Performance of Classification Systems for Age-Related Macular Degeneration in the Rotterdam Study

Abstract

Purpose: To compare frequently used classification systems for age-related macular degeneration (AMD) in their abilty to predict late AMD.

Methods: In total, 9066 participants from the population-based Rotterdam Study were followed up for progression of AMD during a study period up to 30 years. AMD lesions were graded on color fundus photographs after confirmation on other image modalities and grouped at baseline according to six classification systems. Late AMD was defined as geographic atrophy or choroidal neovascularization. Incidence rate (IR) and cumulative incidence (CuI) of late AMD were calculated, and Kaplan-Meier plots and area under the operating characteristics curves (AUCs) were constructed.

Results: A total of 186 persons developed incident late AMD during a mean follow-up time of 8.7 years. The AREDS simplified scale showed the highest IR for late AMD at 104 cases/1000 py for ages <75 years. The Rotterdam classification showed the highest IR at 89 cases/1000 py >75 years. The 3-Continent harmonization classification provided the most stable progression. Drusen area >10% ETDRS grid (hazard ratio 30.05, 95% confidence interval [CI] 19.25-46.91) was most prognostic of progression. The highest AUC of late AMD (0.8372, 95% CI: 0.8070-0.8673) was achieved when all AMD features present at baseline were included.

Conclusions: Highest turnover rates from intermediate to late AMD were provided by the AREDS simplified scale and the Rotterdam classification. The 3-Continent harmonization classification showed the most stable progression. All features, especially drusen area, contribute to late AMD prediction.

Translational relevance: Findings will help stakeholders select appropriate classification systems for screening, deep learning algorithms, or trials.

Keywords: AMD; artificial intelligence; classification systems; clinical trials; screening.

Figure 1.

Incidence rates plotted per category in various AMD classification systems. Numbers of incident cases and person years per subclass are shown below the incidence rate plots, with numbers for ages >75 years post-slash.

Figure 2.

Cumulative incidences of late AMD per subclass and classification system at 2, 3, 5, and 10 years. Different colors represent different subclasses within a system. Subjects aged <75 years old are depicted with a dotted line.

Figure 3.

Incidence rates of late AMD per feature. Incident cases and person years are shown below the plots, ages >75 post-slash.

Figure 4.

Cumulative incidence of late AMD per feature determined at baseline. Subjects aged <75 years old are depicted with a dotted line.

Figure 5.

Cumulative hazard plots per AMD baseline feature for ages <75 years, with number at risk per group shown below plots.

Figure 6.

Cumulative hazard plots per AMD baseline feature for ages >75 years, with number at risk per group shown below plots.

Figure 7.

ROC curve of various prediction models of incident late AMD. M1 = age + sex; M2 = age + sex + drusen area ≥10% in the ETDRS grid; M3 = age + sex + pigment changes; M4 = age + sex + drusen size ≥ 125 µm; M5 = age + sex + drusen area ≥10% in the ETDRS grid + pigment changes + drusen ≥ 125 µm.