Sir,

We note with interest the comments made in the letter from Gray et al, and thank the authors for their interest in our paper.

With reference to our use of anaesthetists, not all of our vitreoretinal cases have LA administered by one anaesthetist. The data we presented were indeed from one anaesthetist (ACW), but this does not represent our entire vitreoretinal workload. Other anaesthetists, both consultant and trainees, are assigned to the Eye theatres and as such have good exposure to ophthalmic anaesthesia.

In addition, much of our out of hours vitreoretinal work is also undertaken using LA, but these data are not included in this series. In cases where an untrained anaesthetist is present, the block is given by the surgeon and when no anaesthetist is available, sub-tenons anaesthesia is used.

With regard to our use of sedation, discussed in paragraph 4, we have not encountered any significant clinical difficulties or problems with patient cooperation and have found it to be extremely useful in some patients. We report high patient satisfaction rates in our paper1 and in our opinion sedation is a useful adjunct in selected cases.

Ophthalmology is becoming an increasingly day-case oriented specialty. Many ophthalmic units face pressure over use of inpatient beds. While it is true that GA can be administered on a day-case basis, the absence of beds in which to recover patients adds pressure to the service and may result in elective cancellations. This is unfortunately a reality, and as such will of course be a factor driving a predominantly LA service. This apart, we find LA to be highly acceptable to both patients and staff alike.

Only 6% of patients (39/518) undergoing retinopexy (with or without vitrectomy) had GA.1 This therefore has little effect on our reported LA rate, as suggested by Gray et al. Our previous work involved taking the opinion of the patient, who being recovered by the anaesthetist, had no preconceived surgical opinion on what does or does not hurt. We found that the laser and cryopexy were more important determinants of discomfort during vitrectomy than other aspects of the surgery, and so these were analysed as one group.2 While we would agree with Gray et al that most retinopexy would not require GA, there are occasional anxious patients who have had failure of treatment at the slit-lamp, and for whom good anaesthesia is as vital as it is in a vitrectomy for retinal detachment.

In response to the comments in paragraph 6, by Gray et al, regarding patient comfort and training issues, we would like to draw attention to the findings in our paper. We noted high patient satisfaction rates, both with the anaesthetic injection and the procedure. We had no cases of globe perforation. Teaching cataract surgery under LA is an experience we have all been through and the principles, when applied to VR surgery are just the same. We have an active vitreoretinal teaching program for both specialist registrars and fellows. We have found LA to be perfectly acceptable for teaching, as many of the procedures reported in our series were performed by trainees.

With regard to examination of the fellow eye, we would agree with Gray et al that LA does pose a disadvantage here. What is not known is on how many occasions fellow eye treatment is needed, and whether or not it could easily be administered as an outpatient, or indeed the acceptability of another LA for the patient.