Ceylon J. Med. Sci. 32 ( N o . 1, June) 1989 pp. 9—21. 
i 

1 Department of Microbiology 2 Department of Economics, University of Peradeniya, Peradeniya. 

A N EVALUATION OF THE USEFULNESS OF DIFFERENT DIAMETERS OF 
SKIN REACTIONS ELICITED BY MYCOBACTERIUM LEPRAE ANTIGENS 

i 

M . R. M. PINTO1, N . B. ERIYAGAMA1, T M . GAMAGE1, 

S. SlVARAJASINGHAM 2 

SUMMARY . The advantages of the transverse, vertical and mean diameters for the measurement of skin 
reactions with M. Leprae antigens were evaluated. It was decided that the vertical and mean diameters 
offered no improvement over the transverse diameter and hence the use of the" latter is recommended. 

INTRODUCTION 
* 

When testing for cell mediated immune responses with skin tests, the indura­
tion reaction produced is usually of a round or oval shape. With tuberculin reactions 
elicited by the Mantoux tests, the authoritatively recommended reading is of the palpated 
transverse diameter ( l ) . However, with skin testing, different workers have also used (or 
recommended), other dimensions in the reading of skin tests: those that are,' or have 
been used, or could be used in the reading of skin tests in man and animals include the 
transverse, vertical and mean diameters, the surface area and increase of skin thickness. 
The measurement of the increase of skin thickness is a practice usually limited to use 
with animals—and in thin and highly sensitive human skin, this procedure is likely to 
cause unacceptable pain. 

Thus, in the reading of skin tests with antigens of Mycobacterium leprae, different 
workers have used different dimensions in the measurement of evoked skin reactions. 
The Leprosy Congresses in Madrid (2) and in Tokyo (3) recommended reading the dia­
meter (presumably the transverse); the latter also recommended that, "When lesions are 
not round,'two measurements should be made and averaged". TheW.H.O. Expert 
Committee in its First Report (4) and in the Second Report (5) made similar recommen­
dations. In the examination of published papers in English with Lepromin skin tests it 
was found that there has been no uniformity in the reading of results. Of 36 papers 
(published between 1934 and 1986) reporting results with Mitsuda reactions, one each used 
the transverse andj greatest diameter ; 6, the mean or average diameter ; IS, read a 
"diameter",-uncertain which ; and 13 made no mention* at all of which measurement 
was read. Of 12 papers reporting results with the Fernandez reaction, one used the 
transverse diameter; 3, the mean or average diameter; 5, a diameter, uncertain which; 
and 3 made no mention of the measurement read. An antigen introduced recently 
for skin testing in leprosy research, the Soluble Protein Antigen (SPA), does not have 
an extensive body of literature available for such examination. 

o 



10 M. R. D. PINTO, N . B. ERIYAGAMA, ET AL. 

0 

In evaluating the results of skin tests with M. leprae antigens the trend has been to 
classify results as positive or negative. While such classification may have advantages 
if uniformly applied in the clinical setting, it may not be as useful in the epidemiological 
setting, where the exact measurement of readings in mms and the graphical presentation 
of the distribution of readings would yield more information. Such presentations would 
be also of value in the comparison of data from different workers, provided the readings 
made were the same. Thus it.is of importance that there be uniformity in the readings 
made by different workers. Such uniformity has been achieved with tuberculin testing (1) 
but unfortunately is not seen in the reported results of skin tests with M. leprae 
antigens. , 

An examination of the literature on skin testing with mycobacterial antigens did 
not reveal any scientific investigation or data analysis made to decide which diameter of 
skin reaction'measurement would give the best results. In this paper is presented an 
analysis of the different diameters which could be used in the evaluation of skin test 
results evoked by Af. leprae antigens. ( 

V 

MATERIALS A N D METHODS 

The results presented here are from skin tests carried out on three population groups 
in different localities, in two geographically different areas. They consisted of adults , 
(12 years of age and above ) , of both sexes, and included both the BCG vaccinated and 
non-vaccinated. The characteristics of the populations and test methodologies have been 
described in 'detail elsewhere (6,7). Briefly the methodology used was as follows. The 
test antigens used were Lepromin A (with a bacillary. content of 3 or 4 x l 0 7 bacilli/ml) 
and a soluble protein antigen of M. leprae (SPA) with a protein content of 10 /ig/ml. 

