Obstructive sleep apnea is associated with liver disease: a population-based cohort study
Tzu-Chieh Chou a,b, Wen-Miin Liang a,c, Chang-Bi Wang a, Trong-Neng Wu a,d*, Liang-Wen Hang e,f,g*
a Department of Public Health, College of Public Health, China Medical University, Taichung, Taiwan
b Department of Health Risk Management, College of Management, China Medical University, Taichung, Taiwan
c Graduate Institute of Biostatistics, Biostatistics Center, College of Management, China Medical University, Taichung, Taiwan
d Department of Nursing, College of Medicine and Nursing, Hungkuang University, Taichung, Taiwan
e Department of Respiratory Therapy, College of Health Care, China Medical University, Taichung, Taiwan
f Sleep Medicine Center, Department of Pulmonary and Critical Care Medicine, China Medical University Hospital, Taichung, Taiwan
g Department of Healthcare Administration, Asia University, Taichung, Taiwan,
* Corresponding author:
Liang-Wen Hang, Department of Respiratory Therapy, College of Health Care, China Medical University, Taichung, Taiwan, Tel: 22062121-1781, Fax: +886-4-22062121-1784, E-mail: [email protected] or Trong-Neng Wu, Department of Nursing, College of Medicine and Nursing, Hungkuang University, Taichung, Taiwan, Tel: +886-4-26318652, Fax: +886-4-26310744, E-mail:
Conflict of interest: None. Article type: Original Study
Keywords: OSA, liver disease, NAFLD, cirrhosis, hepatitis B, hepatitis C Running head: OSA and liver disease
Word count: abstract 206, text 3034, references 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
Abstract Background:
The association between obstructive sleep apnea (OSA) and the risk of liver disease is unclear. Moreover, population-based study on the risk of liver disease among OSA patients has not yet been conducted. This study aimed to investigate the risk of subsequent development of liver disease among OSA patients.
Methods:
Using Taiwan National Health Insurance claims data, this study collected subjects from a cohort of 17,374 patients with OSA diagnosed between 2000 and 2008. A comparison group of 69,496 persons was selected from the same database and matched by age, gender, urbanization, income, and date of initial admission. All subjects were followed up until 2010. Liver disease incidence and risk were calculated.
Results:
The overall risk of liver disease among OSA patients was significantly higher than in the comparison group (aHR = 5.52, p < 0.001). Non-alcoholic fatty liver disease, cirrhosis, and hepatitis C had significant aHRs of 5.29, 7.50, and 7.19 (all at p < 0.001), respectively. In contrast, hepatitis B had the smallest aHR, 3.71.
Conclusions:
The risk of liver disease was more than five times higher among OSA patients
compared with the comparison group, in particular for cirrhosis and hepatitis C. Liver disease is thus a very important health issue among OSA patients.
35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
Introduction
Liver diseases, including viral infections, alcoholic, and non-alcoholic causes, are major health problems around the world. In the Far East, the prevalence of viral and non-alcoholic causes of liver disease is extremely high [1,2]. Fatty liver disease is also a widespread and common health issue around the world, but population-based surveys in China in particular have revealed that fatty liver has become a serious public health problem accompanying improved living standards and increased alcohol consumption [3]. Epidemiological studies of fatty liver disease have revealed some important risk factors. Large population-based surveys in the Asia-Pacific region indicate that the prevalence rates of non-alcoholic fatty liver disease (NAFLD) range from 12% to 24% in population subgroups, depending on age, gender, ethnicity, and geographic location (urban vs. rural) [4]. Most importantly, the major risk factors of NAFLD are obesity and insulin resistance [5]. NAFLD exists along a continuum with various degrees of severity, such as steatosis without inflammation, non-alcoholic steatohepatitis (NASH), and liver cirrhosis or end-stage NAFLD [6]. Insulin
resistance and dyslipidemia are the key promoters of fatty acid deposition in the liver [7,8]. The identified risk factors for the progression of NAFLD are obesity, age of over 45 years, diabetes, hypertriglyceridemia, and hypertension [9,10]. Previous studies have also shown prevalences of hepatitis B and hepatitis C of 13.0% and 4.2%, respectively, in Taiwan [11]. Family history, vertical transmission from mother to child, and age are major risks for hepatic viral infection [12].
