Evaluation of inflammatory markers in obstructive sleep apnea syndrome

Abstract

Objective

This study aimed to investigate the relationship between systemic immune-inflammation indices and the severity of obstructive sleep apnea syndrome (OSAS). Specifically, we evaluated whether the Systemic immune-inflammation index (SII), systemic inflammation response index (SIRI), neutrophil-to-lymphocyte ratio (NLR), and pan-immune inflammation value (PIV) could serve as predictive or supplementary markers for OSAS prognosis and severity.

Methods

A retrospective cross-sectional analysis was conducted on 263 patients diagnosed with OSAS. Based on the apnea-hypopnea index (AHI), patients were categorized into three groups: mild, moderate, and severe OSAS. Inflammatory markers including SII, SIRI, NLR, and PIV were calculated using routine hematological parameters and compared across the groups to determine their association with disease severity.

Results

All examined inflammatory indices (NLR, SII, SIRI, and PIV) showed statistically significant differences across the OSAS severity groups. Specifically, these markers were significantly elevated in the severe OSAS group compared to the mild and moderate groups (p < .05). The results support the hypothesis that increased systemic inflammation is associated with more severe forms of OSAS.

Conclusions

Inflammatory markers such as SII, SIRI, NLR, and PIV may serve as useful tools in assessing OSAS severity and could assist in clinical decision-making. Given their ease of calculation from standard blood tests, these indices may offer practical value in evaluating prognosis and tailoring treatment strategies for patients with OSAS.

Introduction

Obstructive sleep apnea syndrome (OSAS) is characterized by recurrent closure or narrowing of the upper airway and is a significant disorder affecting approximately 10–25% of the world's population, causing intermittent hypoxia, sleep fragmentation, and recurrent nighttime awakenings. ¹ The gold standard method for diagnosing OSAS is polysomnography (PSG). A wide range of parameters are systematically collected during sleep by specialized teams in special centers. ²

Many factors and comorbid conditions affect the pathophysiology of OSAS. Therefore, a heterogeneous condition characterized by a wide variety of clinical findings and respiratory changes emerges. ³ Increased sympathetic activity decreases in oxygen levels in the body during apnea, and inflammation in the upper airways during this time can initiate the systemic inflammation process. Oxidative stress resulting from hypoxia due to repeated apnea causes an increase in many inflammatory and proinflammatory values. There are studies in which increased intracellular adhesion molecule-1, vascular cell adhesion molecule-1, tumor necrosis factor-α (TNF-α), interleukin (IL), and c reactive protein (CRP) were detected in OSAS patients.^4–6 Due to this pathophysiological process, it is thought that the increased systemic inflammatory response in OSAS patients primarily increases the risk of cardiovascular diseases, pulmonary diseases, and neuropsychological diseases. The pan-immune-inflammation value (PIV), the systemic immune-inflammation index (SII) and systemic inflammation response index (SIRI) have recently gained more attention as a novel indicator of inflammation. Studies have shown that the PIV, SII and SIRI are effective in predicting inflammation and prognosis in various disease groups.^7,8

In this study, we aimed to investigate the levels of blood PIV, SIRI and SII as a response to inflammation in OSAS and their relationship with sleep severity.

Methods

This retrospective cross-sectional study was conducted at Baskent University Konya Application and Research Center between May 2021 and September 2024. Patients diagnosed with OSAS who underwent PSG during this period were consecutively included if they met the predefined inclusion and exclusion criteria. Exclusion criteria were systemic inflammatory diseases, malignancy, active infection, recent use of analgesics, uncontrolled hypertension, intracranial space-occupying lesions, recent treatment for anemia or polycythemia, history of bleeding or surgery, blood transfusions within the last 3 months, hematological diseases, iron deficiency anemia, thalassemia, and inflammatory bowel disease.

The study protocol was approved by the Institutional Ethics Committee of Baskent University Institutional Review Board (Approval No: KA25/224, Date: 11/06/2025). The study was conducted in accordance with the principles of the Declaration of Helsinki, as revised in 2024 and reported following the STROBE guidelines.

