MRI assessment of pulmonary vein stenosis predicts outcomes

Key clinical point: Survival after pulmonary vein stenosis repair was adversely affected by smaller upstream cross-sectional area indexed to body surface area.

Major finding: Smaller upstream total cross-sectional area indexed for body surface area (P = .30) and greater number of stenotic pulmonary veins (P = .007) were associated with increased early risk of death.

Data source: Researchers reviewed the outcomes of 31/145 patients who underwent surgical repair of pulmonary stenosis who had preoperative computed tomography and magnetic resonance imaging between 1990 and 2012.

Disclosures: The authors reported that they had no conflicts of interest.

View on the News

Need for multi-institutional registry

“The Toronto group has contributed significantly to our knowledge and management of pulmonary vein stenosis during the past decade. This article by Dr. Lo Rito and coworkers continues that contribution by reinforcing the values of MRI in imaging PVs before intervention and providing a valuable “hint” that preoperative PV size measurements are related to outcome,” Dr. William M. DeCampli wrote in his invited commentary (J Thorac Cardiovasc Surg. 2016;1510:667-8).

“The task of definitively demonstrating this relationship is daunting for any single institution, however, because 1) PVS is relatively rare, 2) MRI and computed tomography are relatively recently used diagnostic modalities, and 3) MRI is not easily used in an important subset of the cohort, small infants.” This limited the study to a small number of covariates,” noted Dr. DeCampli, and prevented the researchers from taking into account a myriad of additional covariates commonly associated with survival in complex congenital heart disease.

Dr. William M. DeCampli

Such covariates included in a sufficiently large model could significantly alter the observed odds ratios otherwise calculated for the included variables in this study, he added, citing a study of PVS by Boston Children’s Hospital (J Thorac Cardiovasc Surg. 2015;150:911-7), which found a different set of covariates associated with death; in that case, age younger than 6 months at operation, weight less than 3 kg at operation, and lesser preoperative right ventricular systolic pressure.

“The challenges in studying PVS encountered by these two high-volume, research-oriented programs leads us to suggest that PVS should be studied in a different way. Perhaps it is time to consider a multi-institutional, mixed or inception cohort registry for PVS. The spring 2015 Society of Thoracic Surgeons Congenital Heart Database report lists 506 cases of PVS repair as the primary procedure between January 2011 and December 2014. If a study were to enroll just one-third of these subjects it would accrue more than 40 subjects per year. Five years hence with an anticipated 50-80 events (deaths), it would be possible to carry out more robust risk-hazard analyses,” Dr. DeCampli suggested.

Dr. DeCampli is a congenital heart surgeon at the department of clinical sciences, University of Central Florida, and the Heart Center at Arnold Palmer Hospital for Children, both in Orlando. He reported having no conflicts.




A retrospective analysis of children who underwent pulmonary vein stenosis repair with preoperative computed tomography and magnetic resonance imaging from 1990 to 2012 showed that smaller upstream or downstream total cross-sectional area indexed (TCSAi) for body surface area led to poorer survival.

The study of 31 patients at a single institution also indicated that early survival seemed especially poor for patients with a greater number of stenotic veins and upstream pulmonary vein (PV) involvement. The study was published in the March issue of the Journal of Thoracic and Cardiovascular Surgery.

Dr. Mauro Lo Rito and his colleagues at The Hospital for Sick Children, Toronto, retrospectively assessed the 31 patients out of 145 who underwent surgical repair who had had preoperative CT and MRI imaging. Complete sutureless repair was done in 18 (58%), single-side sutureless repair in 12 (39%), and pericardial patch reconstruction in 1 (3%). The mean follow-up was 4.3 years; the median patient age at time of operation was 226 days. Stenosis was bilateral in 45% of patients and unilateral in 55 (J Thorac Cardiovasc Surg. 2016;151:657-66).

In-hospital mortality was 9.7%, with an overall survival of 75%, 69%, and 64% at 1, 3, and 5 years, respectively. Univariate analysis showed that a younger age at operation, lower body surface area, smaller upstream TCSAi, and greater number of PV with stenosis/occlusion were associated with an increased risk of death.

Multivariate analysis showed that smaller upstream TCSAi for body surface area (P = .030) and greater number of stenotic PVs (P = .007) were associated with poor early (less than 1 year) survival. There was a nonsignificant tendency for smaller downstream TCSAi to be associated with poor late survival (greater than 1 year). None of the different PV morphologies were found to influence survival, according to Dr. Lo Rito and his colleagues.

Among the 28 hospital survivors, restenosis occurred in 10 patients, 7 of whom did not undergo further surgery (3 of these were alive at last follow-up and 4 died secondary to disease progression). Of the 3 patients who underwent subsequent intervention, 2 were alive at last follow-up.

“Risk stratification for patients with PV stenosis is currently challenging because of the variability in the anatomic configuration and the unknown relationship between these anatomic variants and survival. Our study demonstrates that by using cross-sectional areas, pulmonary vein cross-sectional area indexed to body surface area (PVCSAi) and TCSAi and tabulating the number of stenotic PVs, we can identify high-risk subsets of patients with high predicted mortality.” Dr. Lo Rito and his colleagues stated.

“The upstream total cross-sectional area and the number of stenotic PVs influence early survival and can be used to guide counseling. Smaller downstream cross-sectional area influences late survival, and those patients should be monitored with close follow-up. This methodology could aid in risk stratification for future clinical trials of pharmacologic agents designed to target upstream pulmonary vasculopathy,” the investigators concluded.

The authors reported that they had no conflicts of interest.

A webcast of the original presentation of these results at the 95th American Association for Thoracic Surgery Annual Meeting is available online (

Next Article:

   Comments ()