Beriberi and its History

For centuries in the past, certain populations in the
world were subjected to the ravages of four particular
diseases with well-characterized symptoms: these diseases
are beriberi, scurvy, pellagra and rickets. Progress
in the aetiology of these diseases was hindered by the
dogged belief that a disease must be caused by a
positive factor, i.e. a microorganism or a toxin. The
medical authorities could not envisage that a disease
could be caused by a lack of something in the diet. Although
the aetiologies were not understood until the
early part of the twentieth century, dietary cures were
known much earlier among certain people.

Beriberi
The first clinical description of beriberi is attributable
to a Dutch physician, Jacobus Bonitus, while working
in Java in 1642. No remedies for beriberi were known
until 1882 when Admiral Kanehiro Takaki, Director-
General of the Medical Department of the Japanese
Navy, showed the disease to have a dietary origin. By
simply increasing the allowance of vegetables, fi sh,
meat and barley in a diet consisting predominantly
of polished (milled) rice, Takaki was able to prevent
the disease. Furthermore, Takaki managed to persuade
the Japanese authorities to change the standard
naval ration to one with a higher protein content. The
beriberi problem in the Japanese Navy disappeared
almost entirely following the change. Takaki ascribed
his success to the increase in the nitrogen to carbon
ratio of the diet. This explanation was not accepted by
the medical establishment who, fi nding no causative
microorganism, switched their attention to searching
for a toxin.

In 1886, the Dutch government sent Christiaan
Eijkman, a physician, to Batavia on the island of Java
in the Dutch East Indies (now Djakarta, the capital of
Indonesia) to join a research team in the investigation
of beriberi. The disease was rampant among the native
soldiers and also among the island’s prisoners. Because
of its epidemic character, beriberi was assumed
to have a bacterial origin and the scientists thought
that they had isolated the causative micrococcus from
the tissues of sufferers. They believed that the bacteria
were fl oating in the damp air of barracks and prisons,
so that ventilation and disinfection were the appropriate
preventive actions.

Eijkman tried to infect rabbits and then monkeys
with the micrococcus but the animals showed no
signs of disease. Switching to chickens, because they
were more economical both to buy and maintain,
Eijkman noted that control birds as well as injected
birds developed a paralytic disease, which he named
polyneuritis gallinarum. The disease was characterized
by an unsteady gait leading to an inability to stand
and culminating in death. Autopsies revealed degeneration
of peripheral nerves, most conspicuously in
the limbs, that resembled nerve degeneration seen in
the autopsies of people who had died from beriberi.
Thinking that infection had spread from injected
birds to control birds, Eijkman set up another site,
remote from the fi rst, to investigate whether chickens
would remain healthy unless deliberately infected.
Then a surprising thing happened: the chickens at
the original site began to recover and there were no
new deaths.

Eijkman, at fi rst perplexed, then found out that,
during the fi ve months in which the disease had been
developing, the chickens had been fed cooked rice from
the military hospital. Before and after this period, the
birds had been fed feed-grade uncooked rice. Eijkman
performed a controlled experiment which showed
that the cause of the disease was associated with the
cooked rice diet. His explanation was that ‘cooked rice
favoured conditions for the development of microorganisms
in the intestinal tract, and hence for the formation
of a poison causing nerve degeneration’.
Eijkman turned his attention to the difference between
hospital rice and the raw rice normally used for
Historical events leading to the establishment of vitamins 3
the chicken feed. The latter had been crudely pounded
to remove most of its outer husk, but the adhering
pericarp or ‘silverskin’ (also known as rice polishings)
was still attached. The hospital rice, on the other hand,
had been further processed by polishing to remove its
silverskin. The important difference between the two
types of rice seemed therefore to be the presence or
absence of the silverskin layer. Eijkman proposed two
theories. In the fi rst theory, the polished rice might
cause disease because, once its protective skin has been
removed, pathogens have easier access to the starchy
endosperm and can multiply there during storage. In
the second theory, the silverskin contains substances
indispensable to life and health that are absent or
occur in too low concentrations in the underlying
grain. Eijkman tested the fi rst theory by feeding four
chickens uncooked polished rice freshly processed
each day. Two of the birds developed the polyneuritis,
making this theory the less likely. At the end of many
experiments, Eijkman’s tentative hypothesis was that
the rice polishings contributed an antidote to a nerve
poison produced by the fermentation of starch in the
chicken’s crop.

Having shown the association of avian polyneuritis
with polished rice, Eijkman still had no evidence that
the polyneuritis and human beriberi had a common
origin. He therefore encouraged a medical colleague,
A. G. Vorderman, who was the inspector of prisons, to
study the relation between the incidence of beriberi
in the many prisons in the Dutch East Indies and the
diet and hygienic condition of the prisoners. It was
standard practice to supply a given prison with a particular
type of rice, therefore the inmates were ideal
subjects for a controlled feeding trial. The statistics of
this study dealt with no less than 279 621 individuals,
all of them prisoners at different times. In 37 prisons,
unpolished rice was supplied; only one of these prisons
developed beriberi. In 13 prisons the rice supplied
was polished mixed with unpolished; in six of them
beriberi developed. Out of 51 prisons where polished
rice was supplied, as many as 38 developed cases of the
disease. From these data it was calculated that for each
10 000 of the prison population there was only one
case among those eating unpolished rice, 416 on the
mixed rice diet, and 3900 on polished rice. There was
no signifi cant correlation with the age of the inmates,
the ventilation of their accommodation or degree of
overcrowding.

Despite all of Eijkman’s efforts, beriberi was still
considered to be the result of some kind of infection
and rice was still regarded as a possible carrier of bacterial
toxins. After Eijkman left Java on the grounds of
ill-health, plans to use unpolished rice for all prisons
in the Dutch East Indies were cancelled. Eijkman’s
successor Grijns at the laboratory in Batavia extracted
a water-soluble ‘polyneuritis preventive factor’ from
rice polishings and in a paper published in 1901 he
correctly concluded (the fi rst to do so) that beriberi is
the result of a dietary lack of an essential nutrient.