Although the pathogenesis of NAFLD is not fully understood, a “two-hit” hypothesis has been proposed [13]. In this hypothesis, the first hit represents increased fat accumulation in the liver because of peripheral insulin resistance. The nature of the 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82
second hit is NASH, which results from the progression of NAFLD. Although the pathogenesis remains unclear, the consequences of oxidative stress, including lipid peroxidation, cell degeneration and necrosis, apoptosis, pro-inflammatory cytokine expression, liver stellate cell activation, and fibrogenesis, have all been implicated [14-16]. Oxidative stress also plays an important role in the pathogenesis of viral hepatitis by worsening severity of fibrosis progression in patients [17].
Obstructive sleep apnea (OSA) is a sleep breathing disorder characterized by
recurrent airflow obstruction caused by a total or partial collapse of the upper airway [18]. Untreated OSA can contribute to the development or progression of other disorders that affect multiple daily functions [19]. OSA is also associated with systemic or pulmonary hypertension [20,21], cardiac and cerebrovascular morbidity and mortality [18,22], neuropsychological impairment [23], and increased risk of automobile and occupational accidents [24,25]. Moreover, previous studies have postulated a correlation between OSA and liver disease. A meta-analysis by Musso et al. including 18 cross-sectional studies showed a two-fold increased risk of liver disease among OSA patients, indicating that OSA was associated with liver disease [26]. OSA can induce oxidative stress after intermittent hypoxia (IH) and may damage the liver because of lipid peroxidation [27]. However, studies of NAFLD, cirrhosis, hepatitis B, and hepatitis C infection among OSA patients are limited. To the best of our knowledge, no study exists that has utilized a national database in an Asian population. Most studies of OSA patients correlated with liver disease are also cross-sectional, hospitalized-based study types, and no large-scale, population-based study has been conducted to elucidate the relationship between OSA and liver disease. Given the need for research in this area, as well as the findings supporting the
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107
association of OSA with liver disease, the present study attempted to determine the incidence of NAFLD, NASH, hepatitis B, and hepatitis C infection among OSA patients in a retrospective and population-based study in Taiwan over a 10-year follow-up period, and it investigated the incidence and effects of comorbidities on liver disease.
Materials and methods Data sources
The National Health Insurance (NHI) database, which was published by the Department of Health of Taiwan, covers the period from 1997 to 2010. The NHI program was started in 1995 to finance health care for all Taiwanese residents. The coverage rate was more than 99% for the whole population in 2010. The completeness and accuracy of the NHI Research Database (NHIRD) is guaranteed by the
Department of Health and the NHI Bureau of Taiwan. This retrospective, population-based cohort study screened all subjects. All sampled individuals were followed up until the end of 2010.
Ethics statement
The NHIRD is composed of anonymous secondary data released to the public for research purposes. Thus, this study was exempted from a full review by the local ethics review committee.
Subject collection and study design
This study selected all patients aged >20 years from the database diagnosed with OSA (based on ICD-9-CM codes 780.X, 32720, 32721, 32726, and 32729) between
108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132
January 1, 2000 and December 31, 2008. All OSA patients were diagnosed with OSA and examined with polysomnography (PSG) within one year before or within one year after the index day. The first day of OSA diagnosis was assigned as the index day. Patients who had liver disease before the index day were excluded to avoid confounding effects. The comparison group was selected from the same database, excluding those diagnosed with OSA or liver disease before the index day, and they were matched by age, gender, urbanization status, income, and index day at a ratio of 1:4 (OSA vs. comparison group) during the same time period. Liver disease
diagnosis, including NAFLD (ICD-9-CM code: 571.8), cirrhosis (ICD-9-CM codes: 571.2, 571.5, and 571.6), hepatitis B (ICD-9-CM codes: V02.61, 070.20, 070.22, 070.30, and 070.32), and hepatitis C (ICD-9-CM codes: V02.62, 070.41, 070.44, 070.51, and 070.54) were identified among the OSA patients and the comparison group. Social indicators for each subject were calculated using age, gender, income, and insurance district to ensure matching accuracy. In total, 17,374 patients with OSA and 69,496 controls were enrolled in the study. The cohort was followed until their exit from the NHI program, death, or the end of 2010.
