Sir,

We thank Cheng et al for their interest in our article entitled ‘The Ahmed glaucoma valve in refractory glaucoma: experiences in Indian eyes’.

The fundamental difference between the two studies1, 2 appears to be a variation in the surgical technique. The dissection of the scleral flap was the only major surgical modification of the technique that was different from the procedure described in the studies performed previously.2, 3 Although the scleral dissection was not as deep as it is in nonpenetrating deep sclerectomy (NPDS) in our study,1 in most cases the flap was between two-thirds to three-fourths of the scleral thickness, so as to provide adequate support to the AGV tube. This was the basis of our postulation that egression of aqueous from the scleral flap and bed,4 as is seen in a trabeculectomy, may have contributed to the blunting of the ‘hypertensive’ phase. This, however, remains a nonmeasurable compounding factor, which had no adverse outcome on the postoperative behaviour of the patient's intraocular pressure (IOP). Even if we assume that both, the egression of aqueous from the scleral bed and the drainage through the AGV implant, contributed to the reduction in the IOP, the effect was better control of the same in the postoperative period, which was desirable. However, this query provides food for thought for a future randomized prospective comparative study where the implant is inserted under a scleral flap (measured depth) and under a donor corneoscleral graft so as to come to a solution to this clinical dilemma.

Encapsulated blebs were not encountered in our study as a cause of failure. We have mentioned in the article that this could probably be due to a shorter recorded follow-up period or probably a less aggressive tissue healing process in Indian eyes.1 The latter hypothesis is presumptive and would need substantiation by further randomized trials taking into account the response to surgery in different races. Most of the cases classified as ‘failures’ in our study were patients with refractory and complicated glaucomas (neovascular, aphakic, postuveitic, congenital, etc) and the cause of failure was due to inadequate control of IOP in spite of maximum medical therapy as defined in our success criteria.1 Another important difference between the two Asian studies1, 2 on AGV implantation in refractory glaucomas that we thought should be highlighted is that the patient groups in the two studies were different. The most common diagnosis in the study by Lai et al2 was neovascular glaucoma while that in our study was failed trabeculectomy in primary glaucomas.1 This could also have contributed to a different pattern of cases classified as failures in the two studies.