Eur J Cardiothorac Surg 2008;34:520-524. doi:10.1016/j.ejcts.2008.05.034
Copyright © 2008, European Association for Cardio-thoracic Surgery. Published by Elsevier. All rights reserved.
Treatment of mediastinitis using video-assisted thoracoscopic surgery
Jeong Su Choa,
Yeong Dae Kimb,*,
Hoseok Ib,
Sang Kwon Leeb,
Yeon Joo Jeongc
a National Masan Hospital, Masan, Republic of Korea
b Department of Thoracic and Cardiovascular Surgery, Pusan National University Hospital, Republic of Korea
c Department of Radiology, Pusan National University Hospital, Busan, Republic of Korea
Received 18 January 2008;
received in revised form 10 May 2008;
accepted 19 May 2008.
* Corresponding author. Address: Department of Thoracic and Cardiovascular Surgery, Pusan National University Hospital, #1-10, Ami-dong, Seo-gu, Busan, Republic of Korea. Tel.: +82 51 240 7267; fax: +82 51 243 9389. (Email: domini{at}pnu.edu).
 |
Abstract
|
|---|
Background: Mediastinitis remains a life-threatening disease that is difficult to manage and has a poor prognosis. This is especially true of descending necrotizing mediastinitis, which before the 1990s era had a mortality of approximately 40% despite the use of antibiotics, surgical techniques, and intensive care monitoring. Several authors have recommended that aggressive surgical approaches for mediastinitis are more effective than simple surgical approaches; however the best surgical option for mediastinitis remains controversial. Materials and methods: In a retrospective analysis between January 2000 and June 2006, 17 patients who underwent surgical debridement and drainage using video-assisted thoracoscopic surgery are included in this report. Data extracted from medical records included sex, age, origin of infection, surgical intervention, progress, and outcome including cause of death. There were eight men and nine women. The mean age was 52 years old (range, 20–72). Results: The origins of infection included esophageal perforation in nine patients and odontogenic or peritonsillar abscesses in the remaining eight patients. Among them, two patients required conversion to thoracotomy during operation in cases of mediastinitis due to esophageal perforation, so excluding the two patients, we calculated next five data for 15 patients. The mean duration from onset of symptoms to surgery was 12.4 ± 13.1 days (range, 0–43) and the mean duration from the initial operation to discharge was 43.6 ± 24.4 days (range, 8–113). There was serial operation in one case. There were three cases of postoperative mortality. Conclusion: Mediastinal drainage using video-assisted thoracoscopic surgery with or without cervical drainage can be a feasible and effective surgical option. This less invasive technique seems to have an outcome similar to more aggressive open surgical approaches for patients with mediastinitis previously reported in the literature, although it has not been directly compared.
Key Words: Mediastinitis • Thoracoscopic surgery • Less invasive
|
1. Introduction
|
|---|
Mediastinitis is a surgical emergency with a high mortality rate. Mediastinitis may begin primarily from structures in the mediastinum, or it may be the result of an infection extending downward from the oropharynx, in which case it is called descending necrotizing mediastinitis (DNM). The criteria for the diagnosis of DNM are clearly defined by Estrera and colleagues [1] and include the following: (a) clinical evidence of severe oropharyngeal infection; (b) characteristic roentgenographic features of mediastinitis; (c) documentation of necrotizing mediastinal infection at the time of surgery or at postmortem; and (d) establishment of the relationship between the descending necrotizing mediastinitis and the oropharyngeal infection. The vast majority of mediastinitis are the result of esophageal perforation or an infection of the mediastinum after a transsternal cardiac procedure. A small number of mediastinitis are the result of the spread of an infection arising from the oropharynx and this is the most severe. It has been reported that the overall mortality rate is 19–47%. This is especially true in DNM before 1990 when mortality was approximately 40–50% despite the use of antibiotics, aggressive surgical techniques, and intensive care monitoring [1–3]. In the presence of comorbid illness the mortality rate for patients presenting with established infections may be as high as 67%. Many authors have recommended early diagnosis, adequate antibiotics, and early effective surgical drainage. Brunelli and associates [4] recommended that aggressive surgical approaches had to be performed if the process had diffusely spread below the carina. Since 1990, there has been a decrease in the mortality rate to 15.4%, largely because of the more aggressive approach taken to mediastinitis. In this report we have had a question about whether more aggressive surgical approaches are really necessary for effective drainage in mediastinitis. We think that surgical drainage using video-assisted thoracoscopic surgery (VATS) could be more feasible and effective as a surgical option for mediastinitis; here we report our data using VATS for the treatment of mediastinitis.
