Ultrasound-guided aspiration cytology of retroperitoneal masses with histopathological corroboration: A study of 71 cases

Abstract

Background:

Diagnosis of retroperitoneal lesions is one of the most difficult areas in surgical pathology. Both primary and metastatic tumors can grow silently in retroperitoneum before the appearance of clinical signs and symptoms. Guided aspiration cytology has shown a promising role in diagnosis of lesions in this region.

Aims:

This study was undertaken to evaluate the reliability of guided fine-needle aspiration cytology (FNAC) in the diagnosis of retroperitoneal masses and to study the spectrum of retroperitoneal lesions in a rural tertiary care hospital.

Materials and Methods:

This study was done on 71 patients presenting with retroperitoneal masses over a period of 20 months in a tertiary care hospital. Ultrasound-guided FNAC was done and the smears were stained by MGG and H and E stains. Histopathological corroboration could be done on all neoplastic lesions along with some non-neoplastic lesions.

Results:

Out of 71 cases, 48 cases (67.6%) were found to be neoplastic. Malignant tumors (35 cases) were more common than benign ones (13 cases). Malignant lesions were predominantly composed of lesions of lymph node (17 cases i.e. 48.57%) followed by sarcomas (5 cases) and germ cell tumors. Some uncommon retroperitoneal lesions like adrenal myelolipoma, renal angiomyolipoma, Burkitt's lymphoma and adrenocortical carcinoma were found in our study. Guided FNAC could diagnose all the malignant lesions with 100% accuracy except in 2 cases of poorly differentiated carcinoma which came out to be non-Hodgkin lymphoma on subsequent biopsies.

Conclusion:

Guided FNAC is an inexpensive, rapid and reliable method for diagnosis of retroperitoneal masses.

Introduction

The emergence of fine-needle aspiration cytology (FNAC) as a popular and fairly accurate diagnostic modality has led to its use in diagnosing the lesions in the less accessible sites like the retroperitoneum. The retroperitoneum is a potentially large space and contains, embedded in the meshwork of loose connective tissue important structures like adrenal glands, kidneys and ureters, pancreas, aorta and its branches, inferior vena cava and its tributaries and numerous lymph nodes.[1] The problem with dealing retroperitoneal masses is due primarily to their uninhibited growth with no facial boundaries often leading to the large size attained before the development of symptoms and the establishment of a diagnosis.[2] The proximity to vital vascular and neural structures as well as intra-abdominal organs add to the problem.[2]

Imaging modalities like radiography, lymphography, fluoroscopy, ultrasound, computerized tomography and magnetic resonance imaging are essential to evaluate the retroperitoneal masses and to ensure adequate sampling of the lesions. The added advantages of ultrasonography are that they are relatively less expensive, rapid, do not cause any ionizing radiation hazards, do not require use of contrast medium,[3] and during the procedure of FNAC, the needle tip can be well visualized ensuring sampling from the required area of the lesion.

Here, we present a study on ultrasound-guided aspiration cytology of retroperitoneal masses performed in a rural medical college and tertiary care hospital; the objectives being to determine the prevalence of different benign and malignant lesions of the retroperitoneum and to correlate the cytological diagnoses with the histopathology.

Materials and Methods

The study was carried out following the permission of the Institutional ethics committee and proper consent from the participating patients. It was conducted in the Department of Pathology with the help of the Department of Radio-diagnosis in a tertiary care hospital covering a period of 20 months ( January 2011 to August 2012). A total of 71 patients presenting with retroperitoneal mass lesions were included in the study.

A thorough work up of patients was carried out including detailed history, review of records and clinical examination. FNAC was done under ultrasonographic guidance in the presence of a qualified radiologist with the help of a 22-gauge spinal needle attached to a 20-mL syringe. The air-dried smears were stained with May-Grünwald-Giemsa (MGG) stain while hematoxylin and eosin (H and E) stain was used for the alcohol-fixed smears. Ziehl-Neelsen (ZN) stain was used whenever caseous material or pus was aspirated and whenever there was a clinicoradiological suspicion of tuberculosis. Repeat aspirations were done in cases where the first attempt was inadequate or was inconclusive. The cytological diagnoses were correlated with the histopathological diagnoses as histopathological examination is considered to be the gold standard.

Results

The study population comprised of 71 patients. Most of the patients presented with abdominal pain or discomfort; other symptoms included abdominal mass, weight loss, fever etc. They were clinico-radiologically diagnosed with retroperitoneal masses and were subjected to ultrasound-guided FNAC of their lesions.

In 60 cases, the first aspirate yielded satisfactory material for reporting. In the remaining 11 cases repeat aspiration was required.

All the neoplastic lesions (both benign and malignant) were biopsied and subjected to histopathological examination (n = 48).

