Relationship between Obstructive Sleep Apnea and Liver Abnormalities in Older Patients: A Cross-Sectional Study

Background Older age is a risk factor for obstructive sleep apnea (OSA), which is associated with the development of nonalcoholic fatty liver disease (NAFLD). We aimed to investigate the correlation between OSA and liver injury among older patients. Study Design. This is a cross-sectional study. Methods Consecutive older (≥60 years) snoring patients were included. Subjects were divided into no OSA, mild OSA, moderate OSA, and severe OSA groups according to the apnea–hypopnea index (AHI) and were also separated into liver injury and nonliver injury groups based on liver function. Logistic regression analysis was applied to analyze the independent risk factors for liver injury. Results We studied 227 patients (155 male, 72 female). The prevalence of liver injury exhibited an increasing trend among groups with mild-to-severe OSA. In addition, body mass index, AHI, and TG showed significant differences between the liver injury and nonliver injury groups. Logistic regression analysis revealed that AHI and TG were the major contributing factors for liver injury in older patients (adjusted odds ratio [OR] = 1.055, p=0.013, and OR = 1.485, p=0.039, respectively). Conclusions Older patients with OSA have an increased risk of liver injury and NAFLD, and sleep apnea and high TG are important factors in contributing to the development of liver injury.


Introduction
Obstructive sleep apnea (OSA) comprises a group of chronic sleep-related breathing disorders characterized by the repeated occurrence of partial or complete upper airway obstruction during sleep. Te potential consequences of OSA include cardiovascular disease [1,2], metabolic syndrome [3,4], neurocognitive dysfunction [5], and diminished quality of life [6]. Tese consequences are related to higher mortality in patients and are associated with higher healthcare costs and utilization. Compared with younger people, older people have a higher prevalence, which varied in the range of 13% to 44% [7,8] because of the increased collapsibility of the upper airway, leading to the occurrence of obstructive apnea during sleep [9]. Te prevalence of cardiovascular and cerebrovascular diseases is higher in older men with OSA, which further enhances the risk of hospitalization and health burden [10]. Terefore, it is signifcant to focus on older adults with OSA and its comorbidities.
Nonalcoholic fatty liver disease (NAFLD) is increasingly recognized as a common cause of chronic liver disease worldwide, representing a clinical spectrum of liver disease ranging from nonalcoholic fatty liver (NAFL) and nonalcoholic steatohepatitis (NASH) to cirrhosis and hepatocellular carcinoma [11]. A study conducted by Noureddi et al. [12] found that older NAFLD patients had a higher prevalence of NASH compared with younger patients (72% vs. 56%). Furthermore, older patients with NAFLD had higher rates of advanced fbrosis compared with younger patients with NAFLD, revealing that aging is an independent risk factor for NAFLD. It is also noteworthy that OSA may be another contributor to NAFLD development [13]. Kallwitz et al. [14] observed that the severity of OSA was related to increased alanine aminotransferase (ALT) levels and also suggested that the apnea-hypopnea index (AHI) was a major contributing factor to abnormal levels of ALT or aspartate transaminase (AST), but AHI was not an independent risk factor for NASH [15].
Tere has been little research concerning the correlation between OSA and liver injury as defned by hepatic enzymes in older adults. Our intent was to investigate the characteristics associated with liver injury in older patients with OSA and the infuence of OSA on liver injury in these patients.

Study Design and Participants.
Our investigation was designed as a cross-sectional study. Older subjects (≥60 years) who attended our sleep laboratory because of symptoms of sleep apnea between January 2015 and October 2019 were recruited. We excluded patients who had previously been diagnosed with OSA or used continuous positive airway pressure (CPAP). We also excluded those patients with: (1) absence of liver ultrasound examination; (2) history of secondary fatty liver disease or chronic liver disease including excess alcohol consumption (defned as >20 g/day for men and 10 g/day for women), viral hepatitis, use of liver-damaging drugs, or other causes of chronic liver disease; and (3) documented history of other diseases such as serious cardiopulmonary or renal disease or acute infammatory disease. Te study was approved by the ethics committee of the First Afliated Hospital of Fujian Medical University (Fuzhou, China), and written informed consent was obtained from all patients.

Clinical and Laboratory
Data. Details on clinical and biochemical measurements of all subjects were collected: history of medication (including lipid-lowering drugs, hypoglycemic drugs, and antihypertensive agents), smoking habits, and alcohol consumption. Each subject underwent measurements of weight, height, and waist circumference after the removal of shoes and a heavy coat. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters, and waist circumference was determined at the midpoint between the inferior costal margin and the iliac crest.
Venous blood was drawn in the morning after an overnight fast. Te lipid and lipoprotein profle included total cholesterol (TC), triglycerides (TG), high-density lipoproteincholesterol (HDL-C), and low-density lipoprotein-cholesterol (LDL-C). Te liver enzymes ALT and AST were assayed. All biochemical blood measurements were analyzed using standard laboratory procedures on the Modular P800 autoanalyzer (Roche, Tokyo, Japan).

Polysomnographical Evaluation.
Te diagnosis of OSA was established by full-night polysomnography (P-Series Sleep System, Compumedics, Victoria, Australia) including the following parameters: electrooculography, electroencephalography, electromyography, electrocardiography, snoring, oronasal airfow, thoracic and abdominal respiratory eforts, pulse oxygen saturation, and body position. Te main independent variables were AHI, defned as the number of episodes of apnea and hypopnea per hour of sleep, and three diferent indices of nocturnal hypoxemia: the defned mean arterial oxyhemoglobin saturation during sleep as measured by blood analysis (SaO 2 ), the 3% oxygen desaturation index (ODI), and the percent of sleep time with SaO 2 < 90%. Te following commonly used cut-ofs for AHI were used to assess the severity of OSA: <5 events/h (no OSA), 5-15 events/h (mild OSA), >15-30 events/h (moderate OSA), and >30 events/h (severe OSA).

Defnition of the Liver Injury and NAFLD.
According to the National Health and Nutrition Examination Surveys III (NHANES III) laboratory standard [16], elevated ALT is defned as >40 U/L for males and >31 U/L for females, and elevated AST is defned as >37 U/L for males and >31 U/L for females. Liver injury is defned as elevated ALT or elevated AST. An abdominal ultrasound examination was carried out by an experienced ultrasonographer using the Technos DU8 sonography system with a 3.5 MHz transducer (Esaote SpA, Genoa, Italy). Te diagnosis of NAFLD was established based on the criteria developed by the Chinese Liver Disease Association [17].

Statistical
Analysis. All analyses were conducted using IBM SPSS Statistics for Windows, version 20.0 (IBM Corp., Armonk, NY, USA). First, we assessed whether the data were normally distributed using the Kolmogorov-Smirnov test. Continuous variables were expressed as mean ± standard deviation (SD) or median and interquartile range (IQR), depending on the distribution of data, and categorical data were presented as a number (%). Next, between-group comparisons were made for continuous data using Student's t-test or the Mann-Whitney U-test, and one-way analysis of variance (ANOVA) or Kruskal-Wallis H (K) for multiple-group comparison. Te chi-squared test was performed for categorical variables. Last, a logistic regression model was applied to determine the dependent risk factors for NAFLD in older patients with OSA. A p value <0.05 was accepted as indicative of signifcant diferences.

Clinical Characteristics of the Study Population.
Overall, 227 patients (155 male and 72 female) were enrolled in our study. Te mean age was 68.0 years, and mean BMI was 26.7 kg/m 2 . Of these patients, liver injury and NAFLD were present in 18.5% and 66.1% of the entire cohort, respectively, and in 18.9% and 69.2% of the patients with OSA, respectively.

Comparison of the Main Clinical and Biochemical
Characteristics of Patients with Respect to OSA Severity. Te primary clinical and biochemical characteristics according to the severity of OSA are compared in Tables 1  and 2. Clinical characteristics including age, sex ratio, medical history, and current smoking/drinking did not difer between the four groups (no OSA, mild OSA, moderate OSA, and severe OSA). BMI, neck circumference, waist circumference, AHI, ODI, T90%, and Epworth Sleepiness Score (ESS) increased while the lowest oxygen saturation (LaSO 2 ) decreased with increasing severity of OSA. Te prevalence of NAFLD in the no OSA, mild OSA, moderate OSA, and severe OSA groups was 42.31%, 79.20%, 62.70%, and 68.09%, respectively, and the diference was statistically signifcant (p � 0.012). Furthermore, the severe OSA group had greater prevalence of liver injury (23.4%) compared with the no OSA group (15.4%), mild OSA (10.4%), and moderate OSA group (18.6%), although this was not statistically signifcant.

Comparison of the Main Clinical and Biochemical Characteristics of Patients with Respect to Liver Injury Status.
Te primary clinical and biochemical characteristics according to the liver injury status are compared in Tables 3  and 4. Of the total 227 participants, 201 were patients with OSA (88.55%), including 38 with liver injury and 163 without liver injury. Patients with liver injury tended to be heavier and had signifcantly higher AHI values when compared with the nonliver injury group (BMI, 28 (Figure 1).

