The feasibility and safety of trans-colorectal endoscopic ultrasound-guided fine-needle aspiration: a retrospective study of Japan
Article information
Abstract
Background/Aims
Endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) is an effective diagnostic technique; however, few studies have evaluated the efficacy of trans-colorectal EUS-FNA. This study assessed the feasibility of trans-colorectal EUS-FNA.
Methods
We retrospectively analyzed 76 consecutive patients who underwent trans-colorectal EUS-FNA for pelvic lesions between January 2013 and September 2023.
Results
A total of 76 pelvic lesions were identified. The median number of EUS-FNA punctures was 3 (1–8). The median lesion size was 18.9 (8.2–100.0) mm. The success rate was 98.7% (75/76), with no reported adverse events. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of EUS-FNA were 94.3% (50/53), 100% (22/22), 100% (50/50), 88.0% (22/25), and 96.0% (72/75), respectively. Malignancy was initially suspected in 65 patients before EUS-FNA; however, 25 patients showed benign results. Of these, three were later reexamined and diagnosed with malignancy, three underwent surgery and were found to have benign pathology, and 19 avoided unnecessary surgery.
Conclusions
Trans-colorectal EUS-FNA is a safe and effective diagnostic procedure.
INTRODUCTION
Pelvic lesions, intra-abdominal lymph nodes, and colorectal submucosal tumors can result from a variety of benign and malignant diseases. These lesions are usually detected by ultrasound, computed tomography (CT), or magnetic resonance imaging (MRI); however, the diagnostic accuracy of imaging alone is limited. Moreover, in patients with preexisting malignant disease, future treatment plans often depend on whether newly identified lesions are malignant.
Endoscopic ultrasound-guided fine-needle aspiration and biopsy (EUS-FNA/FNB) is a minimally invasive diagnostic endoscopic technique that differentiates benign from malignant lesions by allowing the acquisition of tissue for cytological and histological examination. EUS-FNA is an established technique for obtaining pathological specimens from submucosal tumors of the digestive tract,1 mediastinal masses,2 pancreatic tumors,3,4 upper abdominal lymph nodes,5 and liver lesions.6 It is commonly employed for the qualitative assessment of lesions in the upper gastrointestinal tract. Although several studies have reported the application of EUS-FNA for pelvic lesions via the lower gastrointestinal approach,7-9 the feasibility and safety of trans-colorectal EUS-FNA remain insufficiently investigated. Therefore, we conducted this retrospective study to assess its feasibility, efficacy, and safety.
METHODS
Patients
This study was conducted at a single academic care center (Aichi Cancer Center Hospital). A database analysis was performed to identify consecutive patients who underwent trans-colorectal EUS-FNA among all EUS-FNA procedures conducted between April 2013 and September 2023. We reviewed 76 patients who underwent trans-colorectal EUS-FNA for intra-abdominal tumors and intra-abdominal lymph nodes. No exclusion criteria were applied.
Study outcomes, reference methods for the final diagnosis, and statistical analysis
The primary outcome of this study was the success rate of EUS-FNA, defined as the ability to puncture the target lesion under EUS guidance and obtain a specimen suitable for histological evaluation. Secondary outcomes included the diagnostic performance and safety of EUS-FNA. The final diagnosis served as the reference standard for comparison with EUS-FNA-based diagnoses. In patients who subsequently underwent surgery, the final diagnosis was determined based on surgical pathological findings. For patients who did not undergo surgery, a malignant diagnosis based on EUS-FNA pathology was considered the final diagnosis. In cases where EUS-FNA pathology indicated a benign lesion, the final diagnosis was established based on at least six months of clinical follow-up, including a review of medical history, clinical symptoms and signs, and imaging findings such as CT and MRI.
Diagnostic success was defined as the successful completion of EUS-FNA targeting the lesion and obtaining specimens that enabled the identification of malignant disease through methods other than EUS-FNA or surgery. The accuracy of EUS-FNA was assessed based on the concordance between the tissue diagnoses obtained from the biopsies and the final diagnoses determined by the aforementioned methods.
The sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of EUS-FNA for diagnosing malignant pelvic space-occupying lesions were calculated. Outcome parameters were expressed with 95% confidence intervals (CIs).
