Mucormycosis (zygomycosis)

INTRODUCTION — 

Mucormycosis is manifested by a variety of different syndromes in humans, particularly in immunocompromised patients and those with diabetes mellitus [1]. Devastating rhino-orbital-cerebral and pulmonary infections are the most common syndromes caused by these fungi.

There is some controversy over the terminology used to refer to infections due to this group of fungi. The term "mucormycosis" was used for years and then was supplanted by "zygomycosis" for several decades. Based on molecular studies, "mucormycosis" is currently again the appropriate term [1-4].

The microbiology, clinical manifestations, diagnosis, and therapy of mucormycosis will be reviewed here. Several specific syndromes that can be caused by these fungi are discussed separately. (See "Fungal rhinosinusitis" and "Non-access-related infections in patients on chronic dialysis".)

MYCOLOGY — 

The genera in the order Mucorales cause most human infection. These organisms are ubiquitous in nature and can be found on decaying vegetation and in the soil. These fungi grow rapidly and release large numbers of spores that can become airborne. Because the agents of mucormycosis are common in the environment, they are relatively frequent contaminants in the clinical microbiology laboratory; all humans have ample exposure to these fungi during day-to-day activities. The fact that mucormycosis is a rare human infection reflects the effectiveness of the intact human immune system. This is further supported by the finding that almost all human infections due to the agents of mucormycosis occur in the presence of some underlying compromising condition.

The genera most commonly found in human infections are Rhizopus, Mucor, and Rhizomucor; Cunninghamella, Absidia (now reclassified as Lichtheimia), Saksenaea, and Apophysomyces are genera that are less commonly implicated in infection [5].

The hyphae of the Mucorales are distinct and allow for a presumptive identification from clinical specimens. The hyphae are broad (5 to 15 micron diameter), irregularly branched, and have rare septations (picture 1). This is in contrast with the hyphae of ascomycetous molds, such as Aspergillus, which are narrower (2 to 5 micron diameter), exhibit regular branching, and have many septations.

The lack of regular septations may contribute to the fragile nature of the hyphae and the difficulty of growing the agents of mucormycosis from clinical specimens. Grinding clinical specimens can cause excessive damage to the hyphae. Thus, finely mincing tissues is preferred for culturing tissue samples that may contain molds.

PATHOGENESIS — 

Rhizopus organisms have an enzyme, ketone reductase, which allows them to thrive in high glucose, acidic conditions. Serum from healthy individuals inhibits growth of Rhizopus, whereas serum from individuals in diabetic ketoacidosis stimulates growth [6].

Rhino-orbital-cerebral and pulmonary mucormycosis are acquired by the inhalation of spores. In healthy individuals, cilia transport these spores to the pharynx, and they are cleared through the gastrointestinal tract. In susceptible individuals, infection usually begins in the nasal turbinates or the alveoli [7]. The agents of mucormycosis are angioinvasive; thus, infarction of infected tissues is a hallmark of invasive disease [8].

Deferoxamine and iron overload — Deferoxamine, which chelates both iron and aluminum, increases the risk of mucormycosis by enhancing growth and pathogenicity [7,9-11]. The deferoxamine-iron chelate, called feroxamine, is a siderophore for the species Rhizopus, increasing iron uptake by the fungus, which stimulates fungal growth and leads to tissue invasion [8,12].

Iron overload itself may predispose to mucormycosis in the absence of deferoxamine therapy [13]. In addition, individuals with diabetic ketoacidosis have elevated concentrations of free iron in their serum, which supports the growth of Rhizopus oryzae at an acidic, but not at an alkaline, pH [14,15].

Deferoxamine was once used commonly as an aluminum chelator in patients with kidney failure; however, aluminum excess is rarely seen today. Currently, patients at risk for deferoxamine-associated mucormycosis are those who have received multiple blood transfusions and are treated with this chelating agent for iron overload. The majority of patients with deferoxamine-associated infection present with disseminated disease that is rapidly fatal, with a mortality rate that approaches 90 percent [11]. (See "Aluminum toxicity in chronic kidney disease", section on 'Adverse effects of deferoxamine'.)

In contrast with deferoxamine, other iron chelating agents, such as deferasirox and deferiprone, do not act as siderophores and therefore do not increase the risk of mucormycosis. Limited studies of their adjunctive use in mice with mucormycosis have suggested benefit, but results in humans have been mixed. This is discussed in greater detail elsewhere. (See 'Therapies not recommended' below.)

RISK FACTORS — 

Almost all patients with invasive mucormycosis have some underlying disease that both predisposes to the infection and influences the clinical presentation. The most common underlying diseases are [5,16-28]:

●Diabetes mellitus, particularly with ketoacidosis

●Treatment with glucocorticoids [23]

●Hematologic malignancies [1,25,29]

●Hematopoietic cell transplantation [5,23,30]

●Solid organ transplantation [30-33]

●Treatment with deferoxamine [9-11,13] (see 'Deferoxamine and iron overload' above and "Iron chelation: Choice of agent, dosing, and adverse effects", section on 'Deferoxamine dosing + AEs (DFO, Desferal)')

●Iron overload [13]

●Recent coronavirus disease 2019 (COVID-19) infection [34-36]

●Acquired immunodeficiency syndrome (AIDS)

●Injection drug use

●Trauma/burns [37-39]

●Malnutrition

EPIDEMIOLOGY — 

The incidence of mucormycosis is difficult to estimate since it is not a reportable disease, and the risk varies widely in different populations. A review of 929 cases of mucormycosis that were reported between 1940 and 2003 noted that diabetes mellitus was the most common risk factor, found in 36 percent of cases, followed by hematologic malignancies (17 percent) and solid organ or hematopoietic cell transplantation (12 percent) [5]. In some patients, mucormycosis was the diabetes-defining illness. In a later study of 101 patients diagnosed with mucormycosis between 2005 and 2007 in France, hematologic malignancy was the most common risk factor, occurring in 50 percent of patients, followed by diabetes in 23 percent and trauma in 18 percent of cases [37].

Malignancy and hematopoietic cell transplantation — Among patients with malignancy, hematologic malignancies are much more frequently associated with mucormycosis than are solid tumors [5,23-25]. Pulmonary infection is more common than rhino-orbital-cerebral infection in this patient population. However, even in patients with hematologic malignancies, mucormycosis appears to occur in less than 1 percent of patients [40]. Among hematopoietic cell transplant recipients, the reported incidence has ranged from 0.1 to 2 percent, with the highest incidence in patients with graft-versus-host disease [30].

