Review Article

An Update on Corneal Biomechanics and Architecture in Diabetes

Table 2

Summary of “in vivo” studies about endothelial status in diabetic patients compared with healthy controls.

Author, yearType of studyStudy groupsTechnologyParametersResults

Shenoy, 2009 [131]Case-control
Prospective
110 diabetic patients (110 eyes)
27 T1D
83 T2D
110 controls (110 eyes)
NIDEK® confoscan 2.ECD, coefficient of variability of cell size of cells showing polymegathism, percentage of hexagonal cells showing pleomorphism.(i) ECD in eyes was negatively associated with the diabetes status.
(ii) The coefficient of variability in endothelial cells with polymegathism was 12 (8 to 16) more among eyes of diabetic patients than that of controls.
(iii) The corneal endothelial cells with pleomorphism were 9% less in controls compared to the diabetic subjects.
Módis, 2010 [132]Case-control
Prospective
21 insulin-dependent T1D patients (41 eyes)
30 patients with non-insulin-dependent T2D (59 eyes).
Control group 1 (22 patients, 40 eyes). Age-matched normal subjects with T1D group)
Control group 2 (30 patients, 60 eyes). Age-matched normal subjects with T2D group)
Wide-field contact specular microscope (Tomey EM-1000, Tokyo, Japan)ECD, mean cell area, CV, CCT, IOPT1D
(i) ECD decreased in T1D in comparison with controls.
(ii) CCT thicker in T1D in comparison with controls.
HbA1c level was inversely correlated with the ECD and correlated with the mean endothelial cell area.
(iii) Positive correlation between glucose level and ECD, endothelial cell area and CCT.
(iv) Negative correlation between ECD and duration of the disease/insulin therapy.
T2D
(i) No differences were found in the evaluated values compared to controls. No correlations were founded
Urban, 2013 [133]Case-control
Prospective
123 children and adolescents with T1D (123 eyes)
124 controls (124 eyes)
Topcon SP-2000P endothelial microscope.ECD and CCT(i) ECD was lower in children-adolescent T1D compared to controls.
(ii) CCT was higher in children-adolescent T1D compared to controls.
(iii) There was no correlation between ECD and metabolic control, HbA1c level and plasma creatinine level.
(iv) Correlation between ECD, CCT, and duration of diabetes was statistically significant.
Storr-Paulsen, 2014 [134]Case-control
Prospective
107 T2D
128 controls
SP 2000P; Topcon, Tokyo, Japan.ECD, CV, hexagonality percentage and CCT(i) No differences between groups.
(ii) Higher HbA1c was associated with lower ECD.
(iii) CCT increased in the T2D group.
Leelawongtawun, 2015 [135]Case-control
Prospective
90 diabetic patients (171 eyes)
(i) 1 patient (two eyes) with severe NP-DR.
(ii) 7 patients (11 eyes) with moderate NP-DR
(iii) 13 patients (24 eyes) with mild NP-DR
(iv) 71 patients (134 eyes) with no DR
90 controls (156 eyes).
Specular microscope (Confoscan4, Nidek)ECD, percentage of polymegathism, and hexagonality percentage.(i) No differences between diabetes and controls
(ii) The over one year diabetic patients had a decreased percentage of hexagonal cell compared to controls.
(iii) The over two years diabetic patients had a decreased percentage of hexagonal cell and an increased percentage of polymegathism compared to controls.
Calvo-maroto, 2015 [13]Retrospective77 noninsulin T2D (77 eyes):
(i) Short-term diabetic subjects (recently diagnosed,<1 year since diagnosis)
(ii) Long-term diabetic subjects (diagnosed and treated for 10 years or more) s80 controls (80 eyes)
Topcon SP-3000P noncontact specular microscopeCCT, ECD(i) CCT higher in long-term diabetic patients when compared with short-term diabetic patients and controls.
(ii) ECD lower in long-term diabetic patients when compared with short-term diabetic patients and controls.
Szalai, 2016 [12]Case-control
Prospective
28 T1D (28 eyes)
18 with DR
10 without DR
17 age-matched controls (17 eyes)
Corneal confocal microscopy with Heidelberg Retina Tomograph III Rostock Cornea Module (HRT III RCM, Heidelberg Engineering GmbH, Heidelberg, Germany)ECD Other (epithelial, stromal density. Subbasal nerve morphology)ECD was lower in T1D with and without DR compared to controls.
Anbar, 2016 [136]Case-control
Prospective
80 T1D children (160 eyes)
40 controls (80 eyes)
Noncontact specular microscope (Topcon SP-1P, Tokyo, Japan).CCT, ECD, polymegathism, and pleomorphism(i) CCT higher in the T1D group.
(ii) ECD lower in the T1D group.
(iii) Percentage of hexagonality lower in the T1D group. Polymegathism higher in the T1D group.
(iv) All changes are correlated only with the duration of diabetes
Leelawongtawun, 2016 [137]Case-control
Prospective
148 diabetes (271 eyes). Divided based on diabetes duration
(i) Below 5 years
(ii) 5 to 10 years
(iii) Over 10 years
(iv) 46 controls (82 eyes)
Specular microscope (Confoscan4 (CS4), Nidek)ECD, percentage of polymegathism and hexagonality percentage(i) ECD was lower in all diabetes groups compared to controls.
(ii) In all groups of diabetes, the polymegathism percentage was more than while the hexagonality percentage was less than controls.
(iii) There were no differences in all endothelial parameters between 3 groups of diabetes.
Galgauskas, 2016 [77]Case-control
Prospective
62 T2D (123 eyes):
(i) 22 (17.9%) eyes with DR
(ii) 10 (8.1%) eyes with macular edema
65 controls (120 eyes)
Noncontact specular microscope (SP-9000; Konan Medical Inc., Hyogo, Japan)CCT, ECD, average size, hexagonality percentage and polymegathism(i) ECD lower in diabetes than in controls.
(ii) CCT higher in diabetes than in controls.
(iii) Hemoglobin A1C and the duration of diabetes not associated with any of the examined parameters
El-agamy, 2017 [79]Case-control
Prospective
57 T2D (57 eyes): 36 eyes without DR
14 eyes with NP-DR
7 eyes with P-DR.
45 controls (45 eyes)
EM-3000 Specular MicroscopeCCT, ECD, CV and hexagonality percentage(i) ECD lower in diabetes than in controls.
(ii) CV higher in diabetes.
Islam, 2017 [138]Case-control
Prospective
149 diabetes (149 eyes)
(i) 52 T1D
(ii) I97 T2D
149 controls(149 eyes)
SP–3000P, Topcon Corporation, JapanECD, average cell size, CV and hexagonality percentage(i) ECD lower in the diabetes group
Diabetes longer than 10 years had significantly lower ECD and larger average size.
(ii) Diabetes duration was correlated with ECD, polymegethism and hexagonality
Qu, 2017 [13]Case-control
Prospective
87 T2D (87 eyes): (i) 48 eyes without cornea fluorescein staining
(ii) 39 eyes with cornea fluorescein staining
51 controls (51 eyes)
Keratograph 5M (K5M; OCULUS Optikgerate GmbH, Wetzlar, Germany)Basal epithelial cell density, subbasal nerve plexus density, langerhans cell density and ECD.No differences in ECD between groups

T1D: type 1 diabetes; T2D: type 2 diabetes; DR: diabetic retinopathy; NP-DR: nonproliferative diabetic retinopathy; P-DR: proliferative diabetic retinopathy; ECD: endothelial cell density; CV: coefficient of variation of cell area; CCT: central corneal thickness; IOP: intraocular pressure.