Airway Resistance in Children with Obstructive Sleep Apnea Syndrome

Abstract

Study Objectives:

Enlarged tonsils and adenoids, the main cause of obstructive sleep apnea syndrome (OSAS) in children, results in upper airway (UA) loading. This contributes to the imbalance between structural and neuromotor factors ultimately leading to UA collapse during sleep. However, it is unknown whether this UA loading can cause elevated airway resistance (AR) during wakefulness. We hypothesized that children with OSAS have elevated AR compared to controls and that this improves after OSAS treatment.

Methods:

Case control study performed at an academic hospital. Children with OSAS and nonsnoring healthy controls underwent baseline polysomnography and spirometry, and AR measurement by body plethysmography while breathing via an orofacial mask. Children with OSAS repeated the previously mentioned tests after adenotonsillectomy.

Results:

31 OSAS participants (mean age ± SD = 9.7 ± 3.0 y, obstructive apnea-hypopnea index (OAHI) median [range] = 14.9 [2–58.7] events/h, body mass index [BMI] z = 1.5 ± 1) and 31 controls (age = 10.5 ± 2.5 y, P = 0.24; OAHI = 0.4 [0–1.4], P < 0.001; BMI z = 0.9 ± 1, P = 0.01) were tested. OSAS AR at baseline was 3.9 [1.5–10.3] cmH2O/L/sec and controls 2.8 [1.4 – 6.2] (P = 0.027). Both groups had similar spirometry results. 20 patients with OSAS were tested 6.4 ± 6.6 mo after adenotonsillectomy. OAHI decreased from 15.2 [2.1–58.7] to 0.5 [0 – 5.1] events/h postoperatively (P < 0.001), and AR decreased from 4.3 [1.5 – 10.3] to 2.8 [1.7 – 4.7] cmH2O/L/sec (P = 0.009).

Conclusions:

Children with OSAS have elevated AR that decreases after treatment. This is likely because of upper airway loading secondary to adenotonsillar hypertrophy and may contribute to the increased frequency of respiratory diseases in untreated children with OSAS.

Citation:

Tapia IE, Marcus CL, McDonough JM, Kim JY, Cornaglia MA, Xiao R, Allen JL. Airway resistance in children with obstructive sleep apnea syndrome. SLEEP 2016;39(4):793–799.

Significance.

Children with obstructive sleep apnea syndrome have impaired arousal responses to inspiratory resistive loading during sleep. However, it is unknown whether they have elevated airway resistance compared to controls and whether this improves after treatment. This study shows that children with obstructive sleep apnea have elevated airway resistance that improves after treatment.

INTRODUCTION

The obstructive sleep apnea syndrome (OSAS) is thought to result from an imbalance between structural or anatomical factors and neuromotor control.^1,2 In children, the main cause of OSAS is enlarged tonsils and adenoid, which may increase airway resistance. In fact, it has been previously shown that children with OSAS have impaired arousal responses to inspiratory resistive loading during sleep compared to controls.³ This may be because, compared with control patients, children with untreated OSAS are used to breathing at higher airway resistance levels. It has also been reported that nasal resistance in children determined by anterior rhinomanometry correlates with OSAS severity as measured by the obstructive apnea-hypopnea index (OAHI).^4,5 However, many children with OSAS do not breathe exclusively through the nose, but favor mouth breathing. The classic oral method of measuring airway resistance excludes the nose as subjects wear nose clips and breathe through a mouthpiece. Hence, it is unknown whether total airway resistance, including nasal and orally measured airway resistance, is increased in children with OSAS compared to controls, nor it is known whether this changes after OSAS treatment. To clarify this, we designed a study to test the hypothesis that children with OSAS have elevated airway resistance compared to control patients and that this would improve after surgical treatment of OSAS with adenotonsillectomy. Airway resistance was measured via plethysmography using an orofacial mask to yield a resistance value that combined both nasal and orally measured airway resistance and thus, approximated the airway resistance from natural breathing.

METHODS

Children with OSAS and age- and height-matched healthy non-snoring control participants underwent baseline polysomnography using pediatric standard techniques and scoring.^6,7 Airway resistance and lung volumes were measured by body plethysmography (Morgan Scientific, Haver-hill, MA) while breathing via an orofacial mask (Respironics Comfort Gel) to include nasal resistance. Subjects also performed spirometry (Morgan Scientific) via a mouthpiece while wearing noseclips. American Thoracic Society/European Respiratory Society standards were followed for both tests.^8,9 Considering there are no z-scores available for airway resistance measurements obtained via an orofacial mask, these values are reported as absolute numbers in cm H 2O/L/sec. Twenty of the children with OSAS were retested 6.4 ± 6.6 mo after surgical treatment: adenotonsillectomy. The Institutional Review Board of The Children's Hospital of Philadelphia approved the study. Informed consent was obtained from the parents/guardians of subjects, and assent from participants aged 7 y and older.

Study Group

Children between 6–16 y of age were included. The younger age limit was selected to exclude children who could not cooperate with testing. The older age limit was chosen to avoid overlap with adult OSAS. Children with OSAS and control participants with a clinical history of asthma were excluded. Patients with OSAS were recruited from the Sleep Center at The Children's Hospital of Philadelphia following a recent clinical polysomnogram. Normal controls were recruited from the general community by means of advertisements. OSAS was defined as having an OAHI ≥ 2/h, and control subjects were included if they were asymptomatic and had an OAHI < 1.5/h.^10–13

Statistical Analysis

Statistical analysis was performed with SAS 9.4 (SAS Institute Inc., Cary, NC) and R 3.1.0 (R Core Team (2014)). The Kolmogorov-Smirnov test was used to test for normality. Categorical variables were summarized as count and percentage, and compared between groups using the Fisher exact test. Continuous variables were summarized as mean ± standard deviation if normally distributed or as median (range) if not, and were compared using the paired or unpaired t-test or Mann-Whitney rank-sum test or Wilcoxon signed-rank test, as appropriate. Correlation between continuous variables was evaluated by Pearson correlation coefficient. To analyze the differences in airway resistance between children with OSAS and control participants, a stepwise regression was performed based on Bayesian Information Criterion (BIC) in order to build the most relevant model from a pool of variables including: participant type (OSAS/Control), obesity, height, sex, and the interaction effects between participant type and these variables.¹⁴ The model with participant type, height, and the interaction between them was found to be the best model according to BIC and therefore, these results were reported here. A linear mixed- effect model was used to estimate the change in airway resistance before and after treatment, adjusting for changes in height, as children were expected to grow during this interval and airway resistance is highly correlated with height, while accounting for the repeated measurements.¹⁵ The stepwise regression and linear mixed-effect models used log-transformed airway resistance to normalize its distribution. A value of P < 0.05 was considered statistically significant.

RESULTS

Study Group

Thirty-one participants with a wide range of OSAS and 31 age-matched nonsnoring healthy control participants were recruited (Table 1). Control participants were thinner than children with OSAS (P = 0.01). Twenty participants with OSAS were reevaluated after surgical treatment. There was no significant correlation between BMI z-scores and OAHI in OSAS (r = −0.02, P = 0.93). Participants did not have evidence of allergic rhinitis at the moment of the study. However, two participants with OSAS were receiving cetirizine, one was prescribed loratadine, and one was prescribed intra-nasal fluticasone by their primary pediatrician for seasonal allergies.

Table 1.

Study group demographics and polysomnography results.

Airway Resistance

Children with OSAS had greater absolute airway resistance values compared to control participants (Table 2 and Figure 1). They also had lower absolute conductance values compared with control participants (Table 2). There was no significant correlation between airway resistance and OAHI in participants with OSAS (r = 0.02, P = 0.9) or in controls (r = 0.24, P = 0.19). The stepwise regression model with participant type, height, and the interaction between them was found to be the best model according to BIC. Of note, obesity did not reach statistical significance on previous steps of this regression model. Using this model, the difference in airway resistance in log-transformed scale between children with OSAS and control participants was statistically significant (P = 0.031) and became even more significant when the interaction between participant type and height was considered (P = 0.004). For example, the airway resistance of participants with OSAS with the average height of 143.9 cm (considering all participants regardless of OSAS status) was 23% higher than that of control participants of the same height. As airway resistance is influenced by height, this percentage was greater for shorter participants and smaller for taller participants.

Table 2.

Airway resistance and pulmonary function data in children with obstructive sleep apnea syndrome versus control participants.

Figure 1.

Airway resistance in children with obstructive sleep apnea syndrome (OSAS) and control subjects. The box represents the interquartile range, the central line represents the median, the whiskers represent the 5^th and 95^th percentiles, and the dots represent the outliers. Children with OSAS had elevated airway resistance z-scores compared to control subjects.

Specific airway resistance (sRaw) and specific airway conductance (sGaw) were similar between participants with OSAS and control participants (Table 2).

