Influence of an Endotracheal Tube with Polyurethane Cuff and Subglottic Secretion Drainage on Pneumonia

Transcription

1 Influence of an Endotracheal Tube with Polyurethane Cuff and Subglottic Secretion Drainage on Pneumonia Leonardo Lorente 1, María Lecuona 2, Alejandro Jiménez 3, María L. Mora 4, and Antonio Sierra 5 1 Department of Critical Care, 2 Department of Microbiology, 3 Research Unit, 4 Department of Critical Care, and 5 Department of Microbiology, Hospital Universitario de Canarias, La Laguna, Santa Cruz de Tenerife, Spain Rationale: Subglottic secretion drainage (SSD) appears to be effective in preventing ventilator-associated pneumonia (VAP), primarily by reducing early-onset pneumonia; but it may not prevent lateonset pneumonia. We tested the hypothesis using an endotracheal tube incorporating an ultrathin polyurethane cuff (which reduces channel formation and fluid leakage from the subglottic area), in addition to an SSD lumen, which would reduce the incidence of lateonset Objectives: To compare the incidence of VAP, using an endotracheal tube with polyurethane cuff and subglottic secretion drainage (ETT- PUC-SSD) versus a conventional endotracheal tube (ETT-C) with polyvinyl cuff, without subglottic secretion drainage. Methods: Clinical randomized trial in a 24-bed medical surgical intensive care unit. Patients expected to require mechanical ventilation for more than 24 hours were randomly assigned to one of two groups: one was ventilated with ETT-PUC-SSD and the other with ETT-C. Measurements and Main Results: Tracheal aspirate samples were obtained during endotracheal intubation, then twice per week and finally on extubation. VAP was found in 31 of 140 (22.1%) patients in the ETT-C group and in 11 of 140 (7.9%) in the ETT-PUC-SSD group (P ). Cox regression analysis showed ETT-C as a risk factor for global VAP (hazard ratio [HR], 3.3; 95% confidence interval [CI], ; P ), early-onset VAP (HR, 3.3; 95% CI, ; P ), and late-onset VAP (HR, 3.5; 95% CI, ; P ). Conclusions: The use of an endotracheal tube with polyurethane cuff and subglottic secretion drainage helps prevent early- and late-onset Clinical trial registered with (NCT ). Keywords: ventilator-associated pneumonia; endotracheal tube; polyurethane cuff; polyvinyl cuff; subglottic secretion drainage Ventilator-associated pneumonia (VAP) continues to be an important cause of morbidity and mortality in critically ill patients (1). Subglottic secretions accumulated above the endotracheal cuff may progress, descending along the channels within the folds of the cuff wall, to the lower respiratory tract, causing VAP (2 6). A preventive strategy to avoid the progression of subglottic secretions into the lower respiratory tract is to remove these secretions by subglottic secretion drainage (SSD), which is accomplished through a separate dorsal lumen that opens directly above the endotracheal tube cuff. SSD has reduced the incidence of VAP in some studies (7 10); but in other studies, it has not decreased the incidence of VAP (11, 12) or airway colonization (13). Several guidelines for the prevention of VAP AT A GLANCE COMMENTARY Scientific Knowledge on the Subject Subglottic secretion drainage appears to be effective in preventing ventilator-associated pneumonia (VAP), primarily by reducing early-onset pneumonia; but it may not prevent late-onset pneumonia. What This Study Adds to the Field An endotracheal tube with polyurethane cuff and subglottic secretion drainage is effective in preventing early-onset and late-onset have recommended the aspiration of subglottic secretions because it can reduce the risk of early-onset VAP (14 17). A metaanalysis published by Dezfulian and coworkers (18) concluded that SSD appears to be effective in preventing VAP in patients expected to require more than 72 hours of mechanical ventilation, primarily by reducing early-onset pneumonia; however, it may not prevent late-onset pneumonia. Thus, it seems reasonable to use an endotracheal tube with SSD in such patients with the purpose of