Obstructive Sleep Apnea Syndrome: Prediction of Lipid Panel in Relation to Apnea- Hypopnea Index

Abstract

The research was designed to predict the relationship between the apnea-hypopnea index (AHI) as the main indicator of severity of obstructive sleep apnea syndrome (OSAS), and lipid panel test results. A cross-sectional observational study was done on 90 patients with suspected sleep-related breathing disorders as assumed by polysomnography. Patients were categorized into three equal groups depending on AHI: mild degree (5–15 events/hour), moderate degree (15–30 events/hour), and severe degree (> 30 events/hour). All patients underwent a comprehensive medical history, PSG, and lipid panel tests, which included measurements of total cholesterol level, low-density lipoprotein (LDL) level, very low-density lipoprotein (VLDL) level, high-density lipoprotein (HDL) level, and triglycerides (TGs) level. Aclose relation was noted between degree of severity of OSA and lipid panel tests that in third group (Severe degree) exhibited significantly higher levels of triglycerides (212 ± 36 mg/dL), total cholesterol (180 ± 54 mg/dL), and LDL (178.1 ± 39.8 mg/dL) compared to those with moderate (TGs: 158 ± 57 mg/dL, total cholesterol: 151 ± 65 mg/dL, LDL: 153.2 ± 31.5 mg/dL) and mild OSA (TGs: 106 ± 37 mg/dL, total cholesterol: 85 ± 12 mg/dL, LDL: 87 ± 9.3 mg/dL), with P-values < 0.001. Conversely, significantly lower level of HDLin the severe OSA group (25.8 ± 3.6 mg/dL) compared to moderate (32.7 ± 4.2 mg/dL) and mild OSA groups (45.8 ± 9.2 mg/dL), with a P-value < 0.001. so analysis of Multivariate linear regression unveiled that both moderate and severe OSA stood as independent predictors for elevated TC, TGs, and LDL levels, alongside diminished HDL levels, according to age, sex, and BMI. Oxygen desaturation due to OSA significantly impacts lipid metabolism, leading to dyslipidemia and increased cardiovascular risk.

Introduction

Sleep disorders are ubiquitously encountered, wherein (OSA) is the predominant manifestation With incessant attacks of partial collapse (hypopnea) or full collapse (apnea) of the upper airway’s collapsible segment during sleep, OSA induces profound daytime drowsiness or relentless fatigue [1]. A wealth of research has demonstrated a consistent association between OSA and heightened risks of vehicular accidents, cardiovascular pathologies and cognitive impairments [2–4].

in a sleep laboratory a Conventional polysomnography (PSG) has been upheld as the gold standard for the diagnosis of OSA [5]. PSG meticulously records a spectrum of physiological variables during slumber, including stages of sleep, respiratory dynamics, oxygen saturation levels, cardiac frequencies, positional changes of the body, and movements of the limbs. These recorded data points are integral to derivation of AHI. Within this framework, apnea denotes complete cessation of airflow for ≥ 10 s, whereas hypopnea is characterized by reduction of respiratory activity leading to ≤ 4% decrease in oxygen saturation. The AHI is a metric quantifying of the frequency of both apneic and hypopneic episodes per hour of sleep, emerges as a fundamental diagnostic parameter for OSA, with a delineation threshold exceeding five events per hour generally indicative of the presence of a sleep disorder [6, 7].

Dyslipidemia and metabolic syndrome are prevalent comorbid conditions in patients afflicted with OSA. Empirical evidence suggests that a significant proportion, specifically one-third, of those diagnosed with OSA also present with metabolic syndrome [8]. Enhanced levels of triglycerides and diminished HDL cholesterol are the most frequently encountered lipid profile deviations in obese patients suffering from OSA [9]. Repeated hypoxic episodes and persistent stimulation of sympathetic nervous system can induce an inflammatory reaction which compromises vascular and myocardial integrity, thereby predisposing individuals to accelerated atherosclerosis and its resultant complications [10]. Atherogenic processes in individuals with OSA begin shortly after the onset of the condition, accentuating the crucial necessity for early diagnosis and preemptive treatment strategies [11].

