Description
Congenital heart disease (CHD) remains the most common type of major congenital malformation
and the leading cause of mortality from birth defects [1-4]. Advances in effective treatment
for these lesions have significantly extended the lifespan of affected patients, especially
for the most complex subtypes of disease. However, these patients are at higher risk of heart
failure (HF) secondary to longer life expectancy. This includes patients with a systemic
right ventricle and a single ventricle circulation palliated by a Fontan procedure [5, 6]. HF
has been documented in up to 30% of patients with a systemic right ventricle and 40% of
patients who have had a Fontan procedure [7].
Ventricular assist devices (VAD) are implanted in patients with HF to improve cardiac output
and prolong life. They remain underutilized in patients with CHD and HF in part due to the
highly variable anatomy in this population. This is true despite outcomes having been shown
to be the same for VAD placement in patients with and without CHD [8-10]. In the absence of
VAD placement, however, wait list mortality for patients with CHD is higher than for those
patients without CHD [11, 12].
Advances in imaging techniques have allowed early diagnosis of CHD as well as anatomic
assessment prior to surgical procedures. Given the significant yet often subtle anatomic
differences between CHD patients, it is a substantial challenge to thoroughly depict all of
the components of a complex patient's cardiac anatomy in a two dimensional imaging dataset.
An innovative technology that is being used with more enthusiasm in the medical field, is
three-dimensional (3D) printing. Our research team has previously reported on the best
technique that should be used to create 3D printed cardiac models from MRI and the subtypes
of complex CHD's for which 3D printing should be utilized [13-16]. 3D printing allows
creation of patient specific physical anatomic models from a patient's own imaging data.
These models provide a physical guide to patient-specific anatomic features that often make
VAD and cannula placement challenging in patients with CHD [17]. Factors such as complex
cardiac anatomic malformations, heavy trabeculations or a severely dilated ventricle can
distort the usual anatomic landmarks used to identify the best position for cannula
placement. Our primary goal is to establish the utility of this advanced imaging technique,
which provides a much more comprehensive understanding of complex congenital cardiac anatomy.
We hypothesize that 3D printed models will allow more informed preoperative planning with a
clearer understanding of the best site for inflow and outflow cannula and VAD placement
leading to better surgical preparedness, less operating room time and improved patient
outcomes.
AIM 1: To assess if a 3D printed cardiac model improves perceived visualization of VAD and
cannula placement sites in CHD-HF patients as compared to 2D imaging. We will prospectively
enroll CHD-HF patients at multiple centers and randomize to Group A (3D printed models will
be used for pre-VAD planning) or Group B (no model-controls). For both Groups, all of the
cardiothoracic surgeons at the participating center will complete a questionnaire after
reviewing 2D imaging data. For Group A, a survey will also be administered after reviewing a
patient specific 3D model. Our primary outcome measure will be better perceived visualization
of cannula and VAD sites. We hypothesize that the 3D model will more clearly demonstrate
sites of cannula and VAD placement as compared to 2D imaging.
AIM 2: To determine if perioperative factors and outcomes improve in CHD-HF patients with use
of a 3D printed model versus traditional imaging in VAD placement planning. Clinical
characteristics will be collected at time of enrollment including primary diagnosis and
indication for VAD. After VAD placement, information regarding the intraoperative and
postoperative course will be collected including surgical cardiopulmonary bypass time (CPB)
and need for cannula repositioning. Longer CPB increases morbidity and mortality and is
associated with intensive care readmission in patients after LVAD placement [18-20]. Our
primary measures of improvement will be CPB. We hypothesize that the improved preoperative
planning using 3D models will lead to a decrease in CPB time.
The skill with which we assess patient specific CHD anatomy for pre-procedural planning must
be improved, especially for the most complex patients. To confirm the clinical benefit of 3D
printed models in pre-surgical planning and justify their use in routine care, multicenter
clinical trials must be conducted. As an expert in the field of 3D imaging in cardiac
disease, I am well poised to lead this body of research. My goal is to become well versed in
conducting high quality multicenter studies and to become facile in survey tool design
through this K23 proposal. I will then design a prospective multicenter study for an
independent R01 proposal focused on assessing the utility of 3D models in pre-procedural
planning for all complex congenital heart diseases. Investigating and reporting on these
findings will result in a paradigm shift in what we consider "standard of care" for advanced
imaging offered to our most complex CHD patients.