Out of 71 patients, 44 patients (61.97%) were male and 27 (38.02%) were female. The majority of patients were in the 30 to 60 year age group (30 cases, 42.25%) with a mean age of 37.5 years. The youngest was a 1.5-year-old boy who was diagnosed with Wilm's tumor and the oldest was a 75-year-old male presenting with non-Hodgkin lymphoma (NHL).

Malignant lesions were more common (35 cases, 49.29%) than the non-neoplastic lesions (23 cases, 32.39%) and benign tumors (13 cases, 18.3%).

Among the 23 non-neoplastic lesions, majority were tuberculous lymphadenitis (14 cases, 60.86%). Acid-fast bacilli (AFB) could be demonstrated in 10 of them by ZN stain. Besides these, some cases of benign cystic lesions, suppurative lesions and pyonephrosis were also diagnosed [Table 1].

Table 1.

Distribution of non-neoplastic retroperitoneal mass lesions: (N = 23)

Among the 13 cases of benign neoplasms in our study, 9 cases (69.23%) were of soft-tissue origin, 2 cases (15.38%) of retroperitoneal teratoma and 1 case (7.69%) each of renal angiomyolipoma and adrenal myelolipoma were found [Table 2].

Table 2.

Distribution of benign retroperitoneal tumorous lesions: (N = 13)

Of the 35 malignant lesions, 17 (48.57%) were of lymph node origin, 5 cases (14.29%) of soft tissue, 4 cases (11.43%) of renal and gonadal origin, 3 cases (8.57%) of pancreas and 2 cases (5.71%) of adrenal origin were found.

In the lymph node, non-Hodgkin lymphoma (NHL) was the most commonly found malignancy (11 cases, 64.70%) followed by metastatic adenocarcinoma (6 cases, 35.29%). One case of Burkitt's lymphoma was found in a 4-year-old sero-positive boy.

Among the 5 soft tissue sarcomas, cytologic diagnosis of pleomorphic sarcoma was rendered in 3 cases. One case each of embryonal rhabdomyosarcoma and malignant peripheral nerve sheath tumor was also diagnosed cytologically, which were corroborated histopathologically later on.

Two cases (50%) each of renal cell carcinoma and Wilms' tumor were found among the renal neoplasms. Of the germ cell category, 3 cases (75%) of seminoma and 1 case (25%) of yolk sac tumor were detected. Three cases of pancreatic adenocarcinoma and 2 cases of adrenocortical carcinoma were also diagnosed cytologically, which were later confirmed by histopathology [Table 3].

Table 3.

Distribution of malignant retroperitoneal neoplasms (N = 35)

Cytological and histopathological correlation done in the 48 cases of neoplastic lesions showed 95.83% correlation as 2 cases of NHL were misdiagnosed as poorly differentiated carcinoma on FNAC.

Discussion

The study was undertaken to look into the efficacy of aspiration cytology in correctly diagnosing the hard-to-reach retroperitoneal masses and thus to avoid unnecessary delay, complication and other hassles of a surgical maneuver.

The predominant age group of the patient was 30 to 60 years with a mean of 37.5 years, which is in contrast to the study of Mangal et al.[4] where the majority of the patients were in 6^th decade but in accordance to Gangopadhyay et al.[5] where the majority of the patients were in the 31- to 45-year age bracket.

The findings of the study were grouped as malignant, benign tumors and non-neoplastic lesions. The malignant lesions predominated (35 cases) i.e. 49.29% of the total cases, which is quite alarming and are comparable with the findings of Mangal et al,[4] Gangopadhyay et al.[5] and Miralles et al.[6]

Among the non-neoplastic lesions, tuberculous lymphadenitis was the most common comprising 60.86% of cases. None of the non-neoplastic lesions were biopsied.

Among the benign tumors, soft tissue tumors predominated (69.23%) with lipomas making the largest single sub-group (4 cases, 44.44%). This was followed by fibromatosis (2 cases, 22.22%) [Figure 1a]. In the remaining 3 cases, benign spindle cell lesion was diagnosed cytologically. One case showed benign spindle cells admixed with intercellular collagen with fibrillar appearance in which a diagnosis of benign spindle cell lesion with a probability of schwannoma was made, which was subsequently confirmed on histopathology. Other two cases were histopathologically diagnosed as one case each of neurofibroma and benign fibrous histiocytoma. Two cases of retroperitoneal teratoma were diagnosed, both of which showed the presence of keratinous debris and anucleate squames. One of them also showed mature squamous epithelium. One case of adrenal angiomyolipoma was seen, which yielded loosely clustered spindle cells in a background of fat droplets from a renal mass. The case of adrenal myelolipoma was also interesting with the aspirate showing a combination of hematopoietic cells including megakaryocytes with fat droplets [Figure 1b]. The efficacy of FNAC in diagnosing tumors of adrenal glands were thus similar to the study of Jhala et al.[7]

Figure 1.

