Although mucormycosis was considered to be a community-acquired disease, nosocomial mucormycosis has been increasingly reported from many hospitals worldwide ( Rammaert et al., 2012; Saegeman et al., 2010). The most common clinical presentations of mucormycosis are rhino-orbito-cerebral (34%) and pulmonary (21%), the lungs being the second most common site of involvement in patients with mucormycosis ( Jeong et al., 2019; Prakash et al., 2019). The main sites of infection are the lungs (19%), skin and soft tissues (19%), paranasal sinus/sino-orbital region (15.8%), and rhino-cerebral (7.9%) ( Pana et al., 2016). A combination of prompt diagnosis, appropriate medical therapy and early aggressive surgical intervention could reduce the mortality to 25% ( Ibrahim et al., 2012; John et al., 2021). However, the mortality of mucormycosis remains high, perhaps because of delayed diagnosis, the high cost of managing the disease, and limited treatment options ( Skiada et al., 2018).

The incidence of mucormycosis is rising globally ( Ambrosioni et al., 2010; Bitar et al., 2009; Pana et al., 2016; Torres-Narbona et al., 2007). Significant risk factors include uncontrolled diabetes mellitus in ketoacidosis, other forms of metabolic acidosis, treatment with glucocorticoids, organ or bone marrow transplantation, neutropenia, trauma and burns, malignant hematologic disorders, and deferoxamine therapy in patients receiving hemodialysis ( Ibrahim et al., 2003; Spellberg et al., 2005; Sugar, 1990). For PM, haematological malignancy is the major risk factor (32–40%), followed by diabetes mellitus (32–56%), haematopoietic stem cell transplant (1–9.8%) and solid organ transplant (6.5–9%), and renal disease (13–18%) ( Feng & Sun, 2018; Tedder et al., 1994). There are recent changes in the epidemiology of the disease, especially in Asia, where tuberculosis and chronic renal failure are emerging risks ( Prakash & Chakrabarti, 2019).

During the pandemic, there was a surge in mucormycosis in patients with COVID-19 and in those who recovered from COVID-19 ( Ghazi et al., 2021). COVID-19-associated PM (CAPM) has been described in patients infected with SARS-CoV-2. Several factors have been implicated in the development of CAPM, including the use of glucocorticoids, poor glycaemic control, and viral-induced lymphopenia ( Pal et al., 2021). Thromboembolic events are also seen in COVID-19, including pulmonary embolism, deep vein thrombosis (DVT), cerebrovascular accidents (CVA), and myocardial infarction ( Abbas et al., 2021). The resultant hyper-coagulability and distal ischemia may well contribute to mucormycosis, although definitive proof for this remains elusive. The prime site of infection depends on the type of Mucorales and the comorbidity of the affected individual ( Jeong et al., 2019; Lanternier et al., 2012; Skiada et al., 2011). Bone and joint infections, peritonitis, and nephritis are rare forms of mucormycosis ( Serris et al., 2019). Endocarditis is also rare and is usually only seen in intravenous drug users ( Serris et al., 2019).

Climate change plays a role in the emergence of diseases ( El-Sayed & Kamel, 2020). It may result in a surge in known fungal diseases and the emergence of new fungal diseases as well ( Garcia-Solache & Casadevall, 2010). Climate change is defined by the United Nations Framework Convention on Climate Change (FCCC) as “a change of climate which is attributed directly or indirectly to human activity that alters the composition of the global atmosphere and which is in addition to natural climate variability observed over comparable time periods” ( Pielke, 2004).

The geographic range of fungal pathogens is expanding because of the increase in temperature at higher latitudes ( Gadre et al., 2022). Many fungal species are capable of developing thermotolerance and, as a result, fungi previously harmless to humans may become pathogenic ( Gadre et al., 2022). Sivagnanam et al. (2017) reported patient clusters with sino-pulmonary mucormycosis associated with changes in climatic conditions, such as changes in precipitation and temperature. Extreme climate events may also exacerbate the incidence of fungal disease - for instance, there are reports of necrotizing fasciitis caused by mucormycosis after the Indian Ocean tsunami of 2004 ( Andresen et al., 2005) and the Missouri tornado of 2011 ( Neblett Fanfair et al., 2012).

