Candida Endophthalmitis: Who Is Looking?

Perfect, John R. MD

Infectious Diseases in Clinical Practice:
doi: 10.1097/IPC.0b013e3180f62aef
Editorial Comment
Author Information

Duke University Medical Center, Durham, NC.

Address correspondence and reprint requests to John R. Perfect, MD, Duke University Medical Center, PO Box 3353, Durham, NC 27710. E-mail:

Article Outline

Guidelines for management of infectious diseases represent ideal vehicles to see if "we practice what we preach." They represent a compilation of evidence-based material held together with the glue from clinical experience and the acumen from experts in the field. Infectious Disease guidelines provide clinicians with reference points for reasonable strategies to specific clinical situations. However, they do not substitute for clinical judgment at the bedside because they simply cannot be written to completely adapt to the variety of clinical scenarios or resources faced daily by clinicians. It is not for patient care improvement to codify these guidelines as the standard of care for each medical community, and there is a disservice to the medical community if they become legal fodder for medical tort. On the other hand, clinicians can receive a reference to a standardized set of principles for their practice; furthermore, the guidelines can be used as a "backdrop" as is done in this article by Popovich et al1 regarding compliance for ophthalmologic evaluation of patients with candidemia to evaluate how we use consultants and medical information.

The 2004 Infectious Disease Society of America Guidelines on Candidiasis recommend that all patients with candidemia undergo at least 1 dilated eye examination by an ophthalmologist.2 It is implied by this recommendation that candida can infiltrate the retina and invade the eye structures in a measurable number of cases, and this event has implications for treatment. Furthermore, it supports that this examination should be performed by those who are expert in ocular evaluations. There is some clinical support for these recommendations. First, it seems that the rate of eye involvement with candidemia is around 4% to 5%.3 Second, with evidence of endophthalmitis (either chorioretinitis or vitreitis), there are additional treatment decisions to be made. For instance, are vitrectomy and local instillation of amphotericin B indicated because of vitreitis?4 How good with limited penetration into ocular structures is echinocandin therapy for endophthalmitis?5 Should the length of therapy for eye involvement be extended for more than 2 weeks as is done with a routine candidemia? In fact, the guidelines recommend consideration of 6- to 12-week courses of antifungals for candida endophthalmitis.2 Furthermore, in previous animal experiments, I was always impressed how long it took to clinically sterilize ocular tissue of viable fungi with antifungal therapy.6 Does an ophthalmologist examination become a better benchmark for future management strategies? These questions must be considered in candida endophthalmitis because the preciseness of the answers has implications for the preservation of vision.

On the other side of the coin, it is likely, for a variety of reasons, that a great many candidemias are not followed by expert eye examinations. First, neutropenic patients do not reveal much eye findings without inflammatory cells, and seriously ill patients are more difficult to examine on the intensive care units and may take ophthalmologists outside their comfort zone. Second, one of the largest and best-studied prospective candidemia trials with eye examinations during and at the end of therapy by Mora-Duarte et al3 showed very little impact of the eye examinations on outcome with the use of amphotericin B or caspofungin as treatment.

In the interface between "to do and not to do" steps the Infectious Disease service. In the Popovich et al1 study, it clearly demonstrates that, in the presence of an Infectious Disease consult, the Infectious Diseases Society of America recommendations are more likely to be followed and formal expert eye examination performed. This extra evaluation is probably a good thing and demonstrates how the Infectious Disease consultant is so helpful in bringing together all services for the common good. Too often, Infectious Disease consultants see issues of high burden of organisms in tissue and protected body sites that complicate management of infections. The consequences for not making the extra effort of completeness can be devastating 1 case at a time. It was also illustrative in this case series that the prediction of eye involvement was not predicated on a particular Candida species such as Candida albicans in the blood. It simply came down to the philosophy of "look and ye shall find."

It is likely that many, many candidemic patients never receive a proper eye examination, and most do well without this added evaluation; however, it only takes an occasional complicated candidemia with eye involvement to appreciate the severe consequences. Generally, the Infectious Disease consultant will follow the recommendation of experts and get the eye examination. In fact, this behavior is one of the important values of an Infectious Disease consultant. He or she will commonly coordinate between evidence-based studies, recommendations/guidelines, personal experience, and pathophysiological principles. The Infectious Disease consultation rarely involves a technical procedure; however, it emphasizes clinical completeness and encourages diagnostic strategies to know the extent of infection, and that is the best way to safely and efficiently eliminate the invading pathogen and reduce host damage.

Back to Top | Article Outline


1. Popovich K, Malani PN, Kauffman CA, et al. Compliance with Infectious Disease Society of American Guidelines for Ophthalmologic Evaluation of Patients with Candidemia. Clin Infect Dis. 2007;15(4):254-256.
2. Pappas PG, Rex JH, Sobel JD, et al. Guidelines for treatment of candidiasis. Clin Infect Dis. 2004;38:161-189.
3. Mora-Duarte J, Betts R, Rotstein C, et al. Comparison of caspofungin vs. amphotericin B for invasive candidiasis. N Engl J Med. 2002;347:2020-2029.
4. Martinez-Vasquez C, Fernandez-Vlloa J, Bordon J, et al. Candida albicans endophthalmitis in brown heroin addicts: response to early vitrectomy preceded and followed by antifungal therapy. Clin Infect Dis. 1998;27:1130-1133.
5. Gauthier GM, Novk TM, Prince R, et al. Subtherapeutic ocular penetration of caspofungin and associated treatment failure in Candida albicans endophthalmitis. Clin Infect Dis. 2005;41:e27-e28.
6. Savani DV, Perfect JR, Cobo LM, et al. Penetration of new azole compounds into the eye and efficacy in experimental candida endophthalmitis. Antimicrob Agents Chemother. 1987;31:6-10.
© 2007 Lippincott Williams & Wilkins, Inc.