Sir,

Phacoemulsification is the commonest procedure, with over 325 471 surgeries being performed in the NHS.1 Identifying and stratifying the pre-operative risk for cataract surgery is probably more relevant now than before. This is especially so because of the relative reduction in ophthalmic training years and the adverse case mix because of case selection by local independent cataract treatment centres. In this context, the recent publication by Narendran et al2 in Eye provides us all with important and germane information.

The main risk indicators identified were increasing age (adjusted odds ratio (OR)=2.37), brunescent or white cataract (adjusted OR=2.99), pseudoexfoliation/phacodonesis (adjusted OR=2.92), small pupil size (adjusted OR=1.45), axial length>26 mm (adjusted OR=1.47), and trainee status (for SHO, adjusted OR=3.73). It is inevitable that there are further factors that may be important, such as shallow anterior chamber depth (ACD).

Analysis of our Medisoft cataract database (Leeds, UK) showed that 8891 eyes had ACD data (from June 2004 to October 2008). In all, 1138 of these eyes had an ACD⩽2.5 mm and 7753 had an ACD>2.5 mm. PCR±vitreous loss was seen in 23 (2.0%) of the eyes with an ACD⩽2.5 mm compared with 95 (1.2%) in the >2.5 mm ACD group (OR=1.66, P=0.04, 95% CI=1.05–2.63), with an overall twofold risk. After correction for covariates such as cataract brunescence, pseudoexfoliation, and small pupil, the adjusted OR was 1.56 (P=0.056, 95% CI=0.98–2.49). Our sample is clearly too small to give more than an indication, and it would be helpful to know from the authors if they have any similar data.

A reduced ACD of less than 2.5 mm has also been previously associated with increased intra-operative complications in patients with pseudoexfoliation.3 Considering ACD may be helpful in risk stratification for cataract surgery and can help us to further improve our outcomes.