Management and Risk Reduction of Rheumatoid Arthritis in Individuals with Obstructive Sleep Apnea: A Nationwide Population-Based Study in Taiwan

Abstract

Study Objectives:

To explore associations between obstructive sleep apnea (OSA) and autoimmune diseases and evaluate whether OSA management reduces the incidence of autoimmune diseases.

Methods:

This was a retrospective cohort study using nationwide database research. The data was from 105,846 adult patients in whom OSA was diagnosed and recorded in the Taiwan National Health Insurance Research Database between 2002 and 2011 were the patients were analyzed retrospectively. Patients with antecedent autoimmune diseases were excluded. A comparison cohort of 423,384 participants without OSA served as age- and sex-matched controls. Multivariable Cox regression analysis was performed on both cohorts to compute risk of autoimmune diseases during follow-up. Time-dependent OSA treatment effect was analyzed among patients with OSA. There were no interventions.

Results:

Among patients with OSA, overall risk for incident autoimmune diseases was significantly higher than that in controls (adjusted hazard ratio [HR] = 1.95, 95% confidence interval [CI] = 1.66–2.27). Risk for individual autoimmune diseases, including rheumatoid arthritis (RA), Sjögren syndrome (SS), and Behçet disease, was significantly higher in patients with OSA than in controls (HRs [95% CI]: RA 1.33 [1.03–1.72, SS 3.45 [2.67–4.45] and Behçet disease 5.33 [2.45–12.66]). Increased risk for systemic lupus erythematosus (HR 1.00 [0.54–1.84]) and systemic sclerosis (HR 1.43 [0.51–3.96]) did not reach statistical significance. Patients with OSA receiving treatment had an overall reduced risk of RA and other autoimmune diseases (time-dependent HRs [95% CI]: 0.22 [0.05–0.94] and 0.51 [0.28–0.92], respectively).

Conclusions:

Patients with OSA are associated with higher risk for developing RA, SS, and Behçet disease. OSA management is associated with reduced risk of RA.

INTRODUCTION

Obstructive sleep apnea (OSA) is a common disorder characterized by recurring total or partial upper airway collapse with apnea/hypopnea and recurrent hypoxia during sleep.^1–3 OSA is associated with many metabolic, endocrine, and especially cardiovascular diseases.^1–3 Patients with OSA are found to have elevated systemic proinflammatory cytokines and inflammatory markers, including tumor necrosis factor-alpha (TNF-α), C-reactive protein (CRP), and interleukin-1b (IL-1b), IL-6, and IL-17.⁴ Therefore, OSA is thought to be a systemic inflammatory disease, and treatment of OSA, such as continuous positive airway pressure (CPAP), could suppress the systemic inflammation in patients with OSA.^5–7

Autoimmune diseases are multifactorial, resulting from dysregulation of the immune system and autoimmune-mediated damage to normal tissue. Triggering factors in a host predisposed by genetic background may result in an aberrant immunological response.⁸ Several different proinflammatory cytokines are known to participate in the pathogenesis of various autoimmune diseases, especially TNF-α.^9–11 Epidemio-logic evidence evaluating the association between OSA and susceptibility to autoimmune diseases is limited to systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), anky-losing spondylitis (AS), and psoriatic arthritis (PsA).^12,13 Other major systemic autoimmune diseases such as Sjögren syndrome (SS), systemic sclerosis (SSc), and systemic vasculitis have not been studied thoroughly.

We hypothesized that OSA would predispose patients to the development of autoimmune diseases and that treatment of OSA may reduce the risk of developing autoimmune diseases. Therefore, this cohort study aimed to explore the relationship between untreated OSA, managed OSA, and risk of autoimmune diseases by utilizing a nationwide database with a large representative sample of patients in Taiwan.

METHODS

Data Sources

The National Health Insurance (NHI) program was initiated in 1995 to provide comprehensive health care for all citizens in Taiwan; enrollment in this program is mandatory and the nationwide coverage rate was approximately 99% by the end of 2006.¹⁴ The NHI program provides coverage for comprehensive medical care, including outpatient, inpatient, emergency, dental, and Traditional Chinese Medicine services. The bureau of NHI in Taiwan maintains a research-oriented database through the Health and Welfare Statistics Application Center of the Ministry of Health and Welfare so that investigators can obtain information for research purposes. The NHI database includes information about basic patient characteristics; diagnoses for all patients admitted to hospitals and treated as outpatients; and detailed claims data for clinical examinations, disease management, and drug prescriptions. The diagnostic codes used are based on the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). As such, the NHI database is one of the largest nationwide population-based data sources in the world. For the current study we collected the data of the Taiwanese population from 2002 to 2012. Because the NHI dataset consists of deidentified secondary data for research purposes, the study was exempted from a full review by the Institutional Review Board of our hospital and, because patients remained anonymous, informed consent was waived.

Study Population and Control Cohort

We conducted a retrospective cohort study using patient data collected from January 1, 2002 to December 31, 2012. Using the International Classification of Diseases, Ninth Revision, Clinical Modification codes (ICD-9-CM codes; 2001 revision), we identified patients with sleep apnea (ICD-9-CM: 327.23, 780.51, 780.53, 780.57) from among 23 million NHI beneficiaries receiving a diagnosis after undergoing polysomnography (ICD-9-CM: 17008A, 17008B) between 2002 and 2011. Those presenting with surveyed autoimmune diseases prior to diagnosis of OSA and diagnosis of OSA before 18 y of age were excluded from the study. Subjects without OSA were used as an age- and sex-matched cohort (nonexposure subjects) from among the 23 million NHI beneficiaries recorded between 2002 and 2011 using the same exclusion criteria. OSA patient and comparison controls were frequency matched at a 1:4 ratio by age (in 5-y bands), sex, and index year. Charlson Comorbidity Index scores were calculated to represent global comorbidity severity at baseline.¹⁵

Outcomes

The primary outcome was the occurrence of autoimmune diseases. Several autoimmune diseases were defined as catastrophic illnesses in the NHI and that certification requires precise fulfillment of related classification criteria as follows: American College of Rheumatology (ACR) 1997 revised criteria for systemic lupus erythematosus (SLE, ICD-9-CM: 710.0)¹⁶; American Rheumatism Association 1987 revised criteria for rheumatoid arthritis (RA, ICD-9-CM: 714.0)¹⁷; ACR criteria for systemic sclerosis (SSc, ICD-9-CM: 710.1)¹⁸; American-European Consensus Group 2002 revised criteria for Sjogren's syndrome (ICD-9-CM: 710.2)¹⁹; Bohan and Peter 1975 criteria for polymyositis and dermatomyositis (ICD-9-CM: 710.3)^20,21; International Study Group 1990 criteria for Behçet disease (ICD-9-CM: 136.1)²²; and ACR 1990 criteria for temporal arteritis (ICD-9-CM: 443.1),²³ granulomatosis polyangiitis (GPA, ICD-9-CM: 446.4),²⁴ and Takayasu arteritis (ICD-9-CM: 446.7).²⁵ After strict verification by a review committee, patients are enrolled in the registry of catastrophic illnesses. The main end point was development of autoimmune diseases, defined by catastrophic illness certification of these diseases. All cases were followed until the development of autoimmune diseases, death, withdrawal from the insurance program, or the end of the study period in 2012. Incidence rates (per 100,000 person-years) and incidence rate ratios were analyzed. The brief summary of the classification criteria of major autoimmune diseases was shown in Table 1.

We also investigated the effects of OSA treatment on the risk of developing autoimmune diseases. The patients with OSA were divided into two groups, those with and without OSA treatment. Treatment included surgery (pharyngeal or nasal surgery) and continuous positive airway pressure (CPAP). The types of pharyngeal surgery included were tonsillectomy, adenoidectomy, adenotonsillectomy, and uvulopalatopharyngoplasty. The nasal surgeries included septoplasty, turbinoplasty, stomatoplasty, and laser turbinoplasty.

