Building Models

When I was a boy, I loved building models. Over the span of a few years, I covered every available horizontal surface with replicas of fighter planes, battleships, galleons, a gruesome assortment of horror movie monsters, and the definitive collection of every rocket in the United States’s fledgling space program. Using a variety of construction toys, I created a vast succession of buildings, robots, and amusement park rides in wood, plastic, and stainless steel. I suspect that many of my future orthopaedic colleagues shared these childhood hobbies, and I know that some have continued to pursue such passions into adulthood. During a recent visit to the home of our chairman, Dave Sisk, I was awestruck by the display of large-scale, fully fitted-out and rigged sailing ships that Dave has constructed during countless hours of jealously guarded “spare time.” Grown-up orthopaedic researchers love to build models too. Experimental models allow researchers to isolate and manipulate parameters in a controlled fashion that would be impossible, or at least unethical, in patients. This issue of AJSM, like most, contains reports of a number of investigations that utilize experimental models. As useful as they are, these models carry their own limitations and ethical concerns. The in vitro cadaver model has been used extensively in orthopaedic sports medicine research. Biomechanical studies in human cadaver joints constituted much of the initial scientific inquiry conducted in our field and have greatly contributed to our knowledge of joint function. “The Effect of Lateral Meniscus Allograft Sizing on Contact Mechanics of the Lateral Tibial Plateau” illustrates the use of this type of model. One advantage of cadaver models is that they employ actual human anatomy and tissues. The experiments are conducted in a carefully regulated environment, in which forces can be applied in a controlled manner. Cadaver models also have a number of weaknesses. Specimens are donated by individuals who were often elderly, arthritic, or inactive at the time of death. Since both age and inactivity can affect tissue properties, these joints do not duplicate those of the younger patients who are often most pertinent to the topics being investigated. Forces must be artificially applied, and it is difficult to verify that all relevant in vivo forces are being reproduced. The weakened state of the available tissues usually precludes using forces of physiologic magnitude. When models of this type are employed to investigate surgical procedures, only “time zero” data are obtained, since healing does not occur. The expense and limited availability of human cadaver joints often results in experiments that depend on a small number of specimens with consequently limited statistical power. Over the years, orthopaedic researchers have strived to improve these models. These improvements have included the addition of forces to simulate muscle contraction and weightbearing. Models have become less constrained, allowing more degrees of freedom of motion. Cyclic loading has been introduced to complement single load-to-failure testing. The rarity of young human cadaver joints has led some researchers to substitute readily available specimens from large food animals. Other researchers have criticized the use of animal bones, which may match the density of young human bone and yet possess structural differences that lead to discrepant results. These findings have led other scientists to propose the use of foam-reinforced bone from elderly humans as a better approximation of the properties of young human bone. The second model that is popular in orthopaedic research is the in vivo animal model.These investigations may involve small animals, typically rabbits or rats, or large animals, such as sheep, goats, or even horses. “Hydrogel Meniscus Replacement in the Sheep Knee” is one such study. These living animal models permit the regulation and manipulation of experimental factors in a controlled manner that might be impossible in humans. The progression of healing or degeneration after an intervention can be monitored over time. In vivo animal studies also have inherent limitations. Although animal anatomy can be strikingly similar to that of humans, there are often important differences. Most animal models utilize quadrupeds, whose joints may differ in weight distribution and functional range of motion from their human counterparts. Physiology, particularly with regard to healing potential, may vary substantially from one species to another. Compliance with postoperative activity restrictions, problematic enough in many human patients, is nearly unattainable in animals. Ethical constraints also restrict experimental investigations in live animals. Finally, the expense of keeping large animals may result in low statistical power, inadequate study length, or the use of a less appropriate but cheaper small animal model. The in vivo human laboratory model, typified by “Pitching Biomechanics as a Pitcher Approaches Muscular Fatigue During a Simulated Baseball Game,” allows investigators to study an approximation of normal human function. Studies of this type have been commonly used to investigate the forces that might predispose certain athletes to injury, particularly involving the anterior cruciate ligament. The laboratory setting permits more precise DOI = 10.177/0363546503262911