Reports from several countries have noted an increase in mucormycosis in patients with hematologic malignancies [1,29,41]. Most of these patients had undergone hematopoietic cell transplant, many had graft-versus-host disease, and almost all were on voriconazole for prophylaxis or treatment [1,29]. A case-control study in a population of patients with hematologic malignancies compared patients with mucormycosis with those who had no fungal infection; voriconazole prophylaxis was an independent risk factor for mucormycosis (odds ratio 10.4, 95% CI 2.1-39) [25].

Although the possible causes of the increase in mucormycosis in patients with hematologic malignancies continue to be debated, reasons that have been proposed include selection of the agents of mucormycosis caused by voriconazole use, increasing use and intensity of immunosuppressive agents, and a decrease in invasive aspergillosis-related mortality following hematopoietic cell transplant, resulting in the emergence of rare fungi during the late post-transplant period [42].

Solid organ transplantation — In a multicenter prospective study of invasive fungal infections in transplant recipients in the United States between 2001 and 2006, the 12-month cumulative incidence of mucormycosis was less than 1 percent among solid organ transplant recipients; only 2 percent of invasive fungal infections in such patients were caused by the agents of mucormycosis [30,43]. Among solid organ transplant recipients, risk factors for mucormycosis include kidney transplantation [31], kidney failure, diabetes, and prior voriconazole or caspofungin use [33].

Diabetes — As noted above, diabetes is a common predisposing condition [5]. Diabetes appears to be more likely than other conditions to predispose to rhino-orbital-cerebral infection, but the reason for this is unknown [44]. The number of reported cases of mucormycosis in patients with diabetes in the United States has declined since the 1990s, a trend that has not been noted in France [41] or in resource-limited countries [26,45]. One hypothesis that has been suggested to explain the decline in the United States is the widespread use of statins, which have inhibitory activity in vitro against a wide range of the agents of mucormycosis [26].

Health care-associated — Health care-associated cases of mucormycosis have been reported. In a study of 169 health care-associated cases of mucormycosis in Europe between 1970 and 2008, the major underlying diseases were solid organ transplantation (in 24 percent), diabetes mellitus (in 22 percent), and severe prematurity (in 21 percent) [46]. Skin was the most common site of infection (in 57 percent), followed by the gastrointestinal tract (15 percent). Portals of entry included surgery, catheters (especially intravascular catheters), and adhesive tape. Outbreaks and clusters have been associated with adhesive bandages, wooden tongue depressors, adjacent building construction, and hospital linens [46-50]. A survey found that Mucorales could be cultured from 47 percent of hospital linen at several cancer and transplant centers [51]. The clinical significance of this finding is unclear. Fatal gastrointestinal mucormycosis has also been reported in a premature neonate who received a probiotic that was contaminated with R. oryzae; probiotics were being given in an attempt to prevent necrotizing enterocolitis [52]. (See "Neonatal necrotizing enterocolitis: Prevention", section on 'Probiotics'.)

Natural disaster-associated — Cases of mucormycosis have occurred following a tornado, a tsunami, and a volcanic eruption [53-57]. As an example, in 2011, an outbreak of necrotizing soft tissue infections causes by Apophysomyces trapeziformis occurred in patients with traumatic injuries resulting from a tornado in Joplin, Missouri, in the United States [53]. A total of 18 suspected cases were identified, of which 13 were confirmed. Two patients had diabetes, and none were immunocompromised. Ten patients required intensive care unit admission, and five died.

In a case-control study of patients injured during the tornado in Joplin, Missouri, the 13 patients with confirmed infection each had a median of five wounds (range one to seven); 11 patients (85 percent) had ≥1 fracture, nine patients (69 percent) had blunt trauma, and five patients (38 percent) had penetrating trauma [56]. All case patients had been located in the zone that sustained the most severe damage during the tornado. On multivariate analysis, infection was associated with penetrating trauma (adjusted odds ratio [aOR] 8.8, 95% CI 1.1-69.2) and an increased number of wounds (aOR 2.0 for each additional wound, 95% CI 1.2-3.2). Presumably, the fungus was introduced into the wounds at the time of trauma and then was able to establish infection. Whole-genome sequencing showed that the Apophysomyces trapeziformis isolates represented four separate strains.

Combat-associated — Invasive fungal infections, including mucormycosis, have been reported in United States military personnel who sustained blast injuries during combat in Afghanistan [58]. Between June 2009 and through December 2010, a total of 37 cases were identified, including 20 proven cases (with histopathologic evidence of angioinvasion), 4 probable cases (with histopathologic evidence of nonvascular tissue invasion), and 13 possible cases (with a positive fungal culture, but without histopathologic evidence). All injuries occurred secondary to explosive blasts, and most injuries occurred during foot patrol (in 34 patients; 92 percent). All patients had extremity injuries, and 29 patients (78 percent) had lower extremity amputation either at the time of the injury or during the first surgery following the injury. Thirty-six patients (97 percent) required large-volume blood transfusion.

Mold isolates were recovered from the wounds of 31 patients (83 percent); the most commonly detected fungi belonged to the order Mucorales (in 16 patients), Aspergillus spp (in 16 patients), and Fusarium spp (in nine patients) [58]. Multiple mold species were isolated from 28 percent of infected wounds. Frequent debridements or amputation revisions were required, and three deaths were thought to be related to the infection (8 percent).

Coronavirus disease 2019-associated — There have been numerous reports of mucormycosis in patients diagnosed with coronavirus disease 2019 (COVID-19) [34-36,59-62]. The majority involve individuals with underlying diabetes mellitus who received steroids for COVID-19 prior to diagnosis with mucormycosis. Most cases were detected two to three weeks after the initial diagnosis of COVID-19.

Rhino-orbital-cerebral mucormycosis is most frequently reported, and symptoms include facial pain, facial or orbital swelling, headache, and/or nasal eschar. Pulmonary mucormycosis infections have been identified in individuals with COVID-19 who develop unexplained worsening pulmonary status or complications. Gastrointestinal mucormycosis has been reported as well [36,63-65].

Reports suggest that onset of mucormycosis symptoms usually occurs 5 to 14 days after admission for COVID-19, but there are reports of individuals receiving both diagnoses at the time of presentation [66,67]. The majority of reports are from India, but cases have been reported from around the world. Overall mortality in case series from India ranges from 15 to 31 percent.

CLINICAL PRESENTATION — 

Mucormycosis is characterized by infarction and necrosis of host tissues that results from invasion of the vasculature by hyphae [27]. The pace is usually fast, but there are rare descriptions of infections with an indolent course [68,69].

