BACKGROUNDAlthough obstructive sleep apnea (OSA) has been recognized as a major risk factor for cardiovascular complications and its clinical features are well characterized, it is difficult to replicate the OSA hypoxic model in humans. We aimed to establish an experimental rabbit model for chronic OSA and to explore its application to measure blood pressure (BP), myocardial systolic function, and oxidative stress.

METHODSThe rabbit model for OSA was established by repeatedly closing the airway and then reopening it. A tube specially designed with a bag that could be alternately inflated and deflated according to a predetermined time schedule, resulting in recurrent airway occlusions and chronic intermittent hypoxia (CIH) imitating OSA patterns in humans, was used. Twenty-four rabbits were randomly divided into obstruction, sham, and control groups, and their upper airways were alternately closed for 15 s and then reopened for 105 s in a 120-s-long cycle, for 8 h each day over 12 consecutive weeks. Before and after the experiment, the BP of each rabbit was monitored. Levels of interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) in the serum, superoxide dismutase (SOD) activity, malondialdehyde (MDA) and reactive oxygen species (ROS) contents, as well as Na+-K+-ATPase/Ca2+-ATPase activities in cardiac muscle were examined. In addition, cardiac functional parameters were measured using echocardiography.

RESULTSAfter 3 months, all rabbits in the obstruction group manifested sleepiness performance similar to that observed in OSA patients. Traces of airflow and SpO2showed that this model mimicked the respiratory events involved in OSA, including increased respiratory effort and decreased oxygen saturation. Gradually, the BP rose each month. CIH led to obvious oxidative stress and injured myocardial systolic performance. The serum levels of IL-6 and TNF-α increased significantly (64.75 ± 9.05 pg/ml vs. 147.00 ± 19.24 pg/ml and 59.38 ± 8.21 pg/ml vs. 264.75 ± 25.54 pg/ml, respectively, both P < 0.001). Compared with the sham and the control groups, myocardial activities of Na+-K+-ATPase/Ca2+-ATPase and SOD in the obstruction group decreased markedly, while ROS and MDA content increased.

CONCLUSIONSThese results show that the rabbit model for OSA simulates the pathophysiological characteristics of OSA in humans, which implies that this animal model is feasible and useful to study the mechanisms involved in the cardiovascular consequences of OSA.