Skip to main content
Journal of the Korean Association of Oral and Maxillofacial Surgeons logoLink to Journal of the Korean Association of Oral and Maxillofacial Surgeons
. 2026 Feb 28;52(1):34–43. doi: 10.5125/jkaoms.2026.52.1.34

Individualized multidisciplinary management of severe odontogenic deep neck infections with and without descending necrotizing mediastinitis: a report of two cases

Shin-Won Hwang 1, Hyun-Jun Park 1, Da-Mi Yoon 1, Jung-Hyun Park 1, Heon-Young Kim 1,
PMCID: PMC12959261  PMID: 41755381

Abstract

Severe odontogenic deep neck infections (DNIs) can rapidly extend along cervical fascial planes and progress to descending necrotizing mediastinitis (DNM) with high mortality, requiring early diagnosis and multidisciplinary management, especially in elderly patients with comorbidities. We report two elderly patients with rapidly progressive odontogenic infections: one with DNM fulfilling Estrera’s criteria (Endo type IIA) and another with extensive DNI without mediastinal spread. Clinical features, serial contrast-enhanced computed tomography (CECT) findings, laboratory risk indicator for necrotizing fasciitis (LRINEC) score trends, microbiological results, surgical approaches were analyzed. Both patients presented with high LRINEC scores (≥8) and polymicrobial infections, involving multidrug-resistant organisms. Case 1 required five staged cervicothoracic debridements, negative pressure wound therapy, and delayed skin grafting. Case 2 underwent three surgeries with gland excision, hemorrhage control. Airway patency was maintained without tracheostomy through coordinated airway monitoring. Targeted antimicrobial adjustments and systemic optimization facilitated infection control. Both patients recovered fully without recurrence. These cases emphasize the importance of early CECT evaluation, cautious interpretation of LRINEC scores in patients with metabolic comorbidities, individualized surgical debridement tailored to anatomical spread, and multidisciplinary collaboration. Personalized reconstructive strategies based on defect size and tissue viability are essential for functional restoration and successful outcomes.

Keywords: Necrotizing fasciitis, Mediastinitis, Ludwig’s angina, Risk assessment

I. Introduction

Odontogenic infections are common causes of soft tissue infections in the head and neck region and most frequently originate from dental caries, pulpitis, pericoronitis, gingivitis, periodontitis, or endodontic infections1. While most odontogenic infections remain localized and can be successfully treated with antibiotics and appropriate dental interventions, delayed or inadequate management may allow spread along the fascial planes into the deep neck and thoracic regions1. The prevalence of maxillofacial space infections (MSIs) of odontogenic origin ranges from 50% to 89%2, with potentially serious complications such as airway obstruction, descending necrotizing mediastinitis (DNM), necrotizing fasciitis (NF), and sepsis1.

Deep neck infections (DNIs) are characterized by the involvement of multiple fascial spaces, most commonly the submandibular space, followed by the buccal, parapharyngeal, and submental spaces3. DNIs may arise from odontogenic infections, pharyngitis, salivary gland infections, trauma, or foreign bodies, and the mortality rate in complicated cases has been reported to be as high as 9.3%4. Historically, tonsillopharyngeal infections have been the predominant cause; however, because of the widespread use of antibiotics, odontogenic infections have become the most common etiology in both younger and older adults5.

When cervical infections progress further, they may extend into the mediastinum and develop into DNM, a life-threatening complication that disseminates rapidly through the danger space, retropharyngeal space, and anterior visceral space. The reported mortality rates remain high, ranging from 12.5% to 37.5%2. Another severe manifestation is NF, a fulminant infection that spreads rapidly along the fascial planes and causes extensive soft tissue necrosis. Early symptoms such as swelling, erythema, and tenderness are nonspecific, but progression is often rapid and destructive6. Recent studies have shown that a considerable proportion of cervical NF cases originate from odontogenic infections, which frequently arise as severe complications of DNI or DNM2.

