| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Department of Traumatology, Osaka University Medical School, 2-15 Yamadaoka, Suita-shi, Osaka 565-0871, Japan.
Received April 23, 2003;
accepted after revision December 10, 2003.
Address correspondence to Y. Nakamori
(nakamori{at}hp-emerg.med.osaka-u.ac.jp).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Thirty-one patients with cervical necrotizing fasciitis and descending necrotizing mediastinitis were included. Twenty consecutive patients were treated by percutaneous catheter drainage. Catheters were introduced into the infected space from the neck, under the guidance of sonography and X-ray fluoroscopy. The results of the treatment were compared with those of 11 patients treated previously by surgical drainage.
RESULTS. In the catheter group, no patient required supplementary surgical drainage. Mortality was 0% in both groups. Comparison of length of stay in the ICU, serial changes in C-reactive protein levels, duration of antibiotic therapy, and duration of mechanical ventilation all showed no statistically significant difference between groups. Secondary infection of the wound and positive culture of antibiotic-resistant bacteria were observed less frequently in the catheter group than in the open surgical group. The total use of plasma infusion and analgesics was significantly less in the catheter group than in the surgery group. Oral feeding was started significantly earlier in the catheter group than in the surgery group.
CONCLUSION. Percutaneous catheter drainage for cervical necrotizing fasciitis and descending necrotizing mediastinitis was less invasive than conventional surgical drainage but produced a similar outcome.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In our ICU, patients with cervical necrotizing fasciitis with or without descending necrotizing mediastinitis were also treated by aggressive surgical débridement. However, in our surgical experience, we recognized that the total amount of necrotic tissue in patients with cervical necrotizing fasciitis and descending necrotizing mediastinitis is small and that the main role of surgical treatment is drainage of the necrotic fluid rather than débridement of the necrotic tissue. In 1998, we began performing percutaneous catheter drainage instead of conventional surgical drainage for the treatment of cervical necrotizing fasciitis and descending necrotizing mediastinitis. Sonographically guided catheter insertion is performed from the cervical space along the route of infection. We treated 20 consecutive patients who had such infections by percutaneous catheter drainage. The mortality rate was 0%. In this study, we determined the clinical utility of percutaneous catheter drainage compared with surgical débridement in the treatment of cervical necrotizing fasciitis and descending necrotizing mediastinitis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Open surgical drainage was performed in 11 consecutive patients who were admitted between January 1995 and September 1998 (group S). Percutaneous catheter drainage was performed in 20 consecutive patients who were admitted between October 1998 and April 2002 (group C). The following patient characteristics were compared for these two groups: sex, age, primary site of infection, initial microbiologic culture, predisposing medical condition, time from the appearance of symptoms to admission in our center, presence or absence of descending necrotizing mediastinitis, initial C-reactive protein level, total leukocyte count, plasma protein level, and blood hemoglobin concentration. The aspects of medical treatment compared between the two groups were total amount of transfusion and plasma infusion, daily infusion during the acute phase, duration of antibiotic therapy, duration of mechanical ventilation, and the use of analgesics. No significant difference appeared between the two groups in the type of patient care, such as inotropic support, infusion, transfusion, mechanical ventilation, or nutrition, except drainage method. Microbiologic culture of wounds, initiation of enteral nutrition and oral feeding, and serial changes in C-reactive protein and plasma protein levels were analyzed to compare the clinical course between the two groups. Clinical outcome in the two groups was also evaluated according to length of stay in the ICU, complications, and mortality rate.
Open Surgical Drainage (Group S)
After diagnosis, patients in this group underwent surgery performed by
trauma surgeons and oral surgeons. Necrotic tissue was subjected to radical
débridement, and all infectious spaces were drained and irrigated with
saline. All layers involved in the infection were desquamated. The
sternocleidomastoid muscle, carotid sheath, trachea, and esophagus were
aggressively separated with Penrose drains. Further drainage from the oral
cavity and tooth extraction were performed if necessary. The wound was left
open and packed with sterile wet gauze. Postoperatively, patients underwent
additional débridement, irrigation, and dressing change under general
anesthesia once or twice daily. After good granulation and absence of
bacterial growth in the wound were confirmed, the wound was closed gradually
with some subcutaneous drains remaining.
