Objective:We aimed to explore the risk factor for diabetic retinopathy(DR)in patients with type 2 diabetes(T2DM)and obstructive sleep apnea syndrome(OSAS).Methods:A cross-sectional study of 455 T2DM patients who underwent overnight polysomnography(PSG)tests and diabetic fundus examination in the First Affiliated Hospital of Xiamen University from March 2013 to December 2017.330 patients who had complete data on clinical,PSG,and diabetic fundus examination were left for the present analysis,of which 217 were diagnosed with OSAS and 113 were without OSAS.Further analysis was performed to determine the differences of anthropometric information,biochemical characteristics,respiratory sleep related indicators and other chronic complications of diabetes between the DR group and non-diabetic retinopathy(NDR)group either with or without OSAS.Multivariable logistic regression analyses were performed to determine the potential risk factors for DR.Results:Among 330 diabetic patients,the prevalence rate of DR is 32.1%.Compared with that in NDR group(n=224),the T2DM subjects in DR group(n=106)were older(p=0.000),and had a longer T2DM duration(p<0.001)and lower educational level(p=0.008).and higher systolic blood pressure(SBP)(p=0.001)Besides,the prevalence rate of diabetes kidney diseases(DKD),diabetic peripheral neuropathy(DPN)and macrovascular diseases were significantly higher in DR group.However,there was no difference in glycated hemoglobin and respiratory disorders related indicators between two groups.Further logistic regression analysis showed that age(OR=0.969,95%CI(0.941-0.999),p=0.042),T2DM duration(OR=1.137,95%CI(1.074-1.205),p<0.001),systolic blood pressure(SBP)(OR=1.028,95%CI(1.008-1.049),p=0.007),diastolic blood pressure(DBP)(OR=0.967,95%CI(0.935-0.999),p=0.050),DKD(OR=2.906,95%CI(1.563-5.404),p=0.001),DPN(OR=2.876,95%CI(1.324-6.427),p=0.0008)were significantly associated with DR.However,there was no significant association between apnea hypopnea index(AHI)and DR(p=0.395).Furthermore,the logistic regression analysis conducted in the diabetic patients without OS AS showed that the T2DM duration,age,middle school were significantly associated with DR,logistic regression analysis showed that age(OR=0.938,95%CI(0.882-0.999),p=0.046).T2DM duration(OR=1.198,95%CI(1.034-1.387),p=0.016),middle school vs.junior school(OR=0.186,95%CI(0.035-0.977),p=0.047)were significantly associated with DR.In addition,in the case of T2DM with OSAS,the subjects in DR group(n=71)were older(p=0.000),and had a longer T2DM duration(p<0.001)and lower educational level(p=0.011)and lower diastolic blood pressure(DBP)(p=0.032)compared with those in NDR group(n=146).Besides,the prevalence rate of DKD,DPN and macrovascular diseases were significantly higher in DR group.The logistic regression analysis showed that T2DM duration(OR=1.139,95%CI(1.062-1.222),p<0.001),education levels(college or above vs.junior school(OR=0.203,95%CI(0.055-0.753),p=0.017),DKD(OR=2.855,95%CI(1.278-6.379),p=0.011)and DNP(OR=3.841,95%CI(1.288-11.459),p=0.016)were the potential risk factors of DR,and AHI was still not associated with DR.Conclusion:In our study,it showed that AHI was not significantly associated with DR.But there were more risk factors in diabetic patients with OSAS than without.It is necessary to focus on the screen of OSAS and other chronic complications,and strengthen diabetic education and blood pressure management. |