Retinal nerve fiber layer (RNFL) loss is commonly observed in patients with diabetes and has correlations with the stage of diabetic retinopathy (DR) (1). However, several reports have shown RNFL loss in patients with diabetes independent of DR, suggesting that the function of neuronal cells in the retina is compromised even before the appearance of overt microvascular changes (2–4). We hypothesized that RNFL loss in patients with diabetes without overt DR may represent a distinct type of diabetic neuropathy. Analyses of the status of peripheral neuropathy and cardiac autonomic neuropathy (CAN) were compared according to the location (superior or inferior) of RNFL loss in patients with type 2 diabetes without DR.
We performed a retrospective review of all patients with type 2 diabetes who underwent assessment of diabetic peripheral neuropathy, CAN, and DR from January 2008 to March 2015 at Seoul St. Mary’s Hospital. The study was conducted in accordance with the ethical standards of the Declaration of Helsinki and with the approval of the Institutional Review Board of Seoul St. Mary’s Hospital. Eligible patients were required to have no signs of a glaucomatous optic disc or DR. All patients had to meet the following inclusion criteria: ≥40 years of age, best-corrected visual acuity >20/30, no history of any retinal disease, a diagnosis of type 2 diabetes (not type 1 diabetes or gestational diabetes mellitus), no acute complications of diabetes at present, and no disease history that might affect neuropathy status.
CAN was assessed by analyzing the heart rate response using Ewing test, which includes five simple, noninvasive cardiovascular autonomic reflex tests. Peripheral neuropathy was evaluated by measuring the current perception threshold (CPT) using a Neurometer.
Evidence of definite or advanced CAN stage was significantly associated with superior RNFL defects (51.6%) compared with inferior RNFL defects (31.8%; P = 0.03). Evaluation of peripheral neuropathy using the CPT grade revealed significantly higher neuropathy grades in the group with inferior RNFL defects than in the group with superior RNFL defects (Table 1). Univariate and multivariate logistic regression analyses revealed that an abnormal CAN stage (β = 1.67; 95% CI 1.03–2.68; P = 0.04) and the CPT grade (β = 1.15; 95% CI 0.99–1.32; P = 0.04) were significantly associated with a superior RNFL defect location. Only the CPT grade was shown to be related to an inferior RNFL defect location in both univariate (β = 1.28; 95% CI 1.09–1.49; P = 0.02) and multivariate (β = 1.26; 95% CI 1.07–21.48; P < 0.01) analyses.
Comparisons of baseline characteristics, laboratory results, and diabetic neuropathy examinations according to the presence and location of RNFL defect in patients with type 2 diabetes
| . | T2D patients with no RNFL defect (n = 70) . | T2D patients with superior RNFL defect (n = 62) . | T2D patients with inferior RNFL defect (n = 44) . | P value . |
|---|---|---|---|---|
| Age, years | 54.2 ± 12.4 | 53.5 ± 11.0 | 57.0 ± 15.3 | 0.19*/0.18† |
| Sex, male:female | 36:34 | 34:28 | 18:26 | 0.24ǂ/0.13ǂ |
| Diabetes duration, years | 8.3 ± 7.2 | 8.7 ± 7.5 | 9.3 ± 6.9 | 0.67*/0.24† |
| Hypertension, n (%) | 44 (62.9) | 28 (45.2) | 20 (45.5) | 0.01ǂ/0.56ǂ |
| Diabetes treatment, n (%) | ||||
| Insulin | 10 (14.3) | 8 (12.9) | 8 (18.2) | 0.53ǂ/0.31ǂ |
| Sulfonylurea | 54 (77.1) | 40 (64.5) | 32 (72.7) | 0.12ǂ/0.24ǂ |
| Antiplatelet use, n (%) | 40 (57.1) | 24 (38.7) | 22 (50.0) | 0.11ǂ/0.16ǂ |
| Laboratory findings | ||||
| Fasting plasma glucose, mg/dL | 142.8 ± 40.0 | 146.1 ± 52.9 | 151.5 ± 84.4 | 0.74*/0.73† |
