Etiology, diagnosis, and management of descending necrotizing mediastinitis: a narrative review

Introduction

Background

First described in 1938 by Dr. Pearse, descending necrotizing mediastinitis (DNM) was recognized for its complex and morbid nature, with Pearse quoting an 85% mortality rate without appropriate source control through aggressive surgical management (1). Since then, significant advancements in recognition, diagnosis, and treatment have considerably lowered the mortality associated with this condition. However, this condition remains a serious requiring aggressive management due to its relatively high complication and mortality rates. Mediastinitis, as outlined by Pearse in his landmark paper, may originate from various sources and have such divergent manifestations that the term “mediastinitis” alone offers little clinical clarity. Among the many types of inflammatory processes that can occur in the mediastinum, although relatively uncommon, DNM remains the most severe.

Rationale and knowledge gap

Despite recent advances in diagnosis and management of DNM, it remains a profoundly morbid condition with no clear consensus on the optimal surgical management. Within the literature, there exists no prospective studies with the amount of retrospective reporting remaining minimal and with ongoing dispute with respect to the optimal surgical management of this disease.

Objective

The goal of the present review is to address the potentially subtle and insidious presentation of DNM, present current approaches to diagnosis and discuss the multiply described techniques to surgical drainage. We present this article in accordance with the Narrative Review reporting checklist (available at https://med.amegroups.com/article/view/10.21037/med-24-29/rc).

Methods

Our search of the literature was conducted between May 5 and June 10, 2024 and included the following databases: PubMed/MEDLINE, Western University Libraries, and Google Scholar databases. Literature used for this review was primarily obtained from PubMed/MEDLINE. A summary of our detailed search strategy is included in Table 1. Articles were screened and included if they either described: (I) the etiology of mediastinitis, including the important anatomy, patient profile, and implicating pathogens; (II) the diagnosis of DNM; and (III) the treatment of mediastinitis with particular focus on the variability in potential options for surgical drainage. Pertinent details of our search strategy are outlined in Table 2.

Content review

Etiology

Mediastinitis refers to any infectious or inflammatory process that occurs within the borders of the mediastinum affecting the connective tissue that fills the interpleural mediastinal space and surrounds the median thoracic organs (2). The process may range from a simple, non-suppurative process such as pericarditis or bronchitis to a severe diffuse purulent infection (1). Of the many types, DNM is the most severe requiring aggressive treatment and, even with advancements in treatments, this condition remains highly morbid. DNM refers to a clinical entity that, by definition, is preceded by an infection of an odontogenic or cervical origin that then spreads downwards to involve the mediastinum.

To understand the development of DNM, one must understand the basics of the anatomy of the cervical and thoracic fascial planes that facilitate the rapid and free spread of infection from the mouth and neck into the mediastinum. Essentially, there exists anatomic continuity between the cervical and mediastinal spaces that contain loose, poorly vascularized areolar tissue which lack immunologically active defense cells, thus serving as potential portals of entry facilitating spread of infection into the mediastinum (2).

Some authors argue that an absolute understanding of these pathways and anatomic planes is essential to fully comprehend the origin of the condition and thus dictate management (3). However, others have postulated that targeting these areas based on the suspected origin and propagation pathways for surgical management may result in incomplete drainage and insufficient source control (2). The main pathways that connect the two spaces allowing the spread of oropharyngeal infection into the mediastinum are the pre-tracheal, perivascular or perivisceral, and retropharyngeal spaces (1,3). Since the original description of the disease, the retropharyngeal space has been thought to be the most important route by which cervical infections progress into the mediastinum and is quoted to facilitate 70% of DNM (1).

The most commonly accepted origin of infection is from odontogenic sources. Often, second or third lower molar abscesses are implicated in deep neck infection (DNI) with subsequent progression to DNM (4). When the infection arises in this manner, the inciting event is known as Ludwig’s angina and is a rare, but potentially severe infection that may lead to DNM (5). However, in a relatively large series reported by Ridder et al., their most common source of DNM originated as pharyngeal infection. Less common sources of primary DNI that can cause DNM include peritonsillar abscess, epiglottitis, or lingual tonsillitis (2). There has been an association reported between radiation therapy to the pharyngeal area and the development of DNI, causing subsequent development of DNM (2).

