A cross-sectional observational study was conducted at Massachusetts Eye and Ear Infirmary, Harvard Medical School, from January 2019 to June 2021. The study was approved by the institutional review board of Mass General Brigham and registered under the identifier 2019P001863. All patients provided written informed consent before enrolling in this study. The conduct of this study adhered to the principles outlined in the Declaration of Helsinki and the regulations governing the Health Insurance Portability and Accountability Act.

Patients with type 1 and 2 DM between the ages of 18 and 90 years, with a minimum Snellen best corrected visual acuity (BCVA) of 20/200 were enrolled to participate in the study between January 2019 and April 2021. We excluded eyes with VA <20/200 (n=22), other ocular comorbidities (epiretinal membrane, n = 16; neovascular glaucoma, n = 5; branch retinal vein occlusion, n = 7; central retinal vein occlusion, n = 3; non-arteritic ischemic optic neuropathy, n = 3; central serous chorioretinopathy, n = 2; lamellar hole, n = 1; history of uveitis, n = 1; history of endophthalmitis following PPV, n = 1; central macular scar, n = 1), and poor image quality (signal strength intensity <7, n = 23, and poor image quality due to artifacts and defocus, n = 8).

All patients received a complete ophthalmic examination including BCVA, intraocular pressure measurement, slitlamp, and dilated fundus examination. Angiograms of 12 × 12 mm centered on the fovea were acquired using the 100-kHz expanded field swept-source OCTA (SS-OCTA) device (PLEX Elite 9000, Carl Zeiss Meditec Inc), which uses a tunable laser of central wavelength between 1040 and 1060 nm with a bandwidth of 100 nm. It has an A-scan depth of 3 mm (in tissue), axial resolution (digital) of 1.95 µm (in tissue), and a transverse resolution of 20 µm. The 12 × 12-mm scans had an approximately 50 ° to 60 ° field of view of the central posterior pole of retina. Spectralis OCT2 B-scan (Heidelberg Engineering) and ultrawide-field color fundus photography (Optos) with or without fundus fluorescein angiography were acquired on the same day. The grading of DR was performed by experienced senior retina faculty (J.B.M., D.G.V., D.H., J.W.M., L.A.K., D.E.) based on the clinical examination and the ancillary imaging using International Clinical DR Disease Severity Scale.

Images with signal strength <7 were excluded to control for image quality. Poor-quality images due to media opacity, defocus, and presence of various artifacts (motion, edge, threshold) were also excluded. Standard parameters and illumination settings were used to evaluate images. A single experienced grader (I.G.) evaluated all scans for quality control, which was further validated by the senior author (J.B.M.). OCTA metrics ( Figure 1 ) like vessel density (VD) and vessel skeletonized density (VSD) on superficial capillary plexus (SCP), deep capillary plexus (DCP), and full-thickness retina slabs, and foveal avascular zone (FAZ) metrics on the superficial retinal layer were quantified using Macular Density v0.7.3 on the ARI Network (Zeiss Portal v5.4-1206), which employs legacy segmentation to define superficial (between the inner limiting membrane and 70% of the distance between it and the outer plexiform layer) and deep slab (between the inner plexiform layer and the outer plexiform layer, situated 110 mm above the retinal pigment epithelium-fit line).

Representative quantitative OCT angiography metrics analyzed on 12 × 12-mm images. Red circle represents foveal avascular zone outline. Blue areas represent nonperfusion areas. OCT = optical coherence tomography.

The deep slabs were generated after the removal of projection artifacts. VD was calculated after binarization of the angiogram layers to generate a black-and-white image and defined as the total area of perfused vasculature per unit area within a region of measurement, ranging from 0 to 1. VSD was calculated after conversion of the binarized vessel to a line of 1 pixel width, per unit area of measurement, and was defined as the total length of skeletonized vessels. The FAZ parameters were assessed as area, perimeter, and circularity (uniformity index with perfect circle having a value of 1). Additionally, superficial slabs of 12 × 12-mm angiograms were used to quantify NPA (mm ²) using our previously published semiautomated algorithm on FIJI (an expanded version of ImageJ: 2.0.0-rc-69/1.52p; National Institutes of Health) by a single experienced grader (I.G.) ( Figure 1 ).

Electronic medical records were reviewed for all patients to record the following parameters: age, gender, race, smoking status, duration of DM, type of DM, body mass index, mean arterial blood pressure, glycated hemoglobin (HbA _1c ), lipid profile, urine albumin, urine creatinine, serum creatinine, and eGFR (on day of imaging or within 6 months pre or post imaging). The ocular parameters collected included BCVA, intraocular pressure, lens status, DR staging, history of prior interventions (laser or intravitreal injections), and presence or absence of diabetic macular edema.

The primary outcomes included presence of albuminuria and CKD. The albumin-to-creatinine ratio (ACR) was calculated from urine albumin and creatinine levels. Patients with ACR more than 30 mg/g was defined as albuminuria as per the international guidelines. Presence of CKD was defined as an eGFR ≤60 mL/min/1.73 m ², or >60 mL/min/1.73 m ² with presence of evidence of CKD (eg, albuminuria). ^, Additionally, we assessed correlation with severity of albuminuria and CKD. ACR was further subclassified into 3 categories based on severity: A1 (<30 mg/g), A2 (30-300 mg/g), and A3 (>300 mg/g). CKD is routinely grouped into 5 categories, G1 to G5 based on eGFR. To account for a smaller percentage of patients in G4 and G5 categories and to enable easier interpretable results, CKD severity was grouped into 3 categories: normal (≥90 mL/min/1.73 m ²), mild decrease (60-89 mL/min/1.73 m ²), and moderate to severe decrease (≤59 mL/min/1.73 m ²).

The secondary outcome was to compare the association between OCTA metrics and the KDIGO prognostic categories for CKD progression and/or mortality. The KDIGO consensus guidelines outlined a prognostic classification of CKD based on both albuminuria and eGFR into low risk (green), moderately increased risk (yellow), high risk (orange), very high risk (red) categories based on longitudinal patient data to study long-term prognosis including kidney failure, cardiovascular complications, and all-cause mortality from 45 studies, assimilated in a meta-analysis. ^, Pairwise comparisons were performed between the 4 groups to assess the role of OCTA metrics as prognostic markers for DN.

Finally, a subgroup analysis was performed in patients with no DR and mild nonproliferative DR (NPDR) to identify whether early microvascular retinal changes were associated with the presence of DN.

Statistical analysis was conducted using R version 4.1.3 (R Foundation for Statistical Computing). Descriptive methods were employed to characterize the population demographics and ocular characteristics. A 2-sided P value of less than.05 was set as the level of significance. Mixed-effects multiple logistic and linear regression models, fitted using restricted maximum likelihood, were employed to account for the correlation between both eyes from the same patients. All multivariate models were adjusted for clinically relevant objective confounding factors for progression of CKD (age, smoking status, HbA _1c , and mean arterial blood pressure).

Our final cohort comprised 375 eyes of 234 patients. The baseline characteristics are summarized in Tables 1 and 2 . Majority of patients were male (53.4%) and of White race (50.9%). Most patients had type 2 DM (83.8%) with a median duration of disease of 16 years. Almost half of patients had albuminuria (47.4%), of which 66 (59.5%) and 45 (40.5%) patients were grouped into the A2 and A3 groups, respectively, and 145 patients had CKD based on eGFR (62%). The CKD categories based on eGFR were roughly equally distributed in normal (72; 30.8%), mild CKD (89; 38%), and moderate-to-severe (73; 31.2%) CKD groups, respectively. Nearly half of patients fell in the KDIGO low-risk and moderate-risk categories. Thirty-four patients had unknown ACR but 3 patients were grouped into very high risk (red) KDIGO group as they had severe reduction in eGFR <29. Among the 375 eyes, 48 eyes (12.8%) had no DR, 190 eyes (50.7%) had NPDR, and 137 eyes (36.5%) had proliferative DR. Approximately one-fourth of eyes received laser and vitrectomy procedures.

Mixed-level multiple logistic regression was performed controlling for age, mean arterial blood pressure, HbA _1c , and smoking status ( Table 3 ). Among the OCTA metrics, increased NPA was associated with both presence of CKD (odds ratio [OR]: 1.060, 95% CI 1.058-1.061; P <.001) and albuminuria (OR: 1.171, 95% CI 1.008-1.359; P =.038).

TABLE 3

Mixed Effects Multiple Logistic Regression Model Results by Presence of CKD and Albuminuria, Adjusted for Age, Mean Arterial Blood Pressure, HbA _1c , and Smoking Status

OCTA Vascular Metrics (12 × 12 mm 2)	Presence of CKD a		Presence of Albuminuria b
	OR (95% CI)	P Value	OR (95% CI)	P Value
NPA	1.060 (1.058-1.061)	<.001	1.171 (1.008-1.359)	.038
VSD (SCP)	0.989 (0.618-1.583)	.962	0.955 (0.465-1.960)	.899
VSD (DCP)	0.980 (0.643-1.493)	.926	0.945 (0.480-1.859)	.868
VSD (retina)	1.00 (0.650-1.537)	.998	0.987 (0.517-1.885)	.969
VD (SCP)	0.413 (<0.001-11039607)	.919	0.102 (<0.001-14920471006)	.862
VD (DCP)	0.517 (<0.001-2240030)	.933	0.020 (<0.001-1135115222)	.757
VD (retina)	0.519 (<0.001-5251372)	.937	0.187 (<0.001-7419777989)	.902
FAZ area	0.105 (0.001-8.085)	.309	0.629 (0.238-1.659)	.349
FAZ perimeter	0.985 (0.803-6.554)	.884	0.869 (0.696-1.087)	.218
FAZ circularity	0.802 (0.001-1182.044)	.953	0.676 (<0.001-9623.9)	.936

Only gold members can continue reading. Log In or Register to continue

Share this:

• X
• Facebook

Related

Related posts:

1. Long-Term Follow-Up of PLACE and SPECTRA Trials: Outcomes After Successful and Unsuccessful Half-Dose Photodynamic Therapy for Chronic Central Serous Chorioretinopathy
2. Ophthalmic Manifestations of Danon Disease: A Systematic Review
3. The Role of Geography in United States Retinoblastoma Care
4. Noninfectious Posterior Uveitides — Atypicals, Variants, and Masquerades: The Jungle of Differential Diagnosis

Stay updated, free articles. Join our Telegram channel

Join