The main findings of this longitudinal cohort study based on the non-obese Chinese population were as follows: (1) GGT/HDL-C ratio was positively correlated with the NAFLD risk among Chinese non-obese people, and the association was stronger in those with elevated SBP. (2) GGT/HDL-C ratio had a nonlinear relationship with NAFLD, and there was a saturation effect. (3) GGT/HDL-C ratio and BMI were comparable in identifying NAFLD and were superior to GGT and HDL-C alone; however, GGT/HDL-C ratio did not appear to have a significant advantage in predicting future NAFLD.
With the change in diet and lifestyle in recent decades, the prevalence of non-obese NAFLD is increasing all over the world [11,12,13, 28], and research on non-obese NAFLD is also increasing [13, 29]. According to some previous research evidence, non-obese patients with NAFLD tended to exhibit worse metabolic status than obese patients with NAFLD, and they may develop more severe non-alcoholic steatohepatitis and advanced fibrosis [14, 30, 31]. Moreover, related studies have also shown that non-obese NAFLD had a higher risk of developing metabolic syndrome and diabetes [12, 32], and a higher risk of all-cause mortality . These findings all suggested the specificity of non-obese NAFLD and that this population needs further attention.
GGT is the most commonly used clinical biochemical index to evaluate liver function, and in addition to identifying the risk of NAFLD prevalence [15,16,17], it is also closely associated with NAFLD-related complications . HDL is a multifunctional structural protein, and HDL functionality represents several performance metrics of HDL, such as antioxidant, anti-inflammatory, and cholesterol efflux activities [35, 36]. Previous studies have shown that the emergence of dysfunctional HDL is usually associated with many acute infectious diseases and chronic aging-related diseases. HDL can be an appropriate biomarker for the diagnosis and progression of many diseases by monitoring the changes in its quantity and quality in terms of antioxidant and anti-inflammatory capacity . It is worth noting that in several recent studies on the relationship between HDL function and NAFLD, researchers have shown that the imbalance of cholesterol efflux activity in HDL function may be the main cause of metabolic NAFLD [37, 38]. The quantity of HDL is expressed as serum HDL-C concentration, and low concentration of HDL-C is a common clinical lipid metabolism disorder [36, 39], which is closely related to the progressive course of NAFLD [20,21,22, 40]. In the current study, after dividing and combining these two indicators, we found that the ratio of GGT/HDL-C was positively related to NAFLD risk in the Chinese non-obese population, and the positive association between them existed stably after further variable adjustment, and the extent of the association changed only slightly. Studies on the association of NAFLD with the GGT/HDL-C ratio have also been described previously. As early as May 2020, Feng et al. first revealed a positive relationship of NAFLD with the GGT/HDL-C ratio through a cross-sectional study of 6,326 general population ; then in a recent study by Xing et al. of 1,434 inpatients with diabetes, they found that the GGT/HDL-C ratio was an independent risk factor for metabolic-associated fatty liver disease in people with diabetes with a BMI ≥ 23 kg/m2 . Although both studies found a relationship of NAFLD with the GGT/HDL-C ratio, it is unclear whether the relationship of the GGT/HDL-C ratio with NAFLD changes over time and whether there is a nonlinear association between the two due to the cross-sectional design. Therefore, we conducted the current study, and our results further revealed the longitudinal correlation between GGT/HDL-C ratio and NAFLD. Additionally, we further evaluated baseline GGT/HDL-C ratio, GGT, HDL-C, and BMI for correlation with future NAFLD and used standardized HR values to present all results; the study found that GGT/HDL-C ratio may be slightly better than GGT, and BMI was better than GGT/HDL-C ratio in assessing future NAFLD risk.
