This was a nationwide, population-based, retrospective study using the National Health Insurance (NHI) claims database in Korea (from 2010 to 2019). The NHI system is one of the largest national insurance services covering more than 98% of the Korean population as it is mandated by law. The database contains claims information on all diseases, providing reliable estimates of prevalence and comorbid diagnosis of certain diseases.

In the NHI database, diagnoses are recorded according to the International Classification of Diseases, 10th revision (ICD-10) (Table 1). Patients were included only if at least 3 documented physician contacts were present under the diagnosis of OSA as the principal diagnosis. Demographic information such as age and sex were obtained from the information provided in the first visit of the diagnosis.

Medical comorbidities highly associated with OSA were extracted using the ICD-10 revision diagnosis code; ischemic heart disease, stroke, heart failure, hypertension, dyslipidemia, diabetes, chronic obstructive pulmonary disease (COPD), chronic kidney disease (CKD), liver cirrhosis, asthma, gastroesophageal reflux disease (GERD). Patients were defined to have a certain comorbidity if it was recorded as a primary diagnosis on at least three clinical visits.

To represent the general population, patients who were recorded to have undergone appendectomy between 2010 and 2019 were included as the control group; however, patients who had a concomitant diagnosis of OSA were excluded.

The baseline characteristics of patients in the OSA and the control group were compared using an independent t-test for continuous variables and the chi-squared test for categorical variables. The Cox proportional hazard ratio (HR) model was used to estimate the effect of OSA on the incidence of medical disorders. Both crude HR and adjusted HR along with 95% confidence intervals (CI) are reported. Subgroup analyses disaggregated by sex and age were also conducted. All statistical analyses were performed using SAS Enterprise Guide version 7.1 (SAS Institute, Cary, NC, USA). This study was approved by the Institutional Review Boards of St. Vincent’s Hospital (VC11QISE0225, Suwon, Korea).

The demographic characteristics of the study population are summarized in Table 2. In total, 103,785 patients with OSA were identified. Among patients who underwent appendectomy, 207,570 age- and sex-matched patients without OSA were included in the control group. The mean age in the OSA group and the control group were 58.36±14.12 years and 57.25±20.31 years, respectively. Male patients accounted for 83.06% and the most common age was 40-59 years in both the OSA and the control groups. There was a notable difference between the OSA group and the control group in terms of insurance coverage. Classification based on insurance type revealed that 3.11% of OSA patients were recipients of medical aid, while this percentage was higher in the control group (5.01%).

Table 3 presents the crude HR and adjusted HR (corrected for age, sex, and socioeconomic status) for medical comorbidities in the OSA group. The prevalence of ischemic heart disease was highest even after adjusting for hypertension, dyslipidemia, and diabetes (adjusted HR, 2.109; 95% CI, 2.088-2.130). The adjusted HR for stroke (1.462; 95% CI, 1.442-1.482) was also higher after further adjusting for hypertension, dyslipidemia, and diabetes mellitus. Heart failure demonstrated a slightly higher adjusted HR (1.188; 95% CI, 1.160-1.218) compared to the control group.

The adjusted HRs for hypertension (1.757; 95% CI, 1.745-1.770), dyslipidemia (1.587; 95% CI, 1.566-1.608), and diabetes (1.561; 95% CI, 1.546-1.575) were also higher in the OSA group compared with the control group.

Additionally, patients with OSA had a significantly higher risk of CKD (adjusted HR, 1.518; 95% CI, 1.483-1.554) and GERD (adjusted HR, 1.445; 95% CI, 1.434-1.455).

However, patients with OSA had a significantly lower risk of COPD (adjusted HR, 0.824; 95% CI, 0.804-0.844). Moreover, liver cirrhosis and asthma were not significantly associated with OSA.

When stratified by sex, the risk of increased medical comorbidities in OSA patients showed a slight variation. OSA was associated with an increased risk of hypertension, dyslipidemia, diabetes, ischemic heart disease, stroke, heart failure, CKD, and GERD in both sexes (Table 4). However, OSA was associated with an increased risk of COPD in female (adjusted HR, 1.854; 95% CI, 1.512-2.274), but a decreased risk of COPD in male (adjusted HR, 0.767; 95% CI, 0.694-0.848). A similar pattern was observed with respect to the risk of asthma in OSA patients (adjusted HR, 1.349; 95% CI, 1.278-1.424 in female and adjusted HR, 0.544; 95% CI, 0.525-0.565 in male). The risk of liver cirrhosis was significantly higher only in female with OSA (adjusted HR, 2.064; 95% CI, 1.550-2.748).

Subgroup analysis by age revealed that OSA was a risk factor for dyslipidemia, diabetes, and GERD in all age groups (Table 5). The increased risk of hypertension and ischemic heart disease in OSA patients was significant in all age groups, except for the 20-39 age group. The risk of stroke and heart failure in OSA patients was higher in the 40-59 and 60-79 years age groups.

In this study, OSA was identified as a risk factor for ischemic heart disease and stroke even after adjusting for diabetes, hypertension, and dyslipidemia. Notably, the adjusted HR was highest for ischemic heart disease. Additionally, patients with OSA also showed a higher risk of heart failure.

Several clinical studies, including a nationwide study based on the US health claims database, have consistently identified a close relationship between OSA and cardiovascular diseases, such as congestive heart failure, stroke, arrhythmia, and ischemic heart disease [11]. Our results align with the previous studies, indicating a significantly increased risk of cardiovascular diseases in Korean patients with OSA.

