Diabetes Onset at 31–45 Years of Age is Associated with an Increased Risk of Diabetic Retinopathy in Type 2 Diabetes

This hospital-based, cross-sectional study investigated the effect of age of diabetes onset on the development of diabetic retinopathy (DR) among Chinese type 2 diabetes mellitus (DM) patients. A total of 5,214 patients with type 2 DM who were referred to the Department of Ophthalmology at the Shanghai First People’s Hospital from 2009 to 2013 was eligible for inclusion. Diabetic retinopathy status was classified using the grading system of the Early Treatment Diabetic Retinopathy Study (ETDRS). Logistic and hierarchical regression analyses were used to identify independent variables affecting the development of DR. Upon multiple logistic regression analysis, patient age at the time of diabetes onset was significantly associated with development of DR. Further, when the risk of retinopathy was stratified by patient age at the onset of diabetes, the risk was highest in patients in whom diabetes developed at an age of 31–45 years (odds ratio [OR] 1.815 [1.139–2.892]; p = 0.012). Furthermore, when patients were divided into four groups based on the duration of diabetes, DR development was maximal at a diabetes onset age of 31–45 years within each group. A diabetes onset age of 31–45 years is an independent risk factor for DR development in Chinese type 2 DM patients.

Risk factors affecting the development of DR. Upon univariate logistic regression analysis, age, age of onset, duration of diabetes, SBP, DBP, HbA1c, MA and eGFR were identified for the potential risk factors. Then we examined a correlation matrix to exclude correlated factors such as age and DBP. Therefore, age of onset, duration of diabetes, SBP, HbA1c, MA and eGFR should be permitted in the multiple logistic regression model. Upon mutiple logistic regression analysis, age of onset, duration of diabetes, SBP, HbA1c, and MA were independent risk factors for DR ( Table 2, Model 1).
To stratify the risk of DR, patients were divided in terms of age at diabetes onset, as follows: ≤ 30 years, 31-45 years, 46-60 years, and ≥ 61 years. A diabetes onset age of 31-45 years was associated with an increased risk of DR. The 31-45-year age group was at the highest risk of DR, thus 1.815-fold (OR 1.815 [1.139-2.892]; p = 0.012) that of patients in the lowest age group (Table 2, Model 2).

Effects of the age of onset of diabetes and diabetes duration on the prevalence of DR. Patients
were divided into four groups according to duration of diabetes: ≤ 5 years, 6-10 years, 11-15 years, and > 15 years. We calculated the prevalence of DR by age at diabetes onset (≤ 30 years, 31-45 years, 46-60 years, and ≥ 61 years) in the groups differing in terms of diabetes duration. When the duration was ≤ 5 years, 375 DR patients (16.3%) were evident among the total of 2,302 patients, the proportions of whom at each age of onset were 24