Demographic data, medical history, body mass index (BMI), and Epworth Sleepiness Scale (ESS) scores were obtained from hospital records. Laboratory parameters were collected from venous blood samples taken within one month prior to the PSG test. All data were recorded and compiled in an Excel database for analysis. ⁹

PSG

Sleep and physiological variables were monitored using the Philips Respironics Alice 5 Diagnostic Sleep System. Recorded parameters included electroencephalography (EEG), electromyography (EMG), electrooculography (EOG), electrocardiography (ECG), oronasal airflow (thermal sensor and nasal pressure transducer), body position, thoracic and abdominal movements (inductance plethysmography), arterial oxygen saturation via pulse oximetry, leg movements, and tracheal sounds. Apnea was defined as a ≥90% reduction in airflow lasting at least 10 s. Hypopnea was defined as a ≥30% reduction in nasal pressure signal accompanied by ≥3% oxygen desaturation or arousal, lasting at least 10 s. OSAS severity was classified based on Apnea-Hypopnea Index (AHI) scores as mild (5–14.9), moderate (15–29.9), and severe (≥30). The AHI, defined as the average number of apnea and hypopnea events per hour of sleep, was used to assess the severity of OSAS. ¹⁰

Calculation of systemic inflammation markers

Complete blood count parameters, including white blood cells, neutrophils, lymphocytes, monocytes, platelets, hemoglobin, and red cell distribution width, were obtained from hospital records. Systemic inflammation markers were calculated as follows:

• Neutrophil-to-lymphocyte ratio (NLR) = Neutrophil count/Lymphocyte count;

• Systemic immune-inflammation index (SII) = (Neutrophil count × Platelet count)/Lymphocyte count;

• Systemic inflammation response index (SIRI) = (Neutrophil count × Monocyte count)/Lymphocyte count;

• Pan-Immune Inflammation Value (PIV) = (Neutrophil count × Platelet count × Monocyte count)/Lymphocyte count.

All blood counts are expressed as ×10³/µL.

Statistical methods

The statistical analyses were performed using SPSS (Statistical Package for Social Sciences) 29.0 package program. Continuous variables were expressed as mean ± standard deviation (SD) and categorical variables as frequencies (n) and percentages (%). For continuous variables, the normality assumption was tested using the Kolmogorov-Smirnov or Shapiro-Wilk test, depending on sample size, while the homogeneity of variances was assessed using Levene's test. If the assumptions of normality and homogeneity of variances were met, parametric tests were used for comparisons. These included the Student's t-test for two-group comparisons and one-way ANOVA for multiple group comparisons, with post-hoc tests applied as necessary. If the assumptions were violated, non-parametric tests were applied, such as the Mann-Whitney U test for two-group comparisons and the Kruskal‒Wallis test for multiple group comparisons. For categorical variables, associations between groups were analyzed using the chi-square test or Fisher's exact test, depending on the distribution of the data and the expected frequencies. A p-value <.05 was considered statistically significant in all analyses. These methods ensured the robustness of the statistical results and provided reliable insights into the relationships between clinical parameters, comorbidities, and OSAS severity.

Results

A total of 263 patients were included in the study. One hundred and fifty-two of the patients were male and 111 were female. Table 1 provides a comprehensive overview of the demographics and clinical characteristics of patients with OSAS. The patient cohort consisted of 57.8% men and 42.2% women, indicating a higher prevalence of OSAS among men. The majority of patients (78.7%) did not have hypertension, while 20.9% were hypertensive, reflecting the known association between OSAS and increased risk of hypertension. Similarly, most patients (84.8%) were non-diabetic, with 14.8% diagnosed with diabetes mellitus, highlighting the potential metabolic impact of OSAS. Hyperlipidemia was present in only 5.3% of patients, whereas 93.9% were free of the condition. Coronary artery disease (CAD) was observed in 9.1% of the cohort, with the remaining 90.5% unaffected, aligning with the established link between OSAS and cardiovascular risk. Chronic obstructive pulmonary disease (COPD) was noted in 40.7% of patients, indicating a significant overlap between these conditions, which may exacerbate respiratory symptoms. All of our patients had at least one of these chronic diseases. Snoring, a hallmark symptom of OSAS, was reported by 85.6% of the patients, while 14.4% did not experience this symptom. Fatigue, another common feature of OSAS, affected 71.5% of the patients, with only 28.5% reporting no tiredness. Regarding disease severity, 46.8% had mild OSAS, 14.1% moderate OSAS, and 39.2% severe OSAS.

Table 1.

Demographics of patients with OSAS.