A nationwide database study in Denmark (2000-2012) also showed an increased risk of heart failure in patients with OSA [12]. The increased risk of heart failure in Danish patients with sleep apnea was observed across all age groups, which is similar to the pattern observed in Korean patients with OSA. An important finding was that application of continuous positive airway pressure (CPAP) significantly reduced the risk of heart failure in patients aged ≥60 years.

Additionally, Peker et al. [13] revealed that patients with coronary disease who had comorbid OSA showed a higher mortality rate (38% vs. 9%; p=0.018). Subsequently, a study conducted in France investigated patients diagnosed with both OSA and coronary artery disease [14]. Patients who received treatment for OSA were found to have a significantly lower risk of cardiovascular death, acute coronary syndrome, hospitalization due to heart failure, and coronary revascularization rate (HR, 0.24; 95% CI, 0.09-0.62) compared to the untreated group.

These studies suggest that treating OSA may help prevent the onset, progression, and exacerbation of cardiovascular disease in OSA patients. Considering the high risk of cardiovascular diseases in Korean patients with OSA, treatment of OSA seems crucial to prevent cardiac events and decrease mortality rates.

In this study, Korean patients with OSA had a higher risk of metabolic syndrome-related diseases, which is consistent with the previous studies including a US nationwide cohort study.

Hypoxia caused by airway obstruction triggers a systemic inflammatory reaction. Repeated hypoxic damage in adipose tissue triggers chronic inflammation, increasing the level of inflammatory cytokines, such as tumor necrosis factor α and interleukin-6 [15]. Systematic inflammatory reaction induces vascular endothelial damage, which in turn causes insulin resistance and glucose intolerance, leading to the onset of metabolic diseases [16].

In this study, the HR for CKD in OSA patients was higher than that in the general population in Korea. This is similar to the results of a large-scale cohort study in the US, where 1.66% of OSA patients were found to have comorbid CKD [11].

COPD, on the other hand, exhibited significantly lower HR in patients with OSA. The results contradict the findings of a population-based study in the United States, where the odds ratio for chronic pulmonary disease was significantly higher in patients with OSA [11]. A cohort study conducted in 2022 revealed that 77% of COPD patients in Korea had the non-chronic bronchitis type COPD [17]. Higher prevalence of non-chronic bronchitis type COPD, along with the lower average body mass index (BMI) in Asian ethnicity could explain the lower HR for COPD in Korean patients with OSA [18].

The HR for GERD, on the other hand, appeared to be high in Korean populations with OSA, aligning with the US studies.

In a subgroup analysis by sex, the HR for liver cirrhosis in male OSA patients was not significantly different compared to the general population, similar to the unstratified analysis. However, in female OSA patients, the HR for liver cirrhosis was significantly higher than that in the matched control population.

Chronic intermittent hypoxia in patients with OSA seems to increase the risk of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis (NASH) by inducing oxidative stress in the liver [19]. Yang et al. [20] reported that in females diagnosed with NASH, the post-menopausal reduction in estrogen level leads to liver fibrosis at a level similar to that seen in male NASH patients. Considering the age of the subjects in this study (approximately 70% of patients over the age of 40 years), it is tempting to speculate that OSA may accelerate the progression to liver cirrhosis in post-menopausal women.

Furthermore, the HRs for COPD and asthma compared to the matched control population were significantly higher in females. This finding raises questions regarding the relationship between chronic pulmonary disease and OSA. Yet, these results are somewhat consistent with the findings of the US study—our study had a sex ratio of 5:1, while the US NHI study had a 1:1 sex ratio—which indicated a high comorbidity rate of chronic pulmonary disease.

Bahammam et al. [21] reported that in female OSA patients, the rapid eye movement (REM)-predominant OSA type is associated with a higher incidence of asthma. Females, with a generally smaller pharyngeal size than males, are prone to pharyngeal wall collapse during REM sleep due to atonia [22]. Similarly, there is a potential increase in comorbid COPD in female OSA patients due to heightened sensitivity to REM sleep. Further study utilizing additional data such as polysomnography and BMI in female OSA patients is warranted.

When stratified by age, the HRs for ischemic heart disease, stroke, hypertension, and CKD increased with age in the OSA group. Considering that the average age at onset of OSA is 36 years [23], we can infer that as the morbidity period of OSA lengthens, the vascular burden is also increased. The lack of significant differences in the HRs for CKD and stroke in OSA patients aged ≥80 years compared to the general population could potentially be biased due to a higher mortality rate [24].

Some limitations of this study should be considered while interpreting the results. First, this study is based solely on insurance benefit-based patient information. The gold standard for the diagnosis of OSA is polysomnography. However, in Korea, polysomnography was not covered by insurance until September 2018. Therefore, there may have been more cases of OSA that were not covered by NHI due to the high cost of polysomnography, which may have introduced an element of selection bias. Second, a study found that the consistency of medical records with residential diseases of claims data was approximately 70%. It is difficult to completely exclude the influence of physician subjectivity related to diagnoses and prescriptions. Third, The ICD-10 diagnosis does not distinguish OSA severity. Indeed, the severity of OSA is a potential confounding factor that may influence the incidence of related comorbidities. Finally, our data lack information about BMI and smoking status, which can influence both OSA and the occurrence of comorbid disease. There may be other confounding factors that were not addressed in this study.