[PDF] Direct E-Test (AB Biodisk) of Respiratory Samples Improves

View - Download Direct E-Test (AB Biodisk) of Respiratory Samples Improves

Previous PDF

Next PDF

382•CID 2007:44 (1 February)•Bouza et al.

MAJOR ARTICLE

Direct E-Test (AB Biodisk) of Respiratory Samples

Improves Antimicrobial Use in Ventilator-Associated

Pneumonia

Emilio Bouza,

1

Marı´a V. Torres,1

Celina Radice,

1

Emilia Cercenado,

1

Roberto de Diego,

2

Carlos Sa´nchez-Carrillo,

1 and Patricia Mun˜oz 1

Departments of

1

Clinical Microbiology and Infectious Diseases and

2 Anesthesia, Hospital General Universitario Gregorio Maran˜o´n, Universidad

Complutense, Madrid, Spain

(See the editorial commentary by Kollef on pages 388-90)Background.Ventilator-associatedpneumoniaisthemostfrequentlyobservednosocomialinfectioninintensive

care units, and it is associated with high morbidity and mortality. Early microbiological diagnosis and the initial

administration of appropriate antimicrobial therapy are associated with decreased mortalityandpotentiallyreduced

costs. Our study evaluates the clinical and financial impact of performing rapid antimicrobial susceptibility tests

directly on samples obtained from the lower respiratory tract. Methods.A prospective, randomized study was performed over a 2-year period. Patients who had a lower

respiratory tract infection that was acquired during mechanical ventilation and for whom samples obtained from

the respiratory tract were sent for culture were randomized to 1 of 2 groups. Samples were cultured for the control

group, and results were reported using standard procedures. Samples were also cultured for the test subject group

using standard procedures, but in addition, a rapid antibiogram was immediately performed by placing E-test

antibiotic strips(ABBiodisk)directlyonrespiratorytractsamples.PatientsintheE-testgroupreceivedapreliminary

laboratory report when it became available. The 2 patient groups were compared according to the following

variables: type and severity of underlying conditions, total days of antimicrobial use, number of defined daily

doses, cost of acquisition of the antimicrobial agent per episode, days of fever, days receivingmechanicalventilation,

days in the intensive care unit, incidence ofClostridium difficile-associated diarrhea, and mortality.

Results.Reporting a rapid E-test was associated with fewer days of fever, fewer days of antibioticadministration

until resolution of the episode of ventilator-associated pneumonia, decreased antibiotic consumption, lessC.

difficile-associated diarrhea, lower costs of antimicrobial agents, and fewer days receiving mechanical ventilation.

Conclusions.A rapid E-test of respiratory tract samples improves antimicrobial use in cases of ventilator-

associated pneumonia.Ventilator-associated pneumonia (VAP) is the most common intensive care unit (ICU)-acquired infection, and it has a mortality rate that ranges from 20% to

50% [1-3]. Early microbiological diagnosis and the ad-

ministration of appropriate initial antibiotic therapy have proven to be associated with decreased rates of mortality [4-10]. Received 28 June 2006; accepted 19 October 2006; electronically published 3

January 2007.

Presented in part: 45th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, D.C., December 2005 (abstract D-51). Reprints or correspondence: Dr. Emilio Bouza, Servicio de Microbiologı´a Clı´nica y Enfermedades Infecciosas, Hospital General Universitario Gregorio Maran˜o´n,Dr. Esquerdo 46, 28007, Madrid, Spain (ebouza@microb.net).

Clinical Infectious Diseases 2007;44:382-7

?2007 by the Infectious Diseases Society of America. All rights reserved.

1058-4838/2007/4403-0012$15.00Results of conventional microbiological reports

based on full bacterial identification and standard an- timicrobial susceptibility tests rarely reach physicians who are attending to patients with VAP within 48-72 h of lower respiratory tract (LRT) sampling. The E-test (AB Biodisk) is a well-known antimicrobial suscepti- bility test that uses antimicr obial agent-inoculated strips and is inoculum size-independent. The teststrips can be used on isolated bacteria or applied directly to clinical samples [11, 12].

Our study evaluatestheclinicalandeconomicimpact

of performing a rapid E-test with 6 key antimicrobial agents by applying the test strips directly to LRT sam- ples and reporting the results. We also examine anti- microbial use and the outcome of patients with VAP

in a large teaching institution.Downloaded from https://academic.oup.com/cid/article/44/3/382/312779 by guest on 25 July 2023

E-Test for Ventilator-Associated Pneumonia•CID 2007:44 (1 February)•383 Figure 1.Schematic representation of our study. The first sample of the episode is a lower respiratory tract (LRT) secretion sample that was collected when ventilator-associated pneumonia (VAP) was first sus- pected. Follow-up samples are the consecutive samples collected after the LRT secretion sample was collected when VAP was first suspected.

