Pathology results of endoscopic ultrasound-guided tissue acquisition in retroperitoneal masses: a multicenter study

Introduction

The retroperitoneum is an area that contains multiple gastrointestinal organs, lymph nodes, and vascular structures. A diverse array of pathologies may manifest within this anatomical region, encompassing benign conditions as well as the more prevalent occurrence of primary or metastatic malignant tumors [1]. Given their location, retroperitoneal tumors can cause a diagnostic and therapeutic problem. To make an appropriate diagnosis, a tissue sample must be obtained using fine-needle aspiration (FNA), guided either by abdominal ultrasound or by contrast-enhanced tomography. Because of the location of the retroperitoneum and the consequent difficulty of access, these modalities carry complications [1].

Given the proximity of the endoscopic ultrasound (EUS) probe to the gastrointestinal wall, retroperitoneal lesions have become more accessible, and a safer approach to tissue acquisition for these types of tumors can be made. However, only limited information has been published on EUS tissue acquisition in retroperitoneal tumors [2-6]. This study evaluated the pathology results of EUS-guided tissue acquisition (EUS-GTA) in patients with non-hepato-pancreatic-biliary or non-adrenal retroperitoneal lesions.

Patients and methods

This was a retrospective study of data collected prospectively from electronic and paper records of adult patients (older than 18 years) with EUS-TGA of retroperitoneal masses. Based on the computed tomography (CT) scan, magnetic resonance imaging (MRI) or EUS findings, patients with masses from adrenal glands, kidneys, ureters or pancreas were excluded. Patients were seen from January 2006 to December 2019. Patients from 5 referral centers in 2 countries were included. The local Institutional Review Board evaluated and authorized the protocol (REF. 3579).

All patients had complete blood count and prothrombin time before the procedure. Patients were continuously monitored throughout the procedure. Patients were placed in the left decubitus position. One anesthetist sedated the patients using a combination of midazolam, propofol and fentanyl. EUS-FNA was performed using a FUJI EG-530UT linear array echoendoscope with an SU-8000 console (Fujifilm Corporation, Minato-Ku, Tokyo, Japan) or an OLYMPUS GF UC140 EUM2 ultrasound gastroscope by 1 of 5 echo endoscopists. All patients were observed for at least 3 hours after the procedure to monitor possible complications.

Standard Echo Tip Ultra 22-G or 19-G needles (Cook Medical, Inc., Winston Salem, North Carolina, USA) were used for FNA, and Acquire needles calibers 19-G, 22-G, or 25-G (Boston Scientific, Inc., Ireland) for FNB. Procedures via D2 were performed using 25-G needles. Those in a different position (D1 or transgastric route) were carried out using a 19-G or 22-G needle, according to the physician’s preference. After 2014, patients underwent EUS-GTA using the fanning technique. Before 2014, no evidence was available about this technique [7].

EUS-GTA technique

First, the transducer was brought into a stable position. The needle was then introduced into the biopsy channel, and when it was inserted completely the Luer-lock handle was firmly screwed into the biopsy channel. The stylet was slightly retracted, the needle was positioned into the mass, negative pressure was connected, and the biopsy needle was moved forward into the lesion under real-time ultrasound control. The syringe piston was locked into this position for permanent suction. The needle was moved back and forth 10-15 times inside the lesion under complete ultrasonic control. With the needle tip still in the lesion, suction was released, and the needle was removed from the working channel.

All specimens were fixed in formalin and processed for histological and cytological analysis. In each center, a single expert pathologist evaluated the tissue samples. The cytological diagnoses were then categorized as positive for malignancy, benign/reactive process, or non-diagnostic. Material reported as suspicious for malignancy, or atypical cells indeterminate for malignancy, were considered negative (failures). The final diagnosis (the gold standard) was based on the results from the surgical specimen, and follow up for at least 6 months in non-operated cases was achieved via global clinical and radiological assessment.

Complications were defined according to the American Society for Gastrointestinal Endoscopy’s lexicon [8]. Immediate complications (intraprocedural and in the recovery area) were evaluated in all patients.

