Public Health Section...control of tuberculosis are fully aware of the develop- ments in this field of medicine. The very character of the tuberculous infection, its chronic nature

74 THE INDIAN MEDICAL GAZETTE [Feb., 1950

Public Health Section | \ 0 t< V

PROTECTIVE VALUE OF BCG

VACCINATION

Review and Digest of Majok Publications

(Issued by the International Tuberculosis Campaign)

Although BCG vaccination, has been brought to the attention of the public a number of times in the course of the last thirty years, only relatively few persons besides the specialists interested in the prevention and control of tuberculosis are fully aware of the develop- ments in this field of medicine.

The very character of the tuberculous infection, its chronic nature as a disease, the distribution in an

endemic form all over the world are such that true

evidence of the efficac}' of the vaccine can be obtained only by statistical studies. A reliable study of this kind should not onfy provide for a follow-up of many years of the material under observation but also for a

control group of non-vaccinated persons exactly com- parable with the group of those vaccinated. It is evident that such experiments, involving hundreds if not thousands of persons to be controlled during a ltJng period of time, are extremely difficult to carry out.

Nevertheless, several studies of this kind have been attempted and the results have justified the use of the vaccine in the fight against tuberculosis throughout the world. Much work has also been done both in laboratories and on animals to improve efficiency and innocuity of the vaccine as well as techniques and indications for vaccination.

The vaccine was used on human beings for the first time by Weill Halle (1925) in 1921 when he adminis- tered it orally to a newborn child, after Calmette and Guerin had studied the properties of BCG, the Bacillus Calmette Guerin as it was called, for more than ten years in vitro and on animals. Soon mass vaccination of children by the oral method, as advocated by Calmette (1928), was carried out on a fairly large scale in several countries. By the end of 1928, more than 116,000 children had received BCG in France alone. This method, however, lost much of its prestige when it appeared that there were no substantial clinical nor

statistical proofs of the efficiency and harmlessness of the vaccine [Wallgren (1927), Greenwood (1928), Rosenfeld (1928), etc.]. The fact that only 5 to 30 per cent of the children became sensitive to tuberculin after receiving BCG orally reduced markedly the

epidemiological value of the vaccination. Furthermore, the Lubeck tragedy, although not caused by BCG, as was definitely established by a German court (1935), made oral vaccination unpopular. The court on the basis of bacteriological examinations, determined that vials of virulent tubercle bacilli of a known strain used for laboratory experiments were accidentally substituted for vials of BCG. To-day the oral method is still used in some countries, mainly for the immunization of newborns in tuberculous families, as in Roumania and in Poland, or on a larger scale as in Brazil (1947).

Parenteral administration of the vaccine was begun around 1927 in Scandinavia; Heimbeck in Norway injected it subcutaneously, Wallgren in Sweden used it intradermally. The latter method has now been very widely adopted although other techniques are also being used to some extent; Rosenthal's multipuncture (1939), scarification by the method of Negre and Bretey (1947).

Wallgren (1934), a pediatrician, was the first to show the importance of BCG in the control of tuberculosis in a community injecting all the children of tuberculous families and those exposed to infectious sources in the town of Gothenburg, in Sweden, and he obtained a

sudden drop in the mortality rates from tuberculosis in the younger age groups. Whenever possible, the children

were isolated from the source of infection for six weeks before vaccination and afterwards until, allergy had developed. Thus a total of 1,069 persons were given BCG intradermally in the period 1927-1937 in Gothenburg alone. Anderson and Belfrage (1939), in 1939, made a.

study of this material and were able to re-examine 905 vaccinated persons. They found that tuberculous disease had developed in two cases only, and in a benign form. Furthermore, a small child had died from a

specific primary infection, with generalization which had developed three weeks after vaccination. This child had not been isolated prior to the vaccination and it must be presumed that BCG had been inoculated in the incubation period. Observations made by Heimbeck (1948) in the years

1924-1926 on probationers entering the nursing school of the IJllevaal Hospital on Oslo had shown that while many of those who did not react to tuberculin became ill with tuberculosis at an early stage of their training, only few cases developed among those found sensitive to tuberculin on admission. Prompted by this experi- ence, Heimbeck began BCG vaccination of his tuber- culin-negative nurses in 1926 and to-day he is still following up most of them. Among the probationers entering the school in the period 1924-1936 for a three- year training course, 668 were found positive to

tuberculin on admittance, 501 were negative and BCG vaccinated, while 284 more negatives were not vaccinated. The following rates of morbidity and mortality were calculated for three groups per 1.000 observation years; for the positive group the rates were 12.4 morbidity, 0.0 mortality (22 cases of tuberculosis, all forms included, and no deaths), for the vaccinated 24.1 and 2.1 (35 cases and 3 deaths) and for the

negative group 141.2 and 14.6 (97 cases and 10 deaths). Both morbidity and mortality appear very high in the latter group and Heimbeck stresses the fact that among the nurses who were vaccinated the incidence of tuberculosis has been reduced to one-sixth.

