NashTest: A New Biomarker for Non-Alcoholic Steatohepatitis
Author Information
Author(s): Poynard Thierry, Ratziu Vlad, Charlotte Frederic, Messous Djamila, Munteanu Mona, Imbert-Bismut Françoise, Massard Julien, Bonyhay Luninita, Tahiri Mohamed, Thabut Dominique, Cadranel Jean François, Le Bail Brigitte, de Ledinghen Victor
Primary Institution: AP-HP Groupe Hospitalier Pitié-Salpêtrière, Paris, France
Hypothesis
Can the NashTest accurately predict non-alcoholic steatohepatitis (NASH) in patients with non-alcoholic fatty liver disease (NAFLD)?
Conclusion
The NashTest is a reliable, non-invasive biomarker for predicting the presence or absence of NASH in patients with NAFLD.
Supporting Evidence
- NashTest showed an AUROC of 0.79 for diagnosing NASH.
- The specificity of NashTest for NASH was 94%.
- NashTest had a sensitivity of 33% for detecting NASH.
- The study included a total of 640 participants across training and validation groups.
- Biopsy results were used as the gold standard for comparison.
- Patients with NASH had higher levels of certain biochemical markers.
- The study demonstrated that NashTest can reduce the need for invasive liver biopsies.
- Discordance rates between NashTest and biopsy were low.
Takeaway
Doctors can use a simple blood test called NashTest to check if someone has a liver problem called NASH without needing a painful liver biopsy.
Methodology
The study included 160 patients in a training group and 97 in a validation group, using liver biopsy as the gold standard for diagnosis.
Potential Biases
There is a risk of bias due to the subjective nature of histological assessments and the variability in biopsy quality.
Limitations
The study's reliance on liver biopsy introduces variability and potential bias in diagnosing NASH.
Participant Demographics
The study included adults with suspected NAFLD, with a mean age of approximately 52 years, and a mix of genders.
Statistical Information
P-Value
0.94
Confidence Interval
95% CI 0.67–0.87
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website