Cohort study
Impact of Preoperative Temporary Mechanical Circulatory Support on Durable Left Ventricular Assist Device (LVAD) Outcomes
- Retrospective cohort study
- registry-based (STS Intermacs)
- multivariable proportional hazards model
- propensity-score matched
Authors' conclusion
Use of HF-MP and IABP were associated with improved survival and equivalent rate of stroke compared to no tMCS.
Study context
- Population
- Adults receiving a fully magnetically levitated LVAD as Intermacs profiles 1–3, 2020 to 20243,863 (66.7%) on temporary mechanical circulatory support (tMCS)
- Participants
- 5,787 adults
- Exposure
- Preoperative tMCS vs no tMCS before durable LVAD
- Outcome
- Mortality and rates of stroke
- Follow-up
- 1 and 2 years
Main finding
As reported in the abstract
Survival
ECMO
70.6%
1-year survival · 2-year survival 67.1%
Intra-aortic balloon pump (IABP)
85.7%
1-year survival · 2-year survival 81.2%
HF-MP
82.9%
1-year survival · 2-year survival 77.5%
No tMCS
82.4%
1-year survival · 2-year survival 77%
Each group vs no tMCSHazard ratio (95% CI)P value
HF-MP
0.71(0.59–0.84)
P = 0.001IABP
0.74(0.63–0.86)
P = 0.001Observational data: associations, not effects.
Further findings
| Outcome | No tMCS | IABP | HF-MP | Extracorporeal membrane oxygenation | P value |
|---|---|---|---|---|---|
| Survival, HF-MP vs no tMCS (PSM cohort) · 1 and 2 years · matched cohort | P = 0.009 | ||||
| 1-year survival | 80.5% | Not reported in the abstract | 85.2% | Not reported in the abstract | |
| 2-year survival | 74.3% | Not reported in the abstract | 79.9% | Not reported in the abstract | |
| Freedom from stroke, HF-MP vs no tMCS (PSM cohort) · 1 year · similar · matched cohort | 93.9% | Not reported in the abstract | 93.1% | Not reported in the abstract | P = 0.33 |
Use of tMCS over the study period
- HF-MP use increased (19.1% to 35.1%)
- P < 0.001
- IABP use decreased (41.2% to 27.3%)
- P < 0.001
Original publication
Impact of Preoperative Temporary Mechanical Circulatory Support on Durable LVAD Outcomes. JACC Heart Fail. Published 8 October 2025. doi:10.1016/j.jchf.2025.102671
Summarised from the published abstract; every number is checked against it