Introduction
Diabetes mellitus is a widely prevalent disease and is one of the leading causes of blindness worldwide.1, 2 The disease affects the generalized micro vasculature and macro vasculature of various structures. It takes years for microvascular complications in diabetes mellitus (DM) such as diabetic retinopathy (DR) and diabetic nephropathy (DN) to develop. Since retinal and renal vessels are exposed to the diabetic milieu, it is assumed that progression of DR and diabetic nephropathy occurs at the same time. Early detection of diabetic retinopathy helps in preventing vision threatening complications. Our study aims to find the correlation between DR and diabetic nephropathy and the influence of other risk factors such as duration of diabetes and hypertension on diabetic retinopathy and diabetic nephropathy changes.3
Methodology
This is a non-randomized, prospective study was conducted at a tertiary care hospital in South India between November 2014 and May 2016. After obtaining approval from the Institutional Ethics Committee, 100 patients with diabetic nephropathy referred from the Nephrourology Department of our hospital for fundoscopy were included in the study. Diabetic nephropathy was defined by presence of either microalbuminuria or macro-albuminuria, in the absence of uncontrolled hypertension, congestive cardiac failure (CCF) and active urinary tract infection (UTI). Patients with hazy media enough to interfere with a detailed fundus examination and management, patients with CCF, UTI or uncontrolled hypertension, retinal proliferative disorders were excluded from the study. After obtaining basic demographic data and a detailed history, all subjects underwent complete ocular examination and systemic evaluation. Visual acuity, fundoscopy by direct and indirect ophthalmoscope, fundus photography and optical coherence tomography were done as and when required. Blood investigations like fasting blood sugar, post prandial blood sugar, glycosylated haemoglobin, haemoglobin levels, serum creatinine and blood urea nitrogen, lipid profile, urine routine and 24-Hour urine albumin were recorded. DR was classified according to ETDRS classification and different grades were compared with grades of nephropathy and association was analysed statistically. Data was analyzed using SPSS version 22 (IBM SPSS Statistics, Somers NY, USA) software and appropriate statistical tools.
Results
Majority of subjects were in the age group 41 to 50 years (33%), followed by 51 to 60 years (31%). Mean age of subjects in the study was 52.64 ± 10.57 years. (Table 1)
Table 1
Mean age of subjects in the study
|
|
Count |
% |
|
|
Age |
<40 years |
13 |
13.0% |
|
41 to 50 years |
33 |
33.0% |
|
|
51 to 60 years |
31 |
31.0% |
|
|
>60 years |
23 |
23.0% |
|
|
Total |
100 |
100.0% |
|
Majority of subjects were males (68%) and 32% were females. (Table 2)
Table 2
Gender distribution of subjects
|
|
Count |
% |
|
|
Gender |
Female |
32 |
32.0% |
|
Male |
68 |
68.0% |
|
|
Total |
100 |
100.0% |
|
Majority of subjects were diabetics and hypertensive from < 5 years duration (44% & 76.8% respectively). (Table 3)
Table 3
Duration of diabetes mellitus and hypertension
|
Duration of DM |
Duration of HTN |
||||
|
Count |
% |
Count |
% |
||
|
Duration of DM |
<5 years |
44 |
44.0% |
63 |
76.8% |
|
6 to 10 years |
31 |
31.0% |
11 |
13.4% |
|
|
11 to 15 years |
11 |
11.0% |
4 |
4.9% |
|
|
>15 years |
14 |
14.0% |
4 |
4.9% |
|
57% of subjects were on Oral hypoglycaemic agents (OHA) and 43% were on insulin in the study. (Table 4)
Table 4
Treatment taken by subjects in the study
|
|
Count |
% |
|
|
Treatment |
Insulin |
43 |
43.0% |
|
OHA |
57 |
57.0% |
|
|
Total |
100 |
100.0% |
|
On fundus examination 23% had proliferative diabetic retinopathy (PDR), 53% had Non proliferative diabetic retinopathy (NPDR) and 24% had no signs of DR.
