CtylotJ.-Mel ScL 29( Nc . 1 June) 1986,pp. 39-52 The Relationship between Reactivities to Lepromin A and Soluble Protein Antigen bt Mycobacterium Leprae and Tuberculin by V. PEMAJAYANTHA (Division of Biometry, Central Agricultural Research Institute, Gamwruwa) and M . R. M . PINTO AND N . B. ERIYAGAMA (Department of Microbiology, University of Peradeniya) SUMMARY The relationship between reactivity to tuberculin (PPD-RT23) and reactivities to antigens of Mycobacterium leprae (Fernandez, Mitsuda and to a Soluble Protein Antigen of M. leprae) was studied using three methods. A clear relationship could be demonstrated with two of the methods. A statistically significant correlation could be demonstrated by the method of regression analysis between tuberculin and all three types of reactivities; however the level of correlation (i-2) was unusually low. This study also revealed au increase of all three types of reactivity with BCG vaccination. INTRODUCTION The relationship between leprosy and tuberculosis has evoked considerable interest from the early days of leprosy research. The available literature on the subject has been comprehensively reviewed from time to time 7 « 1 2 » 1 3 at regular intervals indicating die considerable scientific interest in die subject and its importance. However, the gaps in our knowledge in this area are yet considerable. Some early workers wrote that leprosy and tuberculosis are 'antagonistic diseases1)4. Others have noted that patients with lepromatous disease often tended to develop tuberculous disease as well 2. Further, there is also evidence that the status of hypersensitivity to mycobacterial antigens, or its induction or modification using BCG, may affect the subsequent development of leprosy disease. The earlier studies have all investigated the relationship between reactivities to lepromin of human origin and tuberculin reactivity, and none appear to have examined the relationship between tuberculin and reactivity to Soluble Protein Antigen (SPA) of Mycobacterium leprae, which have now become available for skin tests. In this report is examined the relationship between tubercubn reactivity and reactivity to Lepromin A (of armadillo origin) and that to SPA: 40 V. PEMAJAYANTHA, M. R. M. PINTO AND N . B. ERIYAGAMA MATERIALS AND METHODS The tests done and methods of analysis of relationships between reactions were as described in detail elsewhere. M ( M 1 The methodology was as follows: The method of skin testing used was the standard intradermal technique recomm­ ended for tuberculin testing , the skin tests being carried out on the volar aspect of the forearm, at sites 6 - 7 cms. apart. All individuals tested were adults (12 years of age, and above) in three population groups iii two geographical areas. The antigens used were (1) Lepromin A(of armadillo origin supplied by Dr. Hastings of the National Hansens' Disease Centre, Carvillc, U.S.A., through courtsey of the Chief, Leprosy Section, World Health Organisation, Geneva), (2) a Soluble Protein Antigen (SPA) of M. leprae, prepared by ultrasonic disruption of the organism, with a protein content of 10 /tg/ml (Batch CD 19) - (supplied by Dr. R. J.W. Rees of the IMMLEP M. leprae Bank, Harrow, Middlesex, U.K.), and (3) Tuberculin PPD - RT23, 2 t.u. per dose (Staten Serum mstitute, Copenhagen, Denmark). The tests were read, with Lepromin A - Fernandez reactivity at 48 hours, aud Mitsuda reactivity at 28 days, and SPA and Tuberculin reactivities at 72 hours. The results were read as the maximum transverse diameter of the induration palpable, with Fernandez, SPA and tuberculin reactivities, and of the nodule palpated (observed ) with Mitsuda reactivity. T a b u I. Availability of results at different locations for analysis of the relationships between reictivitics Location of Numbers tested population ' Antigens used in tests Comparison of reactivities Total BCG Status group possible Negative Positive (1) Lepromin A(b.icillary (1) Fcnuudez vs. Tuberculin. 256 147 109 I'ussellawa (1) Lepromin A(b.icillary content 3x107/"") (2) Mitsuda vs. Tubcrcul in 221 124 97 (elevation content 3x107/"") (2) Mitsuda vs. Tubcrcul in 900m) (2) Tuberculin PPD RT 23 (2 t. u./dosc) Nuwara Eliya (1) Fenmdez vs. Tuberculin 120 74 46 -Pedro '(elevation As above (2) Mitsuda vs. Tuberculin .98 56 42 (1950m) Nuwara Eliya' "(1) Soluble Protein '(•I)'SPA Vs. Tube'rculm " ay 53 J4 -Mahagastota Antigen of M. : ; • . • • • ' • (elevation leprae ' (SPA, - • • • •'• 1950m) Protein content .' i0/tg/ml) (2) Tuberculin PPD 'RT.23 (21. u./dosc) R E L A T I O N S H I P B B T W E E N . R E A C T I V I T I E S T O M. LEPRAE A N D T U B E R C U L I N 41- Results were therefore available for reactivites elicited by three different antigens; in three population groups, in two.geographical areas (Table I). The relationships examined here will be that between tuberculin reactivity and Fernandez and Mitsuda and SPA reactivities. Broadly; the relationship between reactivities to antigens o f Mycobacterium leprae. (Lepromin A and SPA) and tuberculin were studied by examina­ tion o f (a) the correlations between reactivities by the use o f regression analysis, (b) the changes in the patterns o f frequency distributions o f a reactivity, where reactions are classified according to immunological status [whether a "reactor" (or "positive") or' 'nonreactor" (or "negative") ] o f another reactivity and (c) the occurrence, o f combinations o f reactions o f different immunological status (as above) with each type o f reactivity in the same" individual. In the analysis o f reactions, the differentiation o f reactions to antigens o f M. leprae into "non-reactors" ("negative") and "reactors" ("positive") could be made relatively easily on the basis o f a simple examination o f frequency, distributions. 1 0," Such interpretation was difficult with the frequency distributions o f tuberculin reactions. Further, it has been suggested that the criteria that may have to be used for classifi­ cation o f "negative" and "positive" tuberculin reactions (to. identify specific infections by Mycobacterium tuberculosis - analogous to the "reactor" component o f distributions o f reactions with antigens o f M. leprae) - may be different in different altitude and geographical zones o f Sri Lanka 9. Thus a reaction size o f 6 m m or more may be acceptable for a "positive" reaction - "reactor", at the highest elevation^nd 10 mm or more at a lower elevation (though with greater error). Therefore, when the immunological status o f tuberculin reactions were used in analysis, both these criteria were used in the same type o f analysis. TABLE 2. Bivari.ce frequency distribution of tuberculin and Fernandez reactions in the same individuals, in the whole population F E R N A N D E Z R E A C T I O N SIZES — '. : W I T H . 0 1-2 .3-4 5-6 7-8 9-10 11-12 13-14 15-16 17- 0 81 18 23 34 15 5 4 4 1-2 1 3 6 8 5 1 1 3 - 4 4 1 7 8 1 1 1 5 -6 4 1 1 4 3 1 7 - 8 5 1 6 3 3 1 I 9-10 4 6 4 5 1 1' I 11—12 9 3 4 2 3 2 13-14 2 4 5 6 1 .2. • 1 15-16 4 1 2 4 5 t 1 17-18 5 2 2 2 I 2 19-20 1 2 I 1 1 I z b— St . - .Individuals not included in .the -Table had reaction sizes -. Tuberculin/ Femindez (respectively) of, 22/14 (two reactions), 24/10,26/10 and 34/22. http://Bivari.ce 42 V . P E M A J A Y A N T H A . , M . B . M . P I N T O A N D N . B , E R I Y A G A M A Results in Tables 2, 3 and 4 are presented bivariate frequency distributions of Tuberculin and Fernandez reactivity, Tuberculin and Mitsuda reactivity and Tuberculin and SPA reactivity (in the same individuals) respectively, in the whole population tested. TABUS 3. Bivariate frequency distribution of tuberculin and Mitsuda reactions in the same individuals, in the whole population MITSUDA REACTION SIZES — WITH 0 1-2 3-4 5-6 7-8 9-10 11-12 0 26 6 32 34 40 5 1 1-2 2 5 8 6 2 3-4 3 2 3 to 3 5-6 1 3 5 3 2 2 7-8 3 1 1 5 4 4 I 9-10 1 2 8 3 3 1 11-12 2 2 6 8 3 1 13-14 4 7 I 3 15-16 3 1 6 4 5 17-18 2 2 4 3 1 19-20 1 2 z a a S f- Individuals not included in the Table had reaction si7es — Tuberculin/ Mitsuda (respectively) of, 22/8, 22/12, 24/6 (two reactions), 26/6 and 28/4. TAME 4. Bivariate frequency distribution of tuberculin and SPA reactions in the s,imc individuals, in the whole population SPA REACTION SIZES WITH 0 0 28 1-2 8 3-4 5-6 7-8 1 9-10 11—12 1 13—14 15-16 1 17-18 19-20 0 1-2 3-4 5-6 7-8 9-10 11-12 13-14 15-16 17-18 19-20 3 1 I 2 4 3 . 