Revisiting the association between metabolic dysfunction-associated steatotic liver disease and sleep: Moving from correlation to causation

Abstract

The retrospective study by Hyatt et al reported that metabolic dysfunction-associated steatotic liver disease (MASLD) with fibrosis was associated with poorer sleep, which in turn correlated with worse life quality in patients with human immunodeficiency virus. While their findings were significant, some limitations warrant consideration. First, the cross-sectional design precludes causal relationships. Second, no subgroup analyses stratified were conducted, leaving it unclear which sleep disturbances may increase MASLD risk. Third, baseline differences between groups may have introduced confounding. To address these limitations, we conducted a Mendelian randomization analysis to minimize confounding and assess the potential causal relationship between sleep traits and MASLD. Mendelian randomization results indicated that sleep apnea significantly increases the risk of MASLD (P = 0.014), whereas no causal associations were identified for other sleep traits. These findings complement and extend those of Hyatt et al, providing further insight into the association between MASLD and sleep.

Key Words: Metabolically-dysfunction-associated steatotic liver disease; Sleep; Mendelian randomization; Sleep apnea; Causation

Core Tip: The recent study by Hyatt et al investigated the prevalence and associations between metabolic dysfunction-associated steatotic liver disease (MASLD), poor sleep, and life quality in patients with human immunodeficiency virus. To further explore the potential causal relationship between specific sleep traits and MASLD, we conducted a Mendelian randomization analysis, which demonstrated that sleep apnea significantly increases the risk of MASLD (P = 0.014), whereas no causal associations were found for other sleep traits (P > 0.05). These findings complement and extend those of Hyatt et al, providing additional insight into the associations between MASLD and sleep.

TO THE EDITOR

We read with great interest the article by Hyatt et al[1], which investigated the prevalence and associations of metabolic dysfunction-associated steatotic liver disease (MASLD), poor sleep, and health-related quality of life among individuals with human immunodeficiency virus[1]. The authors reported that MASLD with fibrosis was associated with poorer sleep quality, and poor sleep was associated with worse health-related quality of life. The relationship between sleep and fatty liver disease has attracted increasing attention. A study utilizing data from the National Health and Nutrition Examination Survey examined the association between sleep traits and MASLD in 4731 older adults. The findings indicated that both short and long sleep durations, as well as the presence of sleep disorders, were significantly associated with MASLD[2]. Another Iranian cohort study of 9151 participants, including 1320 with MASLD, demonstrated that those with MASLD had shorter sleep durations, delayed sleep-wake cycles, reduced sleep efficiency, and more frequent daytime napping[3].

While the findings of Hyatt et al[1] are clinically significant, some limitations warrant further discussion. First, the authors acknowledged that the cross-sectional design of their study precludes the establishment of causal relationships. Second, no subgroup analyses stratified by specific sleep characteristics or disorders were performed, making it unclear which types of sleep disturbances may confer a higher risk of MASLD. Third, baseline characteristics differed significantly between groups, which may have introduced confounding effects. Drug abuse and depression have been shown to cause sleep disturbances in individuals living with human immunodeficiency virus[4]. To address these limitations, we conducted a Mendelian randomization (MR) analysis to minimize confounding and more robustly evaluate the potential causal relationship between sleep traits and MASLD.

MR utilizes single-nucleotide polymorphisms (SNPs) as instrumental variables to infer causality between exposures and outcomes[5]. Due to the random allocation of SNPs at conception, MR analyses are less susceptible to confounding and reverse causation[6]. Genome-wide association study (GWAS) data were sourced from the FinnGen[7] and IEU OpenGWAS databases[8] (Supplementary Table 1). The FinnGen project, involving nearly 500000 Finnish participants, is a genomic initiative providing a valuable resource for investigating genetic variations and their clinical implications[9]. The IEU OpenGWAS database is also a publicly accessible database that integrates large-scale GWAS data[8]. SNPs meeting standard MR assumptions were selected as instrumental variables (Supplementary Table 2), while those associated with confounders or outcomes were excluded using the GWAS catalog[10]. The inverse variance weighted method was employed as the primary analytical approach, supplemented by MR-Egger, weighted median, and weighted mode methods. Heterogeneity was assessed using Cochran’s Q test. Horizontal pleiotropy was evaluated using the MR-Egger intercept test and the MR Pleiotropy RESidual Sum and Outlier test global test.

As shown in Figure 1A, MR analysis revealed that sleep apnea significantly increases the risk of MASLD [odds ratio (OR) = 1.362; 95% confidence interval (CI): 1.065-1.742; P = 0.014]. In contrast, no causal associations were observed for other sleep traits, including trouble falling asleep, sleep duration, daytime dozing, morning/evening person, nap during day, hypersomnia, and disorder of the sleep–wake schedule. No significant heterogeneity or horizontal pleiotropy was detected (P > 0.05). Additionally, reverse MR analysis did not reveal any causal relationship from MASLD to sleep traits (Figure 1B). These results provide valuable complementary evidence suggesting that the management of sleep apnea may reduce MASLD risk. The results are consistent with previous studies. A systematic review by Umbro et al[11] revealed that the prevalence of MASLD increased among individuals with sleep apnea. A cross-sectional study by Huang et al[12] also demonstrated that the prevalence of MAFLD increased with the severity of sleep apnea. The latter study further identified body mass index, oxygen desaturation index, and triglyceride (TG) levels as potential contributing factors to MAFLD in sleep apnea patients.

Figure 1 Causal relationship. A: Causal relationship from sleep traits to metabolic dysfunction-associated steatotic liver disease; B: Causal relationship from metabolic dysfunction-associated steatotic liver disease to sleep traits. OR: Odds ratio; MR: Mendelian randomization; MR-PRESSO: Mendelian randomization-Pleiotropy RESidual Sum and Outlier test.

Chronic intermittent hypoxia (IH), a hallmark of sleep apnea, is considered a critical driver of MASLD progression through several molecular pathways[13]. These include the activation of hypoxia-inducible factors, nuclear factor kappa-B signaling, and the upregulation of proinflammatory cytokines, contributing to hepatic inflammation, fibrogenesis, and metabolic dysfunction. Studies demonstrate that IH impairs glucose homeostasis[14]. Glucose-sensing neurons mediate IH-induced insulin resistance by enhancing sympathetic activity, elevating circulating free fatty acids, and stimulating hepatic glycogenolysis. IH also promotes hyperlipidemia by inhibiting the clearance of TG-rich lipoproteins[15]. Furthermore, IH suppresses lipoprotein lipase activity, leading to increased cholesterol and TG levels[16]. Overall, sleep apnea contributes to MASLD progression through IH-induced insulin resistance, which diverts excess free fatty acids into TG and cholesterol synthesis, thereby promoting hyperlipidemia and disease progression.

In addition, IH contributes to MASLD progression via inflammation. Savransky et al[17] found elevated interleukin (IL)-1 beta, IL-6, and tumor necrosis factor alpha levels in mice exposed to IH, accompanied by hepatic lobular inflammation and fibrosis. Schaefer et al[18] found that IH markedly induced IL-6 expression in hepatocytes and macrophages. IH activates inflammasomes and caspase-1 in lipid-laden hepatocytes, while promoting hepatocyte-Kupffer cell crosstalk via extracellular vesicles, thereby amplifying hepatocellular injury. In a biopsy-confirmed study involving 80 children with MASLD, sleep apnea was associated with increased intestinal permeability, endotoxemia, expansion of adiponectin-deficient progenitor cell populations, and upregulation of Toll-like receptor 4 in hepatocytes, Kupffer cells, and hepatic stellate cells, which were key features of steatohepatitis and fibrosis[19].

The identification of biomarkers linking MASLD and sleep disorders is critical for early detection and disease monitoring. In a retrospective study by Li et al[20], the prevalence of MASLD in patients with sleep apnea increased with higher triglyceride-glucose (TyG) index values (P < 0.05), while metabolic score for insulin resistance (METS-IR) exhibited a non-linear association with MASLD risk (P < 0.05). A dose-response relationship was evident between MASLD risk and METS-IR, and a linear association was found for the TyG index. Notably, METS-IR (under the receiver operating characteristic curve = 0.778; 95%CI: 0.716-0.836) outperformed the TyG index (under the receiver operating characteristic curve = 0.775; 95%CI: 0.711-0.828) in predicting MASLD in sleep apnea patients.

Inflammation-related biomarkers have also drawn attention. Yang et al[21] investigated the independent relationships between trouble sleeping and the systemic immune-inflammation (SII) index in patients with MASLD. Their results indicated that both SII (OR: 1.21; 95%CI: 1.08-1.35) and trouble sleeping (OR: 1.24; 95%CI: 1.05-1.47) were independently associated with MASLD. An additive interaction between SII and sleep trouble was observed, with SII partially mediating this relationship. For patients with MASLD and coexisting sleep apnea, interventions to improve sleep-disordered breathing may offer benefits. Effective strategies include positive airway pressure[22], pharmacological treatments[23], physical activity[24], and bariatric surgery[25]. Zhang et al[25] found that bariatric surgery significantly improved sleep apnea, nocturnal hypoxia, hepatic steatosis, and fibrosis (P < 0.05). Clinical decision-making should balance risks and benefits, incorporating patient preferences and the availability of healthcare resources. In conclusion, our MR results indicated that sleep apnea causally increases the risk of MASLD (P = 0.014). We recommend that screening for sleep apnea be included in MASLD risk assessment protocols. We sincerely appreciate the contributions of Hyatt et al[1]. Building on their pronounced contributions, the MR analysis is expected to provide novel insight into the association between MASLD and sleep.