Clinical methods
Ultrasound endoscopy was performed using the GF-UCT260 and TGF-UC260J models (Olympus) and the EG-580UT model (Fujifilm). Puncture needles included the EZ Shot 3 Plus (Olympus), Acquire TM (Boston Scientific), and Trident TM (Micro-Tech). Needle sizes used were 19-G, 22-G, or 25-G. All EUS-FNA procedures were performed on inpatients. Following preoperative evaluation and bowel preparation with a cleansing enema, patients were positioned in the left lateral decubitus position and administered intravenous sedation or anesthesia. Electrocardiographic monitoring and oxygen supplementation via nasal catheter were maintained throughout the procedure. Lesion characteristics, including size, morphology, location, and anatomical relationships with adjacent organs and tissues, were documented. Lesions smaller than 20 mm, those lacking nearby anatomical landmarks, or those located deeper in the colon beyond the rectum were anticipated to be difficult to visualize with EUS alone; therefore, fluoroscopy was concurrently utilized (Fig. 1). Fluoroscopy was used during EUS-FNA in 53.9% (41/76) of patients. Fine-needle aspiration (FNA) was performed under real-time ultrasound guidance using color Doppler to identify the optimal puncture site along the shortest possible trajectory while avoiding major blood vessels. In cases where the lesion was cystic, puncture was avoided.
Endoscopic ultrasound (EUS)-guided fine-needle aspiration procedure with fluoroscopy. (A) Computed tomography confirmed the lesion (arrow). (B) Fluoroscopy was used to check the endoscope orientation and height. (C) EUS detection of the lesion.
The combination of forward-view EUS (FV-EUS) and fluoroscopy facilitated relatively easy advancement of the endoscope deep into the colon (Fig. 2). EUS-FNB using a 19-G needle was performed without aspiration pressure to minimize the risk of bleeding. In contrast, EUS-FNA/FNB with a 22-G or 25-G needle was conducted using either a 10 or 20 mL negative pressure technique or a slow-pull method to apply minimal negative pressure.
Fluoroscopic image of forward-view ultrasound-guided endoscope inserted into the cecum. A forward-viewing endoscope was inserted deep into the colon with the aid of fluoroscopy.
To improve the diagnostic performance of EUS-FNA, the collected tissue was subjected to rapid on-site cytological evaluation and subsequently fixed in formalin for histological diagnosis.10
Perioperative treatment
Prophylactic intravenous antibiotics were administered on the day of the procedure and the following day. Patients also received intravenous carbazochrome sodium sulfonate for hemostasis and fluid replacement therapy. Immediate adverse events (AEs) were monitored until the day after the procedure. If no complications were observed and blood test results remained stable, oral intake was resumed, and the patient was discharged. AEs were assessed according to the guidelines established by the American Society for Gastrointestinal Endoscopy workshop.11
Ethical statements
This study adhered to the ethical guidelines outlined in the Declaration of Helsinki and was conducted in accordance with the requirements of the Institutional Review Board of Aichi Cancer Center Hospital (IR061026: 2024-0-063). Written informed consent was obtained from all patients prior to the procedure.
RESULTS
Patient characteristics
A total of 76 patients underwent trans-colorectal EUS-FNA at our institution between April 2013 and September 2023. The characteristics of the patients are summarized in Table 1.
Patients’ characteristics
Indications for EUS-FNA
Fifty-seven patients had previously undergone surgical or endoscopic treatment for cancer. Follow-up imaging modalities such as CT, MRI, or colonoscopy revealed findings such as a pelvic mass or enlarged lymph node, leading to the performance of EUS-FNA due to suspected recurrence of malignancy.
Eleven patients had no history of cancer, and EUS-FNA was performed to investigate pelvic masses or enlarged lymph nodes for an initial diagnosis.
Eight patients had active cancer, and CT or MRI revealed pelvic masses or enlarged pelvic lymph nodes; EUS-FNA was performed because distant metastasis or dissemination was suspected.
The indications for EUS-FNA are shown in Table 2.