Out of 53 subjects with NPDR, 19 had Mild, 28 had moderate and 6 subjects had severe NPDR. (Table 5)
Table 5
Fundus findings in the subjects
|
|
Count |
% |
|
|
Fundus |
PDR |
23 |
23.0% |
|
Mild NPDR |
19 |
19.0% |
|
|
Moderate NPDR |
28 |
28.0% |
|
|
Severe NPDR |
6 |
6.0% |
|
|
Normal |
24 |
24.0% |
|
|
Total |
100 |
100.0% |
|
There was significant association between fundal changes and albumin excretion in urine. Among subjects with massive albuminuria, 64.7% of them had PDR. Similarly, among subjects with moderate albuminuria majority had moderate NPDR and subjects with microalbuminuria majority had normal fundus. This shows a strong association between fundal changes and albumin excretion. (Table 6)
Table 6
Association between fundus findings and albumin excretion
No significant association was observed between haemoglobin levels and fundal changes. (Table 7)
Table 7
Association between fundus changes and haemoglobin levels
|
|
Hb |
||||
|
<12 gm/dl |
>12 gm/dl |
||||
|
Count |
% |
Count |
% |
||
|
Fundus |
PDR |
22 |
25.3% |
1 |
7.7% |
|
Mild NPDR |
16 |
18.4% |
3 |
23.1% |
|
|
Moderate NPDR |
25 |
28.7% |
3 |
23.1% |
|
|
Severe NPDR |
6 |
6.9% |
0 |
0.0% |
|
|
Normal |
18 |
20.7% |
6 |
46.2% |
|
|
Total |
87 |
100.0% |
13 |
100.0% |
|
Mean of total cholesterol (TC), High density lipoprotein (HDL), Low density lipoprotein (LDL) and Triglycerides (TG) was compared with the fundal changes. A significant difference was observed for total cholesterol and fundus changes, i.e., higher TC level was seen in PDR subjects than in NPDR. No difference was observed for TG, HDL and LDL levels. (Table 8)
Table 8
Association between fundus changes and lipid profile
Significant association was observed between total cholesterol and fundal changes. Majority of subjects with TC >200 had PDR (41.4%). (Table 9)
Table 9
Association between fundus changes and total cholesterol
|
|
TC |
||||
|
<200 |
>200 |
||||
|
Count |
% |
Count |
% |
||
|
Fundus |
PDR |
11 |
15.5% |
12 |
41.4% |
|
Mild NPDR |
14 |
19.7% |
5 |
17.2% |
|
|
Moderate NPDR |
20 |
28.2% |
8 |
27.6% |
|
|
Severe NPDR |
5 |
7.0% |
1 |
3.4% |
|
|
Normal |
21 |
29.6% |
3 |
10.3% |
|
|
Total |
71 |
100.0% |
29 |
100.0% |
|
No significant association was observed between HDL and Fundal changes. (Table 10)
Table 10
Association between fundus changes and HDL
|
|
HDL |
||||
|
<40 |
>40 |
||||
|
Count |
% |
Count |
% |
||
|
Fundus |
PDR |
12 |
32.4% |
11 |
17.5% |
|
Mild NPDR |
9 |
24.3% |
10 |
15.9% |
|
|
Moderate NPDR |
9 |
24.3% |
19 |
30.2% |
|
|
Severe NPDR |
1 |
2.7% |
5 |
7.9% |
|
|
Normal |
6 |
16.2% |
18 |
28.6% |
|
|
Total |
37 |
100.0% |
63 |
100.0% |
|
No significant association was observed between LDL and Fundus changes. (Table 11)
Table 11
Association between fundus changes and LDL.
