Statistics from Altmetric.com
Editor,—I read with interest the paper by Weinberger et al.1 In this pilot study about ICGA guided photocoagulation of occult choroidal neovascularisation (CNV) in AMD, the authors provide evidence for a beneficial effect on visual prognosis by treating this CNV pattern. However, some issues can be raised about both inclusion criteria and patients' selection and then about results.
On ICGA, all eyes included in the study show a choroidal neovascular network, with CNV size smaller than four disc areas; the authors do not specify how many hot spots, plaques, or mixed lesions are in their sample. Indications for treatment, visual prognosis, and recurrence rate in these three CNV morphological types are quite different.2
Furthermore, a marked disproportion between eyes with pigment epithelial detachment (PED) (two cases) and those without PED (the remaining 19), not reflecting the data provided by Guyer and colleagues in 1000 consecutive eyes,3 characterises the examined population. The authors present the final anatomical and visual outcomes by considering all eyes as a single group; this method is questionable, since vascularised PED and RPE are definitely two distinct entities. Occult CNV with PED has a higher frequency of recurrence, probably due to the greater exudative activity of primary CNV4 and, even if anatomical outcome of laser photocoagulation is satisfactory, the functional result is usually poor.5 Then the encouraging final visual acuity reported in the paper is probably biased by anomalous sample composition and by improper grouping.
In order to draw definite conclusions and provide guidelines about ICGA guided laser treatment of occult CNV, there is a clear need for a randomised prospective, controlled clinical trial, with a larger population and a more realistic proportion between occult CNV with and without PED, and presenting separate final results for the two patterns, with regard to both anatomical and visual variables.
Editor,—We thank Dr Da Pozzo for his interest in our paper. He raised a number of interesting points.
Patient selection for ICG guided laser photocoagulation is extremely crucial. Functional results from different pilot studies on ICG guided laser photocoagulation show various outcomes. This may be explained either by the patient selection or by the indications for ICG guided treatment. Especially, the definition of the choroidal neovascular network in ICG angiograms is crucial since the interpretation of ICG angiograms is still under discussion.
Our interpretation of ICG angiograms for the detection of a CNV is based on the choroidal transit and recirculation phase of ICG dye recorded with a scanning laser ophthalmoscope. We consider this to be more accurate in determining the size, location, and geometry of CNV than the late phase. We have demonstrated that occult CNV defined by the MPS standards could be converted into visible neovascular membranes in up to 50% of cases independent of the presence of PED. Using other imaging techniques, hot spots and plaque hyperfluorescence were used to convert occult CNV into visible CNV. Previous studies on ICG guided laser photocoagulation rely almost exclusively on this interpretation of ICG angiograms. However, it has recently been demonstrated that many eyes with hot spots in late ICG angiograms have polypoidal choroidal vasculopathy (PCV). Since previous studies did not differentiate between eyes with PCV and AMD, a quite significant proportion of eyes included in these studies may have had PCV instead of AMD (Slakter 1999, personal communication). This may have influenced their findings, since it is important to differentiate AMD from PCV because there are significant differences in the demographic risk profile, natural course, visual prognosis, and management of these patients.
Our study involved 175 consecutive patients undergoing ICG angiography for occult CNV secondary to AMD. We performed ICG guided laser photocoagulation only in eyes with occult CNV that demonstrated a visible extrafoveal or juxtafoveal neovascular network in the early ICG angiogram. In these eyes no hot spots or plaque hyperfluorescence were detected in the late angiograms. Only two eyes with PED were included in the study according to the inclusion criteria. This small number allowed no subgroup analysis. However, by analysing the 19 eyes without PED the results are even more promising.
We are aware that the eyes included in our pilot study may represent a special subgroup of eyes with occult CNV as pointed out in the original manuscript. Additionally, this is well demonstrated by the small number of patients included in the study. However, reviewing our quite encouraging final visual results we feel very comfortable with ICG guided laser photocoagulation following our interpretation of ICG angiograms in occult CNV secondary to AMD.
We totally agree that there is a clear need for a randomised prospective, controlled clinical trial to prove the efficacy of ICG guided laser photocoagulation for occult CNV. Based on our promising results we would suggest following our approach for imaging and interpretation of ICG angiograms for this study.