Introduction

Obstructive sleep apnea (OSA) is characterized by recurrent upper airway obstructions during sleep, and it is a frequently encountered disorder affecting approximately 15–30% of the adults in the United States [1, 2]. Apnea-related hypoxia, hypercapnia, and blood pressure fluctuations cause production of various proinflammatory cytokines and trigger a vicious cycle that results in vascular endothelial damage and systemic inflammation in OSA, which forms the basis for a number of comorbidities including cardiovascular, neurodegenerative, and cognitive disorders [3, 4]. Endothelial function is frequently impaired in OSA even in the absence of significant cardiac or vascular disorders [5].

Endocan is a 50 kDa proteoglycan containing a single dermatan sulfate chain called endothelial specific molecule-1 (ESM-1) [6]. Significantly high endocan levels have been reported in disorders going together with vascular endothelial damage including type 2 diabetes mellitus (T2DM), OSA, and septic shock, and endocan has been suggested as a significant endothelial damage marker [7-10].

Serglycin is a proteoglycan and has an essential role in the modulation of activated immune system cells and inflammatory reactions [11]. Inflammatory cells synthesize and store serglycin in granules to react with mediators such as growth factors, cytokines, proteases, and chemokines and when needed [12, 13]. Serglycin may be involved in the development of vascular disorders, including atherosclerosis [14].

In the light of aforementioned information, we have hypothesized that endocan and serglycin may be used as predictive markers in OSA, and we aimed to investigate the serum endocan and serglycin levels in OSA patients. This study is the first one that has investigated serglycin levels in OSA patients.

Material and methods

This prospective study included 171 consecutive patients aged 18–65 years who were admitted to the sleep laboratory of a tertiary referral center with the complaints of daytime sleepiness, witnessed sleep apnea, and/or snoring and who underwent all-night polysomnography (PSG). A total of 93 patients were excluded due to history of hypertension, diabetes mellitus, hematological disorders, chronic inflammatory disorders, hyperlipidemia, stroke, liver conditions, malignancy, myocardial infarction, and coronary artery surgery or refusal to participate in the study. At the end, 78 patients were included in the study.

Blood sampling

Blood samples for endocan and serglycin levels and other standard laboratory tests were obtained on the day of PSG. An automated complete blood analyzer device was used to analyze complete blood count (CBC) parameters. Enzyme-linked immunosorbent assay method was employed to determine plasma serglycin and endocan levels, as described previously [8, 15].

A sleep technician-supervised full-night PSG was recorded with Alice 5 PSG device (Philips Respironics, The Netherlands) during spontaneous sleep, as described previously [4]. A PSG and sleep disorder-certified ENT physician scored all PSG data manually [16]. After PSG, the patients were divided into four groups in relation with their apnea–hypopnea indexes (AHI) as follows: simple snoring (control group) (AHI < 5), mild OSA (5 ≤ AHI < 15), moderate OSA (15 ≤ AHI < 30), and severe OSA (AHI ≥ 30).

The groups were compared for endocan and serglycin levels and their correlations with OSA severity. The correlations with demographic data and PSG findings were also investigated.

Local ethics committee approved the study protocol (Project No. E-16–933). The study was conducted in agreement with the ethical principles of Declaration of Helsinki. All participants provided their informed consents before participating in the study.

For the statistical analyses, IBM-SPSS for Windows v.21.0 software (IBM Corporation, Armonk, NY, USA) program was used. Continuous variables were presented as mean ± standard deviation, and categorical variables were presented as percentages. Data with normal distribution were compared with an independent sample t-test. Two groups with abnormal distributions were compared with the Mann–Whitney U test. Three or more groups without normal distributions were compared with the Kruskal–Wallis test. Pearson’s correlation analysis was employed for correlation analysis. The independent predictors of OSA were determined with multiple logistic regression analysis. Statistical significance was considered as \(p < 0.05\).

Results

Table 1 shows endocan and serglycin levels, CBC parameters, PSG findings, and demographic features of the control and OSA groups. There were 24 (38.7%) female, 38 (61.3%) male, and a total of 62 patients in the OSA group. The control group consisted of 7 (43.75%) male, 9 (56.25%) female, and a total of 16 patients. The participants’ mean age was 47.09 ± 8.44 years in the OSA group and 44.81 ± 4.60 years in control group. Age and gender were similar in two groups (\(p = 0.205\) and \(p = 0.302\), respectively). There was no significant difference between the two groups for CBC parameters (\(p > 0.05\)). BMI value was significantly higher in the OSA group compared to the control group (\(p = 0.021\)). The OSA and control groups had significantly different endocan and serglycin levels (\(p = 0.001\), for both) (Table 1).