|
2. Material and methods
|
|---|
This study is a retrospective medical record review from January 2000 to June 2006 and 47 patients have undergone surgical drainage for mediastinitis at the Pusan National University Hospital (PNUH) in Busan, Republic of Korea. Of these 47 patients we included 17 who underwent pleuromediastinal drainage and debridement using VATS as part of treatment for mediastinitis. Twenty-one patients were excluded because the surgery was limited to cervical drainage alone because the extent of mediastinitis was limited to superior mediastinum and nine patients were excluded because they underwent pleuromediastinal drainage via thoracotomy due to several reasons (Table 1
).
Data obtained from the medical records included sex, age, origin of infection, surgical option, progress, and outcome including cause of death. There were eight men and nine women. The mean age was 52 years old (range, 20–72). All 17 patients had been operated on by 1 surgeon.
All patients were diagnosed by preoperative cervico-thoracic computed tomography (CT) scans (Figs. 1 and 2
) and a cervico-thoracic CT scan was repeated between the second to tenth days after operation according to the patient's clinical status. Empirical broad-spectrum antibiotics were initiated for all patients as soon as mediastinitis was suspected. According to the results of culture and sensitivity tests of debris or pus, we decided to change antibiotic regimens, or not.
View larger version (50K):
[in this window]
[in a new window]
|
Fig. 1. Preoperative computed tomography scan (case 1 in Table 3) showing abscess in posterior mediastinum and medial basal segment of right lower lobe (a), postoperative computed tomography scan of same patient showing disappearance of abscess (b).
|
|
View larger version (142K):
[in this window]
[in a new window]
|
Fig. 2. Preoperative computed tomography scan (case 10 in Table 3) showing multiple abscesses in cervical spaces (a) and anterosuperior and posterior mediastinum and pericardium (b), postoperative computed tomography scan of same patient showing shrinking of the cervical abscesses (c) and mediastinal abscesses (d).
|
|
There were some differences in the details of the surgical approach according to the origin of infection, but basically we did as described below. All operations were performed under general anesthesia. In the cases that required pleuromediastinal drainage using VATS as well as cervical drainage, we performed both procedures without positional change. Patients were positioned supine with their neck extended and tilted down toward the non-operating side at about a 30° angle. A collar incision was performed, and exploration of neck and superior mediastinum was performed and continued through peritracheal, periesophageal, and perivascular spaces. Cervical or upper mediastinal abscess pockets were drained and necrotic materials were debrided through the cervical incision. And then, a 10 mm thoracoscope was inserted for exploration of pleural cavity. If there was pleural effusion or pus, it was evacuated and sent for bacteriological evaluation. Under direct vision of thoracoscope, the mediastinal pleura around all abscess pockets was opened and the pus drained through it. This was assisted by one or two other chest wall openings with trocar. In the cases of patients who underwent pleuromediastinal drainage using VATS alone they were positioned with full lateral decubitus. Other methods were same as the above mentioned. In cases of thoracic esophageal perforation, a primary repair using VATS was performed with interrupted suture. In cases of cervical esophageal perforation a direct repair through their cervical incision was performed. In all patients, chest tubes were inserted via the pleural space and placed in the mediastinum through the opening of mediastinal pleura for drainage of mediastinal pus. The patients were transferred to the intensive care unit.