Characteristics	Group	N	%
Gender	Man	152	57.8
	Woman	111	42.2
HT	No	207	78.7
	Yes	56	20.9
DM	No	223	84.8
	Yes	40	14.8
HL	No	247	93.9
	Yes	16	5.3
CAD	No	238	90.5
	Yes	25	9.1
COPD	No	156	59.3
	Yes	107	40.7
Snoring	No	14.4	14.4
	Yes	85.6	85.6
Tiredness	No	75	28.5
	Yes	188	71.5
OSAS	Mild OSAS	123	46.8
	Moderate OSAS	37	14.1
	Severe OSAS	103	39.2

Table 2 evaluates the sleep recording data, providing detailed insights into various patient characteristics and sleep parameters. The mean age of the patients was 45.99 ± 11.81 years, with a range from 18 to 75 years. The average height was 1.69 ± 0.1 meters, ranging from 1.47 to 1.95 m, and the mean weight was 108.17 ± 107.28 kg, with a notable wide range of 60‒1340 kg. BMI showed an average of 37.82 ± 36.36, ranging from 21.007 to an exceptionally high value of 485.528, indicating significant variability in body composition among the patients. The mean ESS score was 9.05 ± 6.44, ranging from 0 to 24, reflecting varying levels of daytime sleepiness.

Table 2.

Evaluation of sleep recording data.

Variable	Mean ± SD	Minimum	Maximum
Age	45.99 ± 11.81	18	75
Height	1.69 ± 0.1	1.47	1.95
Weight	108.17 ± 107.28	60	1340
BMI	37.82 ± 36.36	21.007	485.528
Epworth score	9.05 ± 6.44	0	24
Total sleep time	378.01 ± 69.59	99	580.5
Sleep efficiency (%)	80.01 ± 13.78	20.1	98.9
Sleep latency start time (min)	33.76 ± 32.21	1.5	184.5
Wake after sleep onset	68.26 ± 89.08	1	922.8
REM latency (min)	193.98 ± 84.88	0	483
N1 (%)	2.65 ± 3.84	0	22.5
N2 (%)	78.95 ± 56.43	31.6	342.5
N3 (%)	36.64 ± 30.94	0	255
REM (%)	14.74 ± 14.53	0	107
Apnea-hypopnea index	25.21 ± 23.92	0	113.4
REM AHI	26.57 ± 28.98	0	106.2
NREM AHI	24.54 ± 23.9	0	113.8
Supine total sleep time (%)	38.23 ± 26.21	0	100
SUPINE AHI	34.67 ± 30.79	0	120.4
Non-supine AHI	18.65 ± 23.85	0	150
Oxygen desaturation index	39.24 ± 30.22	1.5	152.7
Mean SpO2 (%)	91.04 ± 4.73	40	98
Lowest SpO2 (%)	78.87 ± 47.76	20	822

In terms of sleep quality, the total sleep time averaged 378.01 ± 69.59 min, with a range from 99 to 580.5 min. Sleep efficiency, a key indicator of sleep quality, averaged 80.01 ± 13.78%, with values ranging from a low 20.1% to 98.9%. Sleep latency (time to fall asleep) had a mean of 33.76 ± 32.21 min, varying widely between 1.5 and 184.5 min. Wake After Sleep Onset (WASO), which measures sleep interruptions, showed a mean of 68.26 ± 89.08 min, with a wide range of 1 to 922.8 min, indicating substantial differences in sleep continuity among patients. REM latency (time to first REM sleep) had a mean of 193.98 ± 84.88 min, with values ranging from 0 to 483 min.

The distribution of sleep stages revealed an average of 2.65 ± 3.84% in N1 sleep, 78.95 ± 56.43% in N2 sleep, 36.64 ± 30.94% in N3 sleep, and 14.74 ± 14.53% in REM sleep. These findings suggest a wide variation in the proportion of time spent in different sleep stages, which could reflect differing levels of sleep quality and disruption.

The AHI, a critical measure of OSAS severity, averaged 25.21 ± 23.92 events per hour, ranging from 0 to 113.4. REM AHI had a mean of 26.57 ± 28.98, while NREM AHI averaged 24.54 ± 23.9, both showing substantial variation. Supine sleep time accounted for an average of 38.23 ± 26.21% of total sleep, with supine AHI (34.67 ± 30.79) being significantly higher compared to non-supine AHI (18.65 ± 23.85), indicating the positional influence on respiratory events.

The ODI averaged 39.24 ± 30.22, ranging from 1.5 to 152.7, highlighting episodes of oxygen level drops during sleep. Mean oxygen saturation (SpO₂) was 91.04 ± 4.73%, with a minimum value of 40%, while the lowest recorded SpO₂ was 78.87 ± 47.76%, showing extreme variability and potential severity of hypoxic events.

Overall, the data in Table 2 illustrate significant variability in sleep and respiratory parameters among patients with OSAS. The wide ranges in indices such as BMI, AHI, and oxygen saturation underscore the condition's heterogeneous nature, emphasizing the need for individualized diagnostic and therapeutic approaches.

