| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2005;4:189-192. doi:10.1510/icvts.2004.105395 © 2005 European Association of Cardio-Thoracic Surgery
Optimum drainage method in descending necrotizing mediastinitisDepartment of Surgery, Teikyo University School of Medicine, 2-11-1, Kaga, Itabashi-ku, Tokyo, 173-8605 Japan Received 2 January 2005; received in revised form 26 February 2005; accepted 28 February 2005
*Corresponding author. Tel.: +81-3-3964-1231; fax: +81-3-5375-6097.
Descending necrotizing mediastinitis (DNM) is a rare but often fatal disease. Transcervical mediastinal drainage and transthoracic mediastinal drainage are the most commonly employed drainage methods for treating patients with DNM. It remains controversial as to whether transcervical mediastinal drainage alone would be adequate for the treatment of DNM, which is a life-threatening disease. Between 1996 and 2004, 13 patients with DNM were treated at our department. We performed transcervical mediastinal drainage in 6 patients with localized DNM, in whom the infection remained limited to above the level of the carina. A more aggressive approach, that is, transthoracic mediastinal drainage, was employed in the remaining 7 patients who had extensive DNM, with the infection extending below the carina. The overall mortality rate was 8%. All the 6 patients treated by transcervical drainage survived without major postoperative complications. Six out of the 7 patients treated by transthoracic drainage survived, while one died of pneumonia. Our results suggest that transcervical mediastinal drainage may be adequate for treating patients with localized DNM in whom the infection does not extend beyond the carina, while transthoracic mediastinal drainage must be adopted for patients with more extensive disease.
Key Words: Infection; Mediastinitis; Drainage; Surgery
Descending necrotizing mediastinitis (DNM), one of the most virulent forms of mediastinitis, occurs as a rare complication of oropharyngeal or dental infections [1]. The mortality rate associated with this condition has been reported to be in the range of 25–40%, and difficulty and delay in the diagnosis are believed to be the main causes for the high mortality rate [2]. The optimal mode of surgical drainage for DNM remains controversial. The most commonly employed drainage methods are transcervical mediastinal drainage and transthoracic mediastinal drainage. The latter is conducted via a standard thoracotomy or video-assisted thoracic surgery (VATS). Recent reports have suggested that transcervical mediastinal drainage alone may not be sufficient for the treatment of patients with DNM [3–6]. While some authors advocate routine use of combined transcervical and transthoracic mediastinal drainage [3,4,6], others contend that transcervical mediastinal drainage alone may be sufficient for patients whose infection localized to the upper mediastinum above the level of the carina [1,7,8]. However, the latter recommendation is based on the experience in only one or a few cases, and there are no large patient series in which DNM cases have been treated consistently, safely, and successfully by transcervical mediastinal drainage alone. In this study, we review the treatment outcome in DNM patients, focusing on the usefulness of transcervical mediastinal drainage.
Between 1996 and 2004, 13 patients with DNM (8 males, 5 females; average age, 59 years; age range, 40–72 years) were treated at our department (Table 1). Five of these 13 patients have been reported previously [9,10]. DNM occurred as a complication of peritonsillar abscess in 10 patients and dental infection in 3 patients. Four patients had underlying diabetes mellitus, 4 were smokers, and 1 was an alcohol addict.
All patients fulfilled the modified criteria of Estrera et al. for the diagnosis of DNM, i.e., (1) clinical manifestations of severe infection, (2) typical CT features, (3) surgical confirmation of necrotizing mediastinal infection, and (4) relationship between oropharyngeal infection and the necrotizing mediastinal process [1]. We divided our patients into two groups; those with localized DNM and those with extensive DNM, according to the extent of infection as determined by contrast-enhanced computed tomographic (CT) evaluation of the neck and chest. Signs of mediastinal infection on CT included: (A) unencapsulated fluid collections in the mediastinum; (B) mediastinal abscess; (C) mediastinal soft-tissue infiltration with gas bubbles [3]. Localized DNM was defined as infection localized to the upper mediastinum above the level of the carina (Fig. 1), and extensive DNM was defined as infection extending to the lower mediastinum beyond the level of the carina (Fig. 2).
