Leukotrienes Pathway in Chinese Children With Sleep-disordered Breathing

Background:

SDB in children markedly differs from that seen in adults, in particular with respect to clinical manifestations, PSG (Polysomnography) findings, and treatment approaches. Pediatric SDB is due to a combination of increased upper airway resistance and repetitive pharyngeal collapsibility leading to intermittent hypoxemia and arousals during sleep. It is a common and highly prevalent disorder in pediatric age, affecting 4 to 11% of all children. If left untreated, it can result in serious morbid consequences including cardiovascular morbidity and neurocognitive dysfunction.

The etiology and pathophysiological mechanisms leading to SDB in childhood have not been clearly elucidated, but may include a complex interplay between anatomic (mainly adenotonsillar hypertrophy, ATH), neuromuscular factors and an underlying genetic predisposition toward the disease. However, more evidences have identified that both local airway and systemic inflammation also contribute to its pathogenesis.

Among major inflammatory mediators, leukotrienes (LTs) are a class of closely structurally related lipid compounds that derive from the 5-lipoxygenase pathway of arachidonic acid metabolism. The LTs family includes LTA4, LTB4 and LTC4/D4/E4 (cysLTs), all of which are potent leukocytes chemoattractants and activators. LTA4, an unstable intermediate, can be further metabolized to LTB4 or cysLTs. LTC4 and LTD4 both have very short half-lives, whereas LTE4 appears to be most stable. The effects of mediators occur after their interaction with receptors.

Recent investigations in western countries have revealed LTs may be involved in the pathogenesis of SDB, and accelerate the progress of the disease by exacerbating local inflammatory response, then promoting the proliferation of the upper airway lymphoid tissues. And in the later pediatric investigation, it was determined that LTs production emerged disease severity-dependent increases in exhaled breath condensate of SDB patients. Nevertheless, few studies have investigated the role of LTs in systemic inflammation of SDB, and whether LTs productions is a independent risk factor of SDB, or is a obesity-dependent or ATH -dependent risk factor of SDB remains controversial. Moreover, no research work in China has been carried out to investigate the LTs pathway in a large population of Chinese snoring children with SDB.

Participants:

Consecutive children referred to the sleep center for PSG due to symptoms of SDB from December 2009 to June 2010 were enrolled in the study. Normal age-, sex-, and weight-matched children without a history of snoring participated as control subjects, who were recruited from a community-based physical check-up activity.

Grouping:

Five groups of participants were formed:

1. Snoring group:

1. Subjects with snoring and AHI ≥ 20 episodes/h (severe SDB).

2. Snorers with AHI < 20 episodes/h and ≥ 5 episodes/h (moderate SDB)

3. Snorers with AHI < 5 episodes/h and ≥ 1 episodes/h (mild SDB)

2. Non-snoring control group:

- When testing serum and urinal samples, healthy children without snoring will be chosen as controls.

- When testing lymphoid tissue samples, patients with recurrent infectious tonsillitis (at least five tonsillar infections in less than 6 months) but without snoring will be selected as controls before surgery and recruited to the study, because adenotonsillar tissue can't be obtained from normal children for obvious ethical reasons.

The five study groups having at least 45 participants respectively will be compared regarding subjects' characteristics, PSG indices, metabolic and inflammatory variables.

Statistical Methods:

1. Data are presented as mean ± SD or median (interquartile range) depending on the distribution unless stated otherwise. PSG indexes, hsCRP values and LT concentrations will be log-transformed (natural logarithm) and SpO2 will be logit-transformed to correct for skewed distribution.

2. One-way analysis of variance followed by post hoc tests for pair comparisons (Bonferroni test) will be used for continuous variables and X2 test (Yates correction) for categorical characteristics. If data can't be transformed to approach normal distribution, Kruskal-Wallis test followed by nonparametric Bonferroni multiple comparison test will be applied.

3. Correlations will be analyzed without adjustment by using Pearson correlation or spearman rank test depending on data distribution. Partial correlation will be applied to assess the linear association of LTs concentrations with other systemic inflammatory marks and PSG indices after adjustment for covariates including age, glucose, lipids, and BMI z-score.

4. Stepwise multiple regressions will be performed to identify independent predictors of LTs production.

5. All statistical analyses will be conducted using statistical software (version 16.0; SPSS, Chicago, IL, USA). Using Dunnett's test, a two-tailed p value < 0.01 was considered statistically significant.