Lepr Rev (1998) 69, 128-133

Choosing the decolourizer and its strength to stain

Mycobacterium leprae. Does it actually matter?

R. DE SOLDENHOFF, M. HATTA &

T. WELING SIRO
NSL-South Sulawesi Leprosy Control Project; Provincial Health

Services, S'ollth Sulawesi. Indonesia;

*Departll1ent of Microbiology, Medical Facillty of !Ju.\WllIddin

University. Ujwlg Pandang, Indolles;'([

Accepted for pubhcation 15 April 1998

Leprosy bacilli are more easily decolourized during staining than
tuberculosis bacilli. so a weaker concentratIOn of decolourizer is usually recom
mended. In Indonesia. the same 'strong' decolourizer is used for identifying both
urganisms, Tn a study to compare the results llsing different concentrations of
different decolourizcrs, no differcnce could be found in thc bacterial index (HI). It
is suggested that the same staining technique can be used for tuberculosis and leprosy,

SIIII/II/ary

Introduction
Ideally, all leprosy patients should have one skin smear examination before starting
treatment, if reliable facilities are available.' Thc Ziehl-Neelscll method involves slaming
with carbol fuchsin, followed by decolourization then counterstaining with methylene
blue." If thc decolourization process is excessive, bacilli may be rendered invisible. Concerns
regarding which deeolourizer to use, what strength and for how long, have resulted in several
in vestigations. 3-6
In Indo[]esia, the same staining technique is normally usee! for leprosy and tuberculosis,
despite the widely accepted opinion regarding the reduced acid and alcohol 'fastness' of
MycohacteriulIl leprae 2 The objective of this study was agaill to find out if (i) different
decolourizers and (ii) different concentrations of the same decolourizer had an effect on the
B I of skin sl1Icars.
Although the Ziehl-Neelsen stain is used universally to stain /vI. leprae, there are many
modifications. 7 The main difference between the technique for leprosy and tuberculosis is the
strength of the deco!ourizer, but in addition, the duration of decolourizing varies. For
tuberculosis. this is IJsually either 25% sulphuric acid or 3% hydrochloric acid in 70%

Correspondence to: NSL-Sulsel Leprosy lOl1lrol Project, PO 130.\ 10 11. l'JlIng Panoang. Sulawesi. Indrlllesia,
e-mail: richanis(J-i.oupg.lllega.llet.id

128

0305- 7518/98106') 118+06 S I ,(}()

r,oi l.epl<t

Deco!rJ/lri;er {Illd its strellgth ill the Z;V stoill

129

ethanol for 3 min.~ For leprosy, decolourization or differentiation using I % hydrochloric acid
in 70% ethanol is favoured by l11ost,2.79.10 but 5% sulphuric acid by some. I I Dharmendra
surprisingly recommends 3% acid aicohol.12 The countcrstain is llsually mcthylene blue.
although the concentration varies between 02% II and 1%.6 Vettom and Pritze found 10
different techniques from 29 projccts, the most notable variations being in decolourization
time. 13 They ranged from I % acid alcohol for 5 s to 20% acid alcohol for I min and from 59(
sulphuric acid for I min to 25% sulphuric acid for 10-20 min. 1:1
Despite this, several rcsearch laboratories use the same concentration of decolour:ler for
leprosy as for tuberculosis (A. McDougall, personal communication). The Indonesian Icprosy
manual describes two possible Ziehl-Neelsen methods. one with acid alcohol (concentration
not specified) for 3-5 s and the other with 25% sulphuric acid for R s. J.j