Our claim is that intermittent hypoxia from obstructive sleep apnea negatively impacts lipid profiles, leading to dyslipidemia and increased cardiovascular risk. This hypoxia can lead to dysregulation of lipid profiles, resulting in dyslipidemia. Dyslipidemia increases the risk of atherosclerosis and other cardiovascular complications. Consequently, recognizing predictors of heightened cardiovascular risk is essential for the timely diagnosis and appropriate therapeutic intervention to avert future deleterious cardiovascular events [10]. The lipid panel is a blood test in which the amount of certain fat molecules are measured, including different cholesterol types and triglycerides, in the blood.

Hence, the aim of this study is to predict the relationship between AHI and the results of lipid panel tests.

Patients and Methods

Design and Population

Across-sectional international study was carried out in the period from June 2023 to June 2024 at Benha University Hospital, enrolled a cohort of 90 patients referred for PSG owing to suspicion of sleep-related breathing disorders. Following approval by the Institutional Review Board of the Faculty of Medicine at Benha University, A written informed consent was meticulously taken from all patients.

Ethical approval code: RC 8-6-2023.

Inclusion Criteria

Cases aged 18 and above with AHI ≥ 5 as recorded by PSG.

Exclusion Criteria

individuals with unstable vital signs, major neurological or behavioural abnormalities, and those on medications that could impact lipid levels, sleep quality, or autonomic nervous system function during lipid panel testing.

Grouping

Patients were allocated into three main groups according to AHI which is the main indicator to classificy OSAS according to severity [12]. Group 1 (Mild degree) included 30 patients (AHI: 5–15 events/hour), Group 2 (moderate degree) included 30 patients (AHI: 15–30 events/hour), and Group 3 (Severe degree) included 30 patients (AHI: >30 events/hour).

Polysomnography (PSG)

All patients underwent an overnight PSG (SOMNO Screen Plus; SOMNO Medics GmbH, Randersacker, Germany) in a sleep laboratory. PSG monitored the sleep stages, oxygen saturation, respiratory effort, body position, heart rate,, and limb movements. AHI was determined as the mean frequency of apneic and hypopneic events / hour of sleep. noting that, Apnea is the cessation of airflow enduring for a duration of ≥ 10 s, whereas hypopnea is stipulated as a decrement in respiratory effort concomitant with an oxygen desaturation of 4% or more.

Lipid Panel Tests

Blood samples were methodically collected from each subject to intricately examine their lipidomic profiles. This comprehensive array of lipid assays encompassed the meticulous assessment of total cholesterol level, HDL cholesterol level, VLDL cholesterol level, LDL cholesterol level, and TGs level. These diagnostic endeavors were conducted under rigorously controlled fasting conditions to ensure the integrity and accuracy of the analytical outcomes.

Statistical Methods

Data management and statistical analysis were executed utilizing SPSS version 28 (IBM, Armonk, New York, USA). The assessment of the normal distribution of quantitative variables were assessed using a combination of direct data visualization methodologies and the Shapiro-Wilk test. quantitative data were delineated in terms of their means and standard deviations, while categorical data were succinctly summarized in the form of frequencies and corresponding percentages. Intergroup quantitative data were compared using one-way ANOVA, with post hoc analyses adjusted using the Bonferroni method. The Chi-square test was utilized to compare the categorical variables. To predict lipid profile parameters, multivariate linear regression analysis was performed, calculating the regression coefficients along with their 95% CI. All statistical analyses conducted adhered to a two-sided framework, with significance levels set at P value below 0.05 that denoted statistical significance.

Results

Demographics

Patients with mild OSA exhibited significantly lower BMI (23.4 ± 3.4) than those with moderate (28.3 ± 3.9) and severe disease (26.7 ± 4.6) (P < 0.001). Other variables, such as age (P = 0.151), and sex (P = 0.279) did not show significant differences across the groups (Table 1).

Table 1.

