Fusarium infection: report of 26 cases and review of 97 cases from the literature.

Fusarium species is a ubiquitous fungus that causes opportunistic infections. We present 26 cases of invasive fusariosis categorized according to the European Organization for Research and Treatment of Cancer/Mycoses Study Group (EORTC/MSG) criteria of fungal infections. All cases (20 proven and 6 probable) were treated from January 2000 until January 2010. We also review 97 cases reported since 2000. The most important risk factors for invasive fusariosis in our patients were compromised immune system, specifically lung transplantation (n = 6) and hematologic malignancies (n = 5), and burns (n = 7 patients with skin fusariosis), while the most commonly infected site was the skin in 11 of 26 patients. The mortality rates among our patients with disseminated, skin, and pulmonary fusariosis were 50%, 40%, and 37.5%, respectively. Fusarium solani was the most frequent species, isolated from 49% of literature cases. Blood cultures were positive in 82% of both current study and literature patients with disseminated fusariosis, while the remaining 16% had 2 noncontiguous sites of infection but negative blood cultures. Surgical removal of focal lesions was effective in both current study and literature cases. Skin lesions in immunocompromised patients should raise the suspicion for skin or disseminated fusariosis. The combination of medical monotherapy with voriconazole or amphotericin B and surgery in such cases is highly suggested.

INTRODUCTION

Fusarium species, Aspergillus species, and Zygomycetes are the most clinically important molds. Fusarium species isolates are universally found in the environment and cause infection in both humans and plants.29,35,94,95,97 In humans, infection starts with the inhalation of Fusarium conidia or direct contact with materials contaminated with Fusarium conidia. Subsequently, conidia germinate and form filaments that invade the surrounding tissue when a suitable environment is offered.

There is a paucity of reports describing the predisposing factors and clinical characteristics of patients with Fusarium infection.77 The clinical presentation of fusariosis depends on the host’s immune status.140 Invasive infections, such as sinusitis, pneumonia, deep cutaneous infections, and disseminated infections, present in immunocompromised patients and most commonly manifest as fever not responding to antimicrobial medications.96 Specifically, neutropenia, deficits in cellular immunity, induction chemotherapy for leukemia, and hematopoietic cell transplantation are considered risk factors for the development of invasive fusariosis.15,96,98 On the other hand, immunocompetent patients present more frequently with superficial infections, such as keratitis and onychomycosis.19,49,50,60

In the current study, we describe the clinical characteristics of 26 patients with proven or probable invasive fusariosis managed at Massachusetts General Hospital (MGH) during a 10-year period and review the literature of cases with fusariosis published since January 2000, focusing on the therapeutic approach and outcome of patients.

PATIENTS AND METHODS

We identified patients with fusariosis treated at MGH from January 2000 to January 2010 by searching the records of the clinical microbiology laboratory at MGH, Boston, MA. We collected data by reviewing the electronic medical records of the patients; we retrieved their baseline characteristics, underlying diseases, treatment modalities, and outcome. We categorized all patients according to the revised definitions of the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/IFICG and NIAID/MSG) into proven and probable cases.28 We defined disseminated fusariosis as any case with at least 1 positive blood culture or with the concurrent involvement of 2 or more noncontiguous sites. Neutropenia was defined as an absolute neutrophil count ≤500 cells/μL, while steroid therapy was defined as the use of prednisone or prednisone-equivalent at a dose equal to or higher than 10 mg/day. For simplicity, the term “skin infection” was used to describe fusariosis involving the skin with or without involvement of other soft tissues. The study was approved by the institutional review board of MGH.

In our literature review, we identified published cases of fusariosis in the English literature by searching the MEDLINE database (National Library of Medicine, Bethesda, MD) using the terms “Fusarium” or “fusariosis.” Non-English language studies, expert commentaries, and abstracts from scientific meetings were excluded. Of note, in order to focus on invasive fusariosis, we also excluded patients with superficial localized infections, such as keratitis and onychomycosis. Finally, we excluded cases with positive cultures for Fusarium species but without histopathologic confirmation, and cases with limited data.

RESULTS

Case Series

We identified 26 patients with proven (n = 20) or probable (n = 6) invasive fusariosis. The mean age was 38 years (range, 4–68 yr). Table 1 represents the baseline patient characteristics. The most common underlying conditions included burns (n = 7), lung transplantation (n = 6, performed 2 mo, 4 mo, 1 yr, 1 yr, 2 yr, and 15 yr before the diagnosis of fusariosis was established) and hematologic malignancies (n = 5), such as acute myeloid leukemia (n = 3), B-cell lymphoma (n = 1), and Hodgkin lymphoma treated with stem cell transplantation 3.5 years before the index hospitalization (n = 1). Less frequent underlying diseases included focal segmental glomerulosclerosis (n = 1), end-stage liver disease (n = 1), chronic obstructive pulmonary disease (COPD) and tuberculosis (n = 1), lung cancer and COPD (n = 1), and asthma (n = 1). Of note, 3 patients had no comorbidities. Twelve patients were receiving immunosuppressive therapy at the time of admission, including cancer chemotherapy (n = 9), calcineurin inhibitors (n = 5), such as tacrolimus or cyclosporine and steroids (n = 2). Six patients developed fusariosis while receiving antifungal agents for prophylaxis, specifically fluconazole (n = 4), micafungin (n = 1), and voriconazole (n = 1).

TABLE 1

Characteristics of 26 MGH Patients With Fusarium Infection*

Skin was the most common site of infection (n = 11; 8 proven and 3 probable), followed by lung (n = 8 proven) and disseminated infection (n = 7; 4 proven and 3 probable). No speciation of Fusarium species was performed by the clinical laboratory of MGH. The minimum inhibitory concentration (MIC) of antifungal agents was determined for 4 isolates. Two were resistant to voriconazole (≥8 μg/mL) and 2 to both voriconazole and amphotericin B (≥8 μg/mL).

The diagnosis of skin fusariosis was established with cultures from the skin growing Fusarium species (n = 11) and skin biopsy (n = 8). Regarding the therapeutic management of skin fusariosis, 4 patients received medical treatment alone; 3 received combination therapy and died (liposomal amphotericin B/voriconazole: n = 2; and liposomal amphotericin B/micafungin: n = 1); and 1 received monotherapy with voriconazole and was cured. Five patients received both surgical and antifungal therapy; 3 received combination therapy (1 liposomal amphotericin B/micafungin and recovered, 1 voriconazole/micafungin and recovered, and 1 liposomal amphotericin B and voriconazole and died) and 2 received monotherapy (1 liposomal amphotericin B lost to follow-up and 1 voriconazole and recovered). Finally, 2 patients were treated surgically only and recovered. We note that 2 patients received granulocyte colony-stimulating factor (G-CSF) and G-CSF combined with granulocyte macrophage colony-stimulating factor (GM-CSF), respectively, for neutropenia (Table 2). Overall, 4 of 6 patients who received combination therapy died (66.7% mortality), and 1 of them had undergone subsequent surgery. On the other hand, none of the patients who received monotherapy, surgical treatment alone, or monotherapy with surgical treatment died.

TABLE 2

Treatment and Outcome of 26 MGH Patients

Patients with pulmonary fusariosis presented with shortness of breath (n = 6), fever (n = 6), cough (n = 3), fatigue (n = 2), and chest pain (n = 1), and Fusarium species were isolated from bronchial washings (n = 7) and sputum (n = 1). Regarding the treatment administered, 4 patients received monotherapy and recovered (voriconazole: n = 3; itraconazole: n = 1), while 4 received combination therapy (2 liposomal amphotericin B/voriconazole and died, 1 liposomal amphotericin B/caspofungin and died, and 1 inhaled amphotericin B and oral voriconazole and recovered). Overall, 3 of 4 patients who were treated with combination therapy died (75% mortality), while none of the patients treated with monotherapy died. Of note, as shown in Tables 1 and 2, 5 of the 8 cases of pulmonary fusariosis have previously been described.17

Patients with disseminated fusariosis presented with various clinical symptoms, including nausea and vomiting (n = 2), shortness of breath (n = 2), and fever (n = 1), and all of them had blood cultures positive for Fusarium species. Three patients were treated with combination therapy (liposomal amphotericin B/voriconazole, 2 died and 1 recovered), 3 patients received monotherapy with voriconazole and recovered, while 1 patient received no therapy and died (fusariosis was diagnosed after death). Overall, 2 of 3 patients who received combination therapy died (66.7% mortality), while none of the patients treated with monotherapy died. Of note, as shown in Tables 1 and 2, 1 case of disseminated fusariosis has previously been reported.17

Literature Review

We identified 97 cases of invasive fusariosis from the literature published since 2000.1–4,6–13,16,18,20–27,31–33,37–39, 42–44,46,48,50–57,59,61,62,65–67,69–71,74–81,83–85,87,89,91,97,99,100–103,105–108,111–113,115–121,124–126,128–135,139,140 The patients’ mean age was 45 years (range, 3 mo to 85 yr). Fifty-four patients were male and 41 female (sex was not reported in 2 cases). Comorbidities included hematologic malignancies (n = 60); neutropenia (n = 33); transplantation (n = 29) including hematopoietic stem cell transplant (HSCT; n = 18), renal transplant (n = 6), liver transplant (n = 3), and lung transplant (n = 2); diabetes mellitus (n = 10), and burns (n = 2). Regarding the intake of medications at the time of admission, 33 patients were receiving chemotherapy, 24 steroids, and 12 a combination of chemotherapy and steroids.

Speciation was performed in 57 cases. The most common species were Fusarium solani (n = 28), Fusarium oxysporum (n = 12), and Fusarium verticillioides (n = 5). Less common species included Fusarium dimerum (n = 3), Fusarium proliferatum (n = 3), Fusarium moniliforme (n = 2), Fusarium falciforme (n = 2), Fusarium acutatum (n = 1), and Fusarium subglutinans (n = 1).

Regarding the site of infection, 56 patients presented with skin involvement and 27 with disseminated infection, while the rest presented with pulmonary fusariosis (n = 3), infection of the nasal cavity (n = 2), peritonitis (n = 2), sinusitis (n = 1), and liver infection (n = 1). In 5 cases the site of infection was not mentioned. Blood cultures were positive in 22 of the 27 cases with disseminated infection. The MIC to amphotericin B was measured for 14 isolates, and the MIC of voriconazole was measured in 12 of the 14 isolates. Thirteen isolates were susceptible to amphotericin B, but only 8 isolates were susceptible to voriconazole.

Treatment was reported in 70 cases. In 32 cases, the treatment regimen was changed due to treatment failure or drug side effects (Table 3). Among patients with skin fusariosis, 35 received medical therapy alone, consisting of either monotherapy in 17 cases (voriconazole: n = 7, 6 recovered; liposomal amphotericin B: n = 6, 5 recovered; ketoconazole: n = 2, recovered; itraconazole: n = 1, recovered; and fluconazole: n = 1, recovered) or combination therapy in 18 cases (liposomal amphotericin B/voriconazole: n = 12, 7 recovered; liposomal amphotericin B/flucytosine: n = 2, 1 recovered; liposomal amphotericin B/fluconazole: n = 1, recovered; liposomal amphotericin B/itraconazole: n = 1, recovered; itraconazole/terbinafine: n = 1, recovered; and liposomal amphotericin B/posaconazole/terbinafine: n = 1, recovered). Two patients had the combination of surgery and antifungal monotherapy and recovered (amphotericin B: n = 1; and voriconazole: n = 1), while 3 underwent only surgery and recovered. We note that 9 patients received G-CSF because of neutropenia, while 1 received granulocytes, and 1 received the combination of G-CSF and granulocytes. Overall, 5 of the 16 patients who received combination therapy died (31.3% mortality), while 2 of the 19 patients who received monotherapy died (10.5% mortality).

TABLE 3

Treatment and Outcome of 97 Patients From the Literature

Twenty-two patients with disseminated fusariosis received medical therapy only consisting of either monotherapy in 15 cases (liposomal amphotericin B: n = 8, 7 recovered; voriconazole: n = 4, recovered; itraconazole: n = 1, recovered; posaconazole: n = 1, recovered; and caspofungin: n = 1, recovered) or combination therapy in 7 cases (amphotericin B/fluconazole: n = 2, 1 recovered; amphotericin B/voriconazole: n = 2, recovered; amphotericin B/terbinafine: n = 2, 1 recovered; and amphotericin B/posaconazole: n = 1, recovered), while 1 received surgical treatment and monotherapy with amphotericin B and recovered. Ten patients with disseminated disease received G-CSF for neutropenia, 1 received GM-CSF, and 1 received the combination of G-CSF and granulocytes. Overall, 3 of the 7 patients who received combination therapy died (42.9% mortality), while 2 of the 16 patients who received monotherapy died (12.5% mortality).

