| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, White 270, Massachusetts General Hospital, 55 Fruit
St., Boston, MA 02114.
2
Present address: Istituto di Radiologia,
Universitá degli Studi di Napoli
"Federico II," II Policlinico, Via Pansini 5, 80131, Napoli,
Italy.
3
Infectious Disease Division, Infection Control Unit, Jackson 504,
Massachusetts General Hospital, Boston, MA 02114.
Received November 8, 1999;
accepted after revision January 6, 1999.
Address correspondence to P. F. Hahn.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Charts of patients from whom vancomycin-resistant enterococci had been isolated during percutaneous drainage were reviewed to determine patient response to drainage, catheter management, and outcome of treatment.
RESULTS. Twenty-one patients underwent percutaneous drainage of 28 fluid collections from which vancomycin-resistant enterococci were isolated, including 16 intraabdominal abscesses, seven biliary or urinary obstructions, and five empyemas. The drainage of 27 (96%) of 28 collections were technically successful. In seven patients, drainage provided the first isolation of vancomycin-resistant enterococci from the patient. Five patients also had blood cultures with positive findings for vancomycin-resistant enterococci, and 14 collections were coinfected with other bacteria or with fungi. Twenty collections (71%) or obstructions were successfully treated with percutaneous drainage. Drainage was unsuccessful in treating eight collections in seven patients.
CONCLUSION. Despite high-level antibiotic resistance, fluid collections infected with vancomycin-resistant enterococci can be successfully drained percutaneously, resulting in a favorable likelihood of recovery for patients.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because medical therapeutics may have little benefit for such collections, percutaneous intervention can be the primary therapy for patients with vancomycin-resistant enterococci or other multidrug-resistant bacteria. We undertook this retrospective study to determine the efficacy of percutaneous drainage in patients with fluid collections infected with vancomycin-resistant enterococci.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Percutaneous Imaging-Guided Drainage
Since 1991, all percutaneous aspirations and drainages of fluid collections
(performed at our institution) have been prospectively recorded in a database
of interventional procedures maintained on a PC
[2]. Insertion of multiple
catheters into the same or nearby collections performed during the same
sitting are recorded as a single procedure. Transient catheter insertions,
such as those used for thoracentesis or paracentesis, are separately
distinguished and have not been included in our study.
Laboratory Isolation of Vancomycin-Resistant Enterococci
Body fluid or tissue samples submitted to the clinical microbiology
laboratory are routinely cultured on colistin—nalidixic acid sheep blood
agar plates to screen for gram-positive organisms. When growth developed on
any plate, an experienced technician examined for pattern of growth, Gram
stained the growth, and tested it for catalase reactivity. Gram-positive
catalase-negative organisms in pairs and chains with typical colony formation
are defined as enterococci on the basis of positive findings for hydrolysis of
pyrrolidonyl-2-naphthylamide.
Enterococci were classified as resistant or intermediate to vancomycin on the basis of disk-diffusion testing with a zone diameter of less than or equal to 14 mm or 15-16 mm, respectively, as recommended by the National Committee for Clinical Laboratory Standards [3]. Additional disk-diffusion testing was routinely performed for ampicillin, tetracycline, doxycycline, and chloramphenicol. Testing for highlevel resistance to gentamicin was performed on isolates from blood and cerebrospinal fluid and other isolates by request. As part of the hospital nosocomial surveillance program, enterococcal isolates resistant or intermediate to vancomycin were reported to the infection control unit and recorded in a database maintained on a PC.
Case Identification
Lists of hospital medical record numbers from patients with
vancomycin-resistant enterococci and from patients who had undergone
percutaneous catheter drainage, each number listed only once, were combined
electronically and reordered. Numbers appearing twice represented patients on
both lists. For these patients, the hospital information system was queried to
determine whether vancomycin-resistant enterococci were isolated from fluid
obtained during catheter drainage.
Chart Review
Written medical charts and electronic hospital records of all patients with
positive findings for vancomycin-resistant enterococci in percutaneous
drainages were reviewed. General information recorded for each patient
included demographic information, presenting illness, and outcome. Specific
information related to infection and its treatment with antibiotic therapy,
percutaneous drainage, or surgery was determined in each case. Signs of
response to drainage, including changes in pressor or ventilatory support,
fever, or leukocytosis, as assessed clinically in the medical chart, were also
recorded. Queries to admitting physicians provided follow-up information when
necessary.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Demographic Characterization of Patients
Fourteen males and seven females were studied (age range, 5-82 years;
average age, 57.7 years). Underlying diseases included neoplasm (n =
3), pancreatitis (n = 3), pneumonia (n = 3), renal failure
(n = 3), cardiovascular disease (n = 2), trauma (n
= 2), and one each acquired and congenital cirrhosis, lupus, Crohn's disease,
and AIDS. Eighteen patients had recently undergone surgery, either during the
present hospitalization or during a previous admission from which they were
convalescing. Four patients had been recipients of solid organ transplants.
