Keratoconus Natural Progression: A Systematic Review and Meta-analysis of 11 529 Eyes

Abstract

Purpose: We set out to describe the natural history of keratoconus. We included untreated patients, and our key outcome measures were vision, refraction, and corneal curvature.

Clinical relevance: Keratoconus affects 86 in 100 000 people, causing visual loss due to increasing irregular corneal astigmatism, and the quality of life declines in patients. Interventions are used to stabilize the disease or improve vision, including corneal cross-linking (CXL) and grafting, but these carry risks. Detailed knowledge of the natural history of keratoconus is fundamental in making informed decisions on when their benefits outweigh these risks.

Methods: We included prospective or retrospective studies of pediatric or adult patients who reported 1 or more of visual acuity, refraction, and corneal curvature measures: steep keratometry (K₂), mean keratometry (K_mean), or maximum keratometry (K_max), thinnest pachymetry, corneal transplantation rates, corneal scarring incidence, and patient-reported outcome measures (PROMs). Databases analyzed included Medline, Embase, CENTRAL, and CINAHL. Searches were carried out until October 2018. Bias assessment was carried out using the Joanna Briggs Institute model of evidence-based healthcare.

Results: Our search yielded 3950 publication titles, of which 41 were included in our systematic review and 23 were incorporated into the meta-analysis. Younger patients and those with greater K_max demonstrated more steepening of K_max at 12 months. The meta-analysis for K_max demonstrated a significant increase in K_max of 0.7 diopters (D) at 12 months (95% confidence interval [CI], 0.31-1.14; P = 0.003). Our meta-regression model predicted that patients had 0.8 D less K_max steepening over 12 months for every 10-year increase in age (P = 0.01). Patients were predicted to have 1 D greater K_max steepening for every 5 D of greater baseline K_max (P = 0.003). At 12 months, there was a significant increase in the average K_mean of 0.4 D (95% CI, 0.18-0.65; P = 0.004).

Conclusions: We report the first systematic review and meta-analysis of keratoconus natural history data including 11 529 eyes. Younger patients and those with K_max steeper than 55 D at presentation have a significantly greater risk of progression of keratoconus. Closer follow-up and a lower threshold for cross-linking should be adopted in patients younger than 17 years and steeper than 55 D K_max.