Indications for endoscopic ultrasound-guided fine-needle aspiration
Technical performance
The success rate of the procedure was 98.7% (75/76). In one case, the target lesion—a 15 mm periaortic lymph node identified on contrast-enhanced CT and suspected to represent postoperative recurrence of colorectal cancer—could not be visualized with EUS, even with fluoroscopic guidance. Visualization was impeded due to the inability to identify adjacent anatomical landmarks, including the aorta. As a result, EUS-guided puncture could not be performed, and the final diagnosis was obtained through surgical pathology, which revealed a benign lymph node. The EUS-FNA techniques employed are detailed in Table 3.
Endoscopic ultrasound-guided fine-needle aspiration techniques
No AEs (perforation, bleeding, mild abdominal pain and discomfort, aspiration pneumonia, and anesthesia-related) related to the EUS-FNA procedures were recorded.
Pathological diagnosis
In 75 cases, specimens were successfully obtained using EUS-FNA, of which 25 were benign and 50 were malignant. The benign cases included 12 with inflammatory cells, six with normal lymphocytes, two leiomyoma, two specimens of normal colonic mucosal tissue, one case of endometriosis, one case of myometrium-derived tissue, and one case of normal mucus (Table 4). Among the malignant cases, there were 42 adenocarcinomas, two neuroendocrine tumors, one liposarcoma, one squamous cell carcinoma, one serous adenocarcinoma, one mucinous adenocarcinoma, one diffuse large B-cell lymphoma, and one neurofibroma (Table 5).
Benign pathological diagnoses obtained by endoscopic ultrasound-guided fine-needle aspiration
Malignant pathological diagnoses obtained by endoscopic ultrasound-guided fine-needle aspiration
Of the cases with benign EUS-FNA pathology results, 19 were confirmed to be benign after more than six months of clinical and imaging follow-up. In three cases, surgery was performed after imaging studies could not exclude malignancy; surgical pathology confirmed benign disease. Three cases were diagnosed as malignant after more than six months of follow-up: two were diagnosed based on imaging findings, and one by surgical pathology. In all three cases, the puncture targets were lymph nodes measuring 15 mm or less, and a 22-G FNA needle had been used. Although EUS-FNA pathology initially indicated benign lymph nodes, the sample volume was small, making it difficult to definitively distinguish between benign and malignant lesions.
Among the 50 patients with malignant pathology results from EUS-FNA, 38 had final diagnoses confirmed by surgical specimens, and 12 were diagnosed with malignant disease based on clinical course and imaging follow-up. In one case where the lesion could not be delineated by EUS, surgery was performed, and the pathology revealed a benign lymph node. Overall, the final diagnoses were benign in 23 patients and malignant in 53 patients.
Table 6 summarizes the comparison between pre-EUS-FNA imaging diagnoses (CT and MRI), EUS-FNA pathological diagnoses, and final diagnoses.
Imaging and pathology results, final diagnosis
Diagnostic value of EUS-FNA
The sensitivity of trans-colorectal EUS-FNA for diagnosing malignant lesions was 94.3% (50/53), specificity was 100.0% (22/22), positive predictive value was 100.0% (50/50), negative predictive value was 88.0% (22/25), and overall accuracy was 96.0% (72/75) (Table 7).
Diagnostic value of endoscopic ultrasound-guided fine-needle aspiration
DISCUSSION
EUS-FNA is widely recognized for its safety and effectiveness in obtaining tissue samples from lesions within the gastrointestinal wall and surrounding structures. However, limited research has investigated the diagnostic accuracy of trans-colorectal EUS-FNA. For comparison, via the upper gastrointestinal tract, the reported accuracy of EUS-FNA is 70% to 90% for submucosal lesions,12 96% for pancreatic lesions,3 and 98% for upper abdominal lymph nodes.13 Among the patients in this study who achieved technical success with EUS-FNA, the overall accuracy for detecting malignancy was 96.0% (72/75), comparable to previously reported diagnostic capabilities.