Catheter Drainage (Group C)
In all patients, enhanced CT was used to confirm the extent of infection
and to determine the space requiring drainage. After diagnosis, nasotracheal
intubation was performed, guided by bronchial fiberscope. Propofol and
ketamine or pentazocine were injected for general anesthesia. Percutaneous
drainage was performed in a fluoroscopically equipped room under sonographic
guidance (LOGIQ 500 PRO, General Electric Yokogawa Medical Systems). The 6-MHz
microconvex probe (C721, General Electric Yokogawa Medical Systems) was used
because it can detect a wide subcutaneous area clearly and because its size is
suitable for use in the narrow cervical area. Under real-time sonographic
guidance, a fine needle was inserted into the infected space, targeting
low-echoic lesions with fluid collection or hyperechoic foci showing scattered
gas (Fig. 1) after confirmation
by color flow mapping that the route of the needle was free of major vessels.
For catheter drainage, a 7-French catheter kit for percutaneous transhepatic
cholangio drainage (PTCD kit, Create Medic) was used.
|
After the inner stylet of the needle was withdrawn, contrast medium was injected via the needle. If contrast medium spread into the targeted space, an angiography guidewire was inserted from the needle into the infected space and guided to the appropriate position. Either a pigtail or a straight catheter was subsequently inserted over the wire. If the infection had spread into another anatomic space, another catheter was placed by the same procedure until the major part of the involved space was considered to be drained (Fig. 2).
|
Fluid extracted from each catheter was sent to the laboratory for bacterial culture. To examine the efficiency of catheter drainage, we injected contrast medium from each catheter, and CT scanning was performed (Fig. 3A, 3B, 3C, 3D). If the scan revealed any infected space without contrast medium, drainage was considered insufficient and another catheter was added. After sufficient drainage was confirmed, catheters were connected to a collection bag without suction or irrigation. To prevent obstruction of these catheters, we injected 5 mL of saline solution into the drainage route daily. If any residual infected space was identified on CT or sonogram during treatment, catheters were added (Fig. 4A, 4B, 4C, 4D, 4E). Percutaneous catheters were removed under conditions of radiographic improvement and decrease and sterility of the drainage fluid.
|
|
|
|
|
|
|
|
|
Antibiotic Therapy
Benzylpenicillin potassium (20 million U/day) and imipenem–cilastatin
(1 g/day) were initially administered to all patients in both groups.
Antibiotics were changed according to the bacteriologic and antibiotic
sensitivity test results. Termination of the antibiotic therapy was dependent
on improved CT findings and decreased levels of serum C-reactive protein.
Amphotericin B (1,200 mg/day) was also administered to each patient via
nasogastric tube for decontamination of the digestive tract.
Statistical Analysis
Data are presented as mean ± standard deviation. Unpaired Student's
t tests were used for comparisons between the two groups for age,
initial C-reactive protein level, total leukocyte count, serum total protein,
ICU stay, duration of intubation, start of enteral nutrition, start of oral
feeding, transfusion, plasma infusion, infusion during the acute phase,
duration of antibiotic therapy, and the amount of analgesics used. Sex, origin
of infection, and frequency of secondary infection were compared using
chi-square tests. The level for statistical significance was set at p
less than 0.05.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Primary Site of Infection
The initial source of cervical necrotizing fasciitis in group S was dental
infection (n = 6), pharyngitis (n = 2), and other sources
(n = 3). The initial sources of cervical necrotizing fasciitis in
group C were dental infection (n = 9), pharyngitis (n = 3),
tonsillitis (n = 3), and others (n = 5). No statistically
significant difference appeared between the groups in the primary site of
infection.
Microbiology of Primary Culture
In both groups, Streptococcus and Bacteroides or
Prevotella organisms were commonly isolated. Pus or necrotic tissue
for bacterial culture was obtained perioperatively in group S.
Streptococcus (n = 11), Bacteroides or
Prevotella (n = 3), Candida (n = 3), and
Fusobacterium (n = 2) organisms were isolated in group S.