The microbiology implicated in DNM is, not surprisingly, reflective of the normal microflora of the upper aerodigestive tract and become virulent only in certain conditions varying depending on the infectious origin. The most isolated aerobic bacteria tend to be various types of streptococcus species with streptococcus pyogenes being the most common; along with staphylococcus aureus (2). Anaerobic flora tend to consist of bacteroides, peptostreptococcus, and fusobacterium species (2). There have been reports of fungus, mainly candida, isolated in the setting of DNM but this is a rare entity and high suspicion for contamination should be considered unless the patient is otherwise immunocompromised (2). The polymicrobial nature of this infection in combination with the limited number of immune-competent cells found within the cervical mediastinum has been attributed to overall toxic nature of the disease (4).

Presentation

Prompt recognition of DNM can be challenging as the symptoms and signs of the infection present on a spectrum and are often subtle to non-existent until very late in the disease process. The classic symptoms often reported in the literature include chest pain, jugular venous distention, high fever, and crackling on palpation of the cervical area (2). However, the presence of these symptoms in isolation or together are rare with more common, and often earlier, symptoms including sore throat, neck pain, headache, general fatigue, globus sensation, dysphagia or odynophagia (4). Nonetheless, in a patient with oropharyngeal infection and progression of symptoms, clinicians must remain highly suspicious of this disease process with a low threshold for further investigation.

Common comorbidities associated with DNM are generally those that result in reduced tissue level oxygenation. These include diabetes, heart failure, respiratory insufficiency, obesity, peripheral artery disease, or previous radiotherapy of the neck (2). Further, there has been a strong association between DNM and patients who are chronic, heavy smokers, and have a history of drug and/or alcohol abuse (6).

Diagnosis

The diagnostic criteria for DNM was first proposed by Estrera et al. in 1983 and were further refined by Wheatley et al. in 1990 (2,7). These criteria remain in use today and consist of (7):

• Clinical manifestations of severe cervical infection;
• Demonstration of characteristic radiographic features of mediastinitis;
• Documentation of necrotizing mediastinal infection at operation or post-mortem examination (or both);
• Establishment of a relationship between oropharyngeal infection and the development of the necrotizing mediastinal process.

To satisfy criteria one and four, the most important step is a meticulous and thorough history and physical examination, which should include not only the presenting symptoms but also the delineation of the timeline and progression of these symptoms. Additionally, the diagnosis of odontogenic infection or DNI can be difficult and often requires consultation from otolaryngology or dentistry to examine the oral cavity and to directly visualize the larynx and pharynx using fibreoptic nasoendoscopy (4).

With respect to the radiographic features of mediastinitis, the main diagnostic tool accepted to make the diagnosis of mediastinitis (of any kind) is contrast-enhanced computed tomography (CT) scans of the neck and thorax. Common findings in keeping with DNM are a visible, organized fluid collection with or without gas formation, soft tissue thickening and enhancement with loss of the normal fat planes, reactive lymphadenopathy, septic vascular thrombosis, and can even include associated pleural or pericardial effusions (2). For patients with impaired renal function in whom there is significant concern for administration of intravenous contrast, or in children where there is concern for the impact of radiation, magnetic resonance imaging (MRI) is a suitable alternative.

The main classification system to describe DNM and potentially guide management was initially described by Endo et al. in 1999 and is based on the anatomical level of infectious involvement in the mediastinum (8). The classification system is relatively simple with type I infections limited to the area above the carina and type II specific to those that extend below (8). Type II is further classified into those involving the anterior lower mediastinum (IIA), and both the anterior and posterior lower mediastinum (IIB) (8). In 2021, Sugio et al. described a series of 225 patients and identified a previously undescribed subtype of DNM labelling it as type IIC (9). This group of patients were those with lower mediastinal disease only involving the posterior mediastinum (9).

Management

No definitive treatment for DNM has been clearly established. However, one theme that remains consistent in nearly all literature when managing DNM is a multidisciplinary approach involving thoracic surgery, otolaryngology, infectious disease, critical care, microbiology, and other teams, depending on the individual patient’s needs and relevant comorbidities.