|
|
LDL |
||||
|
<100 |
>100 |
||||
|
Count |
% |
Count |
% |
||
|
Fundus |
PDR |
14 |
18.9% |
9 |
34.6% |
|
Mild NPDR |
14 |
18.9% |
5 |
19.2% |
|
|
Moderate NPDR |
21 |
28.4% |
7 |
26.9% |
|
|
Severe NPDR |
5 |
6.8% |
1 |
3.8% |
|
|
Normal |
20 |
27.0% |
4 |
15.4% |
|
|
Total |
74 |
100.0% |
26 |
100.0% |
|
No significant association was observed between TG and Fundus changes. (Table 12)
Table 12
Association between fundus changes and triglycerides
|
|
TG |
||||
|
<150 |
>150 |
||||
|
Count |
% |
Count |
% |
||
|
Fundus |
PDR |
9 |
20.0% |
14 |
25.5% |
|
Mild NPDR |
9 |
20.0% |
10 |
18.2% |
|
|
Moderate NPDR |
13 |
28.9% |
15 |
27.3% |
|
|
Severe NPDR |
4 |
8.9% |
2 |
3.6% |
|
|
Normal |
10 |
22.2% |
14 |
25.5% |
|
|
Total |
45 |
100.0% |
55 |
100.0% |
|
No significant association was observed between Hb% and Albumin excretion. (Table 13)
Table 13
Association between Albumin excretion and Haemoglobin %.
|
|
Albumin Excretion |
||||||
|
Microalbuminuria |
Macroalbuminuria |
Massive Albuminuria |
|||||
|
Count |
% |
Count |
% |
Count |
% |
||
|
Hb |
<12 gm/dl |
33 |
84.6% |
38 |
86.4% |
16 |
94.1% |
|
>12 gm/dl |
6 |
15.4% |
6 |
13.6% |
1 |
5.9% |
|
There was no significant association between HbA1c and Albumin excretion. (Table 14)
Table 14
Association between albumin excretion and HbA1c.
|
|
Albumin Excretion |
||||||
|
Micro albuminuria |
Macro albuminuria |
Massive Albuminuria |
|||||
|
Count |
% |
Count |
% |
Count |
% |
||
|
HbA1c |
<7 |
1 |
2.6% |
3 |
6.8% |
1 |
5.9% |
|
>7 |
38 |
97.4% |
41 |
93.2% |
16 |
94.1% |
|
|
Total |
39 |
100.0% |
44 |
100.0% |
17 |
100.0% |
|
There was no significant association between HbA1c and Diabetic Retinopathy. (Table 15)
Discussion
The microvascular complications of diabetes encompass long-term complications such as damage to the small blood vessels. These classically include retinopathy, nephropathy, and neuropathy. This may have devastating consequences, including blindness and end-stage renal disease. Some authors have identified associations between the complications themselves, and that one complication can serve as a risk factor for another. Recently, studies have shown that the presence of DR itself may increase the risk for diabetic nephropathy.
Retinopathy as a predictor of other diabetic complications, a study done by El Asrar AM4 in the year 2001 concluded that retinopathy, especially the presence of PDR, is an independent predictor for nephropathy. The predictive value of retinopathy for nephropathy is stronger in patients with Insulin Dependent Diabetes Mellitus (IDDM) than in those with Non-Insulin Dependent Diabetes Mellitus (Non-IDDM). Therefore, it was suggested that ophthalmologists should refer patients with retinopathy for regular medical evaluations and vice-versa.
In another study conducted by Thivolet C et al5 in 1990 concluded that, routine analysis of urinary albumin excretion rate in diabetics allows early detection of diabetic nephropathy and emphasizes the need for tight metabolic and blood pressure control.
A study conducted by Savage S, et al6 in1996 concluded that increasing urine albumin excretion rate in Non-IDDM patients was associated with an increased prevalence of diabetic retinopathy, neuropathy, and cardiovascular disease. This suggests that urine albumin excretion rate may be more than an indicator of renal disease in these patients and, in fact, may reflect a state of generalized vascular damage occurring throughout the body.
In our study it was observed that there was significant association between fundus changes and albumin excretion in urine. Among subjects with massive albuminuria, 64.7% of them had PDR. Similarly, among subjects with moderate albuminuria majority had moderate NPDR and subjects with microalbuminuria majority had normal fundus. This shows a strong correlation between fundus changes and albumin excretion.
There was also a significant correlation between total cholesterol and fundal changes, i.e., higher TC was seen in PDR subjects than in NPDR. In our study, no difference was observed for TG, HDL and LDL levels. This is in contrast to a study by Alpana Mathur et al7 in which it was found that triglyceride levels were significantly raised in subjects with DR as compared to those without DR showing a positive correlation. No such association was found between LDL and TC levels with the prevalence of diabetic retinopathy.
