STUDY DESIGN AND PARTICIPANTS

This prospective cohort study was conducted at the Zhongshan Ophthalmic Center (ZOC). Patients diagnosed with type 2 diabetes mellitus (T2DM) were recruited for this study from a selected local community in Guangzhou, a city in Southern China, from December 2018 to September 2020. The DR screening was conducted by a DR-study team from Zhongshan ophthalmic center, a local eye hospital. The inclusion criteria were: (1) diagnosis of type 2 diabetes mellitus; (2) age 40 and older; (3) no history of ocular treatment. The exclusion criteria were as follows: (1) history of other eye disease, such as corneal opacity, ocular trauma, glaucoma, clinically significant cataract and other retinal diseases including high myopia, age-related macular degeneration, macular holes, as well as other retinal conditions reported in the literature that may affect IZ; (2) diabetic macular edema that had rendered the IZ beneath the cystoid spaces undetectable; (3) inability to cooperate with examinations; (4) poor quality of fundus or OCT images. Diabetic patients without DR (NDR) or DR patients graded as mild nonproliferative DR (NPDR) or moderate NPDR were followed up for three years to evaluate the progression of DR. We specifically targeted individuals graded as NDR, mild NPDR or moderate NPDR because focusing on the earlier stages of the disease is more relevant when predicting disease progression. And patients with severe NPDR or PDR at baseline usually needed immediate interventions such as anti-VEGF therapy or laser treatments, which could substantially change the natural progression of the disease. In order to exclude the interference of treatments on the progression of DR, patients initially graded as NDR, mild NPDR, or moderate NPDR who later underwent laser treatments or received intravitreal injections during the follow-up period were excluded. This study conformed to the Declaration of Helsinki and was approved by the Institutional Review Board/Ethics Committee of Zhongshan Ophthalmic Center, Sun Yat-sen University. Informed consent was obtained from all the participants.

OCULAR AND SYSTEMIC EXAMINATION

All subjects underwent a comprehensive ophthalmic examination including best corrected visual acuity (BCVA), intraocular pressure, slit-lamp biomicroscopy, ocular biometry, dilated fundus photography, and OCT imaging. Ocular biometry was obtained by Lenstar LS900 (Haag-Streit AG, Koeniz, Switzerland). Standardized 7-field color retinal images were obtained adhering to the Early Treatment Diabetic Retinopathy Study (EDTRS) criteria using a digital fundus camera (Canon CR-2, Tokyo, Japan) after full pupil dilation. OCT scans were obtained by SD-OCT device (SD-OCT, Heidelberg Engineering, Heidelberg, Germany) after full pupil dilation.

Additionally, we collected medical history of diabetic mellitus along with various systemic parameters, including mean arterial pression (MAP), calculated using the standard formula: MAP = (Systolic Blood Pressure + 2 × Diastolic Blood Pressure). We also obtained HbA1c, lipid profiles, C-reactive protein, estimated glomerular filtration rate (eGFR, calculated using CKD-EPI formula based on serum creatinine), serum uric acid, and urine microalbumin levels for analysis.

DR GRADING AND DR PROGRESSION

The severity of DR was assessed by standardized seven field color retinal images adhering to the modified Airlie House classification system, which grouped DR as mild NPDR, moderate NPDR, severe NPDR and PDR. And the severity score was graded at the baseline and 3-year follow-up using the Airlie House classification system. The system provides a 15-step DR severity scale. Diabetic retinopathy progression was defined as an increase in two or more steps of severity level compared with baseline. Macular edema was defined by retinal images according to ETDRS criteria and confirmed by the OCT scans and the measurements of retinal thickness.

OCT IMAGING AND ANALYSIS OF THE IZ LENGTH

OCT scan was obtained by 9 mm SD-OCT horizontal line scan through the fovea and performed centered on the fovea with the high-resolution setting and imaging averaging of 100 frames with a SD-OCT device (SD-OCT, Heidelberg Engineering, Heidelberg, Germany). Only high-quality images with image quality above 25 dB were collected for further analysis.

