RESUMO
BACKGROUND: Noninvasive tests (NITs) have been proposed as an alternative to liver biopsy for diagnosing liver cirrhosis. The evidence of NIT performance in patients with chronic hepatitis D (CHD) is limited. AIMS: To evaluate the diagnostic performance of liver stiffness measurement (LSM) and other NITs in CHD patients. METHODS: We evaluated the diagnostic performance of LSM by transient elastography for the detection of liver cirrhosis in a retrospective, multicentre cohort of 144 CHD patients with paired (±6 months) LSM and liver biopsies. RESULTS: Cirrhosis was diagnosed histologically in 22 patients (15.3%). Mean LSM was significantly higher in patients with cirrhosis compared to those without fibrosis (23.4 vs 10.2 kPa, p < 0.0001) or with intermediate fibrosis (23.4 vs 13.5 kPa, p < 0.0001). In the detection of liver cirrhosis, LSM was superior to other NITs (AUROCs: 0.89 [LSM], 0.87 [D4FS], 0.74 [APRI], 0.73 [FIB-4], and 0.69 [AAR]). The optimal cut-off for identifying patients with liver cirrhosis was ≥15.2 kPa (Se 91%, Sp 84%, PPV 50%, NPV 98%). The ideal cut-off for diagnosing non-advanced liver fibrosis (Metavir ≤2) was <10.2 kPa (Se 55%, Sp 86%, PPV 90%, NPV 45%), correctly identifying 90% of patients. Data were validated in an independent cohort of 132 CHD patients. CONCLUSIONS: LSM is a useful tool for identifying patients at risk for liver cirrhosis and is superior to other NITs. The cut-offs of <10.2 and < 15.2 kPa reliably diagnose non-advanced liver fibrosis and exclude cirrhosis in the majority of patients. However, LSM cannot completely replace liver biopsy in CHD patients.