Elsevier

Journal of Clinical Anesthesia

Volume 51, December 2018, Pages 49-54
Journal of Clinical Anesthesia

Original Contribution
The relationship between end-expired carbon dioxide tension and severity of venous air embolism during sitting neurosurgical procedures – A contemporary analysis

https://doi.org/10.1016/j.jclinane.2018.08.007Get rights and content

Highlights

  • Venous air embolism (VAE) can be associated with a range of hemodynamic manifestations

  • We evaluated the association between the severity of hemodynamic changes and the magnitude of changes in end-expired carbon dioxide tension during VAE

  • The magnitude of changes in expired carbon dioxide has limited ability to predict severity of hemodynamic changes during VAE

Abstract

Study objective

Determine if changes in expired carbon dioxide tension correlate with the severity of venous air embolism (VAE) associated hemodynamic changes in humans.

Design

Retrospective case series.

Setting

A single academic medical center with high-volume neurosurgical practice.

Patients

One hundred forty seven adult patients having neurosurgical procedures performed with general anesthesia in the sitting position who experienced venous air embolism.

Interventions

Identification of documentation of venous air embolism by either precordial Doppler sonography or transesophageal echocardiography.

Measurement

Retrospective determination of changes in end-expired carbon dioxide (EECO2) changes associated with venous air embolism.

Main results

Greater absolute and relative decreases in end-expired carbon dioxide tension were associated with greater hemodynamic manifestations of venous air embolism. However, based on receiver operating characteristic curve analysis, the absolute and relative changes in EECO2 have moderate utility for predicting the severity of hemodynamic consequences of venous air embolism as area under the curve for absolute and relative carbon dioxide tensions were 0.7654 and 0.7263, respectively.

Conclusions

Greater magnitude of decreases in EECO2 is associated with hemodynamically-significant VAE in mechanically-ventilated patients. However, the magnitude of changes may have limited utility to diagnose VAE or exclude the diagnosis of VAE in patients with unexplained intraoperative hypotension.

Introduction

Venous air embolism (VAE) is a known complication of procedures performed in the sitting position. The incidence of VAE in the sitting position has been reported as high as 25% [[1], [2], [3], [4]]. The severity of VAE events can vary from clinically insignificant to life-threatening hemodynamic collapse, respiratory failure, permanent neurologic deficits, and death [[1], [2], [3], [4]]. Many institutions have chosen to forego utilization of the sitting position for surgical procedures as a result of these documented risks [5,6]. However, the sitting position offers several physiologic, surgical, and anesthetic advantages for neurosurgical operations when compared to the supine position, including improved visualization of the surgical field, decreased brain bulk, and improved postoperative cranial nerve function [1]. As the sitting position continues to be utilized, prediction and early identification of VAE is critical.

Transesophageal echocardiography (TEE) and precordial Doppler sonography are most commonly utilized to monitor for perioperative VAE [1,2]. Though nonspecific, reductions in end-expired carbon dioxide tension (EECO2) may also suggest the presence of VAE. Such changes occur during VAE events due to the development of dead space in the lungs.

VAE events are not limited to neurosurgical procedures performed in the sitting position. In these cases, clinicians may not monitor for VAE events with TEE or precordial Doppler sonography. Consequently, they may be limited to changes in EECO2 and hemodynamic changes to warn of acute VAE. Thus, understanding the utility of changes in EECO2 as a predictor of the severity of VAE in cases where routine monitors for VAE are not employed can serve significant clinical utility.

The correlation between the magnitude of changes in EECO2 and the severity of hemodynamic manifestations of VAE in humans has been sparsely reported in the literature [7]. The present study aims to determine the ability of observed changes in EECO2 for predicting the severity of venous air embolism.

Section snippets

Materials and methods

Following approval from the Institutional Review Board at the Mayo Clinic in Rochester, MN, patients who underwent a neurosurgical procedure performed in the sitting position utilizing general anesthesia between January 1, 2000 and October 8, 2013 were retrospectively identified. Neurosurgical procedures performed in the sitting position were identified from the electronic listing form completed by the surgeon where only procedures performed by neurosurgeons were included and the surgical

Results

During the time interval between January 1, 2000 and October 8, 2013, a total of 1886 patients underwent neurosurgical procedures in the sitting position. One hundred eighty eight patients were excluded due to inconsistent or missing data. As such, VAE events were identified from among 1698 sitting neurosurgical cases that met inclusion criteria. Of these, 168 were identified as having a VAE event that met criteria for inclusion into the study cohort yielding a VAE incidence of 9.9%. Four of

Discussion

In this large retrospective study of clinically diagnosed VAE, we describe an overall VAE incidence of 9.9%, consistent with the previously reported range for sitting neurosurgical procedures [[1], [2], [3], [4]]. We describe the association between the absolute and relative magnitude of decreases in EECO2 and the severity of hemodynamic changes associated with VAE. Specifically, the magnitude of the absolute and relative decreases in EECO2 is greater with VAE associated with significant

Disclosures

None.

References (11)

There are more references available in the full text version of this article.

Cited by (7)

View all citing articles on Scopus
View full text