most of the lesions were hypoechoic irrespective of etiologies, a confirmatory diagnosis is challenging based on only adrenal image findings in EUS. However, a complete assessment involving thoraco-abdominal organs, different group of lymph nodes along with clinical presentations have a much higher cumulative potential to improve certainty of the diagnosis. Further coupling FNAC, cell block preparation, and use of ancillary techniques ensure an accurate diagnosis in majority of the cases. In this series, EUS imaging alone made a correct diagnosis/differential diagnosis in about 1/3^rd of the cases.

Kievit et al. suggested that adrenal incidentaloma has important bearing on the life expectancy of patients, which can be decreased by a mean period of 12 months, if the tumor is undiagnosed or inappropriately treated.[21]

EUS has a potential to identify a normal or minimally enlarged left adrenal gland in almost all the patients compared with only a 69% by trans-abdominal ultrasound.[22] Reversely, for right adrenal lesions trans-abdominal ultrasound is a much better technique compared to EUS. A limited literature is available explaining the role of EUS-FNA in right adrenal lesion and its utility is still questionable. In our series right adrenal aspiration was done in six patients (trans-abdominal ultrasound = 02, EUS = 04), all of which revealed adequate sampling with the diagnosis of malignant pathology in four cases.

Decision on the adequacy of yield and onsite examination by a trained cytopathologist is of utmost importance, which reduces number of non-diagnostic samples and turnaround time of the endoscopic procedure.[23,24,25,26] In our institute, an integrated approach with the active engagement of treating physician, endoscopist and trained pathologist is followed to maintain the high yield and minimum turnaround time for reporting of a case.

Limitations of the study are small number of patients, lack of follow-up which restrain the further elucidation of exact nature of lesion reported as benign adrenal aspirate. No correlation with other imaging modalities like positron emission tomography, CT scan, and MRI limits the appropriate outcome of study. In two cases with right adrenal lesions, a trans-abdominal ultrasound approach was used.

To conclude, EUS-FNA of the adrenal gland is a safe, quick, and sensitive and real-time diagnostic technique, which requires an integrated approach of clinician, endoscopist, and cytopathologist for high precision in diagnosis. Although role of EUS-FNA for right adrenal is not much described, we found adequate sample yield in all the four patients underwent the procedure.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.