INTRODUCTION
The most complex and debated area in head and neck reconstructive surgery is the management of midface defects, which may arise from malignant or non-malignant diseases. This challenge extends beyond disease control or cure, encompassing the determination of optimal approaches to oral and facial reconstruction and rehabilitation. Ablative surgery in the midface often results in substantial psychological and physical trauma for patients and their families. The loss of an eye, for instance, impairs binocular vision, while vision changes from disease— especially when not caused by malignancy—can inflict severe psychological damage, including depression [
1]. These issues, combined with midface defects and associated aesthetic problems, can profoundly and sometimes irreversibly affect mental health and quality of life. Nevertheless, if disease management is effective, patients can achieve a satisfactory quality of life, maintaining good speech and swallowing. Historically, prosthetic options have been used to address maxillectomy defects, but over the past two decades, free-tissue transfer has become increasingly prominent. To date, two primary classification systems—the Memorial Sloan Kettering method and the M.D. Anderson method—have been traditionally employed to manage patients with midface defects. Both systems were originally developed for cancer patients, and their recommended treatments are based on oncologic protocols [
1,
2].
It is important to recognize, however, that key differences exist between cancer patients and those with midface defects from non-malignant diseases such as mucormycosis. In cancer, achieving a clear surgical margin is necessary, whereas, in mucormycosis, only the affected tissue is debrided. Following resection, cancer patients often require adjuvant or neoadjuvant therapies (such as chemotherapy or radiotherapy) and are at risk for recurrence, which can further complicate reconstructive efforts. In contrast, patients with mucormycosis, after successful debridement and definitive treatment, generally do not need further treatment. Another important distinction is that mucormycosis patients frequently have underlying conditions like diabetes, which may affect tissue reconstruction and surgical outcomes due to impaired circulation and immune compromise. Furthermore, the use of prostheses for reconstruction in these patients is associated with increased complications, so their use should be limited [
2,
3].
Mucormycosis often involves both the soft and hard tissues of the midface, sometimes leading to skin defects. However, for most patients, involvement is typically limited to a small area, which further differentiates their management from that of cancer patients. In oncologic surgery, excision of the entire skin of the affected midface region is performed to ensure adequate margins. Advances in reconstructive surgery—including the development of free flaps—highlight the need to distinguish between patients with midface defects from non-malignant diseases and those with cancer. The absence of a comprehensive classification system for non-malignant cases and a lack of tailored treatment strategies is therefore a significant gap. The development of a simplified algorithm is timely and necessary to guide rehabilitation that restores facial symmetry and optimizes oral function [
4]. Management of midface defects is further complicated by factors such as age and comorbidities (e.g., immunodeficiency due to diabetes), often necessitating a nuanced approach that may include avoiding reconstruction to reduce surgery-related morbidity. Informed consent, achieved through collaboration between patients and surgical teams, is critical in presenting rehabilitation options. This article shares the experience of a reconstructive surgery unit in Tehran, Iran, during 2021 to 2022, providing insights into midface defect management and presenting a reasoned algorithm based on a new classification system and corresponding rehabilitation strategies. The classification proposed herein is intended to predict both the functional and aesthetic outcomes associated with different surgical and conservative management approaches for each maxillectomy defect [
5-
7].
Predicting functional and aesthetic outcomes is particularly important as outcome assessments in head and neck reconstruction now place as much emphasis on functional and quality of life measures as on cure rates. The classification system presented in this publication is designed to anticipate the functional and aesthetic results of surgery and align these with the most appropriate obturation or reconstruction methods for each type of maxillectomy defect.
RESULTS
Out of the 107 patients included in this study, 45% were female. Among the patients, 52% had diabetes, 18% had diabetes while concurrently using corticosteroids, and 13% had no underlying disease. The average age of patients in the study was 54.2 years, with the youngest patient being 13 years old and the oldest 85 years old. All subjects exhibited mucormycosis concomitant with COVID-19 infection. Demographic information for the patient cohort is provided in
Table 1. In this study, we observed that 14 patients developed mucormycosis despite having no history of immunodeficiency or corticosteroid use. Among the patients, diabetes was the most frequent disease-related cause of immunodeficiency, affecting 56 individuals. Additionally, one patient had isolated involvement of soft tissue in the brain, and five patients exhibited bilateral facial involvement. This information is detailed separately in
Table 2.
Classification
During the preparation of this article, it became clear that existing classification systems do not adequately address midface defects caused by non-malignant diseases. Therefore, we propose a new classification based on our patient cohort and clinical experience (
Fig. 1). This method is grounded in anatomical principles and guides selection of the most appropriate treatment strategy. Our goal was to create a comprehensive system encompassing the spectrum of mucormycosis-related defects (
Fig. 2). Classes I–IV represent the progressive extent of midface defects, with soft tissue defects (Class I) and bilateral defects (Class IV) specifically included for midface reconstruction. Whereas the Memorial Sloan Kettering and M.D. Anderson methods were designed for cancer-related defects, the complexity of mucormycosis-induced defects—particularly when nasal bones are lost—requires composite reconstruction. The proposed classification addresses the unique challenges and reconstructive options for each defect type.