Statistical Analyses

All statistical analyses were performed using the SAS 9.3 statistical package (SAS Institute Inc., Cary, NC, USA). Differences in continuous variables between the two cohorts were evaluated using the Student t test. Differences in categorical variables of the potential confounders between the two cohorts were evaluated using Pearson χ² test.

A Cox regression model was employed for multivariate adjustment, including age groups and sex. The cumulative incidence of overall and individual autoimmune diseases were compared by Kaplan-Meier method and log-rank test. When focusing on the OSA group, we evaluated whether patients had obtained regular OSA treatment annually during our study period, including surgery and/or CPAP. Treatments of patients with OSA were considered as time-dependent effects in the Cox proportional hazard regression model. The HRs of the treatments were explained as follows: in any given year, if patients received any OSA treatment, the risk of autoimmune diseases would increase on average (HR > 1) / decrease (HR < 1) compared with risk in patients not receiving OSA treatment. The results of all statistical tests were considered significant if the two-sided P value was ≤ 0.05.

RESULTS

Baseline Characteristics of Sleep Apnea and Comparison Cohorts

A total of 113,478 sleep apnea patients were identified from among the 23,000,000 sampling cohort recorded between January 2002 and December 2011. After excluding 398 patients who had autoimmune diseases prior to diagnosis of sleep apnea and 6,641 patients younger than 18 y, the data of 105,846 sleep apnea patients were included in analysis. The median follow-up period of patients with OSA was 4.11 (interquartile range, 3.85) y. The flowchart of study population selection is shown in Figure 1.

The demographic and clinical characteristics of all included participants, including 105,846 OSA patients and 423,384 controls, are shown in Table 2. The controls were well matched for age and sex (both P > 0.9). The mean age was 47.26 ± 13.8 y, and the majority of participants in the two cohorts were male (76.48%). The mean follow-up period was 4.6 ± 2.59 y. Patients with OSA had higher Charlson Comorbidity Index scores, higher incidence of obesity, and received more OSA-associated therapy than did controls.

Higher Incidence of Incident Autoimmune Diseases among Sleep Apnea Patients than among Controls

Among 529,230 subjects, 716 patients (0.14%) received a diagnosis of autoimmune disease, including 234 (0. 22%) from the OSA cohort and 482 (0.11%, P < 0.001) from the matched non-OSA control cohort.

Kaplan-Meier estimates of autoimmune disease-free survival revealed a significantly higher incidence rate (P < 0.001) in the OSA cohort compared to that of the matched control cohort, as shown in Figure 2. The overall cumulative incidence of autoimmune diseases in the entire OSA cohort was significantly higher than that in the matched control cohort, with an aHR of 1.94 (95% CI 1.66–2.27, P < 0.001). For respective autoimmune diseases, the incidence rate of RA (aHR 1.33, 95% CI: 1.03–1.72, P < 0.01), Sjögren disease (aHR 3.54, 95% CI: 2.75–4.56, P < 0.001) and Behçet disease (aHR 5.33, 95% CI: 2.25–12.66, P < 0.001) were significantly higher in the OSA cohort than in the non-OSA control cohort, as shown in Table 3. Kaplan-Meier analyses also revealed that patients with OSA had a higher risk for development of autoimmune diseases (log-rank test, P < 0.001, Figure 2).

Treatment of OSA Patients was Associated with Overall Risk Reduction in RA and Other Autoimmune Diseases

The time-dependent treatment effects among patients with OSA and the risk of developing autoimmune diseases are shown in Table 4. The overall risk of the development of auto-immune disease in patients treated for OSA was significantly lower than in patients with OSA who did not receive treatment (HR:0.51 [0.28–0.92]; P < 0.05). OSA management can reduce the risk of RA; however, no significant changes were found in SS risk. For respective autoimmune diseases, the adjusted HR of time-dependent treatment effects were: RA (0.22[0.05–0.94]; P < 0.05) and SS (0.88[0.42–1.83]; P > 0.05). The HR for Behçet disease is not available; the treatment effects of risk of Behçet disease in patients with OSA could not be calculated due to small case numbers, and further investigation is needed.

