DOI 10.1378/chest.102.2.525 1992;102;525-529Chest

 J Rello, V Ausina, J Castella, A Net and G Prats consciousness with implications for therapy.trauma patients. Influence of level of Nosocomial respiratory tract infections in multiple

  http://chestjournal.chestpubs.org/content/102/2/525

can be found online on the World Wide Web at: The online version of this article, along with updated information and services 

) ISSN:0012-3692http://chestjournal.chestpubs.org/site/misc/reprints.xhtml(without the prior written permission of the copyright holder.reserved. No part of this article or PDF may be reproduced or distributedChest Physicians, 3300 Dundee Road, Northbrook, IL 60062. All rights

ofbeen published monthly since 1935. Copyright1992by the American College is the official journal of the American College of Chest Physicians. It hasChest

 © 1992 American College of Chest Physicians by guest on July 10, 2011chestjournal.chestpubs.orgDownloaded from

[GCS] below 9 during a period greater than 24 h), whileaerobic Gram-negative bacilli were responsible for the

majority of cases in the remaining population studied. Theoverall mortality rate was 19.8 percent, but only five deathswere related to NP. We conclude that nosocomial respiratory tract infections are a frequent problem in multipletrauma patients, especially in those with GCS below 9,although this complication is associated with a relativelylow mortality. Among patients with GCS below 9, S aureus

was a frequent finding; consequently, antimicrobial therapyin this population needs to be different than that for theremaining multiple trauma patients with NP.

(Chest 1992@102:525-29)

IGCS=GlasgowcomascorePSBprotectedspecimenbrushi

A prospective study of 161 multiple trauma patients wascarried out to determine the incidence, the causative agents,and the outcome of nosocomial respiratory tract infectionsin this highly selected population. Thirty-eight (23.6 percent) patients developed a nosocomial pneumonia (NP). Inaddition, there were four superinfections in three patients,representing an incidence of 26 percent (42 of 161). Incidence of NP was significantly greater among comatosepatients (42.2 vs 13.3 percent, p<O.OS). Furthermore,purulent tracheobronchitis was diagnosed in six patients.The causative agent ofNP was identified in 36 (85.7 percent)episodes by means of fiberoptic bronchoscopies with protected specimen brush sampling. Staphylococcus aureus(55.8 percent) was the predominant pathogen isolated in

multiple trauma patients in coma (Glasgow coma score

T raumatic injuries are emerging as a major cause ofmorbidity and mortality in industrialized coun

I Although only a small fraction of injury cases,

the most severe, are admitted to ICUs, the numberhas increased over recent years.2 Alterations in thephysiologic and anatomic defenses from trauma disease predispose to development of pneumonia.3'4Furthermore , treatment with mechanical ventilationincreases a patient's risk of developing pneumonia.5Different prophylactic regimens (selective decontamination of the digestive tract, etc)@°in the preventionof nosocomial pneumonia in mechanically ventilatedpatients have recently been developed. However, theincidence of respiratory tract infection in traumapatients in the ICU, especially in the subgroup ofpatients requiring mechanical ventilation, is uncertain,and the etiology has not yet been clearly established.―Precise etiologic diagnosis requires results of laboratory tests, which may not be immediately available.Empiric therapy is usually required at the onset andshould be based on the broad spectrum of diagnosticpossibilities.

We conducted a prospective study to assess theepidemiology of nosocomial respiratory tract infec

tions in multiple trauma patients hospitalized in ourICU. Because most multiple trauma patients have

cranioencephalic trauma, we examined the influenceof the level of consciousness in the development ofthat complication. Our specific goal was to providecurrent information on the frequency of microbialetiologies so that management of nosocomial pneu

monia in this highly selected population could berationally formulated.

Population Study

PATIENTS AND METHODS

The study was conducted from January 1988 until July 1990, ina 16-bed medical-surgical ICU, at a 1,000-bed teaching hospitalwhich serves as both a referral center and a first-line hospital. Allmultiple trauma patients admitted to the ICU were included. Age,sex, date of admission, underlying diseases, and diagnosis onadmission were recorded. Prior antibiotic therapy; surgery; smoking,and alcohol habits, as well as the presence of chronic obstructivepulmonary disease (COPD), mechanical ventilation, or bronchoaspiration were also recorded. Chest x-ray films and WBC countswere performed dail)@ All patients had a nasogastric tube in place

if they were receiving mechanical ventilation. All patients receivedranitidine or antacids or both to obtain a gastric p11 4. No regimenfor nosocomial pneumonia prophylaxis or selective decontaminationof the digestive tract was used.

