A 10 year review of endophthalmitis seen at Kenyatta National Hospital

Abstract

A 10 year review of endophthalmitis as seen at Kenyatta National Hospital (KNH). OBJECTIVES: To describe the magnitude, pattern and visual outcome of endophthai mitis as seen atKNH over a 10 year period from 1999 to 2008. DESIGN: Retrospective case series. SETTING: Kenyatta National Hospital (KNH), a national referral and teaching hospital in Nairobi, Kenya. SUBJECTS: All patients diagnosed with endophthalmitis seen at KNH within the study period. MATERIALS AND METHODS: Medical records of the subjects were retrieved from hospital registry using ICD 10 classification and analyzed. The data was collected using a structured questionnaire. Subsequently it was entered ans! fUlalyzed usihg the Statistical Package for Social Scientists (SPSS) computer software version 11. RESULTS: 74 cases of endophthalmitis were identified, all unilateral. 39% of the cases were children (<I5 years). The male to female ratio was 2:1 with males being 65% and females 35% of

the patients however the difference was not statistically significant (p=0.74). Prevalence of endophthalmitis in KNH was found to be 0.01%. Average delay from onset of symptoms to admission at KNH was found to be one week. The commonest presenting symptoms were red eye (89.2%), pain (86.5%) and visual loss (73%). Hypopyon (100%) and corneal haze (97.3%) were the main findings on examination. Post-traumatic endophthalmitis (63.5%) was found to be the commonest type in KNH. Others were endogenous (18.9%), post-corneal ulcer (9.5%), postoperative (5.4%) and post-intravitreal triamcinolone endqphthalmitis (2.7%). 41.9% of the cases were done vitreous tap and the positive culture yield was 25.8%. Staphylococcus aureus was the commonest isolated organism. Most organisms isolated were sensitive to ciprofloxacin except Pseudomonas aeruginosa which was sensitive to ceftazidime. 50% of the cases were eviscerated, 47.3%treated conservatively with broad spectrum antibiotics and 2.7% had pars plana vitrectomy. All the subjects presented with poor visual acuity (NPL = 65%, PL = 31.1 %) that eventually resulted in poor visual outcomes. The study found a high rate of loss to follow-up; 36.5%, 79.7% and 97.3% on 1st, 2nd and 3rd follow-up visits respectively. CONCLUSIONS: Endophthalmitis is rare at KNH. Poor visual outcomes were due to poor presenting VA, post-traumatic endophthalmitis and delay in presentation. RECOMMENDATIONS: Prompt referral of suspected cases of endophthalmitis should be advocated.