Using Inter-Session Reliability To Estimate The Minimal Detectable Change For Indices Of Progression In Keratoconus Using Scheimpflug Tomography

Previous studies have used intra-session reliability to estimate measurement error when using scheimpflug tomography to assess keratoconic progression. This fails to account for day-to-day variation in the patient, technician, machine and testing environment. This study aimed to determine inter-session repeatability of several markers of keratoconic progression measured by scheimpflug tomography.

Ophthalmology Department, University Hospital Southampton NHS Foundation Trust, UK

A retrospective study of all keratoconus patients who had a Pentacam® scan performed between 2014 and 2020 at a single tertiary unit with two separate scans performed > 1 day and ≤ 14 days apart. Patients with clinically explainable reason for potential change were excluded. Minimal detectable change (MDC) was calculated for several parameters, including K_max, topography-derived astigmatism (Cyl), corneal thickness at both the apex (CT_apex) and thinnest location (CT_min), as well as anterior and posterior K_mean. MDC is the minimum amount of change needed to for a true change to be considered more likely than that of random measurement error.

45 patients with keratoconus were included in this study (Median age 30; 68.9% Males). Median time between the scans was 13 days. The MDC was found to be 24.66 microns in CT_apex and 28.92 microns in CT_min, 2.68D in Cyl, 2.99D in K_max.

The degree of measurement error estimated was higher than that previously reported by others using intra-session reliability. This likely represents additional noise introduced by day-to-day variation and real-world daily clinical practice. Basing clinical decisions on changes in these markers of progression must be done within the context of this noise.