Introduction

Obstructive sleep apnea syndrome (OSAS) is a common disease in children, characterized by a habitual snoring associated with prolonged, repeated events of partial and/or complete upper airway obstruction and hypercapnia and hypoxemia during sleep [1]. The symptoms of pediatric OSAS such as snoring, pauses or absence of breathing or mouth breathing and night sweats could cause morning fatigue, daytime sleepiness, growth inadequacy, hyperactivity, and low school performance [2]. Untreated sleep disorders can also be the reason for cardiovascular, metabolic, and neurocognitive disorders [3]. Polysomnography (PSG) has been used as a reference standard for the diagnosis of OSAS as it measures the apnea-hypopnea index (AHI) [4]. Due to the disadvantages of PSG such as being time-consuming, expensive, and having difficult accessibility, portable devices can also be used as an alternative diagnostic test to confirm the diagnosis of OSAS [5].

Orthodontic treatment is carried out to correct dental and skeletal discrepancies, and this treatment may be effective on nasorespiratory problems of the growing children with OSAS [6]. Rapid maxillary expansion (RME) is an effective and valid orthodontic treatment approach for managing OSAS in children with maxillary constriction [6, 7]. RME can be used as an alternative orthopedic treatment method to expand the maxilla until providing appropriate and stable maxillary width increase, maxillary and mandibular dental arch coordination, nasal respiratory function, opening the mid-palatal suture, preventing posterior uni- or bilateral cross-bite, and dental crowding among children with OSAS [6, 8,9,10,11].

In order to find out cheaper and effective alternative methods for the diagnosis of OSAS, studies have focused on hypoxia and inflammation-related molecules linking OSAS with metabolic alterations [12]. OSAS is predictably associated with altered protein expression patterns such as orosomucoids (ORMs) which have been implicated in the modulation of a variety of immune responses, endothelial function, and permeability [13, 14]. The fatty acid binding proteins (FABPs) are also potential indicators of OSAS and metabolic alterations [12]. Other proteins which show different protein expression profiles in the urinary samples of children with OSAS are gelsolin, perlecan, and kallikrein (KLK). Gelsolin is a 93-kDa cytosolic protein and it is activated by Ca2+ [15, 16]. Perlecan is a heparan sulphate proteoglycan that belongs to a family of glycosaminoglycans [1]. KLK1 reduction in urine samples emerged as a potentially valuable tool in the prediction of OSAS [1]. Tissue hypoxia due to repeated sleep apneas leads to increased serum levels of uric acid in patients with OSAS [17].

In this study, we aimed to investigate the relationship between the effect of 5-month semi-rapid maxillary expansion (SRME) orthodontic treatment and the magnitude of the changes in serum and urine biomarkers and respiratory parameters in children with OSAS.

Materials and methods

Thirty white children (16 male/14 female) attending the Department of Orthodontics of Ankara University School of Dentistry were enrolled in the study. We enrolled children who have clinical signs of maxillary transverse deficiency, malocclusion (high, narrow palate associated with deep bite, retrusive bite, or cross-bite), and signs and symptoms of OSAS (AHI > 1) including habitual snoring and oral respiration as identified by a parental questionnaire. This population was divided into two groups randomly as control (mean age, 11.46 ± 2.06) and SRME-treated group (mean age, 12.27 ± 1.93). This study was performed with the approval of Ankara University Health Sciences Institutional Review board (2012-37-4), and standard informed consent was obtained from the parents of each child.

Ear, nose, and throat assessment

At the beginning of the treatment, children underwent an ear, nose, and throat examination and every volunteer had an ear nose throat consultation before orthodontic assessment. According to the results of the consultation, patients who require adenotonsillectomy due to adenoid hypertrophy or tonsillar hypertrophy have been excluded from the study. Therefore, the etiological factors in OSAS except for mandibular retrognathia and maxillary construction were eliminated by ear, nose, and throat consultation in this study.

