Article Text

Download PDFPDF

Algorithm to rule out clinically significant portal hypertension combining Shear-wave elastography of liver and spleen: a prospective multicentre study
  1. Christian Jansen1,
  2. Christopher Bogs1,
  3. Wim Verlinden2,
  4. Maja Thiele3,
  5. Philipp Möller1,
  6. Jan Görtzen1,
  7. Jennifer Lehmann1,
  8. Michael Praktiknjo1,
  9. Johannes Chang1,
  10. Aleksander Krag3,
  11. Christian P Strassburg1,
  12. Sven Francque2,
  13. Jonel Trebicka1,3
  1. 1 Department of Internal Medicine I, University of Bonn, Bonn, Germany
  2. 2 Department of Gastroenterology and Hepatology, University Hospital Antwerp, Edegem, Belgium
  3. 3 Department of Gastroenterology and Hepatology, Odense University Hospital, Odense, Denmark
  1. Correspondence to Professor Jonel Trebicka, Department of Internal Medicine I, University of Bonn, Sigmund-Freud Str. 25, Bonn D-53105, Germany; jonel.trebicka{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Recently, the UK guidelines on variceal bleeding1 and other reports have introduced the role of elastography for the diagnosis of oesophageal varices (OVs).2 ,3 Development of OVs is likely when the hepatic venous pressure gradient is higher than 10 mm Hg, which defines clinically significant portal hypertension (CSPH). Baveno VI implemented transient elastography (TE) as a tool to rule in CSPH in viral aetiologies and to rule out varices (need of screening endoscopy for varices).4 Furthermore, CSPH has a strong prognostic value in patients with cirrhosis.

Recent studies introduced Shear-wave elastography of the liver (L-SWE) as a promising tool to diagnose portal hypertension. These studies find good diagnostic accuracy of L-SWE with specificity and sensitivity ranging around 80% and superior to TE. However, in more than 30% of the patients CSPH could not be ruled in or ruled out, since their SWE values were between the cut-offs.

This prospective multicentre study investigated the sequential use of liver and spleen stiffness measured by SWE to rule out CSPH, measured invasively within maximal 48 h. The study included 158 patients, of which 61.4% were men. Main aetiology was alcohol (56%), non-alcoholic steatohepatitis occurred in 17% and chronic viral hepatitis in 8%. Mean Child–Pugh score was 7 (A: 62.7%; B: 28.5%; C: 8.8%), while mean model for end-stage liver disease (MELD) score was 10 (6–28) points. Mild ascites was present in 25 patients and severe ascites in 43 patients. The patients gave their written consent for all procedures and the local ethics committee of the University of Bonn (Nr. 121/14) approved the study in accordance with the Declaration of Helsinki, as well as for Antwerp (15/13/135) and Odense (2599).

The study identified an algorithm using sequential SWE measurements of liver and spleen. Patients with L-SWE ≥16.0 kPa had a high probability of having CSPH, while, also in some patients with L-SWE <16 kPa, CSPH was present. In this latter collective of patients, SWE of the spleen (S-SWE) ≥26.6 kPa was further used to identify patients with CSPH. Indeed, from 109 patients receiving L- and S-SWE measurements in the same session, 76 patients had L-SWE ≥16.0 kPa. A total of 65 of these patients were correctly classified as having CSPH and 11 had no CSPH, therefore being misclassified. Totally, 33 patients had L-SWE <16.0 kPa. In these patients, S-SWE cut-off of 26.6 kPa was used to further classify them. By doing so, 32 patients, of which 26 did not have CSPH and 6 had CSPH, were correctly classified and only 1 patient with CSPH was misclassified using this method (figure 1). These two cut-offs combined had a sensitivity of 98.6% and a specificity of 70.3% (table 1).

Table 1

Diagnostic performance of the algorithm for the diagnosis of CSPH

Figure 1

Algorithm to detect patients with and those without CSPH. CSPH, clinically significant portal hypertension; L-SWE, liver Shear-wave elastography; S-SWE, spleen-Shear-wave elastography.

As presented above, this study offers an easy algorithm with high diagnostic accuracy to rule out CSPH by using sequential L-SWE and S-SWE. The presence of CSPH, as the prerequisite for varices, has prognostic value. Importantly, also the success of therapy for portal hypertension might be dependent on the presence of CSPH, as shown recently that non-selective β-blockers might have a better response in patients with CSPH compared with patients without CSPH. Therefore, the diagnosis of CSPH is important information for clinical decisions.

Interestingly, this algorithm including L-SWE and S-SWE was elaborated to rule out CSPH more accurately with a sensitivity of 98.6%, which clearly surpassed recent results for this purpose. The added value of S-SWE seems to be particularly high in patients with L-SWE <16.0 kPa. In these patients, S-SWE of ≥26.6 kPa is able to detect patients with CSPH with very high accuracy, while S-SWE below 26.6 kPa rules out CSPH with a very high probability.

In conclusion, this study demonstrates for the first time that S-SWE improves accuracy of L-SWE to assess CSPH in cirrhotic patients of different aetiologies.


View Abstract


  • CJ, CB, SF and JT have contributed equally.

  • Contributors The study was designed by JT. Data were obtained by CB, CJ, WV, MT, PM, JG, JL, MP, JC, AK, CPS SF and JT. The report was written by CB, CJ, AK, SF and JT. All authors approved the final version.

  • Funding This study was supported by grants from the Deutsche Forschungsgemeinschaft (SFB TRR57 to P18), the H. J. & W. Hector Stiftung, Ernst-Bertha-Grimmke-Stiftung (6/15) and European Union's Horizon 2020 research and innovation programme (No 668031).

  • Competing interests None declared.

  • Ethics approval Ethics committee of the University of Bonn (Nr. 121/14); ethics committee of the university of Antwerp (15/13/135); ethics committee of the university Odense (51872).

  • Provenance and peer review Not commissioned; internally peer reviewed.