(a) Spindle cells with bland nuclei and mild pleomorphism suggestive of fibromatosis (MGG, ×400); (b) Adrenal myelolipoma showing hematopoietic elements (a megakaryocyte is shown by arrow) in a background containing lipid droplets (MGG, ×400); (c) Burkitt's lymphoma showing dense basophilic cytoplasm with vacuoles (arrow highlighting a mitotic figure) (MGG,×1000); (d) Pleomorphic dyscohesive spindle cells with giant cells (inset) in a case of pleomorphic sarcoma (MGG, ×400)

Among the malignancies, lymphoid neoplasms predominated (17 cases; 48.57% of malignancies). A diagnosis of non-Hodgkin's lymphoma was given in 11 cases with 1 case being diagnosed as Burkitt's lymphoma [Figure 1c]. This was followed by 6 cases of metastatic adenocarcinoma in lymph node and all these cases corroborated with histopathological findings.

Soft tissue sarcomas comprised of 5 cases (14.29% of the malignancies) -3 of the cases diagnosed as pleomorphic sarcomas [Figure 1d] were categorized histopathologically as pleomorphic liposarcomas (2 cases, 40%) and one case of malignant fibrous histiocytoma (MFH). In one case of pleomorphic sarcoma, the probability of pleomorphic liposarcoma was mentioned [Figure 2a], which was confirmed on histopathology. The cases of embryonal rhabdomyosarcoma and malignant peripheral nerve sheath tumor were also correctly diagnosed cytologically.

Figure 2.

(a) Aspirate from pleomorphic sarcoma with giant cells and lipoblast-like cells, suggestie of pleomorphic liposarcoma (MGG, ×400) (b) Chromophobe renal cell carcinoma showing a dual populatin of large cells with abundant eosinophilic cytoplasm with binucleation and another group of smaller cells (MGG, ×400); (c) aspirate from a case of Wilm's tumor showing mesenchymal differentiation (MGG, ×100). Inset shows epithelial differentiation in the form of tubules (H and E, ×400); (d) adrenal corcical carcinoma (MGG, ×100). Inset shows cytoplasmic vacuolation (MGG, ×400);

Among the renal malignancies, one case each of clear cell type renal cell carcinoma and chromophobe renal cell carcinoma was accurately picked up by FNAC [Figure 2b]; the 2 cases of Wilm's tumor showed blastemal component in addition to epithelial (rosettes and tubules) and mesenchymal differentiation [Figure 2c] similar to Dey et al.[8]

Among the germ cell tumors, seminomas (3 cases) were more common than yolk sac tumor in this study. The aspirates from a 7 cm and another 10 cm adrenal mass yielded clusters of pleomorphic cells with vague acini formation, irregular chromatin, mitosis and fragile cytoplasm with vacuolation along with necrosis, which warranted a diagnosis of adrenal cortical carcinoma [Figure 2d]. Both the cases were later confirmed on histopathology. Three cases of well-differentiated mucinous adenocarcinoma of pancreas were also diagnosed.

Organ-wise, lymph node was the most common organ involved (33 cases) of which there were 14 cases of tuberculosis, 2 cases of reactive hyperplasia, 6 cases of metastatic deposits and 11 cases of non-Hodgkin's lymphoma including 1 case of Burkitt's lymphoma. Thus, in lymph node malignancies prevailed over the benign lesions, similar to the some studies[4,5] but different from others.[9] Soft tissue tumors were next with 14 cases but here benign tumors (9 cases) prevailed over the malignant ones (5 cases).

Thus, for the diagnosis of benign tumors, FNAC had the sensitivity and specificity of 100% and 76.92%, respectively. As for the malignant ones, although we could ascertain malignancy in all with an accuracy of 100%, we misdiagnosed 2 cases of non-Hodgkin lymphoma as metastatic poorly differentiated carcinoma cytologically, thus having the specificity of 83.33%.

Thus overall, the sensitivity and specificity of FNAC for diagnosis of retroperitoneal masses were similar to the previous studies of Mangal et al,[4] Gangopadhay et al.[5] and Droese et al.[10]

Cytological and histopathological correlation could be done in 48 cases of neoplastic lesions and there was 95.83% correlation between these two diagnostic modalities further emphasizing the utility of FNAC in the diagnosis of retroperitoneal mass lesions.

Conclusions

The study showed a sufficiently high sensitivity and specificity of guided FNAC in diagnosing retroperitoneal mass lesions. The pitfalls remain the failure to diagnose the sub-types of benign and malignant spindle cell lesions and lack of accuracy in grading of tumors. But the rapidity, cost-effectiveness, safety and acceptable accuracy of FNAC particularly under image guidance avoids unnecessary surgery and helps immensely in management of the patients.