OCT images were all exported in tagged image file format (TIFF) without compression and then analyzed by Image-Pro Plus version 6.0.(Media Cybernetics, USA) for IZ length. IZ length was measured within the central 7000 µm diameter, centered on the fovea, since the EZ and the IZ approached one another and became indistinguishable near the 12-degree (3600-µm) peripheral from the fovea. ^, To objectively evaluate the extend of IZ, the LRPs were utilized, as it objectively confirmed the presence or absence of the IZ. The LRP revealed four prominent peaks corresponding to four hyperreflective bands of the outer retina on OCT images. ^{, , ,} The absence of the third peak corresponding to IZ indicated its absence at this location. Then, the intact IZ was traced by the measurement tool of Image-Pro Plus and the length was automatically measured in pixels and then converted to the actual length according to the scale derived from the OCT image, which in this case was 11.72 µm/pixels. If IZ was disrupted, the length was calculated as the sum of all the segments. And if IZ was disappeared within the central 7000 µm diameter, the length was recorded as zero (See Figure 1 and Supplemental Figure S2).

Measurement of IZ length using SD-OCT imaging. A. Infrared fundus image with the position marker of the OCT scan traversing the foveal center. B. Cross-section B-scan image: the white frame delineates the range of the central 7000 µm retina centered on the fovea, IZ length is measured within the central 7000 µm. C. B-scan image within the central 7000 µm: The presence and absence of IZ was objectively determined with the longitudinal reflectivity profile (LRP). The yellow frame delineates the segment with IZ absent and the blue frame delineates the segment with IZ detectable (upper figure). IZ was highlighted in red (lower figure). IZ was traced and its length was automatically measured. D. The LRPs corresponded to the segments in figure C: the top graph corresponded to the segment in yellow frame where IZ was absent, and the LRP presented only three high-reflectivity peaks; and the bottom graph corresponded to the segment in blue frame where IZ was detectable, and the LRP presented four hyperreflectivity peaks. ELM = external limiting membrane; EZ = ellipsoid zone; IZ = interdigitation zone; RPE = retinal pigment epithelium.

Repeatability of IZ length was assessed in 30 patients randomly selected from the enrolled participants. The IZ length was measured thrice for each eye by the same operator, and the intraclass correlation coefficient (ICC) was calculated to evaluate the repeatability of IZ length measurement. The ICC was 0.994 (95% CI 0.988-0.997) indicating high repeatability.

STATISTICAL ANALYSIS

The research only included data from one eye of each patient for analysis: the worse eye was adopted in patients with different DR gradings in both eyes, and the eye with better OCT image quality was chosen with the same DR gradings in both eyes. The statistical analyses were performed using SPSS Statistics version 24 (IBM, Armonk, NY, USA). Kolmogorov-Smirnov was used to test normal distribution. Differences in baseline characteristics were evaluated using the independent samples t-test for normally distributed data, and the Chi-square test for qualitative data. Linear regression analysis was performed to assess the associations between IZ length and DR severity, the presence of DME, and BCVA. Variates were included in the multivariable linear analysis after the univariable linear regression indicating significant correlation. The multivariable model 1 investigated the relationship between IZ length with DR severity and DME after adjusting for age, duration of diabetes, and HbA1c levels. The multivariable model 2 further adjusted for mean arterial pressure, microalbuminuria levels, eGFR, and LDL-c levels. Sensitivity analyses were performed to exclude the influence of DME cases, which excluding DME cases and reassessed the relationships between IZ length and DR severity, as well as between IZ length and BCVA. Logistic regression models were used to examine the relationship between IZ length and the risk of DR progression after 3-year follow-up. To address potential bias from missing data, a sensitivity analysis was conducted using multiple imputation (MI) for participants lost to follow-up. Twenty imputed datasets were generated, and the relationship between baseline IZ length and DR progression was assessed via logistic regression in each dataset. Results from the imputed datasets were pooled according to Rubin’s rules to derive pooled estimates. Furthermore, ROC curves were also constructed to evaluate the ability of the IZ length in discriminating individuals with high risk of DR progression and DR occurrence among diabetic patients. A two-sided P value less than .05 was considered to represent statistical significance.

