Pulmonary embolism: ventilation-perfusion scans could help diagnose PE.

Clinical bottom line (level 1b)

  1. A third of patients referred for a ventilation-perfusion scan had a pulmonary embolism.
  2. A high probability scan made a pulmonary embolism very likely (~ 85%).
  3. A normal/ near-normal scan made a pulmonary embolism much less likely ( ~ 3%).
  4. Around two-thirds of all scans were low or intermediate probability. Patients with low or intermediate probability scans needed further testing.
The PIOPED Investigators : Journal of the American Medical Association 1990; 263 (20): 2753-2759
Expires October 2003

The study

Setting: six clinical centres, USA

931 patients (aged mean 56 years, 55% female) suspected pulmonary embolism referred for ventilation-perfusion scan

Excluded if
  • <18 years old



    • Independent blinded reference standard, applied in all patients from a consecutive appropriate spectrum.
    Reference standard:
    • pulmonary angiogram or clinical follow-up for 12 months (2.5% incomplete)
    Diagnostic test: six-view ventilation-perfusion scan
    • Clinicians ranked patients as high (> = 80%), moderate (20-79%) or low risk < 20%) for PE based on history, examination, arterial blood gas, chest x-ray, ECG without knowledge of v/q or pulmonary angiogram result.

    The evidence

    pre-test probability of pulmonary embolism: 27%, (95% CI: 24% to 30%)
    pre-test probability of PE in high-risk patients: 68%, (95% CI: 58% to 77%)
    pre-test probability of PE in intermediate risk patients: 30%, (95% CI: 26% to 34%)
    pre-test probability of PE in low-risk patients: 9%, (95% CI: 5.4% to 13%)

    diagnostic test pulmonary embolism no pulmonary embolism LR
    (95% CI)    		 post-test probability
    high probability 102 22 13
    (8.1 to 19)     		82%
    intermediate probability 105 259 1.1
    (0.91 to 1.3)     		29%
    low probability 39 273 0.38
    (0.29 to 0.52)     		12%
    normal/ near normal scan 5 126 0.11
    (0.045 to 0.26)     		4%
    total 251 680


    diagnostic test PE in high-risk patients: V/Q scan result no PE LR
    (95% CI)    		 post-test probability
    high probability 28 1 13
    (1.9 to 93)     		97%
    intermediate probability 27 14 0.92
    (0.56 to 1.5)     		66%
    low probability 6 9 0.32
    (0.15 to 0.68)     		40%
    normal/ near normal scan 0 5 0.0
    (0.0 to 0.28)     		0%
    total 61 29


    diagnostic test PE in intermediate risk patients no PE LR
    (95% CI)    		 post-test probability
    high probability 70 10 16
    (8.7 to 31)     		88%
    intermediate probability 66 170 0.91
    (0.73 to 1.1)     		28%
    low probability 30 161 0.44
    (0.31 to 0.62)     		16%
    normal/ near normal scan 4 58 0.16
    (0.060 to 0.44)     		6%
    total 170 399


    diagnostic test PE in low-risk patients no PE LR
    (95% CI)    		 post-test probability
    high probability 5 4 12
    (3.6 to 42)     		56%
    intermediate probability 11 57 1.9
    (1.2 to 3.0)     		16%
    low probability 4 86 0.46
    (0.19 to 1.1)     		4%
    normal/ near normal scan 1 60 0.16
    (0.024 to 1.1)     		2%
    total 21 207


    diagnostic test PE no PE LR
    (95% CI)    		 post-test probability
    clinical assessment of risk- high 61 29 5.3
    (3.5 to 8.0)     		68%
    moderate 170 399 1.1
    (0.73 to 1.6)     		30%
    low 21 207 0.26
    (0.17 to 0.39)     		9%
    total 252 635

    Comments

    1. 150 patients with low probability or normal scans either did not have an angiogram or the diagnosis of PE was uncertain. On clinical follow-up, it was felt that none of patients had PE.
    2. 50 patients with high and intermediate probability scans did not have an angiogram, or the PE diagnosis was uncertain. Information on outcomes of these patients not included in paper, so it was assumed that they did not have PE when the above calculations were performed.
    3. PIOPED criteria have been reappraised since, but not formally assessed against gold standard of pulmonary angiogram .

    Citation

    1. The PIOPED Investigators , : Value of the ventilation/perfusion scan in acute pulmonary embolism: results of the prospective investigation of pulmonary embolism diagnosis (PIOPED). Journal of the American Medical Association 1990; 263 (20): 2753-2759
    Contributor: Chris Ball and Clare Wotton, October 2000
    Reviewer:

    Clinical Question.
    Patient suspected pulmonary embolism
    Intervention or Exposure ventilation-perfusion scan
    Outcome diagnosis