Corneal grading system for postoperative assessment of manual small incision cataract surgery (MSICS)

G. Strauss, Richard Babington Newsom, J. Surka

Research output: Contribution to journalArticlepeer-review


Background: The post-operative assessment of corneal clarity is widely used to assess the quality of cataract surgical techniques. This is best done with a slit lamp, optical coherence topography or endothelial photography. In settings where a slit lamp is not available, the assessment of corneal oedema still plays an important part of post-operative evaluation, particularly when teaching surgery.

Methods: We developed a system of corneal assessment, which could be used in the absence of a slit lamp to help in the evaluation of patients following manual small incision cataract surgery (MSICS). A hand-held white light source was used to grade corneal clarity from 0–4.

Results: A total of 74 patients were observed day 1 postsurgery. There were 49 women and 25 men, with an average age of 70.8 years (SD 10.7, 95% CI 2.4). Median pre-operative visual acuity was perception of light, improving to 6/36 one day postsurgery. Patients had an average corneal clarity grade of 1.04 (SD 1.02, 95% CI 0.23). Twenty-eight patients had no oedema, 30 had grade 1, nine had grade 2, seven had grade 3 and one had grade 4.

The observers agreed in 81% of readings, with a kappa linear weighting score of 0.85 (95% CI 0.75–0.92).

Conclusion: A rapid method of assessment of corneal clarity was found to be a reliable and consistent tool for the postoperative assessment of cataract surgery. The inter-observer agreement was high and could be used to monitor the quality of cataract surgery in a learning environment.
Original languageEnglish
Pages (from-to)22-25
JournalSA Ophthalmology Journal
Issue number2
Early online date1 Jun 2020
Publication statusPublished - Sept 2020


Dive into the research topics of 'Corneal grading system for postoperative assessment of manual small incision cataract surgery (MSICS)'. Together they form a unique fingerprint.

Cite this