Statistical analysis and liver disease risk analysis
Liver disease developing in the first year after the index day is difficult to detect based on symptoms alone, so we excluded patients diagnosed with liver disease registered in the first year after the index day. Each subject was followed up for a minimum of 2 years and a maximum of 11 years. The demographic data and
comorbidities of these two cohorts were first analyzed. Cumulative incidence analyses were performed using the Kaplan–Meier method, and the differences between the curves were tested with the log-rank test. We measured the hazards ratios (HRs) of 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157
liver disease using the stratified Cox proportional hazards model to determine whether OSA is a risk factor for liver disease development. The model included age, gender, and comorbidities. All data management and calculation of HRs were performed using the SAS System (version 9.3; SAS Institute, Cary, NC). Cumulative incidence was analyzed using the Statistical Package for the Social Sciences (version 10.0; SPSS Inc, Chicago, IL). Statistical significance was set as a two-tailed p value of <0.05.
Results
Demographic data
A total of 17,374 OSA patients were selected as the OSA cohort. The control cohort consisted of 69,496 matched subjects without OSA diagnosis. OSA predominantly affected males (11,852 or 68.2%). Approximately 4,314 (24.8%), 3,414 (19.7%), 1,612 (9.3%), 2,156 (12.4%), and 818 (4.7%) OSA patients and 1,221 (1.8%), 2,616 (3.8%), 1,910 (2.8%), 439 (0.6%), and 170 (0.2%) patients in the comparison group had hypertension, diabetes, hyperlipidemia, congestive heart failure, and atrial fibrillation, respectively. All incidences of comorbidities were higher in patients with OSA compared with the comparison group (p < 0.001) (Table 1).
Incidence of liver disease among subjects
Table 2 shows liver disease incidence in the OSA cohort and the comparison group. The incidence of liver disease was higher among OSA patients compared with the comparison group (11.6% vs. 3.0%). NAFLD and cirrhosis also showed higher incidences among patients with OSA than the comparison group (7.0% vs. 1.8% and 1.8% vs. 0.4% for NAFLD and cirrhosis, respectively). Viral hepatitis showed the 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182
same pattern, and hepatitis B and hepatitis C incidences were nearly three-fold and five-fold higher among OSA patients than in the comparison group, respectively.
Cumulative incidence of liver disease in the OSA and control groups
Kaplan–Meier estimates of the cumulative incidences of liver disease for OSA and the comparison group are shown in Figure 1. The cumulative incidence of all liver disease in the OSA cohort was significantly higher than in the comparison group (p < 0.001, log-rank test, Figure 1A). After the 10-year follow-up, the cumulative incidence was almost 10 times higher among OSA patients than in the comparison group. The greatest difference between the two groups was for NAFLD (p < 0.0001, log-rank test, Figure 1B). Nevertheless, cirrhosis, hepatitis B, and hepatitis C showed smaller but still significant differences between the OSA patients and the comparison group (Figures 1C–1E).
After adjusting the Cox multivariable proportional hazards analysis for age, gender, and comorbidities (Table 3), the risk of liver disease among the OSA patients was significantly higher than in the comparison group (aHR = 5.52, p < 0.001), and the risk was similar with NAFLD (aHR = 5.29, p < 0.001). Cirrhosis had an aHR of 7.50, whereas hepatitis C had an aHR of 7.19. Hepatitis B had the smallest aHR of 3.71, although all liver diseases showed a significant risk increase among OSA patients (all at p < 0.001).
Discussion
Liver disease is a common health problem around the world, especially in the Far East. Studies show that the overall prevalence of NAFLD in Taiwan is approximately 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207
11.4% to 41%; the rates are higher in population subgroups—from 66.4% in healthy taxi drivers to 80% in obese individuals who attended weight reduction programs [28]. Compared with other cross-sectional, hospital-based studies, our research is probably the first large-scale, population-based cohort study to show an increase in liver disease among OSA patients compared with a control group. In our study, the risks of NAFLD and cirrhosis were higher among OSA patients compared with the comparison group (aHR = 5.29 and 7.50, 95% CI, 4.81–5.81 and 6.02–9.34, respectively). This study is also the first one to show a significant risk of viral hepatitis among OSA patients as compared with the comparison group (aHR = 3.71 and 7.19, 95% CI, 3.09–4.46 and 5.41–9.55 in hepatitis B and hepatitis C,
respectively). The use of a large, representative, nationwide, and population-based sample to observe the risk of liver disease among OSA patients increases the validity of the results. The retrospective cohort study design provides a sufficient conclusion that represents the general OSA population.
Previous studies have postulated a relationship between OSA and liver disease since 2005. A study of 83 OSA patients and matched controls suggested a relationship between OSA and the progression of steatosis to steatohepatitis [29]. In a larger study of 218 OSA patients, severe OSA (defined as greater than 50 apneic/hypopneic episodes/hour [AHI]) was associated with increased liver enzymes (odds ratio of 5.9, p < 0.02). Patients with an AHI greater than 50/hour were also more likely to have steatosis, lobular necrosis, and fibrosis by liver biopsy [30]. One study on NASH included 99 patients with PSG data: 77 NASH and 22 non-NASH controls. The results suggested that frequent nocturnal hypoxic episodes among NAFLD patients may be a risk factor for developing NASH [31].
208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232
Nocturnal IH can increase systemic oxidative stress and serum lipid peroxidation [27], which is implicated in the pathogenesis of NAFLD, NASH, and its progression to advanced stages of hepatic fibrosis [32]. Aside from studies demonstrating a higher prevalence of NAFLD in patients with sleep-related breathing disorders [33], a recent study has indicated that a significant correlation exists between OSA and the presence and severity of NAFLD. OSA syndrome was found to be highly common, especially in severe NAFLD cases. The incidences of NAFLD in mild, moderate, and severe OSA were 59%, 58.3%, and 78.2%, respectively. The most highly correlated
parameter to the severity of NAFLD was found to be the duration of hypoxia, so the authors suggested that nocturnal hypoxemia is a key pathophysiologic factor in the development of NAFLD [34].
Data from studies on animal and human models that mimic conditions of IH and sleep fragmentation in OSA have demonstrated the adverse effects of OSA on insulin resistance, glucose intolerance, and diabetes risk [35]. IH has also been shown to induce insulin resistance, independent of obesity, in healthy humans [36]. Insulin resistance is strongly linked to the excessive deposition of triacylglycerol in
hepatocytes, which is the hallmark for NAFLD diagnosis [37]. However, the causal relationship between insulin resistance and NAFLD remains debatable [38]. Given that lab tests showing insulin and glucose levels in subjects were unavailable, our study was unable to include this information in our analysis. In an effort to consider the effects of insulin resistance and glucose intolerance, we included diabetes as a comorbidity in this study. Moreover, all crude HRs showed that patients with
comorbidit ies such as hypertension, diabetes, hyperlipidemia, congestive heart failure, 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257
and atrial fibrillation had significantly high risk for the development of liver diseases (5.19, 2.89, 2.15, 6.73, and 6.70, respectively). In the multivariate analysis, obvious decreased HRs were found for all comorbidities ( aHRs of 1.26, 1.46, 1.13, 1.48, and 1.79, respectively), but still remained statistically significan t (p<0.05) for all
comorbidities except hyperlipidemia (p=0.19) . These results emphasize the promoting effect of comorbidities in developing liver diseases in our study group. Nevertheless, during the observation period after the index day of this study, the incidences of comorbidities were 20.6% vs. 1.7%, 16.4% vs. 7.8%, and 9.0% vs. 5.9% for
hypertension, diabetes, and hyperlipidemia when comparing the OSA group with the comparison group with any liver disease, respectively. As compared with the baseline status of comorbidities (24.8% vs. 1.8%, 19.7% vs. 3.8%, and 9.3% vs. 2.8% for hypertension, diabetes, and hyperlipidemia when comparing the OSA group with the comparison group, respectively, Table 1), a slight decrease in incidences of
comorbidities among OSA patients and a slight increase in incidences of
comorbidities in the comparison group were detected after the index day. This could lead to liver diseases more frequently occurring in the comparison group than in the OSA group. However, in this study, we still found that an obvious risk of liver diseases occurred in OSA patients, indicating that the conclusion drawn from this study could actually be more conservative.