Table 3 presents laboratory variables providing insights into hematological and inflammatory profiles of the patient cohort. The mean platelet count (PLT) was 273.79 ± 64.7 × 10⁹/L, with values ranging from 149 to 490 × 10⁹/L, indicating normal platelet levels for most patients. Hemoglobin (HGB) levels averaged 14.52 ± 1.59 g/dL, with a range from 10.8 to 18.1 g/dL, falling within the typical physiological range.

Table 3.

Laboratory variables.

Variable	Mean ± SD	Minimum	Maximum
Platelets (PLT) × 109/L	273.79 ± 64.7	149	490
Neutrophils (NEU)/L	5.06 ± 2.14	1.45	16.95
Lymphocytes (LYM)/L	2.4 ± 0.94	0.24	5.75
Monocytes/L	0.77 ± 0.36	0.097	2.34
Neutrophil-to-lymphocyte ratio (NLR)	2.72 ± 2.68	0.5017	19.2083
Pan-immune inflammation value (PIV)	2732.11 ± 3989.93	32.8913	19701.24
Systemic inflammatory response index (SIRI)	2.75 ± 3.34	0.1703	27.4679
Systemic immune-inflammation index (SII)	533.42 ± 855.03	1.5376	5839.333

Neutrophil (NEU) counts averaged 5.06 ± 2.14 × 10⁹/L, with values ranging from 1.45 to 16.95 × 10⁹/L, and lymphocyte (LYM) counts had a mean of 2.4 ± 0.94 × 10⁹/L, ranging from 0.24 to 5.75 × 10⁹/L. The Neutrophil-to-Lymphocyte Ratio (NLR), an inflammation marker, showed a mean of 2.72 ± 2.68, with a wide range from 0.5017 to 19.2083, reflecting variability in systemic inflammation levels across patients.

The SII, which combines platelet, neutrophil, and lymphocyte counts, averaged 533.42 ± 855.03, with values spanning from 1.5376 to 5839.333. This wide range suggests significant heterogeneity in immune-inflammatory response among the cohort. Monocyte counts averaged 0.77 ± 0.36 × 10⁹/L, ranging from 0.097 to 2.34 × 10⁹/L.

The platelet-to-immune cell ratio (PIV) exhibited substantial variability, with a mean of 2732.11 ± 3989.93 and a range from 32.8913 to 19701.24, emphasizing large inter-patient differences in platelet-driven immune responses. Similarly, the SIRI, another marker of systemic inflammation, averaged 2.75 ± 3.34, ranging from 0.1703 to 27.4679.

In summary, the laboratory data in Table 3 highlight considerable variability in hematological and inflammatory markers among the patients. Elevated NLR, SII, and SIRI values in some individuals suggest heightened systemic inflammation, commonly associated with various pathological conditions, including OSAS. These findings provide valuable information for understanding the inflammatory and immune profiles of this cohort, offering potential markers for disease severity and prognosis.

Table 4 presents a comparison of BMI across various comorbidities in patients, providing insights into the relationship between BMI and specific conditions associated with OSAS.

Table 4.

BMI comparison across comorbidities.

Comorbidities	Group	Mean ± SD	p
HT	No	38.41 ± 2.85	0.158
	Yes	35.76 ± 0.81
DM	No	38.01 ± 2.65	0.007*
	Yes	36.97 ± 0.85
HL	No	38.04 ± 2.39	0.907
	Yes	34.68 ± 1.58
CAD	No	38.12 ± 2.48	0.858
	Yes	35.21 ± 1.16
COPD	No	37.02 ± 2.44	0.596
	Yes	39.01 ± 4.24
Snoring	No	37.15 ± 0.95	0.002*
	Yes	37.95 ± 2.62
Tiredness	No	40.98 ± 5.03	0.014*
	Yes	36.6 ± 2.44
OSAS	Mild OSAS	37.08 ± 3.07	0.089
	Moderate OSAS	34.37 ± 0.83
	Severe OSAS	40.01 ± 4.44

The * symbol was used for statistically significant values.

For hypertension (HT), the mean BMI was slightly higher in patients without hypertension (38.41 ± 2.85) compared to those with hypertension (35.76 ± 0.81); however, the difference was not statistically significant (p = .158). In the case of diabetes mellitus (DM), a statistically significant difference in BMI was observed (p = .007), with non-diabetic patients having a slightly higher mean BMI (38.01 ± 2.65) than diabetic patients (36.97 ± 0.85). For hyperlipidemia (HL), no significant difference was found (p = .907), with BMI being similar between those without HL (38.04 ± 2.39) and those with HL (34.68 ± 1.58).