Surgical mediastinal drainage was performed in all the patients immediately after the diagnosis of DNM. Transcervical mediastinal drainage alone was performed in patients with localized DNM, while a more aggressive approach, i.e., transthoracic drainage, was adopted in patients with extensive DNM. In transcervical mediastinal drainage, the superior mediastinum and the pretracheal, periesophageal and perivascular spaces were dissected manually through neck incisions. Then, one or more soft tubes or Penrose drains were positioned in the open wounds in the upper mediastinum. In general, we do not prefer to use the common silicone drains, because they are rigid and may erode major blood vessels to cause hemorrhage [11]. Transthoracic mediastinal drainage was performed via a right open thoracotomy or a right VATS approach. The right upper mediastinal pleura was opened longitudinally, and blunt dissection of the trachea, carina, periesophageal spaces, and perivascular spaces was performed. This procedure included radical debridement, decortication, and pleural drainage via chest tubes. Besides mediastinal drainage, all 13 patients underwent local neck drainage as treatment for the primary infection. Tracheostomy was performed in the event of rapidly worsening neck swelling or laryngeal edema. Broad-spectrum antibiotics (clindamycin with a broad-spectrum beta-lactam antibiotic in 10 cases, and carbapenem in the remaining 3 cases) were initiated empirically as soon as the diagnosis was suspected, and the drugs were changed as needed according to the results of bacteriological examination of the mediastinal exudates. Chest CT was repeated after the surgery to check for any signs of residual mediastinitis. The duration of drainage and antibiotic therapy was based on the clinical progress and the results of cultures of the mediastinal exudate and pleural fluid. Statistical analysis was performed using the StatView software, version 5.0 (SPSS, Inc). The unpaired t test was used to evaluate the differences between continuous variables according to the type of DNM. A two-tailed Chi-square test was used to analyze the differences between categoric variables. P<0.05 was considered to denote statistical significance.
In our series, there were 6 patients with localized DNM and 7 with extensive DNM. There were no significant differences in the clinical characteristics between patients with localized and extensive DNM (Table 2).
The mean preoperative serum CRP concentration was 20 mg/dl (range, 6–35) in cases with localized DNM, and 26 mg/dl (range, 4–52) in those with extensive DNM. The differences in values between the two groups were not statistically significant (P=0.4, Table 2). Transcervical mediastinal drainage alone was performed in the 6 patients with localized DNM, while a more aggressive approach, i.e., transthoracic drainage, was employed in the remaining 7 patients with extensive DNM. In 5 out of these 7 patients, mediastinal drainage via a right standard thoracotomy (3 patients) or right VATS (2 patients) was combined with transcervical mediastinal drainage. In the remaining 2, mediastinal drainage via right standard thoracotomy alone was performed. Three patients, all with extensive DNM, had empyema with foul-smelling purulent discharge in the right pleural cavity at the time of diagnosis. One of these three patients and one other patient in our series had reactive left pleural effusion; both were successfully treated by left intercostal drainage. None of the patients had complicating pericardial effusion that needed drainage. Tracheostomy was necessitated in 3 patients (50%) with localized DNM and 6 patients (86%) with extensive DNM. No statistically significant difference was detected between the two groups in terms of the frequency of tracheostomy (P=0.4, Table 2). The outcome was favorable in 12 out of the 13 patients. All of the 6 patients with localized DNM who were treated by transcervical mediastinal drainage alone survived. None required additional mediastinal drainage and there were no postoperative complications. The patients with extensive DNM also had favorable outcomes; 6 out of the 7 patients survived, and the mortality rate was 13%, regardless of the severity of the infection. The overall mortality rate in the series was 8%. There were 2 major postoperative complications; thoracic wound infection and pyothorax. Both occurred in patients with extensive DNM, and both were related to transthoracic mediastinal drainage. They were treated by additional drainage, debridement and irrigation, with highly favorable results. There was one death in this series; the patient was a male and 58 years old, with extensive DNM. He underwent mediastinal drainage via right open thoracotomy combined with transcervical mediastinal drainage and tracheostomy. However, he suffered serious brain damage caused by aspiration. He died of pneumonia on postoperative day 85. The mean duration of hospitalization in the 13 patients after mediastinal drainage was 46 days (range 19–102 days); it was 39 days (range, 25–75) in patients with localized DNM and 53 days (range, 19–102) in patients with extensive DNM. There was no statistically significant difference in terms of the length of hospitalization between the two groups (P=0.4; unpaired t test). Bacteriological examination revealed aerobic organisms in 7 patients, anaerobic organisms in 1 patient, and mixed aerobic and anaerobic organisms in 3 patients; in the remaining 2, the pathogens could not be identified. The most frequently isolated organism was Streptococcus viridans (6 patients).