Semi-rapid maxillary expansion

A detailed personal and family history was obtained for all participants and a general clinical examination was performed. All investigations were carried out before orthodontic treatment (T0), and at the end of the treatment period (T1). Children with OSAS, as the control group, were followed up for 5.38 ± 1.36 months. SRME proceeded 5.01 ± 0.96 months in the treatment group. A modified McNamara RME device was cemented with an expansion screw connecting right and left maxillary dental acrylic segments to each other with a Hyrax type maxi screw (Forestadent) (Fig. 1a–c). The screw was turned twice a day for the first 7 days, then once a day until the palatal cusp of the upper molar came into contact with the buccal cusp of the lower molar [18]. After this initial treatment, when the maxillary arch was sufficiently over-expanded, the screw was fixed with a steel ligature wire.

Fig. 1
figure 1

Oral view of modified McNamara Hyrax appliance (a–c). ApneaLink Plus portable recording device (d)

Full size image

The measurements of the study

Pharyngeal linear widths and pharyngeal area measurements were done by using Dolphin computer software and a digital planimeter device (Ushikata X-Plan 380dIII/460dIII, Tokyo, Japan) on lateral cephalograms (Fig. 2a, b). Each area was measured in triplicate in an average manner.

Fig. 2
figure 2

a Lateral cephalometric landmarks and planes. Eb base of epiglottis, Ba basion, P palate point, Ptm the most superior point of the pterygomaxillary fissure, PNS posterior nasal spine, PPW1 the point where the plane parallel to the palatal plane and passing through PNS intersects the posterior pharyngeal wall, PPW2 the pharyngeal wall point intersecting the plane parallel from the palatal plane and passing through the point P, PPW3 the point where the plane parallel to the palatal plane and passing through Eb point intersects the posterior pharyngeal wall. b Pharyngeal area measurements. Nasopharynx: the area confined within the anterior and posterior pharyngeal wall on the side, the PNS-PPW1 line below, and the region above where the line drawn from Ba point to the sphenoid bone intersects the perpendicular line drawn from Ptm. Superior oropharynx: the area confined within the anterior and posterior pharyngeal wall on the side, the P-PPW2 line below, and the PNS-PPW1 line above. Inferior oropharynx: the area confined within the anterior and posterior pharyngeal wall on the side, the Eb-PPW3 line below, and the P-PPW2 line above

Full size image

Maxillary dental arch measurements (intercanin, interpremolar, intermolar distance) were performed on dental casts twice by using digital compasses (Absolute Digimatic Mitutoyo) before and after SRME treatment (Fig. 3).

Fig. 3
figure 3

Maxillary dental arch measurements. (1) Intercanine width: the distance between the cusp tips of the left and right upper canine tips. (2) Interpremolar width: the distance between the buccal cusp tips of the left and right upper first premolars. (3) Intermolar width: the distance between the mesio-buccal cusp tips of the left and right upper first molars

Full size image

Postero-anterior cephalometric analyses (nasal and maxillary widths) were done and evaluated by using Dolphin computer software (Fig. 4).

Fig. 4
figure 4

Postero-anterior cephalometric measurements. (1) Nasal width (NasW): the distance between the most lateral points of the lower nasal cavity. (2) Maxillary width (MaxW): the distance between the deepest points of the left and right maxilla

Full size image

Sleep analysis by using a multichannel device

The measurements for OSAS in control and SRME-treated groups were performed by a portable multichannel device at the baseline (C0 and T0) and after a 5-month follow-up (C1 and T1). The ApneaLink Plus (ResMed Corporation, Poway, CA) is a multichannel screening tool designed to screen for OSAS as shown in Fig. 1d. The device records air flow via a nasal cannula connected to a nasal pressure transducer. A belt placed around the chest records thoracic respiratory effort and electrocardiogram. Pulse oximetry is also recorded via the digital probe to measure hemoglobin saturation. The compact screening device was attached to the patient’s wrist and it provided information regarding the extent of sleep fragmentation.

