|
|
||||||||
J Thorac Cardiovasc Surg 2008;135:771-776
© 2008 The American Association for Thoracic Surgery
Cardiopulmonary Support and Physiology |
a International Research Center, Ghent University, Gent, Belgium
b Department of Intensive Care Medicine, Ghent University Hospital, Gent, Belgium
c Hogeschool Gent, Healthcare Department, Gent, Belgium
Received for publication February 23, 2007; revisions received August 16, 2007; accepted for publication August 23, 2007. * Address for reprints: Jan Poelaert, MD, PhD, Department of Anesthesiology, Free University Hospital Brussels, Laarbecklaan 101, 1090 Brussels, Belgium. (Email: jan.poelaert{at}uzbrussels.be).
| Abstract |
|---|
|
|
|---|
Methods: In a prospective, single-blind, randomized study, patients scheduled for cardiac surgery were allocated to intubation with a polyurethane cuffed endotracheal tube or the routinely used polyvinyl chloride cuffed endotracheal tube. Patients were scheduled for routine or emergency cardiac surgery and admitted to an 8-bed cardiac surgical intensive care unit of a tertiary care hospital.
Results: A total of 134 patients were available for analysis (67 in each group). Whereas mortality was not different between the groups, the incidence of early postoperative pneumonia and empirical prescription of antibiotic therapy were significantly lower in the polyurethane group than in the polyvinyl chloride group (23% vs 42%, P < .03). Intensive care unit and hospital stays were not significantly different between the two study subsets (3 ± 5 days vs 3 ± 4 days and 16 ± 9 vs 17±11 days, respectively). In a multivariate regression analysis, preoperative serum creatinine levels (odds ratio 1.85, confidence interval 1.02–3.37, P = .04) and perioperative transfusion (odds ratio 1.50, confidence interval 1.08–3.37, P = .015) were independently associated with increased risk of early postoperative pneumonia, whereas use of a polyurethane endotracheal tube was protective (odds ratio 0.31, confidence interval 0.13–0.77, P = .01).
Conclusion: Polyurethane cuffed endotracheal tubes can reduce the frequency of early postoperative pneumonia in cardiac surgical patients.
| Introduction |
|---|
|
|
|---|
Cardiac surgical patients represent an important subset of patients at risk for postoperative pneumonia because of the universal and unavoidable application of invasive mechanical ventilation in supine positioning, the frequent use of intraoperative transesophageal echocardiography (promoting gastroesophageal regurgitation), and the systemic inflammatory response syndrome induced by surgical trauma. Respiratory infection is estimated to complicate the postoperative course of 5% to 20% of cardiac surgical patients.14-16
Because postoperative and ventilator-associated pneumonia have been identified as predictors of worse outcome in this particular patient category, strategies to prevent these pulmonary infections are strongly advocated. We assessed the potentially preventive role of PU cuffed ETs in the development of early postoperative pneumonia in a cohort of cardiac surgical patients in a single-blind, randomized, controlled trial.
| Materials and Methods |
|---|
|
|
|---|
Patients and Procedures
All patients scheduled for routine cardiac surgery or undergoing urgency or emergency cardiac surgery in the postoperative cardiac surgical ICU at our tertiary referral center were eligible for the study. Patients with pneumonia before surgery, with infectious endocarditis, or with current antibiotic therapy were excluded. Randomization occurred according to a computer-generated list, which was kept by the attending anesthesiologists, blinded for the intensivists. Female patients routinely received an ET with 8-mm internal diameter size, whereas male patients were intubated with a 9-mm internal diameter ET, although the final decision was left to the discretion of the attending physician.
Immediately after endotracheal intubation and after closure of the sternum, cuff pressure was assessed by the attending anesthesiologist. Care was taken to maintain cuff pressures between 20 and 26 cm H2O in both subsets. Intraoperative antibiotic prophylaxis consisted of cefazolin (2 g every 8 hours for 24 hours) in all cases.
