Department of Neurosurgery, University of Minnesota Hospital System, Minneapolis, Minnesota
The history of spinal biomechanics has its origins in antiquity. The Edwin Smith surgical papyrus, an Egyptian document written in the 17th century BC, described the difference between cervical sprain, fracture, and fracture-dislocation. By the time of Hippocrates (4th century BC), physical means such as traction or local pressure were being used to correct spinal deformities but the treatments were based on only a rudimentary knowledge of spinal biomechanics. The Renaissance produced the first serious attempts at understanding spinal biomechanics. Leonardo da Vinci (14521519) accurately described the anatomy of the spine and was perhaps the first to investigate spinal stability. The first comprehensive treatise on biomechanics, De Motu Animalium, was published by Giovanni Borelli in 1680, and it contained the first analysis of weight bearing by the spine. In this regard, Borelli can be considered the "Father of Spinal Biomechanics." By the end of the 19th century, the basic biomechanical concepts of spinal alignment and immobilization were well entrenched as therapies for spinal cord injury. Further anatomic delineation of spinal stability was sparked by the anatomic analyses of judicial hangings by Wood-Jones in 1913. By the 1960s, a two-column model of the spine was proposed by Holdsworth. The modern concept of Denis' three-column model of the spine is supported by more sophisticated testing of cadaver spines in modern biomechanical laboratories. The modern explosion of spinal instrumentation stems from a deeper understanding of the load-bearing structures of the spinal column.
(Neurosurgery, 39:657669, 1996)
Key words: Biomechanics, Borelli, Hanging, Hippocrates, History, Spinal devices, Spinal injury, Spine
Spinal biomechanics has a long and fascinating history. For the sake of convenience, its history can be divided into three eras. The ancient era was characterized by only a rudimentary knowledge of spinal anatomy and biomechanics. The premodern era was ushered in by the Renaissance, which brought with it an understanding of the anatomy of the spine and the first detailed inquiries into its mechanical nature. The modern era is characterized by the application of biomechanical principles to the treatment of spinal injuries and deformities. Landmarks in the history of spinal biomechanics are listed in Table 1. To properly trace the history of spinal biomechanics, the relevant history of spinal anatomy and the management of spinal trauma must be discussed because spinal biomechanics is so inextricably entwined in the latter two subjects.
Approximately 500 years later, many words in the papyrus became archaic and a series of glosses that defined difficult words and phrases was added by yet another unknown physician. This revised document was then copied in the 17th century BC by a Theban scribe, a master of penmanship who was more concerned with the beauty of the document than its content. Unfortunately, the scribe did not finish the copy; he abruptly stopped in the middle of a case, in the middle of a sentence, in the middle of a word. The papyrus systematically covers injuries to the body and starts with the head and neck. Sadly, the ancient copyist stopped his work just as the papyrus was beginning to discuss injuries to the thoracic and lumbar spine. The unfinished scroll then lay untouched for 3500 years before it was sold to the American Egyptologist Edwin Smith in 1862. It was posthumously donated to the New York Historical Society in 1906.
Five of the 48 recorded "cases" describe cervical injuries. It is clear that the ancient Egyptians recognized that misalignment of the bony vertebral column could have disastrous consequences. The papyrus' descriptions of spinal injury have not been improved upon for 4500 years (Fig. 1) (6): "Instructions concerning a crushed vertebra in his neck. If thou examinest a man having a crushed vertebra in his neck thou findest that one vertebra has fallen into the next one, while he is voiceless and cannot speak; his head falling downwards, has caused that one vertebra crush into the next one; and shouldest thou find that he is unconscious of his two arms and his two legs because of it Thou shouldest say concerning him 'One having a crushed vertebra in his neck; he is unconscious of his two arms and his two legs and he is speechless. An ailment not to be treated'."
The Egyptians' knowledge of anatomy was greatly advanced in part by the examination of the dead during the extensive mummification process and in part by treating the injured during the construction of the pyramids and temples. They clearly recognized that a fracture-dislocation was associated with a poorer prognosis than a simple fracture of the spine. Remarkably, the author makes a specific point of discussing the specific spinal fracture produced despite that the outcome was seen as hopeless for all spinal injuries. A wenekh (disjoining) of the vertebrae (subluxation) is distinguished from a sehem (crush) of the vertebrae (compression fracture). Furthermore, a sehem of the vertebrae was known to occur with axial loading such as falling on one's head. These statements demonstrate that the original author had a keen interest in pathophysiological correlation.
