Aims/hypothesis: The aim of this study was to determine the incidence and progression rates of diabetic retinopathy and their associations in Japanese individuals with type 2 diabetes.

Methods: This is a part of the Japan Diabetic Complications Study (JDCS), a multi-centred randomised trial of type 2 diabetes patients aged 40-70 years with an 8 year follow-up. There were 1,221 patients without diabetic retinopathy at baseline; incidence of diabetic retinopathy was defined as the development of any diabetic retinopathy. There were 410 patients with mild non-proliferative diabetic retinopathy at baseline; progression of diabetic retinopathy was defined as the development of severe non-proliferative diabetic retinopathy or proliferative diabetic retinopathy. We used multivariate proportional Cox hazard models, and generalised additive models were also applied to identify potential threshold effect.

Results: The incidence and progression rate of diabetic retinopathy was 38.3/1,000 person-years and 21.1/1,000 person-years, respectively. Higher HbA(1c) (adjusted HR [aHR] per 1% [10.9 mmol/mol] 1.36 [95% CI 1.28-1.45]), longer duration of diabetes (aHR per 5 year period 1.26 [95% CI 1.17-1.35]), higher systolic blood pressure (aHR per +10 mmHg 1.01 [95% CI 1.00-1.02]) and higher body mass index (aHR per 1 kg/m(2) 1.05 [95% CI 1.00-1.09]) were associated with incident diabetic retinopathy. The association between HbA(1c) and incident diabetic retinopathy was linear; the association with duration of diabetes increased rapidly between 5 and 10 years. Higher HbA(1c) was also associated with progression of diabetic retinopathy (aHR per 1% [10.9 mmol/mol] 1.66 [95% CI 1.41-1.96]).

Conclusions: Observed incidence and progression rates of diabetic retinopathy seemed lower than that in western populations. HbA(1c) was the only factor associated with both incidence and progression of diabetic retinopathy. The strength of the association between duration of diabetes and incidence of diabetic retinopathy increased rapidly during a period of 5 to 10 years duration of diabetes.

Trial registration: C000000222 ( www.umin.ac.jp )

Funding: This study is supported by the Ministry of Health, Labour and Welfare, Japan.