Candida in Solid Organ Transplant Recipients

Infections due to Candida spp are the most common invasive fungal infections (IFIs) among solid organ transplant recipients, accounting for over half of all IFIs in this population (1). Invasive candidiasis occurs earlier than other invasive mycoses, generally within the first 3 months following transplantation, and is viewed as a classic nosocomial infection (2, 3). However, a substantial number of cases of invasive candidiasis, especially among liver and small bowel transplant recipients, occur well beyond this traditional period of risk (1-3). In contrast to stem cell transplant recipients, where non-albicans Candida spp dominate, the distribution of Candida spp causing invasive disease among organ recipients is similar to the distribution of organism seen among nontransplant hospitalized patients (1, 4). Specifically, Candida albicans is the dominant invasive pathogen, accounting for approximately 50% of isolates. The non-albicans spp. including Candida glabrata, C. tropicalis and C. parapsilosis constitute the majority of other isolates. C. krusei and C. guilliermondii, isolates which are more common among stem cell transplant recipients are far less common among organ recipients (5). The risk of developing invasive candidiasis posttransplantation is related to a number of factors, including the type of transplant and surgical anastomosis (6). For instance, among liver transplant recipients, a choledochojejunostomy is associated with a higher risk of invasive candidiasis compared to a choledocho–choledocho anastomosis (7). Similarly, among pancreas transplant recipients, enteric drainage is associated with a higher risk of invasive candidiasis than bladder drainage (8). In addition, there are other well-established risks that enhance the likelihood of developing invasive candidiasis. These include acute renal failure, recent CMV infection, primary graft failure, early surgical re-exploration and early colonization with Candida spp (9). Overall rates of invasive candidiasis have increased slightly over time based on the most recent data (1). The reason for this trend is unclear, but perhaps reflects the fact that relatively few organ recipients routinely receive perioperative antifungal prophylaxis, and that the period of risk is measured in years following transplantation. A definitive diagnosis of invasive candidiasis is dependent on recovery of an organism from a sterile body site, such as the bloodstream, pleural fluid, intra-abdominal fluid or abscess material. Unfortunately, cultures, especially blood cultures, are an insensitive means of identifying patients with invasive candidiasis. Even with newer blood culture techniques, the overall sensitivity of blood cultures for the isolation of Candida spp is estimated to be 70% (10). Therefore, the development of nonculture based diagnostic methodology is especially important. Presently, there are several FDA-approved assays available, but their use has been limited in clinical practice. Among these, the 1–3 β D glucan assay is probably the most reliable, with the sensitivity and specificity of 70 and 87%, respectively, among patients who have proven invasive candidiasis (11-13). At present, this assay is only approved as an adjunct to the diagnosis of invasive candidiasis. The Platellia Aspergillus ELISA is an FDA-approved test for invasive aspergillosis (14). The assay identifies the presence of galactomannan, a cell wall component of many fungi, but this test is not approved as a diagnostic assay for IC. Other newer diagnostic assays, including PCR-based multiplex assays, remain in development. Identification of Candida isolates to the species level is critically important in selecting antifungal therapy, and to a lesser extent, predicting outcome. The germ tube test is reliable and an inexpensive means of identifying Candida albicans. The recently developed peptide nucleic acid fluorescent in situ hybridization assay (PNA-FISH) can also reliably distinguish C. albicans from non-albicans spp (15). This assay is more rapid than the germ tube, but is considerably more expensive and provides only modest improvement in sensitivity. Chromogenic agar, a specialized media for Candida isolation and identification, is easily used and readily distinguishes C. albicans, C. tropicalis and C. krusei based on distinctive pigments produced by each of these species (16). Susceptibility testing for all clinically significant Candida isolates is not practical for many centers. Generally, antifungal susceptibility can be predicted on the basis of species and local epidemiology (see Table 1). Antifungal susceptibility testing is suggested for clinically significant C. glabrata isolates, in clinical situations where azole resistance is strongly