In general, spinal fractures were not treated because the prognosis was seen as dismal. There is good evidence that the Egyptians realized that the appropriate treatment of long-bone fractures was reduction and immobilization. Splints, dating from the Fifth Dynasty, have been discovered around compound extremity fractures, suggesting that bone setting was an established art in ancient Egypt (38). However, there is no evidence to suggest that splinting was practiced with spinal fractures.
It is probable that Kubja had kyphoscoliosis with the primary curve and the two secondary curves being regarded as the triple deformity. The veracity of the account may, of course, be placed into question. Nevertheless, the fact remains that this passage is the earliest known reference to the concept of axial traction, predating Hippocrates by over a millennium.
Spinal manipulation as a treatment for spinal dislocation or deformity was widely practiced at the time of Hippocrates. Hippocrates recommended that kyphotic deformities be treated by subjecting the body to traction and applying pressure locally to the area of the kyphosis (Fig. 3A) (24): "But the physicians, or some person who is strong, and not uninstructed, should apply the palm of the hand to the hump, and then, having laid the other hand upon the former, he should make pressure, attending whether this force should be applied directly downward, or toward the head, or toward the hips. And there is nothing to prevent a person from placing a foot on the hump, and supporting his weight on it, and making gentle pressure; one of the men who is practiced in the palestra would be a proper person for doing this in a suitable manner."
If the application of the hand or foot was unsatisfactory, Hippocrates recommended another method (Fig. 3B) (24): "But the most powerful of the mechanical means is this: if the hole in the wall, or in the piece of wood fastened into the ground, be made as much below the man's back as may be judged proper, and if a board, made of lime-tree, or any wood, and not too narrow, be put into the hole, then a rag, folded several times or a small leather cushion, should be laid on the hump. When matters are thus adjusted, one person, or two if necessary, must press down at the end of the board, whilst others at the same time make extension and counterextension along the body, as formerly described." Hippocrates' methods were commonly used through the Middle Ages by medieval physicians, such as Henri de Mondeville (12601320) and Guy de Chauliac (13001368).
"Succussion" was a practice condemned by Hippocrates (Fig. 4). The patient was attached with ropes to an elevated ladder. By flinging the patient to the ground in an upside-down position, it was hoped that the jerk produced by sudden deceleration would realign any spinal deformity (24): "Wherefore succussion on a ladder has never straightened anybody, as far as I know, but it is principally practiced by those physicians who seek to astonish the mobfor to such persons these things appear wonderful, for example, if they see a man suspended or thrown down, or the like; and they always extol such practices, and never give themselves any concern whatever may result from the experiment, whether bad or good. But the physicians who follow such practices, as far as I have known them, are all stupid." Despite Hippocrates' warnings regarding the danger of this procedure, succussion was practiced through the 15th century AD.
da Vinci was the first to accurately describe the spine with the correct curvatures, articulations, and number of vertebrae (Fig. 5). Never before had the anatomy of the entire spine been examined with such attention to detail and proportion. He was also the first to suggest that stability to the spine was provided, in part, by the cervical musculature (Fig. 6) (26): "You will first make the spine of the neck with its tendons like the mast of a ship with its side-riggings, this being without the head. Then make the head with its tendons which gives it its movement on its fulcrum."
This statement is taken from da Vinci's later works. By that time, it seemed that he was no longer fascinated by simple dissection. He began to wonder how the body moved and how geometry and mechanics could further unlock the secrets of human physiology. For perhaps the first time, the biomechanics of the spine was contemplated.
In 1653, Borelli became a Professor of Mathematics at the University of Naples, and 5 years later he obtained the Chair of Mathematics at Pisa. Some of Borelli's more notable achievements were the discovery that some comets move in a parabolic path, a description of Jupiter's moons, and a concise review of Euclid's texts. Most notably, he was one of the founders of "iatromechanics," or the application of mechanics to physiology. This field is the forerunner of what we now call biomechanics.
Borelli's interest in human motion seems to have been sparked by Marcello Malphigi, professor of theoretical medicine at the University of Pisa. Apparently, the two men had a close relationship, as Malphigi recalls (30): "What progress I have made in philosophising stems from Borelli. On the other hand, dissecting living animals at his home and observing their parts, I worked hard to satisfy his keen curiosity." Borelli, who was not a physician, needed to work with Malphigi to ensure that his mechanical calculations made biological sense. The relationship between the two men was a forerunner of present collaborations between biomechanical engineers and spinal surgeons. Although Borelli's knowledge of mechanics was restricted to the principle of levers and the triangle of forces, he was able to generate an accurate and comprehensive account of muscle action (1).
His work, De Motu Animalium (3), underwritten by Queen Christina of Sweden, was published posthumously in 1680 and is the first comprehensive text devoted to biomechanics (Fig. 9). The book is split into two parts. The first part examines the external motions of the musculoskeletal system in animals from a mechanical viewpoint. The second part examines internal motions, such as the physiology of muscles and the circulation.
One of the striking mechanical features of the body noted by Borelli was that the muscles act with short lever arms so that the intervening joint transmits a force that is a magnitude greater than the weight of the load. Borelli overturned older concepts of muscle action, which stated that long lever arms allowed weak muscles to move heavy objects (4). "Galen also states that a tendon is like a lever. He thinks that, consequently, a small force of the animal faculty can pull and move heavy weights. This general opinion seems to be so likely that, to my knowledge, nor surprisingly, it has been questioned by nobody. Who indeed would be stupid enough to look for a machine to move a very light weight with a great force, i.e., use a machine or contrivance not to save forces but rather to spend forces? This seems strange and against common sense, I agree, but I can convincingly demonstrate that this is what happens and, given permission, that the upholders of the opposite opinion have been mistaken." This fundamental fact of human biomechanics was not realized again until 1935 when it was rediscovered by Friedrich Pauwels.
Borelli's evaluation of the spine illustrated a remarkable grasp of its biomechanics. He realized that the intervertebral discs acted like a viscoelastic substance by both cushioning the bones and acting like springs. He proposed that the discs must perform some load sharing because his calculations revealed an inability of the spinal musculature alone to support heavy weights. Remarkably, he was able to conduct detailed calculations describing the forces sustained by individual vertebrae when a load was carried on the neck (Fig. 10) (4): "If the spine of a stevedore is bent and supports a load of 120 pounds carried on the neck, the force exerted by Nature in the intervertebral disks and in the extensor muscles of the spine is equal to 25,585 pounds. The force exerted by the muscles alone is not less than 6,404 pounds. Therefore, the sum of the muscular forces GH which control the fifth lumbar vertebra and a third of the resistance of the intervertebral disc is equal to 826 pounds. The muscular forces are equal to 413 pounds and the forces exerted by the disc are equal to 1,239 pounds." These figures, produced by Borelli over 300 years ago, agree quite well with modern experimental calculations of spinal load sharing.
The mechanical philosophy of the 16th and 17th centuries altered the study of medicine altogether. The prevailing view held that man was the center of the universe and above the physical laws. But just as Copernicus subjected the earth to the laws of celestial mechanics, so Borelli subjected the human musculoskeletal system to the laws of physical mechanics. The human body was converted from an inscrutable, perfect creation of nature to an amazing collection of biological systems that followed the laws of the universe. Borelli's contribution to this change was not trivial. He showed that human motion could be explained in mechanical terms, a concept that remains fundamental to spinal biomechanics.
His results reflected the soundness of his thinking. Ten of the sixteen patients treated in this manner were greatly improved. The need for immediate reduction became so important in Burrell's mind that by 1905, he was advocating open reduction if closed reduction failed. "If the kyphos is very marked, or if upon extension, it does not readily reduce, an immediate operation, unless there is some contraindication, such as shock, should be done, that the reduction may take place while the cord is under the surgeon's eye" (7).
Burrell's conclusions are sound and for the most part are followed in the modern management of patients with spinal cord injuries. The major shortcoming of Burrell was his inability to divide spinal fractures into different categories based on the mechanism of injury and the pathological abnormality produced.
Hanging was probably introduced to the British in the 5th century AD by invading Germanic tribes (Angles, Jutes, and Saxons) and was subsequently employed as the preferred method of execution for the common criminal, decapitation being the privilege of the nobility. Early English hangings can only be described as cruel. The prisoner was hoisted off the ground by a rope tied around the neck and left to struggle until strangulation produced death. Sometimes the hangman would pull on the feet of the condemned or climb on his back to hasten death (36). Guidelines for execution were absent; the technical details of hanging were left to the hangman whose " vanity is equaled only by his ignorance" (31). It would take centuries before English hangings became more humane.
The "drop" method of hanging was probably introduced in 1784 at Newgate Gaol in London. Unfortunately, the length of the drop was variable and drops that were too short produced strangulation whereas drops that were too long produced decapitation. It must be remembered that hangings were public spectacles and neither result was satisfactory to the audience; strangulation was too tedious and decapitation too gruesome. It was not until 1875 that the drop length was standardized by Samuel Haughton. A table was calculated that gave the appropriate size of rope and drop length for a given weight of a prisoner.
Standardization of drop length prevented decapitation, but some prisoners still died of suffocation. This prompted an investigation into knot position. John J. deZouche Marshall viewed several hangings and concluded in 1886 that a knot placed under the chin (submental knot) was the most lethal and even devised a contraption that forced the knot to remain below the chin (Fig. 13) (31). Marshall presented his finding to the Committee of Capital Sentences; he was, much to his chagrin, ignored. Although Marshall's observations were correct, he did not study the underlying mechanism or pathological findings in the cervical spine produced by hanging.
A.M. Paterson in 1890 was the first to describe the pathological outcome produced by hanging with a submental knot (34). The findings of a "hangman's fracture" were described. The pedicles of C2 were fractured, and the anterior and posterior longitudinal ligaments were ruptured. The body and odontoid of C2 migrated cephalad with the cranium while the neural arch migrated caudad.
Frederic Wood-Jones, director of anatomy at the London School of Medicine for Women, can be credited with the first systematic study of clinical biomechanics. His interest began in 1908 when he was allowed to examine the exhumed remains of 102 Nubian men executed by hanging in the Upper Nile Valley during the Roman occupation (2000 yr earlier) (45). Many of the victims still had the ropes tied to their necks, but Wood-Jones noted a curious observation; not a single case of cervical fracture was found. The majority of the victims had disruptions of the cranial base as the only finding.
Wood-Jones hypothesized that the knot position may have been the determining factor in the pathological findings produced by hanging. It was his impression that a large knot was placed below the angle of the jaw and mastoid process in the Roman hangings. This knot position, referred to as a subaural knot, transmitted a force through the cranial base as the body jerked upon hanging.
Wood-Jones then had the exceptional fortune, in 1913, to examine other skeletons whose original owners had been executed by hanging. First, he found a description of the skull of Dr. Pritchard, a Scottish prisoner known to be executed in 1865 by a subaural knot. The skull had the same peculiar cranial base fracture as the Nubians, confirming that a subaural knot was used in the ancient Roman hangings (Fig. 14). Again, as in the Nubians, the cervical spine was intact. He next examined five skeletons donated by Captain C.F. Fraser, superintendent of the Rangoon Central Jail. These prisoners were known to have died instantaneously, and Captain Fraser assured Wood-Jones that a submental knot was used. Examination of the cranium revealed no fractures, but all five sets of cervical spines had identical fractures (Fig. 15) (46): "It is to be noted that the odontoid plays no part in producing death, but that the posterior arch of the axis is snapped clean off and remains fixed to the third vertebra, while the atlas, the odontoid process, and the anterior arch of the axis remain fixed to the skull. This lesion is produced by the violent jerk which throws the man's head suddenly backwards and snaps his axis vertebra."
Wood-Jones, like Marshall, argued on the grounds of mercy that the submental knot should be used in all English judicial hangings because it produced death efficiently, cleanly, and rapidly. His report is also a landmark in the history of spinal biomechanics because specific forces, namely hyperextension and distraction, were seen to produce a specific pathological outcome in the spine, in this case, traumatic spondylolisthesis of C2 or "hangman's fracture."
A three-column model was proposed by Francis Denis in 1983 after he studied >400 radiographs of spinal fractures (12). He discovered that burst fractures, considered by Holdsworth to be stable, were actually unstable. This led him to describe an additional column, a middle column consisting of the posterior vertebral body, the posterior annulus fibrosis, and the posterior longitudinal ligament (Fig. 16). Disruption of two columns was required for instability. Therefore, he considered compression fractures as stable and burst fractures, Chance fractures, and fracture-dislocations as unstable, a paradigm consistent with today's treatment strategies. Denis' model has undergone modification by many authors, but the concept of three columns in the spine has withstood more than a decade of scrutiny (33).
The future of this approach lies in its ability to assess the degree of stability imparted to the spine by a construct before implantation. Furthermore, a scenario can be envisioned in which magnetic resonance images can be directly converted into a three-dimensional computer model and subjected to analysis. Currently, the major limitation of the method is the enormous computer time and storage required to faithfully represent the complexities of the spine.