Pulmonary Function Tests

Participants with OSAS and controls had similar forced expiratory volume in 1 sec/forced vital capacity (FEV₁/FVC) ratios, and percent predicted values of forced vital capacity (FVC), forced expiratory volume in 1 sec (FEV₁), forced expiratory flow during 25–75% of the expiration (FEF 25% to 75%), total lung capacity (TLC), functional residual capacity (FRC), residual volume (RV), and RV/TLC (Table 2).

Response to Treatment of OSAS

OSAS was surgically treated as per the participants' clinical indication. Twenty participants were evaluated postoperatively, all of whom had underwent adenotonsillectomies. Of the remaining participants, seven declined further research, one family refused surgical treatment and favored CPAP, one opted for weight loss, and two were unreachable. Participants who were not tested after treatment were older and taller compared to those who were tested, but did not differ in OSAS severity (Table 3). Baseline polysomnography and airway resistance measurements were repeated 6.4 ± 6.6 mo after treatment. Overall, the OAHI improved from 15.2 [2.1–58.7] to 0.5 [0–5.1] events/h (P < 0.001) (Figure 2). However, eight participants (40%) had residual mild obstructive sleep apnea after treatment (OAHI range: 1.6–5.1 events/h).

Table 3.

Comparison between participants with obstructive sleep apnea syndrome who underwent postoperative assessments versus those who did not.

Figure 2.

Change in obstructive apnea-hypopnea index (OAHI) following surgery. Overall, the OAHI decreased significantly in children with obstructive sleep apnea following adenotonsillectomy.

Airway Resistance

Absolute airway resistance decreased after OSAS treatment (Table 4, Figure 3, P = 0.009). Absolute conductance increased after OSAS treatment (Table 4, P = 0.022) Furthermore, specific resistance decreased and specific conductance increased after OSAS treatment (Table 4). The postoperative OAHI did not correlate with airway resistance (r = 0.11, P = 0.65). The linear mixed-effects model used to control for height to further ascertain whether the changes observed in airway resistance were attributable to this or to the postoperative status of participants yielded an airway resistance of P = 0.017 between OSAS and controls. Specifically, the model suggested that adenotonsillectomy lowered airway resistance by 20.7% and that each 1 cm of linear growth furthered lowered airway resistance by 1%.

Table 4.

Airway resistance and pulmonary function data in children with obstructive sleep apnea syndrome before and after treatment (n = 20).

Figure 3.

Airway resistance before and after adenotonsillectomy. The box represents the interquartile range, the central line represents the median, the whiskers represent the 5^th and 95^th percentiles and the dots represent the outliers. Airway resistance decreased significantly in participants with obstructive sleep apnea after adenotonsillectomy.

Pulmonary Function Tests

Spirometry results did not change after OSAS treatment, except that FEV₁/FVC showed a statistically significant decrease but remained within normal limits (Table 3). Lung volumes did not change after adenotonsillectomy (Table 4).

DISCUSSION

This study has shown that, compared with control participants, children with OSAS have higher airway resistance measured by an orofacial mask. This study has also demonstrated that airway resistance improves significantly in participants with OSAS after treatment. In addition, this study confirmed previous findings that OSAS improves significantly after adenotonsillectomy in obese children, based on the OAHI, but does not completely resolve.^16–18

The upper airway includes the nose, paranasal sinuses, pharynx, and the portion of the larynx above the vocal cords, and plays a critical role in OSAS. For instance, several studies have shown that children with OSAS have a more collapsible upper airway and impaired upper airway sensation compared to controls during wakefulness and sleep.^19–22 Upper airway resistance may affect airway patency but there is no consensus regarding the methodology of measuring upper airway resistance. Airway resistance is defined as the resistance of the respiratory tract to airflow during inspiration and expiration and is typically measured by body plethysmography, using a mouthpiece and nose clips to exclude the nose.²³ The major determinants of conventionally measured airway resistance are the diameter of the airway and the characteristic of the airflow, i.e., laminar versus turbulent. Airway resistance increases linearly from the first up to the fifth airway generation of the tracheobronchial tree, and inversely to the total airway cross-sectional area of the airway generation. Airway resistance starts to decrease from the fifth airway generation, until it becomes practically zero at the level of the lower airways.²⁴ Nasal resistance, however, measures mainly the resistance along the nasal cavity and can be measured by anterior or posterior rhinomanometry. Anterior nasal resistance comprises the resistance between the nares and the choanal junction. Posterior rhinomanometry comprises the resistance from the nares to the pressure sensor location, which is typically placed in the oropharynx below the choanal junction.²⁵ Considering that adenotonsillar hyper-trophy is important in the pathogenesis of childhood OSAS,²⁶ neither nasal resistance nor orally measured airway resistance is adequate to estimate airflow resistance in this population because one excludes the effect of the adenoid and the other that of the tonsils. In the current study, we have shown that children with OSAS have elevated airway resistance measured by plethysmography using an orofacial mask. The use of the mask was designed to include both nasal and oral airway resistance in order to mimic normal breathing. It is important to consider that typical measurements of nasal or airway resistance require that subjects modify their breathing route. Hence, a strength of this research is that airway resistance was measured in a state similar to natural breathing.

