Prevention of
Ventilator-Associated
Pneumonia Through
Aspiration of Subglottic
Secretions
A Systematic Review and Meta-Analysis
A. Renee Leasure, PhD, RN, CCRN, CNS; Joan Stirlen, MPH, BSN, RN; Shu Hua Lu, PhD, RNVentilator-associated pneumonia (VAP) is a subset of hospital-acquired
pneumonias and is a serious, sometimes fatal, complication in patients who
need mechanical ventilation. In addition, pay-for-performance initiative has
placed increased emphasis on preventing nosocomial infections including
VAP. Facilities may not be reimbursed for costs associated with prevalence
infections. This article presents a review and meta-analysis of the
prevention of VAP through the aspiration of subglottic secretion.
Keywords: Aspiration, Subglottic secretions, Ventilated-associated
pneumonia
[DIMENSCRITCARENURS. 2012;31(2):102/117]
C
ritically ill patients are particularly prone to infectionseither because of their illness, comorbidities, or invasive devices/procedure associated with their critical manage-ment. Ventilator-associated pneumonia (VAP) is the most
common nosocomial infection in critically ill patients1
and the second most common hospital infection.1
Pay-for-performance initiatives have placed increased
empha-sis on preventing nosocomial infections including VAP.2,3
The Pathogenesis of VAP
Ventilator-associated pneumonia is a subset of
hospital-acquired pneumonias.4The Centers for Disease Control
and Prevention’s National Healthcare Safety Network defines VAP as a pneumonia that develops in patients
who are intubated and ventilated at the time of or who develop a pneumonia within 48 hours of
discontinua-tion of mechanical ventiladiscontinua-tion (MV).5 However, some
definitions consider VAP pneumonia to be a pneumonia that develops after 48 hours of mechanical ventilator with pneumonias developing prior to 48 hours to be either a preexisting pneumonia or the result of aspira-tion during intubaaspira-tion. Ventilator-associated pneumonias are further classified as being either early onset or late onset. Early-onset pneumonias that develop within the first 4 days are often caused by organisms such as Moraxella catarrhalis, Haemophilus, and Staphylococcus pneumoniae. Late-onset pneumonias are caused by agents such as gram-negative bacilli, Staphylococcus aureus,
methicilin-resistant S aureus, and viruses such as
influ-enza A, influinflu-enza B, legionnellae, yeasts, and fungi.5
Patients who require MV are either intubated with an endotracheal tube (ETT) or ventilated through place-ment of a tracheostomy. The longer a patient is
intu-bated, the greater the likelihood of developing VAP.6
Also, later-developing VAPs are more likely to be caused by antibiotic resistant organisms. Ventilator-associated pneumonia increases costs to payors and the health care system as a whole through increased lengths of stay in high-cost intensive care units (ICUs), increased length of
hospital stay, and additional costs of antibiotic therapy.7-10
Ventilator-associated pneumonia is a preventable com-plication of MV and is also associated with increased mortality and morbidity. Pneumonia rates are 6 to 21 times higher in patients receiving MV, and the risk in-creases by as much as 1% per ventilator day.
Several mechanisms have been cited as causative
factors in the pathogenesis of bacterial pneumonia.11,12
They include bacteremia, gastrointestinal colonization, inhalation of pathogens from the environment, and in-troduction of pathogens from the environment such as through suctioning, MV, or colonization of the oral
cav-ity with VAP-associated pathogens.13-15
Although a frequently cited complication of MV, VAP
is not inevitable.16,17As national focus has highlighted the
significance of this problem,18a barrage of interventions has
been proposed in consensus standards, national campaigns,
and in the literature.19-26 It is important to continually
analyze existing science underlying these recommenda-tions so that beneficial intervenrecommenda-tions may be implemented and the need for further studies may be highlighted.
Subglottic Secretion Drainage
The cuff of an ETT serves several purposes. It helps secure the tube in the correct location, helps in the de-livery of appropriate tidal volumes, and aids in the pre-vention of aspiration of secretions from either the oral cavity and/or the stomach. However, over time, secre-tions can accumulate above the endotracheal cuff, and aspiration into the lungs can occur if these fluids remain
in place.27,28 Contamination of the lower respiratory
tract by these secretions can cause VAP.
Aspiration or drainage of subglottic secretions is available through a specialized ETT, which has an ad-ditional lumen above the cuff. This lumen can be con-nected to either continuous or intermittent suction. Manufacturers recommended suction rates that vary from j20 mm Hg to around j100 to j150 mm Hg (Mallinckrodt and Nelcor, both in St Louis, Missouri).
Objective of the Review
The objective of this review was to examine the effective-ness of subglottic secretion aspiration in reducing the
oc-currence of VAP. The sub-objectives of this review were (1) reduction in VAP rates, (2) duration of MV, (3) mor-tality, (4) length of stay, and (5) length of hospital stay.
METHODS
Criteria for Considering Studies for Review
Original studies were included in this review if they fo-cused on subglottic secretion drainage (SSD) and met the following criteria: (1) utilized a prospective design; (2) sampled human subjects, and (3) had a control group. Ultimately, other systematic reviews were identified and included in the discussion of this systematic review.Types of Participants
Participants who served as the focus of this review were human subjects hospitalized in an ICU who were in-tubated and receiving MV.
Types of Interventions
Subglottic secretion drainage comprised the intervention of interest. Studies were included that implemented either intermittent or continuous SSD.
Subglottic secretion drainage
comprises the intervention
of interest.
Types of Outcome Measures
Dichotomous outcomes included (1) the presence or ab-sence of ventilator pneumonia and (2) mortality. Con-tinuous outcomes included (1) incidence of VAP per 1000 ventilator days, (2) days to onset of VAP, (3) duration of MV, (4) length of ICU stay, and (5) length of hospital stay.
Search Strategy for Identification of Studies
Medline (OVID and PubMed) 1448 to March week 1 2011; EMBASE 1980 to 2011 week 16; Medline in-process and other nonindexed citations April 26, 2100; Cochrane Database of Systematic Reviews 2005 to March 2011; EBM ReviewsVACP Journal Club 1991 to March 2011; OVID Nursing; Dissertation Abstracts; BMJ Clinical Evidence; and CINAHL 1981 to April 2011 databases were utilized to locate relevant abstracts for review. Search terms for the outcome variables in-cluded (1) pneumonia (prevention and outcome variables), (2) ventilator-associated pneumonia, and (3) pneumonia,ventilator-associated. Search terms for the intervention included (1) intracheal intubation, (2) endotracheal in-tubation, (3) endotracheal cuff, (4) endotracheal tube, (5) subglottic, and (6) glottis. Searches were then performed using both ‘‘key word’’ searches and also mapping to MeSH (Medical Subject Headings) subject headings. Searches were combined using the Boolean operator ‘‘and’’ and then limited to clinical trials and humans. As studies were retrieved, reference lists were reviewed, and possible relevant studies obtained for review of either the abstract or full review of the article. After variables of interest were combined and limits applied, 259 articles were selected for abstract review. From this list, 38 articles were selected for full review.