Results

Sixty-six patients with retroperitoneal masses were detected, and 23 were excluded (Fig. 1). Thus, 43 patients were included in the final analysis. The patients’ median age was 50.5 (range: 23-83) years, and 22 (51.2%) were female. EUS-GTA was performed in all these 43 patients (100%). A histological report from the EUS-guided tissue samples was available in all cases. Table 1 shows the characteristics of the included population. The mean size of the masses was 61.8±30 mm (median 60 mm, range: 18-150 mm). The most frequent histological diagnosis was adenocarcinoma/carcinoma in 14/43 (32.6%), followed by lymphoma in 10 (23.2%). Table 1 shows all the diagnoses obtained with EUS-TGA. In total, 31 (72%) patients had a malignant etiology: 17 were primary tumors from the retroperitoneum, and 14 were metastases. In these 14 patients with metastases surgery was avoided and medical treatment was started.

Figure 1 Flowchart of the patients included in the study

EUS-GTA, EUS-guided tissue acquisition; CPN, celiac plexus neurolysis

Table 1 Clinical characteristics of the patients included in this study

The positive pathological result rate for the first EUS procedure was 51.1% (22/43). Ten patients had a second EUS-GTA procedure and a diagnosis was achieved in 7 of them, for an overall diagnostic yield of 67.5% (29/43). FNA needles were used in 21 (48.8%) patients and FNB needles in 16 (37.2%). In 6 patients this information was unavailable. The caliber of the needles was 19 G in 10 (23.3%) patients, 22 G in 24 (55.8%) patients, and 25 G in 5 (11.6%) cases. In 4 (14%) cases the information was unavailable. The median number of passes was 2 (range: 1-7).

In 14/43 (32.5%) patients it was impossible to obtain the diagnosis by EUS-GTA: 4 underwent percutaneous biopsy, 4 surgical laparoscopy, 1 patient died before an extra procedure, and 5 patients were lost to follow up. A diagnosis was obtained in 8 of these patients: 4 had lymphoma, 2 retroperitoneal fibrosis associated with IgG4 disease, 1 liposarcoma, and 1 granulomatous lymphadenitis. Table 2 shows the univariate analysis to evaluate characteristics associated with diagnostic EUS-GTA.

Table 2 Univariate analysis to evaluate characteristics associated with a diagnostic EUS-GTA

No adverse events associated with EUS-GTA were reported. No mortality related to endoscopic procedures was reported.

Discussion

According to our results, EUS and EUS-GTA can frequently provide accurate tissue diagnosis, help clarify the diagnosis, and significantly impact subsequent management. Retroperitoneal masses are uncommon findings in general practice, and information about these cases is scarce. When these patients are seen in clinical practice, it is common for physicians to have questions about the best way to obtain tissue for diagnosis. Previously, image-guided or surgical interventions were the most common options. Tissue acquisition can be performed percutaneously, based on guidance from US and CT [9]. However, a randomized comparison between EUS-FNA vs. CT- or US-guided tissue samples for malignant pancreatic tumors has revealed better results for EUS [10]. EUS has emerged as an essential resource for obtaining tissue from all organs surrounding the gastrointestinal tract, including the retroperitoneal area [6] (Fig. 2). Because of this, in some institutions EUS-TGA has become the first-line option for obtaining tissue samples in most of these cases. The present study shows evidence of the utility of EUS-GTA in patients with retroperitoneal masses. EUS is an ambulatory procedure, with low risk and high diagnostic yield, that has been widely probed in different scenarios, mainly pancreaticobiliary [11-13], though evidence in other non-pancreatic scenarios has been reported [14,15].