reducing early-onset Another preventive strategy to avoid the progression of subglottic secretions into the lower respiratory tract is to prevent channel formation within the folds of the endotracheal cuff. High-volume and low-pressure (HVLP) endotracheal tubes with a polyurethane cuff (PUC) have been introduced, with an ultrathin cuff membrane (thickness, 7 mm) compared with the cuff membrane of conventional HVLP endotracheal tubes (thickness, >50 mm), designed to prevent the formation of folds within the cuff and thus to prevent fluid and air leakage (19, 20); nevertheless, there are no data on the prevention We tested the hypothesis that the use of an endotracheal tube incorporating an ultrathin polyurethane cuff (which reduces channel formation and fluid leakage from the subglottic area), in addition to an SSD lumen, would reduce the incidence of lateonset The study was designed to compare the incidence of VAP, using a new commercially available endotracheal tube incorporating two potential strategies to prevent VAP, an ultrathin polyurethane cuff and subglottic secretion drainage (ETT-PUC- SSD), versus a conventional endotracheal tube (ETT-C) with polyvinyl cuff and without SSD. (Received in original form May 23, 2007; accepted in final form September 17, 2007) Correspondence and requests for reprints should be addressed to Leonardo Lorente, M.D., Ph.D., Department of Critical Care, Hospital Universitario de Canarias, Ofra s/n, La Cuesta, La Laguna 38320, Santa Cruz de Tenerife, Spain. lorentemartin@msn.com Am J Respir Crit Care Med Vol 176. pp , 2007 Originally Published in Press as DOI: /rccm OC on October 18, 2007 Internet address: METHODS Design of the Study A randomized clinical trial was performed at the 24-bed medical surgical intensive care unit (ICU) of the Hospital Universitario de Canarias (Tenerife, Spain), a 650-bed tertiary hospital, from March 1, 2006, to October 31, The study was approved by the institutional

2 1080 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL review board and informed consent from the patients or from their legal guardians was obtained. Only patients expected to require mechanical ventilation for more than 24 hours, according to the judgment of attending physicians, were included. Exclusion criteria were as follows: age less than 18 years, pregnancy, infection with human immunodeficiency virus, blood leukocyte count less than 1,000 cells/mm 3, solid or hematological tumor, and/or immunosuppressive therapy. Intervention Patients were randomly assigned to either of two groups at the time of intubation, using a number list generated with Excel software (Microsoft, Redmond, WA): one group was intubated with ETT-C and the other with ETT-PUC-SSD. The ETT-PUC-SSD patients were intubated with a SealGuard Evac endotracheal tube (Mallinckrodt Medical, Athlone, Ireland), which incorporates a dorsal separate lumen ending in the subglottic area, above the polyurethane cuff, for subglottic secretion drainage. This was performed intermittently during 1-hour periods with a 10-ml syringe. Patients who had undergone tracheostomy received a SealGuard Evac tracheostomy tube (Mallinckrodt Medical), which also incorporates a separate dorsal lumen ending above the cuff in the subglottic area for SSD and a polyurethane cuff. The ETT-C patients were intubated with a Hi-Lo endotracheal tube (Mallinckrodt Medical), which does not incorporate a dorsal separate lumen ending in the subglottic area, and has a polyvinyl cuff. Measures to Prevent VAP In both groups, identical measures for the prevention of VAP were established: no routine change of ventilator circuits, tracheal suction by an open system when necessary, periodic verification every 4 hours of semirecumbent body position to maintain an angle of 408, periodic verification every 4 hours of intracuff pressure to maintain a pressure of 25 cm H 2 O, nasogastric tube, continuous enteric nutrition, periodic verification of the residual gastric volume every 6 hours, prophylactic ranitidine for stress ulcers, oral