Rhino-orbital-cerebral mucormycosis — The most common clinical presentation of mucormycosis is rhino-orbital-cerebral infection, which is presumed to start with inhalation of spores into the paranasal sinuses of a susceptible host. Hyperglycemia, usually with an associated metabolic acidosis, is the most common underlying condition [5]. A review of 179 cases of rhino-orbital-cerebral mucormycosis found that 126 (70 percent) of the patients had diabetes mellitus and that most had ketoacidosis at the time of presentation [16]. There are rare reports of rhino-orbital-cerebral mucormycosis in the absence of any apparent risk factors [16,69-72].

The infection usually presents as acute sinusitis with fever, nasal congestion, purulent nasal discharge, headache, and sinus pain. All of the sinuses become involved, and spread to contiguous structures, such as the palate, orbit, and brain, usually progressing rapidly over the course of a few days. However, there have been some reports of rhino-orbital-cerebral mucormycosis with an indolent course that progresses over the course of weeks [68].

The hallmarks of spread beyond the sinuses are tissue necrosis of the palate resulting in palatal eschars, destruction of the turbinates, perinasal swelling, and erythema and cyanosis of the facial skin overlying the involved sinuses and/or orbit (picture 2). A black eschar, which results from necrosis of tissues after vascular invasion by the fungus, may be visible in the nasal mucosa, palate, or skin overlying the orbit (picture 3) [73,74].

Signs of orbital involvement include periorbital edema, proptosis, and blindness. Facial numbness is frequent and results from infarction of sensory branches of the fifth cranial nerve. Spread of the infection from the ethmoid sinus to the frontal lobe results in obtundation. Spread from the sphenoid sinuses to the adjacent cavernous sinus can result in cranial nerve palsies, thrombosis of the sinus, and involvement of the carotid artery. Hematogenous spread to other organs is rare unless the patient has an underlying hematologic malignancy with neutropenia.

A review of 208 cases of rhino-orbital-cerebral mucormycosis published in the literature between 1970 and 1993 found the following frequency of symptoms and signs [75]:

●Fever – 44 percent

●Nasal ulceration or necrosis – 38 percent

●Periorbital or facial swelling – 34 percent

●Decreased vision – 30 percent

●Ophthalmoplegia – 29 percent

●Sinusitis – 26 percent

●Headache – 25 percent

Rhino-orbital-cerebral mucormycosis is most commonly caused by R. oryzae.

Pulmonary mucormycosis — Pulmonary mucormycosis is a rapidly progressive infection that occurs after inhalation of spores into the bronchioles and alveoli (image 1). Pneumonia with infarction and necrosis results, and the infection can spread to contiguous structures, such as the mediastinum and heart [76,77], or disseminate hematogenously to other organs [78,79].

Most patients have fever with hemoptysis that can sometimes be massive [78,79]. The most common underlying conditions have been hematologic malignancies, treatment with glucocorticoids or deferoxamine, and solid organ transplantation; pulmonary infection is less common than rhino-orbital-cerebral infection in patients with diabetes [17,78-83]. In a series of 24 patients with mucormycosis and hematologic malignancy, 71 percent had pulmonary involvement, and only one had rhino-orbital-cerebral involvement; 7 of 12 patients who underwent autopsy had disseminated disease in addition to pulmonary infection [23].

Gastrointestinal mucormycosis — Although unusual, mucormycosis of the gastrointestinal tract may occur as the result of ingestion of spores. One review of 87 cases found that the stomach was the most common site (58 percent), followed by the colon (32 percent) [84]. The ileum and esophagus were rare sites of involvement.

The underlying diseases of patients with gastrointestinal mucormycosis have been diabetes mellitus, solid organ transplantation, treatment with glucocorticoids, and prematurity and/or malnutrition in infants [18,84-87]. Patients present with abdominal pain and hematemesis. The gastrointestinal lesions are necrotic ulcers that can lead to perforation and peritonitis [18]. Bowel infarctions and hemorrhagic shock can result from gastrointestinal mucormycosis [88], and the prognosis for all patients is poor.

Cutaneous mucormycosis — Infection of the skin and soft tissues with the agents of mucormycosis usually results from inoculation of the spores into the dermis. Thus, cutaneous mucormycosis is almost always associated with trauma or wounds. The entry of the fungi into the dermis can result from seemingly innocuous insults, such as the entry site for an intravenous catheter, spider bites, and insulin injection sites [19,20,89-93]. Infection has also been associated with contaminated traumatic wounds, dressings and splints, burns, and surgical sites [19,20,46,53,54,56,58,89-93].

When infection is associated with relatively minor breaks in the skin, the host usually has some underlying disease, such as diabetes mellitus, organ transplantation, neutropenia, or severe prematurity. Infections associated with major trauma or contaminated dressings have been found in otherwise immunocompetent patients [5,93].

Health care-associated, natural disaster-associated, and combat-associated cases of mucormycosis are discussed above. (See 'Health care-associated' above and 'Natural disaster-associated' above and 'Combat-associated' above.)

Cutaneous mucormycosis usually appears as a single, painful, indurated area of cellulitis that develops into an ecthyma-like lesion. Patients who have suffered trauma with an open wound that was contaminated with spores can develop rapidly progressive tissue necrosis, reflecting the presence of ischemic infarction [19,20,54,56,58,90,92,93]. Dissemination and deep tissue involvement are unusual complications of cutaneous mucormycosis [91,93].

Renal mucormycosis — Isolated involvement of the kidneys with mucormycosis has been reported and is presumed to occur via seeding of the kidneys during an episode of fungemia. Almost all patients with isolated renal mucormycosis have risk factors for fungemia, including an intravenous catheter, intravenous drug use, or AIDS [21,22,94-96].

Patients with renal mucormycosis usually present with flank pain and fever. Involvement can be either unilateral or bilateral.

Isolated CNS involvement — Central nervous system (CNS) mucormycosis typically arises from an adjacent paranasal sinus infection. However, there have been more than 30 cases of isolated CNS mucormycosis described in the literature [22,97-101]. Infection is thought to result from seeding of the brain during an episode of fungemia, analogous to kidney involvement. Over two-thirds of the patients with isolated CNS mucormycosis have been people who inject intravenous drugs who presumably have injected material contaminated with fungi directly into the bloodstream. Some of the patients with isolated CNS mucormycosis have had human immunodeficiency virus (HIV) infection in addition to drug use [22,98,102]. However, it is not clear whether HIV infection is an independent risk factor.

Most of the patients with isolated CNS mucormycosis have presented with lethargy and focal neurologic deficits, with the vast majority having involvement of the basal ganglia [97-99]. Isolated involvement of the frontal lobe has also been described [97].

Disseminated disease — Disseminated mucormycosis is rare and occurs most commonly in severely immunocompromised patients, burn patients, premature infants, and individuals who have received deferoxamine [1,5]. In the series of 929 cases of mucormycosis described above, disseminated disease was present in 40 to 50 percent of patients with cerebral or pulmonary mucormycosis [5]. The mortality rate in patients with disseminated mucormycosis was 96 percent.

DIAGNOSIS

Approach to diagnosis — The diagnosis of mucormycosis relies upon the identification of organisms in tissue by histopathology with culture confirmation. However, culture often yields no growth, and histopathologic identification of an organism with a structure typical of Mucorales may provide the only evidence of infection. A clinician must think of this entity in the appropriate clinical setting and pursue invasive testing to establish a diagnosis as early as possible. On the other hand, the agents of mucormycosis can colonize the airways or be contaminants in cultures, and the isolation of these fungi in a culture does not necessarily prove infection. Interpreting the culture results in the context of the patient's signs and symptoms and underlying disease are necessary to determine whether antifungal therapy should be given.

Serum tests, such as the 1,3-beta-D-glucan assay and the Aspergillus galactomannan assay, are being used with increased frequency in patients suspected of having an invasive fungal infection. The agents of mucormycosis do not share these cell wall components and neither test is positive in patients with mucormycosis. (See "Diagnosis of invasive aspergillosis", section on 'Beta-D-glucan assay' and "Diagnosis of invasive aspergillosis", section on 'Galactomannan antigen detection'.)

Polymerase chain reaction (PCR)-based techniques performed on histologic specimens that show fungal elements can confirm the diagnosis of mucormycosis [103-105]. However, the value of these tests on histologic samples with no demonstrable fungal elements is unclear. PCR-based tests on serum or plasma samples would be ideal since it would obviate the need for invasive sampling. While there have been several studies showing the value of serum or plasma PCR tests, their utility in clinical practice is unclear [39,106-112], and their commercial availability remains limited. A positive test may influence the choice of antifungal (for example, changing from voriconazole to amphotericin B), but a negative test cannot rule out mucormycosis.

In addition to traditional culture techniques and PCR with sequencing, matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry can be used to identify the causative species from culture specimens [113-115].

Rhino-orbital-cerebral infection — The presence of mucormycosis should be suspected in high-risk patients, especially those who have diabetes and metabolic acidosis and who present with sinusitis, altered mentation, and/or infarcted tissue in the nose or palate.

Endoscopic evaluation of the sinuses should be performed to look for tissue necrosis and to obtain specimens. The specimens should be inspected for characteristic broad, nonseptate hyphae with right-angle branching using calcofluor white and methenamine silver stains. The presence of the characteristic hyphae in a clinical specimen provides a presumptive diagnosis that should prompt further evaluation (picture 1). However, the absence of hyphae should not dissuade clinicians from the diagnosis of mucormycosis when the clinical picture is highly suggestive.

Further evaluation includes imaging to gauge sinus involvement and to evaluate contiguous structures such as the eyes and brain [116]. We generally perform a computed tomography (CT) scan as the initial imaging study as it can often be obtained quickly and is more sensitive than magnetic resonance imaging (MRI) for detecting bony erosions [117]. Clinicians should have a low threshold for performing an MRI in patients with abnormalities on CT because the MRI will enhance detection of intracranial, intraorbital, and cavernous sinus involvement. In a study of 23 immunocompromised patients with fungal sinusitis, CT findings included severe soft tissue edema of the nasal cavity mucosa (turbinates, lateral nasal wall and floor, and septum) in 21 patients, sinus mucoperiosteal thickening in 21 patients, bone erosion in eight patients, orbital invasion in six patients, facial soft tissue swelling in five patients, and retroantral fat pad thickening in two patients [118].

Pulmonary infection — The diagnosis of pulmonary mucormycosis is difficult because the presentation does not differ from pneumonia due to other angioinvasive molds. Isolating an agent of mucormycosis from respiratory cultures in a high-risk patient with a compatible clinical presentation is an indication for starting empiric treatment. Establishing a definitive diagnosis can be difficult because it requires demonstration of the organism in tissue.

Because obtaining tissue can be difficult in these cases, we often rely on radiographic evidence to support the diagnosis. Chest radiographs or CT scans can have variable findings:

●The most common abnormalities are focal consolidation, masses, pleural effusions, or multiple nodules [80,119]. The presence of a pleural effusion or more than 10 nodules may independently predict mucormycosis and help differentiate it from aspergillosis in immunocompromised individuals [119].

●A halo sign (ground-glass attenuation surrounding a nodule) is highly suggestive of angioinvasive fungi, including mucormycosis.

●Reversed halo signs (focal areas of ground-glass attenuation surrounded by a ring of consolidation) have been reported and may be more common in mucormycosis than other invasive mold infections [120-122]. In one study of 189 patients with fungal pneumonia, a reversed halo sign was seen in 7 of 37 patients with mucormycosis (19 percent), 1 of 132 patients with invasive aspergillosis (<1 percent), and none of 20 patients with fusariosis [120].

●Cavitary lesions with or without air crescent signs are unusual, although cavitation may be more common in COVID-19-associated mucormycosis [80,123].

Sputum or bronchoalveolar lavage (BAL) specimens can show the characteristic broad nonseptate hyphae, which is often the first indicator of mucormycosis [81]. However, in one case series, only 25 percent of sputum or BAL specimens were positive premortem [23]. Hyphae can also be demonstrated on lung biopsy (picture 1).

Other syndromes — The diagnosis of gastrointestinal mucormycosis can be made with endoscopic biopsy of the lesions that show the characteristic hyphae. For isolated kidney involvement, percutaneous biopsy or nephrectomy can establish a diagnosis [21]. Urine cultures are almost always sterile. Imaging of the kidneys with CT can demonstrate either ill-defined areas of low attenuation and diminished enhancement suggestive of pyelonephritis or multiple small foci suggestive of abscesses.

In isolated central nervous system (CNS) involvement, CT scan usually shows poorly enhancing lesions; cerebrospinal fluid cultures are negative. Diagnosis can be made with biopsy or resection of the involved areas.

TREATMENT — 

Treatment of mucormycosis involves a combination of starting antifungal therapy as soon as possible and aggressive surgical debridement of involved tissues, whenever feasible (see 'Aggressive surgical debridement' below). We initiate empiric antifungal therapy and evaluate the patient for surgical candidacy as soon as the diagnosis is considered because establishing a definitive diagnosis is challenging, and early treatment improves outcomes. Predisposing factors (eg, ketoacidosis, neutropenia, immunosuppression) should be reversed whenever possible [2,28].

Aggressive surgical debridement — We perform aggressive surgical debridement as soon as any form of mucormycosis is suspected because extensive surgical resection has been associated with improved survival. Surgical debridement allows for the collection of tissue specimens that can be sent for culture and histopathology to confirm the diagnosis. Additionally, histopathology can evaluate resection margins to identify residual disease leftover after debridement. Repeat debridement is often necessary for maximal resection.

There are no clinical data to support the use of topical amphotericin B irrigation, but some surgeons use amphotericin B irrigation during surgical debridement. Although amphotericin B does not penetrate into host tissues when used topically, some clinicians believe that it may locally stimulate host immune responses in addition to providing some topical debulking of the infection.

If extensive mucormycosis is not amenable to complete resection or there are other factors precluding surgery (eg, severe thrombocytopenia), we focus on prompt antifungal administration and make every effort to optimize predisposing medical conditions. (See 'Correction of predisposing conditions' below.)

Surgical removal of necrotic and infected tissue has been associated with improved survival in anecdotal reviews of rhino-orbital-cerebral and pulmonary mucormycosis [124,125]. Debridement of all necrotic tissue can be disfiguring, requiring removal of the palate, nasal cartilage, and the orbit, although endoscopic debridement with limited tissue removal can be accomplished [125]. Lobectomy cures have been reported for early pulmonary infection [78,79,126]. For isolated renal mucormycosis, cures have been described with amphotericin B deoxycholate alone [21,94] or combined with nephrectomy [21].

Antifungal therapy — Early antifungal therapy improves the outcome of mucormycosis infection. The preferred initial antifungal treatment for mucormycosis is liposomal amphotericin B. In patients who demonstrate clinical response after a few weeks, initial therapy with amphotericin B can be switched to maintenance therapy with intravenous (IV) or oral isavuconazole or posaconazole. Maintenance therapy should continue until radiographic, clinical, and histopathologic signs of disease resolve, and predisposing conditions have been optimized.

Initial therapy with amphoterin B

●Monotherapy for most patients – For initial therapy of suspected or diagnosed mucormycosis, we suggest a lipid formulation of amphotericin B (liposomal amphotericin B or lipid complex amphotericin B) [127-129]. However, for patients with disseminated infection and severe immunosuppression (eg, solid organ or hematopoietic stem cell transplant recipients), some experts start with combination antifungal therapy (eg, amphotericin B plus posaconazole), given the historically very poor outcomes in these patient populations. (See 'Combination antifungal therapy' below.)

We avoid amphotericin B deoxycholate (nonlipid formulation) due to the significant potential for nephrotoxicity unless other options are not available or the patient has isolated renal mucormycosis; lipid formulations of amphotericin B do not penetrate the kidney or achieve adequate concentrations in the urine, and there is little experience with using posaconazole or isavuconazole in this situation [21,94].

●Dosing – For most patients, we use a starting dose of 5 mg/kg liposomal amphotericin B daily. For patients with central nervous system (CNS) involvement and/or transplant recipients, we use a starting dose of 10 mg/kg liposomal amphotericin B daily. For patients who do not improve after at least one week of standard-dose liposomal amphotericin B 5 mg/kg daily, we increase the dose to 10 mg/kg daily (high-dose) and reassess response to therapy in one to two weeks. (See 'Assessing response to therapy' below.)

●Duration of initial therapy – We continue lipid formulation amphotericin B until the patient has shown signs of clinical improvement, which usually takes several weeks.

●Monitoring during initial therapy – Monitoring during initial antifungal therapy consists of frequent laboratory monitoring and hydration to prevent or correct side effects of amphotericin B (nephrotoxicity or electrolyte abnormalities) and to ensure the infection is responding to treatment. Nephrotoxicity is a serious side effect of amphotericin B, although kidney injury is mitigated by use of liposomal formulations. In one study of 33 patients treated with 10 mg/kg/day liposomal amphotericin B for mucormycosis, creatinine doubled in 40 percent of patients; kidney injury resolved in 63 percent of these patients within 12 weeks [130]. Because of the high mortality rate with mucormycosis infection, greater risks of treatment should be tolerated. Adverse effects of amphotericin B are discussed separately. (See "Pharmacology of amphotericin B", section on 'Adverse effects' and "Amphotericin B nephrotoxicity".)

●Managing adverse effects of amphotericin B – Every effort should be made to use liposomal amphotericin B for initial therapy rather than azoles.

In patients who have or develop compromised kidney function but whose infection has not yet clinically improved, we continue with liposomal amphotericin B since the potential survival benefit justifies the possibility of requiring kidney replacement therapy. In patients who cannot tolerate liposomal amphotericin B, monotherapy with posaconazole or isavuconazole is a less effective alternative.

In patients who develop acute kidney injury during initial therapy of mucormycosis with liposomal amphotericin B but who are clinically improving, we switch to posaconazole or isavuconazole. (See 'Maintenance therapy' below.)

●Rationale – The selection of amphotericin B as the preferred initial therapy is mostly based on clinical experience because there have been no high-quality comparative studies of various antifungal regimens in the treatment of mucormycosis. In vitro activity provides further support for amphotericin B formulations as preferred initial therapy; amphotericin B is the most active antifungal agent against Mucorales in vitro, followed by posaconazole and isavuconazole [131].

Itraconazole has variable activity against Mucorales [131]. The antifungal agents voriconazole, fluconazole, flucytosine, and the echinocandins are not effective against the Mucorales [132-136].

The use of combination therapy in the initial treatment of mucormycosis is not routinely recommended given the lack of convincing data on efficacy [137]. The data on combination antifungal therapy are discussed separately. (See 'Combination antifungal therapy' below.)

Assessing response to therapy — We repeat radiographic imaging after one to two weeks of initial therapy (or after escalation of antifungal therapy) to evaluate for signs of disease progression or recurrence that would require additional debridement. In cases of sinus disease, we repeat endoscopic assessment for direct visualization of the tissues with any clinical progression or within the first two weeks of initiation of antifungal therapy and send repeat tissue biopsies for histopathology (to assess resection margins) and fungal culture.

Once patients have clinically improved and no further surgical debridement is planned, we switch to posaconazole or isavuconazole for maintenance therapy. (See 'Maintenance therapy' below.)