suspected, and in cases of treatment failure (17). The treatment of invasive candidiasis among organ transplant recipients is similar to treatment among most other patients based on the recently published 2009 IDSA guidelines for the treatment of candidiasis (17). There are no randomized studies for the treatment of invasive candidiasis among organ transplant recipients, thus, the therapeutic approach to these patients in general is based on large randomized studies in a heterogeneous group of patients which include organ transplant recipients as a small component. A summary of the treatment recommendations is described in Table 1, and a brief description of the relevant antifungal agents is discussed further. Amphotericin B: Most experience with amphotericin B (AmB) is with a deoxycholate preparation (AmB-d). There are three lipid formulations of AmB (LFAmB) including amphotericin B lipid complex (ABLC), amphotericin B colloidal dispersion (ABCD) and liposomal amphotericin B (LAmB). These agents possess the same broad spectrum activity as AmB-d, but differ in terms of pharmacokinetic properties and rates of treatment-related adverse events; they are each less nephrotoxic than AmB-d (18-20). For most forms of invasive candidiasis, the typical intravenous dose of AmB-D is 0.5–0.7 mg/kg daily (21-24). Doses as high as 1 mg/kg daily should be considered for patients with less susceptible species such as C. glabrata and C. krusei. The typical adult dosage of LFAmB is 3–5 mg daily (19, 20, 25). Triazoles: Fluconazole, itraconazole, voriconazole and posaconazole demonstrate similar activity against most Candida spp, although each is less active against C. glabrata and C. krusei (26). The azoles have important pharmacokinetic differences. For example, fluconazole is the only agent concentrated in the urine, and it also achieves cerebrospinal fluid (CSF) concentrations at least 50% of that in serum (27). Voriconazole achieves excellent CSF and vitreous concentrations, has excellent oral bioavailability, but does not achieve meaningful concentrations in urine (28-30). Itraconazole and posaconazole demonstrate excellent in vitro activity against most Candida spp, but both require oral administration, have unpredictable serum levels and neither offer significant therapeutic advantages over fluconazole or voriconazole. All azoles demonstrate significant drug-drug interactions and special attention must be given to dosing adjustments for co-administered drugs, especially the calcineurin inhibitors (31). Echinocandins: Caspofungin, micafungin and anidulafungin are only available as parenteral preparations, but each have excellent in vitro activity against most Candida spp including C. glabrata and C. krusei (32). C. parapsilosis and C. guilliermondii demonstrate less in vitro susceptibility to the echinocandins (33). The echinocandins have few side effects, none require dose adjustment for renal insufficiency or dialysis, and are uncommonly associated with drug–drug interactions (32). A dose reduction from 50 mg to 35 mg/daily is advised for adult patients with significant hepatic impairment who are receiving caspofungin. Children with moderate hepatic insufficiency should have their daily dose reduced by 30%. Therapeutic drug monitoring (TDM): For patients receiving prolonged courses of voriconazole or posaconazole, therapeutic drug monitoring is recommended by some experts, but there is no consensus on this topic (34, 35). The main purpose of TDM is to potentially avoid toxicity that may be observed at higher serum concentrations and to reduce the risk of treatment failure at lower concentrations (36). Candidemia: The selection of any particular agent for the treatment of candidemia should take into account recent azole exposure, a history of intolerance to an antifungal agent and the dominant Candida spp and current susceptibility data in a particular location (37). In addition, the severity of illness, relevant co-morbidities and evidence of metastatic involvement to other organs systems are important considerations. Early initiation of therapy is critical to the successful treatment of candidemia (38, 39). Based on abundant data from clinical trials, fluconazole remains the standard therapy for selected patients with candidemia (21-24). Fluconazole is considered first-line among patients with mild to moderate illness, no recent azole exposure, and in whom C. glabrata is unlikely (17). The echinocandins demonstrate rapid fungicidal activity against all Candida spp, and have demonstrated approximately 75% success in randomized clinical trials (40-42). Due to their efficacy, favorable safety profile and very few drug-drug interactions, the echinocandins are favored as initial therapy for patients with a recent history of azole