cleaning with chlorhexidine every 8 hours, protocol of sedation, protocol for mechanical ventilation weaning, and no selective digestive decontamination. Vigilance against Microbiological Threats Tracheal aspirate samples were obtained during endotracheal intubation, then twice per week and finally on extubation. Definitions The diagnosis of pneumonia was established when all of the following criteria were met: (1) new onset of purulent bronchial sputum, (2) body temperature.388cor,35.58c, (3) white blood cell count.10,000/mm 3 or,4,000/mm 3, (4) chest radiograph showing new or progressive infiltrates, and (5) significant quantitative culture of respiratory secretions by tracheal aspirate (.10 6 cfu/ml). Pneumonia was considered to be VAP when it was diagnosed during mechanical ventilation and was not present at the time mechanical ventilation was established. VAP was considered early onset when it was diagnosed during the first 4 days of mechanical ventilation. VAP was considered late onset when it was diagnosed after 4 days of mechanical ventilation. All randomized patients were considered at risk of early-onset Patients were considered at risk of late-onset VAP when the duration of mechanical ventilation exceeded 5 days. The diagnosis of VAP was made by an expert panel blinded to treatment assignment. Antibiotic Strategy We used the following antibiotic strategy: for antimicrobial prophylaxis after cardiothoracic surgery, neurosurgery, and orthopedic traumatic surgery we used cefazolin, and vancomycin when there was allergy to b-lactam antibiotics. For community-acquired respiratory infection treatment, we used either cefotaxime, ceftriaxone plus levofloxacin, or azithromycin. For empiric antibiotic treatment of early-onset VAP, we used a monotherapy with amoxicillin clavulanic acid or a secondgeneration cephalosporin. For empiric antibiotic treatment of late-onset VAP, we used a combination therapy with an antipseudomonal cephalosporin or antipseudomonal carbepenem or piperacillin tazobactam plus aminoglycoside or fluoroquinolone. Vancomycin was added when an infection for methicillin-resistant Staphylococcus was suspected. Variables Recorded The following variables were recorded for each patient: sex, age, diagnosis group, Acute Physiology and Chronic Health Evaluation (APACHE) II score, duration of mechanical ventilation, antibiotics before VAP onset, use of paralytic agents, tracheotomy, reintubation, enteral nutrition, intracuff pressure, number of tracheal suctioning procedures per day, and mortality. Statistical Analysis We have found that the proportion of patients who developed VAP, receiving more than 24 hours of mechanical ventilation, was 25% when using ETT-C. We expected a 50% reduction of VAP rates with the ETT-PUC-SSD, based on the results of the meta-analysis published by Dezfulian and coworkers (18). For a power of 80% and a 5% type I error rate, we needed 110 patients per group to test the proportion of patients needed to reduce this proportion (25% using ETT-C) to 12% using ETT-PUC-SSD. We assumed a dropout rate of 20% (patients undergoing less than 24 h of mechanical ventilation) per group. With this condition, we needed to include 140 patients per group. Quantitative variables are reported as means 6 standard deviation, and were compared by Student t test. Qualitative variables are reported as percentages, and were compared by chi-square test, or by Fisher s exact test, as appropriate. The proportion of VAP between groups was compared by Kruskal- Wallis test for a single order of classification. The probability of remaining VAP free was determined by the Kaplan-Meier method and comparison between the two groups was performed by log-rank test. The incidence density of VAP (number of events per days of mechanical ventilation) between groups were compared by Poisson regression analysis. Three Cox proportional hazard models were constructed, with the following dependent variables: (1) VAP-free time, (2) early-onset VAPfree time, (3) late-onset VAP-free