In recent decades, advances in diagnostic imaging, surgical techniques, antibiotic therapy, anesthesia, and critical care protocols have led to marked reductions in morbidity and mortality among patients with DNI, DNM, and NF4. Early and accurate diagnosis, followed by prompt and aggressive treatment, is essential to prevent disease progression4. Contrast-enhanced computed tomography (CECT) remains the most reliable diagnostic tool for defining the extent of infection4. Estrera’s diagnostic criteria7 and Endo’s anatomical classification are widely applied to guide clinical judgment and surgical planning8. In addition, the laboratory risk indicator for necrotizing fasciitis (LRINEC) score serves as a useful adjunct for risk stratification and treatment monitoring9. Most importantly, the successful management of these complex infections requires a multidisciplinary approach that integrates expertise from oral and maxillofacial surgery, otolaryngology, thoracic surgery, plastic surgery, and critical care teams4.

This report presents two rare cases of severe odontogenic infections in elderly patients, one extending beyond the deep cervical spaces to involve the thoracic wall, and the other complicated by NF, underscoring the importance of early recognition, multidisciplinary management, and staged surgical intervention to achieve infection control and functional recovery.

II. Cases Report

This study was approved by the Institutional Review Board (IRB) of Ewha Womans University Mokdong Hospital (IRB No. EUMC 2025-05-015-004). Written informed consent for the publication of clinical information and images was obtained from both patients before manuscript preparation. All procedures were performed in accordance with the ethical standards of the IRB and the 1964 declaration of Helsinki and its later amendments. Patient confidentiality was maintained throughout the study.

1. Case 1

A 73-year-old male with a history of pneumoconiosis presented with a two-week history of progressive swelling in the left submandibular region, preceded by more than one month of severe pain in the left maxillary second molar. The tooth had been extracted at a local dental clinic two weeks earlier, but the patient was subsequently admitted to a secondary hospital because of persistent pain and swelling, where he received sequential intravenous antibiotics, including piperacillin/tazobactam, meropenem, and linezolid.

Despite inpatient management at the secondary hospital, his pain and cervical swelling continued to worsen. Consequently, the patient was then transferred to Ewha Womans University Mokdong Hospital to prevent further systemic deterioration, including respiratory distress. On admission, he presented with dysphagia, odynophagia, dyspnea, mild tachycardia (>100 bpm), and hypothermia (35.5°C). Physical examination revealed swelling and tenderness extending to the left temporal region, and a draining fistula was noted in the left buccal mucosa and overlying submandibular skin. CECT of the neck and chest revealed multiple abscesses in the left submandibular, buccal, and masticatory spaces. The infection had spread to the contralateral submandibular space, bilateral anterior cervical spaces, and anterior chest wall, accompanied by significant soft tissue swelling, fluid collection, and gas infiltration.(Fig. 1) Airway patency was maintained; however, the risk of airway compromise remained, and preparedness for airway intervention was warranted.

Fig. 1.

Fig. 1

Initial panoramic radiograph and contrast-enhanced computed tomography (CECT) images of the neck and chest. A. Panoramic radiograph showing multiple dental caries and retained roots in the mandibular molar region. B. Sagittal CECT image showing multifocal abscesses with gas formation extending along the submandibular, parapharyngeal, and retropharyngeal spaces (arrow). C. Coronal CECT image showing abscess extension into the submandibular, parapharyngeal, and carotid space, with marked inflammatory infiltration (arrow). D. Axial CECT image showing anterior mediastinal and anterior chest wall involvement, with soft tissue infiltration extending from the deep cervical spaces into the upper mediastinum (arrow).

Initial laboratory evaluation revealed a white blood cell (WBC) count of 30.36×103/μL, elevated C-reactive protein (CRP) level of 20.29 mg/dL, and hyponatremia (131.0 mmol/L). The LRINEC score was 9, indicating a high risk of NF. Based on the clinical and radiological findings, the case was consistent with Endo Classification type IIA and fulfilled Estrera’s diagnostic criteria.

