Existing Solutions
Human Models
Medical device companies may test their products on cadavers from medical schools. However, the sphincters within the cadavers are closed, which prevents feeding tubes from being inserted past the pylorus, or the lower part of the stomach. This obstacle makes tube placement testing on cadavers fruitless.
Testing on live human subjects is extremely costly and is, therefore, not appropriate for testing early in the design process.
Animal Models
Live animal models can serve as useful tools for testing the effectiveness of a feeding tube design before the product goes to market. Although animal models may prove to be useful in the later stages of product development, companies have a need for a testing model that can be used in earlier stages of development before progressing to animal studies.
Medical companies have used porcine, chimp, and canine models in feeding tube placement tests. Swine are often used as models in studies involving the GI system due to the similarities between swine and humans in the physiology of digestion. Additionally, porcine and human GI tissues are structurally similar. However, one disadvantage of this model is the necessity of oral, rather than nasal, placement due to anatomical differences between the human and porcine nasal cavity. In addition, the pylorus is higher in the pig than in a human, which makes the model an inaccurate anatomical simulation of the human GI system.
Chimp models have been explored, but nasal placement causes inflammation in the tissues, and this trauma prevents multiple tube placements in succession. In addition, some medical companies have now halted testing on primate models due to company policies.
A canine model is more similar to the human anatomy than the pig model, and these methods are being explored further for testing.
Biological/Non-biological Hybrid Models
Some GI models combine the use of animal tissue with manmade materials. These models could be an excised GI tract from an animal model that is forced to be in a specific shape through the use of a plastic housing unit.
In a feasibility study of an endoscopic capsule by Valdastri, a lower gastrointestinal (LGI) tract phantom model was used. The Valdastri model consists of both manmade materials and porcine tissue. The phantom was based on a standard training phantom with manmade abdominal, chest, and pelvic cavities. These cavities contained “fixtures aligned in the shape of human mesenteries for the attachment of ex vivo animal intestine.” In the Valdastri study, porcine colon was used. Although this lower GI model is useful for the testing of endoscopic capsules, it would most likely be less useful for testing nasojejunal feeding tubes which need to pass through parts of the upper GI system like the stomach and duodenum.
Non-Biological Models
Numerous manmade GI models exist, with varied properties and uses. These models have the advantage that they are generally less expensive, are more reusable, and are more portable.
One existing models of interest is the EMS Trainer by The Chamberlain Group, pictured on the left. This silicone model simulates the lumens of the stomach and duodenum, a portion of the colon, and the esophagus in one apparatus. Although this model is portable and geometrically accurate, the material does not accurately model the elasticity of the stomach or the mucus properties in the stomach and duodenum. Additionally, it does not model the lumen of the jejunum.
The Tube Feeding Simulator is another GI model training tool used to practice the insertion of naso/orogastric feeding tubes as well as the insertion of percutaneous endoscopic gastrostomy tubes. The manikin is transparent, so proper tube placement can be visually confirmed. It is also equipped with an opaque outer covering to practice proper placement confirmation by auscultation. Although this model may be sufficient for testing naso/orogastric feeding tube placement, it does not allow for the testing of nasojejunal feeding tube placement.
The Otto Ostomy by VATA Anatomical Healthcare Models is another GI Model designed to aid in the understanding of ileostomies, colostomies, uterosigmoidostomies, and tube cecostomies. This model is based on a CT scan from an actual patient and is therefore anatomically similar to a typical human patient. However, the Otto Ostomy also does not allow for the practice or testing of feeding tube placement due to the lack of anatomic mimicking materials.