1. Background
Obstructive sleep apnea (OSA) is a common breathing disorder among individuals, and its prevalence ranges from 0.1% to 13% (1). The main risk factors of the disease are older age, male gender, and obesity (2). OSA is defined by repeated episodes of airway collapses during sleep time, which can be complete (apnea) or incomplete (hypopnea) (3). Apnea-hypopnea index (AHI) is a measure to determine the severity of the disease. AHI ranges of 5 - 14, 15 - 29, and ≥ 30 refer to mild, moderate, and severe disease conditions (4), respectively. The burden of the disease is high due to both healthcare costs related to OSA alone and its contribution as an independent risk factor for metabolic, cardiovascular, and psychiatric disorders (5), such as arterial hypertension, heart failure, and stroke, as well as several metabolic dysfunctions such as DM, glucose intolerance, and insulin resistance (6-8).
It is well known that the morbidity and mortality rate is higher in patients with severe OSA compared with those with mild to moderate disease, especially in patients with underlying diseases, such as arterial hypertension, coronary artery disease, and stroke (7, 9). There is convincing data to support early treatment effects on the reduction of symptoms and cardiovascular risks in such patients (10). Hence, early diagnosis and management is particularly important in severe cases. However, severe OSA includes a wide range of patients with AHI ≥ 30, and limited studies investigated clinical and polysomnographic characteristics of different subsets of this group. Jurcevic et al. (9), compared patients with AHI ≥ 60 to those with less severe disease and suggested that patients with AHI ≥ 60 are more likely older obese males with lower total sleep time with < 90% O2 saturation (T < 90%). They also demonstrated that this group of patients has an overall higher mortality rate (9). In addition, Rey de Castro et al., also claimed that arterial HTN, neck circumstance, age, and over 10% of T < 90% are associated with AHI ≥ 100 (11). Yet, no comparisons with other associated diseases and polysomnographic characteristics of patients with AHI ≥ 100 are reported. Therefore, it is beneficial to define the prognostic variables for this subset of severe OSA.
2. Objectives
The current retrospective study aimed at investigating whether associated diseases and/or risk factors are higher in patients with AHI ≥ 100 than those with 60 > AHI > 100 was performed.
3. Methods
3.1. Study Design and Participants
The current retrospective study was conducted on adult patients referring to Bamdad Sleep Clinic in Isfahan, Iran, from 2018 to 2020 for the evaluation of suspected OSA. Patients undergoing eight hours of split night polysomnography (PSG) and diagnosed with very severe sleep apnea (AHI ≥ 60 per hour) were selected. Patients with other medical neurodegenerative or psychiatric problems that might affect their sleep quality, such as congestive heart failure, myasthenia gravis, or major depressive disorder, were excluded. The AASM (American Academy of Sleep Medicine) Sleep Scoring Manual version 2.6 was used to score the subjects. Two OSA subgroups were defined as very severe sleep apnea (very severe OSA) (60 ≤ AHI < 100) and extremely severe sleep apnea (extreme OSA) (AHI ≥ 100).
3.2. Procedure and Variable Assessments
In the current study, standard PSG equipment (Embla® N7000 Recording System PSG) for OSA diagnosis was utilized by which parameters, such as electroencephalography, electro-oculography , electromyography, electrocardiography, oxygen saturation, respiratory effort, oronasal airflow and snoring sounds, and body position, were recorded.
The reported parameters included AHI, which is the average number of apnea and hypopnea episodes per hour; minimal O2 saturation, which is the lowest SpO2 during sleep; T < 90, which is the total time of sleep with less than 90% O2 saturation, and central apnea percentage, which is the ratio of central apneas to total apneas in one hour. Sleep scoring criteria for the staging system were determined according to AASM (12).
Clinical characteristics were obtained from the medical records of patients. In some cases, there was no documented medical history, and the data were obtained by self-reporting of drugs taken for hypertension, hypothyroidism, hyperlipidemia, or diabetes.
3.3. Statistical Analysis
Statistical analysis was performed using SPSS version 18 software. Numerical measures were expressed as mean ± standard deviation (SD). Clinical and polysomnographic findings were compared between the two groups using t-test for quantitative and chi-square test for categorical variables. A P-value of < 0.05 was considered the level of significance.
The current study was approved by the Ethics Committee of the Institutional Board Review of Isfahan University of Medical Sciences, Isfahan, Iran (ethical code: IR.MUI.MED.REC.1398.625).
4. Results
Out of the total 114 patients with OSA included in the study, 19 had AHI ≥ 100 and 95 60 ≤ AHI < 100, with the mean AHI of 113.95 in extreme OSA and 75.66 in severe OSA groups.
The mean age of participants was 60.36 years, and the mean BMI 36.27 kg/m2. The baseline characteristics were compared according to AHI among 114 patients, and it was observed that those with extreme OSA were more obese than patients with severe OSA (39.26 ± 5.93 vs. 35.68 ± 6.45 kg/m2, P = 0.025). Obesity hypoventilation syndrome (OHS) was significantly higher in the extreme OSA group (48% vs. 84%, P = 0.04) (Table 1). Moreover, compared to the severe OSA group, the extreme OSA group had a lower minimal O2 saturation (65.68 ± 10.16 vs. 74.10 ± 8.74, P = 0.003) and a higher T < 90 to total sleep time (81.78 ± 22.57 vs. 58.87 ± 33.14, P = 0.01) (Table 2).
