Validation of Fatty Liver Index and Hepatic Steatosis Index for Screening of Non-Alcoholic Fatty Liver Disease in Adults with Obstructive Sleep Apnea Hypopnea Syndrome
Introduction
Obstructive sleep apnea hypopnea syndrome (OSAHS) is a prevalent sleep disorder characterized by repeated episodes of upper airway obstruction during sleep, leading to hypoxia. It affects approximately 2% to 4% of the general population and 35% to 45% of obese individuals. OSAHS is closely associated with metabolic syndrome components, including visceral obesity, hypertension, dyslipidemia, and insulin resistance. Consequently, OSAHS is also linked to non-alcoholic fatty liver disease (NAFLD), a condition characterized by the abnormal accumulation of fat in hepatocytes.
NAFLD has become a significant public health concern due to its high prevalence worldwide, particularly in Asian countries, where it affects over 25% of the population. In China, NAFLD is the leading cause of chronic liver disease and abnormal liver function tests. The prevalence of NAFLD is even higher in individuals with OSAHS, especially in severe cases. Given the strong association between OSAHS and NAFLD, screening OSAHS patients for NAFLD is essential for comprehensive health risk evaluation.
While liver biopsy remains the gold standard for diagnosing NAFLD, it is invasive, expensive, and carries health risks. Non-invasive algorithms, such as the fatty liver index (FLI) and hepatic steatosis index (HSI), have been developed and validated for screening NAFLD in various populations. However, their diagnostic performance in OSAHS patients, who often have worse clinical and laboratory parameters, remains underexplored. This study aims to compare the diagnostic accuracy of FLI and HSI for detecting NAFLD in adults with OSAHS and determine their optimal cut-off values.
Methods
The study was conducted from March 2016 to January 2018 at two sleep laboratories in China. Consecutive adult patients newly diagnosed with OSAHS (apnea-hypopnea index [AHI] ≥5 events/hour) were enrolled. Exclusion criteria included prior OSAHS treatment, excessive alcohol consumption, hepatitis B or C, autoimmune hepatitis, drug-induced liver damage, or other conditions that could affect FLI and HSI levels.
Anthropometric measurements, including body mass index (BMI), waist circumference (WC), and neck circumference (NC), were recorded. Fasting blood samples were collected for biochemical analysis, including glucose, alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyltransferase (GGT), and lipid profiles. Polysomnography (PSG) was performed to assess OSAHS severity, with patients categorized into mild to moderate (AHI 5–30) and severe (AHI ≥30) groups.
NAFLD was diagnosed using abdominal ultrasonography, which identified hepatic steatosis based on increased echogenicity, vascular blurring, and deep-echo attenuation. FLI and HSI were calculated using the following formulas:
- FLI = (e^(0.953 log_e (TG/0.0113) + 0.139 BMI + 0.718 log_e (GGT) + 0.053 WC – 15.745)) / (1 + e^(0.953 log_e (TG/0.0113) + 0.139 BMI + 0.718 log_e (GGT) + 0.053 WC – 15.745)) * 100
- HSI = 8 * (ALT/AST ratio) + BMI (+2 if diabetes mellitus; +2 if female)
Statistical analysis included ROC curve analysis to determine the area under the curve (AUC) and optimal cut-off values for FLI and HSI. Differences between groups were assessed using Student’s t-test, Mann-Whitney U-test, or chi-square test, as appropriate.
Results
A total of 431 subjects were included, with 326 diagnosed with NAFLD and 105 serving as controls. Patients with NAFLD were younger and had significantly higher BMI, WC, NC, fasting glucose, total cholesterol (TC), triglycerides (TG), ALT, AST, ALT/AST ratio, GGT, AHI, oxygen desaturation index (ODI), and percentage of sleep time with SpO2 <90% (T90%). They also had lower high-density lipoprotein cholesterol (HDL-C), lowest O2 saturation (LaSO2), and average SpO2.
Both FLI and HSI values were significantly higher in the NAFLD group. The AUC for FLI and HSI in predicting NAFLD was 0.802 and 0.753, respectively, with FLI demonstrating significantly better diagnostic performance. The optimal cut-off values for FLI and HSI were 60 (sensitivity 66%, specificity 80%) and 35 (sensitivity 81%, specificity 60%), respectively.
In subgroup analysis, the AUC for FLI and HSI in mild to moderate OSAHS patients was 0.770 and 0.732, respectively. In severe OSAHS patients, the AUC was 0.802 for FLI and 0.748 for HSI. No significant differences in diagnostic performance were observed between FLI and HSI in these subgroups.
Discussion
This study validates the use of FLI and HSI as screening tools for NAFLD in OSAHS patients. FLI demonstrated superior diagnostic accuracy compared to HSI, with an AUC of 0.802 versus 0.753. The optimal cut-off values of 60 for FLI and 35 for HSI provided acceptable sensitivity and specificity for detecting NAFLD in this population.
The findings align with previous studies that evaluated FLI and HSI in general populations and specific patient groups. For example, Bedogni et al. reported that FLI <30 could rule out NAFLD with high sensitivity, while FLI ≥60 could rule it in with high specificity. Similarly, Lee et al. found that HSI 36 effectively excluded and detected NAFLD, respectively. However, the diagnostic performance of these indices in OSAHS patients was slightly lower, likely due to the unique clinical and laboratory characteristics of this population.
The study highlights the utility of FLI and HSI as cost-effective, non-invasive screening tools for NAFLD in OSAHS patients. FLI, in particular, offers advantages such as ease of calculation, quantitative assessment, and potential application in predicting cardiovascular and all-cause mortality. However, the study has limitations, including potential selection bias, reliance on ultrasonography for NAFLD diagnosis, and lack of liver biopsy data. Additionally, the sample size was relatively small compared to studies in general populations.
Conclusion
Both FLI and HSI are valuable screening tools for NAFLD in OSAHS patients, with FLI demonstrating better diagnostic performance. The optimal cut-off values of 60 for FLI and 35 for HSI provide acceptable sensitivity and specificity for detecting NAFLD in this population. These indices offer a practical, non-invasive approach to identifying NAFLD in OSAHS patients, facilitating early intervention and comprehensive health risk evaluation.
doi.org/10.1097/CM9.0000000000000503
Was this helpful?
0 / 0