Figure 2 Retroperitoneal mass: (A) oral phase; (B) venous phase; (C) EUS-FNB. Yellow arrow: mass. Red arrow: FNB needle

EUS, endoscopic ultrasound; FNB, fine-needle biopsy

Similarly to other reports [4,9], malignant lesions were the most common etiology for the retroperitoneal mass in this series. In all previous reports but one [4], kidney, pancreas or adrenal gland lesions were included. We excluded those patients in order to get a more homogenous sample. For patients with renal, adrenal or pancreatic masses, the clinician frequently does not have questions about the next step (take biopsies or surgery); moreover, as in kidney masses, patients are seen by other specialists other than gastroenterologists. However, patients who have CT scan reports of “retroperitoneal mass” with “unknown” origin represent a big issue for the clinician. Previous studies have reported that sarcomas are the most frequent primary retroperitoneal tumors, with lymphomas in second place [4]. Our findings are different as regards that point. In our study, malignant carcinomas were the tumors most frequently diagnosed (32%), followed by lymphomas (22%). The difference could be related to results from different populations, or to changes in epidemiology after almost 30 years [5]. The patients included in this study were Hispanic, and may differ from those in previous reports [4]. However, it is important to consider that the Hispanic population in the USA is currently almost 20% of the total, making up the largest minority.

Symptoms of retroperitoneal neoplasms include abdominal discomfort, fatigue, weight loss, occasionally fever or pain radiating to the back and thighs, and symptoms resulting from compression of adjacent organs. In our series, abdominal pain and weight loss were present in about 80% of the cases. Only in 3 patients was the finding of the retroperitoneal mass an incidental finding (in all 3 it was revealed by a CT scan). All included patients had undergone previous imaging before EUS, mainly CT scans (n=33) or MRI (n=7). In contrast, only 5 (11.6%) patients had a biopsy by a different method before EUS-GTA. This could be related to the fact that all patients were seen in referral centers where EUS is an available resource, and physicians are aware of the potential for access to the retroperitoneum using EUS. Furthermore, information about the excellent performance of EUS-GTA in retroperitoneal organs, mainly the pancreas, is widely available, including information from our center [11,12,16,17].

In most cases, EUS-TGA has become the first-line option for obtaining tissue samples in these institutions. Moreover, it is well known that CT often has difficulty visualizing some retroperitoneal lesions; thus, CT-FNA cannot be expected to have a higher yield when visualization of the lesion in question is problematic. Here, it should be emphasized that there are no clinical guidelines with recommendations about the best method for obtaining tissue samples in patients with retroperitoneal masses.

The diagnostic yield obtained in this report was not as excellent as it might be for other retroperitoneal organs, such as the pancreas. This could be related to cytological diagnosis in non-pancreatic retroperitoneal tumors, which can be challenging for cytopathologists. In most of the malignancies we detected in the retroperitoneal space, the final determination of malignancy may need a complete histological examination of the resected specimen. However, being able to get a presurgical diagnosis could be very important, because some patients may need medical treatment (such as lymphomas or metastatic carcinomas) instead of surgery.

Identifying factors that could predict success in getting a diagnostic tissue with EUS-GTA presents a challenge. In this study, we were not able to identify any of these factors. “Malignancy” in the histology was the only factor with a trend towards statistical significance in our analysis, but it was not reached. Interestingly, the use of FNB vs. FNA needles did not show differences. This could be related to the sample size or the technique used during the procedures. Unfortunately, this was a retrospective study from different centers, and we do not have information about the specific technique used in each patient.

As we mentioned, sample size and the study design could be considered the main limitations of this study. However, to our knowledge, this report represents the largest sample reported of patients with retroperitoneal masses who underwent EUS-GTA. Designing a prospective study with appropriate subgroups to study factors such as the type of needle or the technique used (wet suction, fanning technique, slow pull, etc.) could be very difficult in this group of patients, given their uncommon pathology. Another limitation is that not all our patients had IgG4 levels determined. The earliest patients included were seen starting in 2006, at which time IgG4 determination was unavailable in our countries. Only patients seen in the last ~5 years had IgG4 determination available. Rapid on-site evaluation (ROSE) proved helpful in cases where EUS-FNA was used; however, ROSE is not needed since FNB became available. We recommend using FNB needles plus macroscopic on-site evaluation in retroperitoneal masses [18]. Other techniques, such as elastography or contrast during EUS, have not been evaluated in cases of retroperitoneal masses; however, we consider that any of these could change the need for tissue for the correct diagnosis.

In conclusion, our results show that EUS-GTA can frequently provide accurate tissue diagnosis, help clarify the diagnosis, and significantly impact management. The information in this study could be used for future guidelines on this topic.