Table 1.Demographic and Clinical Characteristics of Patients in the Severe OSA and Extreme OSA Groups
| Very Severe OSA, N= 95 | Extreme OSA, N=19 | P-Value | |
|---|---|---|---|
| Age (y) | 60.68 ± 11.10 | 58.74 ± 13.20 | 0.528 |
| Gender (M/F), No. (%) | 53 (55.8)/42 (44.2) | 11 (57.9)/8 (42.1) | 0.866 |
| Body mass index, kg/m2 | 35.68 ± 6.45 | 39.26 ± 5.93 | 0.025a |
| Hypertension, No. (%) | 55 (57.9) | 14 (73.7) | 0.199 |
| Diabetes mellitus, No. (%) | 27 (28.4) | 8 (42.1) | 0.238 |
| Hyperlipidemia, No. (%) | 32 (33.7) | 9 (47.4) | 0.257 |
| PCI, No. (%) | 7 (7.4) | 3 (15.8) | 0.236 |
| Hypothyroidism, No. (%) | 9 (9.5) | 1 (5.3) | 0.554 |
| OHS, No. (%) a | 46 (48%) | 16 (84%) | 0.04 b |
a Obesity hypoventilation syndrome.
b Significant difference.
Table 2.Polysomnographic Characteristic Differences Between the Severe OSA and Extreme OSA Groups
| Very Severe OSA (N= 95) | Extreme OSA (N=19) | P-Value | |
|---|---|---|---|
| Central apnea index, (%) | 7.40 ± 10.31 | 4.88± 6.17 | 0.308 |
| AHI | 75.6 ± 11.1 (73.5-77.9) | 113.9 ± 2.8 (108.1-119.9) | 0.01a |
| Minimal O2 saturation, (%) | 74.10 ± 8.74 | 65.68 ± 10.16 | 0.003a |
| T < 90 to TST, (%) | 58.8 ± 33.14 | 81.78 ± 22.57 | 0.001a |
| CPAP, No. (%) | 57 (60.0) | 7 (36.8 | 0.063 |
| BIPAP, No (%) | 38 (40.0) | 12 (63.2) | 0.063 |
a Significant difference.
Among the 114 study subjects, 69 (60.5%) had hypertension, 35 (30.7%) diabetes mellitus (DM), 41 (36.0%) hyperlipidemia, 10 (8.8%) a history of PCI, and 10 (8.8%) hypothyroidism, although no significant differences were observed regarding these clinical parameters (Table 1).
Of the 19 patients in the extreme OSA group, 12 (63.2%) responded to Bilevel positive airway pressure (PAP) and seven (36.8%) to continuous positive airway pressure therapies. Compared to patients in the very severe group, those with AHI ≥ 100 had a better response to Bilevel PAP, although the difference between the two groups was insignificant (63.2% vs. 40% P = 0.063) (Table 2).
5. Discussion
The main goal of the present retrospective study was to determine the clinical and polysomnographic differences between the two subgroups of severe OSA, as well as the baseline characteristics of the patients, such as age, gender, and obesity. Knowledge of different characteristics of patients with AHI > 100 may increase awareness and improve the diagnosis. The present study assessed the differences in clinical and polysomnographic findings between 19 patients with AHI ≥ 100 and 95 with 60 ≤ AHI < 100, and it was observed that patients with extreme AHI were significantly more obese, had lower minimal O2 saturation during sleep, and relatively more time with O2 saturation < 90%.
Obesity is a major risk factor for OSA presented in 70% of patients with this disorder, according to previous research (13, 14). Lee et al. (15), suggested that obesity might play a pivotal role in the severity of OSA.
The current study results were in agreement with those of previous studies, but the specific subgroup of severe apnea, with AHI > 100, was not distinguished in the literature. One of the very few studies in which patients with AHI > 100 were distinguished from other subgroups of severe OSA was conducted by Rey de Castro et al. (11). They investigated patients with AHI > 100 and, in a bivariate analysis, found an intense association between obesity and extreme OSA. However, the association was lost when neck circumstance was adjusted; it was not observed in the current study.
Previous studies also showed that the degree of nocturnal hypoxemia was higher in patients with severe OSA (11, 16). This parameter is evaluated by indicators, such as T < 90% and minimal O2 saturation. Kulkas et al., demonstrated that the minimal O2 saturation decreases as tending toward severe OSA (17) that was in agreement with the current study findings, and the lowest SpO2 and T < 90 were significantly different in patients with extreme OSA, which should be considered as a great risk factor for cardiovascular events and apnea-induced hypertension in this subgroup (18-20). Gehring et al., evaluated patients with severe OSA and found that nocturnal BP fluctuations, along with very low O2 saturation, can add to cardiovascular events during the night (18). It is well known that intermittent hypoxia in patients with OSA contributes to the pathogenesis of hypertension by activating the rennin-angiotensin system (19). Severity in oxygen deterioration is also related to impaired glucose tolerance in patients with OSA (21). Thus, the mentioned complications indicate the importance of severe nocturnal hypoxia management in patients with extreme OSA.
Despite the fact that patients in the extreme OSA group had a higher response to Bilevel PAP, the difference was not statistically significant, which could be due to the small number of patients in the extreme OSA group.
Higher arterial blood pressure is previously shown in association with AHI >100 compared to moderate (AHI 15 - 29) and severe (AHI 30 - 50) cases (11), but in the current study, hypertension prevalence was not significantly different among the subgroups of severe OSA.
The main limitation of the present study was the small sample size due to a low number of cases referred to the study site and diagnosed by polysomnography with very severe apnea. However, to avoid losing statistical power, the number of controls was increased. Additionally, the clinical characteristics of all the patients could not be evaluated due to the missing documents or lack of self-knowledge of patients.
5.1. Conclusions
In the present study, patients with extreme OSA (AHI > 100) were significantly more obese and had different polysomnographic manifestations, such as lower minimal O2 saturation and more time with < 90% O2 saturation, compared to those with 60 < AHI < 100.
