Table 5.
Retinopathy results
| Study | Study size, No. | Population + age, y | Diabetes duration, y | Mean HbA1c, % | Duration of CGM trace, d | Findings |
|---|---|---|---|---|---|---|
| Picconi et al (2016) (34) cross-sectional | 37 | T1D | 19.0 ± 10.4 | 7.9 ± 1.1 | 3 | Inner nuclear layer thickness correlated with |
| 41.5 ± 10.0 | • CONGA-1: r = 0.40, P = .03 | |||||
| • CONGA-2: r = 0.39, P = .03 | ||||||
| • CONGA-4: r = 0.41, P = .02 | ||||||
| Retinal nerve fiber layer thickness correlated with LBGI: r = –0.38, P = .03 | ||||||
| Sartore et al (2013) (35) cross-sectional | 68 | T1D, T2D | 15.0 ± 8.3 | 8.1 ± 1.6 | 3 | Retinopathy associated with |
| 48.6 ± 13.8 | • SD: OR = 1.03 (1.01-1.06), P = .01a | |||||
| • CONGA-2: OR = 1.02(1.00-1.04), P = .04a | ||||||
| • HBGI: OR = 1.10 (1.01-1.18), P = .03a | ||||||
| Retinopathy not associated with MAGE: OR = 1.74 (0.69-4.40), P = .24a | ||||||
| Šoupal et al (2016) (28) cross-sectional | 32 | T1D | 19.5 ± 5.5 | 8.6 ± 0.9 | 12-14 | Retinopathy associated with SD: 4.1 ± 0.7 vs 3.5 ± 0.8 mmol/L, P = .03a |
| 41.5 ± 11.5 | ||||||
| Stem et al (2016) (36) cross-sectional | 81 | T1D | 14.0 ± 6.7 | 7.9 ± 1.0 | 5 | Neurodegenerative structural retinal changes were associated with |
| 46.5 ± 16.5 | • LBGI: β = –0.47, P = .02, R2 = 0.28b | |||||
| • Area under curve for hypoglycemia: β = –0.45, P = .02, R2 = 0.26b | ||||||
| Neither presence of retinopathy nor neuroretinal function associated with LBGI or area under curve for hypoglycemia | ||||||
| Lu et al (2018) (37) cross-sectional | 3262 | T2D | 8.1 ± 6.8 | 8.9 ± 2.2 | 3 | Retinopathy severity associated with |
| 60.2 ± 12.0 | • Lower TIR: P < .01a | |||||
| • Lower TIR quartiles: r = –0.15, P < .01a | ||||||
| • SD: P < .01a | ||||||
| • CV: P < .01a | ||||||
| • MAGE: P < 0.01a | ||||||
| Any diabetic retinopathy negatively associated with 10% increase in TIR: OR = 0.92 (0.88-0.96), P < .01b | ||||||
| Mild nonproliferative retinopathy negatively associated with | ||||||
| • 10% increase in TIR: OR = 0.93 (0.87-0.99), P = .02b | ||||||
| • Highest compared to lowest quartile TIR: OR = 0.56 (0.36-0.87), P = .01b | ||||||
| Moderate nonproliferative retinopathy negatively associated with | ||||||
| • 10% increase in TIR: OR = 0.91 (0.84-0.98), P = .01b | ||||||
| • Highest compared to lowest quartile TIR: OR = 0.48 (0.27-0.83), P = .01b | ||||||
| Vision-threatening retinopathy negatively associated with | ||||||
| • 10% increase in TIR: OR = 0.91 (0.85-0.98), P = 0.02b | ||||||
| • Highest compared to lowest quartile TIR: OR = 0.53 (0.30-0.91), P = .02b | ||||||
| Lu et al (2019) (38) cross-sectional | 3119 | T2D, LADA | 7.7 ± 6.3 | 8.9 ± 2.1 | 3 | Retinopathy associated with |
| 57.6 ± 10.1 | • SD: OR = 1.15 (1.03-1.29), P = 0.02b | |||||
| • MAGE: OR = 1.21 (1.11-1.31), P < .01a | ||||||
| • CV: OR = 1.16 (1.07-1.26), P < .01a | ||||||
| Retinopathy associated with increasing quartiles of SD and MAGE in T2D: P < .01 | ||||||
| No significant associations for LADA |
Values expressed as mean ± SD or median (interquartile range).
Abbreviations: CGM, continuous glucose monitoring; CONGA, continuous overall net glycemic action; HbA1c, glycated hemoglobin A1c; LADA, latent autoimmune diabetes of adulthood; LBGI, low blood glucose index; SD, SD of blood glucose levels; HBGI, high blood glucose index; MAGE, mean amplitude of glycemic excursions; T1D, type 1 diabetes; T2D, type 2 diabetes; TIR, time in range.
a Univariable analysis.
b Multivariable analysis.