The incidence of mucormycosis in India and Pakistan is a hundred times higher than the rest of the world combined ( van Rhijn & Bromley, 2021). The higher incidence of co-morbidities such as diabetes mellitus accounts for some proportion of this significant difference in incidence ( Ghazi et al., 2021). However, even before the advent of Covid-19, there was an increase in the number of cases of mucormycosis in these South Asian countries, and the higher incidence of diabetes mellitus alone does not account for the increase ( van Rhijn & Bromley, 2021). Climate change in South Asia could be a contributing factor because fungi thrive at higher temperatures and in conditions of increasing humidity ( Pörtner et al., 2022; Safdar et al., 2019; van Rhijn & Bromley, 2021).

Individuals with defective immunity or increased available serum iron are more susceptible to mucormycosis. However, the disease has also been reported in healthy hosts ( Larsen et al., 2003; LeCompte & Meissner, 1947). Symptoms of PM include dyspnea, cough, chest pain, and fever is present in most reported cases ( Tedder et al., 1994). Angioinvasion leads to the necrosis of the pulmonary parenchyma, which may lead to cavitation and hemoptysis. This can be fatal if a major blood vessel is involved ( Harada et al., 1992; Watts, 1983). Although most patients with PM present with a fulminant and rapidly progressive disease, some patients only present with chronic symptoms ( Agarwal et al., 2006).

Early diagnosis is a cornerstone of successful management ( Fernandez et al., 2013). Clinical diagnostic methods have low sensitivity and specificity, which makes the diagnosis of PM challenging ( Skiada et al., 2018). Therefore, diagnosis requires a high index of suspicion, recognition of host factors, and analysis of the clinical presentation. Diplopia in a diabetic patient or pleuritic pain in a patient with neutropenia should arouse the suspicion of mucormycosis. The prompt use of imaging, as well as histology, microbiology, and advanced molecular methods underpin successful diagnosis and management ( Skiada et al., 2018).

Molecular diagnosis

In 2014, Gebremariam et al. investigated host immune responses to Mucorales. They discovered the genes for proteins that encode spore coatings – the CotH genes, which are only found in Mucorales ( Gebremariam et al., 2014). A 2018 study by Baldin et al. showed that CotH could be used as a target for the early diagnosis of mucormycosis ( Baldin et al., 2018). In this study, mice infected with Rhizopus delemar, Lichtheimia corymbifera, Cunninghamella bertholettiae, and Mucor circinelloides, were positive for CotH, detected through polymerase chain reaction (PCR), in 3/3, 1/3, 2/2, and 3/3 urine samples, respectively ( Baldin et al., 2018).

Real-time qualitative polymerase chain reaction (qPCR) can now be used to detect Mucorales DNA in blood samples. qPCR is non-invasive and allows mucormycosis to be diagnosed up to eight days before mycological diagnosis ( Bourcier et al., 2017; Millon et al., 2016; Millon et al., 2013; Springer et al., 2016). In 2022, Alinaghi et al. (2022) found that when mucormycosis occurred as a co-infection alongside COVID-19, the mortality rate was 33.6%. Furthermore, mucormycosis was diagnosed between 3 and 45 days after the diagnosis of COVID-19, and other co-infections such as aspregillosis occasionally occurred ( Alinaghi et al., 2022).

Chest X-ray

Radiology may only show peribronchial ground-glass opacity in the early stages, but later, pulmonary nodules, isolated masses, lobar consolidation, or cavitation may be found ( Kawakami et al., 2004; Lin et al., 2017; McAdams et al., 1997). Wedge-shaped pulmonary infarcts may also be found. These infarcts are usually seen when there is thrombosis of the pulmonary vessels, following fungal angioinvasion ( Marchevsky et al., 1980).

Chest CT scan

High-resolution chest CT scans are the preferred method to detect the extent of PM. These scans may show evidence of a Mucorales infection earlier than a chest X-ray. If an expanding mass is found, or a consolidation across tissue planes, especially towards the great vessels of the mediastinum, it is highly suggestive of mucormycosis ( Reimund & Ramos, 1994). Other suggestive findings include 10 or more nodules with pleural effusion ( Skiada et al., 2018). When a reverse halo sign is found on CT, it is more suggestive of PM than IPA ( Agrawal et al., 2020).