Web of Science Citations was searched for articles that had cited the studies listed in Table 1. The most frequently cited studies and representative number of
studies were Valles et al40(254), Mahul et al37(157), and
Kollef32(108). Abstracts of the 688 articles that cited
studies listed in Table 1 were also reviewed for inclusion in this review.
Sixteen articles, which included original studies (n = 12) and reviews (n = 4), were included in this systematic review. Three studies were written in Chinese and were translated by S.H.L., a doctorally prepared nurse with a critical care background and experience conducting sys-tematic reviews.
Data Collection and Analysis
SELECTION OF STUDIES
Two authors independently searched the literature to lo-cate studies. Studies were included if they examined SSD.
DATA EXTRACTION AND MANAGEMENT
Two reviewers independently extracted data pertaining to outcome variables with no difference in data extrac-tion noted.
Two reviewers independently appraised the
studies and extracted data pertaining
to outcome variables.
ASSESSMENT OF METHODOLOGY
Two reviewers (A.R.L. and J.S.) independently assessed studies for methodologic quality using a standardized checklist. When reviewing studies, we addressed repre-sentativeness of the sample. It was considered desirable when investigators recruited subjects sequentially and
also provided data that reflected how subjects were sim-ilar or different from the accessible population. As VAP prevention practices have changed throughout the last decade and because the development of science has oc-curred in an international setting, we reviewed the de-scription of clinical care in order to clearly identify and
describe the standard of care for this patient group.43-45
Specifically, we looked for standard care measures such as head-of-bed elevation, routine oral care protocols, use of a standardized weaning protocol, maintenance of
endotracheal cuff pressure between 20 and 25 cm H2O
or 20 and 30 cm H2O,46,47and stress ulcer prophylaxis.48
In an ideal world, each study would have used the same criteria for screening for the suspicion of and then
con-firming the presence of VAP.49 However, the
interna-tional settings utilized similar but different criteria; we included only the randomized studies that used a pro-spectively identified criteria and that utilized prospec-tively identified standardized screening and diagnostic tools that were uniformly applied to both the
experi-mental and control groups.50-52
Across studies, 3 components were utilized in es-tablishing the diagnosis of VAP. The first component addressed systemic signs of infection such as fever, tachy-cardia, and leukocytosis. The second component ad-dressed chest x-rays, whereas the third component examined bacteriologic evidence of pulmonary infection
based on culture results.53,54When reviewing studies, we
looked for the use of standardized criteria in determin-ing suspicion of VAP. Serial x-rays were preferred to a single x-ray, and it was preferable if the radiologist was
blinded to group assignment when evaluating the x-rays.49
Although tracheal secretions are easily obtained through endotracheal suctioning, the results often are contaminated by upper respiratory tract pathogens. Bronchoalveolar lavage or protected brush specimen with either calibrated loop or serial dilution techniques for microbiologic evaluation was considered to be the most
desirable measures for confirmation of VAP.55,56
Al-though the visibility of the specialized ETT made it im-possible to blind investigators and clinicians to group assignment, we did evaluate whether the radiologists who read chest x-rays and the microbiologists who an-alyzed the laboratory specimens were blinded to group assignment.
Disagreements regarding study quality were resolved through discussion or through consultation with a third party. In an attempt to avoid either a too restrictive or too lenient approach to inclusion in this systematic re-view, we reviewed and abstracted data from each study prior to discussing the results (Tables 1 and 2). Agreement among the review group was reached in determining which studies to include in the final review.
TABLE 1 Characteri stic of Studies Abstracted and Reviewed Study Author and Design Setting and Accessible Population Intervention (I) and Control ( C) Determination of VAP Findings Comments Bo et al 29 Sur gical ICU in Shang hai. Excl uded were oral surg ery pati ents and those with pneu monia at stud y entry. I = SSD with continuous asp iration of secretion s. VAP diagnose d bas ed on quantitative cultu res obtai ned by PBS. The same organis ms were isola ted in subg lottic secr etions and the lowe r respi ratory tra ct in 61% (14/23) pati ents with VAP . All of the su bj ects rec eived stress ulce r pro phy lax is an d had an NG tu be in place. RCT Include d C = convent ional ET T Bouza et al 30 ICU in a 1750 bed hos pital in Mad rid, Spain. The sampl e was drawn from 1101 pati ents wh o had major heart surgery bet ween May 200 4 and July 200 6 I = SSD with continuous asp iration of secretion s with a negat ive pre ssure main tained bet ween 100 and 150 mm Hg. The CDC def inition was used to scr een for VAP . Patients wi th a clini cal pulmon ary infection score 9 6 were also consi dered to have pneu monia. VAP was con firmed by the is olating of Q1 bac terial cou nts wi th asp irates obtai ned from either endot racheal asp iration or telesc opic brush sam pling of respi ratory secr etions. No compli cations in the SSD grou p wer e obs erved. An tibiotic burden was signifi cantly reduced with a calcu lated savings of appro ximately 21,00 0€ . The extra cost of acq uiring SSD tubes was appr oximately 280 0 € . Following info rmed con sent proced ures 714 patie nts undergo ing MHS were randomi zed by drawing a card from a seal ed envelop e. RCT Of the initial group, 85 pati ents were mechan ically ventilate d Q48 h after undergo ing surgery. This subg roup was included in this meta-analysis. Include d C = convent ional endot racheal tub e All patients rece ived stress ulcer prophylaxis with panto prazole. Girou et al 31 10-B ed ICU in a Fren ch University Hos pital I = semir ecumbent and SSD (N=8 ) Or opharyngeal and tra cheal secr etions were sampl ed and cultu red daily fro m day 1 to 10, extu bation, or death. After day 1 of mec hanical vent ilation, 75% of the SSD and 80% of the con trol group pati ents were colon ized in the trache a. There was no signifi cant differ ence in daily tra cheal and oroph aryngeal bac teria counts bet ween the 2 grou ps. Thus, SSD did not modi fy the level of oroph aryng eal