Human anatomy successfully defies the central principles of mechanical engineering. In most manufactured devices, such as cars and ships, the bulk of weight is concentrated in the lowermost part to achieve maximum stability. Classic examples are the pendulum of a clock or the lead loaded keel of boats—a sudden lateral thrust causes oscillations with little displacement, and the force energy dissipates in friction; the structure swings eventually back into its original position. The mechanics of human anatomy violate this principle.1 In humans the bulk of weight is in the proximal parts of the limbs and trunk: in engineering terms the body consists of a series of inverted pendulums, and its mathematical simulation requires complex algorithms calculated by main frame computers. Such a structure is inherently unstable and prone to falls. The success of the human species with such an “engineering handicap” seems surprising at first, but there are important advantages in bipedal locomotion.2 It allows quick positional changes in all directions and rapid acceleration and deceleration in emergencies, it offers a good view of the surrounding environment and, probably most importantly, it minimises the energy expenditure of movement—brisk walking or jogging consumes only about 0.025 megajoules (6 kilocalories) per minute. These characteristics confer considerable survival advantages in biological selection, and they compensate well for postural instability. Some researchers even regard human gait as an important prerequisite for intellectual development because the upper limbs are free to perform sophisticated tasks; hand dexterity—which corresponds to larger brain size—became a selection advantage.
Summary points
|
The downside of an upright gait is the need for a highly developed apparatus for balance, with rapid integration of multiple sensory input from the eyes, inner ear, and internal position sensors, and accurate regulation of muscle tone and contraction.3 The unstable anatomy with its peculiar weight distribution makes relentless adjustments of joint position and muscle length necessary, and these corrections have to be integrated with spatial orientation to allow purposeful locomotion. The complexity of the process is only beginning to be understood. Unfortunately, a large number of physiological disturbances can impair postural control, ranging from disorders of electrolyte concentration (for example, after diarrhoea or diuretics) to changes of sensory perceptions (for example, disease of the eye or inner ear).4 Another major problem is the progressive decline of “balance reserve” with increasing age.5 In healthy elderly people measurements of horizontal body oscillations (“sway”) after a minor but sudden tilt of a floor platform show an increased amplitude and duration compared with young people. Minor environmental hazards, such as uneven kerbs or pavements, which are easily compensated for by young people often lead to falls in elderly people.6 Many different deficits have been identified as the underlying disorder in ageing, especially alterations in the peripheral and central nervous system and the “hardware” of locomotion such as joints and muscles. Elderly individuals show considerable variations in posture maintenance, and there is debate if these changes are due to disease or part of normal ageing.7
Disease related to falls has become a major health problem owing to demographic changes (around 1 in 6 people is now older than 65 years).8 Over one third of all people aged 70 or older have at least one fall per year, and 1 out of 20 of these sustains serious injury such as head trauma or fracture.9 Consequently fall has become the single most common cause of admission to hospital. The costs to the NHS of this “epidemic” are enormous; the expenditure per year for care of hip fracture exceeds £1bn.10
Increasingly, the need for further research into age related impairment of balance is recognised, and funds are becoming available. There are, however, several problems with research into falls. Firstly, there is lack of an animal model for basic physiological experiments. Humans are the only life form with a purely bipedal gait. Other primates—for example, chimpanzees—have different gait patterns and habitats. Observations in patients and studies in human volunteers have provided important information about the overall performance of bipedal locomotion, but the effects of localised anatomical damage, well defined physiological alterations, or sensory impairments on the various components of the system remain to a large extent unknown. Without such detailed knowledge more accurate clinical classifications and targeted treatments are unlikely to emerge. Furthermore, poor balance (“unsteadiness” or “giddiness”) or its worst consequence, a fall, is an unspecific symptom. Underlying this can be a wide variety of chronic or acute physiological changes or diseases—medical students learn that any known disease can cause falls. This lack of a relation between cause and effect has restricted research mainly to prevention of accident and injury (such as hip protectors) in subgroups of patients.11 Despite some practical success with this approach patient selection and study group size pose considerable problems, and little progress has been made in our general understanding and in finding solutions that could apply to all groups of people who fall.
Researchers into falls have expected to gain insight from the big projects established to create artificial intelligence—machines with intelligent thought and autonomous movement.12 For the layman such enterprises are represented by the android or the human-like machine, as depicted in 1926 by Fritz Lang in the film Metropolis,13 but the expectations of researchers into falls were more humble. They hoped for a basic bipedal model, similar in weight distribution and joint number to humans, that could stand unaided and that could master a couple of steps of upright walking. Such a model would enable researchers to investigate falls and patterns of imbalance by introducing defined faults, thus allowing a more accurate classification of instability states and comparisons with impairments in patients; new treatments could also be assessed, such as changes in weight distribution or adjusted shoe wear.
It was believed that such machines would be available by the end of the century as considerable resources were invested in multicentre research programmes, for example, the Massachusetts Institute of Technology Leg Laboratory and the autonomous walking project of the German Research Society; similar efforts are under way in Japan (for example, the Honda robot). The result of these efforts was a huge number of different types of walking machines, mostly spider-like systems—for example, the TITAN robot series. A comprehensive catalogue has been compiled listing many of the machines capable of independent locomotion.14 It is accessible via the internet and provides detailed information about current research. It leads on to the website of the Massachusetts Institute of Technology, which introduces, for example, Troody, a mechanical copy of a troodon dinosaur built by the institute's leg laboratory. It can be viewed as a short motion picture—its movements have little in common with human walking.
None of the machines built over the past 20 years has the ability to simulate human walking. Tasks that are virtually effortless for us, such as orientation in space and control of upright movement, proved to be insurmountable technical hurdles.15 This failure of modelling, even of apparently simple elements of bipedal locomotion, seems inconceivable 30 years after the Apollo mission. The failure is not due to ineptitude in engineering but rather a reflection of biological complexity. Human evolution needed millions of years to develop a successful bipedal locomotion system, and humans will have to wait a long time before engineers can achieve working models of this. It seems we are just at the beginning of our scientific journey into the understanding of how to remain upright.