Type 1
This category refers to patients with soft tissue involvement (
Fig. 2A) and is subdivided into three groups based on the specific area affected:
· Type 1A: paranasal sinus involvement. Neither soft tissue nor bone involvement is present, and disease is limited to inflammation, resembling sinusitis. Medical management, with or without endoscopic debridement, is generally sufficient. Early surgical intervention combined with systemic antifungal therapy is recommended for rhino-orbito-cerebral mucormycosis. Restoration of immune function and glycemic control are also critical. Liposomal amphotericin B (5–10 mg/kg body weight) is the first-line therapy, with maximal doses suggested in cases involving the brain. Posaconazole serves as an alternative in cases of renal impairment, and may be added to amphotericin B for extensive or progressive disease. Antifungal therapy should continue until imaging shows complete resolution and the host immune system is restored. Once stabilized, intravenous therapy can be transitioned to oral medication. Posaconazole delayed-release tablets are preferred over suspension for efficacy. The total duration of antifungal therapy depends on disease extent and immune status.
· Type 1B: orbital content involvement, further divided into two subgroups: (1) Subtype 1B-a: with an orbital floor defect. Recommended management involves orbital exenteration followed by delayed reconstruction using a bone graft, temporalis muscle flap, and ocular prosthesis; (2) Subtype 1B-b: without an orbital floor defect. After orbital exenteration, delayed reconstruction may include a temporalis muscle flap and ocular prosthesis. Some patients exhibit orbital soft tissue involvement with both floor defect and maxillary involvement, which is addressed in Type 3.
· Type 1C: brain tissue involvement. The primary recommendation is medical therapy, with surgical intervention only indicated for complications such as intracranial hemorrhage (
Fig. 2A).
Type 2
Patients with nasal cavity involvement—including the medial maxilla, septum, and turbinates, with or without ethmoid and sphenoid involvement—are classified here. Treatment recommendations in this group depend on the presence of skin defects. In our study, none of the patients had skin defects; thus, reconstructive surgery was not performed. If the patient is asymptomatic, medical management is sufficient. However, reconstructive surgery, such as septorhinoplasty, may be considered for cosmetic deformities like saddle nose (
Fig. 2B).
Type 3
Maxillary involvement, subdivided into three categories:
· Type 3A: limited involvement of two to four maxillary walls (medial, anterior, lateral, and posterior) horizontally (
Fig. 2C). Neither the orbital floor nor the palate is affected. This corresponds to Type 1 in the Memorial Sloan Kettering (Cordeiro) classification. Based on our experience, and contrary to the Memorial Sloan Kettering approach, if the overlying skin remains intact, medical therapy is sufficient.
· Type 3B: subtotal involvement, divided into two subgroups: (1) Subtype 3B-a: subtotal involvement with an orbital floor defect. Suggested reconstruction is a radial forearm osteocutaneous graft. This matches Type 4 in the Memorial Sloan Kettering classification (orbitomaxillary involvement,
Fig. 2D); (2) Subtype 3B-b: subtotal involvement with a palatal defect. The Memorial Sloan Kettering classification distinguishes between involvement of more than or less than 50% of the palate. In our cohort, mucormycosis led to destruction of more than 50% of the palate (essentially total palate loss) in all cases, so we grouped all such patients together. If the ante-rior maxilla is also destroyed, a radial forearm osteocutaneous flap is recommended. If it is spared, a radial forearm fasciocutaneous flap may suffice (
Fig. 2E).
· Type 3C: total involvement of the maxilla, including both orbital floor and palate defects, further classified into two subtypes: Subtype 3C-a: with a zygoma defect (
Fig 2F). Subtype 3C-b: in the Memorial Sloan Kettering system, these cases correspond to Type 3, subdivided based on whether orbital contents are preserved or evacuated. Our recommended approach is a free musculocutaneous flap combined with bone grafting for orbital floor reconstruction.
Type 4
Bilateral maxillary defects. Such cases are not covered in current classification systems, likely due to the rarity of bilateral facial involvement in malignant disease. Each side should be assessed independently, with treatment tailored to the severity of involvement (
Fig. 2H).
Comparison with existing classifications
The M.D. Anderson classification emphasizes palatal defects and orbital exenteration, but does not account for cases involving the maxillary walls without palate loss or orbital exenteration. Therefore, it is not suitable for many mucormycosis cases. In our proposed system, we include palatal and orbital involvement, but—unlike the Anderson classification—do not further separate palate involvement due to the uniformity of destruction observed in mucormycosis. For patients with mucormycosis-related midface defects, obturators may be used as alternative treatments, similar to ocular prostheses. Although prostheses are artificial and patients with mucormycosis often have immunodeficiency and comorbidities, obturators remain a viable option. Therefore, in reconstructive surgery for these patients, regional flaps (such as temporalis muscle flaps) are our first-line approach. However, the availability of regional flaps is limited. Therefore, after comprehensive preoperative assessment, free flaps may be selected for some patients to achieve optimal reconstruction.
