The trusted source for
healthcare information and
Do Silver-coated Endotracheal Tubes Affect VAP Mortality?
Abstract & Commentary
By Richard Wall, MD, MPH, Pulmonary Critical Care & Sleep Disorders Medicine, Southlake Clinic, Valley Medical Center, Renton, WA, is Associate Editor for Critical Care Alert.
Dr. Wall reports no financial relationship to this field of study.
Synopsis: This exploratory analysis of patients with VAP in a previous trial showed an association between lower mortality and use of a silver-coated tube.
Source: Afessa B, et al. Association between a silver-coated endotracheal tube and reduced mortality in patients with ventilator-associated pneumonia. Chest 2009 Dec 28; Epub ahead of print.
Endotracheal tubes (ETTs) coated with antimicrobial substances can reduce bacterial adhesion on the tube, block biofilm formation, and reduce bacterial burden in tracheal secretions. Various antimicrobial agents have been employed for this purpose. One tube has silver ions in a polymer on both the inner and outer lumens. The ions migrate to the surface and provide a sustained antimicrobial effect. Recently, the North American Silver-Coated Endotracheal Tube (NASCENT) study showed that a silver coating prevents ventilator-associated pneumonia (VAP).1
The NASCENT study was a prospective, randomized controlled trial conducted in 54 North American centers between 2002 and 2006. Adults expected to require mechanical ventilation with an ETT for at least 24 hours were randomly assigned to intubation with either a silver-coated tube or a non-coated tube. VAP diagnosis was based on quantitative cultures of bronchoalveolar lavage fluid. Overall, VAP rates were lower in patients intubated with coated tubes (4.8% vs 7.5%; P = 0.03), with a 36% relative risk reduction in VAP incidence. No significant differences were observed in the duration of intubation, ICU stay, hospital stay, mortality, or adverse events.
In the current paper, the NASCENT data set was re-examined to explore various subsets of patients. First, the authors looked at the subset of 93 patients who were diagnosed with VAP. Next, they looked at the tiny subset of 25 VAP patients who died. They then looked at the subset of 406 patients who died without VAP. Finally, the authors reviewed microbiology and identified VAP cases with potentially multidrug-resistant bacteria. All of this information was entered into a series of multivariate analyses to evaluate the influence of ETT type and other variables on mortality.
After displaying the various analyses in several somewhat confusing tables, the authors focus most of their discussion (and the title of the paper) on the subset of 25 VAP patients that died (see Figure, above). Their conclusion: A silver-coated ETT is associated with lower risk of mortality in patients with VAP.
This study is an exploratory analysis, pure and simple. The authors took subsets of data from a prior randomized trial and performed new multivariate analyses to see if they could uncover any associations. Such post-hoc analyses are frequently performed to generate new hypotheses for future studies. Information gleaned from an exploratory analysis, however, should never impact clinical practice.
An exploratory analysis is handicapped from the start. It inherits all of the problems from its parent trial including imbalances in enrollment, errors in measurement, missing data, unmeasured confounding, biases, and random bad luck. Then it asks new unblinded questions which the data were not originally designed to answer. Using a traditional P-value of 0.05 means that researchers only need to ask 20 exploratory questions before they will uncover a new false-positive "truth."
The original NASCENT study tested a specific intervention (silver-coated ETT) and looked at a specific outcome (VAP incidence). It was a sophisticated and challenging multicenter study, with large numbers of patients excluded for various reasons. The incidence of the primary outcome was low in both the intervention and control arms. Despite randomizing 2003 patients, there were only 93 VAP cases (see Figure).
The current article claims to look at "mortality in patients with VAP." In the NASCENT trial, however, mortality was a secondary outcome, and there were only 25 deaths in the two arms combined. Thus, any death analyses in the current paper are based on comparisons between a group of five and a group of 20. Did I mention that the current study was funded by the manufacturer of the silver-coated tube (Bard; Covington, GA)? Reader beware.
Don't get me wrong. I have enormous respect for the authors of the current paper. They are fantastic clinical researchers and world-class experts on VAP. Given their clinical acumen, their implications may indeed be true. However, this study is merely hypothesis-generating. These analyses are conducted in a subset of a subgroup of a narrowly selected study population. The study provides little useful information for a practicing clinician.
So what do we know? From the NASCENT trial, we know that silver-coated tubes lower the risk of VAP by approximately 35% and also delay the onset of VAP. However, the trial never suggested that silver-coated tubes reduce mortality. Quite the contrary. In the NASCENT trial, mortality rates were higher in the silver-coated group (31% vs 27%; P = 0.08). But yes, I suppose we know one other thing the authors will certainly be back with a larger more rigorous study looking at some of these newly uncovered "truths." Stay tuned.
1. Kollef MH, et al. Silver-coated endotracheal tubes and incidence of ventilator-associated pneumonia. JAMA 2008;300:805-813.