First Author: M.Gkika UK
Co Author(s): N. Vallabh C. Parkes G. Hebbar J. Sahni 0 0 0 0 0 0 0 0 0
Back to previous
To report the near infrared reflectance (NIR-R), near infrared autofluorescence (NIR-AF) and blue-wave autofluorescence (BW-AF) appearance of choroidal naevi using confocal scanning laser ophthalmoscope (cSLO) (Heidelberg Spectralis®, HRA+OCT model) and to correlate them to colour fundus photos (CF) (Topcon TRC.50EX).
St. Paul’s Eye Unit, Royal Liverpool and Broadgreen University Hospitals NHS Trust, UK
Consecutive patients diagnosed with choroidal naevus on fundoscopy underwent CF, optical coherence tomography (OCT) and cSLO (NIR-R, NIR-AF and BW-AF). cSLO images were compared to CF to report the variability in imaging characteristics of the naevi and their possible relation to pigmentation, reflectance, area and margins of the lesion. On CF naevi were graded for pigmentation as dark, moderate or light pigmented based on 2 standard photos selected from the cohort and for demarcation as well-defined if a clear transition between the pigmented area and the normal choroid could be identified for 90% of the circumference of the lesion. Intensity of reflectance of the naevus on NIR-R images was defined as hyper if the reflectance was greater than background and hypo if it was less. Hyper-reflectance was further graded as good or poor based on a standard photo selected from the cohort (good, if ≥ to hyper-reflectance on standard photo, poor if <). If the naevus was not visible, it was deemed to exhibit isoreflectance. On NIR-R, the edge of the naevus was graded as well defined in the same manner as on CF. NIR-AF and BW-AF images were graded and compared to CF in the same way as NIR-R.
42 naevi of 39 patients were imaged. On CF, 21 naevi were graded as dark, 17 as moderate and 4 as light pigmented, while 29 were judged as well-defined. When compared to CF, NIR-R detected naevi in 97.6% of images, as opposed to 31.7% with the BW-AF. On NIR-R 73.8% naevi demonstrated good hyper-reflectance, 19% poor hyper-reflectance, 4.7% were hyporeflective and 2.4% were iso-reflective. NIR-AF images could be only obtained in 22 out of 42 patients and in those naevi were visible in 90.9%. When compared to CF, all naevi graded as darkly pigmented on CF showed hyper-reflectance (71.4% good; 28.5% poor); 76.4% graded as moderately pigmented showed good hyper-reflectance, 1.1% showed poor hyper-reflectance, 0.05% showed hypo-reflectance and 0.05% showed iso-reflectance; 75% of graded as light pigmented showed good hyper-reflectance and 25% (1 naevus) showed hypo-reflectance. BW-AF identified drusen and lipofuscin in 22%. None of the cSLO images showed a larger area of naevus compared to CF.
Variations in NIR-R and NIR-AF of naevi did not appear to correspond to the pigmentation of the lesion on CF. There was also no correspondence between NIR-R and BW-AF. The difference observed between NIR images and those in the visible spectrum (ie CF) could be related to the type of melanin within the lesion. Analysis of the absorbance spectra in dermatology has shown that the levels of eumelanin and pheomelanin increase and decrease, respectively, from dysplastic nevi to invasive melanomas. Diffuse fundus reflectance spectroscopy (NIR-R and NIR-AF) may provide similar insight into the in vivo compostion of the melanin within choroidal naevi and appears to be a promising technique to develop better strategies for the characterization of these melanocytic lesions.