High tidal volume ventilation is associated with ventilator-associated pneumonia in acute cervical spinal cord injury

Abstract

Context/Objective

Pneumonia is the leading cause of death after acute spinal cord injury (SCI). High tidal volume ventilation (HVtV) is used in SCI rehabilitation centers to overcome hypoventilation while weaning patients from the ventilator. Our objective was to determine if HVtV in the acute post-injury period in SCI patients is associated with lower incidence of ventilator-associated pneumonia (VAP) when compared to patients receiving standard tidal volume ventilation.

Design

Cohort study.

Setting

Red Duke Trauma Institute, University of Texas Health Science Center at Houston, TX, USA.

Participants

Adult Acute Cervical SCI Patients, 2011–2018.

Interventions

HVtV.

Outcome measures

VAP, ventilator dependence at discharge, in-hospital mortality.

Results

Of 181 patients, 85 (47%) developed VAP. HVtV was utilized in 22 (12%) patients. Demographics, apart from age, were similar between patients who received HVtV and standard ventilation; patients were younger in the HVtV group. VAP developed in 68% of patients receiving HVtV and in 44% receiving standard tidal volumes (P = 0.06). After adjustment, HVtV was associated with a 1.96 relative risk of VAP development (95% credible interval 1.55–2.17) on Bayesian analysis. These results correlate with a >99% posterior probability that HVtV is associated with increased VAP when compared to standard tidal volumes. HVtV was also associated with increased rates of ventilator dependence.

Conclusions

While limited by sample size and selection bias, our data revealed an association between HVtV and increased VAP. Further investigation into optimal early ventilation settings is needed for SCI patients, who are at a high risk of VAP.

Keywords:

• Spinal cord injury
• High tidal volume ventilation
• Ventilator-associated pneumonia

Acknowledgement

The authors thank Dr. Claudia Pedroza for her review of the statistical analysis presented in this manuscript.

Disclaimer statements

Conflicts of interest The authors report no conflicts of interest.

Disclosures GH and SW are supported by a T32 fellowship [grant number 5T32GM008792] from NIGMS.

Additional information

Funding

This work was supported by the National Institute of General Medical Sciences of the National Institutes of Health [grant number 5T32GM008792].