After cardiac surgery, patients were transferred to the 8-bed cardiac surgical ICU for postoperative care and weaning from mechanical ventilation. Sedation was continued with propofol (0.5–2 mg/[kg · h] intravenously) as previously discussed.17
All patients received standard antacids with intravenous ranitidine (50 mg three times daily). Only vasoactive drugs like nitrates and sometimes isotopic drugs were administered. Cuff pressure was measured immediately on arrival and subsequently each 4 hours until extubation in all cases. Again, cuff pressures were monitored by attending nurses and carefully maintained between 20 and 26 cm H2O. Criteria for weaning and extubation included stable cardiovascular dynamics and metabolism, chest tube drainage less than 1 mL/(kg · h) for at least 2 consecutive hours, adequate gas exchange (defined as PaO
2 >70 mm Hg with inspired oxygen fraction
0.6, peak inspiratory pressure below 20 cm H2O, and peak end-expiratory pressure below 6 cm H2O), and adequate neurologic activity.18
Antibiotic prophylaxis consisted of cefazolin (100 mg/[kg · d]) and lasted 24 hours.
Screening and Diagnosis
After admission to the ICU, patients were screened daily for postoperative nosocomial pneumonia by the attending ICU physician, who was blinded to the ET tube allocation. Diagnosis of postoperative nosocomial pneumonia was defined before onset of the study. Because in our patient population microbial pathogens other than commensal flora are rarely isolated in this setting of early postoperative pneumonia, we aimed for higher specificity in our clinical diagnosis by requiring that all Johanson criteria,19
as well as two additional criteria, be fulfilled.20
Thus postoperative nosocomial pneumonia was defined as the presence of a new or evolving infiltrate on chest radiography within 7 days after surgery, together with the presence of all Johanson criteria19
: temperature greater than 38.2°C, leukocytosis greater than 12,000 cells/mm3, and presence of purulent sputum or endotracheal aspirate. Additional criteria were an increase in C-reactive protein for at least 2 consecutive days after surgery and a deterioration in the ratio of PaO
2 to inspired oxygen fraction of at least 20%. Diagnosis was made during the stay in the ICU by the attending intensivist, who was blinded to the assigned ET. Oropharyngeal and endotracheal aspirates were taken on ICU admission and on clinical suspicion of pneumonia. Microbial evaluation consisted of direct Gram staining and semiquantitative culture. As such, the initiation of antibiotic therapy was left to the discretion of the attending staff ICU physician. At our ICU, surveillance cultures are used to guide empirical antibiotic therapy in cases of suspected pneumonia.21,22
In the absence of surveillance cultures, empirical therapy for early (within 7 days after ICU admission) postoperative or ventilator-associated pneumonia without previous antibiotic exposure consists of cefuroxime or amoxicillin (INN amoxicilline) and clavulanate, with an antipseudomonal betalactam if pneumonia develops after 7 days or after previous antibiotic exposure. This empirical scheme is modified accordingly if surveillance cultures show the presence of multidrug-resistant organisms such as extended-spectrum β-lactamase–carrying Enterobacteriaceae or methicillin-resistant Staphylococcus aureus.
Variables Recorded and Calculated
Two scoring systems were used, EuroSCORE23,24
and Tu score.25,26
Preoperative serum creatinine level and requirement for blood transfusion during or immediately after surgery (within 48 hours) were recorded. The primary end point was the occurrence of pneumonia within 3 days of cardiac surgery, considered to represent early postoperative pneumonia.
Statistical Analysis
The study was performed in two parts. A first, smaller part (n = 30) allowed calculation of statistical power analysis (power .80, type II error .05), yielding a target sample size of 120 patients.