2 5 1 • 1 2 j - 4 1 2 5 u a H—12 I 1 l g 13—14 1 I I Individuals not included in the Table had reaction sizes — Tuberculin/ SPA (respectively) of, 18/24, 22/0,22/12, 22/14, 24/0 and 24/10. RELATIONSHIP BETWEEN REACTIVITIES TO M. LEPRAE AND TUBERCULIN 43 In Table 5 is presented die correlation betweeen reactivities to antigens o f M. leprae and tuberculin in the whole population studied, irrespective o f location or B C G status. (Fernandez and Mitsuda reactivity in relation to tuberculin could be studied at two locations and SPA with tuberculin in one location only). While all correlations are highly statistically significant ( a > 0.01), the strengdis o f correlations (r 2) were low, the closest correlation being between Fernandez and tuberculin reactivity and least between SPA and tuberculin reactivity. TABU 5. Correlation between different types of reactivity to antigens of M. Leprae and tuberculin of the whole population Correlation between Constant (S.E.) Coefficient (S.E.) r2 F value Fernandez vs. 2.8044 0.4517 0.32 48.816** Tuberculin (0.4004) (0.064) Mitsuda vs. 2.5695 0.4581 0.22 16.661** Tuberculin (0.6927) (0.1122) SPA vs. 3.634 0.3478 • 0.15 15.57** Tuberculin (0.779) (0.088) S.E. — Standard error (within parenthesis) ' Statistically significant at an a ^ 0.01 level TABLE 6.* Correlation between different reactivities to antigens of M.-Leprae Fernandez and Mitsuda) and tuberculin,considcred in relation to non-reactor/rcactor status of Fernandez, Mitsuda, and SPA reactivities (levels of significance) Femandezvs. Tuberculin Mitsuda vs. Tuberculin SPA vs. Tuberculin* where where where Subset Fis NR/R Mis NR/R Mis NR/R Fis NR/R SPA is NR/R Of whole population N R NS ** NS NS NS R 90 •* •* ** . ' .NS . Effect of BCG NR J* BCG—ve \ B C G + v e » NS •a NS NS NS NS NS • NS ' status R f BCG—ve ( B C G + v e DO « « NS • « NS • NS * NS • NS o > 1 N R ^ PusscUawa l_ Pedro NS NS * NS NS NS NS NS Effect of 8 fPussellawa I Pedro « »» NS NS NS NS NS NS location N R fPussellawa \ Pedro NS NS 0 « * NS NS + 8 R fPussellawa \ Pedro • NS •• NS _ » 9 . NS • + Results with SPA and Tuberculin were available only for a.single location at Mahagastota; NR-Non reactors R — Reactors; NS — Not significant; * — Significant at ^ 5% level; ** — Significant at ^ 1 %level. F — Fernandez reactivity; M — Mitsuda reactivity V. PF.MAJAYANTHA. M. R. M. PINTO AND N . Ii. ERIYAGAMA In Table 6 is presented a summary o f the analysis o f correlations (levels o f signi­ ficance) between tuberculin and Fernandez, Mitsuda.and SPA reactivity. The level, o f significance o f correlations had been analysed separately in respect of non-reactor and reactor s t a tus 1 0 , 1 1 o f Fernandez and Mitsuda reactivities and SPA reactivity. The reactor groups o f the whole population seem to show more statistically signi­ ficant correlations titan do the non reactor groups. Also, none o f the cohorts studied in the correlation between SPA and Tuberculin showed statistically significant corre­ lations. No definitive trends seem to appear in relation to the other correlations o f other cohorts studied, i.e. with B C G vaccination status and location. The second type o f analysis carried out was an examination o f the variations of the frequency distributions o f reactions with changes in immunological status o f a deter­ mining reactivity (which could be tuberculin, Fernandez, Mitsuda or SPA reactivity) (Figures 1-3). It is noteworthy that in all cases the frequency distributions showed that the essential difference was that the reactor distributions (when compared with the hon-reactor distributions) showed a reduction o f non-reactors (of the distributions being examined) with an increase o f reactors - a '"shift to the right". The major exception to this rule seems to be at Pussellawa, with the distributions o f Mitsuda reactions which seemed to show no markedly apparent change in pattern o f reactions with immunological status o f tuberculin reaction. i reaction size in mms.' Figs. 1 and 2. For legend see page 48 " ' R E L A T I O N S H I P B E T W E E N R E A C T I V I T I E S T O M. LEPRAE A N D T U B E R C U L I N . 4-5 reaction size in mms. Fig. 3. For legend, see page 48. Another way in which some investigators have examined the relationship between tuberculin reactivity and lepromin reactivity is to compare the occurrence of "positives" (viz. reactors, and non-reactors) in the same individuals with both. Such evaluations arc presented in Tables 7, 8 and 9. The most striking change observed here is the effect of BCG vaccination in markedly increasing the number of reactors with all three reactivities. Apart from this, evaluation of these results do not present definite or specific conclusions regarding the concurrence (or not) of immunological status as assessed with different types of reactivity. 46 V . P E M A J A Y A N T H A M . R . M . P I N T O A N D N . B . E R I Y A G A M A 0 4 8 12 16 0 4 8 12 16 reaction size in mms. Fig. 4 . For legend see page 48 . RELATIONSHIP BETWEEN REACTIVITIES TO M. LEPRAE AM) TUBERCULIN 0 4 8 12 16 20 24 reaction size in mms. Fig. 5. For legend sec page 48. 48 M . PEMAJAYANTHA, M. R. M . PINTO A N D N . B. ERIYAGAMA LIST OF FIGURES FIGURE I. Frequency distribution of tuberculin reactions where Fernandez reaction size was S ;2mm( and > 2 mm. ( - ) (a) at Pussellawa (/>) at Pedro FIGURE 2. Ferqnency distribution of tuberculin reactions where Mitsud.i reaction size was ^ 2 mm. (— and ^ 3 mm.( ) ( 6 mm. ( ) at Pedro (rf) ^ 9 mm. ( ) , and ^ 10 mm. ( ) at Pedro FIGURE 4. Frequency distribution of Mitsuda reaction where tuberculin reaction size was (a) ^ 5 mm. ( ) , and ^ 6 mm. ( ) at Pussellawa (4) ^ 9 mm. ( ) , a-»d > 10 mm. ( ) at Pussellawa W 5 mm. ( ) , and ^ 6 mm. ( ) at Pedro (d) ^ 9 mm. ( ) , and ^ 10 mm. (- ) at Pedro FIGURE 5. Frequency distribution (at Mahagastota) of (a) tuberculin reactions where SPA reactions were <; 4 mm. ( ) , and ^ 5 mm. ( — (fcj SPA reactions where tuberculin reactions were ^ 5 mm. ( — — — ) , and ^ 6 mm. (— (i-) SPA reactions where tuberculiu reactions were ^ 9 mm. ( ) , and ^ 10 mm. RELATIONSHIP BETWEEN REACTIVITIES TO Af. LEPRAE AND TUBERCULIN 49 TABLE 7. Occurrence of individuals with different immunological status with different reactivities (association between Fernandez, Mitsuda and tuberculin reactivities) Percentage individuals Status of tuberculin, ' Where tuberculin reactivity was Where tuberculin reactivity was Fernandez and Mitsuda considered non-reactor if considered non-reactor if reaction reactivities respectively reaction size was 5 mm. or less size was 9 mm. or less and reactor and reactor if 6 mm. or more if 10 mm. or more BCG-vc BCG+vc Total" BCG-vc BCG+vc Total RRR 16 36 2 2 14 19 16 RRN J 6 4 2 2 ' 2 R N N 6 . 4 5 3 RNR 9 11 1 0 7 6 7 N N R 16 2 1 1 7 18 26 21 NRR 2 4 17 2 2 2 7 34 2 8 N R N 10 6 9 11 10 11 N N N 16 4 1 2 16 4 12 R -Reactor; N*Non-reactor; RRR - reactor to tuberculin, Fernandez and Mitsuda respectively, NNN, non-reactor to all three, RRN- reactor to Fernandez and tuberculin and no:i-rcactor to Mitsuda respectively and so on. TABIC 8 Occurence of individuals with different immunological status widi different reactivities (association between SPA and tuberculin reactivities) Percentage individuals Status of tuberculin and SPA Where tuberculin reactivity was Where tuberculin reactivity was reactivities respectively considered non-reactor if reaction considered non-reactor if reaction size' was 5 nun. or less and size was 9 mm. or less and reactor reactor if 6 mm. or more if 10 mm. or more f ' BCG-vc BCG+vc Total. BCG -ce BCG+ve Total. RR ; 2 0 ' 32 ' 2 5 . : 15 ' 27 R-N • • 9 - .6 8 - " ' " . 