Patients had suffered from their underlying disease for an average of 34
months (range, 0-228 days). All patients were inpatients at the time of
drainage, with mean hospitalization times before drainage of 37 days (range,
1-154 days). All patients had received parenteral antibiotic therapy during
their hospitalization before drainage. At the time of drainage for
vancomycin-resistant enterococci, 16 of 21 patients were receiving care in an
intensive care unit.
Fluid Collections with Positive Findings for Vancomycin-Resistant
Enterococci
Twenty-eight collections in 21 patients yielded fluid at the time of
drainage, from which vancomycin-resistant enterococci were cultured. These 28
collections were drained a total of 31 times, because three collections were
drained on two different days. The 21 patients underwent 13 additional
drainages, from which vancomycin-resistant enterococci were not isolated, the
findings were sterile (n = 6) or the yield was other organisms
(n = 7). These 13 additional drainages were temporally separate from
the drainage of vancomycin-resistant enterococci collections (2-4 days in two,
at least 1 week in the others) and have not been included in the analysis. Six
patients had more than one collection that revealed positive findings for
vancomycin-resistant enterococci. In seven patients, isolation of
vancomycin-resistant enterococci from a percutaneous drainage was the first
indication that the patient was infected or colonized with
vancomycin-resistant enterococci. Fluid collections that revealed positive
findings for vancomycin-resistant enterococci and were drained percutaneously
included 10 intraperitoneal abscesses, five retroperitoneal abscesses, one
liver abscess, and five empyemas. In the other seven collections, catheters
were inserted to relieve biliary (n = 2), gallbladder (n =
3), or renal (n = 2) obstruction.
Clinical Response to Drainage
Drainage of 14 (50%) of 28 collections produced clinical improvement. Signs
of improvement included diminished pressor or ventilator support (n =
1) and reduced leukocyte count or fever (n = 13). Eight drainages in
seven patients were considered failures of percutaneous intervention,
including one with a good initial response. Three collections in three
patients underwent surgical débridement. The
decision to operate was based on a lack of sustained clinical improvement
after percutaneous drainage. Surgery resulted in clinical improvement beyond
that effected by percutaneous drainage in two patients; however, one patient
experienced no further improvement.
Seven (33%) of 21 patients died during their hospitalization after drainage of a collection or system that revealed positive findings for vancomycin-resistant enterococci. Infection contributed to death in six of these patients. The remaining 14 patients were followed up and were still alive after 2 months (n = 3), 3-6 months (n = 3), or more than 6 months (n = 8). No complications occurred from any of the 31 percutaneous catheter insertions.
Twenty (71%) of 28 collections or obstructed systems that revealed positive findings for vancomycin-resistant enterococci were successfully treated with catheter drainage. The collection was eliminated or the obstructed system completely decompressed by the drainage in 16 of 28 patients. In four other patients, percutaneous drainage ultimately succeeded in satisfactory but subtotal resolution. The catheter was electively withdrawn after the adequate drainage of 15 collections or systems, including one perihepatic collection that resolved after redrainage of a recurrence. One patient was discharged with his catheter, which fell out 51 days after initial drainage (subtotal resolution) but did not need to be replaced during 12 months of follow-up. One patient underwent repeated biliary drainage after failure of a metallic stent. Excluding three catheters in two patients permanently placed for unrelenting biliary or renal obstruction and in one patient who died 17 days after cholecystostomy, the mean length of catheter drainage of successfully treated collections (16 collections) that revealed positive findings for vancomycin-resistant enterococci was 27 days (range, 3-74 days).
Treatment Failure
In eight (29%) of 28 collections in seven (33%) of 21 patients,
percutaneous drainage therapy was unsuccessful. In seven of these collections,
catheter placement was technically successful on the basis of imaging
performed at the time of the drainage. In one collection, a large hemorrhagic
empyema obscured landmarks, resulting in uncertainty about the adequacy of
catheter position. Five collections were acutely hemorrhagic; continuing
hemorrhage may have contributed to the failure of percutaneous drainage in
four of these collections. Two patients with peritonitis did not respond to
drainage of infected ascites. The remaining failure was before terminal
cholecystostomy that produced no detectable delay in the patient's demise.
Microbiologic Features
Cultures of fluid from 14 collections (50%) yielded organisms besides
vancomycin-resistant enterococci, including Staphylococcus organisms
(n = 3), gram-negative bacteria (n = 8), and fungi
(n = 5). On the basis of microbiologic results of drainage,
antibiotics were changed in five patients. Fourteen patients were treated with
an antibiotic regimen specifically directed at vancomycin-resistant
enterococci. Vancomycin-resistant enterococci were sensitive to
chloramphenicol in 25 drainage isolates, resistant in one, and untested in
two. Eleven patients were treated with chloramphenicol, with a mean length of
treatment of 27 days (range, 1-90 days). Three patients received doxycycline,
to which their vancomycin-resistant enterococci were sensitive. Three
patients, two of whom represented drainage failures, were treated with the
combination antibiotic quinupristin and dalfopristin (Synercid; Rhone-Poulenc
Rorer, Collegeville, PA) [4].
