The associations between platelet-to-lymphocyte ratio (PLR) and non-alcoholic fatty liver disease (NAFLD) and cirrhosis are unclear, and there are still no effective means for diagnosing or monitoring disease progression.
Data from the National Health and Nutrition Examination Surveys were collected for analysis. Logistic regression and restricted cubic splines were used to evaluate the associations between PLR and NAFLD and cirrhosis in different populations. The Area Under Curve Receiver Operating Characteristic (AUCROC) was used to distinguish the models. Threshold analysis was performed by constructing a two-piecewise linear regression. Correlation analysis was performed separately on either side of the inflection point.
A total of 5724 adults were included. Logistic regression analysis revealed that the PLR was associated with NAFLD and cirrhosis (AUCROC of NAFLD: 0.803; AUCROC of cirrhosis: 0.851). The AUCROC of the PLR for predicting NAFLD incidence was 0.762 in the diabetic population and 0.804 in the nondiabetic population. High PLR predicted cirrhosis in the diabetic population, with an AUCROC of 0.824, whereas a high PLR was not associated with cirrhosis in the nondiabetic population. The restricted cubic spline revealed a negative linear correlation between the PLR and NAFLD incidence. The inflection point of the PLR for NAFLD was 180.74. A PLR ≤180.74 was statistically significant (odds ratio=0.997, 95% confidence interval=0.995-0.999). In the NAFLD population, the PLR was negatively correlated with cirrhosis at a PLR ≤130.5 (odds ratio=0.987, 95% confidence interval=0.977-0.996) and positively correlated with cirrhosis at a PLR > 130.5 (odds ratio=1.006, 95% confidence interval=1.001-1.012).
The PLR and NAFLD were negatively correlated in the U.S. population. The PLR had a U-shaped relationship with cirrhosis in the NAFLD population. The PLR has potential value in monitoring NAFLD patient progression to cirrhosis.