Several studies of this type, on selected groups highly exposed to infection and .easy to control for a long period, were made in a never-ending effort to improve the conditions of the experiment and to gain new

knowledge on the many problems involved. [Scheel (1935) and Holm (1941) on medical students, Nordwall (1944), Delachaux (1948), Genevrier and Maclouf (1948), etc., on nurses.] R. G. Ferguson's (1946) observations on vaccinations

of general hospital nurses and of sanatoria nurses

showed that the protection given by BCG would apply as well to the groups with a high annual infection rate (sanatoria nurses : 71.8 per cent) as to the groups with a much lower annual infection rate (general hospital nurse : 11.8 per cent). Comparison of the percentage of cases developing a

tuberculous disease in the vaccinated and non-

vaccinated groups shows a ratio of 1 :6.5 for the more exposed nurses and of 1 :4.5 for the less exposed nurses who had received BCG, such a marked reduction can hardly be attributed to mere chance or to the annual decline of the death and case rate in Canada.

In the years 1935-1938, Aronson in the United States started in 13 different Indian reservations, scattered from Arizona to Alaska, what was to become one of the most accurate studies on BCG. One thousand five hundred and fifty-one North American Indians were vaccinated with BCG and 1,457 were

kept as controls. All the subjects chosen for the experiment were in the age group 1 to 20 and the division between those to be vaccinated and those to be followed up as controls was made quite at random. All the subjects were re-examined annually by tuber- culin test and radiography. The statistical analysis of the material was made by Palmer (1946) of the U.S.

FeB-, 1950] PROTECTIVE VALUE OF BCG VACCINATION 75

Table I

BCG vaccination in hospitals and sanatoria of Saskatchewan

(It. G. I' erguson). (1934-1913)

Nurses in general hospital (yearly infection rate of 11.8 per cent)

Positive at admittance ..

Negative, not vaccinated Negative, vaccinated . ?

. .

"

Nurses in sanatoria (yearly aniec-

tion rate of 71.8 per cent) Positive on entrance Negative, not vaccinated Negative, vaccinated

Number of Tuberculosis persons

478 1,368 1,005

293 113 203

cases

5 55 9

11 18 5

Percentage with tuberculosis

1.05 4.02 0.89

3.75 15.9 2.46

Average years observed

2.43 2.43 2.42

1.25 1.06 1.07

Public Health Service. The speci given sub- s-ray films did not know to which gio P j5 tQ

l?ct belonged. The two groups were foun^

^ tQ

similar in age distribution, amoun *

?0jj0W.Up. tuberculous infection and completeness

of fi t

When the results of the Indian .fnn tVicre were

published after G years of obseiva^ >

?^rois as -8 deaths due to tuberculosis ^"omnn" the BCG compared with only 4 such deaths atno ? f 00

vaccinated. This is a rate of 3 4 Person years, with a ratio of 1 .7.7 , losjs that vaccinated. The total incidence of tuberc

^ ̂ tQ the sum of all the cases and all '

^ -n the disease. wna 185 in the control group ^cnfrailJst 47 l"-l 1,UUU ] vaccinated.

Table II

BCG vaccination among American Indians (J. D. Aronson and C. E. Palmer)

in group

Number Per 1,000 person years

Controls (2,550)

2S 185

BCG (1,457>

Controls

(8,977)

4 3.4 40 24.3

BCG Ratio (8,367) BCG controls

0.44 1 : 7.7 4.7 1 :3.2

in <1 1Grmore, Aronson and Palmer mention a decrease sroun -nTber cases tuberculosis in the BCG

the cUlt years following; vaccination, while among ^nstaiT a'S t'ie morbidity rate remained fairly 'that (1 ^CC01'ding to the authors, this might indicate than'j f,r>rotec!'i?n may be greater in the latter rather ' n the earlier years after vaccination

cite?cyH^la'' and Leslie (1945), in Chicago, to

f0,. ni ,

?ne ?f their studies, kept under observation va?;,0l;e ,"lan ten years 1,204 infants who had been

V/ero ]- <'ur^nS their first week of life and 1,213 who caSeS 7.

as conti"ols. Among the vaccinated three

?f ..tuberculosis developed and another child died there c, as?> while among the non-vaccinated controls tuhn,.,^\er?' iu the same period of time, 23 cases of

oerculosis and 4 deaths.