Once a diagnosis of DNM has been made, the fundamentals of sepsis management in the critically ill patient remain consistent: ensuring airway protection, adequate oxygenation, blood pressure support, antibiotics, and source control. However, between each of these broad management concepts, there exists nuance that depends on the etiology, disease course, patient profile, and severity.

Firstly, since DNM primarily originates from oropharyngeal infections leading to DNIs and subsequently progressing to DNM, there is a significant risk of airway compromise. Early consultation with the otolaryngology, thoracic surgery, and critical care teams is therefore crucial in the initial stages of management to secure the airway before proceeding with definitive source control. Generally, these patients benefit from early, rather than late, intubation to ensure appropriate oxygenation and to facilitate the impending surgical management required to achieve source control (4).

There exists debate surrounding the routine use of tracheostomy for patients with DNM. Some authors believe that routine tracheostomy should be performed at the index operation for all patients with DNM due to the likelihood of prolonged necessity of invasive ventilation or potential for upper airway obstruction due to significant post-operative pharyngeal edema (10). However, some authors have called its routine use into question due to the risk of contamination of the tracheostomy site from the cervical wound needed for necessary drainage of DNI (11). Although no large series have specifically examined the routine use of tracheostomy in DNM management, the existing literature generally supports its liberal use in patients with significant medical comorbidities, particularly those with respiratory issues, who are expected to require prolonged mechanical ventilation in the intensive care unit (ICU) (2,4,11).

The optimal antibiotic choice in the management of DNM remains a difficult decision based on the polymicrobial nature of the infection and the necessity to cover aerobic and anaerobic bacteria. However, there is consensus regarding routine use of broad-spectrum intravenous antibiotics which are later tailored based on bacterial cultures obtained from surgical drainage (2,4,7,11). Common choices of empiric antibiotics include second and third generation cephalosporins with the addition of metronidazole or clindamycin for anaerobic coverage (2). An additional option is piperacillin-tazobactam to include coverage for both aerobic and anaerobic bacteria. Although in some reported series, there has been isolated fungal growth, there is no suggestion of routine use of anti-fungal medication in the empiric treatment of DNM (2).

One aspect of the management of DNM that remains universally accepted is that antibiotics alone are insufficient to control the disease necessitating the need for surgical source control. The approach to surgical management, however, varies depending on the origin and extent of disease and has also been a point of contention amongst surgeons treating this condition. The choice of surgical approach is crucial as the main cause of mortality in DNM has been quoted to be inadequate mediastinal drainage (12).

Surgical cervical drainage of DNI is a mandatory and non-controversial part of the management of DNM as inadequate cervical drainage will only continue to perpetuate the disease process. Following cervical surgical drainage, it is recommended to leave multiple, large surgical drains for ongoing source control with many authors opting to leave the wound open with daily packing and irrigation with anti-septic solution (2,4). This approach will also engage our nursing colleagues specializing in wound management as a part of the multi-disciplinary management of this complex disease.

The choice of approach to the thoracic mediastinum, however, remains a debated topic with options for thoracic drainage including trans-cervical, posterolateral thoracotomy, trans-sternal, sub-xiphoid and video-assisted thoracoscopic approaches.

Trans-cervical drainage of the mediastinum is a limited approach that provides access only to the superior mediastinum and may not permit adequate debridement and drainage of all purulent material (2,3,7,10-12). However, prior to 1990, nearly all cases of DNM were managed with transcervical drainage alone which likely contributed to its high mortality (13). Following this, in their retrospective review from 1997, Corsten et al. found a 47% rate of mortality with transcervical drainage alone and a 19% mortality rate with the addition of thoracic mediastinal drainage (14). However, a more recent and much larger retrospective series reported by Ridder et al. with 45 patients showed that the majority of their patients had disease limited to the upper mediastinum that were adequately drained with a trans-cervical mediastinal approach and a mortality rate of 11.1% (2).

