Main

Sir,

Bacterial subretinal abscess is a rare clinical entity, previously reported in association with immunosuppression, septicaemia, and liver abscesses. Implicated organisms have included Streptococcus,1 Klebsiella,2,3 and in one case Pseudomonas was identified in an organ transplant patient.4 Here we describe a subretinal infection resulting from chronic bronchial colonisation with Pseudomonas and presumed transient bacteraemia, in an otherwise immunocompetent host.

Case report

A 75-year-old Asian man presented with pain, foggy vision, and photophobia in the left eye of 24 h duration. He had a history of post-tuberculous cystic bronchiectasis and bronchial colonisation with Pseudomonas aeruginosa, although at presentation there was no clinical evidence of active pulmonary disease.

On ophthalmic examination, visual acuity was 6/6 in both eyes. Fundoscopy of the left eye revealed periphlebitis and a superonasal granulomatous-like haemorrhagic lesion of the choroid and retina. Blood tests, including full blood count, erythrocyte sedimentation rate, and angiotensin-converting enzyme levels, were normal. Syphilis and toxoplasma serology were negative. Blood cultures showed no growth. The patient was started on dexamethasone eyedrops, and as the differential diagnosis included ocular tuberculosis, he was put on a course of oral isoniazid and pyridoxine.

After 3 days, the vision in the left eye had deteriorated to 6/60 and a hypopyon was detected. Fundoscopy showed another granulomatous-like white lesion below the optic disc. A vitreous biopsy was carried out followed by intravitreal injections of 0.4 mg amikacin and 1 mg vancomycin. No organisms were grown from vitreal culture. Over the next 24 h, visual acuity in the left eye dropped to hand movements. No view of the fundus was now possible. The patient was then referred to a tertiary centre for further evaluation and management.

On presentation to Moorfields Eye Hospital, 10 days after the onset of symptoms, the clinical appearance was compatible with a left endophthalmitis. A vitrectomy was performed with injections of intravitreal amikacin and vancomycin. As there was still no improvement, a diagnostic transretinal biopsy was considered necessary. The abscess fluid grew a heavy growth of Gram negative rods, which were later confirmed as being Pseudomonas aeruginosa sensitive to ciprofloxacin. Antituberculous medication was stopped, and treatment continued with oral steroids and oral ciprofloxacin. However, an intense postoperative intraocular inflammatory response developed, resulting in an anterior chamber fibrinous infiltrate. Despite further injections of intravitreal antibiotics over the next 2 months, the visual acuity in the eye deteriorated to no perception of light. A left evisceration had to be performed as a consequence of uncontrolled inflammation and pain.

Currently, over a year after presentation, he has had no further ophthalmic problems. He continues to require cyclical antibiotics for recurrent infective exacerbations of pulmonary disease.

Comment

Bronchiectasis is characterised by irreversible dilatation of the bronchi with susceptibility to increased sputum production and recurrent bronchopulmonary infection. In clinically stable patients with bronchiectasis up to 17% have bronchial colonisation with Pseudomonas species.5 The pathogenesis of chronic bronchial sepsis is usually dependent on impaired mucociliary clearance, by exogenous agents or genetic disease, rather than immunological deficiency. Metastatic infection from bronchiectasis to the eye, in this case presumably the result of transient subclinical bacteraemia, has not previously been reported. This case highlights that chronic bronchial sepsis should be considered as a possible source of ocular disease.

Among patients with Gram negative endogenous endophthalmitis, Pseudomonas is associated with the worst visual prognosis.6 Most reported cases of subretinal abscess have also resulted in evisceration, enucleation, or death from underlying disease. The optimal management of the lesion has yet to be determined. The best visual outcomes have followed attempts to identity the pathogen and achieve sterilisation by early use of antibiotics, vitrectomy, and transretinal drainage, despite the attendant risk of retinal detachment. The only previously reported case of subretinal Pseudomonas abscess, described in an immunosuppressed patient presenting 1 month after bilateral lung transplantation, was successfully treated in this manner and a final visual acuity of hand movements was achieved.4 However, the above case illustrates that even with aggressive medical and surgical therapy, subretinal Pseudomonas infection can follow an accelerated clinical course with rapid vision loss.