Prediction of Pulmonary Embolism in Postoperative Orthopaedic Patients Using End Tidal Carbon Dioxide Measurements

Background:

Pulmonary embolism (PE) is a morbid condition with sequelae including death and chronic pulmonary hypertension.  As a consulting hospitalist in the post-operative period, diagnostic algorithms including d-dimer testing are of limited utility as invasive procedures cause an elevation in d-dimer levels.  Because of its availability, sensitivity and specificity, spiral computed tomography is the test of choice for evaluating patients with suspected PE.  The test has some downsides, however, including radiation exposure and the risk of contrast-induced-nephropathy.

End Tidal CO₂ (ETCO₂) testing has the potential to be used as an indirect measure of PE. Pulmonary embolism results in dead space ventilation at the alveolar surface [1].  This results in reduced measurable CO₂ concentration at end expiration.  Although, some conditions such as advanced chronic obstructive pulmonary disease, or other pulmonary diagnoses may impact ETCO₂, recent surgery should not [1].  In this study, we hypothesize that ETCO₂ will be useful for predicting PE in a post-operative orthopedic surgical population, known to be at elevated risk for this outcome.  

Methods:

In this is single-center, prospective study, all patients aged 18 years of age or older who were admitted for joint or spine orthopedic surgery and whom had a CT pulmonary angiogram (CTPA) performed to evaluate for PE were eligible.  Patients undergoing CTPA also had an ETCO₂ measurement obtained.  Exclusion criteria were an inability to cooperate with either the ETCO₂ test or CT scan.  Neither patients nor testers were blinded, as the protocol dictated that all patients receive the ETCO₂ test and a diagnostic CT regardless of ETCO₂ results.  ETCO₂ was obtained any time within 24 hours of the completed diagnostic test.   Patients were determined to have PE if they were found to have a positive CT PE protocol, or presumed diagnosis by a treating physician associated with a corresponding International Classification of Diseases-9 (ICD 9) code.  Standard descriptive statistics were used to describe the data.

Results:

During a four month period, 46 patients met the inclusion criteria for the study.  Of these patients, 31 had a negative CT scan, 6 had a CT scan positive for subsegmental PE, 4 had a CT scan positive for segmental PE, and 5 had a non-diagnostic CT scan.  The ETCO₂ values with standard deviations (Figure 1/2) were 35.12 mm Hg (5.86), 34.38 mm Hg (7.97), 33.42 mm Hg (5.31), and 36.40 mm Hg (6.34), respectively. 

Conclusions:

This pilot study has demonstrated that ETCO₂ measurements may be a useful tool in identifying post-operative orthopaedic surgery patients with PE.  Currently, our data shows a trend towards lower ETCO₂ values for patients with PE, as expected.  To reach statistical power, additional patients will be recruited to further define this relationship and investigate other patient characteristics.

References:

[1] Hemnes, A.R., et al., Bedside end-tidal CO2 tension as a screening tool to exclude pulmonary embolism. Eur Respir J, 2010. 35(4): p. 735-41.