3D Printing Of The Heart: The Successor To Vesalius, Materialise Reveals
In a session today delegates will hear how printing 3-D models of the heart has become a clinical reality with cutting edge technology being used to create ‘off the shelf’ teaching models, patient-specific cardiovascular (CV) anatomy for assisting in the planning of complicated procedures, and for testing medical devices prior to clinical trials.
The creation of tangible 3-D printed models of the heart, explains Peter Verschueren, who will explore the potential for 3-D printing, represents the next logical step after the development of virtual 3-D models of the heart. “Five hundred years ago there was Vesalius who drew the heart from dissections, one hundred years ago X-rays were introduced, then we had virtual reconstructions of the heart from CT scans and now we’re able to recreate 3-D copies of the heart,” says Verschueren, the CV Team Leader at Materialise, a Belgium company specializing in all aspects of 3-D printing. “In the same way that Vesalius’ anatomical drawings improved understanding of the structure of the heart our models are adding to knowledge and improving clinical outcomes.”
While anatomical specimens provide the best possible way to understand spatial relationships and dimensions of cardiac pathology, in teaching continued use of such specimens leads to gradual deterioration. Materalise has developed a ‘library’ of over 100 off-the shelf educational models representing different pathologies including aortic aneurysms, different types of valve disease, and cerebral aneurysms. They also specialize in paediatric congenital models including Double Outlet Right Ventricle (DORV), Tetralogy of Fallot, and D-Transposition of the Great Arteries.
“Surgical students also use our models to practise making incisions to gain an appreciation of how this feels before going into the theatre for the first time,” says Verschueren. This experience, he adds, can be made even more realistic by having artificial blood pumped through the model. A feature of training ‘surgical’ models is that they can be opened up and have internal parts, such as valve leaflets, continually replaced.
In the ‘Heart Print® Service’ 3-D models of the heart and blood vessels that are bespoke to individual patients are being created by Materialise to provide supplemental information for planning complex surgery and other interventional procedures. “Creating a road map for surgery ultimately results in reduced time on the operating table which delivers the big advantage that patients will be less time on pump, thereby lessening risks of complications,” says Verschueren.
A recent use was the creation of a 3-D model for a patient scheduled for transcatheter aortic valve implantation (TAVI), who had an abnormally small aortic valve, to test whether the Edwards or Medtronic valve offered the best fit.
In R&D settings bench-top 3-D models of the heart are being used for testing novel devices prior to implanting them into patients. The use of 3-D models also would allow ‘bespoke’ cardiac devices to be created for individual patients. “This would be of particular value to position devices when treating aortic aneurysms to ensure you don’t block other lateral vessels,” says Verschueren.
The process of producing a bespoke 3-D model consists of three main phases. First the patient undergoes CT or MRI scanning, then the 3-D model of the heart is virtually reconstructed using the Mimics ® software from the medical images and finally the virtual model is exported to the printer.
Working in much the same way as an inkjet printer, 3-D printers ‘squirt’ out material in predetermined patterns along X,Y and Z axis points building up the model layer by layer, with each layer being around 100 microns thick. Printing hollow structures, such as the heart, poses particular challenges. “Our algorithm calculates where we need to incorporate internal ‘honeycomb’ like support structures on the inside of models to prevent them from collapsing inwards during the printing process,” explains Verschueren.
3-D models can be made from different materials including polyamides, polyurethane and acrylics that can be transparent, rigid or compliant with distensibility properties allowing vessels to expand under pressure. “For some models it’s helpful they’re transparent. For example when you are practicing deploying a stent it’s valuable to see positioning from the outside. In other models flexibility of the vessels is more important to make them feel as realistic as possible,” says Verschueren.
It is even possible to print a combination of flexible and rigid materials in a model to create calcification in anatomy, such as seen in the vessels and valves of older people. “This makes a dummy intervention as realistic as possible,” says Verschueren.
While Materialise would like to initiate trials exploring the efficacy, amount of operating time saved, number of repeat interventions prevented and cost effectiveness of using 3-D models to assist in pre-operative planning, such research is fraught with difficulty. “First it’s hard to recruit large enough numbers for meaningful trials, and second congenital heart disease is anatomically highly variable making it impossible to compare outcomes,” says Verschueren, adding that they have started a registry documenting all cases using the technology.
Undoubtedly the Holy Grail for 3-D printing would be bioprinting using the patient’s own cells. Investigators at Cornell University, led by Jonathan Butcher, have already used 3-D bioprinting technology to create 3-D heart valves from living tissue. In interviews, Butcher has said, the approach allows them to make patient-specific tissue models to learn about disease pathogenesis and screen drug efficacy, and tailor living tissue replacements directly to patient’s individual geometry.