Discussion
Our study suggests that the severity of OSA and level of TG were independently associated with liver injury in older patients with OSA. Te prevalence of liver injury in the severe OSA group was higher than in the non-OSA group, and the AHI was signifcantly higher in the patients with liver injury. Logistic analysis showed that increased AHI was a signifcant risk factor for liver injury in older patients with OSA.
Te prevalence of NAFLD has been increasing rapidly, especially in older individuals, which has the potential to seriously afect the health and quality of life of these patients [18]. Many studies report that OSA is an independent risk factor for NAFLD, leading to an increasing clinical burden. OSA and NAFLD may manifest with phenotypic traits in common, such as obesity and insulin resistance. A "two-hit hypothesis" suggests that insulin resistance acts as a "frst hit" to result in hepatic steatosis, and oxidative injury acts as a "second hit" involved in the development of NAFLD in OSA [19]. A minority of studies [20,21] have confrmed that biochemical evidence of liver injury (including ALT and AST) increased with the aggravation of OSA, indicating that NAFLD might happen in the setting of OSA.
In a study by Turkay et al., nearly 67% of patients referred for a clinical sleep study looking for OSA had NAFLD [22]. A cohort study investigated about 1,300 subjects with suspected OSA and found that patients with severe OSA have a 2.5-fold higher risk for liver fbrosis [23]. Our study enrolled older patients and showed the prevalence of NAFLD in OSA (69.15%) was higher than in the no-OSA group (42.31%). Te incidence of liver injury (23.4%) was highest in the severe OSA group.
Many studies have shown that AHI is an independent predictor for NAFLD [24,25]. Mesarwi et al. [25] examined 35 subjects undergoing bariatric surgery and found that the AHI was higher in those with fbrosis, and severe OSA was more prevalent among those with fbrosis. In line with previous studies, we found that AHI was the independent risk factors for NAFLD. Furthermore, our study also showed that AHI and TG were independent risk factors for liver injury in older patients with OSA. Te study of Kallwitz et al. [14] enrolled 85 obese adults who underwent bariatric surgery and found OSAHS was independently correlated with the increase of ALT. Another study also supported the correlation between OSAHS and liver injury in severely obese adults who underwent weight-loss surgery. In this study, 46.5% of cases showed elevated liver enzymes, and the more severe the condition of OSAHS, the higher the probability of elevated liver enzymes. Multivariate regression showed that OSAHS and men were independently associated with elevated liver enzymes after excluding confounders [15]. However, the association between severity and liver injury was mild. Te reason might involve the fact that we included elderly patients, among whom aging is more likely to be associated with liver injury. In addition, it is notable that our study did not fnd a link between intermittent hypoxia and liver injury, and we assume that sleep fragmentation refected by the index of AHI inducing insulin International Journal of Clinical Practice   International Journal of Clinical Practice resistance and adipose tissue infammation might also be involved in liver injury in older patients with OSA [26,27]. Further research is necessary to characterize the mechanisms of this relationship. In addition, it is generally believed that obesity is a strong risk factor for the development of NAFLD. Studies investigating the prevalence of NAFLD in those patients with OSA whose mean BMI was more than 30 kg/m 2 suggest the prevalence of OSA-complicated NAFLD was about 73%-83% [23,28]. However, our studies did not exhibit the relationship between obesity and liver injury in older patients with OSA. Several explanations might contribute to the result. First, it has been reported that BMI does not demonstrate the same strength of association with NAFLD in older compared with middle-aged adults as the prevalence of obesity in older individuals is lower than in middle-aged adults [29]. Second, studies confrmed that aging plays an important role in inducing NAFLD by increasing lipid accumulation in the liver [30]. Further, OSA-related sleep fragmentation might also play a crucial role in NAFLD and liver injury. Taken together, the potential contribution of aging in OSA to the development of liver injury warrants further study.  International Journal of Clinical Practice Several limitations of our study have to be addressed. First, the study design itself was a cross-sectional design which only suggests a correlation but is insufcient to illustrate a cause-efect relationship. Second, the sample was relatively small because we had only recruited older subjects. Last, the diagnosis of NAFLD was based on liver ultrasonography instead of liver biopsy, which has inevitable limitations because of its low sensitivity for mild steatosis. However, ultrasonographic examination is widely used to screen for NAFLD which enables noninvasive, reliable, and accurate detection of mild-to-moderate hepatic steatosis compared to histopathologic examination. In addition, we also examined the liver function to assess liver injury.

Conclusions
In this study, we found that older patients with OSA had an increased risk of NAFLD and liver injury, and liver injury was linked with sleep apnea and high TG independently of several potential confounding factors. More attention should be given to older patients with OSA who have NAFLD and liver injuries.

Data Availability
Te datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest
Te authors declare that they have no conficts of interest.