Recognition and Diagnosis of Cryptococcus gattii Infections in the United States

To the Editor: An outbreak of Cryptococcus gattii cryptococcosis has been ongoing in the US Pacific Northwest (PNW) since 1999 (1–3). In contrast to C. neoformans infections, which typically cause meningitis in HIV-infected persons, outbreak-associated C. gattii infections occur primarily in persons without HIV and often cause pneumonia (1–3). Sporadic, nonoutbreak-associated C. gattii infections often cause meningitis and have been reported outside the PNW (1–4). The prevalence of both types of C. gattii infection in the United States is unknown because diagnostic practices and awareness vary among physicians. 
 
Some reports indicate that patients with C. gattii infections may respond to treatment more slowly and relapse more frequently than patients with C. neoformans infections and, thus, may require more aggressive clinical management (5–8). Therefore, differentiation of C. gattii from C. neoformans infections may be necessary for optimal patient management. However, cryptococcal infections are often diagnosed by antigen testing, which cannot distinguish between C. gattii and C. neoformans, and differential agar necessary to distinguish species in culture (9) is not uniformly used in clinical laboratories. In addition to possible missed diagnoses caused by the atypical manifestation of outbreak-associated C. gattii, outbreak-associated and sporadic C. gattii infections in the United States are likely being misdiagnosed as C. neoformans infections. 
 
We conducted a survey of US infectious disease physicians to better understand the clinical approach to diagnosing cryptococcal infections, the relative regional frequency of C. gattii, and the capacity of clinical laboratories to differentiate cryptococcal species. To survey physicians, we used the Emerging Infections Network (EIN), a sentinel public health surveillance system of infectious disease clinicians that is supported by the Centers for Disease Control and Prevention and sponsored by the Infectious Diseases Society of America (10). During February–March 2011, web-based surveys were distributed by email or fax to the 1,342 EIN members, of whom 792 (59%) responded. 
 
Analysis was restricted to 286 (36%) respondents (representing 43 states) who treated a cryptococcosis patient during the past year. We compared answers from respondents in the 4 US census regions (Table; Technical Appendix). Results were analyzed by using SAS version 9.2 (SAS Institute Inc., Cary, NC). 
 
 
 
Table 
 
Physician responses, by US region, to a survey about cryptococcosis, February–March 2011* 
 
 
 
The approximate number of reported physician consults for cryptococcosis was similar among respondents from all regions (Table). More respondents from the West (40%), compared with the South (21%), Midwest (22%), and Northeast (19%), reported that >25% of their cryptococcosis patients had pneumonia; this finding may reflect the higher prevalence of outbreak-associated C. gattii infections in the West (1–3). The percentage of respondents who treated cryptococcosis patients without known risk factors for infection (including HIV) during the past 5 years was also higher in the West (53%) compared with other regions (Table). 
 
Most (93%) respondents reported that they were aware of the C. gattii outbreak. However, only 63% of respondents consider Cryptococcus species a factor of interest during diagnosis or treatment, and 54% would consider C. gattii as a differential diagnosis for pneumonia in a patient from the PNW. Although awareness of C. gattii appears high, recognition of infection may be delayed when diagnostic plans do not include species identification. 
 
Of the respondents, 94% reported that they most often use the cryptococcal antigen test for diagnosis, although 73% of respondents report that they commonly request a culture. Furthermore, 76 (27%) of respondents report using a combination of tests (cryptococcal antigen, microscopy, histopathology) that does not include culture. Tests that do not differentiate between cryptococcal species represent missed opportunities for diagnosis of C. gattii infections. When respondents were asked if their clinical laboratory could differentiate C. neoformans from C. gattii isolates, 131 (46%) responded “yes, either routinely or when requested”; 68 (24%) responded “no”; 87 (30%) did not know. When we excluded respondents who did not know, only 66% of respondents from the West indicated that their laboratory could differentiate species. This finding is concerning because outbreak-associated C. gattii is clearly endemic to the region. A better understanding of which laboratories perform this service and which send specimens to a reference laboratory will help identify where additional capacity is needed. 
 