Ultimately the hope is that printers would be able to produce blood vessels for graft during heart bypass surgery and even entire hearts for transplant. But first multiple obstacles need to be overcome including the ability to replicate highly vascularised and innervated functional tissue.
In September the Musica project, a partnership between Materialise and academic institutions including Politechnico di Milano and Imperial College, London, was awarded an EU grant for €3.8 million to develop 3-D bioprinting scaffolds. “We eventually hope to print scaffold structures resembling patient specific anatomies that can be injected with the patient’s own cells. Afterwards the scaffold would be reabsorbed,” explains Verschueren.
3-D models in clinical practice
In their clinical practice Gerald Greil and Tarique Hussain, two paediatric cardiologists running the Congenital Cardiac MRI Imaging Service at St Thomas’ Hospital and King’s College, London, have printed seven 3-D models of complex cardiac morphology to assist cardiac congenital surgeons with preoperative planning.
“Surgeons are finding nothing beats the one to one representation of holding a 3-D model in their hands. It stands to reason that having the opportunity to plan out strategies first prior to surgery leads to better outcomes,” explains Greil.
Surgeons use 3-D models to decide upfront whether procedures are feasible, something that in the past they were not able to appreciate until they opened the patient up. Recently, 3-D models revealed that surgery was practical for a patient with an atrioventricular septal defect (AVSD) with a double outlet right ventricle and an aorta remote from the AVSD; but not for a patient with congenital transposition requiring routing of the VSD to the aorta.
The process requires the whole heart to be imaged with MRI and captured during the still period – either the mid-diastolic period (used for ventricular septal defects) or the end systolic rest period (used for right ventricular outflow tract intervention).
Image acquisition, performed for young children under general anaesthesia, takes around seven minutes. In the next step, the image processing operator undertakes post-processing in which the blood pool is considered as a ‘virtual cast’ to make the 3-D model. “The quality control stage where we ensure the structures we’re interested in are properly represented is critical,” said Hussain, adding it usually takes around two hours for imagers to review the ’virtual’ models. “Even the smallest error would mislead the surgeon.”
To speed up production Greil and Hussain will install a 3-D printer in the department. Printing of the 3-D models, using acrylonitrile butadiene styrene (ABS) plastic, takes around 20 hours. “Practically we will set the machine to print over night,” says Hussain.
Authors: Janet Fricker
Hey, check out all the engineering jobs. Post your resume today!
The creation of tangible 3-D printed models of the heart, explains Peter Verschueren, who will explore the potential for 3-D printing, represents the next logical step after the development of virtual 3-D models of the heart. “Five hundred years ago there was Vesalius who drew the heart from dissections, one hundred years ago X-rays were introduced, then we had virtual reconstructions of the heart from CT scans and now we’re able to recreate 3-D copies of the heart,” says Verschueren, the CV Team Leader at Materialise, a Belgium company specializing in all aspects of 3-D printing. “In the same way that Vesalius’ anatomical drawings improved understanding of the structure of the heart our models are adding to knowledge and improving clinical outcomes.”
While anatomical specimens provide the best possible way to understand spatial relationships and dimensions of cardiac pathology, in teaching continued use of such specimens leads to gradual deterioration. Materalise has developed a ‘library’ of over 100 off-the shelf educational models representing different pathologies including aortic aneurysms, different types of valve disease, and cerebral aneurysms. They also specialize in paediatric congenital models including Double Outlet Right Ventricle (DORV), Tetralogy of Fallot, and D-Transposition of the Great Arteries.
“Surgical students also use our models to practise making incisions to gain an appreciation of how this feels before going into the theatre for the first time,” says Verschueren. This experience, he adds, can be made even more realistic by having artificial blood pumped through the model. A feature of training ‘surgical’ models is that they can be opened up and have internal parts, such as valve leaflets, continually replaced.
In the ‘Heart Print® Service’ 3-D models of the heart and blood vessels that are bespoke to individual patients are being created by Materialise to provide supplemental information for planning complex surgery and other interventional procedures. “Creating a road map for surgery ultimately results in reduced time on the operating table which delivers the big advantage that patients will be less time on pump, thereby lessening risks of complications,” says Verschueren.
A recent use was the creation of a 3-D model for a patient scheduled for transcatheter aortic valve implantation (TAVI), who had an abnormally small aortic valve, to test whether the Edwards or Medtronic valve offered the best fit.
