A recent Ophthalmology digest,
as sent to subscribers

Study Type: In vivo experimental animal study (conditional knockout mouse model)

Key Question: Does targeted deletion of Zeb1 in the corneal endothelium inhibit injury-induced endothelial-mesenchymal transition (EnMT) and retrocorneal membrane (RCM) formation?

Key Findings:

• Conditional Zeb1 knockout (intracameral 4-OHT) significantly attenuated surgical injury-induced corneal oedema: central corneal thickness 106.8 ± 11.1 µm (injury + 4-OHT) vs 182.2 ± 15.5 µm (injury alone), p <0.00001
• Zeb1 targeting suppressed FGF2- and injury-induced upregulation of EnMT markers (COL1, fibronectin, vimentin) and preserved E-cadherin expression
• Conditional Zeb1 deletion inhibited injury-dependent RCM formation in vivo

Clinical Relevance: ZEB1 represents a potential therapeutic target for anterior segment fibrosis and corneal blindness, offering a non-transplant strategy relevant to a NHS corneal graft waiting list context.

Limitations: Findings are limited to a murine model; translation to human corneal endothelial biology requires validation.

Study Type: Prospective cohort study

Key Question: Does 24-hour ambulatory blood pressure predict rates of RNFL loss in primary open-angle glaucoma?

Key Findings:

• Every 10 mmHg reduction in 24-hour minimum MAP, SBP, and DBP was associated with accelerated RNFL loss of −0.54, −0.36, and −0.46 µm/year respectively (all P<0.01), after multivariate adjustment
• Eyes in the lowest quartile of 24-hour MAP (81–90 mmHg) progressed ~0.68 µm/year faster than the highest quartile (P=0.017)
• Minimum nocturnal BP parameters were the strongest predictors, highlighting the relevance of overnight dipping

Clinical Relevance: For UK glaucoma services, this supports consideration of 24-hour ABPM in patients with unexplained progression, particularly those on antihypertensive therapy with potential nocturnal hypotension.

Limitations: Small single-centre sample (42 subjects) limits generalisability across ethnically and clinically diverse NHS glaucoma populations.

Study Type: Systematic review and meta-analysis (observational studies)

Key Question: Is statin use associated with an increased risk of dry eye disease (DED)?

Key Findings:

• Pooled analysis of 6 observational studies (n=560,821) demonstrated a statistically significant association between statin use and DED (OR 1.09, 95% CI 1.05–1.13; p<0.001)
• Absolute risk difference of 10.2 additional DED cases per 1,000 statin users (95% CI 5.5–15.2)
• Evidence rated as **very low certainty** (GRADE) due to serious inconsistency and observational study design limitations; subgroup analyses showed consistent directional association

Clinical Relevance: Statins are widely prescribed in UK primary and secondary care; clinicians should consider counselling patients with pre-existing DED risk factors about potential ocular surface symptoms when initiating statin therapy.

Limitations: All included studies were observational, and very low GRADE certainty limits causal inference; statin dose, type, and duration were not analysable.

Study Type: Retrospective cohort study (nationwide, registry-based)

Key Question: Does anti-VEGF retreatment after a period of loss to follow-up (LTFU) produce meaningful visual and anatomical gains in DME patients who return with stabilised vision?

Key Findings:

• Only 32.7% of patients returning with stabilised VA achieved visual gain on retreatment, versus 62.0% of those returning with vision decline (p<0.001)
• Anatomical response (≥10% CFT reduction) was also lower in the stabilised group (56.4% vs 63.2%; p=0.038)
• Patients with stabilised VA ≥20/50 at return showed no meaningful visual gain despite anatomical improvement, regardless of retreatment intensity

Clinical Relevance: For UK DME services managing patients re-presenting after LTFU with preserved vision, this challenges reflexive anti-VEGF reinitiation and supports considering alternative or combination therapies (e.g., faricimab, steroid implants) in this subgroup.

Limitations: Retrospective registry design limits control for confounders, including reasons for LTFU and treatment adherence history.

Study Type: Retrospective cohort study (target trial emulation) using US EHR data

Key Question: Does levodopa ± carbidopa or DRD2 agonist use reduce the risk of conversion from non-neovascular AMD to neovascular AMD?

Key Findings:

• Levodopa ± carbidopa was associated with a ~31–33% reduced 3-year risk of nAMD conversion vs both active comparators (vs pantoprazole: HR 0.67, 95% CI 0.45–0.98; vs gabapentin: HR 0.69, 95% CI 0.50–0.95)
• DRD2 agonists showed no statistically significant reduction in conversion risk against either comparator
• Findings suggest dopaminergic signalling may differentially influence AMD progression depending on mechanism of action

Clinical Relevance: If replicated prospectively, levodopa use could represent a repurposable therapeutic strategy to delay nAMD conversion in patients already receiving anti-Parkinsonian therapy — relevant to shared AMD/neurology caseloads in NHS practice.

Limitations: Retrospective US EHR data limits causal inference; residual confounding by indication (e.g., Parkinson's disease severity) cannot be excluded despite propensity matching.

Study Type: Retrospective cohort study (propensity score-matched, multi-system US electronic health records)

Key Question: Are routine vaccinations (COVID-19, HPV, varicella, live and recombinant herpes zoster) associated with altered risk of new-onset idiopathic uveitis?

Key Findings:

• All five vaccines were associated with significantly reduced 12-month risk of new-onset idiopathic uveitis versus matched unvaccinated controls
• Largest risk reduction with varicella vaccine (RR 0.29; 95% CI 0.25–0.33); COVID-19 vaccine also substantially protective (RR 0.35; 95% CI 0.33–0.37)
• Protective associations persisted after excluding patients with prior diagnosis of the relevant viral infection, suggesting an effect beyond simply preventing infection-triggered uveitis

Clinical Relevance: Challenges the assumption that vaccination is a meaningful trigger for uveitis; relevant to counselling patients with uveitis history who express vaccine hesitancy.

Limitations: Retrospective observational design limits causal inference; residual confounding likely despite propensity-score matching.

Study Type: Retrospective comparative cohort study (propensity-score matched)

Key Question: Do PCSK9 inhibitors reduce the risk of retinal vascular occlusion compared to other lipid-lowering therapies in hyperlipidaemic patients?

Key Findings:

• PCSK9 inhibitors were associated with significantly lower risk of retinal vascular occlusion at 3 years (RR 0.56), 5 years (RR 0.50), and 7 years (RR 0.46; 95% CI 0.35–0.61)
• Risk reduction extended to both RVO (7-year RR 0.47) and RAO (7-year RR 0.47) individually
• Effect size strengthened over time, suggesting a cumulative protective benefit

Clinical Relevance: As PCSK9 inhibitors become more accessible on the NHS for high-risk lipid patients, these findings raise the possibility of an additional ocular benefit worth considering in shared decision-making for at-risk patients.

Limitations: US-based retrospective EHR data limits generalisability to NHS populations, and residual confounding cannot be excluded despite propensity-score matching.

Study Type: Retrospective comparative cross-sectional study

Key Question: Does CDK4/6 inhibitor therapy cause corneal epithelial changes independently of aromatase inhibitor use or tear film dysfunction?

Key Findings:

• Punctate epitheliopathy (PE) affected 44.4% of CDK4/6 inhibitor + aromatase inhibitor (CDKAI) eyes vs 4.7% (AI alone) and 2.3% (controls); all moderate and severe cases occurred exclusively in the CDKAI group (χ²=34.31; p<0.001)
• Vortex keratopathy was present in 13.3% of CDKAI patients and absent in both comparator groups (p=0.025)
• Tear film parameters (Schirmer, TBUT, OSDI) did not differ between groups, suggesting a direct cytostatic mechanism rather than dry eye

Clinical Relevance: As CDK4/6 inhibitors (ribociclib, palbociclib, abemaciclib) are increasingly used in UK oncology practice, ophthalmologists should anticipate corneal referrals and consider proactive fluorescein slit-lamp examination in this patient group.

Limitations: Retrospective design with incomplete OSDI response rates in the CDKAI group limits assessment of true symptom burden.