exposure, moderately severe to severe illness, a history of allergy or intolerance to the azoles, or high risk for infection due to C. krusei or C. glabrata (17). Following a short course of intravenous echinocandin therapy (3–5 days), fluconazole is a reasonable choice for step-down therapy, provided that the organism is predictably susceptible to fluconazole (C. albicans, C. parapsilosis and C. tropicalis) and the patient is clinically stable (17). There are reports of decreased susceptibility of C. parapsilosis to the echinocandins, but the clinical significance of this is unknown. However, it may be prudent to choose an alternative to an echinocandin as first line therapy for invasive infections due to this organism (43, 44). The echinocandins are sufficiently similar that they are interchangeable. Voriconazole is approved for treatment of candidemia, but clinical trials have not demonstrated a particular advantage compared to other agents (45). The role of voriconazole for the treatment of candidemia is limited to patients who have an infection due to a fluconazole-resistant organism, and who are ready for transition to oral therapy. Examples include infections with C. krusei and fluconazole-resistant, voriconazole-susceptible C. glabrata (17). The role for LFAmB is limited due to potential nephrotoxicity, and is generally reserved for individuals who are intolerant of or refractory to other forms of therapy. Removal of central venous catheters, when feasible, is strongly recommended among patients with candidemia (46). There is debate as to the necessity of removing all intravascular catheters in these patients (47), but most experts agree that removal of these devices is appropriate, particularly if the source of candidemia is unclear. In addition, all patients with candidemia should have a dilated funduscopic exam and repeat blood cultures at 48 to 72 h intervals until blood cultures are negative for Candida spp. The duration of therapy for treatment of candidemia without metastatic complications is (generally 2 weeks following clearance of Candida from the bloodstream and resolution of symptoms attributable to candidemia (17). The treatment of candidemia in neutropenic organ transplant recipients differs somewhat from nonneutropenic patients in that there is a greater emphasis on the use echinocandins and LFAmB (48, 49). Most clinicians prefer these agents over fluconazole based on persistent concerns that a fungicidal agent such as (echinocandin or LFAmB) is preferred over a fungistatic agent (fluconazole or voriconazole), although there are few data to support this approach. Urinary tract infections: In the absence of fever or other evidence of systemic infection, candiduria in the organ transplant recipient does not generally necessitate treatment (50, 51). There are no prospective and comparative trials comparing treatment versus nontreatment in this group, thus treatment in this setting is largely driven by anecdotal experience and personal preference. For purposes of determining selection of an agent and duration of therapy, it is helpful to divide organ recipients with candiduria into asymptomatic and symptomatic categories. Treatment of asymptomatic candiduria is generally discouraged unless the patient is undergoing a urologic procedure or is neutropenic (17). Imaging of the kidneys and collecting system is prudent to exclude abscess, fungus ball or urologic abnormality. Among symptomatic patients with candiduria and suspected disseminated candidiasis, it is appropriate to treat as for candidemia (see earlier). For patients with cystitis due to a fluconazole-susceptible Candida spp, oral fluconazole 200 to 400 mg (pediatric dosing 3–6 mg/(kg dose)) daily for 2 weeks is advisable (17). For patients with fluconazole-resistant organisms, LFAmB or oral flucytosine 25 mg/kg four times daily are recommended (17). AmB-d bladder irrigation is generally not recommended, but might be useful for patients with fluconazole-resistant Candida spp, especially C. glabrata (52). For patients with pyelonephritis, treatment with fluconazole or an echinocandin for at least 2 weeks, as for patients with candidemia, is recommended (17). Pulmonary candidiasis: Isolation of Candida spp from the respiratory tract rarely indicates invasive candidiasis and generally is not treated with antifungal therapy (53-55). An exception exists for lung transplant recipients in whom anastomotic tracheobronchitis due to Candida is a concern. Evidence of Candida tracheobronchitis is based on visual inspection and histologic confirmation, usually with a positive culture from an appropriate specimen. Selection of a specific agent could be based on the same principles as for selecting an agent for treatment of candidemia. There are no specific studies to guide duration of therapy, but it is reasonable to continue treatment until there is bronchoscopically confirmed clinical resolution of the infection. Antifungal prophylaxis against candidiasis in organ transplantation is very controversial. Studies are usually single center with a small number of patients. Few are well-designed and controlled studies have been conducted only in liver transplant recipients. These shortcomings make it difficult to generalize conclusions among different transplanted organs and different centers. Identifying patients at the highest risk of infection is crucial to the development of effective approaches to antifungal prophylaxis. The major points that need to be addressed when deciding if antifungal prophylaxis is warranted and choosing the best prophylactic regime include: (1) general prophylaxis versus targeted prophylaxis; (2) selection of an agent; and (3) the duration of prophylaxis. The prophylactic approach implies that an antifungal agent is administered to all transplant recipients, while targeted prophylaxis applies to the use of an antifungal agent in a subgroup of transplant recipients with predisposing conditions that place them at higher risk of developing invasive candidiasis. The risk factors for invasive candidiasis have been previously described in this document, and some vary according to the transplanted organ. If high-risk patients can be easily identified, and if it is shown that withholding prophylaxis in patients considered low-risk is not associated with a high incidence of invasive candidiasis, then this targeted approach exposes only patients who might clearly benefit from an antifungal agent. The ideal antifungal agent to be used for prophylaxis is one that is proven to be efficacious, safe to the allograft and other organs, with predictable or no drug interactions, ease of administration, with minimal/manageable side effects, and affordable. It is also important to determine if the patient at risk for Candida infection is also at risk for mold infections, particularly due to Aspergillus, as this would require the choice of an agent with good anti-mold activity. Duration of antifungal prophylaxis is not clearly defined, as the studies have not been designed to look at this particular issue. As a general rule, prophylaxis should be maintained for at least 14 days posttransplantation, and longer if predisposing comorbidities persist. Since the risk factors and best choice of antifungal agent vary according to the transplanted organ, each organ will be discussed separately and recommendations are summarized on Table 2. Antifungal prophylaxis against Candida should be given to all adult liver transplant recipients at high risk for development of invasive candidiasis; that is those with ≥2 of the following risk factors: prolonged or repeat operation; retransplantation; renal failure; high transfusion requirement, that is transfusion of ≥40 units of cellular blood products including platelets, packed red blood cells and auto transfusion, choledocojejunostomy and Candida colonization in the perioperative period (1, 2) (II-1). Duration of prophylaxis is not clearly determined, and it has ranged from 5 days to 10 weeks in clinical trials. A duration of up to 4 weeks, or for the period of persistence of risk factors, seems reasonable. The use of fluconazole as a prophylactic antifungal agent should be limited only to patients at high risk for invasive candidiasis. Liver transplant recipients at risk for both candidiasis and aspergillosis should receive an agent with anti-Aspergillus activity. Three prospective randomized controlled trials in adults have shown the efficacy of antifungal prophylaxis of invasive candidiasis. In one study, fluconazole 100 mg/day was compared to oral nystatin in 143 liver transplant recipients. Prophylaxis was given for 4 weeks following liver transplantation. Fluconazole was associated with a reduction in Candida colonization and superficial infections, as well as a trend toward reduction of invasive infections (61). In the second trial, fluconazole 400 mg/day or placebo were administered for 10 weeks after liver transplantation. Antifungal prophylaxis with fluconazole compared to placebo resulted in a decreased rate of proven fungal infection (43% vs. 9%) and invasive infection (23% vs. 6%) (62). Overall survival was not improved. In the third study, itraconazole was compared to placebo, and demonstrated a decrease in the rate of candidiasis from 24% to 4% (63). Studies with LFAmB, including LAmB and ABLC, have used different doses for variable periods of prophylaxis. Risk factors for invasive fungal infection were also not uniform in these trials. These studies have demonstrated that low dose of liposomal amphotericin B (1 mg/(kg day)), administered for as