The effect of obstructive sleep apnea syndrome on serum S100B and NSE levels: a systematic review and meta-analysis of observational studies

Obstructive sleep apnea syndrome (OSAS) is a common disease that manifests as repeated events of nighttime breathing cessation because of upper airway collapse [1]. Epidemiological studies have shown that OSAS has a high incidence in the general population, with a prevalence of 2–4% [2]. There is a correlation between OSAS and increased platelet adhesiveness, vascular endothelial dysfunction and early symptoms of atherosclerosis, indicating an increased risk of vascular effects, such as stroke, in OSAS patients [3]. OSAS is accompanied by structural brain changes, and while how the brain is changed in OSAS remains unclear [4], OSAS likely causes brain injury [5]. Serum neuron-specific enolase (NSE) and astrocytic protein (S100B) concentrations have been examined under both clinical and experimental conditions to explain the relationship between neural cells and astrocytes in pathological situations [6, 7]. An elevated NSE level indicates neuronal injury, whereas an elevated S100B level may reflect either glial injury or astrocytic responses to neural damage [8]. These markers could serve as sensitive indicators of brain injury development [9]. The appearance of a biochemical marker of cerebral injury could serve as a considerable advantage in OSAS for identifying even small brain injuries and improving the efficacy of treatment [10].

The purpose of the present meta-analysis was to explore the effect of OSAS on serum S100B and NSE concentrations in observational studies.

This meta-analysis was done based on the guidelines for the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [11].

A total of 63 studies were found in five databases. After removing duplicated studies, 31 articles were screened (Fig. 1). Out of the 31 studies screened, 21 articles were not relevant and therefore were excluded. Ten studies were evaluated for eligibility and one study identified by a manual search was added. Therefore, 11 studies were screened, 2 of which were excluded (one was an animal study and one did not report the relevant data). Finally, nine studies were included and analyzed in this meta-analysis.

Some characteristics of the studies included in the meta-analysis are shown in Table 1. The studies were published from 2002 to 2017. Two studies were reported in Germany [10, 21], two in Brazil [15, 17], one in the Czech Republic [16], three in Turkey [18‐20], and one in Egypt [22]. Three were cross-sectional studies [10, 16, 17], and six were case-control studies [15, 18‐22]. Other information is shown in Table 1.

The meta-analysis evaluated serum S100B and NSE markers related to brain damages in OSAS patients. It was observed that serum levels of these markers in the patients were significantly higher than the controls, but the level changes were no significant after compared to before sleep.

OSAS can be related to cardiovascular and metabolic disorders, high blood pressure, diabetes, obesity, coronary artery illness [23, 24], stroke, and heart failure [25, 26]. OSAS can also cause cellular injury in the central nervous system (CNS) [27]. This meta-analysis shows that the serum S100B and NSE concentrations were significantly elevated in OSAS patients compared with the controls but that no significant changes occurred in the serum S100B and NSE levels of the patients during sleep. In addition, the S100B level was not significantly correlated with OSAS severity. Several studies [15, 19‐22] reported higher serum S100B levels in OSAS patients than in the controls, indicating a significant difference in four studies [15, 19, 21, 22]. Two studies [15, 18] showed that the serum NSE level was significantly in OSAS patients higher than in the controls. The study by da Silva et al. [17] showed a significant difference in the serum S100B level after sleep compared with before sleep in the patients, but this difference was not confirmed by Sonka et al. [16]. In addition, two studies [10, 17] showed no significant difference in the serum NSE level after sleep compared to before sleep in the patients. In three studies [19, 21, 22] that assessed the S100B level in OSAS patients based on severity grade, there was no significant correlation between the S100B level and OSAS severity.

Studies of the cerebrospinal fluid (CSF) levels of S100B and NSE in patients with neurological injuries have shown a quantitative correlation between the degree of cell damage in the CNS and the level of these markers in CSF [28]. Clinically, elevated levels of NSE and S100B protein have been observed early following ischemic stroke, and elevated levels of S100B protein have been observed after intracerebral and subarachnoid hemorrhage [29]. The serum S100B level may increase not only due to glial injury but merely by opening the blood-brain barrier [30]. Cerebrovascular and neuropsychiatric disorders [31, 32], trauma, and stroke [33] can cause increased serum S100B protein levels due to S100B release from astrocytes.

In patients with mild and severe OSAS without neurological signs or a history of cerebrovascular events, there were no elevations in the serum NSE or S100B level [10]. da Silva et al. [17] reported a correlation between depression and elevated S100B levels in unhealthy obese patients. In addition, older patients are more likely to experience brain injury due to OSAS than younger patients [16]. It also appears that serum S100B levels are not dependent on age or sex [19, 34]. Duru et al. [19] reported no significant relationship between the serum S100B level and related factors of OSAS, including the AHI and patient characteristics, such as age and BMI. However, Sonka et al. [16] showed that the serum S100B concentration was inversely related to the AHI and directly related to both the basal and mean minimal oxygen saturation (SaO2) in line to the results of Riad’s study [22]. It has been confirmed that low mean nighttime SpO₂ significantly associated with an elevated risk of a central nervous system events [35]. In OSAS patients, there is main effect of BMI on S100B level [22]. Duru et al. [19] reported that serum S100B levels did not correlate with age and BMI. Another study [36] showed that there is a negative correlation between serum S100B level and age in patients below 20 years, but after 20 years, S100B level has not appear to vary with age. Braga et al. [15] found a mild effect of age on S100B level, whereas Riad et al. [22] didn’t find any correlation between them. The studies checking SaO2 in OSAS patients compared to the controls showed that S100B [15, 16] and NSE [15] levels had no significant correlation with minimum SaO2 levels. Another study [17] showed a negative correlation between S100B levels and minimum SaO2 levels. Despite a few reported studies, S100B or NSE levels can be correlated with age, sex, AHI, and SaO2 levels. The studies reported different assay methods for S100B and NSE that some methods had a poor inter-assay that it can lead to inconsistent results [21] and reduce the reliability of results of the studies included in the meta-analysis. Therefore, it needs more studies in the future to confirm the effect of S100B or NSE levels on OSAS using unit methods for increasing the accuracy and reliability of the results.

The limitations of the study included the few studies in each analysis and differences among the studies in terms of age, sex, BMI, AHI, and different methods (different cut-offs or inter-assay).