Transcutaneous sonographically guided 18-gauge-needle biopsy of gastrointestinal tract wall lesions

Ultrasound-guided fine needle biopsy (USFNB) is a widely accepted procedure for diagnosis of abdominal and retroperitoneal space-occupying lesions. Gastrointestinal tract lesions, both inflammatory and/or neoplastic, may be accurately visualized by US; although the diagnosis of such lesions is conventionally obtained with endoscopic biopsy, the possibility to accurately obtain the diagnosis of gastrointestinal neoplasia by using a USFNB with 18 or usually 20–22-gauge has been reported. In this study we aimed to evaluate the safety and accuracy of USFNB using a 18-gauge aspiration needle (BIOMOL) in the diagnosis of gastrointestinal tract lesions.

81 patients (M/F 48/33, median age 68yrs, range 45–80) underwent in the last 15 years a percutaneous USFNB with a convex transducer (3.5–5MHz) provided with a biopsy device. All patients were fasting from 8 hours and signed a written informed consent; no bowel preparation or antibiotic prophylaxis was used. The procedure was usually performed without local anaesthesia. Contraindications were only severe coagulation impairment (platelet count=40000/mm³; prothrombine time=40%), and presence of distended bowel loops.

The intestinal USFNB represented the 3.1% (81/2676) of all abdominal USFNB performed over a 15 years period; 27 were made on gastric wall, and 54 on gut wall (30at colon-sigma, and 24at the small bowel). The indications to the procedure were impossibility to perform endoscopy in 20 cases, non-diagnostic endoscopic biopsies in 47 cases, and endoscopically inaccessible lesion in 14 cases, respectively. The gold standard of comparisons were surgery or II level technique imaging. Sensitivity, specificity, diagnostic accuracy, predictive positive and negative values were 90%, 100%, 90%, 100%, and 87%, respectively. The procedure was well tolerated; no mortality has been reported although four serious complications were recorded: one case of small haemoperitoneum non requiring further treatment, one case of sepsis, one of small parietal haematoma and one self-controlled bile peritonitis.

Our study demonstrates that 18-G USFNB of gut wall is a simple, relatively safe, and accurate approach to obtain histologic sampling of gastrointestinal wall lesions, in case of endoscopic failure. By using a larger size needle an higher accuracy has been achieved at the expenses of some self-limiting complications.