few as 5 days, is associated with a significant reduction in invasive candidiasis (64-66). Caspofungin, an echinocandin, given for at least 21 days was shown to be an efficacious and well-tolerated antifungal regimen in high-risk liver transplant recipients in a recent multicenter, noncomparative, open-label trial (67). Its use as a prophylactic agent seems promising due to lack of significant drug interactions with tacrolimus, lack of nephrotoxicity and activity against non-albicans Candida, however no randomized controlled trial has been done so far. A recent meta-analysis demonstrated that antifungal prophylaxis in liver transplant recipients significantly reduced the total episodes of superficial and invasive fungal infection, as well as mortality attributable to fungal infections; however, it did not affect overall mortality or the need for empirical antifungal treatment (68). Compared to controls, patients receiving antifungal prophylaxis experienced a higher proportion of non-albicans Candida infections. Observing liver transplant recipients at low risk for IFIs without antifungal prophylaxis is safe, as demonstrated by a recent multi-center, prospective, observational study, in which 200 liver transplant recipients at low risk for invasive fungal infections did not receive antifungal prophylaxis. In this trial only 7% of the 193 eligible patients developed an invasive fungal infection at 100 days posttransplantation (56). Of those, only 2% were due to Candida sp. and potentially preventable by the use of fluconazole prophylaxis. The use of nonabsorbable agents, such as nystatin, clotrimazole and amphotericin B to achieve selective decontamination of the gastrointestinal tract and oral cavity has shown inconsistent results and not proven to be useful (57-60). Despite an absence of clinical trials in this patient population, antifungal prophylaxis in small bowel adult transplant recipients is routinely practiced and justified by the high rate of Candida infections. Rates of invasive candidiasis have been described to be as high as 28% in small case series (69, 70). Patients at high risk are those with graft rejection or dysfunction, enhanced immunosuppression, anastomotic disruption, abdominal reoperation or multivisceral transplantation. Fluconazole is an acceptable agent. However, LFAmB should be utilized in patients where there is high suspicion of non-albicans Candida spp. Prophylaxis is usually administered for a minimum of 4 weeks, until anastomosis has completely healed, and rejection is not present (II-3). The risk factors for candidiasis among pancreas transplant recipients include enteric drainage, vascular thrombosis and postperfusion pancreatitis (8). The use of prophylactic fluconazole should be considered whenever one of these risk factors is identified. LFAmB is preferred in centers with a high prevalence of non-albicans species. Duration of prophylaxis will depend on reduction of risk factors (II-3). The risk of invasive candidiasis is too low after isolated kidney transplantation to warrant prophylaxis. Candida is commonly isolated from the respiratory tract of lung and heart-lung transplant recipients; however, the occurrence of invasive disease is rare in the presence of prophylaxis. The highest risk for Candida infection is in the first 30 days posttransplantation, and risk factors include the use of broad spectrum antibiotics, duration of antibiotic use, presence of central venous catheters and need for renal replacement therapy (2, 3). There is a wide variation in the practice of antifungal prophylaxis in lung and heart-lung transplant recipients, not only in terms of the antifungal agent, but also on its mode of administration, timing and duration. Due to the high rates of Aspergillus infection following lung and heart-lung transplantation, antifungal prophylaxis should be directed towards the prevention of invasive aspergillosis, and prophylaxis with an agent without adequate anti-Aspergillus activity is not appropriate (II-1). Candida infections are infrequent after heart transplantation, and antifungal prophylaxis is not routinely recommended for these patients (III). There are no infection control measures specifically targeted toward prevention of Candida infections. Measures to reduce the incidence of these infections should include adequate hand hygiene, judicious use of antibiotics and frequent assessments to determine the need for intravascular and urinary catheters. The authors would like to acknowledge Ms. Vicki Noles, who provided administrative support in the preparation of this manuscript. Silveira, F.P.: Grant/Research Support, Pfizer, Luminex Molecular Diagnostics. Pappas P.: Grant Support, Merck, Pfizer, Schering-plough, Astellas; Ad Hoc Advisor, Basilea, Novartis, Semorex, Merck.