Similarly, for CAD, there was no significant difference in BMI (p = .858) between patients without CAD (38.12 ± 2.48) and those with CAD (35.21 ± 1.16). In contrast, for COPD, patients with COPD had a higher mean BMI (39.01 ± 4.24) compared to those without COPD (37.02 ± 2.44), but this difference was not statistically significant (p = .596).

Regarding symptoms, snoring was associated with a statistically significant difference in BMI (p = .002). Patients who reported snoring had a higher mean BMI (37.95 ± 2.62) compared to those who did not snore (37.15 ± 0.95). Similarly, tiredness was significantly associated with BMI (p = .014), with patients without tiredness having a higher BMI (40.98 ± 5.03) compared to those who reported tiredness (36.6 ± 2.44).

For OSAS severity, although not statistically significant (p = .089), the mean BMI was highest among patients with severe OSAS (40.01 ± 4.44), followed by those with mild OSAS (37.08 ± 3.07) and moderate OSAS (34.37 ± 0.83). This trend suggests a possible relationship between higher BMI and increased OSAS severity.

In summary, the data reveal significant associations between BMI and certain conditions, including diabetes mellitus, snoring, and tiredness. Additionally, the trend of higher BMI in severe OSAS patients highlights the potential impact of obesity on OSAS severity and related symptoms. These findings underscore the importance of managing BMI in addressing both OSAS and its comorbidities.

Table 5 compares clinical parameters across different severity levels of OSAS (mild, moderate, and severe) and highlights statistically significant differences in several variables.

Table 5.

Comparison of inflammatory markers among OSAS severity groups.

Variable	Mild OSAS (mean ± SD)	Moderate OSAS (mean ± SD)	Severe OSAS (mean ± SD)	p
Neutrophils (NEU)	4.64 ± 0.13	4.47 ± 0.29	5.78 ± 0.26	.00*
Lymphocytes (LYM)	2.74 ± 0.07	2.85 ± 0.16	1.83 ± 0.08	.00*
Platelets (PLT)	263.60 ± 5.43	252.41 ± 9.32	293.64 ± 6.68	.00*
Monocytes	0.64 ± 0.02	0.73 ± 0.07	0.94 ± 0.03	.00*
Neutrophil-to-lymphocyte ratio (NLR)	1.78 ± 0.06	1.73 ± 0.19	4.19 ± 0.37	.00*
Pan-immune inflammation value (PIV)	290.00 ± 15.14	299.38 ± 38.04	6522.30 ± 406.19	.00*
Systemic inflammatory response index (SIRI)	1.74 ± 0.08	2.08 ± 0.25	4.20 ± 0.48	.00*
Systemic immune-inflammation index (SII)	318.14 ± 44.13	430.07 ± 43.86	1197.83 ± 101.94	.00*

The * symbol was used for statistically significant values. Note. Data are expressed as mean ± standard deviation.

Patients in the severe OSAS group exhibited significantly higher platelet (PLT) counts (293.64 ± 6.68 × 10³/μL) compared to those in the moderate (252.41 ± 9.32 × 10³/μL) and mild OSAS groups (263.60 ± 5.43 × 10³/μL) (p = .00).

Similarly, neutrophil (NEU) counts were elevated in the severe OSAS group (5.78 ± 0.26 × 10³/μL) compared to mild (4.64 ± 0.13 × 10³/μL) and moderate (4.47 ± 0.29 × 10³/μL) groups (p = .00).

In contrast, lymphocyte (LYM) counts were significantly reduced in the severe OSAS group (1.83 ± 0.08 × 10³/μL) when compared to mild (2.74 ± 0.07 × 10³/μL) and moderate (2.85 ± 0.16 × 10³/μL) groups (p <.001). This led to a substantial increase in the neutrophil-to-lymphocyte ratio (NLR) in severe OSAS (4.19 ± 0.37), compared to mild (1.78 ± 0.06) and moderate (1.73 ± 0.19) groups (p = .00).

The SII was also markedly higher in severe OSAS patients (1197.83 ± 101.94) than in moderate (430.07 ± 43.86) and mild OSAS (8.14 ± 0.44) (p = .00), indicating a strong association between OSAS severity and systemic inflammation. In line with this, monocyte counts were elevated in the severe group (0.94 ± 0.03 × 10³/μL) compared to mild (0.64 ± 0.02 × 10³/μL) and moderate (0.73 ± 0.07 × 10³/μL) groups (p = .00).