The patient was immediately administered empirical intravenous ampicillin/sulbactam (3 g, every 8 hours), aggressive fluid resuscitation, and multidisciplinary management. On the following day, extensive incision and drainage were performed under general anesthesia in collaboration with the plastic surgery team. The airway was secured, and tracheostomy was not performed considering the extent and pattern of the infection. Multiple incisions were made in the submandibular, cervical, and chest wall regions, allowing the evacuation of a large amount of purulent material and necrotic tissue. Four drains were placed in the deep space medial to the mandibular angle, along the inner mandibular cortex, within the medial–inferior neck, and into the buccal space via the cutaneous fistula. A follow-up CECT scan on hospital day 8 revealed residual abscesses in the submandibular and anterior cervical spaces and anterior chest wall. Based on these findings, surgical exploration was performed in the operating room under general anesthesia, where the residual cervical abscess was accessed and drained through the previous incision at the level of the left cricoid cartilage. An additional lower incision near the thyroid level was created to reach and evacuate the inferior collection. Bleeding was minimal, and further drainage was completed. Cultures identified Corynebacterium striatum (C. striatum), leading to a change in antibiotics to piperacillin/tazobactam.

On hospital days 18 and 23, further surgical interventions were performed in collaboration with the thoracic surgery team, involving wide debridement of viable tissue margins, followed by the application of negative pressure wound therapy (NPWT) using CuraVAC (CGBio Co.). Based on sensitivity results, the antibiotic regimen was modified to include vancomycin. By hospital day 30, the inflammatory markers had improved, with a CRP of 4.68 mg/dL and WBC count of 9.46×103/µL. Reconstruction of the chest wall defect was performed using a split-thickness skin graft (STSG) harvested from the right thigh combined with mesh coverage and NPWT.

The patient was discharged on hospital day 42 after successful graft intake, infection control, and completion of an additional 14-day course of amoxicillin/clavulanate (1,000 mg twice daily). The decision to prescribe oral antibiotics was based on the finding that, although the patient’s WBC count was within normal limits at the time of transfer, the CRP level remained elevated at 4.68 mg/dL, suggesting residual inflammation. This approach was further supported by existing literature indicating that the duration of antibiotic therapy for DNM cannot be strictly defined, although a 7 to 15-day course following adequate source control is generally regarded as appropriate10. At the 3-month follow-up, clinical and radiological examinations confirmed complete resolution with no recurrence or graft failure.(Fig. 2) At the one-year follow-up, the patient remained in stable condition, with no evidence of infection recurrence or functional impairment. Functional recovery was excellent, with normal swallowing and speech, and the cosmetic results were satisfactory with minimal scarring, allowing the patient to return to normal daily activities. The overall clinical course of Case 1 is summarized in the timeline presented in Fig. 3.

Fig. 2.

Fig. 2

Serial clinical photographs of the anterior chest wall. A. Initial presentation showing skin changes consistent with necrotizing fasciitis. B. Immediate postoperative view after debridement and reconstructive surgery. C. Follow-up view showing complete healing after the split-thickness skin graft.

Fig. 3.

Fig. 3

Initial contrast-enhanced computed tomography (CECT) images of the neck and chest. A. Coronal CECT image showing abscess extension into the submandibular, sublingual, parapharyngeal, and carotid space, with associated inflammatory infiltration (arrow). B. Sagittal CECT image showing multifocal abscesses tracking along the submandibular, parapharyngeal, and retropharyngeal spaces (arrow). C. Axial CECT image showing multifocal abscesses involving the parapharyngeal and carotid spaces, with surrounding soft-tissue inflammation (arrow).

2. Case 2

A 68-year-old male with a history of hypertension, type 2 diabetes mellitus (HbA1c, 6.2%), and chronic renal failure requiring hemodialysis (three times weekly) presented with left submandibular swelling and progressive dyspnea. On admission, he presented with decreased consciousness secondary to significant deterioration in his systemic condition. Physical examination revealed extensive swelling and erythema in the left submandibular and cervical regions accompanied by cervical lymphadenopathy. Trismus, severe dysphagia with drooling, and skin changes extending to the upper chest were also observed.

