Beta cells can be isolated and implanted, requiring only outpatient surgery
for implantation. But foreign beta cells are quickly rejected without
immunosuppressant drugs. Even with the recent advances in drugs, especially
cyclosporin, using immunosuppressants is much more dangerous than living with
diabetes. As a result, beta cell implantation is not currently used to treat
diabetes.

Current research is investigating two general methods of implanting beta
cells without the use of immunosuppressant drugs. The first (immunoisolation)
encapsulates the beta cells within a barrier so that nutrients, glucose, and
insulin can pass freely through the barrier but the proteins which provoke
the immune response, and the cells which respond, cannot pass. The second
(immunoalteration) involves altering the proteins on the surface of the cells
which provoke the immune response. The first human trial began early in 1993
on immunoisolated beta cells, and human trials were scheduled to begin late
in 1993 on immunoaltered beta cells. (As of early 1997, I haven't had the
opportunity to try to locate the followup to these trials.)

An article in the Journal of Clinical Investigation, September 1996,
describes a successful experiment which implanted immunoisolated porcine
(pig) islets into monkeys. An accompanying editorial describes the state of
islet transplantation. Both are online in full, linked from the issue
contents page at

   http://www.jci.org/content/vol98/issue6/

In early 2000, a lot of hype appeared about the "Edmonton protocol" trials.
While an important step, this is still only a small step on a long journey.
They made improvements in technique and graft survival, but no progress on
the serious problems of beta cell supply (each patient needed beta cells
from two cadaver donors) or of immunosuppressant use (they used drugs,
albeit carefully).

Don't expect these treatments to be available on a standard basis any time
soon. I've been reading about this research since the mid-1970s, and the
results are always just around the corner. Serious problems remain to be
solved: safety of the immunoisolated implants, long-term survival, ability to
use beta cells from non-human species or grow usable cells for grafting in
the laboratory, perfection of both techniques -- all
these must be resolved before beta cell implantation moves beyond the
experimental stage. Other problems will likely be encountered along the way,
since this is cutting edge medical research. I'll be surprised if it gets out
of the lab before the year 2005; 2015 is probably a better guess. And it may
fail -- it's always possible that unsolvable problems will yet arise.

Finally, it's not yet clear that even completely normal bG profiles will cure
all the problems of type 1 diabetes. Some may be related to the autoimmune
reaction that is the immediate cause of diabetes. This question cannot be
answered until it is possible to normalize bG levels for a period of many
years.