Bronchoscopy

The use of bronchoscopy in obtaining tissue biopsy for microscopic histological examination and culture seems to be safe in capitating lung lesions, despite its potential risk factors (e.g., pneumothorax) ( Smith & Kauffman, 2012). Bronchoscopic guided Trans-Bronchial Lung Biopsy (TBLB) is the most common modality used to obtain lung tissue for diagnosis ( Krishna et al., 2022). As far back as 1999, Lee, Mossad, and Adal undertook a 30-year review of patients with mucormycosis – of the 35 patients reviewed, bronchoscopy was required in more than 70% for a definitive diagnosis ( Lee et al., 1999). Bronchoscopy remains a mainstay of diagnosis even today. In 2022, the Delphi panel recommended early bronchoscopy in all suspected cases of this life-threatening disease ( Muthu et al., 2022).

Histology, Molecular Tests, and Biopsy

The diagnosis of mucormycosis is based on both histopathological findings of tissue invasion by hyphae and cultures isolating pathogens of the order Mucorales, most commonly Rhizopus, Mucor, and Rhizomucor species ( Smith & Lee, 2022). Many molecular based tests can be used either for detection or identification of mucorales, these tests include polymerase chain reaction (PCR) ( Hsiao et al., 2005), restriction fragment length polymorphism analyses (RFLP) ( Kumar et al., 2022; Page et al., 2021), DNA sequencing of defined gene regions ( Nyilasi et al., 2008; Springer et al., 2016), and melt curve analysis of PCR products ( Kasai et al., 2008).

While transbronchial lung biopsy (TBLB) is the most common method of obtaining lung tissue for diagnosis, specimen tissue can also be obtained by video-assisted thoracoscopic surgery (VATS) and open lung biopsy in patients undergoing surgical treatment ( Krishna et al., 2022). In patients with haematologic disease, it can be a challenge to perform a diagnostic biopsy early. In a study of 28 haematologic patients, Potenza et al. (2011) demonstrated that Mucorales-specific T cells were only detected in patients with proven invasive mucormycosis – these cells neither detected before the onset of infection nor after the infection was fully resolved. Therefore, when early diagnostic biopsy is a challenge, Mucorales-specific T cells can be an effective surrogate diagnostic marker.

Artificial intelligence (AI)

Syed-Abdul et al. (2022) showed that Artificial Intelligence-based models may be able to predict the risk of mucormycosis after COVID-19. AI models may be able to identify high-risk patients and flag predisposing factors such as obesity, anosmia, de novo diabetes, myalgia, and nasal discharge. Such models may enable protocols during and after COVID-19 that prevent co-infection with Mucorales ( Agrawal et al., 2020).

The effective management of depends on early suspicion or recognition of PM. Early diagnosis is critical to effective treatment. The essential components of the management of this disease include the reversal of underlying risk factors, such as correcting hyperglycemia or tapering glucocorticoids; early and aggressive surgical resection, if possible; and optimal anti-fungal therapy ( Pandey et al., 2020; Patel et al., 2020). Amphotericin B is a reliable antifungal that is used for primary anti-fungal therapy, with L-AmB widely in use as a less toxic form of the drug. The triazoles posaconazole and isuvaconazole are also effective against mucormycosis, and are used as primary therapies and in stepdown, maintenance, and salvage therapies ( Smith & Lee, 2022).

Surgical treatment

PM usually progresses rapidly. Surgery is often a necessity because of the massive amount of tissue necrosis that occurs during the disease. Treating the disease and killing these highly virulent fungi may not prevent or reverse the tissue necrosis ( Ibrahim et al., 2005). While surgery and anti-fungal therapies are the primary modalities of treatment, clear-cut guidelines regarding the extent and timing of surgical resection are yet to be defined. However, a successful attempt at methodological management of PM was made by the authors in the previously reported surgical series ( Pulle et al., 2021). Aggressive surgical resection with clear margins should be offered whenever feasible, as anti-fungal therapy alone may have sub-optimal outcomes. Surgical resection should be synergised with peri-operative anti-fungal therapy to improve long-term survival ( Kumar et al., 2022).

Antifungal therapy

Amphotericin B (Am-B) has been in use from 1959, as a reliable antifungal that is effective in treating progressive and potentially fatal fungal infections ( Cavassin et al., 2021). It is still an important option in the treatment of mucormycosis. However, Am-B has a high dose-dependent toxicity that can manifest as nephrotoxicity ( Deray, 2002) or reactions to infusions ( Cavassin et al., 2021). Am-B is only recommended now in settings where resources are limited, or it is the only treatment option available ( Cornely et al., 2019).