Discussion
In this cross-sectional study of Chinese patients with type 2 diabetes, multiple logistic regression showed that the age at diabetes onset was significantly associated with the development of retinopathy, independent of the duration of diabetes, SBP, HbA1c level, and MA; these results are consistent with those of other studies 4-9 . We Previous studies suggested that early onset type 2 diabetes was more aggressive than late-onset disease 2,11,12 . In 2008, Wong et al. 8 reported that an age at type 2 diabetes onset < 45 years was associated with an increased inherent susceptibility to DR; the cited authors matched the duration of diabetes and the extent of glycemic control. Recently, a prospective cross-sectional study of an Asian cohort found that patients with younger-onset type 2 diabetes (diagnosed before the age of 40 years) had higher mean levels of HbA1c, and a greater prevalence of retinopathy, than those with late-onset diabetes (diagnosed at age ≥ 40 years) 10 . No study has yet addressed whether an age group at diabetes onset of < 45 years was associated with a higher risk of retinopathy, independent of disease duration and the extent of hyperglycemia.
In the present study, we sought a definite relationship between age at diabetes onset and DR. We divided our patients into four groups by age at diabetes onset: ≤ 30, 31-45, 46-60, and ≥ 61 years. Next, based on the duration of diabetes, each group was divided into four subgroups: ≤ 5 years, 6-10 years, 11-15 years, and > 15 years. We found that a diabetes onset age of 31-45 years was associated with an increased risk of DR development, independent of the duration of diabetes. However, the underlying mechanism remains unclear. Several possible explanations may be advanced. Some studies have found that the level of vascular endothelial growth factor (VEGF) in diabetes patients varies with age, and VEGF expression after stimulation is higher in younger than older patients 13,14 . When hyperglycemia is in play, VEGF promotes pathological retinal angiogenesis and   fibrovascular proliferation during development of DR 15,16 . Therefore, we speculate that a gene such as that encoding VEGF may be more active in patients with diabetes onset at 31-45 years of age, predisposing such patients to development of DR. In addition, "metabolic memory" may contribute to an increased risk of DR. Patients developing diabetes at 31-45 years may prioritize personal and career development rather than their health, and are usually diagnosed after a long-term history of hyperglycemia. Many studies have shown that prolonged hyperglycemia causes injuries to the retinal vasculature that are not reversed even upon subsequent sustained glycemic control; such impairments may play pivotal roles in "metabolic memory", rendering patients more susceptible to the complications of diabetes [17][18][19][20] . Last but not least, DM individuals diagnosed between 31-45 years of age are usually under high-level psychological pressure [21][22][23] . Such stress may explain why DR is more likely in patients in whom diabetes develops at an age of 31-45 years. Our findings have an important contribution to the monitoring and intervention in type 2 DM individuals in whom the diabetes onset age is 31-45 years of age with their working life at the peak of creation. Our research had several limitations. Firstly, the work was performed in a single hospital and all patients were Chinese. Care should be taken when attempting to extrapolate this data to other patient populations. Secondly, a cross-sectional study cannot identify cause-and-effect relationships. Future multicenter, longitudinal longer-term studies are required to verify our results. Thirdly, Color photographs were acquired using a macula-centered view of fundus photgraph and supplementary fundus photgraphs of lesions. Standard 7-field fundus photographs should be used in our future study.
In summary, a diabetes onset age of 31-45 years is an independent risk factor for development of DR in type 2 DM patients. It is important to conduct stringent monitoring and intervention in type 2 DM individuals in whom the diabetes onset age is 31-45 years, to delay the development and progression of DR. Data Collection. Medical record review was undertaken by a single researcher; demographic details, and physical and biochemical data, were recorded on a form. Demographic details included age, sex, age at diagnosis, duration of diabetes, and the general and ophthalmological medical histories. Physical examination included: systolic blood pressure (SBP), diastolic blood pressure (DBP), waist circumference (WC), and body-mass index (BMI). Laboratory data included the glycosylated hemoglobin (HbA1c) level; MA status; and the levels of fasting plasma glucose (FPG), triglycerides (TG), total cholesterol (TC), C-reactive protein, high-density lipoprotein (HDL), and low-density lipoprotein (LDL); and the estimated glomerular filtration rate (eGFR). The eGFR was calculated using the equation of the Modification of Diet in Renal Disease study 24 .

Methods
Each patient underwent a comprehensive ophthalmologic examination that included a review of ophthalmologic history, measurement of visual acuity and intraocular pressure (IOP), slit lamp biomicroscopy, and fundoscopic examination through dilated pupils via fundus photography and reading center to grade the retinopathy. Color photographs were acquired with a Zeiss Visucam 200 digital fundus camera (Carl Zeiss Meditec AG, Jena, Germany) using a macula-centered field of view. However, supplementary fundus photgraphs of lesions were taken for those who showed any evidence of DR. Diabetic retinopathy status was graded using the system of the Early Treatment Diabetic Retinopathy Study (ETDRS): 1) no DR; 2) nonproliferative disease (mild, moderate, severe); and, 3) proliferative. Whenever the two eyes were graded differently, the more advanced was chosen.