The PIV demonstrated a substantial elevation in the severe OSAS group (6522.30 ± 406.19) compared to both moderate (299.38 ± 38.04) and mild OSAS (290.00 ± 15.14) groups (p = .00). Similarly, the IRI was significantly higher in patients with severe OSAS (4.20 ± 0.48) than in those with moderate (2.08 ± 0.25) and mild OSAS (1.74 ± 0.08) (p = .00).

Taken together, these results show that patients with severe OSAS present marked increases in inflammatory markers, including NLR, SII, PIV, and SIRI, indicating an intensified systemic inflammatory and immune response associated with disease severity. Moreover, changes in red blood cell indices and decreased hemoglobin-to-inflammatory cell ratios support the hypothesis of a pro-inflammatory state in advanced OSAS. These findings underscore the potential utility of inflammatory markers in evaluating OSAS severity and guiding clinical management.

As shown in Table 6, additional clinical parameters varied across OSAS severity groups. Notably, patients with severe OSAS were significantly older (48.94 ± 1.11 years) than those with moderate (46.72 ± 1.79 years) and mild OSAS (42.76 ± 1.14 years) (p = .00), suggesting that age may contribute to disease progression or severity. While height and weight did not differ significantly between groups (p = .43 and p = .29, respectively), BMI tended to increase with OSAS severity. Patients with severe OSAS had the highest mean BMI (40.41 ± 4.76), although this trend did not reach statistical significance (p = .09).

Table 6.

Comparison of clinical parameters among OSAS severity groups.

Variable	Mild OSAS (mean ± SD)	Moderate OSAS (mean ± SD)	Severe OSAS (mean ± SD)	p
Age	42.76 ± 1.14	46.72 ± 1.79	48.94 ± 1.11	.00*
Height	1.699 ± 0.009	1.675 ± 0.014	1.685 ± 0.010	.43
Weight	108.53 ± 11.03	95.25 ± 1.93	113.80 ± 12.64	.29
BMI	37.26 ± 3.31	34.12 ± 0.81	40.41 ± 4.76	.09
Epworth score	7.49 ± 0.59	8.75 ± 0.96	11.05 ± 0.63	.00*
Total sleep time	372.79 ± 6.46	364.27 ± 11.10	397.41 ± 6.42	.02
Sleep efficiency (%)	79.99 ± 1.40	77.71 ± 2.25	82.41 ± 1.05	.19
Sleep latency start time (min)	38.27 ± 2.95	39.33 ± 5.81	23.91 ± 2.42	.00*
Wake after sleep onset (WASO)	55.22 ± 5.23	66.83 ± 7.90	60.90 ± 4.39	.15
REM latency (min)	185.88 ± 7.41	202.47 ± 16.66	200.80 ± 9.04	.32
N1 (%)	2.72 ± 0.38	3.27 ± 0.70	2.23 ± 0.35	.14
N2 (%)	79.75 ± 4.93	87.47 ± 10.61	78.73 ± 6.41	.03
N3 (%)	33.39 ± 2.36	40.60 ± 7.79	39.43 ± 3.30	.14
REM (%)	13.97 ± 1.29	18.41 ± 3.15	16.28 ± 1.41	.06
Apnea-hypopnea index	5.21 ± 0.43	22.78 ± 0.65	50.19 ± 1.92	.00*
REM AHI	9.03 ± 1.67	23.38 ± 3.95	48.27 ± 2.76	.00*
NREM AHI	5.14 ± 0.45	21.56 ± 1.04	48.97 ± 2.04	.00*
Supine total sleep time (%)	35.81 ± 2.43	39.11 ± 3.80	41.34 ± 2.79	.62
SUPINE AHI	11.01 ± 1.33	36.61 ± 3.11	62.59 ± 2.58	.00*
Non-supine AHI	3.36 ± 0.40	16.67 ± 3.40	39.05 ± 2.65	.00*
Oxygen desaturation index	18.56 ± 1.64	33.79 ± 1.85	67.01 ± 2.66	.00*
Mean SpO2 (%)	92.82 ± 0.25	91.58 ± 0.32	89.20 ± 0.41	.00*
Lowest SpO2 (%)	90.13 ± 6.50	76.14 ± 1.16	67.04 ± 1.50	.00*

Note. Comparison of demographic and clinical characteristics of patients with mild, moderate, and severe OSAS. Data are expressed as mean ± standard deviation. The * symbol was used for statistically significant values. OSAS: obstructive sleep apnea syndrome; BMI: body mass index.