The patient's clinical progress was observed until ICU dischargeOI@death. Subsequent bouts of pneumonia after discharge from the

ICU were not studied. Development of septic shock was recorded.The mortality was categorized as directly related when an episodeof nosocomial pneumonia was considered to be the primary causeofdeath and any other potential cause could be identified while theinfection was still active.

Definitions

Nosocomial pneumonia was defined as any lower respiratory tract

*Fn)m the Intensive Care, Microbiology; and Respiratory Departments, Hospital de la Santa Creu i Sant Pau, Universitat Autônoma,Barcelona, Spain.

Manuscript received August 22; revision accepted November 22.

CHEST I 102 I 2 I AUGUST, 1992 525

NosocomialRespiratoryTractInfectionsinMultipleTraumaPatients*Influenceof Levelof ConsciousnesswithImplications for TherapyJordi Rello, M.D.; VicençAusina, M.D.; Joan Gastella, M.D.;Alvar Net, M. D.; and Guillem Prats, M.D.

 © 1992 American College of Chest Physicians by guest on July 10, 2011chestjournal.chestpubs.orgDownloaded from

VariableWholeGroup

(n161)GroupA

(n71)GroupB

(n90)Averageage,yr38.734.840.3Males119

(73.9)52 (73.2)67(74.4)Shock6

(3.7)1 (1.4)5(5.5)COPD5

(3.1)05(5.5)Bronchoaspiration5(3. 1)05(5.5)Corticosteroid

use13 (8)6 (8.4)7(7.7)Renal

failure5 (3.1)5(7)0Pnorsurgeryff35(21.7)5 (7)30(33.7)CET@86

(53.4)59 (83)27(30)Priorneurosurgery37 (22.9)22 (30.9)16(17.7)Mechanical

ventilation108 (67)59 (83.0)49(54.4)Episodesof NPt42 (26)30 (42.2)12(13.3)Averagetimeofonset99.18.8of

NP, days

OrganismEpisodesATBComaResolvedStaphylococcusaureus1661516Pseudsnnonas

aeruginosa3322Serratiamarcescens33. ..2S

aureus,Hemophilus3133influenzaeHinfluenzae2.

..12Streptococcus

pncunwniae2. ..22Hinfluenzae,2.

..22S

pneumoniaeSaureus@H influenzae,2. ..12anaerobic

floraAlcaligenesfaecalis11.

..1Escherichia

coli1. ..11Paeruginosa,11.

..1S

marcescensUnknown6534

Table 1—Characteristicsofl6l MultipleTraumaPatientsStudied*

Multiple trauma was defined as presence of injury to more thanone body area or system.' According to the level of consciousness,patients were classified in two groups. A Glasgow coma score (GCS)below 9 was defined as co'6 Patients who were in coma during aperiod greater than 24 h were classified as group A; patients notmeeting this criterion were classified as group B.

Strategyfor Etiologic Diagnosis

Fiberoptic bronchoscopy examination was performed in cases ofsuspected pneumonia, except in four episodes. To obtain uncontammated lower airway secretions from roentgenographically abnormallobes for bacterial cultures, a protected specimen brush (PSB)samplingwas obtained.'@ Blood, urine, and samples from other siteswere collected for bacterial culture in all patients. Patients receivedantibiotics before cultures were obta,ned in 20 episodes. Afterculture results were known, the antimicrobial regimen was adjustedon the basis of sensitivity testing of the identified organisms.

Bacteriologic Processing

Specimens were sent to the laboratory immediately after collection. A Gram stain of the respiratory specimen was performed inall cases. Two serial 100-folddilutions were made, and 0.1-mlaliquots ofthe original suspension and each dilution were inoculatedon agar plates for aerobic and anaerobic culture; one agar plate withselective buffered charcoal-yeast extract medium for isolation ofLegionella species was@ According to the standardsadopted in previous studies, bacterial counts equal or above 1ff' clii!ml established the threshold to diagnose pulmonary infections.'@Bacterial identification and susceptibility testing were done bystandard methods.@

Statistical Methods

Statistical comparisons were made using the x2 test with Yates'correction or the Fisher's exact test when necessary, with 95 percentconfidence limits.