(l947)?-exceP^ona' are the observations made by Hyge have ]

m a -^anish school for girls, and his findings experir^611 comPared to those of a controlled laboratory ?Pen p

^ on humans. Following the discovery of an Schnni

ase tuberculosis among the pupils, the whole

z-ravp 1 ^0PV'ation of the Aurehoj State School was

again ? tuberculin tested in November 1941 and

200 tii}11 Fef?ruary 1942. At this date 144 girls out of

"ation j negative reactors volunteered for vacci-

c?ntrnl WRr-e subsequently found positive to a

tuberculin test. A new examination took place

in. December 1942 and of the 368 pupils examined 105

were found tuberculin-negative, 130 tuberculin-positive after natural infection and 133 positive after BCG vaccination. About two months later, in January and February 1943, an influenza-like epidemic broke out

among the school girls, beginning with several cases of erythema nodosum. After a renewed, thorough examination the source of infection was found to be a teacher of sciences who held classes in a damp, permanently blacked-out cellar. Some of the classes had not been in contact with the teacher in question so that out of the total 105 negative pupils, 94 had been exposed and 70 of them had become tuberculin- positive (74.5 per cent). Of these inverters 41 showed .r-ray changes of the thoracic organs and 37 had a

positive gastric lavage. In 11 cases, that is in 11.7 per cent of the exposed subjects, a progressive pulmonary

tuberculosis developed, followed by death in one case.

Among the 133 vaccinated 102 had been in contact with the infectious source and only two cases of tuberculosis developed (1.9 per cent). The only girl that had lost allergy after vaccination also suffered from a mild form of the disease. No other cases were found in 5 years of observation in this group. In the group of 130 who were originally tuberculin-positive, 105 had been exposed and four cases with positive gastric lavage were found.

Thus the efficacy of the vaccine has not only been proved beyond doubt on animals but we find that most of the experiments on human beings which can be sub- jected to statistical analysis give a percentage of

protection of the vaccinated varying between 70 and 100 per cent (Irvine, 1949). This evidence has now been accepted universally and the use of vaccination in the fight against tuberculosis is no more a matter of faith as it was in Calmette's days. Even in Great

Britain, where the attitude of some authors towards BCG has been highly critical (Wilson, 1947), vaccina- tion will be made available on a national scale to

particularly exposed groups under the provisions of the National Health Service Act (1948)

The question of whether BCG should be used only in groups particularly exposed to tuberculous infection and in populations with a high incidence of the disease

76 THE INDIAN MEDICAL GAZETTE [Feb., 1950

or if it should ho given indiscriminately to all tuber- culin-negatives whenever possible, still remains open for discussion. It is true that the evidence we now have on the effects of BCG has been gathered mostly from observations of highly exposed groups : nurses

and medical students, North American Indians, children of tuberculous families, etc., and that we know much less about the epidemiological value of vaccination in populations

_ with a low death rate from tuberculosis

and a low infection rate. In Denmark, for example, where the death rate for

pulmonary tuberculosis has dropped to 22 deaths per 100,000 population in 1948 (The National Health Ser- vice of Denmark, 1948), vaccination is not only carried out ou a wide scale but plans are being made to extend it further. On the other hand, some authora flike Myers (1948) in the U.S.] have expressed concern that the introduction of BCG would destroy the

possibility of discovering new contagious cases of tuber- culosis by repeated tuberculin surveys on a mostly negative population. The argument, however, has practical value only in countries where the control of tuberculosis has developed to such an extent that every single bacillary case can be traced and isolated immediately and where, moreover, there is little move- ment of the population requirements that are met with only in very few parts of the world to-day. The situation described by Holm (1946) of the

conditions on the Danish island of Bornliolm, with a

population of about 50,000, is a good example of how BCG can be useful as part of a good tuberculosis con- trol system. After cattle tuberculosis had been eradi- cated from the island in the early twenties and all the open cases of the

