A Mixed-Methods Analysis of a Novel Low-Cost Simulation Eye for Retinal Laser Training and Assessment

Technical Summary: A Mixed-Methods Analysis of a Novel Low-Cost Simulation Eye for Retinal Laser Training and Assessment

Problem Statement

Retinal laser photocoagulation is a critical intervention for sight-threatening conditions such as diabetic retinopathy (DR), retinopathy of prematurity (ROP), and retinal tears. However, inadequate performance of these procedures can lead to blindness and litigation. While simulation-based training (SBT) is established for surgical skills like cataract extraction, there is a significant gap in low-cost, high-fidelity simulation tools specifically for retinal laser training. Existing models are often expensive, rely on inaccessible 3D printing technology, lack the ability to swap fundus patterns for different pathologies, or have not been validated as pedagogical tools. Furthermore, the global burden of retinal blindness disproportionately affects low- and middle-income countries (LMICs), necessitating affordable, accessible training solutions.

Methodology

The study employed a mixed-methods approach divided into two distinct parts to validate a novel "frugal" retinal laser simulation eye:

1. Expert Validation (Part One)

• Participants: 40 ophthalmic experts (16 consultants, 24 trainees) proficient in panretinal photocoagulation (PRP) and retinopexy.
• Procedure: Experts performed laser procedures on the simulation eye using slit-lamp based lasers (532nm green or 577nm yellow) and binocular indirect ophthalmoscopes.
• Data Collection:
  • Quantitative: Experts recorded laser settings (power, pulse duration, spot size) and rated the simulation's fidelity and utility on a 7-point Likert scale.
  • Qualitative: Free-text feedback was collected regarding realism, laser response, and potential improvements.

2. Trainee Training and Assessment (Part Two)

• Participants: 23 novice ophthalmology trainees with no prior retinal laser experience but proficient in retinal examination.
• Intervention: A three-stage training course:
  • Stage 1: Theoretical presentation on laser safety, settings, and indications.
  • Stage 2: Guided practical exercises on the simulation eye focusing on specific skills (spacing, intensity, grid patterns, titration) followed by two weeks of unsupervised practice. Trainees then performed independent assessments on the simulation eye.
  • Stage 3: Supervised clinical sessions where trainees performed PRP and retinopexy on actual patients under trainer supervision.
• Outcomes: Treatment completion rates (complete vs. incomplete) were recorded for both simulation and patient sessions. Trainees also completed a questionnaire assessing confidence in five domains: knowledge, consenting, patient positioning, setting selection, and treatment delivery.

The Simulation Device
The simulation eye utilizes a "reduced" model of an emmetropic eye with linear dimensions increased by a factor of 1.76. This ensures the field of view, magnification, and anatomical landmark positions match a real eye regardless of the optical device used. The 3D fundus is generated by algorithmically "goring" wide-field flat images, which are printed on matte photographic paper using a standard domestic printer. The assembly uses a low-cost Styrofoam template to mount the eye to a laser slit lamp.

Key Results

• Expert Ratings:
  • Fidelity: The simulation eye received a median score of 6/7 (range 5–7). Experts noted the realistic appearance of the retina, the response to laser fluence (faint blanching at low power progressing to white marks and pigmented burns at higher power), and the representativeness of required fluence settings.
  • Utility: The tool received a median score of 7/7 (range 5–7) as a training and assessment tool.
  • Laser Settings: The mean power and pulse duration used by experts on the simulation (e.g., ~590mW for retinopexy, ~556mW for PRP) were comparable to therapeutic settings used in vivo.
• Trainee Performance:
  • Completion Rates: In the final supervised patient sessions, 91.3% of PRP treatments and 95.6% of retinopexy treatments were deemed complete. This represented an improvement from the simulation-only stage (82.6% for PRP, 91.3% for retinopexy).
  • Confidence: Trainees reported high levels of confidence across all assessed domains after completing the course.
• Feedback: Experts praised the realism but suggested improvements, such as a more flexible mount to simulate eye movement and the addition of a cornea to allow for contact lens training and scleral indentation.

Key Contributions

1. Development of a Validated Frugal Tool: The paper introduces a low-cost simulation eye that achieves high fidelity without 3D printing or expensive manufacturing, utilizing standard printers and optical scaling.
2. Modular Training Approach: The study demonstrates a "staggered" training methodology where isolated skills (accuracy, spacing, intensity) are refined on the simulator before combining them into full treatments, eventually transitioning to supervised patient care.
3. Adaptability: The system allows for the easy exchange of fundus images, enabling the simulation of various pathologies (e.g., ROP patterns, retinal tears) and abstract training grids.
4. Global Applicability: By addressing the cost barrier, the tool is positioned to support training in both high-resource and low-resource settings, aiming to improve the standard of care for retinal laser procedures globally.

Significance and Claims

The authors claim that this study validates the new simulation eye as a high-fidelity, adaptable, and affordable tool for ophthalmic education. The significance lies in its ability to address unmet needs in retinal laser training by:

• Providing a safe environment for acquiring and refining skills before patient contact.
• Enabling structured, competency-based assessment and credentialing.
• Potentially reducing the hesitation to administer correct, evidence-based laser dosing by familiarizing trainees with the required intensity and coverage.

The paper concludes that while the model does not yet support contact lens or scleral indentation training (due to the lack of a cornea and malleability), its high utility scores and successful implementation in a staged training program suggest it is a viable solution for improving global retinal laser outcomes. The authors emphasize that simulation offers a critical pathway to lowering the probability of error in patient treatment, particularly in the context of the growing epidemic of retinal blindness.