The diagnostic value of GGT, HDL-C and GGT/HDL-C ratio for NAFLD was previously described by Feng et al. Through ROC analysis, they found that the diagnostic value of GGT combined with HDL-C for NAFLD was significantly higher than that of GGT and HDL-C alone . In order to verify their findings, we have carried out the same analysis in the current cross-sectional data sets, and the results were consistent with the findings of Feng et al. We found that the ratio of GGT/HDL-C was similar to BMI but superior to GGT and HDL-C alone in the diagnosis of NAFLD. In addition, in the current study, we further carried out time-dependent ROC analysis based on longitudinal data. The results show that with the passage of time, although the ratio of GGT/HDL-C seemed to be increasingly valuable in predicting NAFLD and was better than HDL-C in the early stage, it was not superior to BMI and GGT. Taken together, the GGT/HDL-C ratio seemed to be more suitable for epidemiological screening of NAFLD than for prediction of NAFLD, and further studies are still needed to validate the findings based on longitudinal data.
In the current longitudinal cohort, we also examined whether there were differences in the NAFLD risk associated with the GGT/HDL-C ratio in those of different ages, sex, FPG, SBP, and DBP populations. The results showed that the NAFLD risk associated with the GGT/HDL-C ratio was only significant in those with high SBP, while no significant differences were observed in the stratification of DBP, ages, sex, and FPG. It is well known that SBP is an important risk factor for NAFLD, even when elevated within normal levels [41, 42]. We, therefore, suggested that the higher risk of NAFLD associated with the GGT/HDL-C ratio in those with high SBP may be related to the common pathophysiological mechanisms of hypertension and NAFLD. According to the literature, patients with high levels of SBP and those with NAFLD both were more likely to develop insulin resistance (IR), and vascular and adipose tissue inflammation .
Notably, this study also found a nonlinear association of the GGT/HDL-C ratio with NAFLD. Through the dose–response relationship curve, we estimated a saturation threshold of approximately 31.79 U/mmol; when the GGT/HDL-C ratio exceeded the saturation threshold, the risk of developing NAFLD was no longer increased. These new findings provided useful intervention thresholds for the prevention of NAFLD, and to my knowledge, this study is the first to reveal a nonlinear association between the GGT/HDL-C ratio and NAFLD.
The mechanism underlying the correlation between the GGT/HDL-C ratio and new-onset NAFLD risk is unclear, and may be partially explained by the following two reasons: (1) According to the “three strikes theory”, the main pathological changes in the pathogenesis of NAFLD and its pathological progression include steatosis, lipotoxicity, and chronic inflammation ; and high levels of GGT are associated with both hepatic steatosis and chronic inflammation [34, 44]; Additionally, lipid disorders, high levels of GGT accompanying low levels of HDL-C, often lead to an increase in blood fatty acids, which are converted to excess TG when the blood fatty acids exceed the tolerance processing capacity of adipose tissues and various tissues, subsequently causing further organ damage, i.e., lipotoxicity . (2) It is well known that IR is the main pathophysiological mechanism of NAFLD . In order to explore the important role of IR in the relationship between GGT/HDL-C ratio and NAFLD, an mediation analysis was carried out in this study. The findings confirmed that MetS-IR played an intermediary role in the relationship between GGT/HDL-C ratio and NAFLD. These findings provided useful evidence for further revealing the relationship between GGT/HDL-C ratio and NAFLD.
The GGT/HDL-C ratio is an easily accessible clinical parameter, and the results of this study provided useful evidence for risk screening of NAFLD in non-obese individuals. On the basis of the current research results, we put forward several meaningful suggestions for a series of work that may be carried out in the future: (1) According to our results, we should pay more attention to the evaluation of GGT/HDL-C ratio of non-obese individuals, especially those with high SBP. (2) Non-obese people with a high GGT/HDL-C ratio found in clinical practice or routine physical examination should be offered appropriate interventions to prevent NAFLD, such as weight loss and improved dietary patterns .
Some limitations are worth noting: (1) the findings of this study can be applied to the non-obese population in China, but their applicability in other countries and ethnic populations is unclear; (2) the diagnosis of NAFLD in the current study was not performed by liver biopsy, which may increase the false-negative rate of new-onset NAFLD diagnosis, causing the incidence in the study to be lower than the true situation; (3) as with other observational studies, there were still unadjusted covariates that were not measurable or not found (Additional files 3, 4).