Utility of EUS-elastography to Predict Portal Hypertension

Overview

Patients with cirrhosis have structural and functional alterations of the liver. The progressive deposition of hepatic fibrosis is related to the subsequent development of portal hypertension (PH), and PH is associated with mayor complications including ascites, hepatic encephalopathy and development of gastroesophageal varices with a high risk of bleeding. Variceal bleeding is a medical emergency associated with a 6-week mortality rate of approximately 10-20%. Liver biopsy is the gold standard for the assessment of hepatic fibrosis, whereas the measurement of hepatic vein pressure gradient (HVPG) is the standard to evaluate PH and upper endoscopy (UE) is the method of choice to detect the presence and grade of gastroesophageal varices. The last two also estimates the risk of variceal bleeding. Unfortunately, clinical investigation of PH implies HVPG measurement or endoscopy for esophageal varices (EV) screening and grading. The first one is an invasive technique, mainly restricted to tertiary centers, that requires personal training, increased health care costs and patient discomfort. The UE, even though has demonstrated utility to predict HVPG (HVPG value ≥ 10 mmHg predicts the presence of EV and a value ≥ 12 mmHg is predictive for variceal bleeding), has been criticized of being subjective. Because of this, alternative test including elastographic techniques, have been develop to assess the severity of PH, the presence of EVs and the risk of variceal bleeding. Elastography is a technique used to measure tissue elasticity and stiffness in real time, by the application of slight compression using a transducer to the targeted tissue. The principle is that tissue compression produces deformation (strain) and that the strain is smaller in harder tissue as compared to softer tissue. Consequently, by measuring the tissue strain induced by compression, it is possible to estimate the tissue hardness. Fibroscan® (FS) (Echosens, París, Francia) uses the principle of one-dimension transient elastography (TE) for the assessment of tissue stiffness. It was used initially for liver stiffness measurement (LSM) and proved to be reliable for the diagnosis of liver cirrhosis and avoid liver biopsy in 90% of cases. Also LSM by TE accurately correlates with the severity of PH and the presence of esophageal varices.

More recently, the use of Fibroscan for the spleen stiffness measurement (SSM) has become particularly attractive to assess. Previous studies have demonstrated that liver cirrhosis and PH generates some modifications in the spleen such as an increment in size, splenic blood flow, tissue hyperplasia, and fibrosis that determine an increase in the spleen's density and tissue stiffness that may be quantified by elastography. The SSM was found to be a valuable tool for assessing the degree of PH, the presence and severity of EVs and the risk of variceal bleeding among patients with cirrhosis. Moreover, spleen stiffness as compared with liver stiffness better represents the dynamic changes occurring in the advanced stages of cirrhosis and shows higher diagnostic performance in detecting esophageal varices. Elastography can also be applied by endoscopic ultrasound (EUS-E). The EUS elastography allowed qualitative and quantitative evaluation. Qualitative elastography evaluates tissue elasticity by measuring the degree of deformation using a scale in the B-mode image of 1 to 255. This scale is represented by a color map (red- green-blue), wherein hard tissue is shown in dark blue, tissue with intermediate hardness in green, medium soft tissue in yellow, and soft tissue in red. There are two options for quantitative elastography and both are based on the qualitative EUS elastography data, the hue histogram and strain ratio. The hue histogram represents the overall elasticity within a manually selected area witch is the region of interest (ROI). The global hardness is being represent by a hue scale from 0 to 250, where 0 represents the hardest and 255 is the softest. The strain ratio (SR) analyzes the elastographic picture of the target lesion in relation to the surrounding tissues. Two different areas (A and B) are selected. Area A includes as much of the target lesion as possible without including the surrounding tissues. Area B is selected within a soft (red) reference area outside the target lesion, preferably the gut wall. The strain ratio is calculated as the quotient of B/A. EUS is also a technique that has proved to be useful to evaluate cirrhotic patients in many aspects. It allows visualization of a greater part of the portal and portosystemic collateral circulation. It can be used to detect esophageal and gastric varices witch correlate with the presence and severity of liver disease and portal hypertension. It can be use to measure varices size. Large varices (≥ 5 mm) may be risk factors for variceal hemorrhage. It can visualize paramural collaterals and large vessels such as splenic, portal and azygos veins. Previous studies have found a significant increase in azygos vein diameter and blood-flow velocity in patients with varices when compared with control subjects. The evaluation of the azygos vein using EUS can ad useful information about the porto-systemic collateral circulation and could help to determine the treatment efficacy. It has been found a decrease in the azygos vein flow following terlipressin and somatostatin administration, band ligation or sclerotherapy in patients with portal hypertension. EUS can predict variceal recurrence after endoscopic therapy. Finally, cirrhosis causes dilation thoracic duct and thickening of gastric mucosa and submucosa that can be determinate by EUS. MATERIAL AND METHODS Study design: This is a prospective, observational, non-randomized and simple blind study, performed in a Tertiary Academic Center. Setting: Ecuadorian Institute of Digestive Disease (IECED), Omni Hospital Academic Tertiary Center. Patients will be recruited from the gastroenterology unit (IECED), from January 2017 to June 2017. The study protocol and consent form has been approved by the Institutional Review Board and will be conducted according to the declaration of Helsinki. Written informed consent will be obtained from all participants. Portal Hypertension determination: The HVPG will be determinate indirectly by upper endoscopy and clinical examination. A HVPG ≥ 10 mm Hg defines a clinically significant PH (CSPH) and will be determinate by the presence of EV on UE. On the other hand, a HVPG ≥ 12 mm Hg indicates a severe PH and will be determinate by the presence variceal bleeding on UE or ascites and hepatic encephalopathy. The EUS procedure: The procedure will be performed under general anesthesia, with tracheal intubation, with the patient in supine position and in an endoscopy suite. All patients will be initially evaluated by upper endoscopy using a standard gastroscopy in order to avoid any contraindication and for gastro and esophageal varices assessment. Esophageal varices will be graded according to the Baveno VI consensus conference as follows: grade I, varices flattened by insufflation; grade II, varices non-confluent and protruding in the lumen despite insufflation; grade III, confluent varices not flattened by insufflation. Presence of red signs, fibrin plugs on the varices, and blood in the upper gastrointestinal tract will be also recorded. Gastric varices will be classified according to the classification of Sarin and Kumar. The EUS exam will be performed using a 3.8 mm working channel linear-array echoendoscopes (EG 3870UTK; Pentax, Hamburg, Germany), attached to an US console (Avius Hitachi, Tokyo, Japan). Two endoscopists (C.R.M, M.V) with experience in EUS-elastography will perform the procedures. The exam will start by evaluating the left hepatic lobe (the closer hepatic segment to the transducer) transgastrically using the conventional B-mode. Then the elastography software will be activated and the elastographic and B-mode images will be displayed simultaneously side by side, performing a real time qualitative evaluation of the hepatic tissue elasticity. The free movement of the endoscope tip, as well as sporadic motion of the tissues induced by respiratory or heart movements, determines lateral slippage of the target tissue. In order to avoid that, once the scope is located it will be fixed using the brakes. Images obtained with the application of strong pressure lead to wrong measures, so light pressure will be applied. The measurement will be considered adequate when the gastric mucosa is shown red. Once the transducer is correctly located the quantitative evaluation will be performed. For the SR calculation, the area A will be manually selected including as much of hepatic tissue as possible and the area B will be selected on the red gastric mucosa. For the strain histogram measurement, the ROI selected will have a surface of 60 mm2. Both values will be photographically recorded. The same procedure will be repeated 10 times at different points on the left hepatic lobe and finally the mean SR and SH values will be calculated. The procedure will continue by measuring the spleen stiffness. The transducer will be placed as close as possible to the spleen. Patients in whom the distance between the transducer and the spleen is more than 1.5 cm will be excluded from the analysis, but included in the intention to treat. The elastography software will be activated and the previously described sequence will be repeated. Ten measurements will be performed and the mean SR and SH values will be calculated. Finally the azygos vein (AV) will be scanned in all patients and evaluated using EUS Doppler. It will be scanned at a distance of 20-30 cm from the incisors with the EUS probe placed opposite to the posterior wall of the esophagus and to the right of the vertebral column. After the AV is found it will be followed until its arch. Measurements of the AV will be performed at a point just caudal to its arch, which drains into the superior vena cava. The angle of insonation will be less than 60º. The mean velocity (V mean cm/s) and the AV diameter (D) will be measure and the AV blood flow volume index (BFVI) will be calculated \[BFVI (cm3 /s) = Vmean (cm/s) X D2 (cm2)\]. The BFVI has been used before and it has been found to correlate with the blood-flow volume. The Fibroscan procedure: In all patients, the liver stiffness measurements (LSM) will be assesses by transient elastography (TE) using the FibroScan® (FS) (Echosens, Paris, France). Spleen stiffness (SSM) measurement requires special software not approved by the National Administration of Food, Drugs and Technology. For the LSM, a previously described technique will be used. The patients will be in the supine position with arms in maximal abduction. The TE probe will be applied on the right lobe of the liver, via the intercostal spaces, in the right axillary midline and perpendicular to the plane of the skin. The results will be expressed in kilopascals (kPa). At least 10 LSM values will be carried out on each patient. The FS screen will show 3 parameters that have to be taken into account: the success rate calculated as the number of valid measurements divided by the number of total measurements, the median value calculated as average value of the 10 successful measurements and the interquartile range (IQR) calculated as the deviation of the total of valid measurements with respect to the median value. According to the manufacturer's recommendations and previous evidence, an interquartile range (IQR) ≥ 30% of the median value and success rate ≤ 60% will be considered incorrect and these patients will be excluded from the analysis but included in the intention to treat. LSM will be performed by two operators (H.P.L, J.O.A.), with more than 200 procedures performed each, blindly to the EUS-E results. Statistical analysis: Baseline characteristics as age, gender will be compared between case and control group. Categorical variables will be assessed using Chi-square o Fisher Test and continuing variables with Student's t-test or Mann-Whitney Test. The relationships between the parameters will be characterized using the Spearman correlation coefficients and Pearson's correlation coefficient. The liver and spleen stiffness elastographic data will be expressed in kPa. All continuous variables will be expressed in mean ± standard deviation or median (range) as appropriate. The diagnostic performance of LSM and SSM by EUS elastography will be assessed using sensitivity (Se), specificity (Sp), positive predictive value (PPV), negative predictive value (NPV), accuracy, likelihood ratio (LR), Odds ratios (OR) with a 95% confidence intervals (CI) and receiver-operating characteristic (ROC) curves. Sensitivity analyses of patients under NSBB treatment prior elastography examination will be perform separately. Values of P \< 0.05 will be considered significant. All the statistical analysis will be performed using SPSS software suite v.22. (Chicago, IL, USA). Limitations: Since the examiner will have to choose the best EUS-E images from a dynamic sequence, a high examiner bias could be certainly favored. Selection of the ROI will be manually. In case of SH measure, the ROI will have in every case a 60 surface, but for SR measurement of the area A will have to include as tissue as possible with the possibility of differences between measurements. There are factors that have been described, especially with the use of the strain ratio that may influence the results of the elastography calculations. Examples are the presence of fluid (ascites, cysts, vessels) between the transducer and the ROI, the degree of compression by the transducer, the ROI diameter, the distance with the transducer, respiratory and cardiac movements, and the operator's experience. An attempt to reduce this factors will be apply, however they may not be completely eliminated in all the cases. The presence of ascites represents a more advanced stage of liver cirrhosis and a high risk of EV. However, the ascites limits the stiffness measurement by elastography and is considered and exclusion criteria. Therefore, most of the patients will be classified as Child-Pugh score A.