PATIENT DEMOGRAPHICS AND CLINICAL CHARACTERISTICS

A total of 231 eyes of 231 diabetic patients were included in the analysis including 117 eyes without DR and 114 eyes with DR. The demographic and baseline clinical characteristics were presented in Table 1 . One hundred and nineteen patients (51.52%) were female. The average age of the patients was 63.80 ± 6.98 years, and the average duration of diabetes was 11.30 ± 6.81 years. Among NDR patients, 60 patients (51.28%) were female and the average age was 63.62 ± 6.42 years. Among DR patients, 59 patients (51.75%) were female and the average age was 63.99 ± 7.54 years. Patients with DR had a longer duration of diabetes, higher HbA1c levels, lower eGFR, higher microalbuminuria levels, and poorer BCVA (all P < .05) compared to those without DR. Other parameters were similar between the DR patients and patients without DR (all P > .05).

TABLE 1

Demographic and Clinical Features of the Participants.

Characteristics	Overall	NDR	DR	P Value
No. of Subjects	231	117 (50.65)	114 (49.35)	–
Female, %	119 (51.52)	60 (51.28)	59 (51.75)	.943
Age, year	63.80 ± 6.98	63.62 ± 6.42	63.99 ± 7.54	.684
Duration Of Diabetes, Year	11.30 ± 6.81	8.51 ± 5.62	14.17 ± 6.76	<.001
Mean Arterial Pressure, mmHg	92.71 ± 11.12	91.61 ± 10.35	93.85 ± 11.79	.126
HbA1c, %	7.89 ± 1.67	7.15 ± 1.16	8.64 ± 1.78	<.001
Total Cholesterol, mmol/L	5.00 ± 1.29	5.04 ± 1.16	4.97 ± 1.42	.675
eGFR, mL/min/1.73 m 2	89.06 ± 15.20	91.83 ± 13.31	86.23 ± 16.51	.010
HDL-c, mmol/L	1.27 ± 0.36	1.26 ± 0.33	1.28 ± 0.39	.995
LDL-c, mmol/L	3.16 ± 1.08	3.18 ± 0.92	3.13 ± 1.23	.343
Triglycerides, mmol/L	2.31 ± 1.51	2.37 ± 1.71	2.25 ± 1.28	.928
Serum Uric Acid, µmol/L	378.01 ± 113.83	371.69 ± 101.25	384.50 ± 125.55	.516
C-reactive Protein, mg/L	2.28 ± 2.81	2.08 ± 2.14	2.49 ± 3.36	.733
Microalbuminuria, mg/mL	6.35 ± 15.34	3.07 ± 9.26	9.70 ± 19.19	<.001
BCVA, logMAR	0.09 ± 0.11	0.05 ± 0.06	0.13 ± 0.12	<.001
Axial Length, mm	23.36 ± 0.87	23.31 ± 0.80	23.42 ± 0.94	.359

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

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Population-Based Performance of Inflammatory Markers in Giant Cell Arteritis
2. Neurofibromatosis Type 1 (NF1)–Related Ocular Signs: New Insights on Their Prevalence, Incidence, and Genotype-Phenotype Correlation in NF1 Children
3. Adherence and Persistence on Prostaglandin Analogues for Glaucoma: A Systematic Review and Meta-Analysis
4. Comment on: Multifocal Vitelliform Paravascular Retinopathy (MVPR): A New Disorder in the Vitelliform Spectrum

Stay updated, free articles. Join our Telegram channel

Join