Splenic stiffness measurement by transient elastography for predicting varices in children with portal hypertension

Objective To evaluate the diagnostic accuracy of splenic stiffness measurement (SSM) using spleen‐dedicated vibration‐controlled transient elastography (VCTE) for predicting varices and varices needing treatment (VNT) in children with clinically evident portal hypertension (CEPH). Methods In this single‐centre prospective cross‐sectional study, children aged 6 months to 14 years with portal hypertension meeting the criteria for CEPH were eligible. SSM was measured using the VCTE with a spleen‐specific (100 Hz) probe. Esophagogastroduodenoscopy (EGD) was performed to detect varices and risk stratification. Diagnostic performance was assessed using receiver operating characteristic analysis and contingency tables. Results Ninety‐three consecutive children with CEPH underwent EGD, and successful SSM acquisition was achieved in 90 cases. The median (interquartile range) age of presentation was 6 (3–10) years. EGD identified varices in 79 (87.8%) and VNT in 57 (63.3%). Children with varices had higher median SSM values than those without (37.9 vs. 17.9 kPa; p < 0.001). Among those with varices, SSM values were higher in high‐risk than in low‐risk cases (42.9 vs. 21.9 kPa; p < 0.001). SSM demonstrated good accuracy for predicting varices (area under the curve: 0.86, 95% confidence interval: 0.76–0.97) with a sensitivity of 75.9% (65.0%–84.9%) and specificity of 90.9% (58.7%–99.8%) at a threshold of 31.1 kPa. For VNT, accuracy was 0.838 (0.702–0.877), with a sensitivity of 87.7% (76.3%–94.9%) and specificity of 72.7% (54.5%–86.7%) at the SSM cut‐off of 31.4 kPa. Conclusion In children with CEPH, spleen‐dedicated VCTE estimated SSM to have a reasonably high accuracy in predicting varices and VNT. Integrating SSM into clinical practice could assist in the screening and management of portal hypertension in children.