Medicine and Engineering

There is no doubt that an engineering approach to medicine is doomed to be incomplete. We will never know all there is to know about how the body works, and why it goes from one state to another. Nevertheless, medicine has been quite useful for some health problems, and is certain to become even more useful. The reason for the success is that, although the "butterfly effect" is real, most of the butterflies don't count--there are billions of butterflies but only a few hurricanes. Just as there are regions of instability where the smallest butterfly can create a storm cloud, there are regions of stability where even a bird won't cause much change. The fact that we can remain alive even for one day indicates that the homeostatic mechanisms of the body are usually able to keep it in the middle of the calm, "healthy" regions of the hypothetical graph. If every small change caused huge consequences, simply eating a meal would be terribly risky.

In fact, the body's ability to keep itself on track is quite powerful. Alcohol in large doses is a poison, but in slightly smaller quantities it causes only temporary and relatively mild effects. A person's blood sugar can vary by a factor of two or three without them even being aware of it. In rare cases, people have drowned for half an hour and been restored to life. People can be kept alive for hours with their heart and lungs disconnected during open-heart surgery, and if their kidneys fail, dialysis will work for years. Furthermore, the body constantly encounters and compensates for a wide variety of perturbations--depending on the physical demands a person faces, their food intake can vary from 1000 to 8000 calories per day.

If something does go seriously wrong, it can often be put right by a remarkably simple intervention. For example, a heartbeat is a delicately orchestrated ripple of activation that spreads from nerve centers through the heart muscle. When the ripples do not move smoothly (perhaps due to insufficient blood supply during a heart attack), different areas of the heart get out of sync and start to beat out of rhythm. This is called fibrillation, and it is generally fatal. But a single massive electric shock, enough to make the heart convulse in unison, is often enough to allow it to start beating normally (and yes, chaotically) again. That something as intricate as the heart can be restarted by such unsubtle treatment is evidence that the heartbeat is actually quite robust. Continuing the example of the graph, the shock has the effect of jolting the body back into a region of stability.

It seems, then, that a minor perturbation is quite unlikely to cause any disturbance in the body's overall state of health, and even major disturbances can often be tolerated. This is not to say that we can ignore the complexity and chaos inherent in the body. We will always have to be careful of unintended side effects. However, most well-designed treatments will not have any effects that are major, unexpected, and negative. As our medical technology improves, any negative effects will be either immediately noticeable or extremely slow and subtle, giving us plenty of opportunity to detect and correct them before they pose significant threats to health. And as we learn more about where the regions of stability are and how to push the body back to them, we will be able to apply simple "engineering fixes" to more and more problems.

We can assume, then, that medicine will generally be able to correct problems without creating worse problems. It will usually be the case that symptoms of sickness can be traced to improper status of particular systems, and those problems can be corrected using straightforward techniques without causing worse problems in other systems. This is not to say that an automobile-mechanic approach to health is always best--an "integrated" approach will work better for some problems. But the point of this chapter is that even a limited, mechanistic approach can result in greatly increased lifespan.

There are several types of systems in the body. The most obvious are physical systems, such as the bones and the lungs. The physical systems are coordinated by signaling systems--patterns of chemicals or of neural activity that provide control and feedback to the various organs and tissues. DNA can be considered a system whose main purpose is to store information; it interacts with many signaling systems to produce a wide range of proteins. Metabolic systems create and destroy chemicals in order to supply the body with energy and clean up waste. The neural system senses and influences various functions, and provides both short-term and long-term information storage as well as massive information processing. The immune system fights infection, and sometimes attacks body cells as well (which may be good or bad). Some might add soul or spirit to this list. However, any soul or spirit we may have is apparently unaffected by drowning, epileptic fits, and open-heart surgery. We need not consider it in the context of medical intervention--in other words, the presence, absence, or properties of soul or spirit is irrelevant to a discussion of medical techniques.

The body's systems operate on different time scales. DNA never changes, except as a result of mutation, retroviruses, or immune system adaptation (or learning). Physical systems change over a period of days to years. Metabolic systems have a time scale of minutes to hours; signaling, seconds to minutes; immune system, seconds to days; and neural system, milliseconds to decades. A complete medical maintenance program must be able to cope with all of these time scales, as well as the size scales (from molecules to organs) and signaling methods between the various systems. This is a tall order--medicine can't do it yet. But a little advanced technology goes a long way.