Demographic characteristics of the studied groups

		OSA
		Mild (n = 30)	Moderate (n = 30)	Severe (n = 30)	P-value
Age (years)	Mean ± SD	33 ± 5	36 ± 10	36 ± 9	0.151
Sex		14 (46.7)	20 (66.7)	18 (60)	0.279
Males	n (%)	16 (53.3)	10 (33.3)	12 (40)
Females	n (%)
BMI	Mean ± SD	23.4 ± 3.4b, c	28.3 ± 3.9a	26.7 ± 4.6a	< 0.001*

*Statistically Significant at P < 0.05

^aSignificantly different form mild group

^bSignificantly different from moderate group

^cSignificantly different from severe group

OSA Obstructive Sleep Apnea, SD Standard Deviation, BMI Body Mass Index, n Number

Lipid Profile

Triglycerides (TGS) were significantly higher in third group participants (212 ± 36) compared to second group (158 ± 57) and mild OSA (106 ± 37) (P < 0.001). Total cholesterol levels were also significantly elevated in the severe OSA group (180 ± 54) in comparison to moderate (151 ± 65) and mild OSA participants (85 ± 12) (P < 0.001). Similarly, Low-Density Lipoprotein (LDL) levels were significantly higher in severe OSA participants (178.1 ± 39.8) compared to those with moderate (153.2 ± 31.5) and mild OSA (87 ± 9.3) (P < 0.001). Conversely, High-Density Lipoprotein (HDL) levels were significantly lower in the severe OSA group (25.8 ± 3.6) compared to moderate (32.7 ± 4.2) and mild OSA participants (45.8 ± 9.2) (P < 0.001) (Table 2; Fig. 1).

Table 2.

Lipid profile in the studied groups

		OSA
		Mild (n = 30)	Moderate (n = 30)	Severe (n = 30)	P-value
TGs	Mean ± SD	106 ± 37b, c	158 ± 57a, c	212 ± 36a, b	< 0.001*
Total cholesterol	Mean ± SD	85 ± 12b, c	151 ± 65a, c	180 ± 54a, b	< 0.001*
LDL	Mean ± SD	87 ± 9.3b, c	153.2 ± 31.5a, c	178.1 ± 39.8a, b	< 0.001*
HDL	Mean ± SD	45.8 ± 9.2b, c	32.7 ± 4.2a, c	25.8 ± 3.6a, b	< 0.001*

*Statistically Significant at P < 0.05

^aSignificantly different form mild group

^bSignificantly different from moderate group

^cSignificantly different from severe group

OSA Obstructive Sleep Apnea, TGS Triglycerides, SD Standard Deviation, LDL Low-Density Lipoprotein, HDL High-Density Lipoprotein, n Number

Fig. 1.

Lipid profile across the studied groups

Prediction of Lipid Profile

Multivariate linear regression analysis was done to predict different lipid profile parameters. Moderate degree of OSA was significantly associated with ahigher level of Triglycerides (TG) (B = 53.295, 95% CI [27.909–78.681], P < 0.001), total cholesterol (B = 62.034, 95% CI [33.505–90.563], P < 0.001), and LDL (B = 56.094, 95% CI [39.433–72.754], P < 0.001), but with lower levels of HDL (B = −12.877, 95% CI [−16.398 − −9.356], P < 0.001), controlling for BMI, age and sex (Table 3).

Table 3.

Linear regression analysis to predict lipid profile

	TGs		Total cholesterol		LDL		HDL
	B (95% CI)	P	B (95% CI)	P	B (95% CI)	P	B (95% CI)	P
Age (years)	1.203 (−0.011–2.417)	0.052	1.17 (−0.195–2.535)	0.092	0.548 (−0.249–1.345)	0.175	−0.038 (−0.206–0.131)	0.657
Sex	1.916 (−17.355–21.188)	0.844	−3.835 (−25.492–17.822)	0.726	−3.46 (−16.108–9.187)	0.588	3.152 (0.479–5.825)	0.021
Mod. OSA	53.295 (27.909–78.681)	< 0.001*	62.034 (33.505–90.563)	< 0.001*	56.094 (39.433–72.754)	< 0.001*	−12.877 (−16.398 − −9.356)	< 0.001*
Sev. OSA	105.69 (81.709–129.671)	< 0.001*	90.786 (63.837–117.736)	< 0.001*	83.601 (67.863–99.34)	< 0.001*	−19.791 (−23.117 − −16.464)	< 0.001*
BMI	−1.014 (−3.488–1.46)	0.417	−0.078 (−2.858–2.703)	0.956	1.548 (−0.076–3.171)	0.061	0.095 (−0.249–0.438)	0.585