RESULTS

One hundred sixty-one multiple trauma patientswere admitted to our intensive care unit during thestudy period. Some relevant characteristics of thesepatients are summarized in Table 1. One hundredeight (67 percent) patients received mechanical yentilation for more than 48 h, through orotracheal (84)or tracheostomy (24) tubes. Only five patients werepreviously diagnosed with COPD. Eighty-six (53.4percent) were associated with cranioencephalictrauma, 37 of whom underwent neurosurgical procedures. Seventy-one patients (44 percent) showed GCSbelow 9, during a period greater than 24 h, and theywere classified as group A. An episode oflarge volumebronchoaspiration was observed in only five patients.Patients from group A had a higher incidence andseverity of cranioencephalic trauma, and as a consequence, they were associated with a greater percentage of mechanical ventilation or neurosurgery and alower percentage of other surgery. However, we couldnot find significant differences between time of onsetof nosocomial pneumonia, massive bronchoaspiration,or other variables recorded.

Thirty-eight (23.6 percent) patients developed anosocomial pneumonia. In addition, there were foursuperinfections in three patients, representing an

*CET, cranioencephalic trauma; COPD, chronic obstructive pulmonary disease; NP, nosoc()mial pneumonia. Values in parenthesesare percentages.

tExcluding neurosurgery.tp<0.05.

infection developed from 48 h postadmission;° if the infectiondeveloped earlier than this, it was considered to be communityacquired.'@ The criteria for diagnosis of pneumonia were as follows:presence of a new and persistent lung opacity on the chest x-rayfilm, excluding those with clinical evidence ofnoninfectious origin,'4plus two of the following items: (a) fever >38°C, (b) leukocytosis> 10,000/cu mm , (c) punilent respiratory secretions. Superinfectionwas defined on the basis of recurrence of fever, leukocytosis, andincreasing symptoms and signs of infection after definitive clinicaland roentgenographic improvement of a prior 1@Development of fever, leukocytosis, and puntlent respiratory secretions,without lung opacity on chest x-ray film were defined as purulenttracheobronchitis. Systolic blood pressure below 90 mm Hg inassociation with clinical signs ofperipheral hypoperfusion secondaryto pneumonia was considered to indicate septic shock, and COPDwas diagnosed according to standard criteria.' â€A large volumeaspiration was considered present when there was a clinicallydocumented episode of aspiration of food or gastric contents.

Table 2—Characteristics of42 Episodes of NosocomidPneumonia among 161 Multiple Trauma Patients*

*Values are numbers of episodes. ATB prior antimicrobial treatment.

526 Nosocomial Respiratory Tract Infections (ReI!o et a!)

 © 1992 American College of Chest Physicians by guest on July 10, 2011chestjournal.chestpubs.orgDownloaded from

Group A (n=30)

Microorganisms I (%) 11100 EEpisodesGroup

B (n= 12)‘Iital (n@ 42)I

(%) 1/100 EI (‘@%@ 1/1(X)E

incidence of 26 percent (42 of 161). Thirty (42.2percent) episodes developed among 71 patients fromgroup A, compared with 12 (13.3 percent) episodesamong 90 patients from group B (p<O.O5). Averagetime ofonset ofnosocomial pneumonia was nine days.All except three patients who developed nosocomialpneumonia were receiving mechanical ventilation.Two patients had septic shock and another two developed secondary bacteremia; pathogens isolated fromblood were Staphylococcus aureus and Serratia marcescens. Furthermore, six patients had a diagnosis ofpurulent tracheobronchitis. Thirty-two patients (19.8percent) died, but only five deaths were directlyrelated to nosocomial pneumonia. The mean length ofstay of pneumonia survivors was 23.9 days comparedwith 10.5 days for multiple trauma patients withoutpneumonia (p<O.Ol).

Identification of the causative agent was possible in36 (85.7 percent) of these 42 lower respiratory tractinfections; of these 36, nine (25 percent) were polymicrobial infections. Details of these infections can befound in Table 2. Once pneumonia was established,no clinical or roentgenographic features were helpfulin distinguishing the etiology

A total of 46 microorganisms were cultured fromPSB in a significant amount (more than 10@cfulml).All the known etiologic agents are classified accordingto level of consciousness in Table 3. Staphylococcusaureus was the predominant pathogen (21 isolates) in

our series, and all strains were methicillin-sensitive.Strikingly, S aureus accounted for 55.8 percent ofisolates in patients in coma (group A), but it represented only 16.6 percent of the total microorganismsisolated in patients with GCS 9 (group B); in thisgroup of patients, aerobic Gram-negative bacilli, including Hemophilus influenzae, represented 74.9 percent of isolates. Anaerobic flora was present in onlytwo episodes.