and tra cheal coloniza tion. All patients rece ived sucraf ate, which may have affecte d coloniza tion RCT ICU patient s expecte d to be mechan ically vent ilated 9 5d were elig ible for inclusio n. C = supine and con ventional ET T (N = 10) Among the 5 patient s extubat ed in the suction ing group, 2 (40%) develop ed laryng eal edema immed iatel y after extubat ion. Excluded The fre quency of oral care was not reporte d. Kollef et al 32 Subj ects were drawn from the car diothoracic ICU at Barn es Je wish Hospital in St Louis, Mis souri; 343 cardiac surgery pati ents who req uired MV were inclu ded in the stud y. I = SSD with continuous low inte rmitten t suction G 20 mm Hg The Ameri can College of Chest Phys icians crit eria were used to scr een for VAP. A clinical diagnosis of VAP was utilize d that was not bas ed on broncho scopical ly obtai ned cultures of the lower respi ratory tract. Episo des of VAP occurred sta tistically later among the SSD group by 2.7 day s. Birth years were used to assig n patients to groups. RCT Pati en ts in this st udy wer e intubated for a re lati ve ly brief period as com pared wit h ot her st udies. (continues)
TABLE 1 Characteri stic of Studies Abstracted and Reviewe d, continue d Study Author and Design Setting and Accessible Population Intervention (I) and Control ( C) Determination of VAP Findings Comments Excluded C = SSD wi thout suction No com plications relate d to SS D were observed in the interv ention grou p. Outcomes assesso r was blinded to grou p assignm ent Lacherad e et al 33 Subj ects were drawn from 4 medica l/surgic al ICUs in France I = SSD (Hi-Lo Evac) su ctioned man ually wi th a 10-mL syringe wi th an intend ed fre quency of 1 suction per hour . Clin ical suspicion of VAP was bas ed on the presen ce of a recent and per sistent infiltrate on chest x-ray and at least 2 of the following: fever gre ater than 38.3 -Co r hypot her mia of less than 36 -C, whit e blood cell count greater than 10.10 /L or less than 4.10/L, and purul ent trache al secr etions. Co nfirmation of VAP req uired a pos itive quantitative cultu re of eith er a protected telesc oping cathet er sampl e or BAL . Cultures were con sidered positive if the cathet er sampl e or BAL gre w at leas t 10 3 or at least 10 4 colony-for ming units/ mL, respectively, of at leas t 1 m icroo rga nism . Of the cases of VAP, 2 of 169 cases in the experime ntal grou p and 10 of 164 case s in the control group occ urred within 5 days. Late-onset VAP that occur red afte r 5 day s was 23 of 126 cases in the exper imental grou p and 32 of 97 cases in the control group. Analysis included 9 cont rol gr oup and 8 SSD group patients w ho req ui red MV fo rG 48 h. RCT Routine care measures inclu ded ora l ra the r tha n na sa l rou te of intubation of tracheal and gastric tub es, enteral delivery of nutri tional supp ort , ETT cuff pressures between 20 and 30 cm H2 O, and semirec umbent body positio n. No selective digestion decont amination was used. Include d C = Hi-Lo Eva c ETT without asp iration Within th e firs t 48 h of M V 9 SS D and 15 con tro l group pa tients dev eloped pneumon ia. These su bjects were in cl uded in the ana ly sis. Liu et al 34 Sixt y subject s aged Q60 y who were expec ted to req uire mec hanical ventilat ion for 9 48 h were recruit ed fro m an ICU in Shang hai, China I = SSD (continuo us su bglottic secr etion drainage), semir ecumbe nt position, and mos apride ci trate VAP diag nostic criteria as des cribed by B er gma ns et al 35 were followed. Inci dence of VAP the interv ention grou p was lower than that in con trol group SSD was 1 of a 3-part bundl e of interv entions that was compar ed with stand ard care, which included a stand ard endotracheal tube. Th us, it was not possible to determine the contribution of each of the 3 elements of the in tervention in preventing VAP. C = Standa rd endotracheal tube Excluded Ea rly V A P w as co ns id er ed a V A P th at occurred wi thi n 5 d of endotracheal in tub ation wi th m ech an ical ve ntil ation, wh er eas late V AP was th at whic h occurred aft er 5 d (continues)
TABLE 1 Characteristi c of Studies Abstracted and Reviewed, continued Study Author and Design Setting and Accessible Population Intervention (I) and Control ( C) Determination of VAP Findings Comments Lorente et al 36 The setti ng for this study was a 24-bed medica l-surgical ICU in a 650-bed ter tiary hospita l in Tenerife , Spain. Subjects were recruit ed between March 1, 2006, and Oc tober 31, 200 6. I = SSD (Seal-G uard [Covidien , Boulder, Colorado] EVAC Endotrachea l tube or Sea l Guard tra cheostomy ), whic h has a poly urethane cuff. Intermitt ent secre tion drainage every 1 h with a 10-mL syring e Th e diagn os is of pneum on ia was consi dered when all of the fol lowi ng cri teria were pre sent: ne w on se t of puru lent bronchi al spu tum; body tem perature > 38 -Co rG 35 .5 -C; whit e blood ce ll co un t9 10,0 00 or G 40 00/ H L; chest radio gr aph showing new or pr og ressive infi ltrates; and sign ifican t quant itativ e cult ure of respiratory secreti ons by tr ac he al as pi rate (9 10 6col ony-formin g uni ts/mL) Poisson re gre ssion analysis sh owe d a hig her in cidence den sity of V A P in the control group than in th e ex peri men tal gr oup. The investigators found a lower incidence of both early-and late-on set V AP . Stand ard care for both grou ps inclu ded no rout ine cha nge of vent ilator circuits ; trach eal suction by an open system as needed ; semir ecumbent body positi on every 4 h; intracuf f pres sure of 25 cm of H2 Ow hi ch was ve rified eve ry 4 h; N G tube ; con tinuou s enteric nutrition; stress ulcer prophylaxis w ith ranitidine; oral care w ith chlorhex idine eve ry 8 h; and no sele ctiv e dig estive decontamination RCT Th e pr oc ed ur es fo r the co nf irma tio n of the diagnosis of VAP were not described in det ail. Includ ed C = Hi-L o ETT that doe s not incorpo rate a sep arate dorsal lumen for subg lottic secre tion drainage The con firmation of diagnos is was made by an exper t panel bli nded to trea tment assignm ent. Pn eu moni a w as consi dered to be V A P wh en it was di agnosed during mechanical ve ntilation and w as not present at the time mechanical ve ntilator w as establi shed. The inv estigators di d not con sider a pn eu monia th at dev el oped w ithin the first 48 h to be an ex istin g pn eu monia. Mahul et al 37 The sampl e was drawn from 415 admis sions to an ICU in France. I = SSD (70) 34 received aluminum hydroxide, and 36 received su crafate. SGS was per formed hour ly wi th a 10-mL syringe. A new and per sistent infiltrate on CXR occur ring afte r 2 d of intubation was con sidered as nosocom ial pneumon ia. Diag nosis was con firmed with aerobi c microorgan isms on BAL of Q10.5 colony-fo rming units/mL The incid ence of VAP was 50% lowe r in the SGS group (13% ) as compar ed with the no-SGS grou p (29.1%) . Also, the days to on set of VAP were 16.2 in the SGS group and 8.3 in the contr ol grou p. Facto rial design Includ ed C = Regu lar ETT (75), 38 received aluminum hydroxide, and 37 received sucraf ate (continues)
