Dear Editor

I was interested to read the article by Garcia-Arumi and colleagues on "Surgical embolus removal in retinal artery occlusion".[1] The authors claim that "Surgical removal of retinal arterial emboli seems to be an effective and safe treatment for RAO (retinal artery occlusion)".

Briefly, the study was based on 6 eyes with temporal branch retinal artery occlusion (BRAO) and one with central retinal artery occlusion (CRAO). The surgery was performed in BRAO eyes 4, 12, 19, 22, 28, and 33 hours after onset and in CRAO eye 29 hours after visual loss. The first post-operative evaluation, 48 hours after surgery, showed reperfusion of the occluded branch retinal artery in 4 and none in one; in the remaining eye, no evaluation was possible for 2 weeks because of vitreous hemorrhage. In the 6 eyes with BRAO, pre-operative visual acuity was: hand motion, counting fingers, 20/400, 20/200, 20/200 and 20/25, at 48 hours it was 20/200, 20/100, 20/100, 20/200, 20/100 and 20/30 respectively, and final visual acuity was 20/50, 20/30, 20/80, 20/200, 20/25 and 20/25 respectively; in the CRAO eye, visual acuity remained hand motion all along.

It is worth noting, initially, that a conclusion based on 6 eyes is limited. However, the matter of apparent improvement in visual acuity claimed by the authors deserves comment.

1. All 6 of the eyes had temporal BRAO. In the vast majority of temporal BRAO eyes the border between the ischemic and non-ischemic retina passes through the fovea, as their figures 2 and 3 show. Also the fluorescein angiogram in Figure 2 shows a patent cilioretinal artery. These are extremely important facts in determining whether the visual improvement noted was result of removal of the embolus or simply the natural history of the disease. Having studied the natural history of about 200 BRAO eyes, I recently stated: "In cilioretinal artery occlusion, branch retinal artery occlusion and CRAO with cilioretinal artery sparing, the junction between the infarcted and normal retina may often pass through the fovea. In many of these eyes, I have seen marked spontaneous visual acuity improvement occurring within several days or weeks, from almost 20/200 or worse to even normal. This spontaneous improvement is often erroneously attributed to an advocated treatment."[2]

When the junction between the normal and infarcted retina passes through the fovea, in such eyes, the retinal edema associated with retinal infarction most probably also involves to some extent the adjacent normal foveal retina shortly after the occlusion; over the following weeks the normal foveal retina recovers spontaneously, resulting in natural visual improvement. In the series of Garcia-Arumi and colleagues a significant visual improvement at final visit seen in 4 of 6 eyes is not uncommon in such eyes, as a part of the natural history in my experience. Moreover, the eye in figure 2, not only had the junction between the normal and infracted retinal passing through the fovea but also had patent cilioretinal artery, which is an important factor in natural visual acuity improvement not only in BRAO but also in CRAO[3].

2. Having studied more than 450 patients with BRAO and CRAO, I have found another common confounding factor involving visual acuity testing. At the initial visit, because of sudden visual loss, the patient is emotionally upset, and tends to test poorly; for instance a patient with only temporal BRAO with the border between the ischemic and normal retina passing through fovea should not have hand motion or even 20/400 visual acuity, if tested properly because the other half of the macula and retina is still functioning normally. Moreover, later on it is not unusual to find patients with central scotoma learning to fixate eccentrically and show better visual acuity, which may erroneously be interpreted as genuine improvement. In my studies on various ocular vascular occlusive diseases I have found that unless improvement in visual acuity corresponds with improvement in central scotoma, it is not a genuine improvement but due to eccentric fixation.[4]

3. Since visual acuity testing assesses only the function of the fovea and not of the entire involved retina, visual fields, particularly with a Goldmann perimeter, give us much better information about the extent of visual loss and change. In my study, every eye with BRAO had visual fields plotted with a Goldmann perimeter; that showed that the visual fields of BRAO eyes often did show reduction in size of the visual field defect as a part of the natural history. Garcia-Arumi and colleagues state that they recorded the visual fields with Humphrey perimeter but give no information on the visual fields of their cases. Moreover, unlike the Goldmann perimeter, Humphrey perimeter provides information about only the central 240 - 300 and not the entire involved retina.

4. In the series of Garcia-Arumi and colleagues, 6 of 7 eyes had had acute retinal ischemia for 12 to 33 hours and in one for 4 hours. We evaluated the retinal tolerance time to acute ischemia experimentally in rhesus monkeys[5] and found that in CRAO, ischemic retina can recover normal function from acute ischemia of 97 minutes, but after that the longer the ischemia, the more extensive the irreversible damage so that acute ischemia lasting for 240 minutes results in massive irreversible retinal damage. Therefore, it does not seem logical that restoration of circulation in BRAO 4 to 33 hours after the occlusion would restore function in an already irreversibly damaged retina. Moreover, they found restoration of circulation in 4 of the 6 eyes on fluorescein angiography first done 48 hours after the surgery. They argue that "in branch RAO because some degree of perfusion at the macular area may be supplied by the contralateral temporal artery." [1] This may be true, but it may also be another factor in the spontaneous marked visual recovery in such eyes as a part of natural history.

In conclusion, based on my studies on the natural history in BRAO eyes, I feel the visual acuity improvement attributed by Garcia-Arumi and colleagues to embolectomy simply represents natural history.

References

1. Garcia-Arumi J, Martinez-Castillo V, Boixadera A, et al. Surgical embolus removal in retinal artery occlusion. Br J Ophthalmol 2006;90:1252- 5.

2. Hayreh SS. Prevalent misconceptions about acute retinal vascular occlusive disorders. Prog Retin Eye Res 2005;24:493-519.

3. Hayreh SS, Zimmerman B. Central Retinal Artery Occlusion: Visual Outcome. Am J Ophthalmol 2005;140:376-91.

4. Hayreh SS, Zimmerman B, Kardon RH. Visual improvement with corticosteroid therapy in giant cell arteritis. Acta Ophthalmol Scand 2002;80:355-67.

5. Hayreh SS, Zimmerman MB, Kimura A, et al. Central retinal artery occlusion. Retinal survival time. Exp Eye Res 2004;78:723-36.

Yours sincerely,

S.S. Hayreh

Sohan Singh Hayreh, MD, PhD, DSc, FRCS, FRCOphth Professor Emeritus of Ophthalmology & Director Ocular Vascular Clinic