Data are expressed as mean ± SD. Continuous variables were assessed with ANOVA and when appropriate with an independent t test. Categorical variables were analyzed with the
2 test. Backward stepwise logistic regression analysis was performed to elucidate interfering factors for development of early postoperative pneumonia. To avoid spurious associations, only variables with a plausible relationship to the development of pneumonia or variables with a relationship in univariate analysis (P = .20) were entered into the model. Statistical analyses were executed with the SPSS version 10 software package (SPSS Inc, Chicago, Ill).
| Results |
|---|
|
|
|---|
|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
Prevention of early postoperative pneumonia is to be achieved through a multimodality and multidisciplinary effort.11,28
Strategies to disrupt the sequence of oral colonization with pathogenic organisms, followed by dislocation of this contaminated material into the lower airways through defective upper airway barriers and subsequent development of tracheobronchitis and ultimately pneumonia, however, play a central role in each preventive program. Semirecumbent positioning, rather than supine positioning, of ventilated patients (especially when enterally fed) has been conclusively shown to decrease the risk of early postoperative and ventilator-associated pneumonia, as has to a lesser extent the practice of continuous suctioning of subglottic secretions.28,29
A recent multicenter trial reported reduced lower bacterial airway colonization associated with the use of a silver-coated ET, probably through the topical antimicrobial effect of silver ions, although a reduction in prevalence of subsequent pneumonia was not observed.30
From this perspective, decreasing leakage of contaminated upper airway secretions through folds and channels between (partially) inflated ET cuffs and tracheal wall can be considered as part of the central core of preventive measures, especially in circumstances where semirecumbent positioning is not possible, such as during cardiac surgery. Altering the material of the cuff from PVC to PU permits a thinner cuff wall, which therefore expands more completely and follows internal diameter of the trachea more closely when the cuff is inflated.13
Table 6 discloses that the type of ET tube was a stronger predictor than either the number of packed red cell units transfused or preoperative renal function. In a large retrospective study on risk factors for postoperative infection in cardiac surgery, blood bank requirements, among many other factors, were associated with postoperative infection risk.31
and Corwin and associates32
identified transfusion as an independent risk factor for ventilator-associated pneumonia in a general ICU population. Insofar as transfusion policy is dependent on predefined transfusion thresholds, rather than reflecting perioperative complications, this may prove to be a modifiable risk factor as well. Nevertheless, prevention of postoperative pneumonia appears to be far more important.
Some drawbacks and shortcomings of this investigation should be mentioned. A major flaw of this investigation could be the lack of microbiologic confirmation of most episodes of postoperative pneumonia. This may in part have been due to the postoperative setting in which this study was conducted. All patients received intraoperative antibiotic prophylaxis (with a first-generation cephalosporin), which probably has prevented subsequent isolation of microbial pathogens causing early-onset pneumonia, because these are known to be reflective of patients' endogenous flora rather than potentially multidrug-resistant nosocomial flora. In addition, because most patients were already extubated when pneumonia was considered clinically likely, part of this low yield may have been caused by difficulties in obtaining a good quality lower airway microbiologic sample. Alternatively, because a clinical diagnosis of nosocomial pneumonia suffers from lack of specificity,19
it could be argued that some of the episodes of postoperative pneumonia were in fact noninfectious complications mimicking pneumonia, such as aspiration pneumonitis, acute lung injury, or atelectasis. Clearly, there is need for more discriminative diagnostic tests for nosocomial pneumonia, in particular ventilator-associated pneumonia, to reduce unnecessary antibiotic therapy. Because invasive microbiologic sampling was not judged feasible for our fragile, recently extubated patient population, and additional diagnostic tests such as procalcitonin are not yet validated for nosocomial pneumonia33
and were not available to us, a clinical diagnosis of pneumonia was the best we could offer. Moreover, this clinically based diagnosis is reflective of widespread clinical practice, with additional variables increasing specificity to 92%.20,34
It is important to mention that our ICU physicians responsible for postoperative care, including diagnosis and antibiotic treatment of likely pneumonia, were blinded to the patient allocation.