A lower percentage of respondents from the Northeast (10%), Midwest (5%), and South (3%), compared with those from the West (44%), reported having ever consulted on a case of C. gattii infection. This may reflect a low incidence of C. gattii infections in these regions, or it may be a result of decreased clinical suspicion for C. gattii infections outside the PNW. 
 
Results from this study suggest that although most EIN members are aware of C. gattii and the ongoing outbreak in the PNW, missed opportunities for diagnosis still exist. To understand the true incidence of C. gattii inside and outside the PNW, vigilance among physicians nationwide is necessary. Clinicians and laboratorians should be aware of the need to obtain specimens for culture and of the need to develop methods to differentiate cryptococcal species. An accurate diagnosis of cryptococcosis cases in the United States will lead to a better understanding of the epidemiology and incidence of C. gattii in this country and may result in improved treatment.

. In contrast to C. neoformans infections, which typically cause meningitis in HIV-infected persons, outbreakassociated C. gattii infections occur primarily in persons without HIV and often cause pneumonia (1-3). Sporadic, nonoutbreak-associated C. gattii infections often cause meningitis and have been reported outside the PNW (1-4). The prevalence of both types of C. gattii infection in the United States is unknown because diagnostic practices and awareness vary among physicians.
Some reports indicate that patients with C. gattii infections may respond to treatment more slowly and relapse more frequently than patients with C. neoformans infections and, thus, may require more aggressive clinical management (5)(6)(7)(8). Therefore, differentiation of C. gattii from C. neoformans infections may be necessary for optimal patient management. However, cryptococcal infections are often diagnosed by antigen testing, which cannot distinguish between C. gattii and C. neoformans, and differential agar necessary to distinguish species in culture (9) is not uniformly used in clinical laboratories. In addition to possible missed diagnoses caused by the atypical manifestation of outbreakassociated C. gattii, outbreakassociated and sporadic C. gattii infections in the United States are likely being misdiagnosed as C. neoformans infections.
We conducted a survey of US infectious disease physicians to better understand the clinical approach to diagnosing cryptococcal infections, the relative regional frequency of C. gattii, and the capacity of clinical laboratories to differentiate cryptococcal species. To survey physicians, we used the Emerging Infections Network (EIN), a sentinel public health surveillance system of infectious disease clinicians that is supported by the Centers for Disease Control and Prevention and sponsored by the Infectious Diseases Society of America (10). During February-March 2011, web-based surveys were distributed by email or fax to the 1,342 EIN members, of whom 792 (59%) responded.
The approximate number of reported physician consults for cryptococcosis was similar among respondents from all regions (Table). More respondents from the West (40%), compared with the South (21%), Midwest (22%), and Northeast (19%), reported that >25% of their cryptococcosis patients had pneumonia; this fi nding may refl ect the higher prevalence of outbreak-associated C. gattii infections in the West (1- cryptococcosis patients without known risk factors for infection (including HIV) during the past 5 years was also higher in the West (53%) compared with other regions (Table). Most (93%) respondents reported that they were aware of the C. gattii outbreak. However, only 63% of respondents consider Cryptococcus species a factor of interest during diagnosis or treatment, and 54% would consider C. gattii as a differential diagnosis for pneumonia in a patient from the PNW. Although awareness of C. gattii appears high, recognition of infection may be delayed when diagnostic plans do not include species identifi cation.
Of the respondents, 94% reported that they most often use the cryptococcal antigen test for diagnosis, although 73% of respondents report that they commonly request a culture. Furthermore, 76 (27%) of respondents report using a combination of tests (cryptococcal antigen, microscopy, histopathology) that does not include culture. Tests that do not differentiate between cryptococcal species represent missed opportunities for diagnosis of C. gattii infections. When respondents were asked if their clinical laboratory could differentiate C. neoformans from C. gattii isolates, 131 (46%) responded "yes, either routinely or when requested"; 68 (24%) responded "no"; 87 (30%) did not know. When we excluded respondents who did not know, only 66% of respondents from the West indicated that their laboratory could differentiate species. This fi nding is concerning because outbreakassociated C. gattii is clearly endemic to the region. A better understanding of which laboratories perform this service and which send specimens to a reference laboratory will help identify where additional capacity is needed.
A lower percentage of respondents from the Northeast (10%), Midwest (5%), and South (3%), compared with those from the West (44%), reported having ever consulted on a case of C. gattii infection. This may refl ect a low incidence of C. gattii infections in these regions, or it may be a result of decreased clinical suspicion for C. gattii infections outside the PNW.
Results from this study suggest that although most EIN members are aware of C. gattii and the ongoing outbreak in the PNW, missed opportunities for diagnosis still exist. To understand the true incidence of C. gattii inside and outside the PNW, vigilance among physicians nationwide is necessary. Clinicians and laboratorians should be aware of the need to obtain specimens for culture and of the need to develop methods to differentiate cryptococcal species. An accurate diagnosis of cryptococcosis cases in the United States will lead to a better understanding of the epidemiology and incidence of C. gattii in this country and may result in improved treatment.  (1,2). We report a case of coccidioidal endophthalmitis in an immunocompetent person.