_ disease had been isolated, the

percentage of negative reactors to tuberculin was the largest in Denmark, particularly in the younger generation. Nevertheless, cases of tuberculosis still developed and the age distribution was_ the same as in other parts of Denmark, with a peak in the 15 to 35 age group. The cases, mostly of a severe nature, occurred mainly in young negatives that left the island for some other part of the country and came> back when they had contracted the disease. Vaccination of all negatives was begun in a systematic way in 1940 and since then there has been a marked reduction in the number of cases reported. This drop in the morbidity curve is particularly evident in the same 15 to 35 age group where the disease in previous years had had its richest harvest. And, as Holm states, this reduction must be ascribed to the extensive BCG vaccination in these age groups. Although vaccination has been accepted in practice

to such an extent that in some countries like Brazil, Czechoslovakia, France, Yugoslavia and Norway it has been made compulsory by law for certain population groups, there are still many improvements to be made and much information to be gained by further observa- tion and research. The difference in the degree of

protection provided by the vaccine, the possibility that protection may' be different according to race, to the degree of exposure or to other known and unknown variables, and many other problems are of concern

to clinicians and statisticians and spur them to con-

tinually new studies. In the United States, for example, a large research plan on BCG vaccination of school children with a comparable group of controls was started some years ago as part of a larger tuber- culosis control experiment in a community of 100,000 population in Columbus (Georgia, 19-17). Other large- scale studies of vaccination are in progress in Puerto Rico and in Finland, but results will be forthcoming only in some years. The fact that vaccination can be safely used wherever

necessary has been fully demonstrated by the mass campaigns now in progress in many parts of the world under the auspices of the International Tuberculosis Campaign. This programme, which began as a tuber- culosis relief action by the Danish Red Cross in 1947 in Poland. Germany and Hungary, became an inter- national effort in 1918 when it was joined bv

Norwegian Relief for Europe, Swedish Red Cros^ UNICEF (United Nations International Children $

Emergency Fund) and the World Health Organization- As at 15th November, 1949, 17,600,000 persons had been tuberculin-tested and 8,200,000 BCG-vaccinated*

Thus, after many years of study and trials, BCCI vaccination has passed from the experimental phase to practical application in the field of tuberculosis control. The World Health Organization Expert Committee on Tuberculosis (19-19), meeting in July 1949, has expressed itself in the following terms :

'It is considered that the number of individuals who would benefit from BCG vaccination on a large scale would be especially large in communities where tuber- culous infection and disease are frequent. Mass- vaccination with BCG should^ therefore, be recom- mended especially for communities with high tuber- culous infection rates and mortalities. It is, however, recommended that even in countries where tuberculous infection and disease are relatively infrequent, vaccination bo applied to individuals and groups with high degrees of exposure^ to tuberculosis, such as familial contacts and persons with occupational hazards.'

BIBLIOGRAPHY

Anderson, H., and Ten years' experience of Belfrage, H. (1939). BCG vaccination at

Gothenburg. Pcedia- trica, 26, 1-11.

A kon son, J. D., find BCG vaccination among 1\u,tukji, C. 10. (1946). American Indians. Public

Health Reports, Washing- ton, 01, 23, 802, 7th June.

Assis, A. i)to (19-17) .. Diretrizes aluais de vacina- cao BCG no Brasil. Clinica Tisiologica, Rio dc Janeiro, 2, 395-432, Oct,-Dec.

CaijMhtte. A. (1028) .. La vaccination preventive d? la Tuberculose par le BCG (Bacille Calmetle Guerin). Annates Institiit Pasteur, 42, suppl., 1-109, December.

Idem (1928) .. Pratique et resultats par Je BCG en France au ler Juillet 1928. Presse Medi' cale, 36, 1409.

DtiiACHAUX, A. (194X). Morbidite tuberculeuse dans une ecole d infirmieres et premiers resultats de la vaccinal ion systema tique au BCG par scarifications <! 111 a n e e s . Helvetica Medicu Acta, 15, 42-60, Jan.

pErtnusoN, 11. G. (1946). BCG vaccination in Hos- pitals and Sanatoria oi

Saskatchewan. America n Revue, of Tuberculosis, S4, 325, Oct.-Nov.

Geneviuer, j.. and EfTicacite de la vaccination A. C. (1948) par le BCG chez les infir-

mieres do 1 Hopital Saint- Joseph. Critique des indi- cations a b u s i v e 3. 1m Hemaine des Ilopitaux de-

Paris, 24, 2901-2901, 30th Nov.