DISCUSSION

In the review by Baker et al21ofdeaths from trauma

in San Francisco, 53 percent ofthe patients were (k'a(Iat the scene of the injury heftre transport wasavailable, 7.5 percent of patients (lied in the emergency room, and 39.5 percent died in the hospital.Sepsis was the cause of death in the majority of latedeaths (>7 days). The appearance ofsepsis, respiratoryinsufficiency, and multiple system organ failure are

common terminal complications in the course of themultiple trauma victim. The etiology of infection inmultiple trauma series is principally pulmonary orabdominal.@ Using classic diagnostic criteria, we founda mean incidence of 26 percent for nosocomial pneu

monia in multiple trauma patients. Mechanical ventilation is an important risk factor for nosocomialpneumonia and was present in 67 percent of allpatients and in 96. 1 percent of those who developedthis complication. In the literature, the incidence of

ventilator-associated pneumonia varies widely andranges from 9 to 21 percent ofpatients with respiratoryfailure@'2@to more than 70 I)ercent in patients whodied of ARDS.Li@SIn a previous sttidy,@ we found asimilar rate ofincidence for ventilator-associated i@i@etimonia in an unselected population. Our current finding that an important depression oflevel of consciousness significantly increases the percentage ofnosocomial pneumonia in multiple trauma l)atientsagrees with previous reports showing that coma is asignificant risk factor for developing no@ocom ialpneumonia2' in other populations.

Oropharyngeal and gastric carriage followed byaspiration results in colonization of lower airways andmay later lead to nosocomial pnetitn@nia.25@ Multipletrauma is often associated with altered consciousness,abnormal swallowing, and depressed gag reflexes.Such patients may aspirate oropharyngeal contentsduring the process of injury, resuscitation, tI1(l earlyhours in hospital. This is a well—knownphenomenonin experimental models.3' Furthermore, intubationincreases the risk of aspiration and impairs bronchialclearance mechanisms. These mechanisms pr )hal)lvwere the most important factors in developing respi

Table 3—Summary ofCausative Agents*

Staphylococcusaureus 19 (55.8) 63.3 2 (16.6) 16.6 21@45.6!5!)Streptococcuspneurnoniae 4 (11.7) 13.3 0 . . . , . . 414.6)9.5Hemophilusinfluenzae

7 (20.5) 23.3 2 (16.6) 16.6 9(14.5)21.4OtheraerobicGNB3t (8.8) 10 7@(58.3) 50 1021.7)21.4Anaerobic

flora 1 (2.9) 3.3 1 (8.3) 8.3 2(4.3@4.7Unknown3 . . . 10 3 . . . 2.5 6 . . .

*GNB, Gram-negative bacilli; I, isolates; E, episodes ofnosocomial pneumonia. Values in parentheses are percentages olisoLde's.14.2t2Pseudoinonas aeruginosa, 1 Escherichiacoli.:@:4Serratia marcescens 2 P aeruginosa, 1 Alcaligenesfaecalis.

CHEST I 102 I 2 I AUGUST, 1992 527

 © 1992 American College of Chest Physicians by guest on July 10, 2011chestjournal.chestpubs.orgDownloaded from

ratory tract infections in our population. However, inour study, the mean onset of nosocomial pneumoniawas nine days, and no differences were found in thisvariable between groups A and B. Furthermore, wecould identify only clinical signs of large volumebronchoaspiration in five of our patients, althoughsubclinical episodes may occur.

The overall mortality rate was 19.8 percent (32patients), and 80.2 percent (129 patients) survived.However, pneumonia was considered to be the causeof death in only five of our patients. Among survivors,the meanICU staywasovertwiceaslongasthat ofcorresponding survivors without pneumonia. Previousstudies@ found that nosocomial pneumonia significantly prolonged the stay ofpatients in ICU; however,with our current data, we could not conclude ifpneumonia prolongs the stay or if the length of stayincreases risk of pneumonia. Early diagnosis of nosocomial pneumonia in trauma patients and an aggressivestrategy for etiologic diagnosis, allowing the adminis

tration of adequate empiric antimicrobial therapy inthe majority of them, probably contributed to improved prognosis. Understanding the epidemiologiccharacteristics of respiratory tract infections in thishighly selected population of injured persons is thefirst step toward identifying patients at increased riskand taking measures to prevent these complications.

