Inter-observer reliability of alveolar dead space measurements in suspected pulmonary embolism

Objective: Less than 35% of patients suspected of having pulmonary embolism (PE) actually have PE. Safe bedside methods to exclude PE could save health care resources and improve access to diagnostic testing for suspected PE. Recently, interest has been generated in using alveolar dead space measurements to exclude PE at the bedside. Prior to widespread adoption of alveolar dead space measurement in patients with suspected PE the reproducibility of these measurements must be demonstrated. We sought to determine the inter-observer reliability of three previously published techniques of alveolar dead space fraction measurement. Design: Prospective cohort study. Setting: Tertiary care center in Ottawa, Ontario, Canada. Patients: Consecutive inpatients, outpatients or emergency room patients with suspected PE referred to Nuclear Medicine or Radiology for investigation of suspected PE. Interventions: All study patients had alveolar dead space measurement performed by three techniques by two different respiratory therapists blinded to each other's results and outcome (PE or No PE). Main Results: The steady state end tidal alveolar dead space fraction measurement had a Kappa of 1.00 indicating excellent inter-observer agreement. The alveolar plateau dead space fraction measurement and the end expired alveolar dead space fraction measurements had Kappa scores less than 0.3 indicating poor inter-observer agreement. Conclusions: Alveolar plateau and end expired alveolar dead space fraction measurements have poor reproducibility. Steady state end tidal alveolar dead space fraction measurement is a simple reproducible bedside test.