time. The main independent variable in the three models was the type of endotracheal tube used (ETT-C vs. ETT-PUC-SSD). Cox regression model with global VAP-free time was performed controlling for age, sex, APACHE II score, use of antibiotics before VAP, use of paralytic agents, reintubation, tracheotomy, enteral nutrition, intracuff pressure and number of tracheal suctioning procedures per day using partial models (only one covariable was introduced along with the main independent variable each time). With respect to the other two models, early-onset and late-onset, control variables were not included because only 5 and 6 events in the ETT- PUC-SSD were observed, respectively. A P value less than 0.05 was considered statistically significant. For statistical analyses, we used SPSS version (SPSS, Inc., Chicago, IL) and StatXact version (Cytel Software, Cambridge, MA). RESULTS There were no significant differences between the two groups of patients (140 with ETT-C and 140 with ETT-PUC-SSD) in terms of age, sex, diagnosis groups, APACHE II score, pre-vap use of antibiotics, paralytic agents, reintubation, tracheotomy, enteral nutrition, intracuff pressure, and number of tracheal suctioning procedures per day (Table 1). VAP was found in 31 of 140 (22.1%) patients in the ETT-C group and in 11 of 140 (7.9%) patients in the ETT-PUC-SSD group (P ). We found 31 occurrences of VAP in 1,558 days of mechanical ventilation in the ETT-C group and 11 occurrences of VAP in 1,463 days of mechanical ventilation in the ETT-PUC-SSD group. Poisson regression analysis showed a higher incidence density of VAP in the ETT-C group than in the ETT-PUC-SSD group (19.9 vs. 7.5 VAP/1,000 d of mechanical ventilation; odds ratio, 2.65; 95% confidence interval [CI],

3 Lorente, Lecuona, Jiménez, et al.: Polyurethane Cuff and Subglottic Secretion Drainage 1081 TABLE 1. CHARACTERISTICS OF ETT-C AND ETT-PUC-SSD GROUPS ETT-C (n 5 140) ETT-PUC-SSD (n 5 140) P Value Sex (female), n (%) 40 (28.6) 36 (25.7) 0.69 Age, yr, mean 6 SD Diagnostic group, n (%) 0.98 Cardiac surgery 53 (37.9) 55 (39.3) Cardiology 10 (7.1) 7 (5.0) Respiratory 22 (15.7) 22 (15.7) Digestive 19 (13.6) 16 (11.4) Neurologic 16 (11.4) 18 (12.9) Trauma 15 (10.7) 17 (12.1) Intoxication 5 (3.6) 5 (3.6) APACHE II score, mean 6 SD Antibiotics before VAP, n (%) 119 (85.0) 117 (83.6) 0.87 Paralytic agents, n (%) 6 (4.3) 7 (5.0) 0.99 Reintubation, n (%) 8 (5.7) 9 (6.4) 0.99 Tracheotomy, n (%) 13 (9.3) 12 (8.6) 0.99 Enteral nutrition, n (%) 68 (48.6) 75 (53.6) 0.47 Intracuff pressure, cm H 2 O, mean 6 SD Tracheal suction, mean number of suctioning procedures per day, mean 6 SD Duration of MV, d, mean 6 SD ICU stay, d, mean 6 SD ICU mortality, n (%) 32 (22.9) 26 (18.6) 0.46 Definition of abbreviations: APACHE 5 Acute Physiology and Chronic Health Evaluation; ETT-C 5 conventional endotracheal tube; ETT-PUC-SSD 5 endotracheal tube with polyurethane cuff and subglottic secretion drainage; ICU 5 intensive care unit; MV 5 mechanical ventilation; VAP 5 ventilator-associated pneumonia infinite; P ). Kaplan-Meier analysis confirmed a significantly lower incidence of VAP in the ETT-PUC-SSD group than in the ETT-C group (log-rank test, 13.25; P, 0.001) (Figure 1). No differences were found concerning the time of diagnosis of VAP between the ETT-C and ETT-PUC-SSD groups ( vs d; P ). However, when the time of diagnosis of VAP was classified into two categories, early-onset or late-onset VAP, Cox regression analysis showed ETT-C to be a risk factor for both categories (Table 2). Cox regression analysis showed ETT-C as a risk factor for global VAP (hazard ratio [HR], 3.3; 95% CI, ; P ), early-onset VAP (HR, 3.3; 95% CI, ; P ) and late-onset VAP (HR, 3.5; 95% CI, ; P ) (Table 2). Figure 1. Cumulative proportion of patients remaining free of ventilator-associated pneumonia, using an endotracheal tube with polyurethane cuff and subglottic secretion drainage (ETT-PUC-SSD) versus a conventional endotracheal tube (ETT-C). Log-rank test, 13.25; P, Table 3 shows the microorganisms