Three patients with pulmonary fusariosis received monotherapy and recovered (voriconazole: n = 1; amphotericin B deoxycholate: n = 1; and posaconazole: n = 1), 2 patients with nasal involvement received monotherapy with voriconazole and recovered, 2 patients with peritonitis received combination therapy and recovered (amphotericin B/flucytosine: n = 1; amphotericin B/ketoconazole: n = 1), 1 patient with sinusitis received monotherapy with amphotericin B and recovered, and 1 patient with liver involvement received monotherapy with amphotericin B and recovered.

DISCUSSION

Herein we present 26 patients with proven or probable invasive fusariosis treated at a general hospital (MGH patients) and review 97 cases reported since 2000. Hematologic malignancies, solid organ transplantation, HSCT, or immunosuppressive therapy were the predominant underlying conditions, and the skin was the predominant site of infection in both MGH and literature cases. The most frequently identified species among literature patients were members of the Fusarium solani complex (49%; 29 of 59 isolates). The clinical presentation varied depending on the infected site; diagnosis was established with biopsy and culture. Blood cultures were positive in 86% of MGH patients and 82% of literature patients with disseminated fusariosis (82% of all patients with disseminated disease). The most commonly used therapeutic scheme in MGH patients was monotherapy with voriconazole (n = 8, 0% mortality), followed by the combination of amphotericin B/voriconazole (n = 8, 87.5% mortality). On the contrary, literature patients were more frequently treated with monotherapy with amphotericin B (n = 19, 10.5% mortality), followed by the administration of voriconazole as monotherapy (n = 15, 6.7% mortality) and the combination of amphotericin B/voriconazole (n = 14, 35.7% mortality). Overall, the mortality rates of patients receiving combination therapy were higher than those of patients receiving monotherapy in both MGH and literature patients and for all infected sites. Also, patients who underwent surgical intervention exhibited lower mortality rates than patients treated with antifungal agents only.

Fusarium species are ubiquitous in the environment90 and can be found in soil and water.34,36,90,95 In humans, Fusarium species cause a wide spectrum of diseases, ranging from superficial to locally invasive and disseminated infections.14,19,40,41,47,58,64,68,73,82,88,90,93,95,110,114,122,123,127 Invasive and disseminated fusariosis occurs mostly in patients with compromised immune systems, and especially neutropenia.9,15,17,92,95,96,98,104,136 In fact, in our study 61.5% of MGH patients and 100% of literature patients were immunocompromised. The most frequent underlying disease in both the MGH and the published cases was hematologic malignancies (53% of patients overall),5,45,60,92,102 while other major underlying diseases included solid organ transplantation (15% of patients overall) and HSCT (15% of patients overall). Also, the majority of patients (63% of patients overall) were receiving immunosuppressive regimens at the time of diagnosis of fusariosis, and 31% were neutropenic.

Although the clinical presentation of invasive fusariosis is highly variable, high fever that is not responsive to antimicrobial therapy is relatively common in neutropenic patients. Respiratory infections also occur frequently, while skin involvement represents either a primary site of infection or metastatic lesions from disseminated disease11,93,95 and manifests as papules, nodules, areas of necrosis, mycetomas, target lesions, or bullae.93,95,132,138 In the current study, the skin was one of the main sites of infection (54% of patients overall). Since fusariosis leads to fatal outcomes without treatment or when treatment is delayed, skin lesions in immunocompromised patients should be investigated aggressively.11,30,93

The diagnosis of fusariosis is mainly based on culture of infected sites and histopathology, which require more than 5–7 days.95 This can cause a delay in treatment and result in fatalities. Moreover, the microscopic characteristics of Fusarium species are not always distinguishable from other molds, especially Aspergillus species, and therefore identification requires significant expertise.95 In addition, immunocompromised patients often receive prophylaxis with antifungal medications, which turn cultures for Fusarium species negative and render diagnosis more difficult. Thus, non-culture methods such as polymerase chain reaction (PCR) may be superior for detecting Fusarium species.55,56,63,71,72,95,137