Pus was obtained by puncture and aspiration from the infection site in group
C. Streptococcus (n = 17), Bacteroides or
Prevotella (n = 14), Propionibacterium (n
= 2), Staphylococcus (n = 1), and Fusobacterium
(n = 1) organisms were isolated in group C.
Treatment of Wound and Catheter Use
Initial surgical drainage was performed in group S within 131 ± 39
min of admission. Mean operating time was 164 ± 59 min (range,
85–255 min), and mean intraoperative blood loss was 364 ± 237 mL
(range, 80–830 mL). After the initial surgical débridement,
dressing change with débridement and irrigation under general
anesthesia were performed 18 ± 7 times. The wound was closed 25
± 34 days after the initial operation. The mean number of catheters
inserted in each group C patient was 3.3 ± 1.9. In five patients, chest
tubes were placed to drain pleural effusion. The procedure time to complete
percutaneous drainage was between 60 and 120 min. Catheters were removed
completely after 16 ± 10 days. No group C patient required
supplementary surgical drainage. However, in two patients, local skin
débridement was needed for partial skin necrosis.
Airway Management
With one exception in group C, all patients in both groups required airway
management. Nasotracheal intubation was attempted first, and tracheostomy was
avoided as much as possible because of the risk of extending the infection to
the tracheostomy site. The mean duration of intubation was 16.7 ± 21.6
days in group S and 9.4 ± 7.3 days in group C, with no statistically
significant difference between the groups
(Table 1).
|
Nutrition
Enteral nutrition was started as early as possible in both groups. In group
C, enteral nutrition was started in 2.6 ± 1.6 hospital days, and oral
feeding was started in 7.9 ± 4.1 hospital days. In group S, enteral
nutrition was started in 4.3 ± 2.9 hospital days, and oral feeding was
started in 22.7 ± 23.3 hospital days. Oral feeding was started
significantly earlier in group C than in group S
(Table 1).
Infusion and Transfusion
No significant intergroup difference appeared in the amount of infusion
during the acute phase (days 0–5). Nor did a difference appear in the
minimum plasma protein concentrations (group S: 4.9 ± 0.7 g/dL vs group
C: 5.1 ± 0.5 g/dL), but the difference in the total plasma infusion
(group S: 3,409 ± 3,984 mL vs group C: 382 ± 903 mL; p
= 0.003) was significant. No difference appeared in initial hemoglobin
concentration (group S: 12.7g/dL vs group C: 12.7g/dL) or amount of blood
transfusion (Table 1).
Antibiotics
No statistically significant difference appeared in the duration of
antibiotic therapy (group S: 16.0 ± 8.0 days vs group C: 11.8 ±
5.2 days; p = 0.08). Antibiotics were changed to other drugs in eight
group S patients and in seven group C patients. Specimens for microbiologic
testing were obtained from the open wound in group S and from the catheter in
group C. Secondary infection in the primary site was observed comparatively
frequently in group S (group S: 64% vs group C: 30%; p = 0.13). In
group S, antibiotic-resistant bacteria were statistically more frequently
cultured from the wound specimens (methicillin-resistant Staphylococcus
aureus, n = 2; methicillin-resistant Staphylococcus epidermidis,
n = 3). In group C, methicillin-resistant Staphylococcus
epidermidis was detected in only one patient (group S: 45% vs group C:
5%; p = 0.013).
Descending Necrotizing Mediastinitis
Descending necrotizing mediastinitis was diagnosed on cervicothoracic CT.