The Epworth Sleepiness Scale score (ESS) was used to assess daytime sleepiness, ¹¹ a measure of daytime sleepiness, increased significantly with OSAS severity, from 7.49 ± 0.59 in mild OSAS to 11.05 ± 0.63 in severe OSAS (p = .00), indicating worse daytime sleepiness in more severe cases. Total sleep time was significantly longer in severe OSAS (397.41 ± 6.42 min) compared to mild (372.79 ± 6.46 min) and moderate OSAS (364.27 ± 11.10 min) (p = .02). Sleep latency, or time to fall asleep, was significantly shorter in severe OSAS (23.91 ± 2.42 min) compared to the other groups (p = .00), possibly reflecting increased sleep drive due to fragmented sleep.

No significant differences were observed in sleep efficiency (p = .19), WASO (p = .15), REM latency (p = .32), or time spent in N1 (p = .14) and N3 sleep stages (p = .14). However, patients with moderate OSAS spent a significantly higher proportion of time in the N2 sleep stage (87.47 ± 10.61%) compared to mild (79.75 ± 4.93%) and severe OSAS (78.73 ± 6.41%) (p = .03).

The AHI, a primary measure of OSAS severity, increased dramatically with disease severity, from 5.21 ± 0.43 in mild OSAS to 50.19 ± 1.92 in severe OSAS (p = .00). Similar trends were observed in REM AHI and NREM AHI, which were significantly higher in severe OSAS (48.27 ± 2.76 and 48.97 ± 2.04, respectively) compared to the other groups (p = .00). Supine AHI was significantly elevated in severe OSAS (62.59 ± 2.58) compared to mild (11.01 ± 1.33) and moderate OSAS (36.61 ± 3.11) (p = .00), while non-supine AHI showed a similar trend (p = .00).

The ODI was significantly higher in severe OSAS (67.01 ± 2.66) compared to moderate (33.79 ± 1.85) and mild OSAS (18.56 ± 1.64) (p = .00), reflecting increased hypoxic burden. Mean SpO₂ levels decreased significantly with increasing severity, from 92.82 ± 0.25% in mild OSAS to 89.20 ± 0.41% in severe OSAS (p = .00). Similarly, the lowest SpO₂ value was significantly lower in severe OSAS (67.04 ± 1.50%) compared to moderate (76.14 ± 1.16%) and mild OSAS (90.13 ± 6.50%) (p = .00), highlighting more pronounced oxygen desaturation events in severe cases.

These results demonstrate that patients with severe OSAS experience more significant respiratory disturbances, oxygen desaturation, and symptoms like daytime sleepiness compared to those with mild or moderate OSAS. Parameters such as AHI, REM AHI, ODI, and SpO₂ values are particularly indicative of worsening OSAS severity. These findings emphasize the progressive nature of OSAS and its systemic impact, underscoring the importance of timely diagnosis and treatment.

Table 7 shows the prevalence of comorbidities across OSAS severity levels. While conditions such as hypertension, diabetes, hyperlipidemia, CAD, COPD, snoring, and tiredness were more common in severe OSAS, none of the differences were statistically significant (p > .05). This indicates that the presence of these comorbidities does not vary meaningfully across OSAS severity groups in this cohort.

Table 7.

OSAS and comorbidities: cross-tables and analysis results.

Comorbidities	OSAS severity	No (n)	Yes (n)	Total (n)	p
HT	Mild OSAS	101	21	123	.505
	Moderate OSAS	29	8	37
	Severe OSAS	77	26	103
DM	Mild OSAS	107	15	123	.579
	Moderate OSAS	32	5	37
	Severe OSAS	84	19	103
HL	Mild OSAS	114	8	123	.755
	Moderate OSAS	35	2	37
	Severe OSAS	98	4	103
KAH	Mild OSAS	115	7	123	.347
	Moderate OSAS	33	4	37
	Severe OSAS	90	13	103
COPD	Mild OSAS	73	50	123	.723
	Moderate OSAS	24	13	37
	Severe OSAS	59	44	103
Snoring	Mild OSAS	23	100	123	.185
	Moderate OSAS	4	33	37
	Severe OSAS	11	92	103
Tiredness	Mild OSAS	40	83	123	.306
	Moderate OSAS	11	26	37
	Severe OSAS	24	79	103

The * symbol was used for statistically significant values.

Discussion

OSAS is a significant public health concern due to its physical, psychological, and socioeconomic impacts. In this study, we evaluated demographic characteristics, risk factors, OSAS severity using AHI scoring, and changes in novel inflammatory markers in relation to disease severity.