Rzouq et al.8 reported on trans-colorectal EUS-FNA of space-occupying lesions in the pelvic region or bowel walls in 20 patients, finding a sensitivity of 90.0%, specificity of 100.0%, positive predictive value of 100.0%, and negative predictive value of 90.0%. A systematic review and meta-analysis conducted by Han et al.,9 which included ten studies with a total of 236 cases of lower gastrointestinal EUS-FNA for pelvic lesions, reported a sensitivity of 89.0% and specificity of 93.0% in differentiating benign from malignant lesions. The pathological diagnostic performance reported in these studies was similar to our findings and supports the accuracy of trans-colorectal EUS-FNA. Of the 25 patients with benign EUS-FNA pathology results in our study, 19 avoided surgery and demonstrated no changes on subsequent imaging follow-up. Thus, trans-colorectal EUS-FNA has the potential to prevent unnecessary surgeries.
However, among the 25 patients with benign EUS-FNA pathology results, three were later diagnosed with malignancy after more than six months of follow-up. In all three cases, EUS-FNA pathology initially showed benign lymph nodes, but the volume of the specimens was small, making it difficult to definitively diagnose benignity or malignancy. The limited specimen volume obtained may be attributed to the use of a 22-G FNA needle in all cases, selected due to the small size of the target lesions. To obtain larger specimen volumes with EUS-FNA, it is considered essential to monitor the amount of acquired tissue during the procedure and to carefully select the appropriate FNA needle.
The success rate of EUS-FNA was high at 98.7% (75/76). Because there are few anatomical landmarks within the abdominal cavity, lesions can sometimes be difficult to detect using EUS alone. In our study, fluoroscopy was employed in 53.9% (41/76) of EUS-FNA cases, particularly when preoperative radiological imaging identified lesions that appeared difficult to visualize with EUS. Fluoroscopy can assist EUS imaging by aligning the height and orientation of the EUS tip with the lesion’s location as seen on radiological images. Oblique-viewing EUS is difficult to advance beyond the sigmoid colon and is not well-suited for visualizing lesions in the deep colon. In contrast, FV-EUS allows for direct anterior observation and can be performed similarly to conventional colonoscopy. Using fluoroscopic guidance, FV-EUS can be advanced relatively easily into the deeper parts of the colon. In a study by Thinrungroj et al.,14 FNA was successfully performed in all patients using FV-EUS under fluoroscopic guidance, achieving a sensitivity, specificity, and accuracy for detecting malignant lesions of 91%, 100%, and 92%, respectively. No AEs related to trans-colorectal EUS-FNA were observed in this study.
Color Doppler ultrasound was utilized during EUS-FNA to avoid large vessels and to select the shortest and safest puncture route, minimizing the risk of bleeding. Additionally, when using a 19-G FNB needle, the puncture was performed without aspiration pressure to further reduce the bleeding risk.
At our institution, prophylactic antibiotics were administered before and after the EUS-FNA procedure to prevent infection. Furthermore, puncture of cystic lesions was avoided because of the higher associated risk of infection. In the meta-analysis by Han et al.9 on lower gastrointestinal EUS-FNA, AEs were reported in 1.7% (4/236) of cases, including two cases of abscess formation following puncture of cystic lesions, one case of gross hematuria, and one case of bleeding. Although trans-colorectal EUS-FNA is generally a safe procedure with a low incidence of AEs, careful selection of target lesions for puncture is crucial.
This study has several limitations. First, as a retrospective study conducted at a single center, the sample size was relatively small. Second, in some cases, the final diagnosis was based on EUS-FNA pathology findings combined with clinical and imaging follow-up of more than six months. In cases where EUS-FNA pathology suggests a benign lesion, a follow-up period longer than six months may be necessary for a more accurate assessment.
In conclusion, trans-colorectal EUS-FNA demonstrated high diagnostic accuracy and no early procedure-related complications. Therefore, it is a safe and effective diagnostic procedure. However, further validation through multicenter studies with larger sample sizes is necessary to confirm these findings.
Notes
Conflicts of Interest
The authors have no potential conflicts of interest.
Funding
None.
Author Contributions
Conceptualization: TKo, KH, NO; Data curation: TKo, TKu, SH, HK; Formal analysis: TKo, TKu, SH, HK; Investigation: TKo, KH, NO; Methodology: TKo, KH, NO; Project administration: TKo, KH, YY, MU, KO, TO, RK, IM, YM, AAAN; Resources: TKo, KH, NO; Supervision: KH, NO; Validation:TKo, KH, NO; Visualization: TKo, KH, NO; Writing–original draft: TKo, KH, NO; Writing–review & editing: all authors.