Heartflow, Inc. said its P4 trial found coronary CT angiography (CCTA)-planned percutaneous coronary intervention (PCI) took a median 17 minutes less than intravascular ultrasound (IVUS)-guided PCI, with 12-month failure rates of 4.1% and 4.6%, respectively.
The investigator-initiated study was supported by Heartflow, according to the company. It enrolled 1,105 people with stable coronary artery disease at 23 centers in the United States, Europe and Japan. The CT strategy used scans taken before PCI, while clinicians in the comparison group used IVUS during the procedure.
Median procedure time was 46 minutes with CCTA planning and 63 minutes with IVUS. Median contrast use was 150 milliliters and 170 milliliters, respectively. Heartflow said advanced calcium-modification techniques were used in 9.2% of CT-guided procedures and 3.8% of IVUS-guided procedures.
Heartflow described the CT strategy as noninferior on the primary outcome. The P4 design paper groups cardiac death, target-vessel myocardial infarction and ischemia-driven target-vessel revascularization under the 12-month target-vessel failure measure. The company’s announcement gave the event rates but did not provide confidence intervals, a p-value or the noninferiority margin.
P4 used Heartflow FFR CT Analysis and Heartflow Planner, which preceded PCI Navigator. The trial therefore did not test the current Navigator product.
In a separate 203-patient NAVIGATE-PCI registry cohort, Heartflow said clinicians changed their plans in about 70% of cases after reviewing PCI Navigator. The company listed stent length, at 41%, and lesion preparation, at 12%, among the most common changes, but did not state the denominator for those percentages. The observational registry collects procedural and in-hospital information without planned post-discharge follow-up, so the reported changes do not show whether patient outcomes improved.
Enrollment figures vary across records. Heartflow’s announcement counted 1,105 participants, a January 2026 investor presentation described 1,104, and ClinicalTrials.gov lists estimated enrollment of 1,090. The trial record identifies CoreAalst BV as sponsor and showed no results posted in its Jan. 21, 2026, update.