Coccidioidal
In October 2010, a 55-year-old white man in Santa Clarita, California, had severe pneumonia, drenching sweats, and an associated 25-pound weight loss. Three weeks later, when his symptoms had nearly resolved, the man reported having scratched his left eye with his eyeglasses and subsequent development of increasing redness, pain, and progressive vision loss (from 20/10 to 20/60 without correction).
In November 2010, the man sought the care of an ophthalmologist, who noted panuveitis of the left eye. Laboratory testing was performed: the erythrocyte sedimentation rate was 49 mm/h (reference 0-22 mm/h), and test results were negative for human leukocyte antigen B27, angiotensinconverting enzyme, rapid plasma reagin, and antinuclear antibody.
The patient was started on topical corticosteroids and escalated to highdose prednisone soon thereafter without improvement. Pain continued to increase in his left eye, and visual acuity declined to hand motion only.
Thus, in February 2011, the patient was referred to our institution, where an ocular ultrasound showed vitreous opacities (Figure). He underwent vitrectomy with intravitreal injection of empiric antimicrobial drugs, including voriconazole. Aqueous fl uid obtained intraoperatively grew mold, and the patient was admitted to the hospital for systemic antifungal therapy.
The patient's history was unremarkable except for avid mountain biking in the Central Valley of California. His physical examination was notable for left visual acuity limited to hand motion only, limited extraocular movement, conjunctival injection, and hypopyon. His HIV test result was negative. Computed tomography (CT) scanning of the chest showed micronodules in the right upper lobe, suggesting previous pulmonary coccidioidal infection. Intravenous voriconazole (4 mg/kg every 12 hours) was administered along with daily intravitreal injections of voriconazole while the patient was hospitalized. Results for coccidioidal antibody testing were positive by enzyme immunoassay and immunodiffusion but negative by serum complement fi xation. Nucleic acid hybridization testing of aqueous fl uid cultures identifi ed Coccidioides spp. Results of a CT brain scan, lumbar puncture, and bone scan were normal. After 1 week of hospitalization, the patient was discharged on oral voriconazole (4 mg/kg 2×/day). Because of transaminitis, the patient was transitioned to fl uconazole (800 mg/day) 4 weeks later. He underwent 13 subsequent intravitreal injections of amphotericin and voriconazole. Eleven months after discharge, the patient's best-corrected visual acuity was 20/25, and his ocular media were clear and without any lesions.
Coccidioidomycosis often goes undetected because up to 60% of affected patients are asymptomatic (3). When signs and symptoms are present, they vary from subclinical infection to acute pneumonia to disseminated disease (3). The rate of extrapulmonary complications is estimated at 0.5% of infections in white persons, but such complications may occur in Figure. Ocular ultrasound demonstrating hyperechoic, punctate opacities (arrows) within the vitreous chamber (X) of a patient with coccidioidomycosis, California, USA.