Sir,

Dinakaran et al1 raise an interesting question as to success rates in retinal reattachment surgery in their letter published in the May issue of Eye. They refer to earlier work, including audit, conducted by ourselves, so we may perhaps be permitted to comment further.

The authors refer to results quoted from our service and those from Sullivan et al.2 The latter suggested that a success rate of 75% was acceptable. We disagreed on the grounds that there had been no reported improvement in their retinal reattachment success rate since an audit conducted at the same hospital and published 25 years before.3

Retinal reattachment audit in East Anglia was conducted for the first time in four hospitals between 1 January 1989 and 31 December 1990. District general hospitals achieved between 47 and 75% success rate compared with the specialist unit's 88%. All surgery included in that audit was carried out using scleral buckling techniques. As a result, the majority of retinal detachments were thereafter referred primarily to the specialist unit.

A repeat independent audit of the specialist unit was conducted between 1 September 1995 and 31 March 1997. In all, 87% of scleral buckles were successful and for the first time patients treated by vitrectomy with both gas and silicone tamponade were included. In total, 91% of these were successful at the first operation.

Results that we quoted in our letter4 in response to the paper reporting audit results from Sullivan et al were based on independently conducted audit, published later.5 We believe that this is an important aspect of any assessment of results.

Dinakaran and colleagues refer to the improvement in results when the junior was supervised. The ‘junior’ in question was in fact an experienced senior registrar appointed as Fellow to the vitreoretinal service. At the time of the first audit, he operated unsupervised in keeping with practice prevailing at the time and still in current use in some institutions. Patients were largely unselected. The improvement in results from 78 to 94% when supervised by the consultant led to a change of practice in which all surgery was either conducted by or supervised at all times by the consultant. We believe that supervision of Fellows by an experienced vitreoretinal specialist consultant is vital to the achievement of best results.

The results of our most recent audit of 319 unselected consecutive primary detachments between 1 January 2001 and 31 July 2002 covered a period of time when a more graded supervision of inexperienced juniors (SpR and SHO) was undertaken.6 Juniors achieved a 96.6% success rate overall. This included both conventional buckling procedures and vitrectomy using the indirect ophthalmoscope. The overall success rate for the vitreoretinal unit was 88.4% for a single procedure.

Dinakaran and colleagues are to be congratulated on an 81% success rate for primary scleral buckling surgery with a final figure of 90%. The total number of 58 patients operated over a 5-year period does perhaps reflect one of the problems that general ophthalmologists operating on retinal detachments experienced in the past; that of low numbers. We would guess that this was perhaps one of the factors that led them to refer all patients to the regional vitreoretinal service once it had been set up.

We believe that it would be preferable to discuss results in terms of failure rather than success, since patients expect us to get it right all of the time. If we do not succeed at the first operation, then it is failure; however, one dresses up the terminology to comfort ourselves.

Goals should always be set beyond those currently attainable in order to stimulate endeavor and prevent complacency. If we do not have a 0% failure rate after the first operation as our ultimate goal, we shall have failed a significant number of our patients. The way to achieve this is to identify the reasons for failure and to remedy them. Many failures are due to inaccurate surgery2 and some to new retinal breaks. The former is obviously remediable; the latter requires further research into the causation of the very many variety of retinal breaks. A clear understanding of how retinal integrity breaks down leading to detachment still eludes us.