PDR, macular oedema and diabetic trials
Key Takeaways
PDR involves neovascularization with risks of haemorrhage and tractional detachment.
Macular-oedema treatment depends on centre involvement, vision, anatomy, pressure and treatment burden.
DRS, ETDRS and DRCR trials answer different questions with specific populations and follow-up periods.
Proliferative Diabetic Retinopathy (PDR)
Proliferative diabetic retinopathy is defined by the proliferation of fragile extraretinal new vessels (neovascularisation) that penetrate the internal limiting membrane (ILM) to arborize along the posterior hyaloid interface or scaffold onto the cortical vitreous.
- Neovascularisation of the Disc (NVD): New abnormal vessels located on the optic nerve head or within one disc area (1 DD) of the disc margin.
- Neovascularisation Elsewhere (NVE): New abnormal vessels located more than one disc area from the optic nerve head margin.
The DRS High-Risk PDR Criteria
The landmark Diabetic Retinopathy Study (DRS) established the formal definition of High-Risk PDR, which mandates immediate panretinal photocoagulation (PRP) or anti-VEGF therapy to prevent severe visual loss:
- NVD to disc area (even in the complete absence of vitreous or preretinal haemorrhage);
- Any NVD (regardless of area) accompanied by acute vitreous haemorrhage or preretinal (subhyaloid) haemorrhage;
- NVE disc area accompanied by acute vitreous haemorrhage or preretinal haemorrhage.
Diabetic Macular Oedema (DMO): Definitions, Classification & Biomarkers
Diabetic macular oedema is the principal cause of visual acuity decline in patients with diabetic retinopathy, occurring in both NPDR and PDR.
Center-Involving vs. Non-Center-Involving DMO
Modern clinical management has transitioned from historical biomicroscopic definitions to high-resolution Spectral-Domain Optical Coherence Tomography (SD-OCT):
- Center-Involving DMO (CI-DMO): Characterized by retinal thickening, intraretinal cystoid spaces, or subretinal neurosensory fluid involving the central 1-mm foveal subfield. On OCT thickness maps, CI-DMO is defined by central subfield thickness (CST) exceeding normative machine-specific thresholds:
- Heidelberg Spectralis: in women; in men;
- Zeiss Cirrus: in women; in men.
- Non-Center-Involving DMO (NCI-DMO): Retinal thickening or hard exudate rings situated within the ETDRS macular grid but sparing the central 1-mm foveal subfield.
Historical ETDRS Clinically Significant Macular Oedema (CSMO)
Prior to optical coherence tomography, the ETDRS defined CSMO solely based on contact lens slit-lamp biomicroscopy. The presence of any one of the following three features fulfilled the diagnosis of CSMO:
- Retinal thickening situated within of the center of the macula;
- Hard exudates situated within of the center of the macula, provided they are accompanied by adjacent retinal thickening (which may itself lie outside the zone);
- A zone of retinal thickening disc area in size, any portion of which is situated within 1 disc area () of the center of the macula.
OCT Morphological Patterns and Prognostic Biomarkers
Cross-sectional SD-OCT resolves three distinct morphological manifestations of DMO:
- Spongiform / Diffuse Retinal Thickening: Diffuse inner and outer retinal swelling with reduced optical reflectivity, representing intracellular cytotoxic oedema of Müller cells.
- Cystoid Macular Oedema (CMO): Well-defined, hyporeflective round or oval spaces primarily within the inner nuclear layer (INL) and outer plexiform layer (OPL), caused by extracellular fluid pooling and Müller cell septal rupture.
- Subretinal Neurosensory Detachment: Accumulation of hyporeflective serous fluid between the photoreceptor outer segments and the apical RPE surface, observed in approximately 15–30% of eyes with severe DMO.
Two critical prognostic imaging biomarkers must be evaluated on baseline SD-OCT:
- Disorganization of Retinal Inner Layers (DRIL): The inability to demarcate the boundaries separating the ganglion cell layer-inner plexiform layer (GCL-IPL) complex, the inner nuclear layer (INL), and the outer plexiform layer (OPL). DRIL is associated with disruption of bipolar, amacrine, and horizontal cell synaptic architecture. Extensive horizontal DRIL () correlates strongly with poor visual recovery despite complete anatomical resolution of fluid.
- Hyperreflective Foci (HRF): Small (), discrete, punctate intraretinal lesions with reflectivity equal to or exceeding the RPE band, lacking posterior optical shadowing. HRF have several possible substrates, including inflammatory cells and lipid. They are not a validated standalone basis for choosing steroid treatment.