When treating DNM patients, the decision on the type of surgical drainage to be employed is a crucial one. The most commonly employed methods are transcervical mediastinal drainage and transthoracic mediastinal drainage. Estrera et al. proposed that transthoracic mediastinal drainage is necessary once the infection extends beyond the fourth thoracic vertebra posteriorly or the carina anteriorly [1], and that transcervical drainage may be sufficient in other cases. However, recently, the tide of opinion has turned against transcervical mediastinal drainage alone. Marty-Ane et al. reported a favorable treatment outcome with a 16.5% mortality rate, and attributed their results to the early use of transthoracic mediastinal drainage [3]. Corsten et al. reported that the survival rate of patients undergoing combined transcervical and transthoracic mediastinal drainage (19% mortality) was significantly higher than that of patients undergoing transcervical drainage alone (47% mortality, P<0.05) [6]; they concluded that transcervical mediastinal drainage may be insufficient for treating DNM, and that combined transcervical and transthoracic mediastinal drainage should be considered as the standard for all DNM patients. Theoretically, transthoracic mediastinal drainage may be superior to transcervical mediastinal drainage for the treatment of DNM, because thoracotomy provides excellent exposure of all mediastinal compartments and allows radical surgical debridement and drainage. However, it is a more invasive procedure than transcervical mediastinal drainage and also carries the risk of pleural and wound contamination, as seen in two of our cases [9,10]. While the thoracoscopic approach is less invasive than thoracotomy, the risk of thoracic contamination is similar. Anterior mediastinotomy, that is, drainage through a subxiphoidal incision or a median sternotomy may be a useful alternative drainage method for DNM [5,8,12], especially when the abscess is confined to the anterior mediastinum; it has a theoretical advantage over thoracotomy in that pleural contamination and empyema can be avoided. However, anterior mediastinotomy provides unsatisfactory exposure of the visceral compartment and median sternotomy carries the risk of osteomyelitis. We consider that the optimal drainage method should be tailored to each patient's condition and needs further study. In our series, we consistently employed transcervical mediastinal drainage alone in patients with localized DNM, where the infection was localized to the superior mediastinum above the level of the carina. In patients with extensive DNM, however, we adopted the more aggressive approach of transthoracic mediastinal drainage. With this treatment policy, we were able to obtain favorable results in 12 out of the 13 cases. All the 6 patients who underwent transcervical mediastinal drainage alone survived. Their postoperative progress was uneventful, and none required additional mediastinal drainage. The overall mortality rate in the series was 8%. In this study, 2 patients with extensive DNM were treated by transthoracic mediastinal drainage alone, without transcervical mediastinal drainage. In both cases, the infection had extended deep into the mediastinum, beyond the level of the carina. In both cases, the primary infection sites and neck abscesses had already been treated by local drainage and well controlled by the time DNM was diagnosed. Therefore, we performed transthoracic mediastinal drainage alone in these patients. Both recovered fully without any postoperative complications. However, we regard these cases as exceptions, and recommend combined transthoracic and transcervical mediastinal drainage for all patients with extensive DNM. In our series, patients with extensive DNM tended to have higher serum CRP concentrations, require tracheostomy and a longer duration of hospitalization more frequently than those with localized DNM, although this trend did not reach statistical significance (Table 2). We consider this trend as being reflective of the greater virulence of the infection and difficulty in the treatment in cases of extensive DNM. In conclusion, our results suggest that while transcervical mediastinal drainage alone may be sufficient for the treatment of patients with localized DNM, in whom the infection is confined to the superior mediastinum, above the level of the carina. The more aggressive approach of transthoracic drainage must be employed for patients with more extensive infection extending below the level of the carina. However, since this study was conducted as a retrospective study and had only a small sample size, further study is needed to arrive at definitive conclusions. In addition to immediate surgical drainage, early diagnosis of DNM by CT and perioperative adequate broad-spectrum antibiotic therapy are also extremely important to obtain a favorable treatment outcome.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