The antigens were adminstered intradermally using standard techniques, on the volar 
aspect of the forearm at different sites.. The reactions were read, for Fernandez'reac­
tivity at 48 hours, and SPA reactivity at 72 hours, both as the maximum transverse and 
vertical diameters palpated, and Mitsuda reactivity at 28 days as the maximuni'transver'se 
and vertical diameters of the nodule palpated.' The transverse (a) and vertical (p) diameters, 

(a + b) 
and the mean — - — diameter, were used in analysis in this study. The surface area was 

a b 
excluded, as both the mean and surface area ( 7 r x - x - ) are expressions of functions of the 

transverse and vertical diameters, arid theiefdre the decision was to use the mean 
diameter alone, with the more manageable smaller values. 



M. LEPRAE SKIN TEST MEASUREMENTS 11 

The simplest way of examining the relationships between transverse and vertical 
diameters would be to look at bivariate frequency distributions. Such distributions for 
whole populations with Fernandez, Mitsuda and SPA reactivities are presented in Tables 
2 and 3. An examination of this data showed that the reactions tended to be oval 
with the larger reactions, commonly with the vertical diameter exceeding the transverse 
diameter. This change is more marked with Fernandez and SPA reactions and not as 
prominent or hardly seen with Mitsuda reactions, which tended to be smaller than the 
former (Table 4). There was a close correlation between the transverse and vertical 
diameters with all three types of reactivity (Table 5). 

TABLE 1. Bivariate frequency distribution of transverse and vertical diameters of 
Fernandez reactions in the whole population tested. 

Vertical. 

0 

1 

2 

3 

4 

3 

6. 

7 
e 
g . 8 
• 
i 9 

1 0 

t t 

1 2 

13 

1 4 

1 5 

16 

1 7 
1 8 

19 

0 t 2 3 4 3 6 7 B 9 1 0 I I 1 2 ' 3 1 4 1 3 16 1 7 >8 1 ) 20 2 1 2 2 23 24 

\ 0 

% 4 6 1 8 " 

1 ' 2 5 1 9 J 

1 KZJ 1 8 -

3 \ s 3 2 

6 N * 2 5 

\ 30 2 
N V 6 21 9 

4 X 1 4 4 

\ 24 2 

\ t 1 3 4 
\ 

\ 1 0 T 1 

\ 4 - 1 0 1 

\ Z 2 2 2 1 I 

\ t . 1 

\ 
1 I 

I 

RESULTS 



12 M. R. M. PINTO, N. B. ERIYAGAMA ET AL. 

TABLE 2. Bivariate frequency distribution of transverse and vertical diameters of 
Mitsuda reactions in the whole population tested. 

' Vertical 

0 1 * 3 4 5 6 7 8 9 10 11 12 13 14 15 

* 1 
' 1 !0X 1 

t n » 8 

a X » 

N » l 17 

8 ^ I S 1 
1 12 < 

a 10 > i 

i \ • 
i i 

3 
2 
1 

0 . 

\ 
X 

I 
i 
i 

\ 
1 * 



„ Af. LEPRAE SKIN TEST MEASUREMENTS ' 13 

TABLE 3. Bivariate frequency distribution of transverse and vertical diameters of 
' SPA reaction in the whole population tested. 