The most common etiologic agent of nosocomialpneumonia in our study was S aureus (45.6 percent).Hemophilus influenzae was responsible for 19.5 percent and other aerobic Gram-negative bacilli for 21.7percent, and Streptococcus pneumoniae and anaerobicflora represented 8.6 percent and 4.3 percent ofisolates, respectively. Furthermore, a close relationship was observed between etiology and level ofconsciousness, as determined by the GCS. In patientsfrom group A (GCS below 9 during a period greaterthan 24 h), S aureus represented the most commonetiologic agent (55.8 percent), while Gram-negativebacilli (excluding H influenzae) were responsible foronly 8.8 percent ofisolates. In addition, S pneumoniaewas isolated in four (11.7 percent) cases. However, inpatients from group B, aerobic Gram-negative bacillirepresented more than 50 percent and S aureus only16.6 percent of microorganisms isolated. These dataagree with previous authors@'@ who reported thatGram-negative bacilli were the microorganisms mostfrequently isolated, with a range of 40 to 87 percent,while Gram-positive cocci were responsible in lessthan 20 percent of cases@ of hospital-acquired pneumonias. A high incidence of S aureus pneumonia inneurosurgical@ and in head-trauma patients requiringmechanical ventilation37 has previously been described. Rello et aP@recently reported that age below25 years, nonuse of corticosteroids, antecedent oftrauma, and coma were significantly associated with S

aureus nosocomial pneumonia in the univariate analysis in a population of critically ill patients withnosocomial pneumonia. However, a step-forward logistic regression analysis removed antecedents oftrauma and other variables and defined only coma assignificantly influencing the risk ofdeveloping S aureus

nosocomial pneumonia. Further studies are needed toreport causative agents of nosocomial pneumonia incomatose patients and to clarify why patients in comaare at high risk to develop airway colonization by Saureus. The usefulness of nonabsorbable topical antimicrobials (selective decontamination of digestivetract)b03139 in preventing “early―onset nosocomialpneumonia in multiple trauma patients is slight, andsystemic antimicrobials are administered by someauthors.639 However, S aureus is usually resistant tothese antimicrobials, as well as to the antibioticsusually selected for empiric treatment of nosocomialpneumonia. Our results suggest that antimicrobialdrugs active against S aureus should be included inthe empiric antimicrobial regimen for treating nosocomial pneumonia in multiple trauma patients incoma.

As discussed previously in this paper, our datasuggested that a substantial proportion of nosocomialrespiratory tract infections in multiple trauma patientsare caused by S aureus. The virulence factors determining their pathogenicity and the host determinantsunderlying susceptibility to carriage and disease arepoorly understood. A few studies and substantialcircumstantial evidence implicate that level of consciousness,asdetermined by the GCS, influences thebacteriology of pneumonia, and therefore, has implications for therapy. In our view, there is a need toinvestigate the pattern of colonization of critically illcomatose patients and to clarify interactions of Saureus with the microbial and host factors that determine the successful establishment of different pathogens in the respiratory tract. Such studies will also beuseful in developing more effective strategies for theprevention of this life-threatening nosocomial infection.

ACKNOWLEDGMENT: The writers thank Rosa de Vilar for hersecretarial assistance.

REFERENCES

1 Butterworth JF, DeWitt DS. Severe head trauma: pathophysiologv and management. Grit Care Cliii 1989; 5:807-20

2 Ruehsen MM , Abdul-Wahab AWM . The epidemiology of trauma

in an intensive care unit in Bahrain. J Trauma 1989; 29:31-363 Gross PA. Epidemiology ofhospital-acquired pneunionia. Semin

Respir Infect 1987; 2:2-74 Delany IIM, Berlin AW Multiple injuries. In: Tinker J, Rapin

M, eds. Care ofthe critically ill patient. heidelberg, Germany:Springer-Verlag, 1983; 611-27

