Loading...
In patients with candidemia, hematogenous dissemination can lead to ocular abnormalities, including chorioretinitis and endophthalmitis with vitritis. Previous studies of these lesions have involved small patient cohorts.
Now, researchers have analyzed relevant data from an earlier worldwide trial in which 370 nonneutropenic patients with candidemia were randomized to voriconazole or to amphotericin B followed by fluconazole (Lancet 2005; 366:1435). All trial participants were prospectively monitored for ocular manifestations, and the effect of antifungal therapy on the outcome of ocular lesions was assessed. Both the present study and the previous one were manufacturer funded.
In 60 participants (16%), fundoscopic examination revealed abnormalities consistent with probable (n=40) or possible (n=20) ocular candidiasis. (Intravitreal cultures were not performed to confirm the diagnoses.) Ocular involvement occurred in 36 (23%) of 159 patients with fungemia caused by Candida albicans versus 24 (11%) of 211 patients with fungemia caused by other species. Compared to patients without ocular involvement, those with ocular abnormalities were more often infected with C. albicans (P=0.004) and less often infected with Candida parapsilosis (P=0.011). Overall, treatment for the ocular lesions was considered successful in 65% of patients and unsuccessful in 3%; outcomes were not evaluable in 32%.
Among the 40 patients with probable ocular candidiasis, 6 had endophthalmitis, with lesions extending into the vitreous. Five were diagnosed at baseline and one after 18 days of antifungal therapy. Two of these patients died during treatment — before ophthalmologic follow-up; in three of the four remaining patients, the ocular lesions resolved during treatment.
Oude Lashof AML et al. Ocular manifestations of candidemia. Clin Infect Dis 2011 Aug 1; 53:262.
Comment
These findings indicate that, among nonneutropenic patients, endophthalmitis is an uncommon manifestation of candidemia. In fact, incidence in the study population might have been overestimated: Because intravitreal cultures were not performed, some of the fundoscopic abnormalities noted could have had other etiologies.
Although the ocular lesions resolved in most of the evaluable patients, the nature of the study did not allow analysis of specific therapies. As the authors recommend, dilated fundoscopy should be performed in all patients with candidemia within 1 week after therapy initiation.