If radiographic or endoscopic evaluations suggest residual disease, we evaluate whether further surgical debridement is necessary. Serial debridement is often required for mucormycosis infections.

In patients who worsen clinically after at least one week of antifungal therapy, we re-evaluate for surgical debridement, optimize reversal of underlying conditions, and intensify antifungal therapy (eg, increase dosing to 10 mg/kg daily and/or consider combination antifungal therapy). We do not stop liposomal amphotericin B unless toxicities are especially severe. (See 'Aggressive surgical debridement' above.)

Maintenance therapy — For patients who have improved on a lipid formulation of amphotericin B, we suggest switching to posaconazole for maintenance therapy. Isavuconazole is an alternative for patients who cannot tolerate posaconazole (eg, drug-drug interactions, risk of QT prolongation, kidney dysfunction [IV posaconazole should be avoided in patients with creatinine clearance <50 mL/minute]). (See "Pharmacology of azoles", section on 'Drug-specific adverse effects'.)

When switching from amphotericin B to azole therapy, it is important to maintain a period of overlap with both drugs of approximately one week to reach therapeutic azole concentrations (confirmed by blood levels) before discontinuing amphotericin B. Both drugs are available in parenteral and oral formulations [132-134,138]. We prefer the oral route for maintenance therapy after an initial course of amphotericin B. However, in patients with impaired gastrointestinal tract function, critical illness, or inability to tolerate an initial course of amphotericin B, we use IV (instead of oral) azole formulations.

Both posaconazole and isavuconazole are broad-spectrum azoles with clinical efficacy and in vitro activity against Mucorales, but we prefer posaconazole, given lower minimum inhibitory concentrations for most mucor species [131]. There are no controlled trials comparing isavuconazole with posaconazole for the treatment of mucormycosis, and experience with both agents is limited.

●Posaconazole

•Dosing and administration – Both IV and oral delayed-release formulations of posaconazole are given as a loading dose of 300 mg every 12 hours on the first day, followed by a maintenance dose of 300 mg every 24 hours thereafter. When switching to oral posaconazole, we use delayed-release tablets taken with food [139]. We do not use the oral suspension of posaconazole, which has low bioavailability and requires fatty food for absorption. IV posaconazole should be avoided in patients with moderate or severe renal impairment (creatinine clearance <50 mL/minute) due to the potential for accumulation of the betadex sulfobutyl ether sodium (SBECD) vehicle. If assessment of the possible benefits and risks to the patient justifies use of IV posaconazole in patients with renal impairment, serum creatinine should be monitored closely, and, if increases occur, consideration should be given to changing to the extended-release tablet formulation of posaconazole or to isavuconazole (IV or oral). Adverse effects of posaconazole are discussed in more detail separately. (See "Pharmacology of azoles", section on 'Adverse effects' and "Pharmacology of azoles".)

•Checking serum trough levels – A serum trough concentration of posaconazole should be checked after one week of therapy; we suggest a goal trough concentration >3 mcg/mL, since higher levels are preferred for treatment of this serious infection. Therapeutic drug monitoring for posaconazole is discussed elsewhere. (See "Pharmacology of azoles", section on 'Posaconazole'.)

•Efficacy – Data on the efficacy are limited to case series and observational studies, many of which used the older (less bioavailable) oral suspension formulation of posaconazole.

The clinical efficacy of the oral suspension of posaconazole was shown in a salvage study that enrolled 91 patients who had failed or could not tolerate standard therapy [140]. Posaconazole led to complete or partial response in 60 percent of patients; 21 percent had stable disease. Although there are limitations to this salvage study, this series supports a potential role for oral posaconazole for the treatment of mucormycosis refractory to standard therapy.

●Isavuconazole

•Dosing and administration – For patients who cannot receive posaconazole, isavuconazole should be given as a loading dose of 200 mg (equivalent to 372 mg of the prodrug isavuconazonium sulfate) IV or orally every eight hours for the first six doses followed by 200 mg IV or orally every 24 hours thereafter. There are no known concerns about administering the IV formulation to patients with renal impairment because the IV formulation of isavuconazole is highly water soluble and does not contain the SBECD vehicle [141].

•Checking serum trough levels – Although isavuconazole levels are not routinely monitored for the treatment of other infections, we do check levels when isavuconazole is used for the treatment of mucormycosis. We check trough levels five to seven days after treatment (with a loading dose) is initiated, with a goal isavuconazole trough level of 5 to 10 mcg/mL. If a decrease in dose is warranted due to toxicity (eg, hepatotoxicity), we repeat an isavuconazole trough on the new dose to ensure levels above 2 mcg/mL.

More detail regarding adverse effects and therapeutic drug monitoring of isavuconazole is found elsewhere. (See "Pharmacology of azoles", section on 'Isavuconazole'.)

•Efficacy – Isavuconazole is the only US Food and Drug Administration (FDA)-approved antifungal for the treatment of mucormycosis, though it has not been studied in randomized trials. Isavuconazole was evaluated in a multicenter open-label single-arm study (the VITAL study) that included 37 patients with proven or probable mucormycosis [142]. Patients were treated with isavuconazole IV or orally. Median treatment duration was 102 days for patients with primary mucormycosis, 33 days for those with refractory mucormycosis, and 85 days for those with intolerance to other antifungal therapy. All-cause mortality through day 42 was 38 percent, and overall complete or partial response rate at the end of treatment was 32 percent for primary treatment and 36 percent for treatment of mucormycosis refractory to other antifungals. In a matched case-control analysis that compared patients who received isavuconazole for primary therapy of mucormycosis with contemporary controls who received amphotericin B (the majority received a lipid formulation), day 42 crude all-cause mortality was similar in those who received isavuconazole (7 of 21 patients; 33 percent) versus amphotericin B followed by posaconazole (13 of 33 patients; 39 percent) [142]. Weighted all-cause mortality was also similar in those who received isavuconazole versus amphotericin B followed by posaconazole (33 versus 41 percent). These data suggest that isavuconazole has some clinical efficacy in treating mucormycosis, but it is not possible to draw firm conclusions given the nonrandomized study design and the small study size.

Duration of therapy — For initial therapy, we continue lipid formulation of amphotericin B until the patient has shown signs of clinical improvement (which usually takes several weeks) and for whom no further surgical debridement is planned. Maintenance therapy with posaconazole or isavuconazole should continue until clinical signs, symptoms, and radiographic evidence of active disease have resolved, and predisposing conditions have been corrected (eg, optimized glucose control or restoration of immune function) [143]. Therapy often extends for months.

