Findings from a study recently published in the American Journal of Ophthalmology assessed the longitudinal changes in choroidal thickness (CT) and ganglion cell-inner plexiform layer thickness (GC-IPLT) across phenotypes of type 2 diabetes mellitus (T2DM) patients.
Give me some background.
Different subtypes of diabetes mellitus (DM) have demonstrated distinct disease progression profiles and complications risks.
For example: Severe autoimmune diabetes (SAID) and severe insulin deficiency diabetes (SIDD) are associated with a higher risk of diabetic retinopathy (DR), while severe insulin resistance diabetes (SIRD) has the highest risk of diabetic kidney disease.
Which leads us to …
Consequently, a research team based at Zhongshan Ophthalmic Center (ZOC), Sun Yat-Sen University, in Guangzhou, China, used swept-source optical coherence tomography (SS-OCT) to compare changes in CT and GC-IPLT in different DM subtypes.
Now talk about the study.
In this prospective, longitudinal cohort study, investigators included 1,070 patients (59.1% female, aged 35-80 years) with T2DM who underwent annual comprehensive systemic and ophthalmic examinations over 4 years at ZOC.
- Note: Retinal status was graded using fundus photography and SS-OCT imaging.
The setup: T2DM patients were categorized into five groups based on beta-cell function and insulin resistance:
- SAID
- SIDD
- SIRD
- Mild obesity-related diabetes (MOD)
- Mild age-related diabetes (MARD)
Findings?
Over a median 4.11-year follow-up period, CT and GC-IPLT decreased significantly across all T2DM groups.
Choroidal thinning rates were most pronounced in:
- SIDD (−6.5 ± 0.53 µm/year and −3.5 ± 0.24%/year)
- SAID (−6.27 ± 0.8 µm/year and −3.19 ± 0.37%/year).
Meanwhile, MARD demonstrated the slowest thinning rates (−3.63 ± 0.34 µm/year and −1.98 ± 0.25%/year).
What about GC-IPLT thinning rates?
SIRD exhibited the greatest GC-IPLT loss (−0.66 µm/year and −0.91%/year), with the least in SIDD (−0.36 µm/year and −0.49%/year, all P < 0.001).
After adjusting for confounding variables, the SIDD and SAID groups showed faster CT thinning than MARD (−2.57 µm/year [95% confidence interval (CI) −4.16 to −0.97], P = 0.002) and (−2.89 µm/year [95% CI −4.12 to −1.66], P < 0.001), respectively.
GC-IPLT thinning was notably accelerated in SIRD compared to MARD, but slower in SIDD relative to MARD (differences of −0.16 µm/year [95% CI: −0.3 to −0.03], P = 0.015 and 0.15 µm/year [95% CI 0.03-0.27], P = 0.015, respectively).
Expert opinion?
According to the study authors, “Combining [these] findings with the existing literature suggests that accelerated GC-IPLT in diabetic patients may serve as a potential biomarker for conditions such as dementia or peripheral neuropathy.”
Further, they found that metabolic status significantly influenced choroidal microcirculation and retinal nerve integrity in patients with T2DM prior to the onset of clinical DR.
Limitations?
These included:
- Participants were primarily Chinese, potentially limiting the generalizability of these findings.
- Diurnal fluctuations in OCT-measured CT values up to 20-30 µm daily suggest the need for additional considerations in future investigations.
- The researchers did not assess the relationship between CT/GC-IPLT changes and functional ocular changes or if GC-IPLT changes correlated with neurodegenerative processes that led to visual impairment.
Take home.
These findings suggest that microvascular damage in the choroid is associated with SIDD patients, while early signs of retinal neurodegeneration (i.e., GC-IPLT loss) can be observed in SIRD patients.
In addition: Losses in CT and GC-IPLT may precede clinical manifestations of DR in T2DM patients.