Article Text


HLA typing is not predictive of proliferative diabetic retinopathy in patients with younger onset type 2 diabetes mellitus
  1. T Mimura1,2,
  2. S Amano1,
  3. S Kato1,
  4. M Araie1,
  5. H Funatsu2,
  6. S Kitano2,
  7. E Shimizu2,
  8. H Noma3,
  9. O Yoshino4,
  10. S Hori5
  1. 1Department of Ophthalmology, University of Tokyo Graduate School of Medicine, Tokyo, Japan
  2. 2Department of Ophthalmology, Diabetes Center Tokyo Women’s Medical University, Tokyo, Japan
  3. 3Department of Ophthalmology, Hiroshima University School of Medicine, Hiroshima, Japan
  4. 4Department of Obstetrics and Gynecology, University of Tokyo school of Medicine, Tokyo, Japan
  5. 5Department of Ophthalmology, Tokyo Women’s Medical University, Tokyo, Japan
  1. Correspondence to: Tatsuya Mimura MD, Department of Ophthalmology, University of Tokyo Graduate School of Medicine, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8655 Japan;

Statistics from

Chronic hyperglycaemia and the duration of diabetes are the most important factors in retinopathy. However, retinopathy progresses in some patients despite good glycaemia control. Also, poor glycaemia control does not always lead to retinopathy in younger onset patients, while still others develop severe retinopathy that is resistant to retinal photocoagulation.

These facts suggest that the risk factors for diabetes and retinopathy are not necessarily the same, and that the development of severe retinopathy may be influenced by genetic factors.1

Human leucocyte antigen (HLA) status has a significant role in immune responses and immunological tolerance and is a factor in the onset of type 2 diabetes.2–4 DR4, DR8, DR9, and several antigens of the DQ region are related to retinopathy in patients with type 1 diabetes.5,6 In addition, it was reported that HLA-DR was expressed in proliferative retinopathy.7,8 Little is known, however, about the relation between retinopathy with type 2 diabetes and the HLA antigen. Furthermore, most previous studies have not taken into consideration the background of glycaemic control or the duration of the diabetes. A group of younger onset type 2 diabetes patients with PDR, and a group who had no signs of retinopathy despite a long duration of diabetes were compared. Younger patients were studied to reduce the influence of adult diseases such as hyperlipidaemia and hypertension. In addition, clinical background factors were considered when studying the frequency of HLA types.


Following the informed consent of each of the subjects, blood samples were collected. The study was approved by the human studies review board of Tokyo Women’s Medical University and was performed in accordance with the Helsinki Declaration of 1975 and its 1983 revision. The diagnosis of type 2 diabetes was made based on 1985 World Health Organization criteria.9 We excluded subjects who were GAD antibody positive. The patients had been diagnosed as having type 2 diabetes aged under 30 years (range 12–21 years) and type 2 diabetes duration for more than 10 years. Additionally, the patients whose average HbA1C level over 10 years was from 6% to 10%, were selected in this study. All were receiving treatment at Tokyo Women’s Medical University Diabetes Center.

Further, two groups were selected from above patients, as follows. The PDR group consisted of 44 patients, who had undergone vitreous surgery under the age of 40 (mean 28.8 (SD 4.4) years) where surgery had been carried out at the department of ophthalmology, diabetes centre (as above) during the period 1993–9. The non-DR group consisted of 45 patients who had no signs of retinopathy despite having diabetes for more than 10 years. Consequently, the sex, the diabetes duration, and the blood glucose control have been matched between the two groups on the basis of the selection criteria above (table 1). The control group selected for comparison, consisted of 50 healthy patients. The HbA1C level was determined with resin microcolumn technique (HPLC, Kyoto Chemical) (normal range 4.3–5.8%).

Table 1

Clinical profile of the control group, non-DR group, and PDR group

HLA-A, B, C, DR, and DQ typing of blood samples was conducted on all three groups using standard microcytotoxicity methods.

For comparisons between the groups, we applied the χ2 test of independence or Fisher’s exact probability test. The unpaired t test was used for comparing mean values. The level of significance was set at p<0.05. All analyses were performed using the Stat View statistical software package (Abacus Concepts, Berkeley, CA, USA).


The frequencies of HLA-A, B, and Cw antigens in the control group, the non-DR group, and the PDR group are shown in table 2, and those of DR, and DQ antigens in table 3, respectively.

Table 2

HLA-A, B, and C antigen frequencies (%) in each group

Table 3

HLA-DQ and -DR antigen frequencies (%) in each group

There was no significant difference among the three groups in HLA-A, B, and DQ antigens. The non-DR group showed higher frequencies of HLA Cw4 (χ2 = 4.027, p = 0.045) and DR4 (χ2 = 4.398, p = 0.036) than the control group (tables 2 and 3). While there was no significant difference between non-DR group and PDR group in any of the HLA antigens. The PDR group showed higher frequencies of HLA DR4 than the control group (χ2 = 5.937, p = 0.014).


Type 1 diabetes is aetiologically different from type 2 diabetes. Type 1 diabetes is caused by a failure in the autoimmune system, is clearly associated with specific HLA antigens.10 Type 2 diabetes is not autoimmune and has less association or linkage with genes in the HLA region than type 1.

The type 2 diabetic patients in this study showed a typical HLA pattern. Other research has reported that DR3, DR4, and Cw4 increased in patients with type 2 diabetes mellitus.2–4 These findings are consistent with our present results. Additionally an increase in HLA-DR4, which is in linkage disequilibrium with the DQB1*0302 allele has previously been reported in patients with type 2 diabetes.2,3 This increase was mainly reported to be restricted to patients with relative insulin deficiency or antibodies to islet cells or to glutamic acid decarboxylase.4

DR4 was detected in 59.1% of the PDR group, but this was not significantly different from the frequency in the non-DR group (51.1%). DR4 may, therefore, be related to the onset of type 2 diabetes, but not to the development of retinopathy. The HLA-DR4 levels reflected the antibody levels in the pancreatic Langerhans island but not the parameter of diabetic change in the retina.

In summary, our research suggests that HLA antigen investigations may be useful for predicting the prognosis of younger onset type 2 diabetes, but not for retinopathy in these patients. Finally, we must precisely define the alleles or combination of alleles which cause increased susceptibility to PDR.


The authors gratefully acknowledge the assistance of Ms Jayne Simons for critically reviewing the manuscript.


View Abstract


  • The authors have no commercial or proprietary interest in the product or company described in this letter.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.