Table 1 Comparisons of the endocan, serglycin, complete blood count parameters, PSG findings, and demographic features of the control and OSA groups
Full size table

Serglycin and endocan levels in relation with OSA severity

Table 2 shows the comparisons of endocan and serglycin levels and BMI and PSG findings in the control group and the OSA subgroups.

Table 2 Comparisons of endocan and serglycin levels and BMI and PSG findings in the control group and the OSA subgroups
Full size table

The mean endocan levels were 214.43 ± 43.08, 242.57 ± 42.14, 282.07 ± 36.98, and 292.06 ± 61.04 pg/ml, in control, mild, moderate, and severe OSA subgroups, respectively. Endocan levels differed significantly among the groups (\(p = 0.001\), Fig. 1). Except for control–mild OSA and moderate OSA–severe OSA groups, binary comparisons showed significant differences between other subgroups (Table 2).

Fig. 1
figure 1

Comparisons of endocan levels between the OSA subgroups and the control groups

Full size image

The mean serglycin levels were 12.91 ± 2.79, 17.75 ± 4.18, 17.93 ± 4.07, and 20.75 ± 4.29 ng/ml, in control, mild, moderate, and severe OSA subgroups, respectively. Serglycin levels were significantly different in the groups (\(p = 0.001\), Fig. 2). The serglycin level was significantly different in the control group compared to the other OSA subgroups in binary comparisons. On the other hand, binary comparisons of the other OSA subgroups did not reveal any significantly difference for serglycin levels (Table 2).

Fig. 2
figure 2

Comparisons of serglycin levels between the OSA subgroups and the control groups

Full size image

The correlations of serglycin and endocan levels with BMI and PSG findings

The serglycin and endocan levels had positive correlations with AHI, ST90, and BMI (\(p =0.001\) for AHI and ST90, \(p = 0.015\) and \(p = 0.019\) for BMI, respectively) and negative correlations with Min O2 (\(p = 0.001\)) (Table 3 and Figs. 3 and 4).

Table 3 The correlations of serglycin and endocan levels with BMI and PSG findings
Full size table
Fig. 3
figure 3

Correlation of endocan and apnea–hypopnea index

Full size image
Fig. 4
figure 4

Correlation of serglycin and apnea–hypopnea index

Full size image

Multivariate logistic regression analysis for existence of OSA

On univariate logistic regression analysis, it was found that serglycin and endocan levels and BMI were predictors of OSA. Multiple logistic regression analysis showed that endocan and serglycin levels were independent predictors for OSA (\(p = 0.027\) and \(p = 0.003\), respectively) (Table 4).

Table 4 Multivariate logistic regression analysis of parameters in terms of existence of OSA
Full size table

Discussion

In the present study, we have demonstrated that elevated endocan and serglycin levels are predictors for OSA. Furthermore, we have showed for the first time in literature that serglycin is correlated with OSA and is an independent predictor for OSA.

OSA results in systemic inflammation and may be accompanied by various comorbid disorders including cardiovascular conditions, stroke, and cognitive disorders. The systemic inflammation in OSA has been supposed to be triggered by recurrent sleep-related upper airway obstructions, and resulting vascular endothelial injury and the comorbid disorders create a vicious cycle that aggravates OSA [17]. Hypoxia, sympathetic nervous system overactivity, and inflammation due to repetitive upper respiratory tract obstructions trigger excrete of vasoactive substances and cause endothelial dysfunction in OSA [18-20].

Inflammation, endothelial dysfunction, and OSA are strongly linked [21]. Repetitive obstructive sleep apnea events result in endothelial cell activation, signifying onset of an inflammatory response [22], and it has been reported that endothelial dysfunction may occur even without any clinical evidence of cardiovascular disorders in OSA patients [5].

It has been proposed that endocan promotes adhesion of monocytes to endothelial cells and it plays a significant role in endothelial dysfunction in inflammatory disorders [23]. High endocan levels have been reported in various disorders including chronic renal failure, sepsis, transplant rejection, hypertension, and T2DM [8, 24-27]. Balamir et al. investigated endothelial dysfunction indicators in T2DM patients and reported endocan as an endothelial dysfunction indicator positive correlated with the other indicators in T2DM [8].

The relevance of endocan has been investigated in patients with OSA-related comorbidities. However, the researchers excluded the patients with OSA alone, which is a condition that may increase the endocan levels by itself.