Ver t i c i l 

0 1 2 3 4 5 6 7 0 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 . 1 7 I B 1 9 Z Q 2 1 W 23 2+ 2 5 

ON** 
t 

2 
V 

2 V v 

I \ 5 s 
1 \ ? 

a 83 

\ t 4 

a^.17 

e i« a 9 
^ 9 15 3 

5 \ 6 
\ 168 5 

5, I J 1 

\ 5 1 

V 2 2 9 

1 t 
1 



14 M. R. M. PINTO, N . B. ERIYAGAMA ET AL. 

T A B L E 4. Relationship of transverse and vertical diameters of different sized groups, with different types of, 
reactivity 

Transverse 
reaction sizes 

• o f 

Type of 
. reactivity 

Percentage reactions Transverse 
reaction sizes 

• o f 

Type of 
. reactivity 

Where transverse 
and vertical diameter 

are equal 

Where transverse Where transverse 
is greater than diameter is less 
vertical diameter than vertical 

diameter 

< 5 mm. Mitsuda 
Fernandez 
S P A ' 

63 
52 
75 

6 31 
4 • 44 
7 18 

6—10 mm. . Mitsuda' 
Fernandez 
SPA 

61 
39 
45 . 

5 34 
2 59 
5 50 

> I f mm. Mitsuda 
Fernandez 
SPA 

47 
16 
32 

32 21 
5 79 
4 . , 64 

T A B L E 5. Correlation between transverse and vertical diameters with different types of reactivities 

Type of reactivity Coefficient of correlation ( r ) . Probable error Confidence interval 

Fernandez 
(n—647) 

0.75" 0.01 0.71, 0.78 

Mitsuda 
(n—533) 

0.9876 O.O066 0.9852 , 0.9896 

SPA 0.81 0.01 0.78. 0.83 
fn—504) •• . 

(n—number tested and read) 

The major purpose of doing a skin test in an infectious disease is to distinguish 
between those who are "reactors" ("positives") and "non-reactors" ("negatives"). This 
could be best visualised by examining which measurement would give the best separations, 
viz, transverse or vertical diameters, or the mean. Frequency distributions of each with 
different cohorts of tested individuals (from differing localities and of differing BCG 
vaccination status) were therefore constructed in graphic form (Figs la—f, 2a—f and 3a—d). 
The distributions in general tended to be bimodal with a mode of non-reactors (on the left) 
and another of reactors (on the right). They showed that, with Mitsuda reactivity, the 
distributions of transverse, vertical and mean diameters showed no differences at all, 
while with Fernandez and SPA reactivity, there were in a few instances, a shift o f the 
distribution of reactors to the right, of vertical and mean diameters, showing the presence 
of larger readings. The reaction size for the differentiation of negative or non-reactor 
from positive or reactor, would be between the non-reactor and reactor components or 
of the distribution. In order to examine this, enlarged and more detailed graphical re­
presentations of the area of the frequency distributions between 0 to 6 mm for Fernandez 
reactivity and Mitsuda reactivity, and 0 to 10mm for SPA reactivity were constructed 
(Figs 4a—f, 5a—f, 6a—d). These however, did not reveal any significant differences 
between the transverse, vertical, and mean reaction diameters, in the reaction distributions, 
that may affect this differentiation. • . 



Reactcn size in rims. 

Frequency distribution of Fernandez reactions with different diameters (transverse 
vertical , mean — — • — ) in the 

(a) B C G - ve at Pedro 
. (b) B C G + v e at Pedro 

(c ) B C G - ve at Mahagastota 
(dl B C G + v e at Mahagastota 
(e) B C G - ve at Galagedara 
( / ) B C G + v e at Galagedara 



Reaction size n mms. 

Figure 2. Frequency distribution of Mitsuda reactions with different diameters (transverse— 
vertical , mean — ) in the 

(a) B C G - v e a t Pedro (b) B C G + v e at Pedro 
(c) B C G - ve at Mahagastota ' id) B C G + v e at Mahagastota 
(e) B C G - v e a t Galagedara tf) B C G + v e at Galagedara . 





