Assess the reproduciblity of objective measurements of ventilation/perfusion matching in pulmonary embolism

  • Ms Elizabeth Bailey, Department Nuclear Medicine, Royal North Shore Hospital, Australia
  • A/Prof Dale Bailey, Department Nuclear Medicine, Royal North Shore Hospital, Australia
  • Dr Benjamin Harris, Department Respiratory Medicine, Royal North Shore Hospital and Woolcock Institute of Medical Research, Australia
  • Dr Greg King, Department Respiratory Medicine, Royal North Shore Hospital and Woolcock Institute of Medical Research, Australia
  • A/Prof Paul Roach, Department Nuclear Medicine, Royal North Shore Hospital, Australia

    • Background: Methodology to objectively measure ventilation/perfusion (V/Q) relationships may enhance the diagnostic use of V/Q imaging. These measures correlate well with clinical findings of pulmonary embolism (PE). However, the reproducibility of these objective measures has not been determined.

    Aim: To assess the reproducibility of objective measures of V/Q relationships in the setting of PE.

    Methods: Subjects included those diagnosed with PE and being treated with anti-coagulant therapy for at least 3 months following initial diagnosis. Males and females of at least 50 years were recruited. An intial follow-up V/Q SPECT scan performed and repeated at 48 to 72hrs after initial study using either a Phillips SKYLight, GE Hawkeye or Picker 3000XP gamma camera. V/Q SPECT studies reconstructed using OSEM and registered using Multimodality (HERMES Medical Systems) software with a mutual information algorithm. Perfusion SPECT data corrected for ventilation counts using an IDL program developed in-house. The LogSDVQR, LogSDV1/V2 and LogSDQ1Q2 ratios were generated for each study with difference calculated to assess reproducibility.

    Results: At present, total of 7 subjects recruited, with study 2 acquired between 2 and 7 days following initial investigation. The LogSDVQR values for study 1 and 2 were 0.208 and 0.205 respectively, with an average difference of 0.0028. The V1V2 and Q1Q2 ratios were 0.102 and 0.134, with both being centred around 1, indicating good correlation.

    Conclusion: Objective measures of V/Q relationships using SPECT data repeated within 7 days of the initial investigation show high reproducibility, with strong correlation between ventilation and perfusion studies.