TABLE 1 Characteris tic of Studies Abstracted and Reviewe d, continued Study Author and Design Setting and Accessible Population Intervention (I) and Control ( C) Determination of VAP Findings Comments Pneumatikos et al 38 14-B ed general ICU in a uni versity hos pital in Gr eece I = Hi-L o Evac ETT with con tinuous infusion of a sus pension of 3 nonab sorbable anti biotics (p olymyxin , tobr amycin, amphoter icin) VAP sus pected in the pre sence of new and persistent pul monary infilt rates in addi tion to: tem perature of gre ater than 38.3 -C, white blood cell cou nt of gre ater than 12,00 0/ 2 L or less than 400 0/ 2 L; and purul ent tr acheal secretion s. No patie nts with negative bronc hial secr etion cultures develop ed VAP . Ve ntilator-associate d pneumon ia develo ped in 16% of the patient s rec eiving sel ective dec ontamin ation of the subg lottic area and 51% of the patient s who received the placebo. Gastric and tr acheal secretion s w er e ob ta in ed af te r in tu bat io n and every 4 days th ereafter. All patients had a nasogastric tu be in pl ace an d if po ssib le were placed in a se m ire cu m be ntp os iti on ata 30 -to 45-degr ee an gle. Excluded C = Placebo infusion through a Hi-L o Evac ET T Sus pected VAP was confirm ed by a quan titative culture of secr etions in a prot ected speci men collected by a doub le cathet er either bli nd or bron choscopi cally. The diag nosis of VAP was mad e by a chest physician, radio logist, and a phys ician exper ienced in infe ctious disea ses who was bli nded to grou p assignm ent. The team decided on the presenc e or absence of VAP. Smulder s et al 39 12-B ed general ICU in Am sterdam, the Nether lands I = SS D intermit tent 20-s int ervals, 8-s dur ation, 100 Y mm Hg suction VAP was diagnos ed bas ed on a new or progress ive radio graphi c eviden ce for cavitat ion or hist ologic evidenc e of pneumon ia, or positive blood culture findi ng wi thout other source s. RCT All patients rec eived stress ulcer prophylaxis with sucra fate. The sampl e was drawn from 150 patients adm itted to the ICU who were expec ted to rec eive MV 9 72 h. C = stand ard ETT All subject s were orally intub ated. Includ ed A radiolog ist who w as blinde d to grou p assign m ent interpret ed al l C XRs . Valles et al 40 Subjects were recruit ed fro m an ICU in Spain . Those elig ible for inclu sion were those who wer e expected to be intuba ted at lea st 72 h. I = SS D VAP sus pected afte r 72 h of MV if tem perature 9 38.3 -C, WBC 9 12,000/ H Lo r G 400 0/ H L, purul ent secr etions, new or per sistent infiltrat e. Use of SSD red uced the incid ence of VAP by 43.4%. Intracuff pre ssures were monitor ed every 4 h and kept 9 20 mm Hg C = SSD endotracheal tube but did not receive asp iration of subglott ic secr etions. (continues)
TABLE 1 Characteristi c of Studies Abstracted and Reviewed, continued Study Author and Design Setting and Accessible Population Intervention (I) and Control ( C) Determination of VAP Findings Comments RCT 190 Patient s were recruited between 199 0 and 1993, with 153 com pletin g the study. C = SS D endotrac heal tube but did not receive aspiration of subglott ic secr etions. Diagnosis con firmed by po sitive protected bush specimen con taining Q103 colon y-form ing uni ts/mL, BAL Q104 colony-forming units/mL, or good respon se to antibiotic agent s. The mor bidity of VAP in the SSD grou p was 25%, and that for the con trol group was 46.5% (P = .032), and the leng th of time before the onset of VAP in these groups was 7.3 T 4.2 days and 5. 1 T 3d ay s, respectivel y (P = .10). In th eS SD gr ou p, th ev ol um eo ft he subg lotti c secretion s aspi ra ted the first day w as significantly less than th at in pa tients with out V AP (P = .00 6) . Th e m or bid ity of V A P in patients who faile d early aspira tion (the volu me of se cretion s first as pi ra te d e 20 mL) w as signif icantly higher than that in patients in w hom th e asp irat ion w as ef fe ct iv e (P G .01) . All pati ents received stress ulcer prop hylaxis. Based on cl inical presen tation. X-ray showing lun g infiltrat ion. At leas t 2 sym ptoms: body temper ature greater than 38.3 -C, white blood cell count greater than 12.0 10 9 /L or less than 4. 0 10 9 /L. D isc harg e cul tur e posit ive. C hest x-rays were interp reted by a radi ologist who was blinded to group assignm ent. At time of analysis, the inve stigators did analyze data to examine th e impact of an tibiot ic treat ment at time of ra ndo miz at ion on incidence of V AP. Includ ed Determin ation of VAP was det ermined using the simplified version of the clinical pul monary infec tion score (Q 5). Yang et al 41 Patient s mechanic ally ve ntilated in an ICU bet ween Octobe r 2004 and Ap ril 2006 were rand omized to grou ps. I = SSD con tinuous asp iration of subg lottic secretion s There was a signi ficant increase in the percentag e of gram-positive coc ci fr om the lowe r respi ratory tr acts in the con trol grou p (P = .0 04). RCT C = No aspir ation of subg lottic secretion s Zheng et al 42 From Januar y 2005 to June 200 6, patient s with an expec ted dur at ion of m echanical ventil ation fo r m ore than 48 hours and age older than 18 years w er e enrolled and random ized to groups. I = SSD The Nati onal Nos ocomial Infect ion Surveill ance System diagnos tic criteria for VAP wer e foll owed. Comp ared with the con trol grou p, the incidenc e of VAP was si gnifican tly lower (30% vs 51. 6%, P G .05), and the durati on of mechan ical ventilat ion was 7.9 T 2.6 versus 10.4 T 2.6 days. RCT C = stand ard endot racheal tube Abbreviations: BAL, bronchoalveolar lavage; C, control; CSGS, continuous subglottic suctioning; CXR, chest x-ray; ETT, endotracheal tube; I, int ervention; ICU, intensive care unit; € , euro; NG, nasogastric; PBS, pulmonary bronchial suction; CDC, Centers for Disease Control and Prevention; MHS, major hear surgery; MV, m ec ha ni ca l ve nt ila to r; R C T, ra nd om iz ed cl in ic al tr ial; SGS, subglo ttic secr etion suctioning; SS D, subglott ic secret io n drainage; V A P, ventilator-asso ciated pneumo nia; W BC, w hit e bloo d count.