MATERIALS AND METHODS

The ethics committee of Hospital General Universitario Gre- gorio Maran˜o´n (Madrid, Spain) approved the study protocol. Our institution is a general reference hospital with a catchment population of 714,000 inhabitants and 1750 beds. It has 3 dif- ferent ICUs - a medical ICU, a resuscitation service, and a car- diac surgery ICU - with a total of 42 beds. A prospective, ran- domized study was performed among adults who were admitted to 1 of these ICUs during the period December2003-

December 2005.

Study population.Patients who had a suspected LRT in- fection that was acquired during mechanical ventilation were preselected for the study if?1 microorganism was observed by a Gram stain of a sample of the first tracheal aspirate that was obtained (i.e., the LRT sample collected when VAP was first suspected). The result of the LRT sample stain was im- mediately reported by telephone to the attending physicians. Patients were randomized at a ratio of 1:2 using the computer program C4 Study Design Pack: Group of Programs for Ex- periments Development, version 1.1 (GlaxoSmithKline) to re- ceive either a conventional microbiological examination with reporting (control group), or to receive an additional rapid antimicrobial susceptibility E-test (E-test group). All samples obtained from patients with suspected LRT infections (control group and E-test group) were also cultured, and results were reported according to standard procedures. Figure 1 is a sche- matic representation of our study. Samples for microbiological diagnosis were obtained by en- dotracheal aspiration using a 14F sterile probe to a depth of 2 cm from the distal end of the endotracheal tube. The secretions that were obtained were collected in a sterile container (Argyle Lukens Specimen Container; Sherwood Medical) and were transported in sterile packaging to the microbiologylaboratory. Standard microbiological procedures included quantitativecul- turing of a calibrated loop of 0.0025 mL of aspirate in blood agar, chocolate agar, and McConkey agar, and, when required, Legionellaagar (blood charcoal yeast extract) [13, 14]. Iden- tification of the microorganisms and antimicrobial suscepti- bility testing were performed using an automatic system (MicroScan; Dade Behring). Breakpoints were determined in accordance with theClinicalandLaboratoryStandardsInstitute guidelines [15]. Definitive results were reported to the ICU when bacterial identification and antimicrobial susceptibility test results became available. E-test.The rapid test consisted of an antibiogram that was performed by directly applying E-test strips to 150-mm Mueller-Hinton agar plates that wereseededwithLRTsecretion samples. The antibiotic strips tested contained oxacillin, pi- peracillin-tazobactam, cefepime, imipenem, ciprofloxacin, and amikacin. Plates were incubated at 35?-37?C for analysis the

following day with transmitted light. The results wereevaluatedas drug-resistant or drug-susceptible in accordance with the

Clinical and Laboratory Standards Institute guidelines [15]. This information was then immediately provided as a prelim- inary report for the patient's chart, usually within 24 h,without further intervention. VAP definition.VAP was defined as having a Centers for Disease Control and Prevention [16] and/or a Clinical Pul- monary Infection Score?6 [17] when?1 significant micro- organism was isolated by quantitative culturing (

110,000 col-

ony-forming units (cfus) per mL). Variables recorded included.Age, sex, unit of admission, classification of the severity of the underlying condition ac- cording to McCabe/Jackson criteria[18],Charlsoncomorbidity score [19], prior surgery before the episode of VAP during the present admission, number of antibiotics administered to pa- tients before VAP, previous VAP during the present admission, APACHE II score on admission to the ICU [20], bone score for severity of sepsis [21], microbiological isolates, existence of a concomitant infection,Clostridium difficile-associated diar- rhea, mortality of the episode, mortality in the ICU and mor- tality on discharge, days of antibiotherapy per episode, days of adequate therapy (i.e., duration that the antimicrobial covered microorganism sensitivities), defined daily doses (DDDs) re- ceived during the episode, DDDs of adequate therapy (DDDs of an antimicrobial that covered microorganism sensitivities), cost (ini) of the acquisition of antibiotics administeredduring

the episode, days of fever (duration of time that underarmDownloaded from https://academic.oup.com/cid/article/44/3/382/312779 by guest on 25 July 2023

384•CID 2007:44 (1 February)•Bouza et al.

Table 1. Clinical and demographic characteristics of patients with ventilator-associated pneumonia (VAP).

