Wider OCTA Scans Reveal More Diabetic Retinopathy Abnormalities

Key Takeaways

• In treatment-naïve diabetic retinopathy, wider OCTA imaging fields yielded higher measured NPI and IRMA counts.
• Wider imaging detected more neovascularization in proliferative disease.
• The largest OCTA field yielded the highest reported NPI discrimination for both staging comparisons.
• On small-field images, combining NPI with IRMA counts yielded better numerical staging discrimination than NPI alone.

In diabetic retinopathy (DR), does the retinal field sampled by optical coherence tomography angiography (OCTA) affect assessment of the vascular abnormalities used for staging? A restricted view may characterize a different portion of the retina than a wider view. For patients who have not received treatment, the choice of scan field could therefore affect how imaging-based biomarkers distinguish disease stages.

Investigators conducted a cross-sectional OCTA scan-protocol comparison in 110 eyes from 74 treatment-naïve patients with DR. They used 6 × 6 mm, 12 × 12 mm, and 29 × 24 mm fields to measure nonperfusion index (NPI), intraretinal microvascular abnormality (IRMA) counts, and neovascularization (NV) counts. Receiver operating characteristic (ROC) analysis assessed discrimination of proliferative DR (PDR) from nonproliferative DR (NPDR) and advanced from early DR.

As the field widened, measured NPI and IRMA counts increased across DR severities, and wider scans detected more NV in PDR.

On the widest field, NPI had the highest reported ROC estimate for PDR versus NPDR: 0.86 (95% confidence interval [CI], 0.79–0.93).

For advanced versus early DR, NPI on that field likewise had the highest reported estimate: 0.92 (95% CI, 0.88–0.97).

On the smallest field, NPI plus IRMA counts yielded an ROC estimate of 0.87 (95% CI, 0.81–0.94) for advanced versus early DR.

NPI alone yielded 0.81 (95% CI, 0.72–0.89) for the same comparison.

For PDR versus NPDR on the smallest field, NPI plus IRMA counts yielded 0.82 (95% CI, 0.74–0.91).

NPI alone yielded 0.73 (95% CI, 0.62–0.83) for that comparison.

Higher biomarker counts on wider scans reflect a larger sampled retinal area, not documented disease progression within an eye. Cross-sectional discrimination between existing stage groups does not establish improved prospective patient outcomes or a routine imaging recommendation.

The authors concluded that NPI remained useful as a standalone staging biomarker across scan fields, even as the combination with IRMA counts provided better numerical discrimination on the smallest field. [h2] [/h2]

Recommended Reading

• For more on diabetic retinopathy progression: Early Diabetic Retinopathy Regression in Teleophthalmology