At the very least, our study shows that use of a PU cuffed tube is able to decrease early postoperative antibiotic prescription by half. This reduction may have a favorable effect indirectly, because empirical postoperative antibiotic therapy itself has been found to increase the risk for nosocomial infection in cardiac surgical patients.35
Our results may not be readily extrapolated to the general ICU population. All our patients were intubated in the setting of anesthetic induction for cardiac surgery, so risk for significant aspiration during this procedure was kept to a minimum. In an unplanned or emergency resuscitation, significant aspiration of oropharyngeal content into the lower airways may happen before or during airway intubation, diminishing the preventive effect of a tighter tracheal seal once the ET is in place. Because the duration of mechanical ventilation in our patients was short, it has yet to be established whether a PU cuffed ET would limit endotracheal microaspiration during prolonged mechanical ventilation.
Another issue might be that some results are skewed, in particular with respect to ejection fraction and surgery duration (Table 1). A clinical difference was nonexistent, however, and the presence of comparable scoring indices could preclude this potential impact.
In conclusion, use of a PU cuffed ET instead of a standard PVC cuffed ET during cardiac surgery could significantly decrease the occurrence of clinically suspected pneumonia in the early postoperative phase and subsequently reduce the use of empirical antibiotic therapy. A larger randomized trial is desirable to confirm this finding and to examine whether it holds true when a more rigid, microbiologically centered diagnosis of pneumonia is applied. In the absence of a reliable diagnostic strategy for ventilator-associated pneumonia or postoperative pneumonia, however, use of a PU cuffed ET in cardiac surgery can still be advocated as an antibiotic-reducing and cost-effective strategy.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. Lizy, W. Swinnen, S. Labeau, and S. Blot Deviations in Endotracheal Cuff Pressure During Intensive Care Am. J. Crit. Care., November 1, 2011; 20(6): 421 - 422. [Full Text] [PDF] |
||||
![]() |
M. H. Dave, A. Frotzler, C. Madjdpour, N. Koepfer, and M. Weiss Massive Aspiration Past the Tracheal Tube Cuff Caused by Closed Tracheal Suction System J Intensive Care Med, October 1, 2011; 26(5): 326 - 329. [Abstract] [PDF] |
||||
![]() |
L. E. Morrow and M. H. Kollef VAP prevention: pharmacological strategies Nosocomial Pneumonia and Ventilator Associated Pneumonia, September 12, 2011; 74 - 82. [Abstract] [Fulltext] [PDF] |
||||
![]() |
M. Valencia, G. Li Bassi, and A. Torres VAP prevention:non-pharmacological strategies Nosocomial Pneumonia and Ventilator Associated Pneumonia, September 12, 2011; 83 - 97. [Abstract] [Fulltext] [PDF] |
||||
![]() |
J. Poelaert Ventilator-associated Pneumonia and Cuff Shape Am. J. Respir. Crit. Care Med., August 15, 2011; 184(4): 485 - 485. [Full Text] [PDF] |
||||
![]() |
L. Lorente, S. Blot, and J. Rello Ventilator-associated Pneumonia and Cuff Shape Am. J. Respir. Crit. Care Med., August 15, 2011; 184(4): 485 - 486. [Full Text] [PDF] |
||||
![]() |
L. Lorente, S. Blot, and J. Rello New Issues and Controversies in the Prevention of Ventilator-associated Pneumonia Am. J. Respir. Crit. Care Med., October 1, 2010; 182(7): 870 - 876. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. H. Dave, A. Frotzler, N. Spielmann, C. Madjdpour, and M. Weiss Effect of tracheal tube cuff shape on fluid leakage across the cuff: an in vitro study Br. J. Anaesth., October 1, 2010; 105(4): 538 - 543. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Weiss, C. Doell, N. Koepfer, C. Madjdpour, K. Woitzek, and V. Bernet Rapid pressure compensation by automated cuff pressure controllers worsens sealing in tracheal tubes Br. J. Anaesth., February 1, 2009; 102(2): 273 - 278. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |