Discussion
This is the first study to report results according to airway management for patients with DNM secondary to DNI. The airway should be carefully monitored throughout the course of treatment with a flexible laryngoscope. We found that orotracheal intubation should be the first trial and delayed tracheotomy was better than early tracheotomy if needed.
When it comes to DNM secondary to DNI, it is imperative to diagnose the condition with CT (scanned from skull base to diaphragm) as quickly as possible and conduct immediate surgical drainage with a synchronous approach to the mediastinum.4 CT is important in making a diagnose of DNM, in determining the level of infection and the pathways of spread of infections from the neck to the mediastinum, and in planning a successful treatment.5 High risk factors for the development of DNM from DNI include older age (> 65 years), cervical necrotizing fasciitis, involvement of two or more cervical spaces, the presence of comorbidities such as chronic renal failure, diabetes mellitus, cardiovascular and pulmonary diseases, C-reactive protein ≥30 mg/dL, and neutrophil to lymphocyte ratio ≥ 13.6-8 In our study, 50% and 55 % of all patients had DM and pulmonary diseases, respectively.
Propagation of infection in the cervical space readily descends into the mediastinum as a result of gravity, respiration, and negative thoracic pressure. It has been known that inflammation in the cervical space readily descends into the mediastinum passing through potential space such as retropharyngeal, pretracheal, and carotid spaces. It has been reported that over 70% of cases of DNM spread via the retropharyngeal space.9 However, in our case series, 18 (90%) patients had involvement of the pretracheal space. The pretracheal space leads from the thyroid cartilage and the thyroid gland to the pericardium. Infection following this route can lead to suppurative pericarditis and empyema. Therefore, is would be dangerous to undertake tracheotomy if fulminant infection is present in the pretracheal space.
Primary surgical treatment for DNM secondary to DNI is aggressive drainage. This surgical approach to the mediastinum depends on the supra-carinal or the infra-carinal location of the disease.10,11 The operator should have the ability to open all relevant compartments. Failure of drainage can increase the morbidity and mortality. Experienced surgeon could shorten the overall general anesthetic time and decrease the morbidity and mortality.
According to one report including 185 cases of deep neck infection, 10.3% patients had upper airway obstructions.12 It developed more frequently in patients with retropharyngeal space abscess and Ludwig’s angina (36.4% and 27.3%). Three of 19 patients (15.8%) died of airway obstruction. Therefore, airway management in deep neck infection has been known as one of the most important determining factors for mortality. Ludwig’s angina, known as a bilateral inflammation of sublingual, submental, and submandibular spaces, is not rare in severe DNI. The gold standard of airway management in Ludwig’s angina is tracheotomy. However, Ludwig’s angina is not a risk factor for DNM. In addition, not all included patients originated from Ludwig’s angina in our series. Thus, we should consider individualized airway management in DNI. In Ludwig’s angina, neck infection or edema is usually confined to a suprahyoid lesion, rather than a descending infection, and compressed upper airway including the oropharynx and the larynx. On the other hand, in DNM secondary to DNI, neck edema literally descends to the thoracic diaphragm. There has been no report about the incidence of airway problems or management of patients with DNM.
In the airway management of a deep neck infection, tracheotomy has been recommended if fiberoptic intubation is infeasible, if the clinician is not skillful in the use of an awake fiberoptic intubation, or if intubation attempts have failed.13 In the treatment of DNM, tracheotomy is considered mandatory by some authors due to upper airway obstruction and repeated aspiration during swallowing.14 It is also suggested for the purpose of avoiding endotracheal intubation since its accidental dislodgement of the tube can be fatal because of the almost impossible task of re-intubation if massive upper airway edema is present. However, Karkas et al. have the development of tracheal stenosis in two (11.8%) of 17 patients after tracheotomy during DNM treatment.10 We did not experience tracheal stenosis, although the potential of this complication was higher in simultaneous (early) tracheotomy and cervicotomy. A recent report observed that the presence of bacterial biofilm and higher amounts of microbes in the tracheal tissue can increase the rate of laryngotracheal stenosis.15 In addition, tracheotomy is not always necessary because early cervical drainage and anti-inflammatory therapy could prevent upper airway obstruction in DNM.16 The present study also revealed that tracheotomy rather enhanced contamination of the pretracheal space and prolonged the duration of hospital stay. Among 7 (35%) patients with tracheotomy, mortality rates of early tracheotomy and late tracheotomy were 75% and 0%, respectively, with early tracheotomy showing significantly higher mortality (p = 0.032 ). Therefore, tracheotomy should be performed with great caution in patients with DNM.
Tracheotomy should be conducted when a long-term mechanical respiratory support is needed after controlling neck infection. Close clinical and radiological postoperative follow-up investigations are needed with early surgical re-intervention if necessary. An infected wound caused by sputum can be regarded as one of tracheotomy-related complications.17 In the present study, one patient with early tracheotomy developed communication between neck wound and tracheal lumen. It also aggravated pulmonary complications such as pneumonia. This patient could not survive from DNM. However, patients who underwent late tracheotomy did not have any wound infection caused by sputum because neck infection was controlled after the initial operation. In addition, diffusion of sputum onto the neck and the mediastinum can be prevented by suturing the tracheal wall with surrounding tissues during a late tracheotomy.
Keeping orotracheal intubation was safe and adequate after the initial surgery in our cases. The extubation was tried as early as possible (within 12 to 24 h after the surgical procedure) in the ICU. Therefore, endotracheal intubation should be performed as the first airway management in DNM without prolonged intubation. Conducting orotracheal intubation for an extended period is likely to cause complications such as laryngotracheal stenosis, especially in deteriorated patients.
This study has some limitations. The retrospective design of our study was its major inherent
limitation. In addition, statistical analysis was limited owing to our small sample size. This is
because DNM secondary to DNI is a rare disease. To eliminate various biases, a prospective
multicenter study for airway management in DNM is needed to render more reliable results.