Some investigations have focused on antiviral drug use, such as interferon alpha, in patients with hepatitis C and have made correlations with sleep disturbances [39]. Nevertheless, limited studies have discussed the relationship between OSA and viral hepatitis. Hypoxemia with reoxygenation may be analogous to ischemia-reperfusion, and reoxygenation may cause additional damage through the further production of 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282
free radical species that may cause different manifestations, especially in the hepatitides [40]. Other studies have postulated that hypoxemia-induced factors play an important role in the pathogenesis of viral hepatitis, including hepatitis B, hepatitis C, and hepatitis E [41]. Previous investigations illustrate that disrupted sleep appears to have a serious impact on immune function. In healthy humans and animals, sleep disruption or deprivation could lead to changes in immunological reactions [42]. Sleep deprivation or restriction has been demonstrated to inhibit the response of antibodies to vaccination against hepatitis A and hepatitis B [43,44]. OSA patients have poor sleep quality because of frequent sleep interruption and apnea-related arousals [45]. Pro-inflammatory markers, including tumor necrosis factor-a, interleukin (IL)-6, IL-8, and C-reactive protein, are increased among OSA patients [46]. Although this study could not provide direct historical observation of liver disease in both groups, our results for viral hepatitis have given us insight into the response to infections in the context of sleep disorders, especially OSA. Using a prospective cohort study design on a topic related to viral hepatitis incidence among OSA patients and the influence of OSA on viral hepatitis is warranted for future study.
This study is a large, population-based, retrospective cohort study. Nevertheless, it has some limitations that should be addressed. First, the NHIRD does not contain personal information, such as lifestyle, body mass index, or smoking/alcohol use, which may contribute to NAFLD risk. The indices of OSA severity and
apnea/hypopnea as well as the frequency and degree of IH were also unavailable in the database. Second, misclassification of OSA may occur in the NHIRD. Thus, this study enrolled OSA patients with nocturnal PSG evidence and definite diagnoses to 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307
minimize possible bias. Third, liver function tests, liver biopsies, or ultrasonographic data were unavailable. These data can be used to determine the factors that correlate best with the presence of NAFLD, which may add to the knowledge of underlying mechanisms linking NAFLD to OSA. Fourth, while previous studies have reported an overall prevalence rate of ≥2% to 4% in the general population [47], we only found OSA occurring in approximately 0.9% of the subjects in our database. Therefore, OSA being under-diagnosed in Taiwan seems very likely. This scenario leads to the possibility of many undiagnosed OSA patients in the general Taiwanese population, which may dilute the association detected in this study. Moreover, a lack of clinical and basic personal information could limit the interpretation of data and conclusions we can draw from this study. For instance, the high risk of liver disease among OSA patients warrants further verification using other population-based databases.
Conclusions
In conclusion, the risk of liver disease was more than five times higher among OSA patients compared with the comparison group in this large population-based study, particularly for cirrhosis and hepatitis C. A more-than five-fold increased risk of NAFLD and an almost four-fold increased risk of hepatitis B infection among OSA patients were observed, indicating that liver disease is a relevant and very important health issue among OSA patients. Further evaluation of these patients is warranted to ensure the earlier diagnosis and treatment of these health issues. Given the data analysis was limited by a lack of direct evidence for some potentially confounding variables, a prospective cohort study on liver disease among OSA patients is warranted in the future. Moreover, future studies should clarify the multi-factorial nature of liver disease development among OSA patients with sufficient basic 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332
characteristics; it would also be of interest to assess if therapies for OSA, such as continuous positive airway pressure, can decrease morbidity and mortality in the context of liver disease in these patients.
Acknowledgement
The authors would like to thank Mr. Chi-Fung Chen and Ms. Wan-Ting Huang for their assistance in the preparation of this manuscript. This work was supported by grants from China Medical University and Hospital (grant number: CMU 100-SR-66 and CMU 100-TS-09).
References
[1] Amarapurkar D, Kamani P, Patel N, Gupte P, Kumar P, Agal S, Baijal R, Lala S, Chaudhary D, Deshpande A. Prevalence of non-alcoholic fatty liver disease: population based study. Ann Hepatol. 2007;6(3):161-163.
[2] Perz JF, Armstrong GL, Farrington LA, Hutin YJ, Bell BP. The contributions of hepatitis B virus and hepatitis C virus infections to cirrhosis and primary liver cancer worldwide. J Hepatol. 2006;45(4):529-538.
[3] Yan H, Luo JY, Zhou XY. Epidemiological analysis of alcoholic and
nonalcoholic fatty liver in Shanxi and Gansu province. Weichang Bingxue he Ganbing Xue Zazhi. 2007:16:347-350.
