Continuous pulse oximetry and the diagnosis of pulmonary embolism in critically ill trauma patients

Collin E.M. Brathwaite, Keith F. O’Malley, Steven E. Ross, Peter Pappas, James Alexander, Richard K. Spence

Research output: Contribution to journalArticlepeer-review

15 Scopus citations


The diagnosis of pulmonary embolism (PE) may be difficult to establish in trauma patients, particularly those who are unresponsive or mechanically ventilated. Based on a prior retrospective study, we hypothesized that patients monitored by continuous pulse oximetry who experienced a 10% or greater sudden sustained drop in arterial oxygen saturation (Sao2) without a change in static lung compliance (Cst) were most likely to have had a PE. We followed Sao2 in 972 patients admitted to our trauma ICU during the 18-month period ending in December 1990. Forty-eight patients (5%) with Sao2 changes, but no Cst changes, were evaluated for suspected PE using pulmonary arteriography (PA). Of these, 21 (44%) had a positive PA study. All patients with a positive PA had either clear chest roentgenograms or no change in underlying pulmonary pathologic processes. Of the remainder, 26 had evidence of a new pathologic entity on chest roentgenograms and only one patient had a Sao2 decrease, no change in Cst, and a negative PA. All mechanically ventilated trauma patients should have Sao2 monitored continuously. Patients with a >10% drop in Sao2 with no change in Cst and no new roentgenographic chest findings should undergo PA. Based on our experience, this approach would yield a sensitivity, specificity, and predictive value of 100%, 99.9%, and 95%, respectively, for the diagnosis of clinically significant PE.

Original languageEnglish (US)
Pages (from-to)528-531
Number of pages4
JournalJournal of Trauma - Injury, Infection and Critical Care
Issue number4
StatePublished - Oct 1992
Externally publishedYes

ASJC Scopus subject areas

  • Surgery
  • Critical Care and Intensive Care Medicine


Dive into the research topics of 'Continuous pulse oximetry and the diagnosis of pulmonary embolism in critically ill trauma patients'. Together they form a unique fingerprint.

Cite this