Comparison of the effect of oral care with four different antiseptics to prevent ventilator-associated pneumonia in adults: protocol for a network meta-analysis

Zhigang Zhang, Yuying Hou, Jun Zhang, Bo Wang, Juxia Zhang, Ailing Yang, Ge Li, Jinhui TianSystematic 2017 · DOI

Abstract

Ventilator-associated pneumonia (VAP) remains a significant complication in intensive care units (ICUs), contributing to increased mortality and healthcare costs. While oral antiseptics are standard preventative measures, the relative efficacy of different agents remains unclear due to a lack of head-to-head clinical trials. This protocol outlines a Bayesian network meta-analysis (NMA) designed to compare and rank four major oral care antiseptics—chlorhexidine, povidone-iodine, sodium bicarbonate, and herbal extracts—to identify the most effective intervention for VAP prevention.

Introduction

VAP affects approximately 10–30% of patients intubated for more than 48 hours, with associated mortality rates ranging from 15% to 45%. Beyond the human cost, VAP imposes a substantial economic burden, estimated at $50,000–$57,000 per case. Current clinical guidelines advocate for oral care to reduce bacterial colonization, yet there is no consensus on the optimal antiseptic solution. While chlorhexidine is widely utilized, alternative agents like povidone-iodine and sodium bicarbonate are also employed, often without robust comparative evidence.

Methodology

To address the evidence gap, this study employs a Bayesian Network Meta-Analysis. Unlike traditional pairwise meta-analyses, NMA allows for the simultaneous comparison of multiple treatments by integrating both direct evidence (from head-to-head trials) and indirect evidence (via common comparators).

Search Strategy and Study Selection

The researchers will conduct a comprehensive systematic search across seven major databases: PubMed, Web of Science, Cochrane Library, EMBASE, CNKI, WanFang, and SinoMed. The inclusion criteria focus on randomized controlled trials (RCTs) involving adult patients receiving mechanical ventilation for over 48 hours.

Statistical Analysis

The analysis will utilize a Bayesian framework implemented via WinBUGS V1.4.3 and Stata 13.0. Key features include:

  • Surface Under the Cumulative Ranking (SUCRA): This metric will provide a probability-based ranking of the four antiseptics, identifying which is most likely to be the "gold standard."
  • Inconsistency Assessment: A Z-test will be applied to evaluate potential discrepancies between direct and indirect evidence, ensuring the reliability of the network's conclusions.
  • Risk of Bias: The Cochrane "risk of bias" tool will be used to assess the quality of included studies.

Expected Outcomes

The primary outcome of interest is the incidence of VAP. Secondary outcomes include the duration of mechanical ventilation, length of ICU stay, and total hospital stay duration. By synthesizing data from a wide range of RCTs, the study aims to provide clinicians with a clear hierarchy of antiseptic efficacy.

Discussion and Limitations

The strength of this protocol lies in its ability to resolve clinical uncertainty where direct trial data is sparse. However, the validity of the NMA is inherently tied to the quality of the primary RCTs. Subgroup analyses will be performed to account for variations in ICU settings and ventilation duration, acknowledging that patient-specific factors may influence antiseptic performance.

Conclusion

This protocol establishes a rigorous framework for determining the most effective oral antiseptic for VAP prevention. The resulting network meta-analysis will offer a data-driven foundation for ICU oral care protocols, potentially improving patient outcomes and reducing healthcare expenditures.

Summary updated

Abstract

PROSPERO CRD42016038088.
BibTeX
@inproceedings{zhang2017comparison,
  author = {Zhigang Zhang and Yuying Hou and Jun Zhang and Bo Wang and Juxia Zhang and Ailing Yang and Ge Li and Jinhui Tian},
  title = {{Comparison of the effect of oral care with four different antiseptics to prevent ventilator-associated pneumonia in adults: protocol for a network meta-analysis}},
  booktitle = {Systematic},
  year = {2017},
  doi = {10.1186/s13643-017-0496-5},
}
Figure