There is sufficient evidence in the literature to merit the early initiation of the lipid-based amphotericin B formulation, liposomal amphotericin-B (L-AmB), instead of Am-B, as the first line of treatment, in doses of 5 to 10 mg per kilogram per day ( Kontoyiannis & Lewis, 2006; Lewis & Wiederhold, 2003). Nephrotoxicity and infusion-related reactions are significantly less with L-AmB ( Hamill, 2013). Its peak plasma level is much higher than Am-B and the area under the concentration-time curve is larger ( Hamill, 2013). It is smaller in size and carries a negative charge, so it avoids substantial recognition and uptake by the mononuclear phagocyte system ( Hamill, 2013).

Recently, posaconazole and isuvaconazole have been used as primary therapies, usually in conjunction with L-AmB. Intravenous isuvaconazole, intravenous posaconazole, or delayed release posaconazole tablets are also recommended as salvage treatments, in moderate doses ( Cornely et al., 2019; Kohno et al., 2023). They are also used as stepdown or maintenance therapies, or as alternative therapies in patients who are unable to tolerate amphotericin B in any form ( Smith & Lee, 2022). It is worth noting that fluconazole, voriconazole and itraconazole have no activity on Mucorales ( Sun et al., 2002).

Iron chelating agents

Like many bacteria and fungi, Mucorales produce compounds known as siderophores that have a high affinity for iron – these compounds provide the fungus with the ability to acquire iron from the host ( Tahiri et al., 2023). Rhizoferrin is the most specific siderophore that Mucorales produce ( Tahiri et al., 2023). A recent experimental study in a murine model showed that rhizoferrin may increase fungal virulence ( Alejandre-Castañeda et al., 2022). Mucorales also produce permeases that have a high affinity for iron ( Stearman et al., 1996). These permeases can form copper-oxidase-ferrous-permease complexes to capture soluble iron and make it available to the fungus ( Stearman et al., 1996).

Iron chelators are medications that bind iron. There is increasing evidence that they may have a role to play in the management of mucormycosis, in starving the fungus of iron, which is essential for its high virulence and growth. ( Boelaert et al., 1994; Symeonidis, 2009). Mucorales species have a very high affinity for iron. Therefore, iron chelators that do not capture and bind iron as efficiently as the pathogen, may “donate” iron to the fungus, thereby increasing a patient’s susceptibility to mucormycosis. Indeed, haemodialysis patients treated with the iron chelator deferoxamine (DFO) are more susceptible to mucormycosis ( Mahalmani et al., 2021). However, there are new iron chelators, deferiprone and deferasirox, are even more efficient at binding iron than Mucorales fungi. In clinical practice, these new medications are used to treat patients with conditions such as thalassemia without increasing the susceptibility to mucormycosis ( Divakar Jose et al., 2021; Neufeld, 2006).

Deferiprone and deferasirox not only exhibit a very high affinity for iron, they also have a stable chemical structure and low molecular mass – these properties render them impervious to fungal iron uptake systems ( Boelaert et al., 1994; Symeonidis, 2009). The fungus is unable to “steal” iron from these molecules the way it is able to with older iron chelators like DFO ( Boelaert et al., 1994; Symeonidis, 2009). It is possible that these new iron chelators bind iron so strongly that they are actually able to detach iron from fungal iron uptake to inhibit fungal growth ( Symeonidis, 2009). This has been proven in vivo, using animal models ( Symeonidis, 2009). There is anecdotal support as well – a case report in 2006 showed that salvage treatment with deferasirox had clear therapeutic benefits in rhino-cerebral mucormycosis ( Reed et al., 2006). However, Spellberg et al. (2012) found that a combination of L-AmB and deferasirox increased mortality in haematological malignancies. Even so, Sipsas et al. (2018) note thatdeferasirox may be beneficial in diabetic patients with mucormycosis, especially in ketoacidosis where an acidic pH increases unbound tissue iron. Further study is warranted to evaluate the role of iron chelators in the management of mucormycosis.

Cytokines, Granulocyte Transfusions and Hyperbaric Oxygen

Cytokines, granulocyte transfusions, and hyperbaric oxygen are promising new therapies. Cytokines such as IFN-γ, G-CSF, and GM-CSF enhance host immunity by activating granulocytes that damage fungal cells ( Spellberg et al., 2009). IFN- gamma and GM-CSF are also known to enhance the activity of polymorphonuclear leukocytes to augment the damage to fungal hyphae ( Gil-Lamaignere et al., 2005). In neutropenic hosts, granulocyte transfusions can help, as they can deliver polymorphonuclear leukocytes to the site of the infection ( Spellberg et al., 2009). L-AmB acts synergistically with polymorphonuclear leukocytes to damage the hyphae of some Rhizopus species ( Simitsopoulou et al., 2008).

