Allied health professional-led keratoconus service shows safe outcomes in UK tertiary centre

A retrospective study published in Eye evaluates the clinical outcomes of an allied health professional (AHP)-led keratoconus service at Queen's Medical Centre, Nottingham, a single tertiary ophthalmic unit in the United Kingdom, covering the full patient pathway from diagnosis through collagen cross-linking (CXL) and post-operative follow-up.¹ The study describes a service model in which optometrists perform routine assessment and monitoring while nurse practitioners (NPs) perform CXL, with consultant oversight maintained throughout.

Background

Keratoconus is the most common bilateral corneal ectatic disorder and predominantly affects young adults, with onset typically during adolescence and gradual progression marked by increasing irregular astigmatism and deteriorating vision.¹ CXL is currently the only intervention with evidence supporting stabilisation of disease progression, acting by strengthening the anterior corneal stroma through riboflavin activated by UVA light.¹ The demand for monitoring and CXL creates significant pressure on corneal subspecialist clinics in the National Health Service (NHS), with ophthalmology comprising 10% of all UK outpatient referrals.¹ The authors note that approximately half of corneal services in the UK already employ optometrists in extended roles.¹

Study design and service model

The study was registered as a service evaluation and quality improvement project. Participants were identified from patients referred to the AHP-led keratoconus clinic at Queen's Medical Centre. Keratoconus was confirmed by slit-lamp examination and corneal topography. Exclusion criteria included other corneal pathologies such as scars and patients who had previously undergone CXL.

The service operated with a single optometrist experienced in corneal pathology assessment. Each clinic session offered 6 appointments alongside a consultant-led corneal clinic. The optometrist performed ophthalmic history, slit-lamp examination, subjective refraction and corneal topography, then discussed the management plan with a corneal consultant before implementation. The service protocol was developed by 2 corneal consultants and approved through local governance procedures.

CXL was performed using an accelerated epi-off protocol: central corneal epithelial debridement with 20% alcohol, riboflavin 0.1% applied every 2 minutes for 10 minutes and UVA exposure at 7.20 J/cm² over 12 minutes. Postoperative follow-up occurred at 1 week, then at 1, 3, 6, 12 and 24 months. Corneal topography was measured using the Oculus Pentacam AXL. Statistical analysis used Mann-Whitney U tests and chi-squared tests.

Results

A total of 271 eyes from 177 patients were included. Of these, 128 eyes (92 patients) underwent CXL and 143 eyes (85 patients) were monitored without CXL. The median age at presentation was 24.50 years in the CXL group and 30.00 years in the monitored group.

Of the 128 CXL eyes, 82 (64.1%) were performed by an NP and 42 (32.8%) by an ophthalmologist; 4 eyes were excluded from this comparison due to unidentifiable surgeon records. Best-corrected visual acuity (BCVA) significantly improved at 24 months in patients treated by an NP (P = .034, Mann-Whitney U test). K1 and K2 were reduced at 12 months in NP-treated eyes (P = .040 and P = .015, respectively) but this reduction was not sustained at 24 months. Anterior higher-order aberrations were significantly elevated at 12 months (P = .042) but this elevation had resolved and was no longer statistically significant by 24 months.

Eyes treated by ophthalmologists showed no statistically significant change in BCVA at 12 or 24 months. No significant differences in BCVA or corneal topography were found when comparing CXL performed by NPs versus ophthalmologists (P > .05 for all, Mann-Whitney U test). The authors note that more advanced disease and a greater number of different surgeons in the ophthalmologist group may have contributed to this difference.

Stromal haze developed in 3 of 82 NP-performed CXL eyes (3.7%), with 1 eye demonstrating persistent haze at 24 months. In the ophthalmologist group, 6 of 42 eyes (14.3%) developed stromal haze, with 2 eyes having persistent haze at 24 months. The difference in initial haze incidence was statistically significant (P < .05, chi-squared test). The authors note that selection bias—with more advanced cases directed to ophthalmologist-led CXL—may have contributed to this finding. In all cases of haze, prompt identification by the optometrist led to timely escalation and treatment with additional topical steroids.

In the monitoring group, 6 eyes (4.1%) demonstrated topographic progression and were listed for CXL. Visual acuity improved during the monitoring period in the remaining patients (P = .035). No significant changes in corneal topography or subjective refraction were observed over the monitoring period.

Limitations and conclusions

The authors identified several limitations, including non-uniform follow-up intervals due to individualised patient risk assessment, potential selection bias from directing more advanced cases to consultant-led clinics and the impact of the COVID-19 pandemic on appointment attendance and continuity of follow-up. The retrospective single-centre design also limits broader generalisability.

The authors describe this as the first study to analyse outcomes in patients monitored primarily by optometrists following CXL and the first to report outcomes of CXL performed by NPs.¹ The authors conclude that the service model produced acceptable clinical outcomes and offers a practical approach to augmenting specialist corneal services within the NHS, contingent on appropriate staff training, governance oversight and defined pathways for escalation to consultant ophthalmologists.

Reference

1. Boddy LE, Woo S, Murch P, Said D, Dua HS. Evaluation of an allied health professionals-led keratoconus service in a tertiary UK centre. Eye (Lond). doi:10.1038/s41433-026-04482-5