Background: Although endoscopic ultrasonography (EUS) is the best modality in the diagnosis of gastric submucosal tumours (G-SMT), it is still invasive and expensive. We evaluated the usefulness of transabdominal ultrasonography (US) as an alternative to EUS in the assessment and clinical management of G-SMT.
Methods: This is a prospective study of 156 successive patients who had been diagnosed with G-SMT by direct endoscopic visualization. For each patient, US was performed prior to EUS by an examiner who had not been informed about the site or the size of the patient's G-SMT beforehand. US diagnoses were compared with those ascertained by EUS. Diagnostic grade by US was divided into three groups: diagnosed (G-SMT was clearly demonstrated including location, site and type), detected (G-SMT was demonstrable but the type was not clear) and undetected (G-SMT was not demonstrated by US).
Results: We found that US can be an alternative to EUS for: 1) diagnosis of extramural compression, 2) diagnosis of G-SMT with size >30 mm in diameter, 3) detection and measurement of the size of a G-SMT from 21 to 30 mm in diameter, and 4) detection and measurement of the size of a G-SMT at both the gastric angle and cardia, regardless of the size of the tumour. Overall sensitivity and specificity for the detection of G-SMT were 82.5% and 100%, respectively. The diagnostic rates of G-SMT in each group were 60.1% (86/143) for diagnosed, 22.4% (32/143) for detected and 17.5% (25/143) for undetected. Approximate 95% (21/22) of G-SMT over 20 mm in diameter were at least detected, and 97% (30/31) of G-SMT over 30 mm in diameter were diagnosed by US.
Conclusion: US can be an alternative method in the assessment of G-SMT, especially in the follow-up of patients already diagnosed.