|
3. Results
|
|---|
The origins of infection included esophageal perforation in nine patients and oropharyngeal abscess in eight patients (Table 2
). The summary of 17 patients is presented in Table 3
. Among them, 2 patients required conversion to thoracotomy during operation (case 5, case 9), so excluding the 2 patients, we calculated next 5 data for 15 patients. The mean duration from onset of symptoms to initial operation was 12.4 ± 13.1 days (range, 0–43). The mean duration from the initial operation to discharge was 43.6 ± 24.4 days (range, 8–80), and chest tubes were continued during 23.0 ± 18.0 days. Antibiotics duration was 38.6 ± 22.2 days, and the count of WBC was normalized on average 10.7 ± 10.6 day after the initial operation. Upon postoperative CT scan residual infection was found in one patient (case 16), so this patient underwent right thoracotomy (exploration, debridement and drainage) at day 2 after the initial operation. Two patients had postoperative pneumonia and recovered. In cases of esophageal perforation, two patients underwent a primary repair using VATS, three patients underwent repair through their cervical incision, and three could not be repaired due to diffuse necrosis of esophageal muscle layer. In one case we could not find the site of perforation because the hole was too small. There were no leakage on the repaired site in the cases (case 1, case 3, case 4, case 5, and case 7), and spontaneous healing of perforating site happened to cases that could not be repaired (case 2 and case 6).
There were three postoperative deaths; infection at case 8 and case 15 could not be controlled and they died due to septic shock on seventh and tenth postoperative day. ARDS with sepsis happened to case 16. ARDS could not be controlled regardless of ventilator care; she died due to ARDS on the 53rd day after the initial operation.
|
4. Discussion
|
|---|
Mediastinitis may begin primarily from structures in the mediastinum, or it may be the result of an infection extending downward from the oropharynx. Our report consisted of mediastinitis due to esophageal perforation and DNM (Table 2).
In the patients with mediastinitis that began primarily from structures in the mediastinum, more than 90% of cases are caused by esophageal perforation. Common causes of esophageal perforation include medical instrumentation, foreign-body ingestion, and trauma. Esophageal perforation is a serious condition with a high mortality rate. Burnett and colleagues [5] recently analyzed the treatment and outcome of a group of patients with acute posterior mediastinitis due to esophageal perforation and found a mortality rate of 38%. Good outcome depended on the time elapsed between rupture and diagnosis, the size of the rupture, and the underlying health of the patient. In particular early surgical intervention was a predictor of good outcome. Thorough mediastinal debridement and wide mediastinal drainage is important in improving survival of patients with life-threatening mediastinitis due to esophageal perforation. Although several surgical options are available, in the most cases primary repair is recommended regardless of the amount of time that has elapsed since the occurrence of the esophageal perforation [6–9]. We agree with the aforementioned policy. In our data of esophageal perforation, we tried to perform a primary repair and particularly using VATS in the two cases of thoracic esophageal perforation. Nguyen and co-workers [10] have summarized the fundamental goals in thoracoscopic approach to esophageal perforation as follows: (a) identification of the esophageal perforation, (b) debridement of necrotic debris, (c) control of the leak (primary closure or T-tube placement), and (d) wide drainage of mediastinum.
There was conversion to thoracotomy in two cases of mediastinitis due to esophageal perforation. The reasons for conversion were uncontrolled bleeding from azygos vein which was so weak due to serious inflammation in one case and poor visualization of mediastinal anatomy through the thoracoscope because of diffuse inflammation secondary to phlegmonous esophagitis. As described above, if situations such as severe perivascular inflammation, predicted bleeding or distorted structure are expected before operation, thoracotomy might be chosen instead of VATS as pleuromediastinal drainage.
