Algorithm of the major and minor diagnostic criteria for active myopic choroidal neovascularization
Paolo Milani,Marco Mazzola,Mario Cigada,Amedeo Massacesi,Marco Setaccioli,Stefania Moschini,Stefano Ciaccia,Fabrizio Scotti,Elena Mantovani,Davide Soranna,Antonella Zambon,Fulvio Bergamini
DOI: https://doi.org/10.1007/s00417-022-05599-5
2022-04-18
Abstract:PurposeTo propose an algorithm of the major and minor diagnostic criteria for macular myopic choroidal neovascularization (mCNV).MethodsThis single-center, retrospective, cross-sectional study was based in Istituto Auxologico Italiano, Milan, Italy. Two authors evaluated the clinical and imaging parameters of eyes with high myopia (spherical equivalent of -6D or less) and suspected to have naïve, recurrent, or inactive mCNV. Recordings of the eyes that met the inclusion criteria were then independently evaluated by two other senior retinal specialists. Fluorescein angiography (FA), spectral domain optical coherence tomography (SD-OCT), and OCT angiography were used for multimodal imaging.ResultsOne-hundred and twenty-two eyes (n = 107; 39 men, 68 women) were included in the study. The mean patient age was 66 years (range, 22–89 years). There were 83 and 39 eyes in the active mCNV and control groups, respectively.The best diagnostic algorithm had positive- and negative-predictive values of 89% and 85%, respectively, and was based on four criteria: leakage/staining on FA, retinal thickening, fuzzy area on SD-OCT, and recent metamorphopsia. When excluding FA-derived findings, retinal pigment epithelium (RPE) features played a diagnostic role in 33 eyes (27%). Twenty-seven eyes with active mCNV (32%) did not have the fuzzy area.Taken singularly, no clinical or imaging parameter had both sensitivity and specificity greater than 78%. Matching of 2 or 3 biomarkers did not yield a sensitivity or specificity greater than 79%. Sensitivities and specificities ≥ 90% were found in ten criteria combinations that included four to five biomarkers. The most frequent were metamorphopsia, fuzzy area, retinal thickening, and leakage. Less frequently, they included hemorrhage, staining, and RPE features such as elevation, flattening, and focal interruption.For all the parameters, the agreement between the investigators was good (Cohen k ≥ 0.66), being the lowest when detecting the ELM interruption within the lesion.ConclusionsA combination of at least four clinical and biological markers yielded the highest positive- and negative-predictive values. More (“major”) and less (“minor”) frequent diagnostic criteria are proposed.
ophthalmology