Related Experiment Video
Updated: Jun 16, 2026

Three-Dimensional Modeling of the Left Atrium and Pulmonary Veins with a Precise Intracardiac Echocardiography Approach
Published on: June 30, 2023
A 3-dimensional modeling workflow to guide left pulmonary artery reimplantation
Dominic P Recco1,2, Shannen B Kizilski1,2, Noah E Schulz1
1Department of Cardiac Surgery, Boston Children's Hospital, Boston, Mass.
Objective:
Patients with a left pulmonary artery (LPA) sling often require surgical reimplantation to address pulmonary flow maldistribution and tracheal compression. We developed a virtual modeling workflow to determine what length of LPA should be retained for reimplantation and provided these data for intraoperative guidance.
Methods:
From March 2024 to May 2025, 5 patients diagnosed with an LPA sling with (n = 3) or without (n = 2) tracheal stenosis underwent surgical LPA reimplantation with or without slide tracheoplasty at median age 1.8 years [range, 0.4-3.6 years]. All patients had preoperative surgical planning using measurements from a segmented 3-dimensional model using a digital workflow. Model measurements included distances from the main pulmonary artery reimplantation site to nearby landmarks and the length of pressurized, prestretched LPA that should be retained for reimplantation based on the distance between the reimplant site and anticipated extent of distal LPA mobilization. These measurements were used as intraoperative guidance.
Results:
The planned main pulmonary artery reimplant site varied across surgeons, and recommended LPA length ranged from 42% to 71% of the total native length. Model recommendations were closely followed in all patients with LPA sling. Median hospital length of stay was 7 days [6-15]. At discharge, 3 patients had no/trivial LPA stenosis and 2 had mild (peak gradient 23 mm Hg [20, 26]). At median follow-up of 17.6 months [6.5-22.1], all patients had no/trivial stenosis and were asymptomatic with 100% freedom from reintervention. Postoperative lung-perfusion and computed tomography scans demonstrated left/right split of 40/60% ± 2% (n = 3) and LPA diameter z score of -0.23 [-0.82, +0.31] (n = 2).
Conclusions:
Three-dimensional models provide a platform for quantitative surgical planning to help achieve targeted reconstruction with good initial clinical follow-up.