Initial laboratory findings included a WBC count of 19.07×103/μL and elevated CRP level of 24.48 mg/dL, hemoglobin of 11.1 g/dL, sodium level of 137.0 mmol/L, and creatinine level of 2.97 mg/dL. The LRINEC score was 8, indicating a high risk of NF. CECT demonstrated cellulitis of the left submandibular space with early involvement of the deep cervical fascial spaces but no evidence of mediastinal spread.(Fig. 4) Therefore, this case was considered as a severe odontogenic DNI, and the Endo classification for DNM was not applicable.

Fig. 4.

Fig. 4

Timeline of clinical management in Case 1. This figure summarizes the chronological sequence of key clinical events in Case 1, including initial presentation, surgical drainage and debridement, identification of Corynebacterium striatum, subsequent antibiotic modifications, application of NPWT, chest wall reconstruction, and the overall course until discharge. (NPWT: negative pressure wound therapy, STSG: split-thickness skin graft)

The patient was immediately administered empirical intravenous ampicillin/sulbactam (3 g every 8 hours), and incision and drainage of the submandibular space were performed under general anesthesia with extraction of the presumed infectious sources, the mandibular third molar and second premolar. Hemodialysis scheduling was adjusted in collaboration with the nephrology team. Cultures of the drainage specimens obtained during surgery revealed Streptococcus constellatus (S. constellatus), Fusobacterium nucleatum (F. nucleatum), and Gram-positive cocci in chains. Based on infectious disease consultation, the antibiotic regimen was modified to piperacillin/tazobactam, which led to an improvement in fever and mental status.

Follow-up CECT on hospital day 6 revealed newly visible abscesses in the left masticatory, parapharyngeal, submandibular, and anterior cervical spaces. On the following day, extensive surgical debridement was performed by an oral and maxillofacial surgical team. Complete excision of the left submandibular gland and wide debridement of the necrotic tissue across multiple fascial spaces were performed, and drains were placed in the submandibular, anterior cervical, sublingual, parapharyngeal, buccal, and prevertebral spaces. Intraoperative cultures revealed S. constellatus, Staphylococcus epidermidis, Klebsiella pneumoniae (K. pneumoniae, extended spectrum beta-lactamase [ESBL]-positive), and Enterococcus faecium (E. faecium, vancomycin-resistant).

On hospital day 9, persistent bleeding from the surgical site was observed. CECT revealed a large irregular contrast-filled leak in the left submandibular space, highly suggestive of active bleeding, along with interval aggravation of the left parapharyngeal space abscess, a newly developed pterygoid muscle abscess, and persistent collections in the left masseter and anterior cervical spaces.(Fig. 5) The patient underwent an emergency reoperation for hemostasis with vessel ligation, additional debridement, and open-wound management. Based on the infectious disease consultation, the antibiotic regimen was modified to ertapenem (500 mg once daily) and vancomycin (500 mg every three days, adjusted for dialysis). Between days 17 and 28 of hospitalization, secondary wound healing progressed without the need for reconstructive procedures. On hospital day 28, the odontogenic infection resolved, and the patient was transferred to the Internal Medicine Department for further management. The overall clinical course of Case 2 is summarized in the timeline presented in Fig. 6.

Fig. 5.

Fig. 5

Timeline of clinical management in Case 2. This figure summarizes the chronological sequence of clinical events in Case 2, including initial surgical drainage, identification of Streptococcus constellatus (S. constellatus) and Fusobacterium nucleatum (F. nucleatum), detection of new abscesses, subsequent surgical drainage and debridement, isolation of Staphylococcus epidermidis (S. epidermidis), Klebsiella pneumoniae (K. pneumoniae), and Enterococcus faecium (E. faecium), modifications in antibiotic therapy, and confirmation of secondary wound healing prior to discharge on hospital day 28.

Fig. 6.

