Patient details

Fifteen patients aged over 65 years (range 67–86) were recruited. Symptoms of poor or distorted vision ranged from 1 month to about 11 months. Fluorescein angiography in all cases confirmed a classic subfoveal neovascular membrane (NVM). Pretreatment visual acuities ranged from 6/12 (Snellen chart) to counting fingers (CF).

Method of treatment

All patients were positioned supine on the treatment table. Head fixation was achieved using either a customised moulded Perspex mask or a bite block (Figs 1 and 2). The first eight patients were treated using a Perspex mask while patients 9–15 used a bite block device.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

A patient undergoing radiotherapy using a customised Perspex mask to maintain head fixation. The mask is bolted to the treatment table.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Head fixation is maintained during radiotherapy by the patient biting on the moulded rubber impression which is attached to a metal arm fixed to the head of the table.

A lateral approach with an angled half beam blocking technique was used to spare both the ipsilateral and contralateral lens. A 4 × 4 cm or 4 × 3 cm treatment field was marked using the lateral canthus of the affected eye as a guide for the anterior border. The field was angled 5 degrees posteriorly from the lateral canthus such that the 90% isodose curve encompassed the macula. A simulator confirmed the correct positioning of the patient and skin marks plus laser beam alignment were used to ensure reproducibility of the treatment field at each visit. A 6 MV linear accelerator was used to deliver the ionising radiation and 2 Gy was given at the 90% isodose for each fraction. All patients received a total of 13.3 Gy (12 Gy to the 90% isodose curve) given as 6 × 2.22 Gy fractions over a 9 day period.

Follow up visits were scheduled for 2, 4, 8, 12, 18, and 24 months following radiotherapy. All patients were seen and had their visual acuities tested by the same ophthalmologist (AJC) at each visit (using the same chart) in order to minimise observer differences. We used visual acuity as a measure of success of the treatment although other variables, not discussed here, were measured.

Results and discussion

Follow up data are complete for all 15 patients over 24 months.

TREATMENT VERSUS NO TREATMENT

To determine whether radiotherapy was better than conservative management we compared our post-treatment visual acuities with data compiled from natural history studies.1-3 Stable or improved visual acuity was scored where there was less than a two line deterioration in Snellen acuity. We found that 24 months after radiotherapy 8/15 eyes (53%) maintained pretreatment visual acuities or showed an improvement on the Snellen chart and 7/15 (47%) showed a deterioration. If left untreated only 28% individuals would be expected to maintain their visual acuity 24 months after diagnosis. This suggests that treatment is better than no treatment.

IS CT SCANNING ESSENTIAL?

In our study we did not feel it was necessary to use CT scanning to visualise the macula and its relation to the entry point of the field particularly as the optimum dose of radiotherapy has yet to be established. The distance, however, from the temple to the macula does not vary greatly from person to person and is likely to be between 3–4 cm from the beam entry point on the skin. The dose at 3 cm for a 6 MV beam is 93.3% of the entry dose and 88.4% at 4 cm. This fall off in dose is the same whatever single field technique is used so the macula will receive approximately 90% of the entry dose.

In order to test whether the use of CT scanning was essential we compared the post-treatment visual acuities from our study with those from the pilot study.1 Clearly patients were not identical but useful conclusions can still be drawn. It is important to note that the time interval from onset of symptoms to time of treatment did not appear to affect the outcome in our study (data not shown) and no patients developed significant lens opacification during the period of follow up. Using the same criteria for success, in the pilot study, where CT scanning was used to align treatment to the macula, 68% of individuals showed stable vision after 12 months. We have demonstrated that 53% of individuals showed stable visual acuity 24 months after treatment. Given that visual deterioration occurs with time even after radiotherapy, we feel our results compare favourably with those from the pilot study and question whether the cost of CT scanning is justified.

HEAD FIXATION DEVICES

To determine the effectiveness of head fixation we compared head movement during treatment and post-treatment visual acuities in the two groups using either a mask or bite block. Head movement during treatment was found to be slightly greater in those patients using a bite block fixation device but was within 1–3 mm of the set up position and in the directional plane away from the table in line with the 90% isodose curve. This would not be expected to significantly affect the dosage of radiation received by the macula. This was confirmed by careful comparison of repeated on treatment check films captured during each fraction of radiation from which it was felt that the small degree of movement did not remove the macula from the treatment field in any case. Visual acuity comparison of eight moulded masks versus seven bite block head fixation devices showed that after 24 months 5/8 (67%) in the mask group showed stable acuities compared with 3/7 (43%) in the bite block group. These results are considerably better than the 28% expected to maintain vision after 24 months if left untreated. If we regard both fixation devices as equally effective in maintaining head position during treatment there must be an explanation as to why the group using masks appeared to fare better. The answer may simply be that the numbers in each group are so small that a deterioration in just one individual has a major effect on the results. However, for completeness, we looked to see if the initial visual acuities were markedly different or the original size of the membrane or the time taken to treatment in the two groups. We found that the initial visual acuities were slightly better in the bite block group in that 3/7 had acuities of 6/24 or better compared with only 1/8 in the mask group. It could be argued that a small increase in the size of the membrane, or atrophy of the pigment epithelium, would have a greater effect on an acuity of 6/12 compared with one starting at 6/60. The effect of this would become more obvious as follow up increased so that those individuals with a better initial acuity would be expected to fare worse. Interestingly, this was seen in that after only 12 months 5/7 (71%) of the bite block group had stable acuities reducing to 3/7 (43%) by 24 months. This reduction was not observed in the group using masks. Our hypothesis that individuals with better initial visual acuities may not fare so well over time as those with poor initial vision is partly supported by data from other studies, although follow up is relatively short. Hollick et al reported that 53% of individuals with an initial acuity of 6/24 or better had stable acuities at 1 year after irradiation.4 Using the same criteria for success Chakravarthy et al showed that 68% of individuals with an initial acuity of 6/24 or worse had stable acuities 1 year after radiotherapy.1

This study was stimulated by clear patient dissatisfaction at the method of head fixation using the masks. Bite blocks, however, were well tolerated by all our patients and were less than 17% of the cost of a mask to manufacture. Moulding a bite block is simple and quick using dental materials to create an impression of the individual’s teeth. This is then attached to a metal arm that is, in turn, fixed to the head of the treatment table. The preparation of a mask requires two visits by the patient who must remain still while strips of plaster are applied to the face. A Perspex copy of the mask is then made with a treatment window removed and screw holes drilled. During the treatment session head fixation is achieved by bolting the mask to the table. Not surprisingly the patients all found this experience claustrophobic and admitted that having the mask made was the worst part of the whole treatment.

We feel that this study demonstrates that the visual prognosis following the diagnosis of an age related subfoveal neovascular membrane can be improved by the use of radiotherapy. Furthermore this can be achieved without the use of CT scanning. Although the numbers in this pilot study are too small for useful statistical analysis, we have demonstrated that the bite block gives fair results in terms of visual outcome, is more cost effective, and better tolerated by our patients than the Perspex masks. Larger studies are now needed to directly compare these methods of head fixation.