Clinical subtyping of severe acute pancreatitis reveals heterogeneous associations with early management strategies: a multicenter retrospective cohort study
Authors' conclusion
This study identified three SAP subphenotypes with significantly heterogeneous clinical features and prognoses based on early routine indicators and developed a high-performance classifier for bedside use.
Study context
- Population
- Patients with severe acute pancreatitis (SAP), Chinese Jiangxi cohort and US MIMIC-IV cohortJiangxi n = 3,022
- Participants
- 1,446 / 317 / 1,259
- Exposure
- Clinical subphenotypes K2 and K3 vs K1
- K1n = 1,446reference
- K2n = 317
- K3n = 1,259
- Outcome
- In-hospital mortality by subphenotype
Methodological detailsconsensus K-means clustering, external validation, multivariable Cox regression
- Consensus K-means clustering
- External validation
- Multivariable Cox regression
Main finding
As reported in the abstract
P < 0.001
Observational data: associations, not effects.
Further findings
| Outcome | aHR for death (95% CI) | P value | |
|---|---|---|---|
| 28-day mortality, K2 vs K1 (multivariable Cox) | 6.70 (4.94–9.08) | ||
| LMWH therapy and mortality, K2 (subgroup)LMWH vs No LMWH · survival benefit only in K2 | 0.66 (0.45–0.96) | P = 0.031 | |
| 0.10.31310AHR for death (log scale) |
| Outcome | Training set | Test set | External validation set |
|---|---|---|---|
| XGBoost subphenotype classifier discrimination | |||
| Macro-average AUC | 0.997 | 0.994 | 0.958 |
Fluid overload (total intake ≥ 2.5 L within the first days) and mortality, K3 (subgroup)
- Fluid overload significantly associated with increased mortality specifically in K3
Original publication
Clinical subtyping of severe acute pancreatitis reveals heterogeneous associations with early management strategies: a multicenter retrospective cohort study. Crit Care. Published 2 April 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →