Natural history and predictors for progression in pediatric keratoconus

Abstract

We studied the demographic and clinical predictors associated with keratoconus progression in a pediatric population. Retrospective cohort study. We evaluated 305 eyes without previous surgeries from 168 patients, 9 to < 18 years old, and with a minimum 36-month follow-up in a hospital corneal ambulatory. We used Kaplan-Meyer survival curves; the dependent variable (main outcome measure) was the interval time (months) until the event, defined as an increase of 1.5 D in the maximum keratometry (Kmax), obtained with Pentacam. We evaluated the predictors: age (< or ≥ 14 years), sex, keratoconus familial history, allergy medical history, and the baseline tomographic parameters: mean keratometry (Km), Kmax (< or ≥ 55 D); and thinnest pachymetry (TP). We used log-rank tests and compared median survival times for right (RE)/left eyes (LE) and better (BE)/worse eyes (WE). A p value < 0.05 was considered significant. The patients' mean ± SD age was 15.1 ± 2.3 years old; 67% were boys, 30% were < 14 years, 15% had keratoconus familial history, and 70% were allergic. The general Kaplan-Meyer curves showed no differences between RE/LE or BE/WE. RE with allergy and LE with Kmax ≥ 55 D had smaller survival times ((95%CI 9.67-32.1), p 0.031 and (95%CI 10.1-44.1), p 0.042, respectively). For BE and WE, Kmax ≥ 55 D had smaller survival times ((95% CI 6.42- ), p 0.031 and (95%CI 8.75-31.8), p 0.043, respectively). Keratoconus progression was similar between RE/LE and BE/WE. Steepest corneas are predictors of faster progression. Allergy is also a predictor of keratoconus progression in RE.

Figure 1

Examples of patients who were or were not included in this survival analysis. Patient 1: analyzed from time points A (T0) to B; (event or censoring) Patient 2: censored after the first appointment; Patient 3: excluded due to not having the possibility of being evaluated during a minimum period of at least 36 months; Patient 4: analyzed from time points C (T0) to D (censoring); Patients 5 and 6: included.

Figure 2

Pediatric keratoconus: Distribution of the eyes, showing the reasons for exclusion and the final number included.

Figure 3

Kaplan Meyer empirical curves of time until progression, in months. (a) Right and left eyes, overall progression (Right Eyes (n = 156); 88 events; Median of 21.9 CI95% (14.5; 39.7). Left Eyes (n = 149); 76 events; Median of 22.4 CI95% (13.4; 50.0). p value = 0.76); (b) Right eyes, by the presence of allergy (p value 0.032); and (c). Left eyes, by Kmax severity (p value = 0.043).

Figure 4

Better and worse eyes: Kaplan Meyer empirical curves of time until progression, in months. The eye with the highest Km between both eyes was considered the worse eye. Better (n = 136); 72 events; Median of 24.9 CI95% (15.5, 61.1). Worse (n = 136); 77 events; Median of 19.0; CI95% (11.0, 31.8). p value = 0.301.

Figure 5

Better and worse eyes: Kaplan Meyer empirical curves of time until progression, in months, by sex, age, familial history of keratoconus, and the presence of allergy.

Figure 6

Better and worse eyes: Kaplan Meyer empirical curves of time until progression, in months, by Km, (divided into four groups), by Kmax severity (< or ≥ 55D), and by thinnest pachymetry (TP) (divided into four groups).