Venous Air Embolism

Abstract

Background

Venous air embolism (VAE) is an understudied entity. Herein, we summarize VAE case reports and small case series reported in the literature.

Research Question

What are the clinical features, diagnostic approaches, and clinical outcomes of VAE and how do surgery-related VAEs compare with non-surgery-related VAEs?

Study Design and Methods

Using the search terms air, gas, venous, and embolism, 437 articles were identified. After applying predetermined exclusion criteria, we included the 164 articles describing cases of isolated VAE. We extracted data pertaining to patient demographics and clinical presentations; VAE characteristics, for example, cause and clinical context; diagnostic testing and time to diagnosis; and clinical management and outcomes. We used the Shapiro-Wilk test to assess data distribution (ie, normally vs nonnormally distributed), the Pearson χ² test for categorical variables, and the Mann-Whitney U test and t test for continuous variables.

Results

We collated 174 patients; 108 patients (62.1%) were male. Most VAE episodes (n = 160 [92%]) were iatrogenic. Eighty-two patients (47%) experienced respiratory, cardiac, or neurologic symptoms, or a combination thereof, whereas 15 patients (8.6%) were asymptomatic; the remaining patients (n = 77 [44.3%]) had collapsed or been intubated before VAE diagnosis. Most patients (56.9%) were hemodynamically unstable on presentation. Diagnostic and management approaches varied considerably across reports. Of management strategies, oxygen supplementation (Fio₂ = 1.0) and body repositioning were implemented most frequently. Seventy-nine patients (45%) received ICU level of care, 13 patients (7.5%) underwent endotracheal intubation, 39 patients (22.4%) received inotropic support, and 32 patients (18.4%) died. Compared with patients with non-surgery-related VAEs, those with surgery-related VAEs underwent end-tidal CO₂ measurement more frequently (50% vs 3%; P < .001) and showed lower all-cause mortality (11.2% vs 24.5%; P = .01). Time to diagnosis was nonsignificantly shorter in surgery-related episodes. Publication bias is one of our study's limitations.

Interpretation

Approximately one-half of VAEs are nonsurgical. Diagnostic and management strategies varied widely across reports, reflecting disease heterogeneity and inconsistent clinical approach. All-cause mortality was higher for non-surgery-related episodes than for surgery-related episodes. Considering the comparable age, sex distribution, and comorbidities between these two groups, this finding deserves further study.