EXCLUDED STUDIES
The study by Girou and colleagues31 was excluded
be-cause patients were randomized to either semirecumbent position and continuous SSD or prone position and standard ETT. Because the influence of head-of-bed elevation could not be separated from the effectiveness of SSD in reducing VAP, this study was excluded from the meta-analysis of pooled results.
Liu and colleagues34utilized SSD as one of a 3-part
‘‘bundle’’ of interventions, which was compared with
standard care that included a standard ETT. Because of the intervention with a bundle rather than a direct causal comparison between an ETT that did and did not use SSD, this study was not included in the meta-analysis of findings.
Pneumatikos and colleagues38 intubated patients
with an ETT capable of SSD (Hi-Lo Evac, Boulder, Colorado; Mallinckrodt, St Louis, Missouri). The con-tinuous infusion of a suspension of polymyxin, tobra-mycin, and amphotericin B. The control group received
TABLE 2 ICU Impact of Subglottic Secretion Drainage on Ventilator-Associated Pneumonia Outcomes
Study
Sample
Size (n) VAP Mortality
Incidence of VAP per 1000 Ventilator Days Days to Onset of VAP Mean Duration of MV Length of ICU Stay Length of Hospital Stay Bo et al29 SSD, 35 8 14.8T 8 Control, 33 15 6.4T 4 Bouza et al30 SSD, 45 12 20 31.5 Control, 40 19 21 51.6 Girou et al31 SSD, 8 5 Control, 10 6 Kollef et al32 SSD, 160 8 6 34.3 5.6T 2.3 1.5T 3.3 3.7T 4.6 11T 11.2 Control, 183 13 8 43.2 2.9T 1.2 1.9T 5.1 3.2T 4.5 12.4T 14.2 Lacherade et al33 SSD, 169 25 80 17.0 10.5T 11.12 Control, 164 42 84 34.0 7.2T 5.30 Liu et al34 SSD, 41 4 Control, 45 9 Lorente et al36 SSD, 140 11 26 7.5 10.5T 11.12 10.5T 15.91 14.1 T 17.91 Control, 140 31 32 19.9 7.2T 5.3 11.0T 15.19 15.5 T 19.93 Mahul et al37 SSD, 70 9 11 16.2T 11 Control, 75 21 16 8.3T 5 Pneumatikos et al38 SSD, 31 5 5 12.4 Control, 30 16 7 36.44 Smulders et al39 SSD, 49 2 9 6.4 7.9T 9.7 11.9T 8.8 32.1T 25.1 Control, 56 10 10 21.3 7.1T 3.4 14.2T 11.1 32.8T 31.6 Valles et al40 SSD, 76 14 30 19.6 12T 7.1 11T 1 19T 4 Control, 77 25 28 39.6 5.9T 2.1 13T 1 22T 2 Yang et al41 SSD, 48 12 32 7.3T 4.2 8.1T 7.5 Control, 43 20 29 5.1T 3.0 8.4T 6.0 Zheng et al42 SSD, 30 9 8 6.5T 1.3 7.9T 2.5 9.3T 2.9 Control, 31 16 12 5.5T 0.6 10.4T 0.9 12.3T 5.7
a continuous infusion of a placebo. Because this study focused on the decontamination of the subglottic area rather than the SSD, this study was excluded.
RESULTS
The summary and appraisal of studies are presented in Table 1. Table 2 contains summary data of the outcomes variables. The pooled analyses across studies are graphi-cally represented in Tables 3 to 8. The center vertical line indicates that the estimated effects are the same for both the interventions and control groups and is often called the line of no difference. Values to the left of the center line favor SSD and those to the right favor the control. The diamond on the lower aspect of the graph near the
horizontal line represents pooled values.57
The Effectiveness of Subglottic Secretion
Aspiration in Reducing VAP Rates
Across the studies, there were 848 cases in the ex-perimental group and 861 in the controlled group. The pooled results examining the effectiveness of SSD in re-ducing the incidence of VAP demonstrated a 52% reduc-tion (risk ratio, 0.52; 95% confidence interval, 0.43-0.64)
in rates (Table 3).58
The Effectiveness of Subglottic Secretion
Aspiration in Reducing the Duration of MV
Both Bouza and colleagues30 and Lacherade and
col-leagues33reported median rather than days’ duration of
mechanical ventilation. Median durations of ventilation rates in the experimental groups were 3 and 8 days, and
TABLE 3 Effectiveness of Subglottic Drainage on Reducing the Incidence of
those for the control group were 7 and 7 days, respec-tively. When pooled, the results across studies that ex-amined the impact of subglottic secretion drainage on
days of mechanical ventilation were#2= 14.73, df = 5
(P G .01), I2
= 66% (Table 6). An assumption of a systematic review is that the effect of the treatment be-ing studied across patients is the same. This can be
examined visually when all studies consistently demon-strate similar findings (favors either treatment or con-trol). Consistency of treatment effect can also be examined by tests of heterogeneity with a low P value, indicating dif-ferences in underlying effects across studies. Thus, the
PG .0001 would indicate the need for caution in
inter-preting the findings. The I2 statistic is an estimate of
TABLE 5 Effectiveness of Subglottic Drainage on Increasing Days of Ventilator-Associated Pneumonia
variability across studies, with an I2 greater than 0.5
indicating large variability. In this study, the I2of 66%
precludes confidence in the pooled analysis of study
find-ings because of high levels of heterogeneity across studies.
When the study by Kollef et al32 was removed, the I2
dropped to 0 with z = 13.29, P G .00001, indicating a
TABLE 6 Effectiveness of Subglottic Drainage on Duration of Mechanical Ventilation
shorter duration of mechanical ventilation by a mean of 1.99 days in the group who received subglottic secretion drainage.
An assumption of a systematic review is that
the effect of the treatment being studied
across patients is the same.
The Effectiveness of Subglottic Secretion
Aspiration in Reducing Mortality
Eight hundred thirteen intervention cases and 828 con-trol cases were pooled to examine the impact on mortality.
Across studies the tests of heterogeneity were#2= 2.76,
df = 8 (P = .95), I2= 0. The overall effect was z = 1.13 (P =
.26), risk ratio = 0.93 (confidence interval, 0.81-1.06), indicating no significant difference in mortality rates be-tween patients who did and did not receive subglottic secretion drainage (Table 4).
The Effectiveness of Subglottic Secretion
Aspiration in Reducing the Length of ICU Stay
Tests for heterogeneity were#2= 29.22; df = 4, P = .0001,
I2= 0.86, demonstrating that significant heterogeneity
across studies would impact the reliability of pooled
analysis. The test for overall effect was z = 4.82 (PG .01)
(Table 7). When the study by Kollef and colleagues32
was removed, the I2= 0 with z = 7.14, PG .00001, which
indicated adequate homogeneity for pooling of study re-sults. The mean length of ICU stay was 2.95 days shorter than that in the group that received subglottic drainage.