In R&D settings bench-top 3-D models of the heart are being used for testing novel devices prior to implanting them into patients. The use of 3-D models also would allow ‘bespoke’ cardiac devices to be created for individual patients. “This would be of particular value to position devices when treating aortic aneurysms to ensure you don’t block other lateral vessels,” says Verschueren.
The process of producing a bespoke 3-D model consists of three main phases. First the patient undergoes CT or MRI scanning, then the 3-D model of the heart is virtually reconstructed using the Mimics ® software from the medical images and finally the virtual model is exported to the printer.
Working in much the same way as an inkjet printer, 3-D printers ‘squirt’ out material in predetermined patterns along X,Y and Z axis points building up the model layer by layer, with each layer being around 100 microns thick. Printing hollow structures, such as the heart, poses particular challenges. “Our algorithm calculates where we need to incorporate internal ‘honeycomb’ like support structures on the inside of models to prevent them from collapsing inwards during the printing process,” explains Verschueren.
3-D models can be made from different materials including polyamides, polyurethane and acrylics that can be transparent, rigid or compliant with distensibility properties allowing vessels to expand under pressure. “For some models it’s helpful they’re transparent. For example when you are practicing deploying a stent it’s valuable to see positioning from the outside. In other models flexibility of the vessels is more important to make them feel as realistic as possible,” says Verschueren.
It is even possible to print a combination of flexible and rigid materials in a model to create calcification in anatomy, such as seen in the vessels and valves of older people. “This makes a dummy intervention as realistic as possible,” says Verschueren.
While Materialise would like to initiate trials exploring the efficacy, amount of operating time saved, number of repeat interventions prevented and cost effectiveness of using 3-D models to assist in pre-operative planning, such research is fraught with difficulty. “First it’s hard to recruit large enough numbers for meaningful trials, and second congenital heart disease is anatomically highly variable making it impossible to compare outcomes,” says Verschueren, adding that they have started a registry documenting all cases using the technology.
Undoubtedly the Holy Grail for 3-D printing would be bioprinting using the patient’s own cells. Investigators at Cornell University, led by Jonathan Butcher, have already used 3-D bioprinting technology to create 3-D heart valves from living tissue. In interviews, Butcher has said, the approach allows them to make patient-specific tissue models to learn about disease pathogenesis and screen drug efficacy, and tailor living tissue replacements directly to patient’s individual geometry.
Ultimately the hope is that printers would be able to produce blood vessels for graft during heart bypass surgery and even entire hearts for transplant. But first multiple obstacles need to be overcome including the ability to replicate highly vascularised and innervated functional tissue.
In September the Musica project, a partnership between Materialise and academic institutions including Politechnico di Milano and Imperial College, London, was awarded an EU grant for €3.8 million to develop 3-D bioprinting scaffolds. “We eventually hope to print scaffold structures resembling patient specific anatomies that can be injected with the patient’s own cells. Afterwards the scaffold would be reabsorbed,” explains Verschueren.
3-D models in clinical practice
In their clinical practice Gerald Greil and Tarique Hussain, two paediatric cardiologists running the Congenital Cardiac MRI Imaging Service at St Thomas’ Hospital and King’s College, London, have printed seven 3-D models of complex cardiac morphology to assist cardiac congenital surgeons with preoperative planning.
“Surgeons are finding nothing beats the one to one representation of holding a 3-D model in their hands. It stands to reason that having the opportunity to plan out strategies first prior to surgery leads to better outcomes,” explains Greil.
Surgeons use 3-D models to decide upfront whether procedures are feasible, something that in the past they were not able to appreciate until they opened the patient up. Recently, 3-D models revealed that surgery was practical for a patient with an atrioventricular septal defect (AVSD) with a double outlet right ventricle and an aorta remote from the AVSD; but not for a patient with congenital transposition requiring routing of the VSD to the aorta.
The process requires the whole heart to be imaged with MRI and captured during the still period – either the mid-diastolic period (used for ventricular septal defects) or the end systolic rest period (used for right ventricular outflow tract intervention).
Image acquisition, performed for young children under general anaesthesia, takes around seven minutes. In the next step, the image processing operator undertakes post-processing in which the blood pool is considered as a ‘virtual cast’ to make the 3-D model. “The quality control stage where we ensure the structures we’re interested in are properly represented is critical,” said Hussain, adding it usually takes around two hours for imagers to review the ’virtual’ models. “Even the smallest error would mislead the surgeon.”
To speed up production Greil and Hussain will install a 3-D printer in the department. Printing of the 3-D models, using acrylonitrile butadiene styrene (ABS) plastic, takes around 20 hours. “Practically we will set the machine to print over night,” says Hussain.
Authors: Janet Fricker
Hey, check out all the engineering jobs. Post your resume today!