Trichosporon species are ubiquitous fungi that have been rarely reported to cause opportunistic infections in the immunocompromised host.1–5 Trichosporons are found in soil and colonize the skin and respiratory and gastrointestinal tracts in humans.6 Patients most susceptible to Trichosporon infection include those receiving immunosuppressive medications and those with either absolute or functional neutropenia. Invasive Trichosporon beigelii has been described most commonly.7–9 A recent case study of 2 chronic granulomatous disease (CGD) patients infected with Trichosporon inkin also has been published.2 In CGD, affected individuals’ neutrophils fail to mount an adequate oxidative burst toward encapsulated and catalase-positive bacteria, as well as a variety of fungi, rendering these patients susceptible to serious opportunistic infections including trichosporosis.
Reports of invasive Trichosporon pullulans infection are rare but are increasing in frequency. The first case was described in a renal transplant patient receiving immunosuppression with corticosteroids, azathioprine and cyclosporin.10 The majority of subsequent cases have involved patients undergoing cancer chemotherapy.1 These patients shared common features of indwelling venous catheterization, prior treatment with broad spectrum antibiotics and neutropenia. In addition, T. pullulans has been reported in a patient with adult T cell leukemia5 and has been isolated from the oropharynx of a patient with acquired immunodeficiency syndrome.3 Recently T. pullulans was reported in 2 cases of CGD.11 The methods used to identify the causative organism in that study, however, have been called into question.12,13
In this case report, we describe T. pullulans infection in a CGD patient undergoing immunosuppressive therapy for severe, Crohn's-like inflammatory bowel disease (IBD). Granulomatous colitis is a known complication of CGD, with estimated prevalence rates as high as 60%.14 For patients with severe refractory colitis, substantial immunosuppression is sometimes considered.15 It is likely that intensive antiinflammatory regimens place CGD patients at higher risk of serious opportunistic infections.
A 19-year-old young man with X-linked CGD, diagnosed at age 3 months, was hospitalized for respiratory distress. At the time of presentation, he was febrile to 38.4°C and had been experiencing progressive shortness of breath with nonproductive cough. Initial oxygen saturations were 88% on room air. He also complained of some nausea, loose stools and abdominal pain. Physical examination was significant only for decreased breath sounds in the right lower lung base, consistent with a history of right lower lobectomy and some diffuse abdominal tenderness. Initial laboratory evaluation revealed a C-reactive protein level of 25.9 mg/dL and a white blood cell count of 3730 cells/μL, with a differential of 58% segmented neutrophils, 25% band forms, 16% lymphocytes and 1% monocytes. Serum chemistries were noncontributory.
The patient's chronic disease course had been complicated by gastrointestinal symptoms beginning at age 2 years, including loose bloody stools, multiple chronic perirectal abscesses and fistulae-in-ano requiring surgical interventions. Most recently, he had been diagnosed with Crohn-like IBD, for which he had been treated with an immunosuppressive regimen of azathioprine, prednisone and mesalamine. In the 2 months before admission, infliximab had also been initiated for severe refractory fistulae-in-ano. This therapy was discontinued when he began to develop fevers in the weeks leading up to admission. His infectious disease history included right lower lobectomy for presumed fungal pneumonia 3.5 years prior, for which he had been treated with a 2-week course of liposomal amphotericin and placed on daily itraconazole prophylaxis. Two years before, he also was found to have Aspergillus-specific IgG and IgA levels 5.5 and 4.8 SD above the mean (normal range, <2 SD), respectively, with an IgM level within the normal range. Serum Aspergillus antigen enzyme immunoassay and galactomannan levels, however, were normal at the time of admission. Three days before admission, his prophylactic itraconazole regimen of 200 mg twice daily was changed from capsule to liquid form in an attempt to improve absorption. Although serum itraconazole levels 2 weeks before had been essentially undetectable, a level drawn at the time of admission had improved to 1.2 μg/mL (therapeutic range, 1.9–2.2 μg/mL).