[4] Chen CH, Huang MH, Yang JC, Nien CK, Yang CC, Yeh YH, Yueh SK. Prevalence and risk factors of nonalcoholic fatty liver disease in an adult
population of taiwan: metabolic significance of nonalcoholic fatty liver disease in nonobese adults. J Clin Gastroenterol. 2006;40(8):745-752.
[5] Bellentani S, Saccoccio G, Masutti F, et al. Prevalence of and risk factors for 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357
hepatic steatosis in Northern Italy. Ann Intern Med. 2000;132(2):112-117. [6] Browning JD, Horton JD. Molecular mediators of hepatic steatosis and liver
injury. J Clin Invest. 2004;114(2):147-152.
[7] Clark JM. The epidemiology of nonalcoholic fatty liver disease in adults. J Clin Gastroenterol. 2006;40 Suppl 1:S5-10.
[8] McCullough AJ. Pathophysiology of nonalcoholic steatohepatitis. J Clin Gastroenterol. 2006;40 Suppl 1:S17-29.
[9] Clark JM, Diehl AM. Nonalcoholic fatty liver disease: an underrecognized cause of cryptogenic cirrhosis. JAMA. 2003;289(22):3000-3004.
[10]Marchesini G, Bugianesi E, Forlani G, et al. Nonalcoholic fatty liver,
steatohepatitis, and the metabolic syndrome. Hepatology. 2003;37(4):917-923. [11]Li WC, Lee YY, Chen IC, Sun C, Chiu FH, Chuang CH. Association between the
hepatitis B and C viruses and metabolic diseases in patients stratified by age. Liver Int. 2013;33(8):1194-1202.
[12]Wang CS, Chang TT, Yao WJ, Chou P. Comparison of hepatitis B virus and hepatitis C virus prevalence and risk factors in a community-based study. Am J Trop Med Hyg. 2002;66(4):389-393.
[13]Bugianesi E, Leone N, Vanni E, et al. Expanding the natural history of nonalcoholic steatohepatitis: from cryptogenic cirrhosis to hepatocellular carcinoma. Gastroenterology. 2002;123(1):134-140.
[14]Dixon JB, Bhathal PS, O'Brien PE. Nonalcoholic fatty liver disease: predictors of nonalcoholic steatohepatitis and liver fibrosis in the severely obese.
Gastroenterology. 2001;121(1):91-100.
[15]Kohler M, Stradling JR. Mechanisms of vascular damage in obstructive sleep apnea. Nat Rev Cardiol. 2010;7(12):677-685.
358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382
[16]Neuschwander-Tetri BA, Caldwell SH. Nonalcoholic steatohepatitis: summary of an AASLD Single Topic Conference. Hepatology. 2003;37(5):1202-1219.
[17]Poli G. Pathogenesis of liver fibrosis: role of oxidative stress. Mol Aspects Med. 2000;21(3):49-98.
[18]Punjabi NM, Caffo BS, Goodwin JL,et al. Sleep-disordered breathing and mortality: a prospective cohort study. PLoS Med. 2009;6(8):e1000132. [19]Cartwright RD, Knight S. Silent partners: the wives of sleep apneic patients.
Sleep. 1987;10(3):244-248.
[20]Blankfield RP, Hudgel DW, Tapolyai AA, Zyzanski SJ. Bilateral leg edema, obesity, pulmonary hypertension, and obstructive sleep apnea. Arch Intern Med. 2000;160(15):2357-2362.
[21]Sajkov D, Wang T, Saunders NA, Bune AJ, McEvoy RD. Continuous positive airway pressure treatment improves pulmonary hemodynamics in patients with obstructive sleep apnea. Am J Respir Crit Care Med. 2002;165(2):152-158. [22]Sun H, Shi J, Li M, Chen X. Impact of continuous positive airway pressure
treatment on left ventricular ejection fraction in patients with obstructive sleep apnea: a meta-analysis of randomized controlled trials. PLoS One.
2013;8(5):e62298.
[23]Engleman HM, Kingshott RN, Martin SE, Douglas NJ. Cognitive function in the sleep apnea/hypopnea syndrome (SAHS). Sleep. 2000;23 Suppl 4:S102-108. [24]George CF, Smiley A. Sleep apnea & automobile crashes. Sleep.
