Pulmonary vascular disease & thromboembolism
Diagnostic management of acute pulmonary embolism: a prediction model based on a patient data meta-analysis
Authors' conclusion
The present model provides an absolute, individualized probability of PE presence in a broad population of patients with suspected PE, with very good discrimination and calibration.
Study context
- Population
- Adults with clinically suspected acute pulmonary embolism (PE) in primary care, emergency care, hospital and nursing homes
- Participants
- 28,305 adults · 16 studies
- Comparison
- Prediction model vs Wells score plus D-dimer algorithms
- Outcome
- Objectively confirmed PE at baseline or VTE during follow-up
- Follow-up
- 30 to 90 days
Methodological detailsprediction model development and validation, 16 cross-sectional or prospective studies, multilevel logistic regression, multiple imputation, internal-external cross-validation
- Prediction model development and validation
- 16 cross-sectional or prospective studies
- Multilevel logistic regression
- Multiple imputation
- Internal-external cross-validation
Pooled result
As reported in the abstract
Discrimination and calibration (internal-external cross-validation)
D-dimer threshold based on age
0.73
c-statistic · 95% CI 0.70–0.75
Structured clinical pretest probability
0.79
c-statistic · 95% CI 0.76–0.81
- Current model: pooled c-statistic
- 0.87 (0.85-0.89; 0.77-0.93) (95% CI; 95% PI)
- Current model: pooled O:E ratio
- 0.99 (0.87-1.14; 0.55-1.79) (95% CI; 95% PI)
Original publication
Diagnostic management of acute pulmonary embolism: a prediction model based on a patient data meta-analysis. Eur Heart J. Published 15 July 2023. doi:10.
Summarised from the published abstract; every number is checked against it