3 ' ' "3 " nr.;- . . . . .9. 2 4 1 5 . - 19 32 2 4 NN 61. . 3 2 5 2 ' 61 3 8 ' ' 52 '.' '. R - Reactor; N- Non-reactor; RR - reactor to tuberculin aid SPA respectively, NN - Non-reactor to both, RN - Reactor to tuberculin and Non-reactor to SPA, NR - Non-reactor to tuberculin and reactor to SPA 50 V . P B M A J A Y A N T H A , M . R. M . P I N T O A N D N , B. E R I Y A G A M A TABLE 9 Occurnce ot' individuals witli d ft'erem immunological status to different reactivities-analysis of variability of results Percentage Individuals When a tuberculin non-reactor When a tuberculin non-reactor is a reaction of 5 mm. or is a reaction of 9 mm. or less, and reactor is 6 mm. less, and reactor mm. or more or more BCG-vc BCG+ve Total BCG-vc BCG+vc Total Status of Tuberculin, Fernandez and Mitsuda reactivities together - where status of each Agree R 16 36 34 14 19 29 N 16 4 17 4 • Disagree R 18 17 66 14 8 71 N 50 44 55 69 Status of Tuberculin and Fernandez reactivities only - where status of each Agree R 18 41 6G 16 21 50 N 16 • 25 34 29 Disagree R 32 11 34 12 6 50 N 34 23 38 44 Status of Tuberculin and Mitsuda Agree R 25 48 57 21 25 45 N 26 9 27 13 Disagree R 9 6 43 6 2 55 N 40 37 44 61 .V and y — here refers to non-reactor or reactor status of tuberculin reactivity- It has been shown that the smaller reaction sizes to tuberculin are due to "non­ specific sensitisation" - the cross reaction elicited, due to infection by some other my- cobacterium. The problem o f nonspecific sensitisation is complex, with probably multiple factors influencing its occurrence. O f tuberculin reactions, the intermediate reatcion sizes (of2-9 mm) are considered to be largely due to such.non-specific reactions. The variation o f Fernandez and Mitsuda reaction patterns with such non-specific 'sensitisation. was examined by the comparison o f frequency distributions o f the latter in those whose tuberculin reactions were (a) ^ 2 , mm(b) 3-9 mm, and(c) s* 10 mm. These patterns were examined with cohorts o f different B C G status, where numbers permitted, at Pussellawa and Pedro. With Fernandez reactivity, which could be examined in both areas, with increase in tuberculin reaction size, there seemed to be an increase in Fernandez reaction size as well but there was no change which could be said to, be peculliar to (or specific to) the non-specific sensitisation group. With Mitsuda reactivity, the comparison was possible only at Pussellawa, where the Mitsuda pattern was similar in all three groups o f tuberculin reactivity. RELATIONSHIP BETWEEN REACTIVITIES TO M. LEI'RAE AND TUBERCULIN. 51 DISCUSSION ' The results presented here show a statistically significant, though at a low level(r 2) o f correlation, between tuberculin reactivity and reactivity to antigens o f M. leprae. For correlations between tuberculin and antigens o f M. leprae,!2 was somewhat lower than that for correlations between reactivities elicited by antigens o f M. leprae. The relationships .between tuberculin and other reactivities were also confirmed by the examination o f frequency distributions o f different reactivities in relation to immunolo­ gical status (non-reactor/reactor) o f another type o f reactivity. The examination o f the concurrence o f the immunological status o f different types o f reactivities however failed to show any definite results except to demonstrate the influence o f B C G vaccina­ tion status in increasing the occurrence o f reactor status. Further it was