Relationship between MR intra-abdominal fat parameters and cardiovascular risk parameters in obstructive sleep apnoea

Abstract

Objectives To evaluate the relationship between abdominal fat parameters quantified on MR with cardiovascular risk factors in obstructive sleep apnoea. Materials and methods Seventy-five consecutive men attending sleep centre at the Queen Mary Hospital without history of diabetes mellitus, hypertension or hyperlipidemia were recruited. All patients underwent abdominal MR examination using a 1.5T magnet within 10 days of polysomnogram and metabolic studies. Axial spin echo (SE) scans (TR 350 ms, TE 12 ms, slice thickness 10 mm) were obtained from inferior endplate of L4 to inferior endplate of S1. Digital data was transferred to a computer and axial image through the middle of the umbilicus was selected. Threshold values were used (analyze 5.0) to segment out intraabdominal (IAF) and subcutaneous fat (SCF) area. Total abdominal fat (TAF) area was obtained by summation of the intraabdominal and subcutaneous fat areas. Cardiovascular risk parameters were obtained in each patient: systolic and diastolic blood pressure (BP); fasting serum cholesterol, triglicerides, glucose, high density lipid (HDL), low density lipid (LDL); insulin sensitivity (HOMA); and apnoea-hypopnoea index, arousal index, duration of O2 saturation <90% and minimum O2 saturation. Fat parameters were correlated with cardiovascular risk parameters, age adjusted. Linear regression analysis was performed for all parameters to determine which fat parameter was best predictive for each of the cardiovascular risk parameters. Results SCF area was correlated with HOMA (r = 0.44, p < 0.001), cholesterol/HDL (r = 0.28, p = 0.017), apnoea-hypopnoea index (r = 0.34, p = 0.003), duration of O2 saturation <90% (r = 0.37, p = 0.001), and minimum O2 saturation (r = -0.24, p = 0.038). IAF area was correlated with diastolic BP (r = 0.28, p = 0.017), HOMA (r = 0.52, p < 0.001), cholesterol/HDL (r = 0.27, p = 0.021), apnoea-hypopnoea index (r = 0.41, p < 0.001), duration of O2 saturation <90% (r = 0.48, p < 0.001), minimum O2 saturation (r = -0.46, p < 0.001). TAF area was correlated with diastolic BP (r = 0.23, p < 0.05), HOMA (r = 0.54, p < 0.001), cholesterol/HDL (r = 0.32, p = 0.006), apnoea-hypopnoea index (r = 0.42, p < 0.001), duration of O2 saturation <90% (r = 0.47, p < 0.001), minimum O2 saturation (r = -0.37, p = 0.001). After linear regression, IAF area was best predictive of diastolic blood pressure, cholesterol, triglycerides and sleep parameters. TAF area was best predictive of HOMA, LDL and cholesterol/HDL. Conclusion IAF and TAF areas are predictive of cardiovascular risk parameters including sleep parameters. SF area is less predictive.