Rationale: Obstructive sleep apnea syndrome (OSAS) and nonalcoholic fatty liver disease (NAFLD) are frequently encountered in obese children. Whether OSAS and intermittent hypoxia are associated with liver injury in pediatric NAFLD is unknown.
Objectives: To assess the relationship of OSAS with liver injury in pediatric NAFLD.
Methods: Sixty-five consecutive children with biopsy-proven NAFLD (age, mean ± SD, 11.7 ± 2.1 yr; 58% boys; body mass index z score, 1.93 ± 0.61) underwent a clinical-biochemical assessment and a standard polysomnography. Insulin sensitivity, circulating proinflammatory cytokines, markers of hepatocyte apoptosis (cytokeratin-18 fragments), and hepatic fibrogenesis (hyaluronic acid) were measured. Liver inflammatory infiltrate was characterized by immunohistochemistry for CD45, CD3, and CD163, surface markers of leukocytes, T cells, and activated macrophage/Kupffer cells, respectively. OSAS was defined by an apnea/hypopnea index (AHI) greater than or equal to 1 event/h, and severe OSAS was defined by an AHI greater than or equal to 5 events/h.
Measurements and main results: Fifty-five percent of children with NAFLD had nonalcoholic steatohepatitis (NASH), and 34% had significant (stage F ≥ 2) fibrosis. OSAS affected 60% of children with NAFLD; the presence and severity of OSAS were associated with the presence of NASH (odds ratio, 4.89; 95% confidence interval, 3.08-5.98; P = 0.0001), significant fibrosis (odds ratio, 5.91; 95% confidence interval, 3.23-7.42; P = 0.0001), and NAFLD activity score (β, 0.347; P = 0.029), independently of body mass index, abdominal adiposity, metabolic syndrome, and insulin resistance. This relationship held also in nonobese children with NAFLD. The duration of hemoglobin desaturation (Sa(O2) < 90%) correlated with increased intrahepatic leukocytes and activated macrophages/Kupffer cells and with circulating markers of hepatocyte apoptosis and fibrogenesis.
Conclusions: In pediatric NAFLD, OSAS is associated with biochemical, immunohistochemical, and histological features of NASH and fibrosis. The impact of hypoxemia correction on liver disease severity warrants evaluation in future trials.