In group S, five patients presented with cervical necrotizing fasciitis
complicated by descending necrotizing mediastinitis (all were men, with a mean
age of 50.2 ± 20.0 years). In group C, six patients presented with
cervical necrotizing fasciitis complicated by descending necrotizing
mediastinitis (all were male, with a mean age of 59.0 ± 10.8 years). No
patient in either group acquired descending necrotizing mediastinitis during
the course of treatment at our center. No statistically significant difference
appeared in indexes of infection between these descending necrotizing
mediastinitis patients in the two groups (C-reactive protein: group S: 21.1
± 8.9 mg/dL vs group C: 35.8 ± 16.1 mg/dL; total leukocyte
count: group S: 13,300 ± 3,800/µL vs group C: 13,600 ±
5,900/µL). In general, mediastinal infection was drained transcervically in
group S. However, one patient required further transthoracic drainage of a
residual infection. In group C, mediastinal drainage was performed by
percutaneous catheters introduced from the neck or anterior chest wall. No
statistically significant difference appeared in the clinical outcome in these
patients. However, the hospital stay was shorter in group C than group S (23.3
± 13.7 days vs 54.6 ± 56.8 days; p = 0.22). The amounts
of transfusion and plasma infusion were also smaller in group C than in group
S (blood transfusion: 78 ± 121 mL vs 945 ± 1,324 mL, p
= 0.14; plasma infusion: 386 ± 309 mL vs 5,212 ± 5,316 mL,
p = 0.052).
Analgesic Drugs
IV pentazocine and buprenorphine hydrochloride were the primary analgesics
used in both groups. Analgesics were administered more frequently in group S
than in group C (Table 1).
Clinical Outcome
The mortality rate was 0% in both groups. Patients in group S were
discharged after the open wound was closed, and patients in group C were
discharged after the catheter was removed. No statistically significant
difference appeared in the length of ICU stay between the two groups
(Table 1). Serum levels of
C-reactive protein in hospital day 9 did not differ significantly between the
two groups (group S: 4.9 ± 4.3 mg/dL vs group C: 4.8 ± 3.8
mg/dL).
Two patients in group C had complications during the catheter insertion procedure. One group C patient had an esophageal puncture when the catheter was introduced from the neck into the posterior mediastinum. The puncture wound was treated conservatively without further complication. The other group C patient had a mediastinal hemorrhage that stopped spontaneously. One group S patient had mediastinal hemorrhage from the thoracic aorta on day 16 after surgery. The emergent aortogram showed that the hemorrhage occurred from the contact region of the descending aorta with mediastinal drain and that no infected aneurysm was present. The patient was hemodynamically unstable, and emergent placement of a covered stent-graft was successfully performed. His course after this episode was uneventful. No other severe complications were observed in either group.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patients who have severe cervical necrotizing fasciitis and descending necrotizing mediastinitis are transferred from a wide area of Japan to our critical care center. Therefore, we have abundant opportunities to diagnose and treat these infections. Our experiences with cervical necrotizing fasciitis and descending necrotizing mediastinitis suggest the possibility that less invasive catheter drainage would be sufficient to control the infection in patients. Because our experience with surgical drainage for these infections confirm that only a small amount of necrotic tissue is removed, we recognize that the primary purpose of the surgery is opening the infected area and draining pus. In the early stage of cervical necrotizing fasciitis, infection spares underlying muscle and myonecrosis rarely occurs. Even if the infection progresses to the muscle, only a small amount of muscle exists in the cervical area [2, 11]. Also, these infections spread from their primary site to the sublingual, retropharyngeal, and parapharyngeal spaces. Infection can extend into several cervical spaces and sometimes reaches the mediastinum. Once the appropriate cervical space is punctured, a catheter can be introduced with either retrograde or antegrade insertion into every infected space.
We first used catheter drainage for cervical necrotizing fasciitis in 1998. When we had treated several patients successfully by percutaneous catheter drainage, the benefit of this less invasive strategy became clear. Prospective randomized clinical investigation of cervical necrotizing fasciitis and descending necrotizing mediastinitis for comparison of the two methods is difficult because the disorders are rare. Therefore, beginning in October 1998, all patients with such infections admitted to our center were treated with percutaneous catheter drainage, and the results were compared with historic controls treated by surgical drainage between 1995 and 1998.
We used CT to identify the origin and extent of the infection and to determine the point of percutaneous puncture and the layer in which the catheter could be guided to the area of severe infection and necrosis. Sonography of the neck in these patients revealed that fluid collection is sometimes detectable. However, in most cases, fluid collection cannot be detected clearly because of interference by the scattered gas artifact, and the scattered gas sonogram is the target for puncture. Adequacy of the puncture can be confirmed using X-ray fluorography to view the expansion of contrast medium injected through the catheter. If contrast medium spreads along the infection route predicted after initial CT, an angled guidewire can be inserted, even into the deep mediastinum, without technical difficulty.