During hospitalization, the patient was treated with empiric broad spectrum antibiotics, metronidazole, voriconazole and liposomal amphotericin, and his immunosuppressive regimen was markedly reduced. Admission blood cultures were negative for aerobic or anaerobic bacteria. Bronchoalveolar lavage (BAL) was negative for Aspergillus, Pneumocystis carinii, Mycoplasma, acid-fast bacilli, Chlamydia trachomatis and viruses. Within 2 days, yeast was detected growing from a central venous catheter (CVC) culture obtained on admission, eventually identified as T. pullulans (see Laboratory Methods). Caspofungin was added to the antimicrobial regimen, and the patient's CVC was removed. The patient remained febrile and hypoxic, however, progressing to respiratory failure and fatal septic shock. T. pullulans was recovered from multiple sites, including 2 CVC cultures, a peripheral blood culture, 2 arterial catheter cultures and BAL culture.
Blood culture specimens were obtained into BacT/ALERT culture medium bottles (bioMérieux, Durham, NC) for routine cultures, or into yellow-topped Isolator tubes (Wampole Laboratories, Princeton, NJ) for fungal cultures. BacT/ALERT cultures were incubated at 35–36°C and monitored using automated colorimetry. Positive cultures were Gram-stained; those demonstrating yeast were plated onto blood, eosin-methylene blue, or chocolate (CHA) agar tissue culture dishes and incubated at 35°C. Colonies were then transferred onto cornmeal (CMA) agar and incubated at room temperature, and morphology was determined by visual inspection. Fungal cultures were directly plated onto CHA and incubated at 35°C or into brain-heart infusion mediu and incubated at 30°C and identified on CMA. BAL specimens were inoculated onto blood agar, eosin-methylene blue agar and CHA, as well as on inhibitory mold agar (BD Diagnostic Systems, Sparks, MD) at 30°C and identified on CMA. Fungal specimens for a Yeast Biochemical Card (bioMérieux) analysis were grown on Sabouraud agar at 30°C.
A Yeast Biochemical Card from both blood and BAL samples resulted in glycosylation patterns consistent with T. pullulans of 93–99% certainty. Morphology was determined for every positive culture with direct visualization on CMA. Initial cultures were analyzed by 2 experienced technicians, who confirmed the presence first of pseudohyphae and then true hyphae and absence of arthroconidia, for up to 5 days of incubation. Later cultures were inspected and verified as morphologically consistent with the initial identification. No cryptococcal forms were identified from any of the specimens. The initial blood culture obtained from the patient's CVC became positive for yeast after 39 hours of incubation. Later cultures became positive at times ranging from 2.3 to 4.7 days. Cultures from BAL were positive for yeast within 2 days of collection.
The specific neutrophilic defects of chronic granulomatous disease place patients at increased risk for infection from a wide variety of organisms. Phagocytes control the proliferation of yeasts in vivo through both oxidative and nonoxidative mechanisms. Defects in neutrophilic free radical generation result in a significant deficit in the innate immune response to fungi.16 For some high risk CGD patients, therefore, antifungal chemoprophylaxis is occasionally instituted.
Itraconazole has been shown to be effective for long term prophylaxis against fungal infection in CGD.17 A known limitation of oral itraconazole, however, is variable bioavailability, particularly in capsule form.18 Bioavailability can be compromised further by the malabsorption characteristic of IBD. Although the serum itraconazole level in our patient improved with the liquid preparation, it remained subtherapeutic and may therefore have impacted the efficacy of prophylaxis. Furthermore T. pullulans fungemia has been reported in neutropenic patients receiving itraconazole prophylaxis.1
Few reports and no controlled trials exist regarding the efficacy of broad spectrum antifungals such as amphotericin and caspofungin against T. pullulans. One small study demonstrated minimum inhibitory concentrations (MICs) for amphotericin B, liposomal amphotericin and fluconazole against 2 strains of T. pullulans that were below defined resistance breakpoints.19 However, MICs for other Trichosporon species were also lower than reported elsewhere.4,8,9 Another study showed resistance of 1 isolate of T. pullulans to amphotericin B and fluconazole.20 The medical literature pertaining to other invasive Trichosporon infections (primarily T. beigelii) suggests that MICs of amphotericin are usually supratherapeutic,1,4,9 and that in vitro susceptibilities to newer triazoles are generally greater.4,8,9,20,21 One article did report success with voriconazole in a case of disseminated T. asahii.7 The clinical response of most Trichosporon species to antifungals is typically poor, however, and does not necessarily correlate to in vitro susceptibility results.2 Finally extrapolation of in vitro data for other Trichosporon species to in vivo infection with T. pullulans might not be valid. On the basis of the available evidence, we conclude that amphotericin monotherapy likely does not represent optimal treatment of invasive T. pullulans. Additionally achieving clinical efficacy may be difficult regardless of the antifungal agent used. Our patient failed to improve despite early, broad spectrum antifungal therapy, including voriconazole.
One obvious question is the role immunosuppressants might have played in placing this patient at increased risk for opportunistic infection. Accepted regimens for treatment of severe IBD in non-CGD patients include combination therapy with a variety of agents including corticosteroids and immunomodulators. Our patient had chronic, multiple perianal fistulae that eventually became refractory to surgery and low dose steroid therapy. For such patients, infliximab, an anti-tumor necrosis factor-α monoclonal antibody indicated for treatment of Crohn's disease, is becoming increasingly used with good response. Unfortunately infliximab has been associated with serious infectious complications including reactivation tuberculosis and invasive fungal infections.22 The mechanism by which infliximab might predispose to fungal infection is not clear. Binding of hyphal antigen to the Toll-like receptor on phagocytes initiates a signaling cascade that results in cellular activation as well as generation of molecules that up-regulate T cell activity.16 It is theoretically possible that infliximab interferes with this signaling cascade. Corticosteroids further predispose to invasive fungal infection through diminished T cell proliferation and activity. Fever and infection in this patient developed within weeks of initiating infliximab, suggesting a temporal, although not necessarily causal, relationship. Furthermore given that trichosporosis has not been reported previously in patients receiving tumor necrosis factor-α antagonist therapy, it remains to be seen whether CGD patients represent a population at particular risk. Given the importance of oxidative mechanisms in fungal killing and the inability of CGD patients to generate a sufficient oxidative response, it is plausible that further suppression of nonoxidative neutrophil and T cell function affects these patients disproportionately. In general, it is likely that the combination of immunosuppressants placed our patient at additional risk of serious fungal disease through suppression of both innate and adaptive immunity.
With regard to the previous report of T. pullulans in 2 CGD patients,11 a recently published response appropriately questioned the methods used for identification of the organism.12,13 Specifically the case report apparently relied solely on biochemical identification using an automated system for determining yeast sugar assimilation patterns. Although our microbiology laboratory uses the same automated system for initial identification, several additional factors would seem to support a diagnosis of T. pullulans infection in our patient: (1) as recommended by Holland et al,12,13 our laboratory routinely uses morphologic analysis by experienced technicians for confirmation of all yeast cultures before a final report is issued; (2) multiple specimens obtained over several days and from various sites continued to produce an organism consistent with T. pullulans; (3) original cultures were positive for fungus at 39 hours, making it unlikely to have been a laboratory contaminant; (4) unlike the CGD patients described by Moylett et al, the severity of the infection observed in our patient and the lack of response to multidrug antifungal therapy are more consistent with previously reported T. pullulans infections; finally (5) the lack of other identifiable organisms strongly suggests T. pullulans was responsible for our patient's fatal infection. Unfortunately specimens from our patient had been discarded after 3 months, precluding further confirmation of the initial identification.
Given the direct and indirect evidence as presented, we suggest that this case represents the first significant infection in a CGD patient attributable to T. pullulans. We support recommendations in the medical literature calling for implementation of more meticulous methods for the correct identification of this organism. Finally we reiterate that aggressive immunosuppressive treatment of severe colitis places CGD patients at increased risk for life-threatening opportunistic infection.
We thank Dr Ronald Ferdman for critical review of the manuscript, Hisae Nakayama and Aida Mangahis for assistance with the Laboratory Methods section and the Fungus Testing Laboratory at the University of Texas Health Science Center at San Antonio for review of and comments regarding the Laboratory Methods section.
