Mohammad AskarPublic working copy · medical review pending

Treating macular oedema · 178

Macular oedema: sustained treatment with proportionate burden

Start with the cause, central vision and OCT, then agree a treatment and monitoring schedule the patient can attend.

Public working copy · medical review pending. Source, translation and medical checks remain open. Read the limitations in the text and references. These educational materials do not replace individual medical assessment. Published: 2026-11-18T19:45:52+01:00. Internal project references lead to this notice. They are not publicly accessible primary sources; the source limitations remain stated in the text.

DME and RVO share VEGF-mediated leakage but have different evidence, labels and courses. Trial interval extension is not a guarantee for every eye or proof of equal rare-event safety.

Terminology and draft status

Terminology review is not complete. No new WHO UMD entry verification is claimed for this copy; pending entries and limits are recorded in terminology.md.

01Decide whether central vision needs treatment

Diabetic macular oedema (DME) requires assessment of centre involvement, acuity, OCT and diabetic retinopathy stage. Anti-VEGF is a principal option for vision-impairing centre-involving oedema; centre involvement alone does not mandate immediate injections in every eye with good acuity. Selected observation with rescue criteria can be appropriate, whereas non-central oedema may be considered for focal/grid laser according to its pattern. 1

The source cites German BVA/RG/DOG 2019 advice for 6 monthly initial DME injections and 3 initial RVO injections, with another 3 when relevant oedema persists. These are historical cited schedules, not independently verified current universal requirements. Product-specific loading, response, vision and the current local label must guide the actual plan. 1

02Make PRN and treat-and-extend explicit

Pro re nata (PRN) means monitoring with treatment when activity warrants it; a monitoring visit need not contain an injection. Treat-and-extend (T&E) usually treats at each planned treatment visit and lengthens or shortens the next interval using specified activity criteria. The original 2- or 4-week steps are protocol examples, not interchangeable rules. 1

Choose a feasible schedule with bilateral disease, transport, other illness and clinic capacity in mind. The original preference for T&E in unilateral disease is an expert observation, not evidence of unique superiority. Fewer injections and fewer monitoring visits are different outcomes, and extending cannot replace checking for activity. 1

03Ask why trial results are not reached in practice

Routine-care visual gains may be smaller than trial gains, but injection count alone does not prove the reason: disease severity, selection, follow-up loss, capacity and access can confound comparison. The original Freiburg estimate of 44% insufficient adherence in the first year lacks a verified study population and definition here. 1

Original survey figures—more than 50% unable to relax, 25% sleep disturbance and 42% preferring fewer injections with unchanged results—describe unverified questionnaires, not all patients. Explain the procedure, discuss anxiety, arrange appointments and address barriers without blaming patients or promising the same result with fewer visits. 1

04Compare drugs within the trial actually performed

KESTREL/KITE studied brolucizumab in DME. Longer intervals and similar mean visual outcomes under those protocols do not remove the risk of intraocular inflammation, retinal vasculitis or vascular occlusion. The original two Heidelberg cases are unverified anecdotes without a denominator. Examine for inflammation before injection and urgently assess new symptoms; do not inject into active intraocular inflammation. 1

YOSEMITE/RHINE compared faricimab fixed dosing or personalised intervals with fixed aflibercept 2 mg dosing in DME. Their non-inferiority results concern the specified visual endpoint and comparator; a comparator unable to extend does not answer flexible T&E versus flexible T&E. The cited German supplementary statement still needs its complete source/version checked. 1

05Read PHOTON with its denominators and safety limits

PHOTON compared aflibercept 8 mg at assigned 12- or 16-week intervals with 2 mg every 8 weeks after protocol loading, with adjustment criteria. Its 48-week primary endpoint and 96-week extension are distinct; the original description as exclusively treatment-naïve and its early-2024 German availability statement need separate eligibility and local regulatory checks. 2 3

PHOTON is a separate randomised DME trial: 660 randomised, 658 treated. At week 48, 8q12/8q16/2q8 gains were 8.8/7.9/9.2 letters; adjusted differences −0.57 (95% CI −2.26 to 1.13) and −1.44 (−3.27 to 0.39) met a 4-letter margin. At week 96, 534 completed, gains were 8.8/7.5/8.4 and completer injections 9.5/7.8/13.8. This Regeneron/Bayer-funded trial does not transfer its DME outcome or selected extensions to every nAMD/RVO eye. 2 3

The original nearly 90% maintaining ≥12 weeks and 44% reaching ≥20 weeks by week 96 need the exact arm, denominator and last-assigned versus continuously maintained definition. They are not a promise for every treated patient. Similar adverse-event rates in a trial do not establish identical rare-event safety or absence of transient volume-related IOP effects; the original ≥35 mmHg comparison needs the complete safety table. 2 3

Read PHOTON with its denominators and safety limits
Outcome2 mg q8 comparator8 mg q12/q16 and limit
BCVAFixed comparator after loading48-week non-inferiority; 96-week gains are separate, not superiority
CRTAnatomical reductionReduction does not itself establish visual benefit or equal efficacy
Year-1 injectionsOriginal about 8Original about 5 q16; source denominator needs verification
Through week 96Completer mean 13.8Completer means 9.5 q12/7.8 q16; original about 6 fewer is rounded, not every patient
Interval durabilityFixed 8-week comparatorOriginal nearly 90% ≥12 weeks needs exact arm and maintained/assigned definition
Extension by week 96No comparable flexible extensionOriginal 44% ≥20 weeks needs denominator; not a guaranteed routine schedule

06Plan RVO follow-up and possible cessation cautiously

RVO is a common retinal vascular cause of visual loss; original Germany 300000 and Birmingham about one-third developing oedema require dated populations. BRVO is more common and CRVO often more severe. Treat vision-impairing oedema promptly while also examining iris/angle neovascularisation and systemic vascular risk. An acute onset does not remove chronic risk or ensure cure. 1

The source reports CRYSTAL median 15 injections in 24 months and a retrospective 5-year cohort in which 60% received ≥30. These source-specific counts remain unverified and cannot prescribe an individual lifetime load. Loss-to-follow-up series over 6 months may show visual loss despite anatomical improvement after return; selection and untreated progression prevent a universal claim of irreversible loss in every interruption. 1

The original Fight Retinal Blindness! summary gives +10 letters at 3 years, 44% gaining ≥15, 69% monotherapy and 19% switching. Registry comparisons and AURIGA switch outcomes are observational: similar gains with fewer injections do not prove that switching caused benefit, that the drugs are interchangeable or that treatment intensity alone explains outcome. Full cohort definitions and missingness remain open. 1

Early OCT improvement and good acuity around 6 months may be prognostic, not validated stopping rules. Some RVO eyes can cease injections after sustained stability with monitoring, whereas others relapse or need years of treatment. Do not inject every venous-stasis eye without relevant oedema. QUASAR is described as ongoing in the historical export; a 2026 design publication was located, but no fresh outcome or current German RVO 8 mg label is inferred from it. 1

What are the main learning points?

DME and RVO share VEGF-mediated leakage but have different evidence, labels and courses. Trial interval extension is not a guarantee for every eye or proof of equal rare-event safety.

When is assessment urgent?

New pain, redness, photophobia or marked visual loss after injection needs immediate assessment for infection or inflammation. New retinal arterial symptoms require the emergency stroke pathway.

References and access status

  1. 1
    Immutable original export; DME/RVO treatment-burden lecture

    Complete export read; original attribution and any recording/transcript or appointment not independently verified.

  2. 2
    PHOTON 48 weeks, 2024

    Historical review 138/139 primary-abstract appraisal of 15 November 2026 reused; not newly read, full report/protocol/figures unread.

  3. 3
    PHOTON 96 weeks, 2026

    Historical review 138/139 primary-abstract appraisal of 15 November 2026 reused; not newly read, full report/protocol/figures unread.