DISCUSSION

To the best of our knowledge, the current study evaluated data from the largest cohort used to investigate epidemiological associations between OSA and the development of autoimmune diseases. Because the whole database of all NHI enrollees (23 million people) was used rather than a random sample of one million subjects from the registry, this study has greater power than previous studies using the same nationwide database.^12,13 Therefore, this is the first study that was able to demonstrate the effect of OSA management on the incidence of autoimmune diseases in an OSA population. In addition, patients were enrolled in the registry of catastrophic illnesses after strict verification by a review committee, so defining autoimmune diseases as catastrophic illness in the current study makes the diagnosis even more accurate than in previous studies. Although the etiology of autoimmune diseases is not well understood, results of the current study suggest that OSA may be a risk factor for development of autoimmune diseases and management of OSA may reduce that risk.

Previously published studies have mainly focused on patients with SLE, RA, AS, PsA, and Ps, and showed that OSA was associated with increased risk for the development of RA, PsA, and Ps.^12,13 However, whether or not individuals with OSA are more susceptible to the development of other major autoimmune diseases remains unanswered. In the current study, we surveyed most of the major autoimmune diseases, including SLE, RA, SS, systemic sclerosis, inflammatory myositis, and vasculitis. Consistent with previous studies,¹² patients with OSA in the current study also had a higher risk of the development of RA. This is the first study to demonstrate the link between OSA and subsequent occurrence of SS and Behçet disease by using a large, nationwide database.

The mechanisms underlying the epidemiological association between OSA and autoimmune diseases remain unclear. One possible theory is that OSA may contribute to systemic inflammation. Numerous studies have demonstrated elevated levels of circulating proinflammatory cytokines, chemokines, and adhesion molecules in patients with OSA in comparison with those in matched control patients.^{4,9–11,26,27} In particular, the potent proinflammatory cytokine TNF-α has been evaluated by several case-control studies that have consistently shown elevated levels in patients with OSA compared with those in control patients.^4,9–11 TNF-α is the most influential inflammatory cytokine in the pathogenesis of RA, and the effectiveness of anti-TNF-α therapy has been demonstrated in RA.²⁸ Also, in one study of associations between OSA and the autoimmune diseases psoriatric arthritis and ankylosing spondylitis, the presence of OSA in patients receiving TNF-α inhibitors was significantly lower than OSA prevalence in patients not receiving this treatment.²⁹ Additionally, a relationship is shown between oxidative stress and dental diseases in patients with OSA, particularly periodontitis.^30,31 Considerable evidence has demonstrated that periodontitis is associated with RA and may be a potential risk factor.^32,33 Periodontal therapy reduces the severity of active RA.³⁴

Adhesion molecules and vascular growth factor may be involved in the pathogenesis of SS.³⁵ In addition, elevated inter-cellular adhesion molecule-1 levels were found both in patients with OSA⁴ and in minor salivary gland mononuclear cells in patients with SS.³⁶ Patients with primary SS may have increased saliva surface tension but are not shown to have abnormal upper airway collapsibility via the upper airway collapsibility index.³⁷ Results of an observational study of 28 pSS patients demonstrated that these patients had twice the frequency of apnea/hypopnea compared to non-pSS control patients, suggesting that OSA may be a useful therapeutic target to improve both pSS and quality of life in these patients.³⁸

Recently another important inflammatory cytokine, IL-17, was found to be associated with several autoimmune diseases, including SLE, RA, inflammatory bowel disease, and Behçet disease.³⁹ Systemic inflammation may be amplified by the overt production of IL-17 through the activation of autoantibody production and an enhanced immune response at inflammation sites.^40,41 Briefly, OSA may induce systemic inflammation, including increased levels of TNF-α, intercellular adhesion molecule-1, and IL-17, thereby predisposing patients with OSA to the occurrence of autoimmune diseases. It follows, then, that risk reduction for autoimmune diseases may be related to reducing levels of systemic proinflammatory cytokines and inflammatory markers known to be elevated in OSA, including TNF-α, CRP, and interleukins IL-1b, IL-6, and IL-17.^4,9–11

Oxidative stress is a potential causal factor for autoimmune hemolytic anemia and SLE.⁴² It has been reported that oxidative stress increases in rheumatoid inflammation due to impaired antioxidant systems caused by free radicals, which have a role in the etiology of RA.⁴³ In the current study, the significant risk reduction of RA was found in patients with OSA receiving treatment. One possible explanation is that treatment of OSA may reduce related inflammation and oxidative stress. Much evidence demonstrates that treatment of OSA, including CPAP or surgery, can reduce inflammation and oxidative stress.^7,44,45 A recent meta-analysis of 35 studies involving 1,985 patients with OSA also confirmed that CPAP therapy could partially suppress systemic inflammation and there are differences among the various inflammatory markers such as CRP, IL-6, IL-8, and TNF-α.⁷

However, CPAP is underused as a therapy for OSA even though investigators have demonstrated that CPAP may reduce the risk of atrial fibrillation or stroke.⁴⁶ Results of the current study also revealed that CPAP may reduce the risk of RA in patients with OSA depending on the extent of decrease of inflammatory mediators and oxidative stress.

In addition, immortal time bias can sometimes generate an illusion of treatment effectiveness in observational studies. To avoid immortal time bias, we used the Cox proportional hazard model to analyze the time-dependent effect of the treatment of patients with OSA.

One dominant strength of the current study is its matched-control cohort study design. A powerful conclusion can be made based on the large sample size of subjects in this study. In addition, the NHI is a single-payer system and a mandatory insurance program, so most events could be traced with a minimum of referral bias.

The current study has several limitations. First, we utilized a nationwide medical registry database to conduct a population cohort study with retrospective analysis of data. Diagnoses of OSA and autoimmune diseases that rely on administrative claims data may be less accurate than diagnoses made by standardized criteria, which may be an inherent weakness for all database research. Nevertheless, many highly regarded studies in various medical disciplines have been published using NHIRD data for analysis.⁴⁷ In the current study, we selected only patients who received an OSA diagnosis following polysomnography. This served to prevent tentative diagnosis and to improve the diagnostic accuracy. Moreover, the potential for misclassification is possible because OSA still remains underdiagnosed. However, if more cases of OSA were included in the study group, the result would be more compelling to verify our hypothesis. Second, the NHIRD database lacks family history, and certain anthropo-metric and lifestyle information, including body mass index, body weight, smoking status, and alcohol consumption. Even though the etiology of most autoimmune diseases is largely unknown, cigarette smoking, for instance, has been reported to predispose individuals to the occurrence of RA. Body weight change may influence the symptoms of OSA; whether weight change is associated with the development of autoimmune diseases remains uncertain. Weight was not included as a factor in the current study; i.e., the OSA and control groups were not matched for weight. Third, the NHIRD database did not allow us to determine the severity of OSA among patients in the OSA cohort. In addition, CPAP compliance also cannot be determined from the registry data. Consequently, it is unclear whether patients with OSA who received long-term CPAP had lower risk of RA compared to patients with similar severity of OSA but with short-term CPAP use. Yang et al.⁴⁸ had reported that the rate of CPAP adherence in Taiwan is 64%, which is similar to rates reported in previous studies.^49–51 However, it is still possible that some patients receiving CPAP treatment were nonadherent and this would suggest that we underestimated the benefits of CPAP. Decision-making regarding CPAP was carried out by physicians without a uniform rule, and a selection bias may exist. Last, autoimmune diseases are complex and heterogeneous; genetic and environmental factors should still be considered, but were not included in this study, which may call for further research to elucidate their differential influence on the development of autoimmunity.

CONCLUSIONS

Patients with OSA are associated with a higher risk of the development of autoimmune diseases, including RA, SS and Behçet disease. The significant risk reduction of RA incidence was associated with treatment of OSA patients. Further investigation is required to elucidate the mechanism underlying the link between OSA and risk of development of autoimmune diseases.

REFERENCES