If immune function cannot be restored, patients require extended courses of maintenance antifungal therapy, and some remain on antifungal therapy for life. Ideally, chemotherapy or further immunosuppression should be held until after antifungal therapy is completed (or, if that is not possible, at the very least until patients are changed to maintenance therapy). (See 'Correction of predisposing conditions' below.)

Combination antifungal therapy — Combination therapy should be considered in patients who are worsening after at least one week of high-dose liposomal amphotericin B therapy (10 mg/kg daily); some experts also use combination therapy as initial therapy in patients with disseminated disease who are severely immunosuppressed (eg, hematopoietic cell transplant recipients).

Combination antifungal therapy consists of adding a second antifungal agent (either posaconazole, isavuconazole, or an echinocandin) to amphotericin B.

●Choice of second drug – When initiating combination antifungal therapy, we suggest adding posaconazole due to its broad in vitro activity against Mucorales. If posaconazole is not an option, isavuconazole is a reasonable alternative. If azoles cannot be used (eg, major drug-drug interaction or significant intolerance), we use an echinocandin as the second agent.

Posaconazole or isavuconazole are preferred over an echinocandin because these azole agents have intrinsic activity against Mucorales, whereas echinocandins do not. Moreover, patients who respond to combination therapy will ultimately switch to azole monotherapy for maintenance therapy; using an azole as part of combination therapy facilitates the transition to maintenance therapy.

Patients should be assessed for response to therapy after at least a week of initiating combination therapy. Assessing response to therapy is discussed elsewhere. (See 'Assessing response to therapy' above.)

●Rationale for combination therapy – Although no convincing data support combination therapy for mucormycosis, we use combination therapy in patients who are not improving on monotherapy and in severely immunocompromised patients with disseminated disease because of the high mortality rate and minimal additional risks of the second agents. The additive toxicity of an azole or echinocandin when given with amphotericin B is relatively small, and antifungal antagonism does not appear to be a significant concern with these combinations.

Echinocandin antifungal agents may have clinical utility when combined with amphotericin B [144], and successful clinical outcomes have been reported with this combination for the treatment of mucormycosis [145]. In vitro, echinocandins do not exhibit activity against the Mucorales [146-148] but display potential synergism with amphotericin B [149]. There is no role for combining an echinocandin with an azole in the treatment of mucormycosis.

In a review of 106 patients with mucormycosis and hematologic malignancies treated at a single center, similar six-week mortality was observed in the patients initially treated with combination therapy (liposomal amphotericin B with either caspofungin or posaconazole) compared with patients treated with liposomal amphotericin B alone [137]. Another single-center review of 101 cases of invasive mucormycosis showed similar rates of 90-day survival in the group of patients diagnosed between 1995 and 2003 (a period when 5 percent received combination amphotericin B plus an echinocandin) compared with the group diagnosed between 2004 and 2011 (a period when 31 percent received combination therapy with amphotericin B plus an echinocandin) [150]. In an observational review from 12 medical centers in India of 465 cases of mucormycosis, 82 percent of patients received amphotericin B monotherapy, and 11 percent of patients received initial combination therapy with amphotericin B and posaconazole; mortality was similar between groups [151].

No convincing clinical data support combination therapy for mucormycosis, and larger studies are needed to establish efficacy [152]. Mixed results from mostly observational data are limited by heterogeneous patient populations (different affected organs and extent of disease), surgical strategies, and antifungal regimens (monotherapy versus combination therapy with various agents, formulations, bioavailability, and dosing). For example, many older studies used the low bioavailability oral suspension of posaconazole, and within studies, some patients may have received liposomal formulation amphotericin B while others received amphotericin B deoxycholate. Additionally, observations in patients with disparate predisposing conditions may not be generalizable; for example, echinocandins are postulated to modulate the host neutrophil response to filamentous fungi in patients with diabetes, but the same effects may not occur in patients with neutropenia [153,154].

●Investigational therapies – Fosmanogepix and ibrexafungerp are antifungal agents under investigation, and both have in vitro activity against some of the Mucorales, and fosmanogepix has shown activity in some animal models of mucormycosis. There are no reports of these agents being used in clinical cases of mucormycosis.

Correction of predisposing conditions — Correction of predisposing factors for infection is critical for recovery from infection with mucormycosis. Predisposing conditions include hyperglycemia, metabolic acidosis, neutropenia, and other immunocompromising conditions. In patients with mucormycosis who also require iron chelation, we do not use deferoxamine because it has been associated with increased mortality in patients with mucormycosis.

Therapies not recommended

●Deferasirox – Deferasirox as an adjunctive agent for mucormycosis has not been adequately studied in humans, and therefore it should not be used. In contrast with the iron chelator deferoxamine, which increases the risk of mucormycosis (see 'Deferoxamine and iron overload' above), the iron chelating agents deferasirox and deferiprone do not act as siderophores and therefore do not increase the risk of mucormycosis. Studies in mice with mucormycosis found that these agents might actually be beneficial (eg, improve survival and reduce the tissue fungal burden) [155,156].

The possible utility of adjunctive deferasirox has been evaluated in small studies with mixed results. In an open-label study of deferasirox in combination with antifungal therapy, seven of eight patients survived [157]. In a small trial of 20 patients with proven or probable mucormycosis randomly assigned to receive liposomal amphotericin B plus either deferasirox or placebo for the first 14 days of therapy, 90-day mortality was higher in the deferasirox group (82 versus 22 percent) [158]. Reported adverse events were similar between the groups. One possible reason for the worse outcomes in the patients who received deferasirox is that more patients with hematologic malignancy, neutropenia, and/or pulmonary involvement received this agent; all of these conditions are associated with poor outcomes.

Further study is necessary to determine the possible benefits or harms of deferasirox, and we do not recommend its use.

●Hyperbaric oxygen – Hyperbaric oxygen has been used in some patients with mucormycosis, but the benefit of this therapy has not been established [75,159,160].

OUTCOMES — 

Despite early diagnosis and aggressive combined surgical and medical therapy, the prognosis for recovery from mucormycosis is poor (table 1). An exception is cutaneous mucormycosis, which rarely disseminates. Independent risk factors for mortality include disseminated infection, kidney failure, and infection with Cunninghamella species, while the use of surgery and administration of any antifungal agent are associated with a better outcome [5]. Even with treatment, mortality varies by site of disease and is highest for disseminated (96 percent), pulmonary (60 to 87 percent), and cerebral (62 to 79 percent) infection [5].