In our cases of mediastinitis due to esophageal perforation, a 70-year-old female (case 8) died postoperative day 7. The etiology was an esophageal ulcer. Preoperatively she was already septic, leukopenic and had bilateral pneumonia and diabetes mellitus as a comorbidity and postoperatively acute renal failure developed. The remaining eight patients had uneventful postoperative courses.
In DNM the general incidence is low, but the mortality is so high. Oropharyngeal infection spreads to mediastinum through the loose anatomical structure of the pretracheal, perivascular, and retropharyngeal spaces [11]. Assisted by gravity and the negative pressure in mediastinum and pleural cavities during inspiration, pus in the cervical space readily descends into the mediastinum. In general, the management of DNM included early diagnosis, adequate antibiotics, and effective surgical drainage of pus and debridement of infected tissue. In order to determine on surgical option, Endo and colleagues [12] classified DNM into three types (I, IIA, or IIB) according to the extent of infection and suggested different surgical option based on this classification. Endo and colleagues [12] said also that cases of type I DNM might not always require aggressive mediastinal drainage. Diffuse DNM-type IIB demanded complete mediastinal drainage with debridement via thoracotomy, and subxiphoidal mediastinal drainage without sternotomy might provide adequate drainage in type IIA DNM. Many authors have reported the advantage of drainage and debridement through more aggressive open surgical approaches in the management of patients with DNM [13–16]. However, these aggressive approaches may worsen the condition of patients with sepsis and also increase the length of hospital stay, increase morbidity, and mortality. Particularly, because median sternotomy [15] or clamshell incisions [16] may cause subsequent osteomyelitis and dehiscence of sternum, these procedures seem detrimental in patients with DNM.
In our study, all cases belonged to type IIB (classified by Endo and colleagues [12]) (Table 3), and underwent cervical drainage with pleuromediastinal drainage using VATS. Postoperative complications occurred in two patients including nosocomial pneumonia and aspirated pneumonia. These two patients recovered.
There were two cases of postoperative death. These patients suffered uncontrolled sepsis both of which had methicillin resistant Staphylococcus aureus (MRSA) in clinical specimens (blood or pus). One of these patients underwent serial surgical drainage via thoracotomy for residual abscess pockets and ongoing sepsis, but sepsis could not be controlled, and she expired on the 53rd day after initial operation due to acute respiratory distress syndrome (ARDS). In this case one could speculate that this serial operation was related to trial of VATS approach as the initial operation instead of thoracotomy approach. We did not compare this directly but we think that this mortality is related to not effectiveness of pleuromediastinal drainage but severity of infection. If necessary, serial operations have been recommended in patients with thoracotomy approach, as well [17].
In summary, both mediastinitis that arises primarily from structures in the mediastinum and DNM remain a life-threatening infection and necessitates radical cervicotomy followed by pleuromediastinal drainage. In agreement with Roberts et al. [18] we support that pleuromediastinal drainage using VATS with or without cervical drainage can be a feasible and effective surgical option, although it has not been directly compared with a more aggressive open surgical approaches for patients with mediastinitis. A minimally invasive approach using thoracoscope seems to be a viable option for the treatment of mediastinitis. Continued critical review of the accumulating experience in VATS techniques will redefine the surgical indications for VATS for the treatment of mediastinitis.