Fig. 6

Contrast-enhanced computed tomography (CECT) images showing progression of infection after initial drainage. A. Coronal CECT image demonstrating persistent and worsening inflammatory infiltration in the submandibular and submental spaces, with continued soft-tissue swelling and poorly resolving abscess components (arrow). B. Sagittal CECT image showing progression of cellulitis and inflammatory extension along the submandibular, parapharyngeal, and upper cervical spaces, indicating inadequate response to initial drainage (arrow).

A comparative summary of the demographic profiles, comorbidities, treatment courses, and sequential LRINEC parameters for Cases 1 and 2 is provided in Table 1.

Table 1.

Demographic characteristics, comorbidities, clinical course, microbiology results, treatments, and serial LRINEC score of the two cases

Case 1 Case 2
Characteristic
Age (yr)/sex 73/Male 68/Male
Comorbidities Pneumoconiosis Diabetes, HTN, CKD on dialysis
No. of surgeries 5 3
Total hospitalization (day) 42 28
Total ICU stay (day) 19 26
Estrera classification
A primary oropharyngeal infection
Radiographic evidence of mediastinal extension
Surgical/postmortem confirmation of mediastinal necrosis
Demonstrable causal relationship between the two sites
Yes
O
O
O
O
No
O
X
X
X
Endo classification Type IIA Not applicable
Main cultured organisms Corynebacterium striatum Streptococcus constellatus, Fusobacterium nucleatum, Enterococcus faecium, Klebsiella pneumoniae
Antibiotics administered Amp/Sul→Pip/Tazo→Pip/Tazo+Vancomycin Amp/Sul→Pip/Tazo→Ertapenem+Vancomycin
Reconstruction NPWT+delayed STSG Secondary healing
Outcome Discharged; no recurrence at 3-month follow-up Discharged; no recurrence at 1-month follow-up
LRINEC score parameters
Initial (admission)
CRP (mg/dL) 20.29 4 24.48 4
WBC (×103/μL) 30.36 2 19.07 1
Hb (g/dL) 12.9 1 11.1 1
Na (mmol/L) 131.0 2 137.0 0
Cr (mg/dL) 0.44 0 2.97 2
Glucose (mg/dL) 102.0 0 120.0 0
LRINEC score 9 (High risk) 8 (High risk)
After 1st operation
CRP (mg/dL) 17.43 4 28.40 4
WBC (×103/μL) 13.98 0 18.96 1
Hb (g/dL) 8.0 2 9.0 2
Na (mmol/L) 128.0 2 138.0 0
Cr (mg/dL) 0.30 0 2.30 2
Glucose (mg/dL) 141.0 0 98.0 0
LRINEC score 8 (High risk) 7 (Medium risk)
After 2nd operation
CRP (mg/dL) 6.89 0 15.05 4
WBC (×103/μL) 9.09 0 14.34 0
Hb (g/dL) 7.3 2 6.6 2
Na (mmol/L) 134.0 2 134.0 2
Cr (mg/dL) 0.56 0 2.37 2
Glucose (mg/dL) 112.0 0 175.0 0
LRINEC score 4 (Intermediate risk) 10 (High risk)
Before discharge
CRP (mg/dL) 4.68 0 8.10 0
WBC (×103/μL) 6.47 0 11.54 0
Hb (g/dL) 9.4 2 8.0 2
Na (mmol/L) 130.0 2 143.0 0
Cr (mg/dL) 0.52 0 2.28 2
Glucose (mg/dL) 75.0 0 137.0 0
LRINEC score 4 (Intermediate risk) 4 (Intermediate risk)

(LRINEC: laboratory risk indicator for necrotizing fasciitis, HTN: hypertension, CKD: chronic kidney disease, ICU: intensive care unit, Amp/Sul: ampicillin/sulbactam, Pip/Tazo: piperacillin/tazobactam, NPWT: negative pressure wound therapy, STSG: split-thickness skin graft, CRP: C-reactive protein, WBC: white blood cell count, Hb: hemoglobin, Na: sodium, Cr: creatinine)

III. Discussion

Two patients presented with severe odontogenic infections with rapid cervical extension. One case progressed to DNM, fulfilling Estrera’s diagnostic criteria and was classified as Endo type IIA. The other patient presented with severe DNI without mediastinal spread. The DNM patient required chest wall reconstruction and the other achieving healing by secondary intention. Both patients initially had high LRINEC scores, necessitating repeated surgical interventions, broad-spectrum antibiotics, and intensive care.