Hyperbaric oxygen (HBO) enhances the oxidative action of amphotericin B by correcting lactic acidosis, at pressures over 10 atmospheres absolute (ATA) ( Gebremariam et al., 2014; Siddiqui et al., 1997). HBO may enhance leukocyte-mediated phagocytoses and tissue repair ( Chakrabarti & Singh, 2020). However, there is a paucity of data on the use of HBO in PM ( Grigull et al., 2006; John et al., 2021; Spellberg et al., 2009). It has been used as adjunctive therapy for rhino-orbital-cerebral mucormycosis and soft tissue infections, but it is much less commonly used for PM and other manifestations of the disease ( John et al., 2021). The recommended dose of hyperbaric oxygen therapy consists of two sessions of 90–120-min each, at 2–3 ATA, per day ( Chakrabarti & Singh, 2020). HBO reportedly benefits patients with diabetes or in trauma, but is less effective in stem cell transplantation and in patients with haematological malignancies ( Chakrabarti & Singh, 2020). Randomized controlled trials may prove the efficacy of HBO in mucormycosis ( Brunet & Rammaert, 2020).

This systematic review was designed to answer the following research question: ○

What recent treatment protocols have been reported in the literature that can be deployed in the management of pulmonary mucormycosis?

All articles relating to treatment protocols for PM were included. The concept is treatment protocols in the context of PM. The population of interest is the human population. The following inclusion criteria were applied: •

Articles published in English.

Articles that focus on limited treatment regimens or topics that are not relevant to the research question, were excluded. For example, articles focusing on the following topics were excluded: •

Articles that only mention limited treatment regimens for PM, such as the use of amphotericin B only.

Four electronic search systems were identified: Scopus, PubMed, ScienceDirect, and Web of Science. Scopus offers excellent coverage ( Falagas et al., 2008; Kulkarni et al., 2009) and is the most effective search engine for the overview of a topic ( Tober, 2011). ScienceDirect is also a highly effective, second only to Scopus as a search engine ( Tober, 2011). PubMed, a subset of Scopus, is the optimal search system for biomedical topics - it is the most frequently updated and includes early releases of articles ( Falagas et al., 2008; Kulkarni et al., 2009; Samadzadeh et al., 2013). PubMed was included for this reason, to ensure a more precise and in-depth search.

Science Direct is second only to PubMed in terms of precision, recall, and relevance of search results ( Samadzadeh et al., 2013). Therefore, ScienceDirect was also selected as an appropriate search engine for this systematic review. Web of Science has a well-documented performance in tracking citation counts ( Falagas et al., 2008; Kulkarni et al., 2009) and was therefore selected as a suitable search engine for this review. While Google Scholar is also effective in terms of recall and relevance, the search results it provides are inconsistent, and citation information is often inadequate or not updated ( Falagas et al., 2008; Kulkarni et al., 2009). Therefore, Google Scholar was not selected for this systematic review.

For each of the four search systems selected, search strategy/search terms were defined ( Table 1). Within these search systems, all databases were included in the search. After the search results were assessed and articles relevant to the research question identified, citation tracking (snowball search) was also undertaken – a search of references within the relevant articles identified.

The second and third authors reviewed this. The extracted data was cross-checked by all authors to minimise personal bias ( Page et al., 2021). Any disagreements on data extraction and the categorisation of articles were resolved through detailed discussions, leading to a consensus between the authors.

All three authors developed the concept and idea for this systematic review. The first author provided expert guidance on this review and verified the methods and results. The second and third authors conducted the search, identified, and removed duplicates from search results, and read the abstract and titles of the remaining articles, reading the full text of those articles where abstracts were unavailable. The second and third authors then applied the exclusion and inclusion criteria to the remaining articles, to ensure that only relevant articles were selected.

The selected articles were analysed to identify broad qualitative themes. A descriptive analysis was also undertaken to: •

identify recent treatment protocols mentioned in the literature reviewed; and

The authors confirm that this paper is an accurate and transparent account of the systematic review conducted. The review’s design and analysis were not preregistered.