5 Espersen F, Gabrielsen J. Pneumonia due to Staphylococcusaureus during mechanical ventilation. J Infect Dis 1981; 144:19-23

528 Nosocomial Respiratory Tract Infections (Reio et a!)

 © 1992 American College of Chest Physicians by guest on July 10, 2011chestjournal.chestpubs.orgDownloaded from

6 Craven DE, Daschner FD. Nosocomial pneumonia in theintubated patient: role of gastric colonization. Eur J ClinMicrobiol Infect Dis 1989; 9:40-50

7 Johanson WG. Infection prevention by selective decontamination in intensive care. Intensive Care Med 1989; 15:417-19

8 Weinsten BA. Selective intestinal decontamination: an infectioncontrol measure whose time has come? Ann Intern Med 1989;110:853-55

9 Microbial selective decontamination in intensive care patients[editorial]. Lancet 1988; 1:803-04

10 Stoutenbeck CP, van Saene HKF, Miranda DR. Zandstra DF,Langrehr D. The effect oforopharyngeal decontamination usingtopical nonabsorbable antibiotics on the incidence of nosocomialrespiratory tract infections in multiple trauma patients. J Trauma1987; 27:357-64

11 Comitéde Consensus. V Conference de consensus en réanimation et medicine d'urgence: diagnostic des pneumopathiesnosoc()miales en reanimation. Paris: Le Kremlin-Bicêtre, Oct13, 1989

12 Sanford JP, Pierce AK. Lower respiratory tract infections. In:Bennet JV, Brachman PS, eds. Hospital infections. Boston:Little, Brown & Go, 1979; 255-86

13 Centers for Disease Control. National nosocomial infectionsstudy: site definitions manual. Atlanta: Centers for Disease

Control, 1972; 82-8414 Esteban A, Fernández-Segoviano P. Oliete 5, Ruiz-Santana 5,

Castelló J, de Ia Cal MA. Radiographic findings for the adultrespiratory distress syndrome in patients with peritonitis. GritCare Med 1983; 11:880-84

15 Grayhill JR. Maughall LW, Charache P, Wallace CK, MelvinVB. Nosocmial pneumonia: a continuing major problem. AmRev Respir Dis 1973; 108:1130-40

16 Rimel RW, Tyson GW The neurologic examination in patientswith central nervous system trauma. J Neurosurg Nurs 1979;

12: 148-5217 Wimberley NW, Bass JB, Boyd BW, Kirkpatrick MB, Serio BA,

Pollock H M . Use of a bronchoscopic protected catheter brushfor the diagnosis of pulmonary infections. Chest 1982; 81:556-

6218 Edelstein P11. Improved semiselective medium for isolation of

Legiondlla pneuinophila from contaminated clinical and environmental specimens. J Glin Microbiol 1981; 14:298-303

19 Chastre J, Viau F, Bnin P. Pierre J, Dange MG, Bouchame A.Prospective evaluation of the protected specimen brush for thediagnosis ofpulmonary infections in ventilated patients. Am RevRespir Dis 1984; 130:924-29

20 Lennette EH, Bullows A, Hausler WJ Jr, Sadhomy HJ. Manualof clinical microbiology, 4th ed. Washington: American Societyfor Microbiology, 1985

21 Baker CC, Oppenheimer L, Stephens B, Lewis FR, TrunkeyDD. Epidemiology oftrauma deaths. Am J Surg 1980; 140:144-50

22 Meduri GU. Ventilator-associated pneumonia in patients withrespiratory failure. Chest 1990; 97: 1205-19

23 Graven DE, Kunches LM, Kilinsky V, Lichtenberg DA, MakeBJ, McCabe WR. Risk factors for pneumonia and fatality inpatients receiving continuous mechanical ventilation. Am Rev

Respir Dis 1986; 133:792-9624 Fagon JI, ChastreJ, Hance AJ, Guiguet M , Tronillet JL, Domart

Y, et al. Detection of nosocomial lung infection in ventilatedpatients: use of a protected specimen bnish and quantitativeculture techniques in 147 patients. Am Rev Respir Dis 1988;138:110-16

25 Bell RC, Coalson JJ, Smith JD, Johanson WG Jr. Multiple organsystem failure and infection in adult respiratory distress syndrome. Ann Intern Med 1983; 99:293-98