Landmark Clinical Trials & Evidence-Based Therapeutics
| Clinical Trial / Protocol | Study Cohort & Comparison Arms | Primary Findings & Clinical Impact |
|---|---|---|
| ETDRS (Early Treatment Diabetic Retinopathy Study) | Evaluated focal/grid argon laser vs. observation for CSMO | Focal laser to leaking microaneurysms and grid laser to appropriate areas of retinal thickening reduced the 3-year risk of moderate visual loss ( letters) by 50% (from 24% down to 12%). Defined standard laser parameters. |
| DRS (Diabetic Retinopathy Study) | Evaluated panretinal photocoagulation (PRP) vs. observation in PDR | Panretinal photocoagulation (1200–1600 burns, spot size) reduced the 2-year incidence of severe visual loss () in high-risk PDR by 50% (from 28% down to 14%). |
| DRCR.net Protocol T | Evaluated intravitreal aflibercept (2.0 mg) vs. ranibizumab (0.3 mg) vs. bevacizumab (1.25 mg) in CI-DMO | Baseline VA 20/40 or better: No significant between-agent difference was demonstrated in that subgroup (~+8 letters at 1 year). Baseline VA 20/50 or worse: Aflibercept was statistically and clinically superior (+18.9 letters) compared to ranibizumab (+14.2 letters) and bevacizumab (+11.8 letters) at 1 year. |
| DRCR.net Protocol S | Evaluated ranibizumab (0.5 mg) vs. PRP for proliferative diabetic retinopathy | Ranibizumab was non-inferior to PRP at 2 and 5 years for visual acuity, with less peripheral visual field loss, reduced rates of DMO, and fewer vitrectomies, though requiring continuous, intensive monitoring and repeated injections. |
| DRCR.net Protocol W | Evaluated aflibercept vs. sham in severe NPDR without CI-DMO | Prophylactic aflibercept significantly reduced the development of PDR and CI-DMO over 4 years, but showed no significant visual acuity benefit over observation followed by prompt treatment upon development of PDR/DMO. |
Intravitreal Corticosteroid Implants
Intravitreal steroids downregulate VEGF gene expression, inhibit leukostasis, stabilize endothelial tight junction claudins, and suppress inflammatory cytokines (IL-6, TNF-):
- Intravitreal steroid implants may be considered for selected DMO, including pseudophakic eyes or insufficient/unsuitable other treatment according to the local label. Cataract and pressure elevation are important risks. A recent cardiovascular event or vitrectomy does not automatically prohibit anti-VEGF or mandate steroids. Consider glaucoma, capsule integrity, migration risk, patient goals and monitoring. Different implants have different release profiles and authorisations.
Indications for Pars Plana Vitrectomy (PPV) in Diabetic Retinopathy
Surgical intervention with 23-, 25-, or 27-gauge pars plana vitrectomy is indicated in advanced diabetic vitreoretinopathy for:
- Non-clearing vitreous haemorrhage: Indicated after 1 to 3 months of observation in type 2 diabetes, or earlier (within 2 to 4 weeks) in type 1 diabetes, bilaterally affected individuals, or monocular patients.
- Tractional Retinal Detachment (TRD): Recent or progressive fibrovascular traction involving or directly threatening the fovea.
- Combined Tractional and Rhegmatogenous Retinal Detachment (TRRD): Caused by fibrovascular contraction tearing the ischaemic, atrophic retina near a vascular arcade.
- Dense, persistent premacular subhyaloid haemorrhage: Trapped retrohyaloid blood over the fovea predisposing to permanent macular epiretinal membrane formation.
- Refractory DMO with vitreomacular traction (VMT) or taut posterior hyaloid face: Relieves anteroposterior and tangential mechanical traction.
Systemic Care and Evidence Boundaries
DCCT in type 1 diabetes and UKPDS in type 2 established the value of systemic glycaemic management; blood-pressure and renal care also influence retinopathy risk. Coordinate control rather than treating the retina in isolation. Rapid metabolic improvement and pregnancy can require closer ophthalmic surveillance. Screening intervals follow disease and the local programme, and symptomatic change requires assessment outside screening.
Protocol T results depend on baseline acuity and time point: aflibercept outperformed both comparators in the worse-vision subgroup at one year, but its two-year advantage over ranibizumab was not statistically significant. Protocol V supports initial observation with rescue criteria in selected centre-involving DMO with good acuity and reliable follow-up. Protocol S supports anti-VEGF as an alternative to PRP in selected PDR, with sustained follow-up critical; missed injections can be dangerous. DRVS supported early vitrectomy in selected severe vitreous haemorrhage, especially type 1 diabetes, but modern timing depends on traction, retinal visibility, vision and other eye status rather than a fixed waiting period.
A 62-year-old patient with center-involving diabetic macular oedema (CI-DMO) in the right eye has a baseline best-corrected visual acuity of 20/60 (Snellen equivalent, 63 ETDRS letters). According to the evidence generated by the landmark DRCR.net Protocol T trial, what comparative visual outcome should guide treatment selection?
All three anti-VEGF agents (aflibercept, ranibizumab, and bevacizumab) provide strictly equivalent visual acuity gains regardless of baseline visual acuity
Bevacizumab demonstrates non-inferiority to aflibercept and ranibizumab while exhibiting a significantly lower rate of systemic thromboembolic events
Focal/grid argon macular laser photocoagulation is superior to all intravitreal anti-VEGF agents when baseline visual acuity is 20/50 or worse
Aflibercept provides statistically and clinically superior visual acuity gains compared to both ranibizumab and bevacizumab at 1 year
Which of the following clinical findings definitively meets the Diabetic Retinopathy Study (DRS) criteria for 'High-Risk Proliferative Diabetic Retinopathy' requiring urgent panretinal photocoagulation?
Neovascularisation of the disc (NVD) measuring 1/3 disc area in the absence of preretinal or vitreous haemorrhage
Neovascularisation elsewhere (NVE) measuring 1/4 disc area in the absence of vitreous haemorrhage
Intraretinal microvascular abnormalities (IRMA) covering 2 disc areas in the mid-periphery
Prominent venous beading in four quadrants accompanied by extensive cotton-wool spots
Sections you finish are checked off in the contents.