Five collections that revealed positive findings for vancomycin-resistant
enterococci were drained within 2 weeks of documented vancomycin-resistant
enterococci bacteremia. Three of these collections were successfully treated.
Microbiologic features related to outcome are displayed in
Table 1.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Serious infections with enterococci are difficult to treat with antibiotics and they usually require a combination regimen of ampicillin or vancomycin together with an aminoglycoside for bactericidal activity [6]. Antibiotic therapy of polymicrobial infections selects enterococci by eradicating other more susceptible organisms [7]. Enterococci with high-level resistance to gentamicin began to appear in the early 1980s [8]. Multidrug-resistant enterococci have become an increasing problem since the emergence of vancomycin resistance, first reported in 1988 [9, 10]. Risk factors include prolonged hospitalization, treatment with broad-spectrum antibiotics, and chronic antibiotic prophylaxis [11, 12]. Vancomycin-resistant enterococci now represent the third most common nosocomial cause of bacteremia in the intensive care unit population [11]. In some institutions, more than one quarter of bloodstream enterococcal isolates are resistant to vancomycin [13]. Unlike patients infected with antibiotic-sensitive enterococci, patients with vancomycin-resistant enterococci usually acquire the organism through nosocomial transmission [14]. Consequently, patients recognized as colonized or infected with vancomycin-resistant enterococci should be treated in isolation, with strict adherence to contact precautions.
In 1997, Landman and Quale [15] reviewed the many antibiotic strategies reported in attempts to treat vancomycin-resistant enterococci. No antibiotic regimen can be considered reliable or universally recommended. Although many vancomycin-resistant enterococci reveal in vitro sensitivity to chloramphenicol, its clinical effectiveness has been limited [16]. In September 1999, a combination of the two semisynthetic streptogramin antibiotics, quinupristin and dalfopristin, was approved by the United States Food and Drug Administration for use against some vancomycin-resistant enterococci. Synercid has been reported to be 63% effective against vancomycin-resistant Enterococcus faecium bacteremia [4] and to improve in-hospital survival [17]. This drug was used to treat three patients in our series while the drug was still under investigation.
Lai [6] examined the treatment and outcome of 28 patients with vancomycin-resistant enterococci infections. In the absence of effective antibiotic options, the most important therapies involved the removal of infected foreign bodies and drainage of infected collections. Curiously, all the patients who required drainage procedures underwent surgical drainage. This approach is echoed in a study by Landman and Quale [15], who recommend surgical drainage of deep collections infected with multidrug-resistant enterococci.
Although several studies have detailed the success of imaging-guided percutaneous drainage in large series of patients, no previous report has presented the results or efficacy of percutaneous drainage in cases of infection complicated by advanced antibiotic resistance [18, 19]. We treated 28 fluid collections that revealed positive findings for vancomycin-resistant enterococci during the course of 2311 percutaneous imaging-guided catheter drainages over a 4-year period. The 21 patients who underwent drainage of fluid infected with vancomycin-resistant enterococci were typical of other series of patients infected with vancomycin-resistant enterococci in terms of the need for intensive care, continuous hospitalization, and severity of disease. Although we were asked to treat a specific abnormal fluid collection or obstructed biliary or urinary system, these abnormalities were only part of our patients' poor medical condition.
Overall results of percutaneous drainage are reported as 80-85% successful. In our very specialized series, 50% of patients responded to a single percutaneous drainage, and 77% of patients were ultimately treated with percutaneous drainage of a fluid collection or obstructed biliary or renal system. When only abscesses and empyemas are considered, percutaneous catheter drainage was successful in treating 13 (62%) of 21 collections. This somewhat lower success rate can probably be explained by the generally poor overall physical condition of this patient group, rather than by the antibiotic resistance of the infecting organism per se. Moreover, from a diagnostic standpoint, drainage provided the first vancomycin-resistant enterococci isolation from seven patients and proved bacterial or fungal coinfection in 14 collections. Identification of vancomycin-resistant enterococci may effect clinical treatment of individual patients but is equally important for preventing nosocomial spread.
In conclusion, imaging-guided percutaneous catheter drainage can be an effective treatment of collections infected with highly antibiotic-resistant organisms. The benefits of drainage are conferred despite the advanced illness prevalent in our population of patients.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. C. Varghese, P. F. Hahn, M. G. Harisinghani, S. M. Hayat, D. A. Gervais, D. C. Hooper, and P. R. Mueller Fungus-infected Fluid Collections in Thorax or Abdomen: Effectiveness of Percutaneous Catheter Drainage Radiology, August 1, 2005; 236(2): 730 - 738. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. F. Hahn, D. A. Gervais, M. J. O'Neill, and P. R. Mueller Nonvascular Interventional Procedures: Analysis of a 10-year Database Containing More than 21,000 Cases Radiology, September 1, 2001; 220(3): 730 - 736. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