responsible for VAP and classified according to the onset of DISCUSSION In this study, we found a reduction in the incidence of early- and late-onset VAP in patients treated with an endotracheal tube with an ultrathin polyurethane cuff and a lumen for subglottic secretion drainage as compared with a conventional endotracheal tube with polyvinyl cuff and without subglottic secretion drainage. To our knowledge, this is the first study to report data about ETT-PUC-SSD and the incidence of In relation to the first potential advantage of the ETT-PUC- SSD, as well as the polyurethane cuff, it is necessary to remember that initial low-volume, high-pressure (LVHP) cuffs required a pressure of more than 60 cm H 2 O to achieve a clinical seal, and that a pressure higher than 50 cm H 2 O may ultimately stop capillary mucosal blood flow; thus, they frequently induced tracheal injury after prolonged use (21). It is important to distinguish between the pressure inside the cuff and the pressure applied to the tracheal wall; the tracheal wall pressure at a given cuff volume has been calculated as the difference between intracuff pressures at that volume when the cuff is inflated inside the trachea and when it is inflated while suspended freely in air (22). In the 1970s, high-volume, low-pressure (HVLP) cuffs, which can achieve clinical seals at pressures below 30 cm H 2 O, were designed and introduced in an attempt to reduce the incidence of tracheal mucosal damage due to ischemic mucosa produced by the LVHP cuffs. When fully inflated, these HVLP cuffs have diameters times the diameter of the average adult trachea (2). When HVLP cuffs are inflated in a trachea to achieve a clinical seal, the excess material folds over itself, developing channels. Subglottic secretions accumulated above the endotracheal cuff may descend along the channels within the folds of the cuff wall to the lower respiratory tract. This progression is easier with HVLP than with LVHP cuffs (2 6) and, accordingly, the risk of VAP increases. In addition, in one study with anesthetized patients, increasing the pressure in the HVLP cuff beyond

4 1082 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL TABLE 2. COMPARISON OF VAP PROPORTIONS AND HAZARD RATIOS BETWEEN GROUPS ETT-C (n 5 140) n(%) ETT-PUC-SSD (n 5 140) n(%) ETT-C vs. ETT-PUC-SSD HR (95% CI) P Value* VAP 31 of 140 (22.1%) 11 of 140 (7.9%) 3.3 ( ) Early-onset VAP 15 of 140 (10.7%) 5 of 140 (3.6%) 3.3 ( ) 0.02 Late-onset VAP 16 of 60 (26.7%) 6 of 63 (9.5%) 3.5 ( ) 0.01 Definition of abbreviations: CI5 confidence interval; HR 5 hazard ratio; ETT-C 5 conventional endotracheal tube; ETT-PUC- SSD 5 endotracheal tube with polyurethane cuff and subglottic secretion drainage; VAP 5 ventilator-associated pneumonia. * P values are from the Cox regression model. Patients at risk: All the patients randomized. Patients at risk: Patients with duration of mechanical ventilation longer than 5 days. clinical seal, to 50 cm H 2 O, did not prevent aspiration, possibly because of not being able to avoid the formation of dye-filled cuff folds (2). In some laboratory studies, it was found that while using HVLP cuffs, a rapid leakage of fluid occurred from above the cuff into the trachea, unless the tracheal pressure was greater than the height of fluid in the column above the cuff (5, 6). In the study by Blunt and coworkers (23), it was found that lubrication of HVLP cuffs with a water-soluble gel reduced dye leakage in anesthetized patients; however, this effect was temporary, because in critically ill patients the leakage of dye occurred after an average period of 48 hours (range, h). Thus, fluid leakage past the tracheal tube remained an unresolved problem with HVLP cuffs. However, new HVLP ultrathin polyurethane cuffs have shown lower fluid leakage in vitro (19), and lower air leakage in vivo (20), than conventional HVLP cuffs of polyvinyl. Dullenkopf and coworkers (19) compared in vitro fluid leakage past the tube cuff, using conventional HVLP endotracheal tubes of polyvinyl from various