Age related macular degeneration (AMD) is the major cause of irreversible blindness in elderly patients worldwide. The pathogenesis of this disease has been investigated from genetic,1histological,2 and haemodynamic perspectives.3-6 AMD is classified as non-exudative—the dry type, or exudative neovascular—the wet type.7 8Choroidal neovascularisation (CNV) in the macula causes severe visual impairment in AMD. Several angiogenic factors (vascular endothelial growth factor (VEGF) and basic fibroblast growth factor), which are induced by hypoxia and ischaemia, may play a part in the development of CNV.9-11 Choroidal haemodynamics are important for CNV development in AMD.

Chen and colleagues report abnormal choroidal blood flow in asymmetric exudative AMD by measuring pulsatile ocular blood flow (POBF) using an OBF tonometer in 37 patients with asymmetric exudative AMD in this issue of the BJO (p 1411). Their results demonstrated that POBF was significantly higher in eyes with CNV than in the contralateral eyes with drusen. Eyes with disciform scars had lower POBF than the contralateral eyes with drusen. There was no significant correlation between the POBF and the size of the CNV lesion. These authors concluded that haemodynamic differences between fellow eyes in individuals are relevant to CNV development and the formation of disciform scars.

We recently studied ocular haemodynamic abnormalities by examining the choroidal blood flow using POBF in patients with AMD.3 We used a Langham OBF computerised tonometer in 10 patients with non-exudative AMD, 11 patients with exudative AMD, and 69 age matched control subjects. We reported that POBF and pulse amplitude are lower in the patients with exudative AMD than in the patients with non-exudative AMD and normal subjects. Several angiogenic factors increased in the CNV membranes and vitreous fluid in the patients with AMD.10 12 In an experimental model of CNV, the expression of VEGF in the accumulating macrophages, migrating retinal pigment epithelial (RPE) cells, and Mueller cells increased.11 The RPE cells can promote endothelial proliferation via production and release of VEGF during hypoxia.13 These reports suggest that angiogenic factors may play a part in the formation of CNV. In our recent study, POBF in patients with AMD decreased by about 30% compared with controls and patients with non-exudative AMD.3 The decreased total choroidal blood flow may induce CNV via angiogenic factors resulting from hypoxia in patients with exudative AMD. We concluded that decreased choroidal blood flow may play a part in the development of CNV in AMD.

Several authors investigated choroidal haemodynamic abnormalities in patients with AMD using different blood flow measurements. Ciulla and coworkers reported that the blood velocities of the short posterior ciliary artery are lower in patients with AMD.4 Their results indicate that the total choroidal blood flow is lower in patients with AMD compared with controls. Grunwald and coworkers reported that the choroidal blood flow in the centre of the fovea was lower in the patients with non-exudative AMD than in age matched controls, primarily because of decreased blood volume measured by laser Doppler flowmetry.5 Friedman and coworkers reported that scleral rigidity increases and choroidal blood flow decreases in patients with AMD6 14 15 and concluded that the eyes of patients with AMD had decreased choroidal blood flow and increased resistance in the choroidal vasculature. There is no doubt that choroidal haemodynamic abnormalities exist in patients with AMD.

Recent treatments, such as laser photocoagulation, photodynamic therapy, and surgery, are efficacious. However, the number of patients who recover good visual acuity after treatment is limited. Thus, there is need for pharmacological therapy that prevents the development of CNV. An understanding of choroidal haemodynamics abnormalities may be important for the development of pharmacological therapy in AMD. For example, we reported that losartan, an angiotensin receptor antagonist, inhibited development of CNV in an experimental model.16Choroidal haemodynamics may have to be assessed to evaluate the inhibitory effect of several drugs on CNV development in patients with AMD.

The study of Chen and coworkers showed that POBF was significantly higher in eyes with CNV than in the contralateral eyes with drusen, and eyes with disciform scars had lower POBF than the contralateral eyes with drusen in exudative AMD. Our recent study published in the BJO showed that POBF in patients with exudative AMD is lower than in patients with non-exudative AMD and normal subjects.3 There seem to be differences in these results using POBF. Chen and coworkers did not conduct a group comparison but an interindividual study in which they evaluated only patients with exudative AMD. In another of our studies, axial length was a major factor affecting POBF in normal subjects.17 When evaluating choroidal blood flow by measuring POBF, the effect of axial length should be considered. However, the severity in exudative AMD was not considered in our group comparison study. Therefore, the results of these two studies cannot be compared. Long term follow up and large group comparison studies are needed in the future to gain an understanding of choroidal haemodynamic abnormalities in AMD using POBF and another ocular blood flow measurements.