Altıntas et al. reported high endocan levels in the patients with OSA and revealed that continuous positive airway pressure (CPAP) therapy caused improvement in the serum endocan levels after 3 months. They also indicated endocan as a useful marker for monitoring treatment response in OSA [10]. Another study reported endocan as an independent predictor for severe OSA [9]. Sun et al. have reported that endocan may be employed as a biomarker to monitor development and progression of coronary artery disease (CAD) in OSA [23]. Kanbay et al. found significantly higher endocan levels and lower flow-mediated dilatation (FMD) values in OSA patients when compared to healthy controls and suggested that serum endocan level and FMD were correlated. They also showed that BMI was positively correlated with endocan levels [21]. In our study, the endocan levels were higher in OSA patients compared to the controls, and endocan levels showed a positive correlation with OSA severity. Our results indicated endocan as an independent predictor for presence of OSA.

Obesity is a significant predisposing issue for OSA [28], and AHI increases as the weight of the patient increases [29]. In our study, we found that BMI was correlated with presence and severity of OSA. Univariate analysis showed correlation of BMI with presence of OSA; however, multivariate analysis did not show any correlation.

Recently, Yazan et al. reported significant correlation of serum endocan levels with AHI in hypertensive OSA patients although this correlation was not evident in normotensive ones [27]. Although we excluded hypertensive patients, we found higher endocan levels in OSA patients in our study. This result offers evidence that endothelial dysfunction in OSA is multifactorial and cannot be attributed to hypertension alone. Furthermore, we may conclude that endothelial dysfunction may occur in OSA even in the absence of clinically evident cardiovascular disease, and cardiovascular disorders may be diagnosed at an early stage by determining the endocan levels in OSA patients.

Serglycin is mainly a hematopoietic cell proteoglycan and is released by hematopoietic cells such as mast cells, natural killer cells (NK), platelets, and macrophages as well as non-hematopoietic endothelial cells. Serglycin release is triggered by cytokines in endothelial cells and monocytes, and this plays a significant role in atheroma formation and other inflammatory processes [12, 30-34].

OSA patients have higher inflammatory mediator levels including intracellular adhesion molecule-1 (ICAM-1), IL-6, IL-8, and tumor necrosis factor-alpha (TNF-α) compared to the normal population [35, 36]. Liposaccharide, TNF-α, and interleukin 1-β are critical inflammatory mediators that upregulate synthesis of serglycin [32]. This signifies that serglycin plays a role in the augmentation of inflammatory response. Serglycin plays an important part in endothelial functions. Higher serglycin synthesis and secretion was reported in activated endothelial cells compared to the inactive endothelial cells [37]. In fact, serglycin plays various functions and physiological roles in reproduction, apoptosis, cell growth, immunity, and hemostasis and is also involved in several inflammatory conditions.

Considering that serglycin may play a role in vascular disorders and coronary artery disease, Bolayir et al. performed a study and reported that plasma serglycin level was correlated with the existence of coronary artery disease and the severity of coronary stenosis in patients with stable angina pectoris [38]. Ilgın et al. found elevated serglycin levels in ST-segment elevation myocardial infarction (STEMI) patients and reported a positive relationship with troponin and CRP levels, and they suggested serglycin as a useful marker in patients with STEMI [34]. Kundi et al. reported significantly and independently higher serglycin levels in patients with coronary artery ectasia (CAE), and they suggested that inflammation might play a role in the development of CAE. In a recent study, Doncheva et al. investigated the effect of serglycin expression on diet-induced adipose tissue inflammation and obesity. They found that serglycin played a role in the occurrence of obesity-induced adipose tissue inflammation [39]. In our study, we found higher serglycin levels in patients with OSA compared to the control group. In addition, we showed serglycin as an independent predictor for presence of OSA. In our study, high BMIs of the OSA patients might have affected the results concerning serglycin levels. Therefore, apart from multifactorial endothelial damage and inflammation caused by OSA, high BMI values might have caused high serglycin levels in our study.

Our study has some limitations. First, serglycin and endocan levels were not measured after CPAP therapy. Second, this is a single-center study. Third, we did not evaluate the smoking status or CRP and TNF-α levels. However, we excluded the patients with inflammatory conditions since those conditions could affect serum serglycin and endocan levels, and this strengthens our results. Further multicenter studies are required on larger and more homogeneous study populations.

In conclusion, to the best of our knowledge, this is the first study that investigated the correlation of serglycin with OSA. We found that high serglycin and endocan levels were correlated with the presence and severity of OSA. Serglycin and endocan might be useful early markers for cardiovascular comorbidities in OSA, before the onset of clinically apparent disease. More comprehensive studies are needed to further bring underlying mechanisms and the predictive value of serglycin and endocan to light in patients with OSA.