The Effectiveness of Subglottic Secretion
Aspiration in Increasing Days to Onset of VAP
When the studies were pooled, analysis demonstrated signif-icantly delayed onset of developing VAP in the SSD group. However, pooled tests of heterogeneity indicated significantvariability across studies.34Consistency of treatment
ef-fect can be examined by tests of heterogeneity with a low P value, indicating differences in underlying effects across
studies. Thus, P G .0001 would indicate differences in
underlying-effect studies and require caution in
interpret-ing the results. The I2statistic is an estimate of variability
across studies with an I2precluding confidence in the pooled
analysis of the findings because of high levels of heterogene-ity across studies. Even though there was a statistically significant difference between the 2 groups, heterogeneity across studies would prohibit confidence in pooled findings. Even though there was a statistically significant difference
between the 2 groups in the study of Kollef and colleagues32
included in the analysis (Table 5), heterogeneity across studies would prohibit confidence in the pooled findings.
The Effectiveness of Subglottic Secretion
Aspiration in Reducing the Length of
Hospital Stay
The tests for heterogeneity were#20.00, df = 1, P = .98,
I2 = 0, indicating the results could be pooled. Test for
overall effect z = 1.08, P = .28 (Table 8). The results of the analysis indicate no significant difference in length of hospital stay between the treatment and control groups.
Benefits and Harms
Although there were reports of airway complications attri-buted to intervention with an ETT that included SSD, compli-cation rates were not routinely reported across studies. Types and rates of complications when intubated with any type of
ETT could be used as an outcome variable in future studies.59
Potential Bias in the Review
As no efforts were made to locate either studies with neg-ative results or unpublished studies, this systematic review
may be at risk of publication bias.60The older studies were
more likely to use intermittent SSD, whereas more recent studies utilized continuous aspiration, which is congruent with the manufacturer’s recommendations. We included studies that utilized both intermittent and continuous SSD. As our expertise in the use of technology increases, recommendations for device use may change over time.
Consideration for Future Studies
Standard practices in the care of the critically ill change rapidly. It was most helpful when investigators included a description of standard-care practices for both groups whether it was the presence of a heeding tube, head-of-bed
elevation, or ETT cuff pressure and monitoring. The CONSORT guidelines provide a helpful framework for standardizing the information collected and reported in
clinical studies.61
Although the ‘‘bundled’’ approach to reducing infec-tions has been nationally embraced, there is still a need for randomized trials that examine the effectiveness of
individual interventions.62By varying a single
interven-tion at a time, causal relainterven-tionships can be established. An ETT with SSD could be compared with ETTs with
an ultrathin cuff membrane.63Investigators can
contrib-ute to the body of critical care knowledge by examin-ing the effectiveness of this device as well as comparexamin-ing ETTs with SSD to silver-coated ETTs while using con-sistent methods of screening in verifying the presence of VAP. The need at this time is to examine which of the specialized tubes is most effective in method com-parison studies rather than to compare specialized ETTs
to standard tubes.64-66In addition, there is the continuing
need for systematic reviews focusing on an economic anal-ysis of the cost and benefits of the various types of spe-cialized ETTs used in the care of the critically ill patient.
DISCUSSION
The goal of this study was to examine the effectiveness of subglottic secretion aspiration in reducing the occur-rence of VAP. The findings of this systematic review demonstrated a 50% reduction in VAP rates when an ETT with SSD was compared with an ETT without SSD. Across studies, subjects in the experimental group did experience a shorter duration of MV by approximately 2 days compared with control subjects.
Subglottic secretion drainage did not have a significant impact on mortality rates. Individuals requiring MV have underlying medical conditions, which may have a greater impact on mortality rates than the more subtle impact of the use of an ETT with the capabilities for SSD. Several investigators have established the positive relationship of VAP and mortality rates, which may well have a more direct causal relationship than the use of an ETT that facilitates removal of subglottic secretions.
Critically ill patients who received subglottic drain-age were in the ICU at an averdrain-age of 3 days less that those who did not. This is likely related to the fact that they were extubated an average of 2 days sooner. Thus, transfer from the ICU was likely linked to extubation. There was no difference in hospital length of stay. Crit-ically ill patients who require MV represent the ‘‘sicker of the sick’’ among critically ill patients. It is likely that this underlying illness has greater impact on length of ICU stay than the type of ETT used is a more subtle cofactor. Thus, a larger sample would be needed for this smaller effect size as the impact is likely more subtle.
CONCLUSION
The results of this systematic review support level 1
recommendation58 for use of ETTs with SSD for
re-ducing the incidence of VAP.
Acknowledgments
The authors acknowledge the American Association of Critical Care Nurses for its support of this project through both an AACN Evidence Based Practice Grant but more importantly for their belief in this project.
References
1. Bonten MJ. Ventilator-associated pneumonia: preventing the inevitable. Healthc Epidemiol. 2011;52(1):115-121.
2. National Nosocomial Infections Surveillance (NNIS). System report, data summary from January 1992 through June 2004, issued October 2004. Am J Infect Control. 2004;32(8):470-485. 3. Kollef M. SMART approaches for reducing nosocomial
infec-tions in the ICU. Chest. 2008;134(2):447-485.
4. Kollef MH, Morrow LE, Baughman RP, et al. Health care-associated pneumonia (HCAP): a critical appraisal to improve identification, management, and outcomes-proceedings of the HCAP summit. Clin Infect Dis. 2008;46:S296-S334. 5. Ventilator-associated pneumonia (VAP) event. Device Assoc
Events. 2011;6:1-4. www.cdc.gov.nhsn/PDRs/pcsManual6/ pscVAPcurrent.pdf. Accessed July 5, 2011.
6. Combes A, Figliolini C, Trouillet J, et al. Factors predicting ventilator-associated pneumonia recurrence. Crit Care Med. 2003;31(4):1102-1107.
7. Chawla R, Chawla R. Epidemiology, etiology, and diagnosis of hospital-acquired pneumonia and ventilator-associated pneumonia in Asian countries. Am J Infect Control. 2008;36(suppl 4):S93-S100. 8. Combes A, Luyt C, Fagon J, Wolff M, Trouillet J, Chastre J. Early predictors for infection recurrence and death in patients with ventilator-associated pneumonia. Crit Care Med. 2007;35(1):146-154. 9. Collard HR, Said S, Matthay MA. Prevention of ventilator-associated pneumonia: an evidence-based systematic review. Ann Intern Med. 2003;138(6):494-501.
10. Dodek P, Keenan S, Cook D, et al. Clinical guidelines. Evidence-based clinical practice guideline for the prevention of ventilator-associated pneumonia. Ann Intern Med. 2004;141(4):305-313. 11. Tablan OC, Anderson LJ, Beser R, Bridges C, Hajjeh R. Guide-lines for preventing health-careYassociated pneumonia, 2003: recommendations of CDC and the Healthcare Infection Con-trol Practices Advisory Committee. MMWR Morb Mortal Wkly Rep. 2004;52(RR-3):1-36.
12. Zanella A, Bellani G, Presenti A. Airway pressure and flow monitoring. Curr Opin Crit Care. 2010;16:255-260. 13. Adair CG, Gorman SP, Feron BM, et al. Implications of
en-dotracheal tube biofilm for ventilator-associated pneumonia. Intensive Care Med. 1999;25(10):1072-1076.
14. Apostolopoulou E, Bakakos P, Katostaras T, Gregorakos L. Incidence and risk factors for ventilator-associated pnuemonia in 4 multidisciplinary intensive care units in Athens, Greece. Respir Care. 2003;48(7):681-688.
15. Brennan MT, Bahrani-Mougeot F, Fox PC, et al. The role of oral microbial colonization in ventilator-associated pneumonia. Oral Surg Oral Med Oral Pathol Oral Radiol Endo. 2004;98(6):665-672. 16. Abbott CA, Dremsa T, Stewart DW, Mark DD, Swift CC.
Adop-tion of a ventilator-associated pneumonia clinical practice guide-line. Worldviews Evid Based Nurs. 2006;3(4):139-152. 17. Set a goal of zero central line and VAP infections: determining
what is really preventable. Hosp Peer Rev. 2009;33(1):4-5. 18. Berenholtz SM, Milanovich S, Faircloth A, et al. Improving
care for the ventilated patient. Joint Comm J Qual Saf. 2004; 39(4):195-204.
19. Bassi GL, Zanella A, Cressoni M, Styliano M, Kobolow T. Following tracheal intubation, mucus flow is reversed in the semi-recumbent position: possible role in the pathogenesis of ventilator-associated pneumonia. Crit Care Med. 2008;36(2):518-525. 20. Bastin AJ, Ryanna KB. Use of selective decontamination of
the digestive tract in United Kingdom intensive care units. Anaesthesia. 2009;64(1):46-49.
21. Bench S. Humidification in the long-term ventilated patient: a systematic review. Intensive Crit Care Nurs. 2003;19(2):75-84. 22. Beraldo CC, Andrade D, Beraldo CC, Andrade DD. Oral hygiene with chlorhexidine in preventing pneumonia associated with mechanical ventilation. J Bras Pneumol. 2008;34(9):707-714. 23. Bucknall T. Review: several interventions prevent ventilator
as-sociated pneumonia in critically ill patients. Evid Based Nurs. 2003;6(4):112.
24. O’Reilly M. Oral care of the critically ill: a review of the literature and guidelines for practice. Aust Crit Care. 2003;16(3):101-110. 25. Palmer LB. Aerosolized antibiotics in critically ill ventilated
patients. Curr Opin Crit Care. 2009;15(5):413-418. 26. Silvestri L, van Saene HK, de la Cal MA, Sarginson RE,
Thomann C, van Saene HKF. Prevention of ventilator-associated pneumonia by selective decontamination of the digestive tract. Eur Respir J. 2008;32(1):241-243.
27. O’Neal PV. Factors Associated With Subglottic Secretion Ac-cumulation and Removal. Richmond, VA: Adult Health, Virginia Commonwealth University; 2000.
28. O’Neal PV. Factors associated with subglottic secretion viscosity and evacuation efficiency. Biol Res Nurs. 2007;8(3):202-209. 29. Bo H, He L, Qu J. Influence of the subglottic secretion
drain-age on the morbidity ventilator associated pneumonia in me-chanically ventilated patients. Chung-Hua Chieh Ho Ho Hu His Tsa Chih Chinese (J Tuberc Respir Dis). 2006;29(1):19-22. 30. Bouza E, Hortal J, Munoz P, et al. Postoperative infections after major heart surgery and prevention of ventilator-associated pneu-monia: a one-day European prevalence study (ESGNI-008). J Hosp Infect. 2006;64(3):224-230.
31. Girou E, Buu-Hoi A, Stephan F, et al. Airway colonization in long-term mechanically ventilated patients. Effect of semi-recumbent position and continuous subglottic suctioning. Inten-sive Care Med. 2004;30(2):225-233.
32. Kollef MH, Skubas NJ, Sundt TM. A randomized clinical trial of continuous aspiration of subglottic secretions in cardiac sur-gery patients. Chest. 1999;116:1339-1346.
33. Lacherade JC, De Jonghe B, Guezennec P, et al. Intermittent sub-glottic secretion drainage and ventilator-associated pneumonia: a multicenter trial. Am J Respir Crit Care Med. 2010;182(7):910-917. 34. Liu SH, Yan XX, Cao SQ, An S, Zhang L. The effect of sub-glottic secretion drainage on prevention of ventilator associ-ated lower airway infection. Chung-Hua Chiek Ho Ho Hu Hsi Tsa Chih Chinese (J Tuberc Respir Dis). 2006;29(1):19-22. 35. Bergmans DC, Bonten JM, Gaillard CA, et al. Prevention of
ventilator-associated pneumonia by oral decontamination: a prospective, randomized, double-blind, placebo-controlled study. Am J Respir Crit Care Med. 2001;164(3):338-339.
36. Lorente L, Lecuona M, Jimenez A, Mora ML, Sierr A. In-fluence of an endotracheal tube with polyurethane cuff and subglottic secretion drainage on pneumonia. Am J Respir Crit Care Med. 2007;176(11):1079-1083.
37. Mahul P, Auboyer C, Jospe R, et al. Prevention to nosocomial pneumonia in intubated patients: respective role of mechani-cal subglottic secretions drainage and stress ulcer prophylaxis. Intensive Care Med. 1992;18:20-25.
38. Pneumatikos I, Kloulouras V, Nathanail C, Goe D, Nakos G. Selective decontamination of subglottic area in mechanically ventilated patients with multiple trauma. Intensive Care Med. 2002;28(4):432-437.
39. Smulders K, van der Hoeven H, Weers-Pothoff I, Vanderbroucke-Grauls C. A randomized clinical trial of intermittent subglottic secretion drainage in patients receiving mechanical ventilation. Chest. 2001;121(3):858-862.
40. Valles J, Artigas A, Rello J, et al. Continuous aspiration of subglottic secretions in preventing ventilator-associated pneu-monia. Ann Intern Med. 1995;122(3):179-186.
41. Yang CS, Qiu HB, Zhu YP. Effect of continuous aspiration of subglottic secretions on the prevention of ventilator-associated pneumonia in mechanically ventilated patients. A prospective, randomized controlled trial. Chung-Hua Chieh Ho Ho Hy Has Tsa Chih Chinese (J Tuberc Respir Dis). 2008;47:625-629. 42. Zheng RQ, Lin H, Shao J. A clinical study of subglottic secre-tion drainage for prevensecre-tion of ventilasecre-tion associated pneu-monia. Chin Crit Care Med. 2008;20:338-340.