?The following countries are participating in the

I.T.C.?Europe: Austria, Czechoslovakia, Finland. Greece, Italy, Poland, Yugoslavia: Asia: India.

Pakistan, Ceylon; Middle, East: Egypt, Lebanon,

Israel; North Africa: Algeria, Morocco, Tunisia- The I.T.C. Campaign is also scheduled for extension to other countries in these areas and in- Latin-America in

1950, with a total goal of 50,000.000 to be tested and about 20.000,000 to be vaccinated.

Feb-? 1950] FIFTY YEARS AGO 77

""?KMvonn. M. (1928). Professor Calmette's statis- tical study of BCG vac-

cination. Brit. Med. Jour., 3514, 2, 793, 800.

^imhkck. J. (19-is) .. BCG vaccination of nurses.

Tubercle, 29, 4, 84, April. ?f-M. .Tons. (1941) .. The significance of increased

exposure to infection with

i tubercle bacilli : compara- tive studies on medical and polytechnical studies. Acta Tubercolosea Scancli- iiavica, 15, 370-395.

Id,.m (1910) .. BCG vaccination in Denmark. Public Health Reports, Washington, 61, 36, 1298-

1315.

(1949) W.H.O. T.B.C., 15, 15th

II August.

U,K- r. V. (1947) .. Tuberculosis epidemy in a

Statesehool. Acta _ Tuber-

colosea Scandinavica, 21,

1-56.

idem (1949) The efficacy of BCG vac-

cination. Ibid., 23, 153-

1 i55.

"X,N'K. IC. N. (1949) .. BCG vaccination in theory and practice. Blackwell Scientific Publ., Oxford,

i v. p- "6.

it'"1 VinoI'gsscatork, Epilog zur Lubeck-Katas- " 1') trophe. Medizinische Welt,

5, 113. ' ( >n (1935). .. Bakteriologische Untersuch-

ungen. Die Sauglingstuber- kulo.se in Lubeck. R c ichsf/es undheitsam t, Ju I.

w Springer Verlag, 69.

A. (1948) .. BCG vaccination. Amer.

Revue of Tuberculosis, 57,

N'rgrf- 107-109, Jan.

j /jqL; a,u' 1 tretry, Vaccination par le BCG par ? scarifications cutanees.

Massan and Cie, Edit.,

Y Paris, Ilnd Ed.

ri,^ai.l, TT. (1944) Influence of BCG vaccina- tion on the tuberculosis frequency of the Sophiah- iemmet School for Nurses in Stockholm. Acta Tuber- eojosea Scandinavica, 18,

45' ' S. (1928) .. Der statistische Beweis fur

die Immunisierung Neuge- borener mit BCG. Wiener kliniscJi c Wochenschrift,

j, 41, S00'

Jiiai,, S. R. (1939). The multiple puncture method of BCG vaccina- tion. American Review of Tuberculosis, 39, 128-134.

^ M'ui u, s. i> f jjf UiJ) Ten years' experience with

(104<A E. T. BCG. J own. Pedialr., 26,

S0 170-180.

heel, 0. (1935) .. La tuberculosa parmi lea

etudiants en medecine a

Oslo et sa prevention per

la vaccination au BCG.

Revue do la Tuberculose,

Ti,,. v 1'

SiWr/C?I0NAl' ^t;ALTU (Preliminary abridged com-

oW) ?F munication) Medico Statis- tical Informations from

Denmark for the year 1048.

Ho

Wallgren, A. (1927) .. Observations critiques eur la vaccination anti-tubercule- use de Calmette. Acta Pcedialrica, 7, 120.

Idem (1934) .. Value of Calmette vaccina- tion in prevention of tuberculosis in childhood. ?lour. Amer. Med. Assoc., 103, 2, 1341.

Wkill-IIalle, B., and Premiers essais de vaccina-

TtTRPTN, Tv. (1925). lion anti-tuberculeuse de l'enfant per le Bacille Calmette-Gue'rin (BCG). Bull, and Mem. Soc. Med. des IIop. de Paris, 49, 1589-1601, 18th December.

Wilson, G. P. (1947) .. The value of BCG vaccina- tion in control of tuber- culosis. Brit. Med. Jour.,

855-859, 29th Nov.

Idem (1947) .. Columbus selected for long range BCG study. Jour.

Amer. Med. Assoc., 134, 89.

Idem (1049) .. BCG vaccination. Lancet, p. 29, 2nd July.