M . R, M. PINTO, N . B. ERIYAGAMA ET AL. 

Figure 4. Frequency distribulion of Fernandez reactions of the smaller sizes (0—6 mm), with 
different diameters (transverse , vertical , mean — ) 

{a) BCG 

(c) BCG 

(e) BCG 

- ve at Pedro 

- ve at Mahagastota 

- ve at Galagedara 

(b) B C G + v e at Pedro 

(d) BCG + v e at Mahagastota 

( 0 B C G + v e at Galagedara 



M. LEPRAE SKIN TEST MEASUREMENTS 

(a) BCG 
(c) BCG 
(e) BCG 

- ve at Pedro 

- ve at Mahagastota 

- v e a t Galagedara 

(6) B C G + v e at Pedro 

(d) B C G + v e at Mahagastota 

( / ) B C G + v e at Galagedara 



M. R. M. PINTO, N. B. ERIYAGAMA ET AL. 

Reaction size in mms. 

figure 6. Frequency distributions of SPA reactions of the smaller sizes (0—10 mm and 0—8),. with 
different diameters (transverse , vertical , mean ) in the 

(a) BCG - ve at Mahagastota (6) B C G + v e at Mahagastota 
(c) BCG - ve at Galagedara (d) B C G + v e at Galagedara 



M. LEPARE SKIN TEST MEASUREMENTS 21 

DISCUSSION 

The results presented here all indicate that, with Fernandez and SPA reactivity, 
with the larger reaction sizes, the Vertical diameters tended to be larger than transverse 
diameters. The use of the vertical diameter, however, did not in any way contribute to a 
better interpretation of the skin test result, or to a better differentiation between "negative" 
and "positive". Further, the use of the mean reaction size did not in anyway provide 
any improvement over the use of the transverse diameter alone. Thus, the effort 
spent in calculation of the mean was not worthwhile. With Mitsuda reactivity, the 
reactions tended to be rounded and smaller and hence the difference between transverse 
and vertical was not noticeable. 

In conclusion, we recommend that the measurement of the transverse diameter alone 
would be adequate in the reading of skin reactions with M. leprae antigens, as is the 
practice with tuberculin reactions. 

ACKNOWLEDGEMENTS 

This investigation was supported by grants from the Leprosy Section, World Health 
Organisation, Geneva, and the University of Peradeniya. 

We wish to thank Ms. M . Wijekoon of the Department of Microbiology, Faculty 
of Medicine, University of Peradeniya, Perandeniya, for clerical assistance. 

REFERENCES 

1. World Health Organisation. The WHO Standard Tuberculin Test. WHO/TB/Techn. Guide/3. 
Geneva : World Health Organisation 1963. 

2. Transactins of the VI International Congress of Leprology, Madrid, October 1953. Madrid: Memoria 
del VI Congreso International de Leprologia 1953 ; 102—103. 

3. Transactions of the VII International Congress of Leprology, Tokyo, November 1958. Tokyo: 
Tofu Kjokai 1959 : 464—465. 

4. World Health Organisation Expert Committee on Leprosy, First Report, Technical Report - Series 
No. 71. 
Geneva : World Health Organisation 1953 : 25—26. 

5. World Health Organisation Expert Committee on Leprosy, Second Report, Technical Report Series 
No. 189. 
Geneva : World Health Organisation 1960 : 8. 

6. Pinto MRM, Eriyagama NB, Pemajayantha V. Studies of reactivity of some Sri Lankan population 
groups to antigens of Mycobacterium leprae. I. Reactivity to Lepromin A. Leprosy Review 1987; 
58 : 105—118. 

7. Pinto MRM, Eriyagama NB, Pemajayantha V. Studies of reactivity of some Sri Lankan population 
groups to antigens of Mycobacterium leprae. II . Reactivity to soluble protein antigen of Mycobac­
terium leprae. Leprosy Review 1987; 58: 216—226.