Uveitis: an emerging clinical form of Bartonella infection

Abstract

Listen

Translate article

A few reports have now established that the cat scratch disease agent Bartonella henselae is responsible for uveitis [1]. Bartonella grahamii has also been recognized as being responsible for uveitis [2]. However, the prevalence of Bartonella uveitis remains poorly known, as is the potential role of other Bartonella species. Prospective evaluation of a standardized laboratory protocol for the diagnosis of uveitis of unknown aetiology [3] showed that Bartonella species were responsible for most cases of bacterial uveitis. Prominent clinical features and laboratory diagnosis data of this large series are reported herein. This study was approved by the Ethics Committee of our institution; patients with a known aetiological diagnosis and patients with ophthalmological features pathognomonic of an aetiology, such as toxoplasmosis, were not eligible. A questionnaire was used for the collection of data, and was anonymized before the study according to French law. Informed consent for inclusion was obtained from patients and parents or legal guardians of children (i.e. age <18 years). Uveitis was classified according to the international uveitis study group definition and classification. Five millilitres of serum and ocular fluid were collected aseptically, by anterior chamber paracentesis or vitreous tap ⁄ vitrectomy. Serum was tested for the presence of antibody against B. henselae Houston and Marseille serotypes, Bartonella quintana and B. grahamii using microimmunofluorescence and Western blot [4]. After lysis, nucleic acids were isolated using the tissue kit and the Magna-Pure apparatus, according to the instructions of the supplier (Roche Applied Science, Indianapolis, IN, USA). Intra-ocular specimens from deceased patients without infection from the Marseilles area as well as water and mix samples were used as negative controls. One negative control was used for eight specimens, and any false-positive control annulled the validity of the assay. A nested PCR targeting the Bartonella hbpE gene, using primer pair 5¢-GAGAGTGCTTCACCTAAATAG-3¢ and 5¢-CCACCAATCTGTCCTCCAAA-3¢, was performed. The diagnosis of Bartonella uveitis was considered as definite if: (i) positive PCR with an identifying sequence was obtained from ocular fluid; and (ii) the antibody level was ‡1 : 100 by microimmunofluorescence or there were ‡2 specific bands in Western blot for Bartonella spp. Between January 2001 and April 2008, 1520 intra-ocular and serum specimens collected from 1417 patients were analysed. Bartonella spp. infection was diagnosed in 31 (2.6%) patients, by specific molecular testing of intra-ocular specimens only in four patients, and by Western blot only in ten patients. Uveitis was due to B. henselae in 17 patients, B. quintana in seven patients, B. grahami in four patients, and undetermined Bartonella spp. in three patients (Table 1). All cases of B. henselae uveitis were due to the Houston genotype as determined by sequencing PCR products or by Western blot analysis. In our series, Bartonella spp. ranked second after spirochaetes among the fastidious pathogens responsible for uveitis. Characteristic stellar retinitis was found in nine of 31 (29%) patients, but clinical features were otherwise non-specific, with uveitis being posterior in 18 of 31 (58%) patients, anterior in nine of 31 (29%), intermediate in two of 31 (6.5%), and panuveitis in two of 31 (6.5%). Patients presented with non-granulomatous uveitis in 18 of 31 (58%) cases and granulomatous uveitis in 13 of 31 (42%) cases. Contacts with animals were notable in 11 of 31 (35.5%) patients. In most patients, the diagnosis was made by Corresponding author and reprint requests: Michel Drancourt, URMITE, Faculte de Medecine, 27 Boulevard Jean Moulin, 13385, Marseille cedex 9, France E-mail: michel.drancourt@univmed.fr