1999;22(6):790-795.
[25]Ulfberg J, Carter N, Edling C. Sleep-disordered breathing and occupational accidents. Scand J Work Environ Health. 2000;26(3):237-242.
[26]Musso G, Cassader M, Olivetti C, Rosina F, Carnone G, Gambino R. Association 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407
of obstructive sleep apnoea with the presence and severity of non-alcoholic fatty liver disease. A systematic revire and meta-analysis. Obes Rev. 2013;14:417-431. [27]Jun J, Savransky V, Nanayakkara A, et al. Intermittent hypoxia has
organ-specific effects on oxidative stress. Am J Physiol Regul Integr Comp Physiol. 2008;295(4):R1274-1281.
[28]Lin TJ, Lin CL, Wang CS, Liu SO, Liao LY. Prevalence of HFE mutations and relation to serum iron status in patients with chronic hepatitis C and patients with nonalcoholic fatty liver disease in Taiwan. World J Gastroenterol.
2005;11(25):3905-3908.
[29]Tatsumi K, Saibara T. Effects of obstructive sleep apnea syndrome on hepatic steatosis and nonalcoholic steatohepatitis. Hepatol Res. 2005;33(2):100-104. [30]Tanne F, Gagnadoux F, Chazouilleres O, et al. Chronic liver injury during
obstructive sleep apnea. Hepatology. 2005;41(6):1290-1296.
[31]Mishra P, Nugent C, Afendy A, et al. Apnoeic-hypopnoeic episodes during obstructive sleep apnoea are associated with histological nonalcoholic steatohepatitis. Liver Int. 2008;28(8):1080-1086.
[32]Edmison J, McCullough AJ. Pathogenesis of non-alcoholic steatohepatitis: human data. Clin Liver Dis. 2007;11(1):75-104, ix.
[33]Singh H, Pollock R, Uhanova J, Kryger M, Hawkins K, Minuk GY. Symptoms of obstructive sleep apnea in patients with nonalcoholic fatty liver disease. Dig Dis Sci. 2005;50(12):2338-2343.
[34]Turkay C, Ozol D, Kasapoglu B, Kirbas I, Yildirim Z, Yigitoglu R. Influence of obstructive sleep apnea on fatty liver disease: role of chronic intermittent
hypoxia. Respir Care. 2012;57(2):244-249.
[35]Pamidi S, Tasali E. Obstructive sleep apnea and type 2 diabetes: is there a link? 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432
Front Neurol. 2012;3:126.
[36]Louis M, Punjabi NM. Effects of acute intermittent hypoxia on glucose
metabolism in awake healthy volunteers. J Appl Physiol. 2009;106(5):1538-1544. [37]Abdelmalek MF, Diehl AM. Nonalcoholic fatty liver disease as a complication of
insulin resistance. Med Clin North Am. 2007;91(6):1125-1149.
[38]Postic C, Girard J. Contribution of de novo fatty acid synthesis to hepatic
steatosis and insulin resistance: lessons from genetically engineered mice. J Clin Invest. 2008;118(3):829-838.
[39]Sockalingam S, Abbey SE, Alosaimi F, Novak M. A review of sleep disturbance in hepatitis C. J Clin Gastroenterol. 2010;44(1):38-45.
[40]Samarasinghe DA, Tapner M, Farrell GC. Role of oxidative stress in hypoxia-reoxygenation injury to cultured rat hepatic sinusoidal endothelial cells. Hepatology. 2000;31(1):160-165.
[41]Nath B, Szabo G. Hypoxia and hypoxia inducible factors: Diverse roles in liver diseases; 2012.
[42]Gomez-Gonzalez B, Dominguez-Salazar E, Hurtado-Alvarado G, et al. Role of sleep in the regulation of the immune system and the pituitary hormones. Ann N Y Acad Sci. 2012;1261:97-106.
[43]Lange T, Perras B, Fehm HL, Born J. Sleep enhances the human antibody response to hepatitis A vaccination. Psychosom Med. 2003;65(5):831-835. [44]Prather AA, Hall M, Fury JM, et al. Sleep and antibody response to hepatitis B
vaccination. Sleep. 2012;35(8):1063-1069.
[45]Deegan PC, McNicholas WT. Pathophysiology of obstructive sleep apnoea. Eur Respir J. 1995;8(7):1161-1178.
[46]Arnardottir ES, Mackiewicz M, Gislason T, Teff KL, Pack AI. Molecular 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457
signatures of obstructive sleep apnea in adults: a review and perspective. Sleep. 2009;32(4):447-470.
[47] Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence ofsleep-disordered breathing among middle-aged adults. N Engl J Med. 1993;328(17):1230-1235. 458 459 460 461 462 463 464
Table 1
Comparison of baseline characteristics and comorbidities among patients with and without obstructive sleep apnea (OSA)
OSA patients N = 17,374 Comparison group N = 69,496 p-value* Age 0.99 <30 1,184 (6.8%) 4,759 (6.9%) 30-59 7,106 (40.9%) 28,404 (40.9%) >=60 9,084 (52.3%) 36,333 (52.3%) Gender 1.00 Male 11,852 (68.2%) 47,408 (68.2%) Female 5,522 (31.8%) 22,088 (31.8%) Urbanization area 1.00 Area1 6,086 (35.0%) 24,344 (35.0%) Area2 3,861 (22.2%) 15,444 (22.2%) Area3 2,941 (16.9%) 11,764 (16.9%) Area4 2,530 (14.6%) 10,120 (14.6%) Area5 1,956 (11.3%) 7,824 (11.3%)
Income per month (NTD) 1.00
Less than 10,000 7,554 (43.5%) 30,216 (43.5%) 10,000~20,000 5,601 (32.2%) 22,404 (32.2%) 20,001~30,000 1,270 (7.3%) 5,080 (7.3%) 30,001~40,000 893 (5.1%) 3,572 (5.1%) 40,001~50,000 949 (5.5%) 3,796 (5.5%) Over 50,000 1,107 (6.4%) 4,428 (6.4%) Comorbidities Hypertension 4,314 (24.8%) 1,221 (1.8%) < 0.001 Diabetes 3,414 (19.7%) 2,616 (3.8%) < 0.001 Hyperlipidemia 1,621 (9.3%) 1,910 (2.8%) < 0.001 Congestive heart failure 2,156 (12.4%) 439 (0.6%) < 0.001 Atrial fibrillation 818 (4.7%) 170 (0.2%) < 0.001 *: by chi-square test 465 466 467 468 469
Table 2
Incidence of all liver diseases, non-alcoholic fatty liver disease, cirrhosis, hepatitis B, and hepatitis C among patients with and without OSA
OSA patients N = 17,374
Comparison group N = 69,496
p-value*
All liver diseases 2,008 (11.6%) 2,049 (3.0%) < 0.001 Non-alcoholic fatty
liver disease 1,217 (7.0%) 1,226 (1.8%) < 0.001
Cirrhosis 310 (1.8%) 249 (0.4%) < 0.001
Hepatitis B 255 (1.5%) 352 (0.5%) < 0.001
Hepatitis C 182 (1.1%) 153 (0.2%) < 0.001
*by chi-square test 470 471 472 473 474 475 476
Table 3
Cox proportional-hazards regression estimated hazard ratios (HR) in all liver diseases, non-alcoholic fatty liver disease, cirrhosis, hepatitis B and C among patients with and without OSA Crude HR (95%CI) p-value Adjusted HR (95%CI)* p-value
All liver diseases 6.24 (5.82-6.70) < 0.001 5.52 (5.12-5.96) < 0.001
Non-alcoholic fatty
liver disease 5.83 (5.34-6.37) < 0.0001 5.29 (4.81-5.81) < 0.001
Cirrhosis 8.84 (7.23-10.80) < 0.0001 7.50 (6.02-9.34) < 0.001
Hepatitis B 3.89 (3.27-4.62) < 0.0001 3.71 (3.09-4.46) < 0.001
Hepatitis C 8.47 (6.55-10.96) < 0.0001 7.19 (5.41-9.55) < 0.001
*: adjusted for age, gender, and comorbidities 477 478 479 480 481 482 483
Figure Legend
Fig 1. The estimated cumulative incidence of liver disease among patients with and without OSA by Kaplan–Meier analysis. A. All liver diseases; B. Non-alcoholic fatty liver disease; C. Cirrhosis; D. Hepatitis B; E. Hepatitis C.
484 485 486 487 488 489
