摘要

Background: Arterial gas embolism (AGE) is a major cause of morbidity and mortality in self-contained underwater breathing apparatus (SCUBA) diving and certain medical procedures. There are currently no well-defined criteria to diagnose AGE. Emergency physicians often find themselves facing a decision whether or not a patient with dive-related symptoms has an AGE and needs to be transferred to a hyperbaric facility. Objectives: The objective of this study was to test the accuracy of diagnostic criteria developed by the San Diego Hyperbaric Oxygen Group (SANDHOG) for the diagnosis of AGE. Methods: This was a retrospective review of consultations completed by the Hyperbaric Medicine Department (HBO) at the University of California San Diego where the diagnosis of AGE was considered. HBO staff blinded to the purpose of the study identified potential cases of AGE. The criterion standard was the final diagnosis by a panel of HBO specialists also blinded to the purpose of this study. Descriptive statistics and comparisons evaluating SANDHOG criteria compared to the criterion standard were performed. Results: Twenty-six patients were identified for inclusion. Twenty-three of 26 (88%) were SCUBA divers, 2 had intravascular gas injections, and 1 patient had a military training chamber accident. Nineteen of 26 (73%) patients were diagnosed with AGE. A SANDHOG score of 2 had 94.7% sensitivity (95% confidence interval [CI] 71.9-99.7), 85.7% specificity (95% CI 42.0-99.2), positive likelihood ratio of 6.6 (95% CI 1.1-40.8), and negative likelihood ratio of 0.06 (95% CI 0.01-0.43) for AGE. A SANDHOG score < 2 had a negative predictive value of 100% for AGE. Conclusion: The SANDHOG criteria appear to be reliable in diagnosing AGE. AGE is unlikely with SANDHOG scores < 2, whereas SANDHOG scores >= 2 resulted in high sensitivity and specificity for AGE. Emergency physicians may find this tool useful in evaluating patients for suspected AGE, and it may assist in determining whether to transfer the patient to a hyperbaric facility. Future studies should be performed to further examine and validate the accuracy and inter-rater reliability of this tool.

  • 出版日期2015-11

全文