Ultrasound Pressure Estimation for Diagnosing Portal Hypertension in Patients Undergoing Dialysis for Chronic Kidney Disease. (1st December 2021)

• Record Type: Journal Article
• Authors: Machado, Priscilla; Gupta, Ipshita; Fenkel, Jonathan M.; Gummadi, Sriharsha; Stanczak, Maria; Wessner, Corinne E.; Shaw, Colette M.; Schultz, Susan; Soulen, Michael C.; Wallace, Kirk; Eisenbrey, John R.; Forsberg, Flemming
• Is Part Of: Journal of ultrasound in medicine. Volume 41:Number 9(2022)
• Page Start: 2181
• Page End: 2189
• Publication Date: 2021-12-01
• Subjects: chronic kidney disease -- contrast‐enhanced ultrasound -- dialysis -- portal hypertension -- subharmonic imaging -- subharmonic pressure estimation; Ultrasonics in medicine -- Periodicals; Ultrasonics; Ultrasonography; Ultrasonics in medicine; Electronic journals; Periodicals; Periodicals; 616.07543
• DOI: 10.1002/jum.15897
• Languages: English
• ISSNs: 0278-4297

Abstract

Abstract : Objectives: Hepatic venous pressure gradient (HVPG) is considered the standard in quantifying portal hypertension, but can be unreliable in dialysis patients. A noninvasive ultrasound technique, subharmonic‐aided pressure estimation (SHAPE), may be a valuable surrogate of these pressure estimates. This study compared SHAPE and HVPG with pathology findings for fibrosis in dialysis patients. Methods: This was a subgroup study from an IRB‐approved trial that included 20 patients on dialysis undergoing SHAPE examinations of portal and hepatic veins using a modified Logiq 9 scanner (GE, Waukesha, WI), during infusion of Sonazoid (GE Healthcare, Oslo, Norway). SHAPE was compared to HVPG and pathology findings using the Ludwig‐Batts scoring system for fibrosis. Logistic regression, ROC analysis, and t ‐tests were used to compare HVPG and SHAPE with pathological findings of fibrosis. Results: Of 20 cases, 5 had HVPG values corresponding to subclinical and clinical portal hypertension (≥6 and ≥10 mmHg, respectively) while 15 had normal HVPG values (≤5 mmHg). SHAPE and HVPG correlated moderately ( r = 0.45; P = .047). SHAPE showed a trend toward correlating with fibrosis ( r = 0.42; P = .068), while HVPG did not ( r = 0.18; P = .45). SHAPE could differentiate between mild (stage 0–1) and moderate to severe (stage 2–4) fibrosis (−10.4 ± 4.9 dB versus −5.4 ± 3.2 dB; P = .035), HVPG could not (3.0 ± 0.6 mmHg versus 4.8 ± 0.7 mmHg; P = .30). ROC curves showed a diagnostic accuracy for SHAPE of 80%, while HVPG reached 76%. Conclusion: Liver fibrosis staging in dialysis patients evaluated for portal hypertension appears to be more accurately predicted by SHAPE than by HVPG; albeit in a small sample size.