Background: Obstructive sleep apnea syndrome (OSAS) may induce metabolic abnormalities through intermittent hypoxemia and simpathetic activation. It is difficult to demonstrate an independent role of OSAS in the occurrence of metabolic abnormalities, as obesity represents an important risk factor for both OSAS and metabolic abnormalities.
Aim: to assess the relations between insulin resistance (IR), insulin sensitivity (IS), OSAS severity and nocturnal oxyhaemoglobin levels in obese, nondiabetic patients with daytime sleepiness.
Material and methods: We evaluated 99 consecutive, obese, nondiabetic patients (fasting glycemia < 126 mg/dL, no hypoglycemic or hypolipemiant medication) diagnosed with OSAS (AHI > 5/hour and daytime sleepiness) by an ambulatory six channel cardio-respiratory polygraphy. Hight, weight serum triglycerides (TG), high density lipoprotein-cholesterol (HDL-C) levels were evaluated. Correlations between Apneea Hypopnea Index (AHI), Oxygen Desaturation Index (ODI), average and lowest oxyhaemoglobin saturation (SaO), body mass index (BMI) and insulin resistance or sensitivity were assesed. IR was defined as a TG/ HDL-Cratio > 3, and insulin sensitivity (IS) as a TG/HDL-C ratio < 2.
Results: 64 patients (out of 99) had lR and 18 IS. In the IR group (44 men and 20 women), the mean age was 52 +/- 10.6 years, mean BMI: 38.54 +/- 6.67 Kg/m2 (30-60), TG/HDL-C:5, 27 +/- 2.03 (3.02-11.1), mean AHI: 49.65 +/- 25.55/hour (7-110), mean ODI: 4769 +/- 24.95/hour (4-98), mean average SaO2 89.42 +/- 4.6 and mean lowest SaO2 68.4% +/- 13.8% (32-88%). 48 patients had severe, 7 moderate and 9 mild OSAS. In the IS group (10 men and 8 women), the mean age was 58.4 +/- 8.2years, mean BMI: 35.4 +/- 4.29 Kg/m2 (30-46), TG/ HDL-C: 1.64 +/- 0.29 (1.13-1.95), mean AHI: 45.8 +/- 30.3/hour (9-131), mean ODI: 39.9 +/- 32.2/hour (2-133), mean average SaO2 90.8 +/- 8.2 (81-95) and mean lowest SaO2: 74% +/- 10.8% (52-87%). 12 patients had severe, 3 moderate and 3 mild OSAS. Insulin sensitivity positively correlated with mean average SaO2 (r: 0.49; p: 0.037) and negatively with ODI (r: - 0,56; p: 0.014). Insulin resistance negatively correlated with mean lowest SaO2 (r: -0,25; p: 0.045). Mean lowest SaO2 values were significant lower in patients with IR than in those with IS (p: 0.042). No statistically significant difference was found for BMI, AHI or ODI between IR and IS patients.
Conclusions: nocturnal oxyhaemoglobin levels rather than OSAS severity (expressed as AHI or ODI) may be involved in the occurrence of metabolic abnormalities in obese nondiabetic patients. Preserving insulin sensitivity is more likely when oxyhaemoglobin levels are higher and ODI is lower. Mean lowest nocturnal SaO2 levels seems to be independently involved in the development of insulin resistance as no statistically significant differences were found for BMI between the two groups.