CharacteristicE-test group

(np167)Control group (np83)Total (np250)P

Age, mean years?SD 60?16.9 56.9?17.4 58.9?17 .2

Mean Charlson comorbidity score?SD 3.3?2.7 3.6?2.2 3.43?2.5 .4 Mean no. of antibiotics before VAP?SD 1.83?1.9 1.7?2.1 1.8?2.7 Mean APACHE II score on admission to the ICU?SD 11.9?4.6 11?3.6 11.6?4.3 .1 Sex

Male123 (73.7) 59 (71.1) 182 (72.8)

Female44 (26.3) 24 (28.9) 68 (27.2) .7

ICU of admission

Medical ICU61 (36.5) 43 (51.8) 104 (41.6)

Resuscitation service76 (45.5) 27 (32.5) 103 (41.2) .1

Cardiac surgery ICU30 (18) 13 (15.7) 43 (17.2)

McCabe/Jackson classification

Rapidly fatal0 (0) 0 (0)0 (0)

Moderately fatal63 (37.7) 31 (37.3) 94 (37.6) 1.0

Nonfatal104 (62.3) 52 (62.7) 156 (62.4)

Surgery before VAP106 (63.5) 42 (50.6) 148 (59.2) .1 Previous episodes of VAP29 (17.4) 16 (19.3) 45 (18) .6

Bone score for severity of sepsis

Systemic inflammatory response syndrome3 (1.8) 2 (2.4) 5 (2)

Sepsis63 (37.7) 33 (39.8) 96 (38.4) .9

Severe sepsis61 (36.5) 26 (31.3) 87 (34.8)

Septic shock33 (19.8) 18 (21.7) 51 (20.4)

Multiple organ failure7 (4.2) 4 (4.8) 11 (4.4)

Concomitant infection88 (52.7) 50 (60.2) 138 (55.2) .3

Polymicrobial56 (33.5) 20 (24.1) 76 (30.4) .2

NOTE.Data are no. (%) of patients, unless otherwise indicated. ICU, intensive care unit; VAP, ventilator-associatedpneumonia.

temperature was138.5?C), days receiving mechanical ventila- tion, and duration of stay in the ICU. Statistical analysis.Relationships between baseline vari- ables were evaluated for the randomized groups. Qualitative variables appear with their frequency distribution.Quantitative variables are summarized as mean?SD or as median (inter- quartile range). Relationships between variables were evaluated using thex 2 test for categorical variables, and the Student'st test for normally distributed continuous variables. The level of significance was set at for all tests. The statisticalanalysisP !.05 was performed using SPSS software, version 12.0 (SPSS).

RESULTS

During the study period (December 2003-December 2005),

4829 patients were admitted to the adult ICUs at our institu-

tion. We collected a total of 2779 LRT samples from 1220 patients. Microorganisms were observed by Gram staining in

387 samples sent to the laboratory that were obtained from

patients who had suspected LRT infection. The resulting data from gram staining were transmitted to ICU staff via telephone within a median of 111 min?73 min (range, 8 min-6 h).

Of the 387 episodes of suspected LRT infection, 250 patientsfulfilled the diagnostic criteria of VAP. We excluded the follow-

ing from the study: 55 episodes of ventilator-associated trach- eobronchitis, 39 episodes of severe pneumonia not caused by mechanical ventilation, and 43 cases of mere colonization. Of the 250 patients with bacteriologically confirmed VAP,167were enrolled in the E-test group and 83 were enrolled in the control group. These cases constituted the basis of our report. Table 1 summarizes the characteristics of the patients with VAP in the E-test group and in the control group. Univariate analysis did not reveal any differences in the underlying char- acteristics of the patients in either group. The population in- cluded 182 men and 68 women with a mean age of 58.9?

17 years. Underlying disease was moderately fatal or rapidly

fatal in 37.6% of the patients, and the mean Charlson com- orbidity score was 3.43?2.5. Previous surgery was the reason for admission to the ICU for 148 patients (59.2%). The mean APACHE II score on admission to the ICU was 11.6?4.3, and 138 (55.2%) of the cases of VAP were accompanied by a concomitant infection. Regarding the severity of sepsis in pa- tients with VAP, 62 (24.8%) experienced septic shock or mul- tiple organ failure. Previous episodes of VAP had occurred in

18% of the patients, and the mean consumption of antibioticsDownloaded from https://academic.oup.com/cid/article/44/3/382/312779 by guest on 25 July 2023

E-Test for Ventilator-Associated Pneumonia•CID 2007:44 (1 February)•385 Table 2. Microorganisms isolated from 250 patients suspected of having venti- lator-associated pneumonia (VAP).quotesdbs_dbs22.pdfusesText_28

[PDF] Brevet des Collèges DNB 2015 Pondichéry - Collège Jacques

[PDF] Série PRO - ABAC

[PDF] Série SPINN - ABAC

[PDF] Calendrier universitaire 2017-2018 (sauf MBA)-3

[PDF] Abandon d inscription - ULB

[PDF] Abaque ab7pdf

[PDF] Abaque ab7pdf

[PDF] Annexe 1 Calcul de la rétention des EP

[PDF] Mesure de l 'humidité dans les installations - CS Instruments

[PDF] ABATS lapin aBaTS VOlaillE - Saint Aubin d 'Aubigné

[PDF] Vous êtesEmployeur

[PDF] Deutschkurs ABC - deutschfit

[PDF] correction Devoir libre 15 3èmes - Blogs de l 'académie d 'Amiens

[PDF] Contrôle n #730 4

[PDF] ABC est un triangle rectangle en A tel que : AB = 12 cm AC = 16 cm