DISCUSSION
While mucormycosis is exceedingly rare in healthy individuals, we observed in this study that 14 patients with mucormycosis were otherwise healthy. Various immunocompromised conditions significantly increase susceptibility, including uncontrolled diabetes mellitus with or without diabetic ketoacidosis, hematological and other malignancies (and associated chemotherapy), prolonged neutropenia, immunosuppressive and cor-ticosteroid therapy, severe burns, AIDS, intravenous drug use, malnutrition, and open wounds due to trauma. Mucormycosis can affect multiple anatomical regions, such as the nose, sinuses, orbit, central nervous system, lungs (pulmonary), gastrointestinal tract, skin, jaw bones, joints, heart, kidneys, and, in invasive cases, the mediastinum. However, the most commonly encountered clinical form worldwide is rhino-orbital-cerebral mucormycosis [
8-
10].
Timely diagnosis and intervention remain the most critical factors in achieving improved outcomes. In most cases, diagnosis is only confirmed post-mortem, with up to 81% of patients relying solely on autopsy for definitive identification [
6]. Certain clinical symptoms—such as diplopia, ophthalmoplegia, proptosis, ocular pain, periocular cellulitis, acute vision loss, and the presence of naso-antral or cutaneous eschars in predisposed individuals—are recognized as highly predictive. A study found that indications for surgical debridement include ophthalmoplegia, proptosis, cranial nerve palsy, ocular involvement, periorbital and orbital cellulitis, and fever [
9,
11].
In 2021, Singh et al. [
8] published a systematic review encompassing global and Indian cases, aggregating data from 101 case reports of patients with both COVID-19 and mucormycosis. Consistent with our findings, hyperglycemia, new-onset diabetes, or diabetic ketoacidosis emerged as the predominant risk factors for mucormycosis in the study of Singh et al. Nasal and sinus involvement was the most common manifestation (88.9%), followed by rhino-orbital (56.7%) and rhino-orbitalcerebral mucormycosis (22.2%) [
8,
12].
Antifungal therapy may result in atypical histopathological features, making it difficult to distinguish
Mucorales from other filamentous fungi. Therefore, evaluation by an experienced pathologist is essential. Even if biopsy results are negative but the disease is clinically progressive, surgical debridement should not be delayed. CT imaging is a key tool for early detection of invasive disease and assessing bony involvement. However, a negative CT scan cannot exclude invasion [
13-
15].
A comprehensive classification system is essential for timely and appropriate treatment of each disease entity. The system proposed here facilitates a better understanding of the diverse presentations and conditions, allowing for more individualized treatment recommendations.
The acceptance of any classification depends on its utility, clarity, and simplicity. It should be logical, easy to remember, and minimize confusion when comparing different patient groups. In head and neck surgery, after survival, patients’ principal concerns center on function and aesthetics. Presently, there is no established classification for midface defects arising from non-malignant diseases. The proposed system is the first to incorporate both functional and aesthetic outcomes and their impact on rehabilitation choices for patients with midface defects due to mucormycosis. Attempts to construct highly detailed classifications for maxillectomy defects often lead to systems that are overly complex and impractical for clinical use. Although earlier systems attempted to account for factors such as the extent and level of maxillectomy, palate or nasal resection, skull base involvement, and anterior or posterior maxillectomy, the resulting classifications were unwieldy and provided little advantage over the more practical system proposed here [
5-
7,
16].
The extent of soft palate resection is particularly important for functional outcomes. In most maxillectomies resulting in an oro-antral fistula, preservation of the posterior rim of the soft palate maintains velopharyngeal competence and allows for effective obturation. In contrast, patients who undergo soft palate resection, including the posterior rim, and have full palate involvement due to mucormycosis, are left with a complete maxillectomy defect, frequently leading to persistent fistula and challenges in oral and dental rehabilitation.
In summary, this classification is specifically tailored for cases where the overlying skin remains intact, lesions are more localized, and nearly all patients have underlying diseases—factors that significantly affect the reconstructive approach and choice of flaps. Due to these unique features, we have adopted a simplified and pragmatic reconstructive strategy, recommending only two types of flaps: (1) pedicled temporalis muscle flap: suitable for smaller defects, especially in patients with significant comorbidities, as it provides reliable soft tissue coverage without the need for microvascular surgery; (2) free radial forearm fasciocutaneous or osteocutaneous flap: preferred for more complex defects requiring both soft tissue and bony reconstruction, offering versatility and minimal donor site morbidity.
In conclusion, this study represents a significant advancement in the treatment of mucormycosis, introducing a specific classification method designed to meet the unique needs of these patients for the first time. Patients with mucormycosis can achieve improved life expectancy when their underlying condition is properly managed. To assist surgeons in making optimal intraoperative decisions, an appropriate classification system for patients with mucormycosis is essential.