1. Krcmery V Jr, Oravcova E, Spanik S, et al. Nosocomial breakthrough fungaemia during antifungal prophylaxis or empirical antifungal therapy in 41 cancer patients receiving antineoplastic chemotherapy: analysis of aetiology risk factors and outcome. J Antimicrob Chemother
2. Wynne SM, Kwon-Chung KJ, Shea YR, et al. Invasive infection with Trichosporon inkin
in 2 siblings with chronic granulomatous disease
. J Allergy Clin Immunol
3. Neves RP, deq Ueiroz Cavalcanti MA, Chaves GM, et al. Trichosporon pullulans
(Lidner) Diddens and Lodder isolated from the oral cavity of AIDS patient. Braz J Microbiol
4. Paphitou NI, Ostrosky-Zeichner L, Paetznick VL, et al. In vitro antifungal susceptibilities of Trichosporon
species. Antimicrob Agents Chemother
5. Shigehara K, Takahashi K, Tsunematsu K, et al. A case of Trichosporon pullulans
infection of the lung with adult T-cell leukemia. Jpn J Med
6. Gueho E, Improvisi L, de Hoog GS, et al. Trichosporon
on humans: a practical account. Mycoses
7. Fournier S, Pavageau W, Feuillhade M, et al. Use of voriconazole to successfully treat disseminated Trichosporon asahii
in a patient with acute myeloid leukaemia. Eur J Clin Microbiol Infect Dis
8. Wolf DG, Falk R, Hacham M, et al. Multidrug-resistant Trichosporon asahii
infection of nongranulocytopenic patients in three intensive care units. J Clin Microbiol
9. Walsh TJ, Melcher GP, Rinaldi MG, et al. Trichosporon beigelii
, an emerging pathogen resistant to amphotericin B. J Clin Microbiol
10. Hughes CE, Serstock D, Wilson BD, et al. Infection with Trichosporon pullulans
. Ann Intern Med
11. Moylett EH, Chinen J, Shearer WT. Trichosporon pullulans
infection in 2 patients with chronic granulomatous disease
: an emerging pathogen and review of the literature. J Allergy Clin Immunol
12. Holland SM, Shea YR, Kwon-Chung J. Regarding “Trichosporon pullulans
infection in 2 patients with chronic granulomatous disease
.”J Allergy Clin Immunol
13. Moylett EH, Chinen J, Wickes B, et al. Author reply. J Allergy Clin Immunol
14. Marciano BE, Rosenzweig SD, Kleiner DE, et al. Gastrointestinal involvement in chronic granulomatous disease
15. Rosh JR, Tang HB, Mayer L, et al. Treatment of intractable gastrointestinal manifestations of chronic granulomatous disease
with cyclosporine. J Pediatr
16. Romani L. Immunity to fungal infections. Nat Rev Immunol
17. Gallin JI, Alling DW, Malech HL, et al. Itraconazole to prevent fungal infections in chronic granulomatous disease
. N Engl J Med
18. Glasmacher A, Prentice A, Gorschluter M, et al. Itraconazole prevents invasive fungal infections in neutropenic patients treated for hematologic malignancies: evidence from a meta-analysis of 3,597 patients. J Clin Oncol
19. Carillo-Munoz A-J, Quindos G, Tur C, et al. Comparative in vitro antifungal activity of amphotericin B lipid complex, amphotericin B and fluconazole. Chemotherapy
20. McGinnis MR, Pasarell L, Sutton DA, et al. In vitro activity of voriconazole against selected fungi. Med Mycol
21. Falk R, Wolf DG, Shapiro M, et al. Multidrug-resistant Trichosporon asahii
isolates are susceptible to voriconazole. J Clin Microbiol
22. De Rosa FG, Shaz D, Campagna AC, et al. Invasive pulmonary aspergillosis soon after therapy with infliximab, a tumor necrosis factor-α-neutralizing antibody: a possible healthcare-associated case? Infect Control Hosp Epidemiol