FOOD-BORNE PATHOGENIC BACTERIA
Many cases of so-called food poisoning are due to the presence of
pathogenic bacteria in the food. In some instances, as in the typical
meat poisoning epidemics, symptoms develop so soon after eating that the
particular food involved is immediately suspected and laid hands on. In
other cases the guilty article of food is difficult to trace. Certain
cases of tuberculosis are undoubtedly caused by swallowing tubercle
bacilli in the food, but the precise source and date of infection can be
rarely, if ever, certainly established.
The presence of pathogenic bacteria in food is usually due either to the
contamination of the food by infected human beings during the process of
preparation or serving, or to an infection of the animal from which the
food is derived. The relative importance of these two factors is quite
different in the various infections.
TYPHOID FOOD INFECTION
The typhoid bacillus does not attack any of the domestic animals;
consequently all food-borne typhoid is caused more or less directly by
human contamination. A remarkable instance of typhoid infection due to
food was reported in 1914 in Hanford, California, where ninety-three
typhoid cases were caused by eating Spanish spaghetti served at a public
dinner.[49] Investigation showed that this dish was prepared by a woman
typhoid-carrier who was harboring living typhoid bacilli at the time
she mixed the sauce for the spaghetti before baking. Further laboratory
experiments indicated that the ordinary baking temperature at which the
spaghetti was cooked was not only not sufficient to sterilize the food,
but afforded a favorable opportunity for the bacteria in the interior of
the mass to multiply. The infection of the food was consequently heavy
and involved a very large proportion (57 per cent) of those present at
the dinner.
Merited celebrity attaches to the exploits of the typhoid-carrier, Mary
Malloy, who, in pursuing her career as cook in and about New York City,
is known to have caused at least seven typhoid outbreaks in various
families in which she worked and one extensive hospital epidemic.
Similar cases of typhoid food infection by employees in restaurants and
public institutions are by no means uncommon, and show the necessity of
protecting food from contamination during the whole process of
preparation and serving. Acting on this principle, the Department of
Health of New York City has inaugurated a comprehensive examination of
the cooks and waiters (approximately 90,000) employed in the public
restaurants and dining-rooms in that city. Results have been obtained in
the discovery of typhoid-carriers and of cases of communicable disease
that amply justify this procedure as an important measure for protecting
the community against the dissemination of infection.
Some foods by their origin are exposed more than others to typhoid
contamination. Such vegetables as lettuce, celery, radishes, and
watercress, which are commonly eaten without cooking, are more likely
to convey typhoid than peas, beans, and potatoes. A typhoid outbreak
apparently due to watercress has been reported from Philadelphia.[50] At
a wedding breakfast to forty-three guests on June 24, 1913, watercress
sandwiches were served, and subsequent inquiry showed that nineteen of
the guests partook of these sandwiches. Eighteen of this number became
ill with typhoid fever within a month, the illness developing in most
cases after the guests had scattered to their summer homes. Those who
did not eat watercress sandwiches were not affected. Typhoid infection
by uncooked celery has also been reported.[51]
The practice of using human excreta as fertilizer in truck gardens is
sometimes responsible for a dangerous contamination of the soil, which
is communicated to the growing plants and persists for a long time.[52]
Even scrupulous washing of vegetables is not sufficient to render them
bacterially clean. In the future the danger to the community from this
source is likely to become increasingly serious unless the growing use
of this method of soil enrichment is definitely checked.
In 1915 an increasing number of typhoid cases in South Philadelphia led
to an investigation by the state health department.[53] This disclosed
the fact that the majority of the cases were clustered in and about
three public markets.
These are all curb markets--fruits, vegetables, pastry, clothing,
and miscellaneous merchandise of every description are dumped on
push-carts and pavements without regard for any sanitary
precautions. The patrons of these markets handle and pick over the
exposed foodstuffs, thus giving every opportunity for the
transmission of disease....
The greatest number of cases occurred in the immediate vicinity of
the Christian Street Market. This market is largely patronized by
the inhabitants of the section known as "Little Italy." The patrons
of the South Street Market are principally Hebrews, while the
Seventh Street Market is patronized in the main by Hebrews
and Poles.
The following conclusion was reached regarding the particularly large
number of cases among persons of one nationality:
Our inspectors have found that the different methods used by the
Italians and Hebrews in the preparation of their food are
responsible for the larger number of cases being found in the
vicinity of the Christian Street Market in Little Italy. It is the
custom of the Italians to eat many of the fruits and vegetables raw,
while the Hebrews cook the greater portion of their food. It is
presumably due to this custom that the members of the Italian
colony have suffered to a greater extent than the other residents
of the district.