Several centuries passed before any significant improvement was made to the scamnum like devices. Taylor (41) in 1929 was among the earliest to use controlled cervical traction for the reduction and immobilization of cervical injuries. He used a halter device that used the occipital protuberance and mandible as purchase points (Fig. 19). Crutchfield (9), in 1933, was forced to use an alternate means of traction in a patient that presented with a C2C3 dislocation and a compound fracture of the mandible. Crutchfield inserted Edmonton extension tongs into the outer table of the cranium and attached the tongs to a pulley and weight system (Fig. 20). Over the next several years, Crutchfield made several modifications to the system and his instrument became the standard for cervical traction (2, 10, 28). One major disadvantage of Crutchfield tongs was that the pins needed to be placed near the cranial vertex, which limited the amount of traction that could be safely applied. Another inconvenience was the requirement of a limited burr hole through the outer table to obtain sufficient bony purchase.
Gardner (19), in 1973, devised a set of tongs that improved upon Crutchfield's system. The tongs were larger, which allowed placement of the pins below the cranial equator. The pins were directed toward the vertex, and heavy traction would only drive the pins further into the cranium. The pin design was particularly noteworthy. Sharp pins were used to allow for direct placement through the skin. Penetration of the inner table was prevented by the taper; as the pin was driven further into the cranium, an exponential increase in the metal-bone surface area limited further entry. The system became known as Gardner-Wells tongs (Fig. 21), a name that credited the surgeon and the manufacturer (Trent Wells).
Halo immobilization was introduced by Perry and Nickel (35) in 1959 as an alternative to Minerva jackets in patients undergoing total cervical fusion to treat head instability secondary to poliomyelitis (Fig. 22). The halo produced better postoperative immobilization, was more comfortable, and allowed positional control of the head in all three directions.
Harrington (22), in 1962, developed a rod and hook system that represented a major technical advance in spinal surgery (Fig. 23). Harrington instrumentation, like the halo device, was invented in an effort to better treat patients with polio. Scoliosis of the thoracic spine was treated by carefully studying the preoperative roentgenograms and then applying the biomechanical principle of three-point bending to achieve the desired result. The current explosion in spinal segmental fixation devices is a direct result of enhanced understanding of spinal biomechanics and increased sophistication in applying these principles to treat spinal deformities.
Received, March 25, 1996.
Accepted, April 12, 1996.
Reprint requests: Abhay Sanan, M.D., UMHC Box 96, D429 Mayo Memorial Building, 420 Delaware Street S.E., Minneapolis, MN 55455.
Edward C. Benzel
Albuquerque, New Mexico
Sanan and Rengachary present a special and interesting article on the historical development of concepts in spinal mechanics. Starting with antiquity, they write of the remarkable understanding our early brethren had of the significance of injuries to the spinal column, some so severe that treatment need not be performed. Reviewing the work of our Renaissance predecessors, we find remarkably accurate drawings of the spine and its structures. By the 17th century, people like Borelli were working out significant concepts on the biomechanics of the spine. Within this article are a number of wonderful anecdotes; one that immediately comes to mind is the discussion of hanging and the evolution of how high to place the criminal and where to place the knot. A "hangman's fracture" will never have the same meaning for me after reading this article. The authors go on to provide an excellent historical review of recent 20th century contributions to the treatment and stabilization of the spine. All of this has been illustrated to provide visual clues of the development of this subject for the reader. This is an article all neurosurgeons should read, enjoy, and learn from, and it should be kept very close at hand for our colleagues in training.
James T. Goodrich
Bronx, New York
When this article first came across my desk, I groaned, expecting an hour or so of boring reading. How wrong I was. After several paragraphs, I realized that I could not put it down. Until I read this historical review, I thought that the study of biomechanics began in the mid-20th century. This engrossing article shows how much we owe to early investigators, such as da Vinci, Borelli, Hildanus, Burrell, and others who paved the way for the modern practice of spinal surgery. The title of this scholarly article is a bit of a misnomer, because the authors have chosen to define biomechanics in the broadest possible sense of the term. In fact, this is more a mini-history of the management of spinal disease with an emphasis on the contributions of anatomists, early clinicians who made accurate observations of the nature of spinal disease (especially trauma), and those who treated spinal disease without operation. A disproportionately small amount of space is given to the enormous contributions made by a variety of investigators in the last 30 or 40 years. This is appropriate, because the more recent advances are familiar to most neurosurgeons and, although more relevant in our contemporary management of patients, they are perhaps less interesting than what transpired in the several millenia that preceded the 20th century.
Paul R. Cooper
New York, New York
Return to the Neurosurgery Home Page
