Successful Monitoring and Treatment of Intraocular Dissemination of Acanthamoeba

Intraocular spread of Acanthamoeba after keratitis is rare, being previously documented and microbiologically confirmed in 1 case of chorioretinitis¹ and 1 case of endophthalmitis.² Both occurred several months following penetrating keratoplasty (PK), were initially treated as suspected sterile inflammations, did not receive specific oral or intraocular treatment for Acanthamoeba, and finally required evisceration.

We report the first documented case, to our knowledge, of successful management of intraocular Acanthamoeba dissemination. Intraocular and systemic treatment was monitored through microbiological testing and polymerase chain reaction analysis of serial aqueous taps.

A 56-year-old man with Acanthamoeba keratitis and keratoconus underwent tectonic PK to treat keratitis that was unresponsive to chlorhexidine gluconate and eventually resulted in corneal perforation. Treatment after PK included topical chlorhexidine, moxifloxacin, cyclopentolate hydrochloride, and dexamethasone sodium phosphate as well as oral prednisolone (60 mg once daily) and tacrolimus (1 mg twice daily).

Five days after surgery, the patient reported blurred vision. The graft showed inferior edema, Descemet folds, mutton-fat keratic precipitates, and 4+ cells in the anterior chamber (AC) with a hypopyon (Figure 1A). Taps of the AC were taken and intravitreal vancomycin hydrochloride and ceftazidime were administered. Also, 100 μL of 1-μg/μL voriconazole solution was injected in the vitreous and AC owing to suspected Acanthamoeba dissemination. The AC tap result was positive for T4 Acanthamoeba by culture and polymerase chain reaction.³ Direct microscopical examination of an aliquot of the aqueous sample previously homogenized by gentle inversion showed 10⁴ trophozoites/mL using a Coulter Z automatic cell counter and 90% viability as assessed using trypan blue dye exclusion.³ The voriconazole sensitivity test showed mean (SD) 90% inhibitory concentrations³ of 10.63 (0.46) μg/mL for cysts and 11.28 (0.32) μg/mL for trophozoites. Oral immunosuppression was stopped and treatment with topical voriconazole, 1%, oral voriconazole (200 mg twice daily), and trimethoprim/sulfamethoxazole (800/160 mg twice daily) was started.

The clinical picture worsened during the next 3 days, then improved until clinical resolution 1 month after onset. Ocular echography ruled out vitreous involvement during follow-up. Second and third AC taps followed by voriconazole intraocular injection were performed at days 4 and 11 following the first tap; both were cultured and were positive for Acanthamoeba by polymerase chain reaction. A fourth tap without voriconazole injection confirmed microbiological cure at day 31 on culture and polymerase chain reaction analysis. Figure 2 shows the AC total and viable Acanthamoeba concentrations during follow-up.

The patient's maintenance regimen has been topical chlorhexidine, 0.02%, and voriconazole, 1%, 4 times daily with 200 mg of oral voriconazole twice daily for the last 6 months, without recurrence (Figure 1B). The corneal graft shows early signs of failure, and Descemet-stripping automated endothelial keratoplasty plus phacoemulsification is planned once clinical remission reaches 12 months.

We report the first case, to our knowledge, of a successful outcome in a patient with microbiological evidence of intraocular dissemination of Acanthamoeba. We believe success was based on 3 points that were not emphasized in previously reported cases: (1) early diagnosis; (2) successful combination of topical, oral, and intraocular therapy with drugs that can reach therapeutic levels in aqueous and vitreous^4,5 and are effective in vivo⁶ for other Acanthamoeba infections; and (3) guiding treatment by effective monitoring of the response by Acanthamoeba.

We think oral and topical administration of voriconazole must have achieved a sustained therapeutic dose and frequent administration of intraocular voriconazole produced high peak levels, increasing effectiveness. Topical chlorhexidine was used before the PK but the keratitis worsened, raising the question of its effectiveness in our patient. It is unknown whether topical chlorhexidine can reach aqueous therapeutic levels; however, rabbit studies have shown that frequent instillation of chlorhexidine, 0.02%, in epithelialized corneas⁷ produces concentrations 10 to 40 times lower than voriconazole but, in our experience, a similar 90% inhibitory concentration. Moreover, in the other described cases, topical antiseptics such as chlorhexidine used after PK did not prevent endophthalmitis. Therefore, we believe chlorhexidine did not play a major role in our case. The susceptibility of Acanthamoeba to trimethoprim/sulfamethoxazole, also used in our patient, is based on a few reports⁸; we have not tested the susceptibility of the patient's strain and cannot be sure of its real contribution.

Correspondence: Dr Arnalich-Montiel, Servicio de Oftalmología, Hospital Ramón y Cajal de Madrid, Carretera de Colmenar Viejo km 9.100, 28034 Madrid, Spain (farnalich@gmail.com).

Funding/Support: This work was supported in part by project RICET (project RD06/0021/0005 of the Programme of Redes Temáticas de Investigación Cooperativa, Fondo de Investigación Sanitaria), Spanish Ministry of Health, and the project “Protozoosis emergentes por amebas de vida libre: aislamiento y caracterizacion molecular, identificacion de cepas transportadas de otros agentes patogenos y busqueda de quimioterapias” PO10/01298, Spanish Ministry of Science and Innovation. This work was also supported by grants to Dr Arnalich-Montiel from DGTATX, Ministerio de Sanidad y Consumo (Proyecto TRA-036), and Fundación Mutua Madrileña. Dr Lorenzo-Morales was supported by the Ramón y Cajal subprogram RYC-2011-08863 of the Spanish Ministry of Science and Innovation.