All 20 patients with cervical necrotizing fasciitis, including the six patients with cervical necrotizing fasciitis complicated by descending necrotizing mediastinitis, were treated successfully by means of percutaneous catheter drainage. We found several advantages of percutaneous catheter drainage compared with open surgical drainage in this study. Antibiotic-resistant bacteria, including methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis, were detected less frequently when percutaneous catheter drainage rather than open surgical drainage was used, suggesting that percutaneous catheter drainage with closed-wound management can protect patients from secondary bacterial infection.
Although no significant difference appeared between the two groups in the minimum plasma protein concentrations, less plasma infusion was needed when the catheter was used than when surgery was performed. This finding suggests that the leakage of protein from the wound is significantly reduced when a catheter is used. Oral feeding could be started earlier in patients who underwent percutaneous catheter drainage, which shows the reduced cervical invasiveness of catheter drainage. The total use of analgesics was lower in patients treated with catheter drainage than in patients in whom surgery was performed. Large amounts of anesthetics were required for the daily débridement and irrigation of patients treated surgically, and it is clear that the patients treated with catheter drainage experienced significantly less pain and stress. In the length of hospital stay, duration of mechanical ventilation, total amount of transfusion, and duration of antibiotic therapy, no statistically significant differences appeared between surgical drainage and catheter drainage. With respect to the parameters compared, however, the results of catheter drainage were better than those of surgical drainage.
Extended airway management is needed in cases of cervical necrotizing fasciitis because of airway obstruction due to pharyngeal edema and mucous discharge in the oral cavity. Whitesides et al. [5] suggested that the airways of patients with cervical necrotizing fasciitis should be treated by tracheostomy performed during the initial surgery. We used nasotracheal intubation instead because the point of catheter drainage is to control the infection site by a closed drainage system. Tracheostomy introduces the risk of secondary infection at the tracheostomy site. The duration of intubation was only 9.4 ± 7.3 days in our catheter drainage series.
We chose to use benzylpenicillin potassium and imipenem–cilastatin for the initial antibiotic therapy before the pathogens were identified. Benzylpenicillin was used to target gram-positive cocci, including Streptococcus, which are most frequently isolated in these infections. Imipenem–cilastatin was used to target gram-negative bacteria, particularly anaerobes. Mohammedi et al. [3] reported that although Streptococcus bacteria were most commonly isolated, Prevotella bacteria were the important cervical necrotizing fasciitis pathogens. Therefore, they recommended amoxicillin and clavulanate potassium in combination with aminoglycoside for initial antimicrobial therapy. Streptococcus was found in 90% of our cases, and Bacteroides or Prevotella were found in 51% of our cases. Our data suggest that antibiotic therapy for cervical necrotizing fasciitis must target gram-positive and gram-negative bacteria simultaneously.
We experienced two complications of mediastinal bleeding and esophageal injury in group C. These complications occurred during insertion of the dilator and the catheter over the guidewire. Although both complications were resolved after removal of the dilator or the catheter, insertion of the dilator and the catheter to the mediastinum should be performed with caution under fluoroscopic guidance to avoid traumatic injuries.
We conclude that percutaneous catheter drainage strategy is less invasive than surgical drainage, and it controls the infection effectively. Moreover, percutaneous catheter drainage areas are less likely to become secondarily infected by antibiotic-resistant bacteria, and percutaneous catheter drainage seems superior to surgical drainage in pain control and in preventing protein leakage from the wound. Percutaneous catheter drainage may replace conventional surgical drainage as the treatment of choice for cervical necrotizing fasciitis and descending necrotizing mediastinitis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. Sumi, H. Ogura, Y. Nakamori, I. Ukai, O. Tasaki, Y. Kuwagata, T. Shimazu, H. Tanaka, and H. Sugimoto Nonoperative Catheter Management for Cervical Necrotizing Fasciitis With and Without Descending Necrotizing Mediastinitis Arch Otolaryngol Head Neck Surg, July 1, 2008; 134(7): 750 - 756. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
