Felix H Lui1, Patrick Ma1, Mohammad Helmy2, Ke-Qin Hu3. 1. Department of Gastroenterology/Hepatology, School of Medicine, University of California, Irvine, 101 the City Drive, Building 22C, Room 1503, Orange, CA, 92868, USA. 2. Department of Radiology, School of Medicine, University of California, Irvine, Orange, CA, USA. 3. Department of Gastroenterology/Hepatology, School of Medicine, University of California, Irvine, 101 the City Drive, Building 22C, Room 1503, Orange, CA, 92868, USA. Kqhu@hs.uci.edu.
Abstract
BACKGROUND: Splenomegaly measured by spleen length has been an imaging evidence for cirrhosis. However, data remains lacking on the value of other US findings for diagnosing cirrhosis. This study evaluated the value of spleen two-dimensional measurements (2D, i.e., length × thickness) in diagnosing cirrhosis by comparing with other US parameters. METHODS: A retrospective study on 297 cohort 1 patients with clinical/imaging diagnosis of cirrhosis was conducted. Spleen length and thickness were measured via US imaging and compared with other US parameters using statistical analysis to assess their value in diagnosing cirrhosis. A separate 161 cohort 2 patients with histological fibrosis staging was used to validate the findings from the cohort 1. RESULTS: Using 297 cohort 1 patients, US findings of spleen length > 12 cm (50.6% vs. 9.6%, p < 0.001); spleen thickness > 4 cm (78.2% vs. 21%, p < 0.001); and spleen 2D > 46 cm2 (81.6% vs. 15.3%, p < 0.001) were significantly associated with, but only spleen 2D > 46 cm2 (95% CI 7.9-92.8, p < 0.001) was independently associated with clinical/imaging evidence of cirrhosis on multivariate analysis. We further analyzed 161 cohort 2 patients and validated that US finding of spleen 2D > 46 cm2 carried the best sensitivity and specificity (93.5% and 95.3%) and was the only US parameter independently associated with histological stage 3-4 fibrosis, i.e., cirrhosis (95% CI 3.1-87, p = 0.006). CONCLUSION: Using both testing and validation cohorts, we demonstrated that spleen 2D > 46 cm2 carries 93.5% sensitivity and 95.3% specificity and is superior to other US parameters in diagnosing cirrhosis.
BACKGROUND:Splenomegaly measured by spleen length has been an imaging evidence for cirrhosis. However, data remains lacking on the value of other US findings for diagnosing cirrhosis. This study evaluated the value of spleen two-dimensional measurements (2D, i.e., length × thickness) in diagnosing cirrhosis by comparing with other US parameters. METHODS: A retrospective study on 297 cohort 1 patients with clinical/imaging diagnosis of cirrhosis was conducted. Spleen length and thickness were measured via US imaging and compared with other US parameters using statistical analysis to assess their value in diagnosing cirrhosis. A separate 161 cohort 2 patients with histological fibrosis staging was used to validate the findings from the cohort 1. RESULTS: Using 297 cohort 1 patients, US findings of spleen length > 12 cm (50.6% vs. 9.6%, p < 0.001); spleen thickness > 4 cm (78.2% vs. 21%, p < 0.001); and spleen 2D > 46 cm2 (81.6% vs. 15.3%, p < 0.001) were significantly associated with, but only spleen 2D > 46 cm2 (95% CI 7.9-92.8, p < 0.001) was independently associated with clinical/imaging evidence of cirrhosis on multivariate analysis. We further analyzed 161 cohort 2 patients and validated that US finding of spleen 2D > 46 cm2 carried the best sensitivity and specificity (93.5% and 95.3%) and was the only US parameter independently associated with histological stage 3-4 fibrosis, i.e., cirrhosis (95% CI 3.1-87, p = 0.006). CONCLUSION: Using both testing and validation cohorts, we demonstrated that spleen 2D > 46 cm2 carries 93.5% sensitivity and 95.3% specificity and is superior to other US parameters in diagnosing cirrhosis.