A bacterial examination of various kinds of vegetables obtained from
push-carts and curb markets led to the finding of the typhoid bacillus
upon some of the celery. It would naturally be difficult to determine in
such cases whether the typhoid bacilli were derived from infected soil
in which the celery was grown or whether the contamination occurred
through improper handling.
Bread, when marketed unwrapped, is subject to contamination from flies
and from uncleanly handling. Katherine Howell[54] has shown that
unwrapped loaves of bread sold in Chicago were more or less thickly
smeared with bacteria and were coated on the average with a much larger
number than wrapped loaves. In some cases typhoid fever has been
directly traced to bread. Hinton[55] has recorded the occurrence of
seven typhoid cases in the Elgin (Illinois) State Hospital, which were
apparently due to a typhoid-carrier whose duty it was as attendant to
slice the bread before serving. When this typhoid-bearing attendant was
transferred to another department where she handled no uncooked food,
cases of typhoid ceased to appear.[56]
Food such as milk that is not only eaten customarily without cooking,
but is also suitable for the growth of typhoid bacilli, needs to be
particularly safeguarded. It is noteworthy that the compulsory
pasteurization of milk in New York, Chicago, and other large American
cities has been accompanied by a great diminution in the prevalence of
typhoid fever. Until recent years milk-borne typhoid in the United
States has been common and hundreds of typhoid epidemics have been
traced to this source.
[Illustration: FIG. 5.--Bacteria left by fly passing over gelatin plate.
(By courtesy of Doubleday, Page & Company.)]
One food animal, the oyster, frequently eaten raw, has been connected on
good evidence with certain typhoid outbreaks.[57] The number of
well-established oyster typhoid epidemics is not great, however, and
the danger from this source has sometimes been exaggerated. The source
of oyster contamination is in sewage pollution either of the shellfish
beds or of the brackish water in which the oyster is sometimes placed to
"fatten" before it is marketed. State and federal supervision of the
oyster industry in the United States in recent years has largely done
away with the taking of oysters from infected waters, and although
oysters--and clams and mussels as well--must be steadily safeguarded
against sewage contamination, the actual occurrence of oyster infection
at the present time is believed to be relatively rare.
Probably the most effective method of preventing typhoid food infection
is to investigate every case of typhoid fever and trace it, so far as
practicable, to its origin. In this way typhoid-carriers may be
discovered and other foci of infection brought to light. Carriers, once
found, may be given proper advice and warned that they constitute a
danger to others; the complete control of typhoid-carriers who are not
disposed to act as advised is a difficult problem and one not yet solved
by public health authorities.
ASIATIC CHOLERA
With Asiatic cholera, just as with typhoid fever, domestic animals are
not susceptible to the disease, all cases of infection having a direct
human origin. Drinking-water is the usual vehicle of cholera infection,
and even in countries where the disease is endemic, food-borne outbreaks
of this disease are far less common than those of typhoid fever.
Occasional instances of Asiatic cholera due to milk supply and to
contaminated fruits or lettuce are on record, but these are exceptional
and cannot be regarded as exemplifying a common mode of spread of this
disease. The extent, however, to which dwellers in tropical
countries--and indeed in all lands--are at the mercy of their household
helpers is illustrated by the following experience of the English
bacteriologist, Hankin. "I have seen," he says, "a cook cooling a jelly
by standing it in a small irrigation ditch that ran in front of his
cookhouse. The water running in this drain came from a well in which I
had detected the cholera microbe. He cleaned a spoon by dipping it in
the drain and rubbing it with his fingers; then he used it to stir the
jelly."[58]
TUBERCULOSIS
Animal experiments have shown that both meat and milk derived from
tuberculous cattle are capable of conveying infection. The precise
degree of danger to human beings from the use of these foods under
modern conditions is still in dispute. Since the tubercle bacillus of
bovine origin differs from the tubercle bacillus of human origin in
certain well-defined particulars, it is possible by careful study to
distinguish the human infections caused by the bovine bacillus from
those caused by the so-called human tubercle bacillus. Additional
comparative investigations are needed in this field, and these may
enable us to estimate eventually more fully than is possible at present
the extent of human tuberculous infection derived from bovine sources.