Odontogenic infections are now recognized as the leading cause of DNI, replacing tonsillopharyngeal infections as the predominant source of infection because of the widespread use of antibiotics5. The literature reports a wide variation in etiologic factors, with odontogenic origin accounting for 5%-63% of DNIs and up to 89% of severe multi-space infections4. These infections can progress to DNM or NF, both of which are associated with high mortality rates ranging from 11% to 40% for DNM2 and up to 76% for NF2. As the parapharyngeal and retropharyngeal spaces are anatomically continuous with the mediastinum, infection can readily spread downward, aided by gravity, respiration, intrathoracic negative pressure, and a lack of fascial barriers11.

Considering the rapid progression of deep neck and descending infections, early radiological evaluation is essential to assess the extent of the disease and guide timely surgical intervention4. Among the available modalities, CECT has been established as the gold standard for acute settings because of its speed, availability, and ability to clearly delineate anatomical structures4,12. Although magnetic resonance imaging and ultrasonography have emerged as supplementary tools, CECT remains the most reliable method for identifying abscesses, evaluating airway compromise, and defining the anatomical spread of infection4. Moreover, CECT findings allow clinicians to apply diagnostic frameworks, such as Estrera’s criteria7 and Endo’s classification8, thereby facilitating staging and determining the appropriate surgical approach. In the postoperative setting, follow-up CECT is invaluable for monitoring the progression of healing, assessing the treatment response, and identifying the need for additional surgical procedures4. In both cases, initial CT scans confirmed the spread of abscesses into the cervical and superior mediastinal spaces, and follow-up CECT clearly demonstrated interval changes in the infection, thus proving valuable for monitoring the treatment response.(Fig. 1, 4, 5)

Among hematologic markers, the LRINEC score has been widely used as an adjunctive tool for risk stratification. Wong et al.13 originally proposed this scoring system incorporating six variables (CRP, WBC, hemoglobin, sodium, creatinine, and glucose), suggesting that a score of ≥8 indicates a high risk of NF. While Wong et al.13 reported a positive predictive value (PPV) of 92%, subsequent reports have shown more variable results. For example, Tarricone et al.9 analyzed four studies and found that the diagnostic sensitivity of an LRINEC score >6 ranged from 36% to 77%, with a specificity between 72% and 93%. The pooled results yielded a sensitivity of 49.39%, specificity of 83.17%, PPV of 34.91%, and a negative predictive value of 89.99%. These findings indicate that the LRINEC score has limited sensitivity for early diagnosis and confirmation of NF but is useful for excluding NF and monitoring treatment response due to its relatively high negative predictive value. Both cases initially showed LRINEC scores ≥8, indicating a high risk of NF, but the scores decreased to 4 before discharge following surgery and antibiotic therapy.(Table 1; Fig. 7) This decline objectively reflected clinical improvement and resolution of the inflammatory process, suggesting that the LRINEC score can serve as a useful adjunct not only for initial risk stratification, but also for monitoring treatment response. Given these characteristics, the LRINEC score should be interpreted as a complementary rather than a standalone diagnostic tool, alongside thorough clinical assessment and definitive radiologic findings, especially in the early stages of rapidly progressing DNIs.

Fig. 7.

Fig. 7

Changes in laboratory risk indicator for necrotizing fasciitis (LRINEC) scores in Case 1 and Case 2. Line graph showing the LRINEC score trends of both cases during hospitalization. Case 1 (blue) and Case 2 (orange) both presented with high initial scores (≥8), followed by gradual declines after surgical management and antibiotic therapy, reaching 4 before discharge.