Since there are not enough data to support a solid evidence-based approach to the treatment of invasive fusariosis, improvement in this domain is mandatory. Medical intervention remains the mainstay of treatment.17 Amphotericin B monotherapy is considered an effective therapeutic approach in immunocompromised patients, according to previously published studies. Specifically, in a retrospective study, amphotericin B cured or stabilized 15 of 26 patients with invasive fusariosis.109 Moreover, posaconazole has been reported as a successful salvage treatment for patients with invasive fusariosis.111 Most clinicians start with either amphotericin B or voriconazole as first-line therapy.86 In the current study, amphotericin B was used as monotherapy in 1 MGH patient, who recovered, and in 19.6% of literature patients with a mortality rate of 10.5%. Voriconazole was also commonly administered (30.8% of MGH patients and 15.5% of literature cases, with mortality rates of 0% and 6.7%, respectively). Finally, the combination of amphotericin B/voriconazole was used in 30.8% of MGH patients and 14.4% of literature cases with mortality rates of 87.5% and 35.7%, respectively. These results show that monotherapy with voriconazole and monotherapy with amphotericin B are both successful in treating invasive fusariosis, since the difference in mortality is not significant.

Among patients with skin fusariosis, for MGH patients the mortality for those treated with combination therapy was 66.7% compared with 0% mortality for those treated with monotherapy, and for literature patients the mortality was 31.3% for those treated with combination therapy compared with 10.5% for those treated with monotherapy. MGH patients with pulmonary fusariosis had a mortality rate of 75%, while none of the literature patients treated with monotherapy died. Finally, disseminated disease led to a mortality rate of 66.7% in MGH patients treated with combination therapy compared to 0% in those treated with monotherapy, while the mortality rates for literature patients were 42.9% and 12.5%, respectively. The higher mortality rates of patients who received combination therapy may be explained by the fact that more severely ill patients tend to receive more medications and thus are more likely to be treated with combination therapy. Overall, the most effective antifungal therapy should be based on Fusarium species susceptibility testing. Although in the present series, the MIC of antifungal agents was determined for only 6 isolates, the resistance profiles of some of these isolates highlight the emerging resistance of Fusarium species to traditional antifungal regimens.60

Finally, surgery was performed in 7 MGH patients with skin fusariosis (14.3% mortality), 5 literature patients with skin fusariosis (0% mortality), and 1 literature patient with disseminated infection (0% mortality). The mortality rates of patients who underwent surgery were lower compared to those who did not have surgery (rates of 75%, 19.4%, and 8.7%, respectively). This finding suggests that the surgical removal of focal lesions may result in better outcomes and thus should be considered for patients with skin or disseminated infections.

Despite medical intervention, invasive fusariosis has been linked to a high mortality rate. In the current report, the mortality rate was 50% among the MGH patients with disseminated fusariosis, 40% among patients with skin fusariosis, and 37.5% among patients with lung fusariosis. On the other hand, the mortality rates of the literature patients with disseminated, skin, and pulmonary fusariosis were 30%, 18%, and 0%, respectively. The higher mortality rates of the MGH patients in the current series may be attributed to their comorbidities, and the lower mortality rates of the literature patients may be due to a “publication bias.” A representative example of more comorbidities among MGH patients is the presence of burns (7.7%, n = 7 MGH patients vs 2.1%, n = 2 literature patients). This difference may lead to the observation of a higher mortality in the MGH patient population and may be explained by the fact that most case reports in the literature present patients with a favorable outcome and thus with fewer comorbidities. We note that 6 of 7 patients with burns had skin fusariosis. However, a true association between burns and skin fusariosis is difficult to establish, because of the small sample size of the MGH cohort.

In conclusion, the results of the current study suggest that monotherapy with voriconazole is equally effective to monotherapy with amphotericin B for the treatment of invasive fusariosis, and that the addition of surgery leads to better outcomes in patients with skin and disseminated fusariosis. More studies with larger sample sizes that will allow for stratification of the outcomes by disease severity are required to investigate the difference in the effectiveness of combination medical therapy versus monotherapy, with the low prevalence of invasive fusariosis being a limiting factor.