Meat is a less likely source of infection than milk, chiefly because it
is rarely eaten without cooking. Opinion regarding the actual frequency
of the transmission of tuberculosis by means of the meat of tuberculous
cattle has been widely at variance in the past, and must even now be
based on indirect evidence. There is no well-established instance of
human infection from the use of the flesh of tuberculous cattle. The
significance of this fact, however, is diminished by the observation
that tubercle bacilli can pass through the intestinal wall without
leaving any trace of their passage and can make their way to the lungs
or to other distant organs where they find opportunity for growth. This,
together with the long period which usually elapses between the actual
occurrence of infection and the discovery of the existence of infection,
makes the difficulty of securing valid evidence peculiarly great.
Opposed to any very frequent occurrence of meat-borne tuberculosis are
the facts that the tubercle bacillus is not commonly or abundantly
present in the masses of muscle usually marketed as "meat," that the
tubercle germ itself is not a spore-bearer and is killed by ordinary
cooking, and that the reported cases of the finding of tubercle bacilli
of bovine origin in adults over sixteen years of age are extremely rare.
This latter fact is perhaps the strongest evidence indicating that
tuberculous meat infection, although theoretically possible, is at least
not of common occurrence.
Most of the commissions and official agencies that have considered the
precautions to be taken against possible tuberculous meat infection are
agreed that the entire carcass of an animal should be condemned when the
tuberculous lesions are generalized or when the lesions are extensive in
one or both body cavities as well as when the lesions are "multiple,
acute, and actively progressive." Any organ showing evidence of
tuberculous lesions is obviously not to be passed as food. On the other
hand, it is considered that portions of properly inspected animals may
be put on the market if the tuberculous lesion is local and limited and
the main part of the body is unaffected; in such cases contamination of
the meat in dressing must be avoided. It is the general belief that
when such precautionary measures are taken the danger of tuberculous
infection through properly cooked meat is so slight as to be negligible.
Milk is a much more likely vehicle than meat for the transmission of
tuberculosis. Freshly drawn raw milk from tuberculous cattle may contain
enormous numbers of tubercle bacilli, especially if the udder is
diseased. Contamination of milk by the manure of tuberculous cows can
also occur. Observers in England, Germany, France, and the United States
have found tubercle bacilli in varying numbers in market milk, and have
proved that such milk is infectious for laboratory animals. Although, as
pointed out with reference to meat infection, the difficulties of
tracing any particular case of tuberculosis to its source are very
great, there are a number of instances on record in which the
circumstantial evidence strongly indicates that milk was the vehicle of
infection. Especially convincing are the observations on the relative
frequency of infection with bovine and human tubercle bacilli at
different ages as shown in the following tabulation:[59]
====================================================================
|Adults Sixteen|Children Five to|Children under
| Years Old | Sixteen Years | Five Years
| and Over | Old |
--------------------+--------------+----------------+--------------
Human tubercle | 677 | 99 | 161
bacilli found | | |
Bovine tubercle | 9 | 33 | 59
bacilli found | | |
-------------------------------------------------------------------
The large proportion of bovine tubercle bacillus infections in children
stands in all probability in causal relation to the relatively extensive
use of raw milk in the child's dietary.
The proper pasteurization of milk affords a safe and reasonably
satisfactory means of preventing tuberculous infection from this source.
The general introduction of the pasteurizing process in most American
cities has ample justification from the standpoint of the prevention of
infection.
VARIOUS MILK-BORNE INFECTIONS
The facts related in the foregoing pages indicate that of all foods milk
is the most likely to convey disease germs into the human body. This is
partly due to the fact that milk is sometimes obtained from diseased
animals, and partly to the fact that unless great care is taken it may
readily become contaminated during the process of collection and
transportation; if milk is once seeded with dangerous bacteria these can
multiply in the excellent culture medium it affords. It is also partly
because milk is commonly taken into the alimentary tract without being
cooked. For these reasons the amount of illness traceable to raw milk
far exceeds that ascribable to any other food.
There are several infections that may be communicated by milk, but are
rarely if ever due to other foodstuffs. Diphtheria and scarlet fever are
perhaps the best known of these. Both diseases have been repeatedly
traced to the use of particular milk supplies, although various forms of
individual contact also play a large role in their dissemination.
Milk-borne scarlet fever and diphtheria seem to be generally, if not
always, due to the direct contamination of the milk from human sources.
It is considered possible, however, by some investigators that the cow
may sometimes become infected from human sources with the virus of
scarlet fever or diphtheria and may herself occasionally contribute
directly to the infection of the milk.