Several important limitations of the LRINEC score must be acknowledged, particularly in patients with underlying renal dysfunction. Because serum creatinine is one of the six variables, baseline elevation in patients with chronic kidney disease may artificially inflate the score independent of necrotizing infection severity—a consideration particularly relevant to Case 2, where renal failure was present. Prior studies have similarly warned that the LRINEC may be less reliable in patient groups prone to metabolic or electrolyte abnormalities14. Although we attempted to minimize variability by excluding the laboratory values obtained at the time of initial presentation and instead calculating LRINEC scores using the remaining fasting morning samples measured at consistent time points in both cases, the inherent structure of the scoring system still allows for potential overestimation or underestimation. These issues have prompted the development of modified scoring systems such as the m-LRINEC, which incorporates renal disease as an additional variable and has demonstrated improved diagnostic performance15. Therefore, LRINEC scores should be used as a complementary, rather than standalone, diagnostic tool and interpreted in conjunction with clinical and radiologic assessments, especially in patients with pre-existing metabolic or systemic comorbidities that may influence laboratory parameters.

The surgical approach for odontogenic infections extending deep into the neck and thorax is highly challenging and remains a subject of debate4. These infections often cross multiple fascial planes and are rarely confined to a single compartment, making complete eradication with a single operation difficult and frequently necessitating repeated incisions, drainage, and debridement of the necrotic tissue12. In our cases, the extent of debridement was determined based on preoperative CT imaging and intraoperative findings, with the goal of completely removing all devitalized tissue and ensuring adequate drainage of all involved spaces16. Debridement was continued until healthy, bleeding tissue was encountered. Re-operation decisions were guided by clinical, laboratory, and radiologic criteria16. Follow-up contrast-enhanced CT was performed 48-72 hours after initial debridement, or earlier if clinical deterioration occurred. Indications for re-debridement included new or worsening fluid collections on imaging, rising or persistently elevated inflammatory markers, persistent sepsis or hemodynamic instability, and clinical signs suggestive of ongoing infection. These same criteria were applied consistently in both of our cases16. Outcomes are further influenced by host-related factors, such as poorly controlled diabetes, chronic renal failure, obesity, human immunodeficiency virus infection, and other immunocompromised states, which can exacerbate disease severity and delay recovery12. Therefore, such comorbidities must be carefully considered when planning surgical strategies and predicting the prognosis. Among our patients, Case 1 underwent five surgical procedures, with a hospital stay of 42 days and an intensive care unit (ICU) stay of 19 days, whereas Case 2 underwent three surgeries, with a hospital stay of 28 days and an ICU stay of 26 days. Advanced age and multiple comorbidities in both patients may have contributed to their prolonged clinical course. In addition, in Case 1, prolonged hospitalization was required because discharge could not be considered until all infectious exudates from the thoracic drainage had completely resolved.

Airway management is another critical issue. Some authors advocate early prophylactic tracheostomy to prevent airway compromise, while others recommend reserving the procedure in cases of evident obstruction or impending failure4. In our cases, tracheostomy was avoided because the plastic and thoracic surgery teams considered it unsafe due to the potential risk of purulent discharge from the drainage catheter tracking into a tracheostomy site. With the support of the anesthesiology team, airway patency was successfully maintained without tracheostomy; however, full preparations for emergent airway intervention were in place. These considerations align with previous reports suggesting that tracheostomy should be avoided whenever possible when infection involves the anterior cervical or pretracheal spaces, as the procedure may increase the risk of anterior mediastinitis4,17. Collectively, these findings demonstrate that both surgical and airway management require an individualized approach tailored to the clinical presentation, radiologic findings, and patient’s underlying conditions.