43. Hurley JC. Profound effect of study design factors on ventilator-associated pneumonia incidence of prevention stud-ies: benchmarking the literature experience. J Antimicrob Chemother. 2008;61:1154-1161.
44. Manojlovich M, Antonakos CL, Ronis DL. The relationship between hospital size and ICU type on select adverse patient outcomes. Hosp Top. 2010;88(2):33-42.
45. Lunario RA. The Relationship Between Frequent Suctioning and the Risk of VAP. Buffalo, NY: D’Youville College; 2004. 46. Petring OU, Adelhoj B, Jensen BN, Pederson NO, Lomholt M.
Prevention of silent aspiration due to leaks around cuffs of endotracheal tubes. Anesth Analg. 1986;65:777-780. 47. Terrer M, Torres MA. Maintenance of tracheal tube cuff
pres-sure: where are the limits? Crit Care. 2008;12:106.
48. Lucangelo U, Zin WA, Antonaglia V, et al. Effect of positive expiratory pressure and type of tracheal cuff on the incidence of aspiration in mechanically ventilated patients in an intensive care unit. Crit Care Med. 2008;36(2):409-413.
49. Koenig SM, Truwit JD. Ventilator-associated pneumonia: di-agnosis, treatment, and prevention. Clin Microbiol Rev. 2006; 19(4):637-657.
50. Niederman MS. Hospital-acquired pneumonia, health care-associated pneumonia, ventilator-care-associated pneumonia, and ventilator-associated tracheobronchitis: definitions and chal-lenges in trial design. Clin Infect Dis. 2010;51(S1):S12-S17. 51. Lisboa T, Rello J. Diagnosis of ventilator-associated
pneumo-nia: is there a gold stand and a simple approach? Curr Opin Infect Dis. 2008;21:174-178.
52. Koulenti D, Lisboa T, Brun-Buisson C, et al. Spectrum of prac-tice in the diagnosis of nosocomial pneumonia in patients re-quiring mechanical ventilation in European intensive care units. Crit Care Med. 2009;37(8):2360-2368.
53. Berton DC, Kalil AC, Cavalicanti M, Teixeira PJZ. Quantita-tive versus qualitaQuantita-tive cultures of respiratory secretions for clin-ical outcomes in patients with ventilator associated pneumonia. Cochrane Database Syst Rev. 2008;4:CD006482.
54. Bergmans DC, Bonten M, Leeuw PW, Stobberingh EE. Reproduc-ibility of quantitative culture of endotracheal aspirates from mech-anically ventilated patients. J Clin Microbiol. 1997;35(3):796-798. 55. Chastre J, Trouillet J, Combes A, Luyt C. Diagnostic
tech-niques and procedures for establishing the microbial etiology of ventilator-associated pneumonia for clinical trials: the pros for quantitative culture. Clin Infect Dis. 2010;2010(suppl):S88-S92. 56. Afessa B, Hubmayr RD, Vetter EA, et al. Bronchoscopy in
ventilator-associated pneumonia: agreement of calibrated loop and serial dilution. Am J Respir Crit Care Med. 2006;173(11): 1229-1232.
57. Montoni V, Ioannides J, Cook D, Guyatt CG. Fixed-effects and random-effects models. In: Guyatt GH, Rennie D, Meade MO, Cook D, eds. Users’ Guide to the Medical Literature: A Manual for Evidence-Based Clinical Practice. 2nd ed. New York, NY: Mc-Graw Hill; 2008.
58. Guyatt G, Rennie D, Meade MO, Cook DJ. Users’ Guide to the Medical Literature: A Manual for Evidence Based Clinical Practice. 2nd ed. New York, NY: McGraw-Hill Companies, Inc; 2008. 59. Harvey RC, Miller PR, Lee JA, Bowton DL, MacGregor DA.
Potential mucosal injury related to continuous aspiration of subglottic secretion device. Anesthesiol Clin North Am. 2007; 107:666-669.
60. Montori VM, Ioannides J, Guyatt GH. Chapter 20.1. Reporting bias. In: Guyatt GH, Rennie D, Meade MO, Cook DJ, eds. Users’ Guide to the Medical Literature: A Manual for Evidence-Based Clinical Practice. 2nd ed. New York, NY: McGraw-Hill; 2008. 61. CONSORT statement, 2010. www.consort.org. Accessed
November 19, 2011.
62. Isakow W, Kollef MH. Preventing ventilator-associated pneu-monia: an evidence-based approach of modifiable risk factors. Semin Respir Crit Care Med. 2006;27(1):5-17.
63. Lorente L, Blot S, Rello J. New issues and controversies in the prevention of ventilator-associated pneumonia. Am J Respir Crit Care Med. 2010;182(7):870-876.
64. Safdar N, Dezfulian C, Collard HR, Saint S. Clinical and eco-nomic consequences of ventilator-associated pneumonia: a sys-tematic review. Crit Care Med. 2005;33(10):2184-2193. 65. Speroni KG, Lucas J, Dugan L, et al. Comparative effectiveness
of standard endotracheal tubes vs endotracheal tubes with con-tinuous subglottic suctioning on ventilator-associated pneumo-nia rates. Nurs Econ. 2011;29(1):15-37.
66. Hallis C, Merle V, Guitard P, et al. Is continuous subglottic suctioning cost-effective for the prevention of ventilator-associated pneumonia? Infect Control Hosp Epidemiol. 2011;32(2):131-135. 67. Montori VM, Hatala F, Ionnides J, Meade MO, Wyer PC, Guyatt GH. Making sense of variability in study results.
In: Guyatt GH, Rennie D, Meade MO, Cook DJ, eds. Users’ Guide to the Medical Literature: A Manual for Evidence-Based Clinical Practice. 2nd ed. New York, NY: McGraw-Hill; 2008.
ABOUT THE AUTHORS
A. Renee Leasure, PhD, RN, CCRN, CNS, is an associate professor at the University of Oklahoma Health Sciences Center and research investigator at the Department of Veterans Affair Medical Center, Oklahoma City. Dr Leasure is a reviewer of manuscripts for DCCN. Joan Stirlen, MPH, BSN, RN, is a performance measures coordinator at the Office of Performance and Quality, Department of Veterans Affairs Medical Center, Oklahoma City.
Shu Hua Lu, PhD, RN, is an assistant professor at School of Nursing, China Medical University, Taichung, Taiwan.
The authors have disclosed that they have no significant relationship with, or financial interest in, any commercial companies pertaining to this article. Address correspondence and reprint requests to: A. Renee Leasure, PhD, RN, CCRN, CNS, CNB 316, 1100 N Stonewall, Oklahoma City, OK 73117 (Reneeleasure@yahoo.com).