OSAS affects approximately 4% of middle-aged men and 2% of middle-aged women in developed countries and is associated with increased morbidity and mortality. ¹² Previous studies have demonstrated heightened risks of traffic accidents, ¹³ hypertension, ¹⁴ cardiovascular morbidity, ⁴ and impaired quality of life in OSAS patients. Consistent with the literature, our cohort commonly presented with comorbidities including Diabetes Mellitus, Hypertension, Hyperlipidemia, ischemic heart disease, and COPD.

The pathophysiology of OSAS involves complex inflammatory processes. Sleep fragmentation, intermittent hypoxia, and intrathoracic pressure changes contribute to systemic inflammation and oxidative stress. ¹⁵ These factors promote increased oxidative stress and sympathetic nervous system activation, leading to elevated inflammatory marker expression and endothelial dysfunction, all of which may contribute to cardiovascular and cerebrovascular disease development.^16–18

Biological markers reflecting systemic inflammation, such as NLR, SII, SIRI, and PIV, are emerging as practical and cost-effective tools in clinical assessment. ¹⁹ Our findings demonstrated higher NLR levels in severe OSAS patients compared to mild and moderate groups, aligning with previous meta-analyses and studies indicating NLR as a potential marker of systemic inflammation and disease severity.^20,21 The observed increases in neutrophils and decreases in lymphocytes are consistent with physiological stress responses involving cortisol and catecholamine-mediated immune modulation.^22,23

Importantly, inflammatory marker elevations appear to exhibit a dose-response relationship with OSAS severity. While mild and moderate cases may not induce marked inflammatory changes, patients with severe OSAS showed significantly elevated markers such as CRP, IL-6, NLR, and PLR.^24,25 Our results concur with this pattern, showing statistically significant increases in these markers in the high AHI group.

SII and SIRI, relatively novel indices, integrate multiple immune cell counts and have shown prognostic relevance in various inflammatory and cardiovascular conditions.^26–29 Previous studies have noted increased SII in OSAS and its association with disease severity.^30,31 Our findings support these observations and extend them by demonstrating elevated SIRI levels correlating with OSAS severity, an area with limited prior research.

Similarly, PIV, encompassing neutrophils, platelets, monocytes, and lymphocytes, has emerged as a comprehensive biomarker in oncology and inflammatory diseases.^32–35 To our knowledge, this is the first study investigating PIV in OSAS, revealing higher levels associated with greater disease severity. This suggests that PIV may offer additional prognostic value beyond traditional markers.

Limitations of the study

Single-center design

The study was conducted in a single tertiary care center, which may limit the generalizability of the findings. The demographic and clinical characteristics of the study population may not fully represent the broader OSAS patient population.

Retrospective study design

Due to its retrospective nature, the study is inherently prone to selection bias, incomplete data, and potential inaccuracies in medical records. Moreover, causal inferences cannot be established, and only associations can be evaluated.

Inability to fully control for confounding factors

Inflammatory markers such as CRP, NLR, and PLR can be influenced by a wide range of comorbid conditions (e.g. infections, autoimmune diseases, malignancies) and medications. Given the retrospective design, it was not feasible to fully account for or exclude all potential confounding variables.

Lack of association with common comorbidities

The study did not identify a statistically significant correlation between the severity of obstructive sleep apnea and common comorbid conditions such as hypertension, diabetes mellitus, and CAD. This may be attributed to limitations such as insufficient sample size, heterogeneity within patient groups, or the retrospective nature of comorbidity data collection, which may have introduced misclassification bias. Therefore, findings regarding comorbidities should be interpreted with caution.

Conclusion

This study's findings suggest that NLR, SII, PIV, and SIRI may be useful in predicting prognosis in patients with OSAS. The potential clinical application of these biomarkers could contribute to improved prognostic assessment and management of these patients. However, further research is warranted to clarify the mechanistic roles of PIV, SII, and SIRI in OSAS and to determine the most appropriate ways to apply these indicators in clinical practice.

We believe that identifying inflammation using these easily accessible, measurable, and cost-effective markers may help estimate the risk of complications such as cardiovascular disease in OSAS patients and guide preventive strategies. Our findings indicate that the combined assessment of four inflammatory indices (SII, SIRI, NLR, and PIV) might serve as a potential tool in evaluating OSAS prognosis.

Additional studies with larger sample sizes and longitudinal designs are necessary to validate these findings and support their translation into clinical decision-making.