not possible to show any special influence o f "non-specific sensitisation" as demonstrated by tuberculin testing on the patterns o f Fernandez and Mitsuda reactivity. Direct comparison o f the data o f this survey with that o f other workers is not possible, because o f differences o f the criteria used for evaluation (different criteria used for reading and or delineating "positive" and "negative") and also o f the antigens used. However the conclusions reached by different workers have been variable. With "early lepromin reactions" read at 48 lir, as compared with tuberculin reactions, Fernandez3 reported agreement and disagreement, as regards "positivity" o f each type o f reactivity, in equal numbers; Kupfer 6 similarly found no correlation. Widi the late lepromin reaction, a.correlation was reported by Kupfer 5 and Doull et al2 while Kooij and Rutgers 3 found no correlation. ACKNOWLEDGEMENTS This investigation was supported in part by a grant from the Leprosy Section, World Health Organisation, Geneva. W e wish to thank Mr. D . W . Dissanayake, Medical Assistant, Mahagastota State Plantation, Nuwara Eliya, and Ms. M. G. Kusumalatha, Medical Assistant, Trafford Hill State Plantation, Galagedera, for assistance in carrying out this study; and Ms. T . M . Gamage and Ms. M.Wijekoon o f the Department o f Microbiology, Faculty o f Medicine, University o f Peradeniya, Peradeniya, for assistance in data analysis. 52 M . P E M A J A Y A N T H A M.tt.M P I N T O A N D N . 0 . E R I Y A G A M A R E F E R E N C E S 1. CHACSSNAND, R . (1953). ' Tuberculosis and Leprosy- : Mutually antagonistic diseases. Leprosy Review. 23 ,90-97 2 . Douu, J.A., GUINTO, R . S. and MABALAY, M . C. (1959) . "The origin of natural reactivity to lepromin. The association between die Mitsuda reaction and reactions to graded doses of tuberculin. Imertuuioiial Jounuil ol Leprosy 27, 1 ,31-42. 3- FERNANDEZ. J. M . M . (1940) . The early reaction induced bv lepromin. International Journal of Leprosy 8, ( 1 ) , ! -M. 4. FERNANDEZ, J. M . M . (1957). Leprosy aid tuberculosis. Antagonistic diseases. A. XI. A. Archive: of • Dermatology 73 , 101-106. ' 5-' KOOIJ, R . and RUTGERS, A. W . F. (1958) . Leprosy and Tuberculosis. A comparative study with the aid of skin tests with tuberculin, killed BCG, and the Dlnrincndra lepromin in South ' African llanrus. Inuruiitional Journal of Leprosy 26, ( l ) . 24 -40 . 6- KuiT-EK, S.W.A. (1955) . Tuberculin and lepiomiu sc.isitivitv in the South African Bantu. A pilot survey. Lrnca 996-1001. 7- LOWE,J and MCNULTY, F. (1953) . Tuberculosis and Leprosy: Immunological studies. Leprosy Review 24, 61-90 8. PEMAJAYANTHA, V . , PINTO, M . R . M. and EUIYAGAMA, N . B. (1986).. Correlation between reactivities to Lepromin A aud Soluble Protein. Antigen of Mycobacterium leprae. Asian Pacific Journal of Allergy and Immunology (in press). 9- P w r o , . M . R . M*(1974) ' . A Study of some aspects ofiheMycobacterioscs in Sri Luika. Ph.D. Thesis. University of Ceylon, Peradeniya, Sri Lanka. 10. PINTO, M . R . M . , ERIYAGAMA, N . B. and PEMAJAYANTIIA, V . (1986) . Studies of reactivity to antigens of Mycobacterium leprae anions some Sri Lankac population groups. I . Reactivity to Lepromin A. Leprosy Review (in press). 11. PINTO, M . R . M. , ERIYAGAMA, N . B. and PsMAjAYANrHA, V . (1986). Studies of reactivity to antigens uf Mycobacterium leprae among some Sri La'ikau population groups. I I . Reactivity to a Soluble Protein Antigen of Mycobacterium leprae. Leprosy Review (in press). 12. REES, R . J . W . (1964) . The significance of die lepromin reaction in nun. Progress in Allergy 8,224-258- 13. WADE, H . W . (1950) . Reactions to tuberculin in leprosy. International Journal oj Lcpivsy 18, 373-388. 14. WORLD HEALTH ORGANISATION (1963) . The WHO siandari tuberculin test. World Health OrganisJticn Document W . H . O . ' / T B 7 Technical Guide/3. Geneva: W H O .