This study has shown that children with OSAS have higher airway resistance but similar specific airway resistances measured by an orofacial mask compared to control participants. Previous research in adults has shown elevated airway resistance in obese subjects, and attributed this primarily to changes in lung volume.²⁷ The effects of obesity on airway resistance in children with OSAS are complex, as illustrated here. That Gaw and Raw were different, but sGaw and sRaw were similar in the OSAS and control groups suggests that at least in part, the resistance and conductance differences seen were related to lung volume differences between the two groups. However, no significant lung volume differences were found. Therefore, we believe it is probable that the major differences in Gaw and Raw were due to the upper airway resistance, not lung volume changes. This speculation is further supported by the posttreatment changes: both airway resistance and specific airway resistance improved significantly in participants with OSAS after treatment without significant changes in lung volumes.

Spirometry and lung volumes were similar in children with OSAS and control participants. This is different from recent research by Van Eyck et al.²⁸ They studied a cohort of overweight and obese children with and without OSAS and reported decreased FEV₁ and lung volumes in the OSAS group. Further research in obese and nonobese children with OSAS is needed.

To the best of our knowledge, there are no data in the adult literature reporting airway resistance in OSAS compared to controls. However, studies in adults have estimated upper airway resistance during sleep by measuring esophageal pressure and inspiratory flows. Several calculations have been used, such as airway resistance at the esophageal pressure nadir, at peak flow, as an average resistance and using complex polynomial equations.^29–31 Similar invasive studies have not been performed in children but it has been reported that children with OSAS have elevated nasal resistance during wakefulness and that this correlates with the OAHI in obese children.^4,5,32 It is important to point out that previous studies in children have not evaluated total airway resistance. The research procedure used in the current study, not meant to isolate upper airway resistance but to assess global airway resistance, provides further insight into global airflow resistance in children with OSAS. Importantly, untreated OSAS has been associated with wheezing, increased frequency of respiratory diseases, and elevated airway inflammatory markers.^33–37 Hence, it is plausible that these conditions would result in elevated global airway resistance, consistent with these data. It is also possible that the elevated airway resistance reported here may be mechanical due to enlarged tonsils and adenoid, as it decreased substantially after adenotonsillectomy. Of note, the participants in this study were otherwise healthy and without asthma, and inflammatory markers were not measured.

Four participants with OSAS were receiving antihistaminic/ anti-inflammatory medications prescribed by their primary doctor, but none had evidence of allergic rhinitis at the time of testing.

It is possible that the postoperative improvement in airway resistance was due to changes in lower airway resistance. However, this is unlikely as participants did not have known pulmonary disease such as asthma. In addition, participants' postoperative spirometry and lung volume values were unchanged except for the FEV₁/FVC that decreased slightly, but remained within normal limits. We believe that AR did not correlate with the OAHI because the OAHI is not a perfect measure. Specifically, the number of apneas and hypopneas are counted without consideration of duration and/or associated oxyhemoglobin desaturation, e.g., a 10-sec hypopnea with 4% desaturation counts the same as a 30-sec apnea with 8% desaturation. Therefore, there may be a threshold relationship between OSAS and AR rather than a linear one.

Limitations

A limitation of this study is that control subjects were leaner than participants with OSAS. However, obesity was not significant in the stepwise regression model, and the effects of obesity were also evaluated by the measurements of specific resistance and conductance. Lung volumes were measured by an unconventional method in order to assess the full upper airway of children with OSAS. However, they were measured consistently between participants with OSAS and control participants, and participants with OSAS before and after treatment.

CONCLUSIONS

In conclusion, children with OSAS have elevated airway resistance that improves after treatment of OSAS. This is likely due to upper airway loading secondary to adenotonsillar hypertrophy and may contribute to the increased frequency of respiratory diseases in untreated childhood OSAS.

DISCLOSURE STATEMENT

This was not an industry supported study. Support was provided by AHA 10CRP376001, NIH UL1RR024134, and Research Electronic Data Capture (RedCap). Dr. Marcus has research support from Philips Respironics in the form of loaned equipment only for investigator-initiated studies, not relevant to the current manuscript. The other authors have indicated no financial conflicts of interest.