manufacturers (cuff membrane thickness, >50 mm) versus HVLP endotracheal tubes with ultrathin polyurethane cuff (cuff membrane thickness, 7 mm). A vertical polyvinylchloride trachea model with an internal diameter of 20 mm was intubated, and cuffs were inflated from 10 to 60 cm H 2 O. Colored water (5 ml) was added to the top of the cuff. Fluid leakage past tube cuffs occurred within 5 minutes in all conventional endotracheal tubes at cuff pressures up to 60 cm H 2 O. In the polyurethane cuff, fluid leakage was not observed at cuff pressures of 20 cm H 2 O. In addition, in the study by TABLE 3. MICROORGANISMS RESPONSIBLE FOR VENTILATOR- ASSOCIATED PNEUMONIA AND CLASSIFIED ACCORDING TO ONSET OF VENTILATOR-ASSOCIATED PNEUMONIA Microorganisms ETT-C ETT-PUC-SSD Total GPC 10 (8 EO and 2 LO) 2 (1 EO and 1 LO) MSSA 6 (5 EO and 1 LO) 1 (1 EO) MRSA 2 (1 EO and 1 LO) 1 (1 LO) Streptococcus pneumoniae 2(2EO) 0 Total GNB 21 (7 EO and 14 LO) 9 (4 EO and 5 LO) Pseudomonas aeruginosa 4 (1 EO and 3 LO) 4 (2 EO and 2 LO) Stenotrophomonas maltophilia 2(2LO) 0 Acinetobacter spp. 3 (3 LO) 1 (1 LO) Escherichia coli 5 (2 EO and 3 LO) 0 Klebsiella spp. 2 (1 EO and 1 LO) 0 Enterobacter spp. 2 (1 EO and 1 LO) 2 (1 EO and 1 LO) Serratia marcescens 0 1 (1 LO) Proteus mirabilis 1(1EO) 0 Citrobacter koseri 1(1LO) 0 Haemophilus influenzae 1(1EO) 1(1EO) Total 31 (15 EO and 16 LO) 11 (5 EO and 6 LO) Definition of abbreviations:eo5 early-onset pneumonia; ETT-C 5 conventional endotracheal tube; ETT-PUC-SSD 5 endotracheal tube with polyurethane cuff and subglottic secretion drainage; GNB 5 gram-negative bacilli; GPC 5 grampositive cocci; LO 5 late-onset pneumonia; MRSA 5 methicillin-resistant Staphylococcus aureus; MSSA 5 methicillin-sensitive Staphylococcus aureus. Dullenkopf and coworkers (19) computed tomography was performed after bathing the cuffs in contrast medium and inserting them into the polyvinylchloride trachea model at cuff pressures of 20 cm H 2 O; all conventional HVLP endotracheal tubes showed additional contrast enhancement within the cuff area due to folds. Another study by Dullenkopf and coworkers (20) compared the cuff pressures required to prevent air leakage, using a conventional HVLP endotracheal tube of polyvinyl from various manufacturers versus an HVLP endotracheal tube with an ultrathin polyurethane cuff. Fifty patients were randomly assigned to receive endotracheal intubation with a conventional HVLP endotracheal tube or an HVLP endotracheal tube with a polyurethane cuff. Cuff pressure to prevent air leakage at standardized ventilator settings (peak inspiratory pressure, 20 cm H 2 O; positive end-expiratory pressure, 5 cm H 2 O; respiratory rate, 15 breaths/min) was assessed by auscultation of audible sounds at the mouth. The HVLP endotracheal tube with a polyurethane cuff required significantly lower sealing pressures (9.5 [8 12] cm H 2 O) compared with the other brands of HVLP endotracheal tube (19.1 [8 42] cm H 2 O). In relation to the second potential advantage of the newer endotracheal tube and the possibility of subglottic secretion drainage, it is necessary to remember that SSD has reduced the incidence of VAP in some studies (7 10); but in other studies it has not decreased the incidence of VAP (11, 12) or airway colonization (13). In a meta-analysis published by Dezfulian and coworkers in 2005 (18), which evaluated 896 patients from five studies (8 12), SSD appears effective in preventing VAP (relative risk, 0.51; 95% CI, ) in patients expected to require more than 72 hours of mechanical ventilation. We performed subglottic drainage by intermittent aspiration because continuous subglottic drainage has been found to be injurious to the tracheal mucosa in some studies (13, 24). In the study by Berra and coworkers (24), 22 intubated sheep randomly received an endotracheal tube with or without continuous subglottic secretion drainage, and after 72 hours of mechanical ventilation were slaughtered and underwent autopsy. None of the sheep intubated with only a conventional endotracheal tube showed gross or microscopic findings in the trachea. All those sheep intubated with continuous subglottic secretion drainage showed gross and microscopic findings in the trachea, including 21% of mucosal necrosis and exposed cartilage, although the authors only found blood in the aspiration catheter in one case during the first minutes. In addition, in a study by Girou and coworkers (13), it was found that 40% of the patients with continuous subglottic drainage developed laryngeal edema immediately after extubation. It is possible that intermittent subglottic secretion drainage is less injurious to the tracheal mucosa than continuous subglottic secretion drainage, although this hypothesis has not been studied. However, intermittent subglottic drainage also seems to be less effective than continuous subglottic drainage in preventing

5 Lorente, Lecuona, Jiménez, et al.: Polyurethane Cuff and Subglottic Secretion Drainage 1083 the leakage of oropharyngeal secretion and thus, the risk of VAP, although this hypothesis has not been studied either. The main contribution of our study is the finding that ETT- PUC-SSD, besides preventing early-onset VAP, also prevents late-onset The meta-analysis by Dezfulian and coworkers (18) concluded that SSD appears primarily to reduce early-onset VAP; however, it is not clear why it may not prevent late-onset Presumably, some of the VAP cases (perhaps those of later onset) that still occurred in that study were due to secretions slipping past the ETT cuff and the dorsal suction lumen. This tube, in theory, should prevent leakage long enough to permit it to pool around the dorsal lumen and be aspirated. It is possible that, while using a polyvinyl cuff during SSD, there is going to be some subglottic secretion leakage within the folds of the cuff for the following reasons: first, when SSD is intermittent, there could be subglottic secretion leakage during the periods without suctioning; second, when SSD is continuous the suctioning port may sometimes be occluded because of suctioning of the tracheal mucosa, sometimes causing SSD failure and, subsequently, subglottic secretion leakage. The use of an ultrathin polyurethane cuff (which reduces channel formation into the cuff) during SSD may minimize this subglottic secretion leakage within the folds of the cuff and thus reduce the risk of Our study has some limitations. First, we did not perform an assessment of fluid leakage (we assumed the reliability of the data reported by the manufacturer). Another limitation is that the patients were only assigned to two groups: ETT-C and ETT- PUC-SSD; thus, we were not able to discriminate the independent influence of SSD and polyurethane cuff in the incidence of Another limitation is that we did not compare injury to the tracheal mucosa in the two groups. Another point is that the study was performed within a single ICU, and the results may therefore not be applicable to other ICUs. A further limitation is that the VAP diagnostic procedure was not invasive and we used only tracheal aspirate samples; however, a randomized clinical trial found no significant differences in clinical outcomes and use of antibiotics when using a diagnostic strategy for ventilatorassociated pneumonia based on quantitative culture of bronchoalveolar lavage fluid and nonquantitative culture of endotracheal aspirate (25). Another limitation is the blinding process; because ETT-C and ETT-PUC-SSD are visually different (ETT- PUC-SSD has a different cuff and a separate dorsal lumen) the study could not be blinded for the attending physicians; however, the kind of ETT was blinded for the expert panel, who established the diagnosis of Conclusions The use of an endotracheal tube with polyurethane cuff and intermittent subglottic secretion drainage helps prevent earlyand late-onset Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. References 1. Safdar N, Dezfulian C, Collard HR, Saint S. Clinical and economic consequences of ventilator-associated pneumonia: a systematic review. Crit Care Med 2005;33: Seegobin RD, van Hasselt GL. Aspiration beyond endotracheal cuffs. Can Anaesth Soc J 1986;33: Pavlin EG, VanNimwegan D, Hornbein TF. Failure of a high-compliance low-pressure cuff to prevent aspiration. Anesthesiology 1975;42: Petring OU, Adelhoj B, Jensen BN, Pedersen NO, Lomholt N. Prevention of silent aspiration due to leaks around cuffs of endotracheal tubes. Anesth Analg 1986;65: Young PJ, Rollinson M, Downward G, Henderson S. Leakage of fluid past the tracheal tube cuff in a benchtop model. Br J Anaesth 1997;78: Oikkonen M, Aromaa U. Leakage of fluid around low-pressure tracheal tube cuffs. Anaesthesia 1997;52: Rello J, Sonora R, Jubert P, Artigas A, Rue M, Valles J. Pneumonia in intubated patients: role of respiratory airway care. Am J Respir Crit Care Med 1996;154: Mahul Ph, Auboyer C, Jospe R, Ros A, Guerin C, El Khouri Z, Galliez M, Dumont A, Gaudin O. Prevention of nosocomial pneumonia in intubated patients: respective role of mechanical subglottic secretions and stress ulcer prophylaxis. Intensive Care Med 1992;18: Bo H, He L, Qu J. Influence of the subglottic secretion drainage on the morbidity of ventilator associated pneumonia in mechanically ventilated patients. Zhonghua Jie He He Hu Xi Za Zhi 2000;23: Smulders K, van der Hoeven H, Weers-Pothoff I, Vandenbroucke- Grauls C. A randomized clinical trial of intermittent subglottic secretion drainage in patients receiving mechanical ventilation. Chest 2002;121: Vallés J, Artigas A, Rello J, Bonsoms N, Fontanals D, Blanch L, Fernandez R, Baigorri F, Mestre J. Continuous aspiration of subglottic secretions in preventing ventilator-associated pneumonia. Ann Intern Med 1995;122: Kollef MH, Skubas NJ, Sundt TM. A randomized clinical trial of continuous aspiration of subglottic secretions in cardiac surgery patients. Chest 1999;116: Girou E, Buu-Hoi A, Stephan F, Novara A, Gutmann L, Safar M, Fagon YJ. Airway colonisation in long-term mechanically ventilated patients: effect of semi-recumbent position and continuous subglottic suctioning. Intensive Care Med 2004;30: Torres A, Carlet J; European Task Force on Ventilator-associated Pneumonia. Ventilator-associated pneumonia. Eur Respir J 2001;17: Centers for Disease Control and Prevention (CDC). Guidelines for preventing health-care associated pneumonia 2003: recommendations of CDC and the Healthcare Infection Control Practices Advisory Committee [Internet] [accessed September 2007]. MMWR Recomm Rep 2004;53:1 36. Available from mmwrhtml/rr5303a1.htm 16. Dodek P, Keenan S, Cook D, Heyland D, Jacka M, Hand L, Muscedere J, Foster D, Mehta N, Hall R, et al.; Canadian Critical Care Trials Group; Canadian Critical Care Society. Evidence-based clinical practice guideline for the prevention of ventilator-associated pneumonia. Ann Intern Med 2004;141: American Thoracic Society; Infectious Diseases Society of America. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med 2005;171: Dezfulian C, Shojania K, Collard HR, Kim HM, Matthay MA, Saint S. Subglottic secretion drainage for preventing ventilator-associated pneumonia: a meta-analysis. Am J Med 2005;118: Dullenkopf A, Gerber A, Weiss M. Fluid leakage past tracheal tube cuffs: evaluation of the new Microcuff endotracheal tube. Intensive Care Med 2003;29: Dullenkopf A, Schmitz A, Frei M, Gerber AC, Weiss M. Air leakage around endotracheal tube cuffs. Eur J Anaesthesiol 2004;21: Lewis FR Jr, Schiobohm RM, Thomas AN. Prevention of complications from prolonged tracheal intubation. Am J Surg 1978;135: Black AM, Seegobin RD. Pressures on endotracheal tube cuffs. Anaesthesia 1981;36: Blunt MC, Young PJ, Patil A, Haddock A. Gel lubrication of the tracheal tube cuff reduces pulmonary aspiration. Anesthesiology 2001; 95: Berra L, De Marchi L, Panigada M, Yu ZX, Baccarelli A, Kolobow T. Evaluation of continuous aspiration of subglottic secretion in an in vivo study. Crit Care Med 2004;32: Canadian Critical Care Trials Group. A randomized trial of diagnostic techniques for ventilator-associated pneumonia. N Engl J Med 2006; 355: