STATE
OF HAWAII
DEPARTMENT
OF HEALTH
P.0.
B0X 3378
HONOLULU,
HAWAII 96801
January
31, 2002
In
reply, please refer to:
HEER
OFFICE
Coqui3
Mr. Sydney
Ross Singer
Medical
Anthropologist
P.O. Box
1880
Pahoa,
Hawaii 96778
Dear Mr.
Singer:
Thank you
for your letter of January 7,2002. We appreciate your
mention of your credentials in that letter. We did not
mean to insult your intelligence in past letters when
speaking in generalities. We now appreciate what level
of detail you require. The Deputy Director for
Environmental Health, Gary Gill, asked me to reply to
you for him.
"Skin
permeation can be measured in living humans or
animals, or in vitro by using excised (human) skin in
diffusion cells. Studies with excised skin seem
feasible, since passage through the skin is a passive
diffusion process and the primary barrier, the stratum
comeum, is composed of nonliving tissue." (Robert
L. Bronaugh, Raymond F. Stewart, Elaine R Congdon, and
Albert L. Giles, Jr., "Methods of in Vitro
Percutaneous Absorption Studies. I. Comparison with in
Vivo Results," Toxicology and Applied
Pharmacology 62: 474-480(1982).)
The first
thing to note is the passive diffusion, not active
transport, through the nonliving stratum comeum layer.
Based on this statement by experts, sir, would you
agree that the skin of a human being, a land mammal
whose skin has adapted a stratum comeum to protect
against water loss due to existence in relatively arid
land environments, would be much more protective
against water intrusion than the thin skin of a frog
which had originally adapted to existence in an
aquatic environment? An Olympic swimmer who trains in
a pool for hours does not absorb pool water or
chlorine beyond his skin into his bloodstream and
emerge waterlogged or edematous. Also, if you recall
dissecting frogs in Comparative Anatomy class, you may
remember how thin a frog’s skin is, with no stratum
comeum, and how easy it is to snip with tiny scissors,
compared to the difficult scalpel work in cutting the
thick skin of a cat. One would expect that a frog’s
skin absorption of water or of an aqueous solution of
caffeine would be much greater than that of a land
mammal. The greater absorption of caffeine by a frog
would account for the toxic accumulation and greater
effect of caffeine on a frog than on a human.
Also, the
in vivo experiments in the aforementioned study used [14C-labeled]
caffeine and demonstrated the removal of caffeine from
the bloodstream by the kidney and its excretion via
the urine. As you know, the accumulated concentration
of caffeine in a tiny frog with a small internal
volume of water would be much greater per unit of body
weight, compared to the concentration of a heavy dog
or human with a large internal volume of water. The
concentration of caffeine in a mammal would then be
reduced by the kidney’s removal and excretion
actions.
The
aforementioned study employed four compounds that had
been previously studied by other researchers,
including caffeine, to verify that certain in vitro
methods using rat skin gave similar results to test
methods using live animals, in viva. For example,
caffeine was found to have permeability constants of
2.1 x 10’4 cm/hr in vivo and 3.1 x 10 cm/hr in vitro
for percutaneous absorption through shaved rat skin.
The
permeability constant, kp3, is
defined as the steady-state rate of absorption
(amount/cm2/hour) divided by the
concentration of solute applied to the skin (amount/cm3),
which is then expressed in centimeters per hour
(cm/hr). The permeability constant accounts for
applying different concentrations of caffeine to the
skin. These authors verified that caffeine is absorbed
through the skin at a very limited rate by passive
diffusion, not active transport.
The
principal author. Dr. Robert Bronaugh, wrote a book
with another nationally recognized expert. Dr. Howard
Maibach of the University of California medical school
in San Francisco, titled Percutaneous Absorption
(Marcel Dekker publishers, 1989). Chapter 32 of this
book, pp. 555-565, is titled "Facilitated
Percutaneous Absorption of Charged Drugs," and is
written by Jonathan Hadgraft, Philip G. Green, and
Paul K. Wotton. In this chapter, the permeability
constant (also called permeability coefficient) of
caffeine is compared to other substances. The test
system used fall-thickness human skin in vitro that
had been pretreated with ethanol and Ethomeen S-12 to
increase skin absorption, and 0.01M pH 5 aqueous
solutions of various substances. Pentachlorophenol, a
pesticide notorious for being absorbed through the
skin, had a permeability coefficient of 0.7 cm/hr. Two
other substances which have been found to be skin
absorbed, benzene and toluene, had coefficients of
0.02 cm/hr and 0.05 cm/hr. Metal ions, such as iron,
chromium, lead, and mercury, had coefficients of
approximately 0.001 cm/hr. Caffeine had a permeability
coefficient of 0.002 cm/hr, which possibly implies
that caffeine may be as unlikely to be absorbed
through the skin as metals which are commonly found in
Hawaiian soil.
These
experimental results confirm the theoretical
application of the log octanol-water partition
coefficient (Kow) which was
mentioned in a previous letter. There are supporting
studies, such as one by A-L. Bunge and R.L. Cleek,
"A new method for estimating dermal absorption
from chemical exposure: 2. Effect of molecular weight
and octanol-water partitioning," Pharmacology
Research 12(1): 88-95, Jan. 1995, or more than 50
studies coauthored by either Jonathan Hadgraft or
Philip Green, who appear to be two more well-published
authorities on percutaneous absorption.
You stated
that caffeine and nicotine have "similar chemical
properties." We agree that caffeine and nicotine
are both water-soluble, but beyond that, they do not
appear to be similar. We said that the solubility of
caffeine in oils and lipids determines the
percutaneous absorption, not the water solubility.
Nicotine has a Kow of 1.17, and
besides being soluble in water, it is very soluble in
oils, kerosene, ether, chloroform, and alcohol. This
accounts for its easy absorption through the skin.
After sitting on a bench wet with a 40% nicotine
solution for 15 minutes, a florist showed signs of
toxicity (Faulkner JM, "Nicotine poisoning by
absorption through the skin," Journal of the
American Medical Association 100: 1664-65, 1933). A
look at its chemical structure in any edition of The
Merck Index shows pyrrolidine rings with one attached
methyl group, which accounts for its nonpolar
character and its solubility in nonpolar oils and
lipids, as found in the skin. Caffeine, however, has a
Kow of -0.07, is very soluble in
polar solvents such as water or alcohol, but is only
slightly soluble in ether and other nonpolar solvents.
The Merck Index shows its structure as purine rings
with two attached oxygen atoms, which make it polar
and thus insoluble in nonpolar oils and lipids. Also,
"Aqueous solutions of caffeine salts dissociate
quickly," (The Merck Index. Tenth Edition, 1983)
which makes caffeine even less absorbable through the
skin.
We still
believe that caffeine is unlikely to be absorbed
through the skin in quantities sufficient to pose a
public health threat. If you are still unconvinced by
our reasoning, backed up by the scientific evidence in
our citations, we welcome your statements to the
contrary, backed up by your citations.
Sincerely,
LESLIE K.L.
AU, M.Sc.
Toxicologist,
HEER Office