Empirical antibiotics are selected based on statistical probability criteria, typically involving a single broad-spectrum agent or a combination regimen to treat both anaerobic and aerobic bacteria4. However, the increasing prevalence of multidrug-resistant organisms has made this selection difficult. In our cases, empirical antibiotics were initially administered; however, subsequent cultures identified diverse pathogens, including C. striatum, S. constellatus, F. nucleatum, E. faecium, and K. pneumoniae, prompting a switch to targeted antibiotic therapy, including piperacillin–tazobactam, ertapenem, and vancomycin. Severe odontogenic infections are often polymicrobial and may involve resistant organisms. Therefore, even after the initiation of empirical antibiotic therapy, prompt culture and susceptibility testing are essential to enable a timely transition to targeted therapy1. This approach is critical for achieving infection control and minimizing the risk of antimicrobial resistance and treatment failure. The importance of this approach is even greater in elderly patients with multiple comorbidities in whom resistant organisms are detected.

In advanced cervicothoracic infections, reconstruction should only be performed after complete resolution of the underlying infectious process and must be individualized according to the size and complexity of the residual defect12. In our cases, the wound conditions differed considerably after extensive debridement, necessitating distinct reconstructive strategies. In Case 1, the post-debridement defect was large, contaminated, and lacked sufficient viable soft tissue to support immediate primary closure or STSG. Therefore, NPWT was applied for approximately one week to promote granulation tissue formation, reduce wound contamination, and allow serial assessment of tissue viability18. Once adequate granulation was achieved and the wound bed had stabilized, delayed STSG was successfully performed. In contrast, Case 2 presented with a smaller residual defect and preserved soft tissue viability. NPWT or immediate STSG was not required, as infection control and granulation progression allowed for predictable closure through secondary intention. These cases underscore the need for individualized wound-closure strategies rather than a uniform approach, with decisions guided by the defect size, tissue viability, and degree of infection control19. For larger or more complex defects, particularly those involving the exposure of vital structures, the recruitment of vascularized tissue should be strongly considered using regional rotational flaps, such as the trapezius or pectoralis major, or even free flaps when necessary12. Overall, reconstructive planning must be closely integrated with infection control to achieve both functional recovery and long-term stability.

In our cases, successful treatment was achieved through multidisciplinary collaboration. The Department of Oral and Maxillofacial Surgery performed the extraction of the infectious source and extensive drainage, while the Department of Otolaryngology contributed to cervical drainage and airway evaluation. Thoracic surgeons managed mediastinal and chest wall involvement, plastic surgeons performed reconstructive procedures, and infectious disease specialists guided antibiotic adjustments based on culture and susceptibility results. Critical care and nephrology teams optimized the systemic conditions, particularly in the presence of significant comorbidities.

Several studies have reported that establishing a multidisciplinary treatment system reduces surgical delays, limits complications, and improves survival outcomes4. In particular, Prado-Calleros et al.20 presented a flowchart outlining the stepwise management of DNM, providing a practical algorithm that assists clinicians in making timely and structured decisions. Our experience demonstrates that even in elderly patients with multidrug-resistant infections, coordinated multidisciplinary management can achieve complete recovery. Therefore, individualized debridement tailored to the anatomical spread, culture-directed antibiotic therapy, and close interdepartmental collaboration are indispensable for treating deep cervical and descending infections. Future efforts should focus on enhancing early diagnostic algorithms and developing rapid and precise pathogen identification methods to optimize both empirical and targeted antibiotic strategies for severe polymicrobial infections.

Footnotes

Authors’ Contributions

S.W.H. and H.Y.K. obtained the data and wrote the manuscript. H.J.P. and D.M.Y. designed the study. J.H.P. and H.Y.K. reviewed and revised the manuscript. All the authors have read and approved the final version of the manuscript.

Funding

No funding to declare.

Ethics Approval and Consent to Participate

This study was approved by the Institutional Review Board (IRB) of Ewha Womans University Mokdong Hospital (IRB No. EUMC 2025-05-015-004). Written informed consent for the publication of clinical information and images was obtained from both patients.

Consent for Publishing Photographs

Written informed consent was obtained from the patients for publication of this article and accompanying images.

Conflict of Interest

No potential conflict of interest relevant to this article was reported.

References


Articles from Journal of the Korean Association of Oral and Maxillofacial Surgeons are provided here courtesy of Korean Association of Oral and Maxillofacial Surgeons

RESOURCES