A serious milk-borne disease, which has lately been conspicuous in
Boston, Chicago, Baltimore, and other American cities under the name of
"septic sore throat" or "streptococcus sore throat," originates
apparently in some cases from infection of the udder of the cow by an
infected milker; in other cases the milk has seemingly been directly
infected by a human "carrier." The specific germ is thought to have been
isolated and its connection with the disease demonstrated in the
laboratory. This disease, like diphtheria and scarlet fever, is
sometimes due to contact. It is not known to be caused by any food
except milk.
Foot-and-mouth disease of cattle is transmissible to man through the
milk of infected cattle, but this infection in man is not very common or
as a rule very serious. So far as known, it is not communicated to man
in any other way except through the use of uncooked milk.
Such cases of infection or "poisoning" by milk may be prevented, as
already stated, by the exclusive use of heated milk. The possible
occurrence of nutritional disturbances (e.g., scurvy) in a small
proportion of the children fed on pasteurized or boiled milk is
considered by many physicians to be easily remedied and to possess much
less practical importance than the avoidance of infection.
POSSIBLE INFECTION WITH B. PROTEUS
One widely distributed organism known as _Bacillus proteus_ has been
several times held responsible for food poisoning outbreaks, but it is
not yet certain how far this accusation is justified. _B. proteus_ is
related to _B. coli_, but most varieties do not ferment lactose and are
much more actively proteolytic than the latter organism, as shown by
their ability to liquefy gelatin and casein. Like _B. coli_, they form
indol and ferment dextrose with gas production. Varieties of _B.
proteus_ are found widely distributed in decomposing organic matter of
all sorts.
The evidence upon which this bacillus is regarded as the cause of food
poisoning is not altogether convincing. The outbreak described by
Pfuhl[60] is typical. Eighty-one soldiers in a garrison at Hanover were
suddenly attacked with acute gastro-enteritis four to twelve hours after
eating sausage meat. The meat was found to contain _B. proteus_ in large
numbers, although it was prepared with ordinary care and was entirely
normal in appearance, taste, and smell. Rats and mice fed with the
sausage became ill and _B. proteus_ was isolated from the blood and
internal organs. But these animals sometimes die when fed with quite
normal meat, and _B. proteus_ and other common intestinal bacteria are
often isolated from the body after death. _B. proteus_, in fact, is
found in many animal foods and in the apparently normal human intestine.
Like _B. coli_, it frequently invades the internal organs after or
shortly before death. Finding _B. proteus_ in food or in the internal
organs does not therefore constitute definite proof of any causal
relationship. The evidence attributing other outbreaks to infection with
_B. proteus_ is similarly inconclusive.
It is equally uncertain whether the production of a poison in food by
this species can in any degree be held responsible for meat poisoning.
_B. proteus_ is common enough in decomposing food material and under
certain circumstances is known to generate substances that are toxic for
man. It is possibly true that toxic substances are produced in the early
stages of decomposition by this organism. In the opinion of Mandel[61]
and others, if any injurious effect at all is to be attributed to _B.
proteus_, it is in the nature of an intoxication and not an infection
(see chapter viii). So far as the existing evidence goes, the question
of the responsibility of this organism for food poisoning is still an
open one.
FOOTNOTES:
[49] Sawyer, _Jour. Amer. Med. Assoc._, LXIII (1914), 1537.
[50] _Eng. News_, LXX (1913), 322.
[51] Morse, _Report of State Board of Health of Mass._, 1899, p. 761.
[52] R. H. Creel, _Reprint from Public Health Reports, No. 72_,
Washington, 1912.
[53] _Health Bull. No. 76, Pennsylvania State Department of Health_,
December, 1915.
[54] _Amer. Jour. Public Health_, II (1912), 321.
[55] _Institution Quarterly_, III (1912), 18.
[56] See also a similar instance reported by Lumsden, _Hyg. Lab., U.S.
Public Health and Marine Hosp. Service, Bull. 78_, p. 165.
[57] For a discussion of the oyster question see G. W. Fuller, _Jour. of
Franklin Institute_, August, 1905; _N.Y. City Dept. of Health, Monthly
Bull._, November, 1913, and May, 1915; H. S. Cumming, _U.S. Public
Health Service, Pub. Health Bull. 74_, March, 1916.
[58] _Lancet_, II (1895), 46.
[59] Park and Krumwiede, _Jour. Med. Research_, N.S., XVIII (1910), 363.
[60] _Ztschr. f. Hyg._, XXXV (1900), 265.
[61] _Centralbl. f. Bakt._, I, Orig., LXVI (1912), 194.