This poster is published under an
open license. Please read the
disclaimer for further details.
Keywords:
Abdomen, Liver, Ultrasound, Diagnostic procedure, Haematologic diseases
Authors:
L. Cevasco, F. Paparo, M. Revelli, C. Puppo, E. Aleo, L. Bacigalupo, G. Forni, G. A. Rollandi; Genova/IT
DOI:
10.1594/ecr2014/B-0413
Results
Sixty-seven patients met the inclusion criteria.
Seven patients were excluded: both TE and RTE were unsuccessful because of narrow intercostal spaces in 2 cases,
and BMI > 27.5 kg/m2in other 3 cases; in the remaining 2 patients TE measurements had an IQR > 30%.
The resulting 60 patients were 34 males (56.7%),
26 females (43.3%) with a median age of 42 (21–76) years and a mean BMI of 23.83±3.67.
They included 37 adult homozygous-β-thalassemic patients (thalassemia major),
13 patients withβ-thalassemia intermedia,
6 patients with primary (genetic or type 1) hemochromatosis,
and 4 patients with myelodysplastic syndrome.
The overall prevalence of hepatitis C infection in our patient’s cohort was 18/60 (30%).
Mean serum ferritin level was 1690 (890–5242) ng/mL.
Patient’s characteristics are reported in Table 1.
T2* values did not follow the normal distribution,
while both TE values and RTE elastic ratios followed the normal distribution.
The mean TE value of each patient was converted in its corresponding METAVIR fibrosis stage,
thus resulting in 28 (46.6%) F0/1,
12 (20%) F2,
12 (20%) F3 and 8 (13.4%) F4 patients (Table 2).
Neither TE measurements (r=−0.0820,
95%CI−0.329–0.176,P= 0.533),
nor RTE elastic ratiosr=−0.170,
95%CI−0.406–0.0877,P= 0.1942) were significantly associated to MRI T2* values of liver iron overload.
A significant increase in elastic ratios was observed with increasing stiffness values measured by TE (r=−0.645,
95% CI 0.468–0.772,P< 0.0001).
The mean elastic ratios for each METAVIR group were as follows: F0/1 = 1.9±0.4; F2 = 2.2±0.4; F3 = 2.9±0.5; F4 = 3.2±0.4.
The one-way analysis of variance (ANOVA) test demonstrated that mean elastic ratios of F0/1 and F2 METAVIR groups were significantly different from those of F3 and F4 groups (P< 0.05).
Box-and-Whisker plots for measurements of elastic ratios for each fibrosis stage are shown in Fig.
3.
The diagnostic accuracy of RTE forF≥2 evaluated by AUC-ROC analysis was 0.798 (95%CI 0.674–0.890).
At a cut-off of elastic ratio≥2.01,
RTE showed a sensitivity of 80% (95% CI 64.4–90.9),
a specificity of 75% (95%CI 50.9–91.3),
a positive predictive value (PPV) of 86.5% (95% CI71.2–95.5) and a negative predictive value (NPV) of 65.2% (95% CI42.2–84)(Fig.
4A).
The diagnostic accuracy of RTE forF≥3 eval-uated by AUC-ROC analysis was 0.909 (95% CI 0.806–0.968).
At a cut-off of elastic ratio≥2.75,
RTE showed a sensitivity of 70% (95%CI 45.7–88.1),
a specificity of 97.5% (95% CI 86.8–99.9),
a PPV of93.3% (95% CI 66.8–99.9) and a NPV of 86.7% (95% CI 73.2–94.9)(Fig.
4B).
Despite the small sample size of F4 patients,
the diagnostic accuracy of RTE forF= 4 evaluated by AUC-ROC analysis was 0.906 (95% CI 0.803–0.966).
The best elastic ratio cut-off for detectingF= 4 was the same ofF≥3 (i.e.≥2.75).
Using this cut-off value for detecting F4 fibrosis stage,
RTE showed a sensitivity of 87.5% (95% CI 47.3–99.7),
a specificity of 84.62% (95% CI 71.9–93.1),
a PPV of 46.7% (95% CI 21.3–73.4) and a NPV of 97.8% (95% CI 88.2–99.9).