*Significant P-value at P < 0.05

TGs Triglycerides, LDL Low-Density Lipoprotein, HDL High-Density Lipoprotein, B Regression Coefficient, CI Confidence Interval, P P-value, OSA Obstructive Sleep Apnea, BMI Body Mass Index

Similarly, severe OSA showed significant associations with higher TG (B = 105.69, 95% CI [81.709–129.671], P < 0.001), total Cholesterol (B = 90.786, 95% CI [63.837–117.736], P < 0.001), and LDL (B = 83.601, 95% CI [67.863–99.34], P < 0.001), and lower HDL levels (B = −19.791, 95% CI [−23.117 − −16.464], P < 0.001), regarding age, sex, and BMI (Table 3).

Discussion

Obstructive sleep apnea syndrome (OSAS) is known to significantly impact various metabolic processes, particularly lipid metabolism [13]. The intermittent hypoxia that characterizes OSAS contributes to a cascade of physiological disturbances, resulting in dyslipidemia [14, 15]. Our study sought to explore this relationship by examining the lipid profiles of patients with varying severities of OSAS.

In the current work, patients with mild OSA exhibited significantly lower BMI than those with moderate and severe disease.no significant differences were noted across the groups regarding age or sex. Supporting this study, Fang et al. [16] reported that patients with severe OSA had higher BMI and worse lipid profiles compared to those with mild or moderate OSA. Specifically, their study found a BMI of 30.57 ± 5.99 kg/m² in severe OSA patients, compared to 27.43 ± 4.04 kg/m² and 28.49 ± 5.31 kg/m² in mild and moderate OSA patients, respectively (p < 0.001). Contrastingly, they observed a higher percentage of males in the severe OSA group (82.93%) compared to the mild (62.50%) and moderate (79.56%) groups, The variation in results may be ascribed to the disparities in demographic parameters and sample size between the two investigations.

This study demonstrated that patients with severe OSAS exhibited elevated levels of triglycerides (212 ± 36 mg/dL), total cholesterol (180 ± 54 mg/dL), and LDL cholesterol (178.1 ± 39.8 mg/dL), along with decreased levels of HDL cholesterol (25.8 ± 3.6 mg/dL), compared to those with mild or moderate OSAS. In harmony with our findings, Hendy et al. [13] found that as OSA severity increased, there were significant changes in lipid profiles. Specifically, BMI, LDL, and HDL levels varied significantly across mild, moderate, and severe OSA groups, with p-values of 0.024, < 0.001, and 0.005, respectively. However, TG showed borderline significance across the groups (p = 0.059). Comparably, Fang et al. [16] observed higher levels of TC, TG, and LDL cholesterol (LDL-C), and lower HDL cholesterol (HDL-C) in severe OSA patients compared to those with mild or moderate OSA (all p-values < 0.05). In addition, Gunduz et al. [17] demonstrated that AHI independently predicts lipid levels, with significant elevations in total cholesterol, LDL cholesterol, and triglycerides, while decreased HDL cholesterol in patients with severe OSAS.

Additionally, in the Sleep Heart Health Study cohort, as elucidated by Newman et al. [18], established a correlation between severe OSA and elevated levels of TC and TG, particularly in males ≤ 65 years old. Togeiro et al. [19] investigated a cohort of 1,042 participants who underwent PSG. This study revealed that subjects with moderate to severe OSA were just older, exhibited higher degrees of obesity, and presented with significantly elevated TG levels compared to their counterparts with mild or no OSA (P < 0.001).

Our investigation found that both moderate and severe OSA were independently linked to elevated levels of TGs, LDL, total cholesterol, as well as reduced levels of HDL. This association persisted even after meticulous adjustment of variables including age, sex and BMI.