●Rhino-orbital-cerebral infection – A review of 208 patients with rhino-orbital-cerebral infection showed that the most significant factors associated with death were delayed diagnosis, the presence of hemiparesis/hemiplegia, bilateral sinus involvement, leukemia, kidney disease, and treatment with deferoxamine [75]. Overall mortality from rhino-orbital-cerebral mucormycosis ranges from 25 to 62 percent, with the best prognosis in patients with infection confined to the sinuses [5,161]. The prognosis is especially poor for patients with brain, cavernous sinus, or carotid involvement, although some patients with these complications have been cured of the infection [162-164]. Among COVID-19-associated cases, severe COVID-19 with need for mechanical ventilation appears to predict increased mortality [161].

●Pulmonary mucormycosis – The outcome in patients with pulmonary mucormycosis is worse than for patients with rhino-orbital-cerebral involvement, with mortality rates as high as 87 percent [5,31,78]. This may be in part due to underlying conditions and the inability to widely excise the involved tissues. Widely disseminated mucormycosis carries a mortality rate of 90 to 100 percent [5].

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Mucormycosis".)

SUMMARY AND RECOMMENDATIONS

●Microbiology – Mucorales are ubiquitous in nature and can be found on decaying vegetation and in the soil. The genera most commonly found in human infections are Rhizopus, Mucor, and Rhizomucor. Less commonly implicated genera include Cunninghamella, Saksenaea, and Apophysomyces. (See 'Mycology' above.)

●Pathogenesis – Invasive disease occurs when hyphae invade vasculature, which causes infarction and necrosis of host tissues. The organism typically enters via inhalation of spores, although other routes of infection can occur. (See 'Risk factors' above and 'Epidemiology' above and 'Clinical presentation' above.)

●Risk factors – At least one predisposing condition exists in almost all patients with invasive mucormycosis. The most common risk factors are diabetes mellitus (especially with ketoacidosis), glucocorticoid treatment, hematologic malignancy, hematopoietic or solid organ transplantation, deferoxamine treatment, iron overload, recent COVID-19 infection, trauma, and injection drug use. (See 'Risk factors' above.)

●Clinical syndromes – Signs and symptoms are usually rapidly progressive and vary depending on the organ(s) involved. (See 'Clinical presentation' above.)

•Rhino-orbital-cerebral infection – This is the most common clinical presentation of mucormycosis and is most commonly seen in patients with diabetes. Initial symptoms of acute sinusitis progress as the organism invades adjacent tissues including the palate, orbit, and brain. Besides sinusitis, signs include perinasal facial edema; necrosis of the palate, nasal mucosa, or skin; and orbital edema, proptosis, blindness, or ophthalmoplegia. (See 'Rhino-orbital-cerebral mucormycosis' above.)

•Pulmonary infection – Pulmonary mucormycosis is a rapidly progressive infection that causes pneumonia with infarction and necrosis. Patients typically present with classic symptoms of pneumonia plus hemoptysis. The infection can spread to contiguous structures, such as the mediastinum and heart, or disseminate hematogenously to other organs (see 'Pulmonary mucormycosis' above). Patients with hematologic malignancies or other forms of severe immunosuppression are most likely to present with pulmonary infection.

•Other sites of infection – Mucormycosis can also cause gastrointestinal, cutaneous, renal, and disseminated disease. (See 'Clinical presentation' above.)

●Diagnostic tests – To establish a diagnosis as soon as possible, mucormycosis should be considered early in the appropriate clinical setting, and invasive testing should be aggressively pursued. The diagnosis relies upon identification of organisms in tissue by histopathology with culture confirmation. However, culture often yields no growth, and histopathologic identification of an organism may provide the only evidence of infection. Polymerase chain reaction (PCR) may be useful for identifying the causative species when histopathology is positive but cultures are negative. (See 'Diagnosis' above.)

●Treatment – A combination of surgical debridement, antifungal therapy, and elimination of predisposing factors is usually necessary. Early initiation of antifungal therapy improves outcomes.

•Surgical debridement – Aggressive surgical debridement of involved tissues should be pursued as soon as the diagnosis of mucormycosis is suspected. (See 'Aggressive surgical debridement' above.)

•Initial therapy – For initial therapy of mucormycosis, we suggest a lipid formulation of amphotericin B (Grade 2C). For severely immunocompromised patients (eg, solid organ or hematopoietic cell transplant recipients) with disseminated infection, some experts use amphotericin B in combination with posaconazole, isavuconazole, or an echinocandin, given the historically very poor outcomes in this patient group.

We use 5 mg/kg daily of liposomal amphotericin B for most patients and use 10 mg/kg daily for patients with central nervous system (CNS) involvement or history of transplantation. Patients receiving liposomal amphotericin B should be monitored for side effects, including nephrotoxicity.

We continue amphotericin B therapy until the patient demonstrates clinical and radiographic improvement. (See 'Assessing response to therapy' above.)

•Maintenance therapy – For patients who have improved on a lipid formulation of amphotericin B, we suggest switching to posaconazole for maintenance therapy. Isavuconazole is an alternative for patients who cannot tolerate posaconazole (eg, drug-drug interactions, risk of QT prolongation, kidney dysfunction [intravenous (IV) posaconazole should be avoided in patients with creatinine clearance <50 mL/minute]). (See "Pharmacology of azoles", section on 'Drug-specific adverse effects'.)

When switching from amphotericin B to azole therapy, it is important to maintain a period of overlap with both drugs of approximately one week to reach therapeutic azole concentrations (confirmed by blood levels) before discontinuing amphotericin B.

•Duration of therapy – For initial therapy, we continue lipid formulation of amphotericin B until the patient has shown signs of clinical improvement (which usually takes several weeks) and for whom no further surgical debridement is planned. Maintenance therapy with posaconazole or isavuconazole should continue until clinical signs and symptoms and radiographic evidence of active disease have resolved, and predisposing conditions have been corrected (eg, optimized glucose control or restoration of immune function) [143]. Therapy often extends for months. (See 'Duration of therapy' above.)

•Correction of predisposing conditions – Hyperglycemia, metabolic acidosis, deferoxamine administration, and immunosuppressive conditions or medications should be resolved if at all possible. (See 'Correction of predisposing conditions' above.)

●Mortality – Overall mortality from rhino-orbital-cerebral mucormycosis ranges from 25 to 62 percent, with the best prognosis in patients with infection confined to the sinuses. The prognosis is especially poor for patients with brain, cavernous sinus, or carotid involvement, although some patients with these complications have been cured of the infection. The outcome in patients with pulmonary mucormycosis is worse than for patients with rhino-orbital-cerebral involvement, with mortality rates as high as 87 percent. (See 'Outcomes' above.)