|
References
|
|---|
- Estrera AS, Landay MJ, Grisham JM, Sinn DP, Platt MR. Descending necrotizing mediastinitis. Surg Gynecol Obstet 1983;157(6):545-552.[Medline]
- Guardia SN, Cameron R, Phillips A. Fatal necrotizing mediastinitis secondary to acute suppurative parotitis. J Otolaryngol 1991 Feb;20(1):54-56.[Medline]
- Levine TM, Wurster CF, Krepsi YP. Mediastinitis occurring as a complication of odontogenic infection. Laryngoscope 1986 Jul;96(7):747-750.[Medline]
- Brunelli A, Sabbatini A, Catalini G, Fianchini A. Descending necrotizing mediastinitis. Surgical drainage and tracheostomy. Arch Otolaryngol Head Neck Surg 1996 Dec;122(12):1326-1329.[Abstract/Free Full Text]
- Burnett CM, Rosemurgy AS, Pfeiffer EA. Life-threatening acute posterior mediastinitis due to esophageal perforation. Ann Thorac Surg 1990 Jun;49(6):979-983.[Abstract]
- Ohri SK, Liakakos TA, Pathi V, Townsend ER, Fountain SW. Primary repair of iatrogenic thoracic esophageal perforation and Boerhaave's syndrome. Ann Thorac Surg 1993 Mar;55(3):603-606.[Abstract]
- Whyte RI, Iannettoni, MD, Orringer MB. Intrathoracic esophageal perforation. The merit of primary repair. J Thorac Cardiovasc Surg 1995;109(1):140-144discussion 144–6.[Abstract/Free Full Text]
- Wright CD, Mathisen DJ, Wain JC, Moncure AC, Hilgenberg AD, Grillo HC. Reinforced primary repair of thoracic esophageal perforation. Ann Thorac Surg 1995;60(2):245-248discussion 248–9.[Abstract/Free Full Text]
- Port JL, Kent MS, Korst RJ, Bacchetta M, Altorki NK. Thoracic esophageal perforations: a decade of experience. Ann Thorac Surg 2003;75(4):1071-1074.[Abstract/Free Full Text]
- Nguyen NT, Follette DM, Roberts PF, Goodnight Jr. JE. Thoracoscopic management of postoperative esophageal leak. J Thorac Cardiovasc Surg 2001;121(2):391-392.[CrossRef][Medline]
- Kiernan PD, Hernandez A, Byrne WD, Bloom R, Dicicco B, Hetrick V, Graling P, Vaughan B. Descending cervical mediastinitis. Ann Thorac Surg 1999;68(4):1443.[Free Full Text]
- Endo S, Murayama F, Hasegawa T, Yamamoto S, Yamaguchi T, Sohara Y, Fuse K, Miyata M, Nishino H. Guideline of surgical management based on diffusion of descending necrotizing mediastinitis. Jpn J Thorac Cardiovasc Surg 1999;47:14-19.[Medline]
- Marty-Ané CH, Berthet JP, Alric P, Pegis JD, Rouvière P, Mary H. Management of descending necrotizing mediastinitis: an aggressive treatment for an aggressive disease. Ann Thorac Surg 1999;68(1):212-217.[Abstract/Free Full Text]
- Corsten MJ, Shamji FM, Odell PF, Frederico JA, Laframboise GG, Reid KR, Vallieres E, Matzinger F. Optimal treatment of descending necrotizing mediastinitis. Thorax 1997;52(8):702-708.[Abstract]
- Izumoto H, Komoda K, Okada O, Kamata J, Kawazoe K. Successful utilization of the median sternotomy approach in the management of descending necrotizing mediastinitis: report of a case. Surg Today 1996;26(4):286-288.[CrossRef][Medline]
- Ris HB, Banic A, Furrer M, Caversaccio M, Cerny A, Zbären P. Descending necrotizing mediastinitis: surgical treatment via clamshell approach. Ann Thorac Surg 1996;62(6):1650-1654.[Abstract/Free Full Text]
- Freeman RK, Vallieres E, Verrier ED, Karmy-Jones R, Wood DE. Descending necrotizing mediastinitis: an analysis of the effects of serial surgical debridement on patient mortality. J Thorac Cardiovasc Surg 2000 Feb;119(2):260-267.[Abstract/Free Full Text]
- Roberts JR, Smythe WR, Weber RW, Lanutti M, Rosengard BR, Kaiser LR. Thoracoscopic management of descending necrotizing mediastinitis. Chest 1997;112(3):850-854.[CrossRef][Medline]
Top
Abstract
1. Introduction