Consistently, Roche et al. [20] found that moderate to severe OSAS was associated with low HDL-C levels independently in an elderly population (p < 0.03) after adjusting for multiple assiciated factors. Hendy et al. [13] reported that Lipid abnormalities are associated with degree of OSAS severity, as they found that LDL and TG were positively correlated with AHI, while HDL was negatively related to AHI with all correlations being statistically significant. In another multisite investigation, Trzepizur et al. [21], showed that the severity of OSA was correlated with elevated TG levels and reduced levels of high-density lipoprotein cholesterol (HDL-C), even after adjusting of associated variables. Likewise, an investigation conducted by Wang et al. [22], in a Chinese cohort identified a robust correlation between an increased triglyceride glucose index and OSA.

Fang et al. [16] intriguingly elucidated a direct linear relationship between TC/HDL-C ratio and the propensity for severe OSAS. Kawano et al. [23] demonstrated that as OSA severity escalates, significant perturbations occur in the lipid profile, notably the ratio of LDL-C/HDL-C. They reported that the LDL-C/HDL-C ratio showed a positive relation to AHI (ρ = 0.28, p < 0.001) and, through multivariate regression analysis, determined that AHI was independently associated with LDL-C/HDL-C. Additionally, Togeiro et al. [19] identified triglyceride levels to be the prominent indicator of episodic hypoxia, solely linked to AHI.

Furthermore, Nadeem et al. [24] conducted a meta-analysis of 1,958 subjects, indicating that CPAP treatment in OSAS patients improves dyslipidemia by reduction of total cholesterol and LDL levels and increasing levels of HDL. This suggests that treating sleep apnea may improve heart disease risk factors by positively affecting obesity, hypertension, diabetes, and metabolic syndrome. Similarly, Börgel et al. Mentioned that AHI was associated with levels of HDL-C and that bilevel or CPAP therapy significantly improved abnormal lipid levels in OSA patients [25]. Nevertheless, there existed notable disparities among the studies, with certain investigations yielding positive outcomes while others reported no discernible impact of CPAP therapy on lipid profiles [26–28].

Contrasting our findings, Sharma et al. [27] found that OSAS has no independent relation to lipid abnormalities. The discrepancy can be justified by differences in study design and population. In a recent inquiry concerning pediatric populations undertaken by Lei et al. [29], it was discerned that serum lipid parameters did not demonstrate any discernible correlation with the magnitude of AHI.

Finally, we had some limitations in this study. Firstly, its Across-sectional design which prevents the achievement of causality between OSA severity and lipid profile alterations. Secondly, the small sample size which was derived from one center, which may limit the globalization of the findings. Additionally, this study did not follow patients over time, precluding the assessment of long-term effects of episodic hypoxia on lipid metabolism. Other potential associated factors as dietary habits, physical activities, and genetic predispositions, were not controlled for, which could influence the lipid profiles independently of OSA. Future studies with larger, more diverse populations and longitudinal designs are needed to validate our findings.

Conclusions

Intermittent hypoxia due to OSAS significantly impacts lipid metabolism, leading to dyslipidemia. This study demonstrates that severe OSA is associated with elevated triglycerides, total cholesterol, and LDL levels, along with reduced HDL levels. Therefore, monitoring lipid profiles in patients with OSAS is crucial to mitigate the increased cardiovascular risk. Effective management of OSA can improve lipid profiles and reduce associated cardiovascular morbidity.

Author Contributions

R.T.: protocol/project development, manuscript writing/editing. R.T. and O.A.: data collection and management, data analysis, administrative, technical, and material support. R.T.: manuscript writing/editing. R.T. and O.A.: protocol/project development.

Data Availability

The data underpinning these findings and interpretations of our study can be obtained from the corresponding author on request.

Declarations

Ethical Approval

This study was approved by the Research Ethics Committee, Faculty of Medicine, Benha University, Egypt (Approval no: Rc 8-6-2023).

Informed Consent

Informed consent was obtained from all patients included in the study.

Conflict of interest

No conflict of interest.