Each ApneaLink study was scored manually by the same investigator based on pediatric criteria. AHI, flow limitation, snoring, blood oxygen saturation, and desaturation data were obtained by ApneaLink software. Respiratory events were scored in the presence of at least 3% of desaturation, when there is no movement and change causing snoring and pulse changes. All desaturations which are decreased more than 3% from baseline arterial oxygen saturation (SaO2) were quantified. OSAS was identified as an AHI > 1 [19].

ELISA for determination of OSAS-related parameters in serum and urine

The blood samples in serum separator tubes and first-morning urine samples were collected on the day after the sleep analysis. Serum and urine specimens were stored at − 80 °C until analysis. Serum and urine levels of ORM2, FABP4, perlecan, gelsolin, KLK1, and uric acid were assayed with commercially available ELISA kits used according to the manufacturer’s instructions; human ORM2, with a sensitivity of 0.05 ng/mL, and linearity between 0.05 and 15 ng/mL; human FABP4; human perlecan; human gelsolin; human KLK1 (Sunred Biological Technology Co., China), which has a sensitivity of 0.083 ng/mL, exhibits linearity between 0.1 and 30 ng/mL, and has inter- and intra-assay coefficients of variability 12 and 10%, respectively; and human uric acid (Cell Biolabs, Inc., San Diego, CA), which has a sensitivity of 0.5 μM.

Statistical analysis

Data are expressed as means ± SE (standard error). The changes before and after control/treatment were compared by Student’s t test and Mann-Whitney U test. The Pearson correlation test and Spearman’s rank correlation were applied to determine the relationship between parameters. p values less than 0.05 were considered as statistically significant.

Results

Study population

The orthodontic evaluation and parental questionnaire results demonstrated that all children had dental malocclusion with narrow maxilla and snoring. The characteristics of control and treatment groups were shown in Table 1. There were no significant differences in age, sex, and BMI distribution between groups. Half of the children (T0, n = 15) completed the 5-month SRME treatment (T1), and other 15 children (C0, n = 15) with OSAS were accepted as the control group and subjected to a follow-up for 5 months (C1). None of the volunteers experienced adverse effects when SRME was applied. The alterations in sleep parameters, pharyngeal area, dental arch, and posterio-anterior widths were measured before and after SRME treatment.

Table 1 SRME-induced alterations in demographic data, respiratory parameters, and biomarkers in children with OSAS
Full size table

According to the sleep test results, AHI decreased significantly in the treatment group from T0 to T1 and the control group from C0 to C1 (Table 1, p < 0.05). Oxygen desaturation index (ODI) and SaO2 values were not altered significantly in groups before and after the 5-month control/treatment duration (Table 1). There were also no differences in sleep parameters between before and after treatment (T1-T0) and control (C1-C0) (Table 1).

As shown in Table 2, a significant increase in nasopharyngeal (p < 0.05) and total pharyngeal area (p < 0.01) was observed after SRME in children with OSAS. Oropharynx area did not change before and after treatment. In the dental model measurement, intercanine, interpremolar, and intermolar area showed an increase after SRME treatment (Table 2, p < 0.001). The nasal, maxillary, and intermolar widths were increased by treatment (Table 2, p < 0.001 and p < 0.05).

Table 2 The changes of pharyngeal area, maxillary dental arch, and postero-anterior widths before and after SRME treatment
Full size table

The change in protein levels associated with OSAS in serum and urine and its relationship between arch widths

