30
The Body Electric
weeks if the ends are held close to each other without movement. Occa-
sionally, however, a bone will refuse to knit despite a year or more of
casts and surgery. This is a disaster for the patient and a bitter defeat for
the doctor, who must amputate the arm or leg and fit a prosthetic sub-
stitute.
Throughout this century, most biologists have been sure only chemi-
cal processes were involved in growth and healing. As a result, most
work on nonunions has concentrated on calcium metabolism and hor-
mone relationships. Surgeons have also "freshened," or scraped, the frac-
ture surfaces and devised ever more complicated plates and screws to
hold the bone ends rigidly in place. These approaches seemed superficial
to me. I doubted that we would ever understand the failure to heal
unless we truly understood healing itself.
When I began my research career in 1958, we already knew a lot
about the logistics of bone mending. It seemed to involve two separate
processes, one of which looked altogether different from healing else-
where in the human body. But we lacked any idea of what set these
processes in motion and controlled them to produce a bone bridge across
the break.
STAGES OF FRACTURE HEALING
Every bone is wrapped in a layer of tough, fibrous collagen, a protein
that's a major ingredient of bone itself and also forms the connective
tissue or "glue" that fastens all our cells to each other. Underneath the
collagen envelope are the cells that produce it, right next ro the bone;
together the two layers form the periosteum. When a bone breaks, these
periosteal cells divide in a particular way. One of the daughter cells stays
where it is, while the other one migrates into the blood clot surrounding