Given this, however, multiple authors have condemned the isolated trans-cervical approach and routinely advocate for the addition of thoracic drainage, usually in the form of posterolateral thoracotomy (2). Ridder et al. argued that for disease confined to the superior mediastinum the additional surgical stress of posterolateral thoracotomy and intraoperative position changes may outweigh the risks of the potentially inadequate drainage from isolated trans-cervical drainage (2). The posterolateral thoracotomy offers access to the entire hemithorax, including all ipsilateral mediastinal structures and compartments allowing for adequate drainage (12). One potential disadvantage of posterolateral thoracotomy for drainage of the mediastinum is the lack of access to the contralateral hemithorax presenting challenges for patients with extensive, bilateral contamination (3). An early adaptation to this limitation, described by Ris et al., is the clamshell incision (15). This approach offers a wide surgical field with access to bilateral hemithoraces for extensive exploration and drainage, however, it is extremely aggressive and has a significant impact on ventilation mechanics in already compromised patients (3). Additionally, there is added risk of phrenic nerve hyperextension, diaphragmatic paralysis, and carries risk of sternal osteomyelitis (similar to a sternotomy for mediastinal drainage) (3).

Another way to achieve access to bilateral pleural cavities, with the addition of significantly less surgical morbidity is a video-assisted thoracoscopic surgery (VATS) approach to mediastinal drainage. In a series of 11 patients with a 0% mortality rate, Wakahara et al. approached DNM with the traditional cervicotomy for drainage of DNI and a “mini-thoracotomy” to facilitate thoracoscopic assistance to achieve a minimally invasive and efficient drainage (11). In a larger systematic review, 28 patients underwent mediastinal drainage with a VATS approach with an 8% mortality rate (16).

Min et al. in their 2004 case series reported four cases using cervicotomy to obtain control of DNI and VATS for mediastinal drainage. Although no control group was present, they showed a 0% mortality and the only post-operative complications were persistent pleural effusion with no need for return to the operating room (7).

In a larger retrospective cohort study, Tanaka et al. compared patients who underwent mediastinal drainage via VATS or thoracotomy over a 4-year study period with the primary outcome defined as 90-day mortality and the adjusted risk difference between VATS and thoracotomy (17). They found that for patients with a poor performance status pre-operatively, there was tendency towards a VATS approach to mediastinal drainage whereas patients with infection extending to both the anterior and posterior lower mediastinum, there was a tendency to approach this via thoracotomy (17). With respect to their primary outcome, there was no statistically significant difference noted with 90-day mortality rates of 4.8% and 8.6% in the VATS and thoracotomy groups, respectively, and an adjusted risk difference of nearly zero (17). Additionally, there was no difference in post-operative outcomes including volume of blood loss, length of hospital stay in days, indwelling time of mediastinal drainage (days), rates or duration of mechanical ventilation (17).

Although there exists limited literature on the use of VATS in the drainage of DNM, preliminary studies show promising outcomes with similar outcomes without the associated morbidity of thoracotomy. However, in approaching DNM with minimally invasive approaches, one must remember that the disease progresses rapidly and infection can often extend beyond the region identified on pre-operative imaging and therefore, in cases when sufficient drainage is questioned using a VATS approach, thoracotomy should be performed (17).

Although reported in the literature, sub-xiphoid and parasternal or trans-sternal approaches to mediastinal drainage have resulted in less favorable outcomes.

In their systematic review including a total of 480 patients, Prado-Calleros et al. used their findings to propose a management decision making algorithm based on the Endo type of DNM (16). Multiple authors advocate for decision making based on the Endo classification with type I being drained with cervicotomy alone and type II having the addition of, usually, thoracotomy. However, it is imperative to recognize that no consensus has been reached among surgeons regarding the optimal surgical management, and the debate continues. Regardless of the surgeon’s preference for drainage, it is crucial to consider the individual patient’s ability to tolerate extensive surgery and weigh this against the risk of potentially inadequate drainage.

Regardless of the approach used in the surgical management of DNM, the most important aspect of the care provided to these patients is adequate and often, aggressive, source control. In the setting of cervicotomy, this often necessitates large, sometimes bilateral incisions to accommodate the finger of the surgeon to perform blunt dissection to explore the area and drain all recesses that harbor purulence. Further, the dissection of the mediastinum, if performed from the neck, should start at the base of the mandible and extend down to the area of the tracheal bifurcation. If the origin of infection is a result of a dental infection, all infected or necrotic must be removed to attain appropriate source control. In short, in the face of a life-threatening infection, the surgeon must perform careful, yet often radical, surgery to achieve adequate source control from the mouth, neck, or chest.

Despite extensive reporting of various techniques for sepsis control in the setting of DNM and the potential utility of classification systems in guiding management, the importance of an individualized, patient centered, approach to the surgical management of DNM cannot be understated. As outlined by Coltro et al., all decisions made in the management of this complex condition should be individualized to the patient from empiric antibiotic choice, surgical approach, the choice of tracheostomy, and approaches to reconstruction, especially in the face of potential surgical complications (18). For example, in their presented case, they chose a median sternotomy and cervicotomy as their initial approach which allowed for appropriate sepsis control and stabilization of the patient, however, it was complicated by sternal dehiscence for which an individual decision was made for unilateral pectoralis major muscle flap was used for reconstruction to good effect (18).

Following surgical drainage of the mediastinum, patients generally require ongoing invasive ventilation in the ICU setting with an expected protracted hospital course. Some authors advocate for routine 48–72 hour post-operative contrast-enhanced CT scan of the neck and thorax to reassess the potential ongoing extent of disease and plan further management based on this (16). If there is expansion of the abscess or persistent localized abscess with worsening or ongoing septic shock, then re-exploration of the neck and/or mediastinum would be necessary (16). Another proposed approach is the liberal use of CT only if the patient’s condition deteriorates with operative decision making planned around this repeat imaging (2). Nonetheless, consensus in the literature acknowledges the likelihood of these patients requiring repeat drainage procedures for adequate neck or mediastinal sepsis control with a generally liberal approach to re-operation if the patient’s condition does not improve or significantly deteriorates.

As mentioned, these patients are expected to have a long, protracted course in ICU and hospital. While the multi-disciplinary management of DNM is essential for obtaining appropriate infection control, post-operative multi-disciplinary involvement is equally crucial. This includes the important addition of allied healthcare providers such as physiotherapy, occupational therapy, and speech-language pathology to address the expected post-operative deconditioning.

Limitations

In this study, several limitations warrant careful consideration of the findings. Primarily, the retrospective design of the included studies presents inherent challenges, as it may introduce biases that can influence the outcomes. These biases are often difficult to control or account for, potentially affecting the reliability of the results. Furthermore, the findings should be interpreted with caution, given that the scope of biases is extensive and beyond the authors’ capacity to manage effectively. Therefore, while the results provide valuable insights, they should be viewed within this context, encouraging further research to validate and expand upon these conclusions.

Conclusions

DNM remains a formidable and potentially fatal condition despite advancements in medical and surgical interventions. The complexity of DNM, originating primarily from oropharyngeal infections, necessitates a high index of suspicion for timely diagnosis and management. The evolution from initial infection to mediastinitis underscores the importance of understanding the anatomical pathways involved. The consensus underscores the critical role of multidisciplinary management, emphasizing early and aggressive surgical intervention combined with broad-spectrum antibiotic therapy tailored to the polymicrobial nature of the infection.

Optimal management strategies continue to be debated, particularly concerning the surgical approach. While trans-cervical drainage remains a cornerstone for cervical infections, its adequacy for mediastinal involvement is often supplemented by thoracic drainage techniques such as posterolateral thoracotomy or minimally invasive methods like VATS. Each method carries distinct benefits and risks, underscoring the necessity for individualized patient assessment and tailored intervention strategies.

The main focus of future work in this disease should focus on the optimal surgical approach to drainage with special consideration given to minimally invasive techniques such as VATS. Despite the current challenges, continued advancements in understanding and managing this condition offer hope for reducing its high morbidity and mortality rates.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://med.amegroups.com/article/view/10.21037/med-24-29/rc

Peer Review File: Available at https://med.amegroups.com/article/view/10.21037/med-24-29/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://med.amegroups.com/article/view/10.21037/med-24-29/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Pearse HE. Mediastinitis following cervical suppuration. Ann Surg 1938;108:588-611. [Crossref] [PubMed]
2. Ridder GJ, Maier W, Kinzer S, et al. Descending necrotizing mediastinitis: contemporary trends in etiology, diagnosis, management, and outcome. Ann Surg 2010;251:528-34. [Crossref] [PubMed]
3. Singhal P, Kejriwal N, Lin Z, et al. Optimal surgical management of descending necrotising mediastinitis: our experience and review of literature. Heart Lung Circ 2008;17:124-8. [Crossref] [PubMed]
4. Reuter TC, Korell V, Pfeiffer J, et al. Descending necrotizing mediastinitis: etiopathogenesis, diagnosis, treatment and long-term consequences-a retrospective follow-up study. Eur Arch Otorhinolaryngol 2023;280:1983-90. [Crossref] [PubMed]
5. Albacete Neto A, Coltro PS, Horácio GS, et al. Unilateral pectoralis major muscle flap for the treatment of sternal wounds due to Ludwig's angina. Int Wound J 2018;15:174-7. [Crossref] [PubMed]
6. Oruc M, Oruc K, Meteroglu F, et al. Descending Necrotizing Mediastinitis: Evaluation of 30 Cases. Indian J Surg 2023;85:465-70. [Crossref]
7. Min HK, Choi YS, Shim YM, et al. Descending necrotizing mediastinitis: a minimally invasive approach using video-assisted thoracoscopic surgery. Ann Thorac Surg 2004;77:306-10. [Crossref] [PubMed]
8. Endo S, Murayama F, Hasegawa T, et al. Guideline of surgical management based on diffusion of descending necrotizing mediastinitis. Jpn J Thorac Cardiovasc Surg 1999;47:14-9. [Crossref] [PubMed]
9. Sugio K, Okamoto T, Maniwa Y, et al. Descending necrotizing mediastinitis and the proposal of a new classification. JTCVS Open 2021;8:633-47. [Crossref] [PubMed]
10. Roccia F, Pecorari GC, Oliaro A, et al. Ten years of descending necrotizing mediastinitis: management of 23 cases. J Oral Maxillofac Surg 2007;65:1716-24. [Crossref] [PubMed]
11. Wakahara T, Tanaka Y, Maniwa Y, et al. Successful management of descending necrotizing mediastinitis. Asian Cardiovasc Thorac Ann 2011;19:228-31. [Crossref] [PubMed]
12. De Freitas RP, Fahy CP, Brooker DS, et al. Descending necrotising mediastinitis: a safe treatment algorithm. Eur Arch Otorhinolaryngol 2007;264:181-7. [Crossref] [PubMed]
13. Wheatley MJ, Stirling MC, Kirsh MM, et al. Descending necrotizing mediastinitis: transcervical drainage is not enough. Ann Thorac Surg 1990;49:780-4. [Crossref] [PubMed]
14. Corsten MJ, Shamji FM, Odell PF, et al. Optimal treatment of descending necrotising mediastinitis. Thorax 1997;52:702-8. [Crossref] [PubMed]
15. Ris HB, Banic A, Furrer M, et al. Descending necrotizing mediastinitis: surgical treatment via clamshell approach. Ann Thorac Surg 1996;62:1650-4. [Crossref] [PubMed]
16. Prado-Calleros HM, Jiménez-Fuentes E, Jiménez-Escobar I. Descending necrotizing mediastinitis: Systematic review on its treatment in the last 6 years, 75 years after its description. Head Neck 2016;38:E2275-83. [Crossref] [PubMed]
17. Tanaka Y, Maniwa Y, Sugio K, et al. The efficacy of thoracoscopic surgery for descending necrotizing mediastinitis. Interdiscip Cardiovasc Thorac Surg 2023;36:ivad053. [Crossref] [PubMed]
18. Coltro PS, Farina JA Junior. The importance of individualized surgical treatment of descending necrotizing mediastinitis. Transl Pediatr 2022;11:1283-4. [Crossref] [PubMed]