KLK1 levels were decreased significantly in the serum samples of children with OSAS after SRME treatment (Table 1, p < 0.05). An increase in serum ORM2 (p < 0.05) and a decrease in urine KLK1 and perlecan levels (p < 0.05 and p < 0.01) were observed after 5 months of follow-up in the control group as shown in Table 1. Changes in serum levels of ORM2 before and after treatment (T1-T0) also showed a significant difference compared to changes in serum levels of ORM2 before and after the follow-up period in the control group (C1-C0) (Table 1, p < 0.05). Although urinary uric acid levels were increased slightly in the control group and decreased in the treatment group, the alteration was not statistically significant at the end of the control and treatment period (Table 1). Also, FABP4 and gelsolin levels were not altered in serum after control/treatment (Table 1). Besides a significant negative correlation found between serum ORM2, perlecan, gelsolin, KLK1 levels, and intercanin width, a negative correlation at the p < 0.01 level was observed between serum uric acid levels and interpremolar width (Table 3). We also found an inverse relationship between serum ORM2, KLK1, and uric acid levels and intermolar width (Table 3, p < 0.05).

Table 3 Pearson correlation analysis between OSAS-related protein levels and measurements on the dental model after SRME treatment
Full size table

Discussion

In this study, we evaluated the beneficial effect of SRME orthodontic treatment in children with OSAS by comparing the levels of biomarkers as well as sleep test data, pharyngeal area, maxillary dental arch, and postero-anterior widths before and after treatment. We observed that pharyngeal airway space and maxillary dental arch width (intercanine, interpremolar, and intermolar) and postero-anterior width (nasal and maxillary) increased after SRME. AHI, an indicator of sleep apnea severity, was decreased following the treatment. The SRME-treated group showed a statistically significant decrease in serum KLK1 levels. In the control group, we observed a decrease in perlecan levels of urine and an increase in serum ORM2 levels at the fifth month. There was also a significant negative correlation between serum levels of OSAS-associated markers, ORM2, KLK1, gelsolin and perlecan, and intercanin width. Serum uric acid levels and interpremolar/intermolar width also demonstrated an inverse relationship in the study.

According to the results, the significant increase in pharyngeal airway space, intercanine, interpremolar, intermolar, and postero-anterior width can be considered as a positive progress induced by SRME. A study by Ashok et al. showed that RME is a useful treatment option for widening of maxilla [20]. AHI measured by an ApneaLink portable device was significantly decreased after control/treatment duration, while one of the most important markers of intermittent hypoxia, ODI, and desaturation (SaO2 < 95%) did not show any alteration before/after control and treatment. Similarly, Villa et al. reported statistically significant decreases in AHI of children with OSAS by RME [7, 8, 21]. An improvement in sleep parameters after RME was shown in children with sleep-disordered breathing [20]. Also, a recent study showed that RME treatment reduced AHI in children with OSAS [22]. Previous studies reported that a high-sensitive and specific determination of AHI could be provided by ApneaLink Plus portable devices [23, 24]. We can suggest that the clinical symptoms of OSAS may be resolved by expanding the pharyngeal airway, dental arch, and posterio-anterior widths and decreasing AHI at the end of the treatment period.

As a potential OSAS biomarker, serum ORM2 levels showed an increase in the control group after 5 months from the first specimen collection. Serum ORM2 levels showed an inverse correlation with intercanin width. Similarly, a previous study by Gozal et al. showed that urine ORM2 levels were increased in children with OSAS [1]. Magid et al. showed that ORM2 levels were increased in urine in association with acute inflammation [25]. Increased expression of ORM2 in the serum samples of children with OSAS may also induce the propagation of inflammation-related markers like FABP4.

FABP4 is a cytosolic lipid chaperone, and our present data showed that serum levels of FABP4 were increased in OSAS and decreased at a clinical level following SRME treatment. As a supporting evidence by Bhushan, children with OSAS showed an increased morning plasma FABP4 levels [26]. Several studies also reported a higher serum FABP4 levels in adults with OSAS [27, 28]. Lam et al. have reported that plasma FABP4 levels were significantly associated with sleep hypoxemia parameters, including duration of oxygen desaturation and minimal oxygen saturation, independent of age and obesity [28]. FABP4 levels may be upregulated markedly in subjects with high AHI and severe degree of hypoxemia. In our study, the sleep apnea in children with OSAS may not be severe enough to cause tissue hypoxia. The different intensity of intermittent hypoxia in OSAS subjects results in different expression profiles of FABP4. We also suggested that increased FABP4 levels may also be associated with inflammation arisen from increased hypoxia during OSAS.

At present study, urine perlecan levels were decreased at the fifth month in the control group, while we did not observe any change in the treatment group. The inflammatory events can induce an increase in perlecan levels [29]. OSAS is a disease which has an inflammatory component [30]. No significant change of perlecan levels of the treatment group may be the result of the beneficial effect of RME on OSAS. In addition, the decreased urine perlecan levels of the control group may be related to the mild severity of OSAS. The urine perlecan level has been suggested as a potential clinical marker for many diseases. Serum contains low levels of perlecan relative to those found in urine [31, 32]. Krishna et al. [33] reported an increased expression of perlecan in the urinary proteome of children with OSAS. Considering a negative relationship between serum perlecan levels and intercanin width in this study, we can suggest that perlecan levels, as a hypoxia-related marker, can be affected by dental arch width. The marker of perlecan can be used to understand if the hypoxic condition is still present in children with OSAS or not.

In the present study, unchanged gelsolin levels before and after control/treatment may be related to the mild severity of OSAS. We suggest that the mild OSAS could not induce hypoxia and inflammation-related changes like protein permeability, catabolism, and excretion in children. The role of gelsolin in inflammation associated with OSAS is still unknown. A study by Zhang et al. [34] showed that treatment with gelsolin reduces brain inflammation in mice. Another study showed that gelsolin expression is necessary for the development of pulmonary inflammation [35]. A significant negative relationship between serum gelsolin levels and intercanin width may suggest that an increase in intercanin width may prevent hypoxia through decreasing gelsolin levels in patients with OSAS. Our data showed that KLK1 levels, another important marker for OSAS, were decreased in both the control and SRME-treated groups. Similarly, Gozal et al. [1] showed that KLK1 levels were significantly reduced in urine samples of children with OSAS. Previous studies showed that tissue KLK1 protected against retinal and neuronal ischemic damage in different animal models [36, 37]. We can suggest that a decrease in serum KLK1 levels can exert inflammatory effects in OSAS. SRME treatment may prevent hypoxia and inflammation related with OSAS by increasing KLK1 levels.

According to our results, a slightly lower uric acid level of urine in response to SRME is an important clinical finding, despite the low number of individuals in study groups. Several studies demonstrated a positive correlation between the severity of sleep apnea and increased levels of serum uric acid levels as a reflection of oxidative stress and hypoxia [38,39,40]. Oxidative stress is the result of a hypoxia-reoxygenation phenomenon and is associated with elevated serum uric acid levels [17, 40]. Van Hoorenbeeck et al. have reported that improvements in uric acid levels were correlated with improvements in sleep parameters such as ODI [39]. We can suggest that sleep parameters such as ODI and SaO2 values can show positive changes depending on uric acid levels. The findings of studies may vary because of different skeletal and dental expansion rates in different age groups. Our findings indicated that there was a decrease in uric acid excretion due to maxillar expansion and decreased airway blockage by SRME treatment.

Although this study is the first to evaluate the relation between SRME treatment and biochemical markers, it has some limitations. Therefore, further studies are needed to make certain interpretations about the role of relevant biochemical markers in patients with OSAS. A limitation of our study was the short follow-up, the small number of subjects, and the mild degree of OSAS. Our results pointed to the alteration of some biomarker levels (perlecan, KLK1, uric acid) in snoring children with SRME treatment. Our findings warrant further studies based on larger numbers of subjects and a population with severe OSAS. In conclusion, the 5-month personalized SRME treatment showed beneficial effects on children with OSAS and may be considered as a useful approach in children with OSAS to deal with abnormal breathing during sleep. Future studies which aim to characterize the protein maps in OSAS are required to provide reliable and convenient alternative clinical approaches in the treatment of patients with OSAS.