Sir,

After reading this excellent article (Eye 2003; 17: 506-512), it is quite clear that ‘sterile endophthalmitis’ could be due to these residual debris, but it is not clear that the high incidence of postsurgical endophthalmitis (PE) in 1999 is only due to residual debris.

If the sterilising procedures were correct and the rate of phacoemulsification surgeries were similar in 1998 and 1999 (although having different PE annual incidence), the proven infectious PE of 1999 could be due, for instance, to an insufficient surgical prophylaxis (data about the hospitals’ prophylaxis protocols are not provided), or to an accumulation of patients with a higher risk of a bad outcome1, or/and to some specific factor associated with the end of 1999.

On the other hand, it is difficult to keep on accepting as ‘current PE incidence’ that given in 1991 for Kattan2 and Javitt et al3 for the following reasons:

  1. 1)

    Their PE incidences refer to cataract surgery using extracapsular technique.

  2. 2)

    The Kattan et al incidence excludes those PE cases not proven by culture. However, 5 years later the Endophthalmitis Vitrectomy Study4 described 69% of PE cases proven by culture among their 420 intraocular biopsies.

  3. 3)

    The Javitt et al incidence excluded those patients younger than 65 years; those having diabetic retinopathy; those who underwent cataract extraction combined with corneal, retina, and glaucoma procedures; and those having a secondary implant.

  4. 4)

    The Kattan and Javitt et al studies were retrospective, while a prospective national study 5was published in 1991, which described a 0.31% PE incidence in France; and, another prospective national study 6 gave a 0.3% PE incidence in England, in 1993. Years later, much higher PE incidences were published.7, 8, 9, 10

There seems to be enough information for considering a redefining of the ‘normal PE incidence in cataract surgery’, especially, taking into account the knowledge achieved since 1991 about PE risk factors.