| HbA1c, % (mmol/mol) | 7.8 ± 1.7 (62 ± 8) | 7.7 ± 1.8 (61 ± 9) | 7.7 ± 1.6 (61 ± 8) | 0.83*/0.94† |
| eGFR, mL/min/1.73 m2 | 95.0 ± 27.8 | 90.6 ± 20.8 | 89.7 ± 36.2 | 0.55*/0.83† |
| ACR, μg/mg creatinine | 13.0 ± 14.7 | 27.9 ± 63.2 | 22.5 ± 34.9 | 0.05*/0.70† |
| CAN staging, n (%) | ||||
| Normal–borderline | 40 (57.1) | 30 (48.4) | 30 (68.2) | 0.53ǂ/0.03ǂ |
| Definite–severe | 30 (42.9) | 32 (51.6) | 14 (31.8) | |
| Diabetic peripheral neuropathy | ||||
| CPT grade (median/ulnar nerve) | 0.9 ± 1.9 | 1.2 ± 3.2 | 2.6 ± 3.2 | 0.01*/<0.01† |
| CPT grade (peroneal nerve) | 1.6 ± 2.4 | 2.3 ± 3.0 | 4.6 ± 3.1 | <0.01*/<0.01† |
| Mean neuropathy grade | 1.3 ± 0.5 | 1.7 ± 1.2 | 2.8 ± 1.9 | <0.01*/0.03† |
| . | T2D patients with no RNFL defect (n = 70) . | T2D patients with superior RNFL defect (n = 62) . | T2D patients with inferior RNFL defect (n = 44) . | P value . |
|---|---|---|---|---|
| Age, years | 54.2 ± 12.4 | 53.5 ± 11.0 | 57.0 ± 15.3 | 0.19*/0.18† |
| Sex, male:female | 36:34 | 34:28 | 18:26 | 0.24ǂ/0.13ǂ |
| Diabetes duration, years | 8.3 ± 7.2 | 8.7 ± 7.5 | 9.3 ± 6.9 | 0.67*/0.24† |
| Hypertension, n (%) | 44 (62.9) | 28 (45.2) | 20 (45.5) | 0.01ǂ/0.56ǂ |
| Diabetes treatment, n (%) | ||||
| Insulin | 10 (14.3) | 8 (12.9) | 8 (18.2) | 0.53ǂ/0.31ǂ |
| Sulfonylurea | 54 (77.1) | 40 (64.5) | 32 (72.7) | 0.12ǂ/0.24ǂ |
| Antiplatelet use, n (%) | 40 (57.1) | 24 (38.7) | 22 (50.0) | 0.11ǂ/0.16ǂ |
| Laboratory findings | ||||
| Fasting plasma glucose, mg/dL | 142.8 ± 40.0 | 146.1 ± 52.9 | 151.5 ± 84.4 | 0.74*/0.73† |
| HbA1c, % (mmol/mol) | 7.8 ± 1.7 (62 ± 8) | 7.7 ± 1.8 (61 ± 9) | 7.7 ± 1.6 (61 ± 8) | 0.83*/0.94† |
| eGFR, mL/min/1.73 m2 | 95.0 ± 27.8 | 90.6 ± 20.8 | 89.7 ± 36.2 | 0.55*/0.83† |
| ACR, μg/mg creatinine | 13.0 ± 14.7 | 27.9 ± 63.2 | 22.5 ± 34.9 | 0.05*/0.70† |
| CAN staging, n (%) | ||||
| Normal–borderline | 40 (57.1) | 30 (48.4) | 30 (68.2) | 0.53ǂ/0.03ǂ |
| Definite–severe | 30 (42.9) | 32 (51.6) | 14 (31.8) | |
| Diabetic peripheral neuropathy | ||||
| CPT grade (median/ulnar nerve) | 0.9 ± 1.9 | 1.2 ± 3.2 | 2.6 ± 3.2 | 0.01*/<0.01† |
| CPT grade (peroneal nerve) | 1.6 ± 2.4 | 2.3 ± 3.0 | 4.6 ± 3.1 | <0.01*/<0.01† |
| Mean neuropathy grade | 1.3 ± 0.5 | 1.7 ± 1.2 | 2.8 ± 1.9 | <0.01*/0.03† |
Data are mean ± SD unless otherwise indicated. T2D, type 2 diabetes.
One-way ANOVA between all groups.
Multiple comparisons of ANOVA between patients with type 2 diabetes with superior versus inferior RNFL defect.
χ2 test.
Our study revealed significant associations between diabetic peripheral neuropathy and inferior RNFL defects in patients with type 2 diabetes. Consistent with our findings, inferior RNFL thinning was found in patients with diabetes with peripheral neuropathy at risk for foot ulceration (5). Abnormal CAN staging was associated with the presence of superior RNFL defects. The location of RNFL loss in patients with diabetes may represent different aspects of neuropathy that involve the eye independent of DR. Our study shows that it may be possible to predict the involved type of diabetic neuropathy by examining the location of RNFL loss in the retina.
Article Information
Funding. This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning (NRF-2014R1A1A3049403).
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
Author Contributions. H.-Y.L.P., J.S., and J.H.L. researched data, contributed to discussion, and reviewed and edited the manuscript. C.K.P. reviewed and edited the manuscript. H.-Y.L.P. wrote the manuscript. C.K.P. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
