Perioperative medicine & outcomes
Prediction models for postoperative pulmonary complications: a systematic review and meta-analysis
Authors' conclusion
Many prediction models postoperative pulmonary complications have been developed, but the clinical utility of the vast majority remains uncertain.
Study context
- Population
- Studies developing or validating prediction models for postoperative pulmonary complications (PPCs) that reported c-statistic (Cochrane, Embase, PubMed up to June 2024)14 models in meta-analysis
- Participants
- 1,004,029 participants
- Comparison
- ARISCAT PPC score vs Xue's model vs CARDOT score vs DAGDA score vs Wang's model vs Jin's model vs Yoon's model vs Nizamuddin's model
- Outcome
- c-statistic of prediction models for composite PPCs
Methodological detailsBayesian meta-analysis, CHARMS data extraction, PROBAST risk of bias
- Bayesian meta-analysis
- CHARMS data extraction
- PROBAST risk of bias
Pooled result
As reported in the abstract
Composite PPCs, summary c-statistic
ARISCAT PPC score
0.76
summary c-statistic · 95% CI 0.67–0.86
Xue's model
0.82
summary c-statistic · 95% CI 0.75–0.89
CARDOT score
0.73
summary c-statistic · 95% CI 0.61–0.85
- C-statistic of all models ranged (median)
- 0.614 to 0.996 (median 0.80)
- Models self-reporting good (c-statistic >0.8) discrimination
- 50%
Further pooled results
| Outcome | Values |
|---|---|
| Postoperative pneumonia, summary c-statistic | DAGDA score summary c-statistic 0.81 (95% CI 0.74–0.88) · Wang's model summary c-statistic 0.78 (95% CI 0.70–0.86) · Jin's model summary c-statistic 0.75 (95% CI 0.68–0.82) |
| Postoperative respiratory failure, summary c-statistic | Yoon's model summary c-statistic 0.90 (95% CI 0.84–0.96) · Nizamuddin's model summary c-statistic 0.85 (95% CI 0.78–0.92) |
Risk of bias (PROBAST)
- Models assessed as high risk of bias
- 90.2%
Original publication
Prediction models for postoperative pulmonary complications: a systematic review and meta-analysis. Br J Anaesth. Published 4 June 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →