The contribution of daytime sleepiness to impairment of quality of life in NAFLD in an ethnically diverse population

This study, in an ethnically diverse urban population showed that high Epworth Sleepiness Score, indicative of excessive daytime sleepiness, is a significant independent predictor of impairment HRQoL in patients with NAFLD. Overall, people with NAFLD had lower quality of life (particularly physical health scores) compared to the general population, in keeping with previous reports8,9,10. We found that ethnicity is associated with the age of onset of decline in physical health scores in people with NAFLD which was significantly earlier in patients of South Asian compared to White ethnicity. However, no significant differences in OSA prevalence or ESS scores were observed.

Other than NAFLD, previous studies have demonstrated variability in the prevalence of daytime sleepiness and OSA in different ethnicities, associated with a range of conditions including obesity, type 2 diabetes, cardiovascular disease and depression. In an atherosclerosis study of 2230 participants, higher prevalence of daytime sleepiness and undiagnosed OSA were reported in those of Black, Hispanic and Chinese ethnicities compared to White ethnicity17. The National Health and Wellness Survey from 7239 diabetic patients found the strongest predictors of sleep disturbance and daytime sleepiness were obesity, White ethnicity, female gender, low income, and smoking18. Longitudinal analysis of 41,094 participants of the UK Biobank found obesity, ethnicity (Black, Asian and mixed), depression, and social deprivation were the main factors associated with poor sleep and symptoms19. In Asian populations, daytime sleepiness and OSA were associated with subclinical atherosclerosis and diabetes irrespective of BMI20,21. The variability in these observations are not fully understood, but are likely to be multifactorial, associated with genetic, biological, social and environmental factors. Recent genetic studies have identified methylation sites in multiple genes specifically associated with daytime sleepiness in African Americans22. Differences in craniofacial structures have been linked to elevated risk of OSA and associated symptoms despite lower rates of obesity in Asian populations compared to White and Black ethnicities23,24.

In our study, 13.5% of participants had a diagnosis of OSA at enrolment, but a significantly higher proportion (24.9%) recorded ESS scores > 10. To our knowledge, this is the first report that shows ESS score > 10 is a significant independent predictor of physical HRQoL, in addition to factors known to be associated with impaired physical health scores in NAFLD; Older age, female gender, BMI, presence of T2DM, degree of fibrosis including cirrhosis8,9,10,25,26,27. In our study, ESS score > 10 was associated with metabolic co-morbidites (obesity, T2DM), and liver stiffness score by transient elastography. Our study was not designed to detect associations of HRQoL or sleepiness with histology and the lack of statistical associations may be related to the low proportion of patients in our study who had liver biopsy as part of routine clinical care (21%). Traveling, our findings were congruent with findings by Newton et al.28, which reported perceived fatigue experienced by patients with NAFLD was associated with daytime sleepiness, but not liver disease incision. However, a Swiss study of sleep disturbance in patients with biopsy-proven NAFLD, insulin resistance and elevated serum transaminase levels29 showed an association between sleepiness and NASH on biopsy albeit the Swiss cohort differed in that patients had lower mean BMI, ESS and fibrosis scores, and higher mean ALT. Continuous positive pressure ventilation (CPAP) is the primary treatment for OSA, associated with improvements in OSA symptoms, HRQoL30 and reduction in ESS scores31. However, a randomised controlled trial of CPAP, for 6 months in people with NAFLD and OSA did not demonstrate regression of hepatic steatosis or fibrosis32 Suggesting that mechanisms beyond hypoxaemia and fibrosis regression influence HRQoL in patients with NAFLD and OSA. In addition, further work is needed to assess the impact of daytime sleepiness on key lifestyle determinants, such as exercise capacity, mood, appetite, eating patterns and other activities or daily living, which may reduce the ability of people with NAFLD to make substantive lifestyle changes. Daytime sleepiness in people with NAFLD without obesity was high in our study (26.7% of all participants with ESS score >10 did not have obesity), keeping with similar findings in a non-morbidly obese Italian NAFLD cohort33. Future studies with polysomnography to confirm OSA are warranted in this subpopulation of NAFLD with high ESS scores.

To date, the majority of studies assessing HRQoL in NAFLD have been conducted in the United States with preponderance of White ethnicity (>73%)8,9,10,26,27. In patients with advanced NASH fibrosis participants enrolled in the multinational phase 3 STELLAR trials of selonsertib, there were significant in HRQoL scores between White, Asian and Black groups of differences, although only 1.5% of the total cohort were of Black ethnicities9. We previously reported that patients of South Asian ethnicity have a greater risk of developing NAFLD at a younger age (by approximately a decade), with a lower BMI compared to White patients with comparable disease stages34. This is mirrored in our current study by the in physical health scores despite lower BMI, and together, this signifies that South Asian patients with NAFLD suffers physical impairment from a much younger age compared to White patients. The causes underlying for this are likely to be multifactorial, though higher rates of diabetes were observed in South Asian people and Chawla et al. reported a significant reduction in physical health scores in people with NASH and diabetes compared to non-diabetic NASH patients25. Further work is required to understand the impact of South Asian ethnicity on the natural history of NAFLD and why it impacts HRQoL at younger ages. While higher rates of diabetes may play a role, other factors may include diet, genetics, cultural health behaviors and social deprivation, which were not explored in this study. Our collective work highlights the importance of ethnicity when evaluating NAFLD severity from a clinician as well as a patient perspective.

This study has both strengths and limitations. Firstly, the patients enrolled reflect real world NAFLD in clinical practice, not restricted by stringent clinical trial criteria. Secondly, our study population is more ethnically diverse compared to previous studies, with a significant proportion of South Asian patients. Thirdly, SF-36v2 questionnaire was used to evaluate HRQoL in this study. Although this questionnaire is not specifically designed to evaluate HRQoL in chronic liver disease, unlike the Chronic Liver Disease Questionnaire (CLDQ), it is a widely used HRQoL assessment tool validated in different global populations that allows for comparisons with different disease states and the general population , which is not possible with CLDQ. Furthermore CLDQ was developed and validated using SF-36 as the gold standard25. ESS alone has modest discriminatory ability as a screening tool for OSA (sensitivity ranging from 38 to 66% and specificity ranging from 48 to 79% with ESS score > 10 as threshold)35,36. Although the focus of our study was to elicit the symptoms of daytime sleepiness rather than to formally diagnose OSA, one limitation of our study is that polysomnography was not used to confirm those suspected of OSA based on the ESS scores. High ESS score reflects a state of fatigue, which is also 1 of 8 major components measured in the SF-36 questionnaire. A limitation in our study is the absence of control group(s) to assess whether liver disease or other associated factors contribute to fatigue in patients with NAFLD, but again, this was not an aim of our study. Finally, the cross-sectional nature of this study prevents causal conclusions to be drawn from the reported correlations. Very few longitudinal HRQoL studies have been conducted to date in NAFLD. The PIVENS trial for NASH reported no significant changes in SF-36 scores between the study arms after 96 weeks37. In patients with NASH F2 – 3 fibrosis treated for 24 weeks with selonsertib in a phase 2 clinical trial, patients in whom there was improvement in a surrogate marker of fibrosis (percentage collagen) but not histological NAFLD activity score also recorded improvement in physical health scores38. Future prospective longitudinal studies are warranted to investigate strategies that can improve physical impairment in people with NAFLD of different ethnicities.

Leave a Comment