Sir,

We thank Vasumathy Vedantham for his interest in our paper. In reply to his comments:

  1. 1

    The British Ophthalmological Surveillance Unit (BOSU) is now well established in the UK to assist in the investigation of clinically important or rare eye conditions such as endophthalmitis. The BOSU-reporting scheme is dependent on voluntary reporting and therefore under-reporting of cases is a potential source of error for incidence calculations. A national registry is ideal for monitoring of postoperative endophthalmitis cases but would require the support of the vast majority of ophthalmologists in the UK or India.

  2. 2

    The incidence of diabetes in the endophthalmitis cases was 10%. This figure was mentioned in our paper under the heading of cataract surgery details. Comparison of outcome between diabetic and nondiabetic eyes was not analysed in this paper.

  3. 3

    Pseudomonas was cultured from the aqueous sample but not the vitreous in 3/5 Pseudomonas isolates. This was a surprising result as vitreous is a better culture medium for bacteria than aqueous;1 however, cases with a positive aqueous tap and negative vitreous tap have been described.1, 2 As this study was dependent on individual UK centres providing specific information about endophthalmitis cases, we cannot comment on the vitreous/aqueous sampling technique or sample processing at these centres.

  4. 4

    In contrast to several Indian endophthalmitis series,3, 4, 5 no cases of fungi were isolated in our UK survey. Fungal endophthalmitis following cataract surgery has been described as having a prolonged latency period of weeks to months after intraocular inoculation.5, 6 For our UK survey, we only included cases diagnosed within 6 weeks of cataract surgery, excluding potential delayed onset cases due to fungal infection. Differences in climate, operating theatre conditions or the sample processing technique may also be important. Information on the antibiotic susceptibility of ocular isolates was not requested for our study.

  5. 5

    The role of systemic antibiotics in the treatment of endophthalmitis remains unclear. Patients randomised to the intravenous (IV) antibiotic group in the Endophthalmitis Vitrectomy Study (EVS)7 received ceftazidime and amikacin (oral ciprofloxacin instead of ceftazidime if the patient was allergic to penicillin). There was no statistically significant difference in outcomes between the IV antibiotic group and the control group. Animal studies have shown inadequate intraocular drug levels of amikacin in inflamed, aphakic, vitrectomized eyes following IV treatment.8 IV vancomycin or cefazolin may provide better Gram-positive coverage in aphakic, inflamed eyes than ceftazidime or amikacin.9 Therefore, the results of systemic antibiotics in the EVS may not apply to other drug combinations.