Esteban P Dardanelli1, Maria Eugenia Orozco2, Juliana Lostra2, Clarisa Laprida2, Solange Lulkin2, Andrea P Bosaleh3, Carolina Cernadas4, José E Lipsich2. 1. Department of Radiology, Hospital de Pediatría J. P. Garrahan Buenos Aires, Combate de los Pozos 1881 (C 1245 AAM), Buenos Aires, Argentina. estebandardanelli@yahoo.ar. 2. Department of Radiology, Hospital de Pediatría J. P. Garrahan Buenos Aires, Combate de los Pozos 1881 (C 1245 AAM), Buenos Aires, Argentina. 3. Department of Pathology, Hospital de Pediatría J. P. Garrahan Buenos Aires, Buenos Aires, Argentina. 4. Department of Research and Development, Hospital de Pediatría J. P. Garrahan Buenos Aires, Buenos Aires, Argentina.
Abstract
BACKGROUND: A limited number of publications correlate bidimensional shear-wave elastography (2-D SWE) and stages of liver fibrosis in children. OBJECTIVE: To correlate liver elastography values using 2-D SWE and liver biopsy classified by Knodell-Ishak score to evaluate fibrosis in pediatric patients with liver disease, and to propose values of 2-D SWE (kPa) correlating with Knodell-Ishak score, which have not been defined in the literature. MATERIALS AND METHODS: We conducted a prospective cross-sectional observational study on the performance of diagnostic tests. Between June 2016 and June 2018, elastography was performed in 213 children and young adults who had undergone liver biopsy. B mode, Doppler and 2-D SWE were performed using an Aixplorer (SuperSonic Imagine, Aix-en-Provence, France). Histology samples were classified using the Knodell-Ishak score. We evaluated performance by assessing sensitivity, specificity, positive predictive value and negative predictive value. To determine cut-off points for the continuous variables, we used receiver operating characteristic (ROC) curves. All the cutoff values we established apply only to the SuperSonic Aixplorer system. RESULTS: Measurement with 2-D SWE was successful, with a good correlation with fibrosis stage. The area under the curve (AUC) to differentiate between early (Stages 1-2) and moderate (Stages 3-4) fibrosis was 0.91 (95% confidence interval [CI]: 0.87-0.96), with a sensitivity of 92% and specificity of 86%, with a cutoff value 12 kPa (2 m/s). The AUC of severe fibrosis (early stages of cirrhosis; Stage 5) was 0.95 (95% CI: 0.92-0.97), with a sensitivity of 94% and specificity of 90%, with a cutoff value 18.5 kPa (2.48 m/s). In two patients with hematopoietic stem cell transplantation and suspicion of graft versus host disease we found high 2-D SWE values in correlation with the fibrosis stages (Stage 0 with a median of 13 kPa [2.08 m/s] with hemosiderosis Grade 2 in one child and Stage 2 with a median of 46 kPa [3.91 m/s] and hemosiderosis Grade 4 in the other). CONCLUSION: Our study shows the usefulness and accuracy of 2-D SWE for detecting liver fibrosis in pediatric patients. We propose reference values for Knodell-Ishak Stages 1 and 5. We found hemosiderosis as a possible confounding factor that hasn't been described with 2-D SWE.
BACKGROUND: A limited number of publications correlate bidimensional shear-wave elastography (2-D SWE) and stages of liver fibrosis in children. OBJECTIVE: To correlate liver elastography values using 2-D SWE and liver biopsy classified by Knodell-Ishak score to evaluate fibrosis in pediatric patients with liver disease, and to propose values of 2-D SWE (kPa) correlating with Knodell-Ishak score, which have not been defined in the literature. MATERIALS AND METHODS: We conducted a prospective cross-sectional observational study on the performance of diagnostic tests. Between June 2016 and June 2018, elastography was performed in 213 children and young adults who had undergone liver biopsy. B mode, Doppler and 2-D SWE were performed using an Aixplorer (SuperSonic Imagine, Aix-en-Provence, France). Histology samples were classified using the Knodell-Ishak score. We evaluated performance by assessing sensitivity, specificity, positive predictive value and negative predictive value. To determine cut-off points for the continuous variables, we used receiver operating characteristic (ROC) curves. All the cutoff values we established apply only to the SuperSonic Aixplorer system. RESULTS: Measurement with 2-D SWE was successful, with a good correlation with fibrosis stage. The area under the curve (AUC) to differentiate between early (Stages 1-2) and moderate (Stages 3-4) fibrosis was 0.91 (95% confidence interval [CI]: 0.87-0.96), with a sensitivity of 92% and specificity of 86%, with a cutoff value 12 kPa (2 m/s). The AUC of severe fibrosis (early stages of cirrhosis; Stage 5) was 0.95 (95% CI: 0.92-0.97), with a sensitivity of 94% and specificity of 90%, with a cutoff value 18.5 kPa (2.48 m/s). In two patients with hematopoietic stem cell transplantation and suspicion of graft versus host disease we found high 2-D SWE values in correlation with the fibrosis stages (Stage 0 with a median of 13 kPa [2.08 m/s] with hemosiderosis Grade 2 in one child and Stage 2 with a median of 46 kPa [3.91 m/s] and hemosiderosis Grade 4 in the other). CONCLUSION: Our study shows the usefulness and accuracy of 2-D SWE for detecting liver fibrosis in pediatric patients. We propose reference values for Knodell-Ishak Stages 1 and 5. We found hemosiderosis as a possible confounding factor that hasn't been described with 2-D SWE.