Diagnostic Accuracy of Splenic Stiffness Measurement for Gastro-Oesophageal Varices
No Thumbnail Available
Authors
Mootien A.
Yogasivam D.
Sharma V.
Chen F.
MacCallum, P.
Check for full-text access
Issue Date
2025
Type
Article
Language
Keywords
Alternative Title
Abstract
Portal hypertension, a life-threatening complication of splanchnic vein thrombosis (SVT), frequently leads to high-risk gastro- oesophageal varices (GOV). Current reliance on invasive gastroscopy for surveillance burdens both patients and healthcare, highlighting an urgent need for reliable, non-invasive screening. While liver elastography is established, it may not adequately assess portal hypertension in complex SVT, suggesting splenic stiffness as a valuable complementary tool. Splenic stiffness measurement (SSM) shows promise as a non-invasive method for assessing portal hypertension and its sequelae. This multi-centre diagnostic accuracy study evaluated SSM's utility as a non-invasive predictor for GOV in 52 patients with confirmed SVT (32 males, 20 females; mean age 55 years). The cohort included 18 cirrhotic and 34 non-cirrhotic patients. Each patient underwent splenic stiffness assessment using a FibroScan with a loaned novel Echosens spleen probe (100 Hz), measuring values up to 100 kPa. Median liver stiffness was also measured. All patients underwent gastroscopy by experienced endoscopists, with abnormal gastroscopy (presence of gastro-oesophageal varices or isolated gastric varices) serving as the gold standard. A splenic stiffness cutoff of >27 kPa was applied. Diagnostic accuracy metrics (sensitivity, specificity, PPV, NPV) were calculated. SSM demonstrated a sensitivity of 80% and a specificity of 40.6% for identifying GOV, with a PPV of 55.2% and an NPV of 82.6%. For patients with splenomegaly (n=24), sensitivity and PPV were notably higher (85.7% and 66.7%), while NPV was lower (66.7%). Overall median liver stiffness was 11.9 kPa but varied significantly across SVT aetiologies in patients with abnormal gastroscopies (e.g., 7.7 kPa to 35.3 kPa). These results indicate that while SSM effectively identifies most patients with varices, its ability to confidently rule in the condition with a positive result is moderate. Conversely, its high overall NPV strongly correlates with the true absence of varices, offering substantial reassurance. Splenic stiffness measurement is a valuable screening tool for non-invasive GOV prediction in SVT. It shows good sensitivity and a strong negative predictive value, with limited specificity and variably moderate to higher positive predictive value. Guiding gastroscopy decisions with SSM requires nuance. In splenomegaly, higher PPV means positive results are more likely true varices, warranting careful consideration. For patients without splenomegaly, SSM offers a pragmatic approach to safely identify those unlikely to harbour varices, streamlining care and reducing unnecessary gastroscopies. Given liver stiffness limitations, this study supports SSM integration into SVT diagnostic algorithms, advocating for its role in optimising patient management and resource utilisation.
Description
Citation
Publisher
License
Journal
Gut