26 Rello J, Quintana E, Ausina V, Castella J, Luquin M, Net A, etal. Incidence, etiology and outcome of nosocomial pneumoniain mechanically ventilated patients. Chest 1991; 100:439-44

27 Gelis R, Torres A, Gatell JM, Almela M, Rodriguez-Roisin R,.@gusti-Vidal A. Nosocomial pneumonia: a multivariate analysisofrisk and prognosis. Chest 1990; 97: 1208-19

28 La Force FM . Hospital acquired Gram-negative rod pneumonia:an overview. Am J Med 1981; 70:664-68

29 Graven DE, Daschner FD. Nosocomial pneumonia in theintubated patient: role of gastric colonization. Eur J GunMicrobiol Infect Dis 1989; 9:40-50

30 van Saene HKF, Stoutenbeck CP, Zandstra DF. Pathophysiologyof nosocomial pneumonia in intensive care units. Rean SoinsIntens Med Urg 1990; 6:100-02

31 Johanson WG. Conclusions. In: van Saene HKF, StoutenheckCP, Lawn P, Ledingham IMA, eds. Update in intensive careand emergency medicine: infection control by selective decontamination. Heidelberg, Germany: Springer.Verlag, 1989; 215-19

32 Craig CP, Connelly S. Effect of intensive care unit nosocomialpneumonia on duration of stay and mortality. Infect Control1984; 12:233-38

33 Daschner FD, Frey P. WolfG, Bauman PC, Suter P Nosocomialinfections in intensive care wards: a multicenter prospectivestudy. Intensive Care Med 1982; 8:5-9

34 Winterbauer RH, Hutchinson JF, Reinhardt GN, Sumider SE,Dearden B, Thomas CA, et al. The use of quantitative culturesand antibody coating ofbacteria to diagnose bacterial pneumoniaby fiberoptic bronchoscopy. Am Rev Respir Dis 1983; 128:98-103

35 Pollock HM, Hawjins EL, Bonner JR. Sparkman T, Bass JB.Diagnosis of bacterial pulmonary infections with quantitativeprotected catheter cultures obtained during bronchoscopy.J Clin Microbiol1983;17:255-59

36 Baugham RP, Thorpe JE, Staneck J, Bashkin M, Frame PT. Useof the protected specimen brush with endotracheal or tracheostomy tubes. Chest 1987; 91:233-36

37 Braun SR, Levin AB, Clark KL. Role of corticosteroids in the

development of pneumonia in mechanically ventilated headtrauma victims. Grit Care Med 1986; 14:198-201

38 Rello J, Quintana E, Ausina V. Puzo C, Net A, Prats C. Riskfactors for Staphylococcus aureus nosocomial pneumonia incritically ill patients. Am Rev Respir Dis 1990; 142:1320-24

39 van Saene HKF, Stoutenbeck CP, Lawn P. Ledingham IMA.Update: intensive care and emergency medicine—infectioncontrol by selective decontamination. Heidelberg, Germany:

Springer-Verlag, 1989

CHEST I 102 I 2 I AUGUST, 1992 529

 © 1992 American College of Chest Physicians by guest on July 10, 2011chestjournal.chestpubs.orgDownloaded from

DOI 10.1378/chest.102.2.525 1992;102; 525-529Chest

J Rello, V Ausina, J Castella, A Net and G Pratsof level of consciousness with implications for therapy.

Nosocomial respiratory tract infections in multiple trauma patients. Influence

 July 10, 2011This information is current as of

 

http://chestjournal.chestpubs.org/content/102/2/525Updated Information and services can be found at:

Updated Information & Services

http://chestjournal.chestpubs.org/content/102/2/525#related-urlsThis article has been cited by 14 HighWire-hosted articles:

Cited Bys

http://www.chestpubs.org/site/misc/reprints.xhtmlonline at: Information about reproducing this article in parts (figures, tables) or in its entirety can be foundPermissions & Licensing

http://www.chestpubs.org/site/misc/reprints.xhtmlInformation about ordering reprints can be found online:

Reprints

the right of the online article.Receive free e-mail alerts when new articles cite this article. To sign up, select the "Services" link to

Citation Alerts

slide format. See any online figure for directions. articles can be downloaded for teaching purposes in PowerPointCHESTFigures that appear in Images in PowerPoint format

 © 1992 American College of Chest Physicians by guest on July 10, 2011chestjournal.chestpubs.orgDownloaded from