Materials and methods

Skin smcars were takcn from 40 l11ultibacillary (MB) leprosy patients. with a range of Bl. The
selection was by the technician who took the smears, did the staining and rcad the results. The
patients were registered cases and were on WHO ivlB multidrug therapy (MDT). I All patients
had smears taken from four standard sites with four identical slides being made from eClch
patient, each slide with the foul' sites on it.
Apart j'rolll the decolourizcr, the same Ziehl-Neelscn mcthod was used. ls The decolour
izers were I 'k hydrochloric acid in 70'k ethyl alcohol (HCI). 3 ck hydrochloric acid in 70%
ethyl Cllcohol, 5'k sulphuric acid (H 2 S0 4 ) and 25% sulphuric acid. Slides were stained in
batches. Carbo I fuchsin was filtcred onto the slide, allowed to act for 2 min, heated gently
until steam rosc, allowed to act for a furthcr 10 min. washed with tap watcr until clean,
transfcrred to a staining rack and dipped in decolourizer for 8 s. If any slide was still red, this
was repeated, washed, and counterstained with O'3'k methylene blue for I min. Most slides
only rcceived one nllIllersion ill the decolourizer.
The intention was that the examination of the smears would be blind, but it was possible
to distinguish immediately between thc slides decolourized with HCl and those with H 2 S0 4
by microscopy of the smear. It was not possible to distinguish between I r,; and 39c HCI or
between 5% and 25% H 2 S0 4
The BI was reported for each of thc four smears on the slide and the avcrage was
calculated. The morphological index (MI) is not routincly used in Indonesia.
The results were compared using the Student's paired t-test](i Differences were
considered signiflcant at the 95% level of confidence.

Results
Forty paticnts had slit skin smears taken, resulting in 160 slides. Each slide had four smears,
except in the casc of patient 6. who fainted aftcr smears had been taken from two sites (ears).
Thirty-four patients were positive with at least onc staining technique. An average BI was
cCllculated for each of the 160 slides (~ee Table I). Comparison was also made lIsing
individual smcars,
Eleven patients had a higher BI with 1% HCI than with 3%. Twelve patients had a lower
BI with 1% HClthan with 3%. Eight patients had a higher BI with 5% H 2 SO.j than with 25%.
Thirteen patients had a lower BI with 5% H 2 SO. 1 than with 25%.

130

R. de So/dell hoff et ai.

When looking at low BIs (i.e. >0 but <2), the same pattern was obsened. El.;:ven patients
had a low BI using 1% He!. In two cases, the BI was greater using 1%, and in five cases, the
BI was less. Twelve patients had a low BI using 5% H::>SO-\. In three cases, the BI was greater
using 5% and in five cases. the B1 was lower with 5%.

Tabl!! 1. BI

Patient

r~sul!s

I c,;. acid
alcohol

for .)(J patients using four decolourizers

3<;;C acid
alcohol

Difference

Yk sulphuric
acid

_._-------------- - - -

.~~-

----~.-"--

...

25'1lc sulphu!ic

acid

1,7)

225

2
3
4

1
'),75
1

075

45

-\

(J'25
0

5
6

()

()

()

.)

.)

175

375
45
15
2'75
()'5
1
0
45
175
25
125

0
()'25
025
075

Il
3

4
275
275
I

()

()

()

()

-125

()

4'~5

35

~75,

-125
225

(lS

05

().)

{l

I"
025

35
0

')

10
11
12
13
14
IS
16
17
18
19
20
21
:22
23
24
25
26

()

()'l)

()

-0<'5

125

()

275

--I
05

()

125
0
32.'
325
2'5
I

()
(J

325
-3,25
-25

025
[)'75
0
0
I
-0'25
()S

()'25

I
0
-0'25

()

-15
175
25
()'25

IJ

()

{l75
-2

0
1'~5

O'~'5

025

()

(j

(lS

()

I)

()

()

()

3~5

-1
225
375
.+75
375

0
-0:'5

025
0
()'25

()

()

()

0
375

(J

()

l.~

35
3
-125

375
4-5
225
25
4'25

025
0
3'75
375
0'75

075

.)

375

(j

3(,

3'5

37

:2

38

0
05
475

-1'25
35
125
05
-125

p value

()

I
--()25

-025

45

()

SO

()

()

-_.,----.._. - ----

()

:;
OS

A\'erag~

()

[)

025
025

0
0

125

Total

05
-I
-()25

4
4
125
1'75
0'5
125

125
4
425

025

()

375
475
35
35
4
5

40

.)

-"-------

..

.+75

28

39

(J'25
-I
-0'75

175
075
425
075
0

()

~9

(J

-025

.+75

27

30
31
32
33
34
35

35
375
175
25
i 5
175

-05

Difference

--~------

Di n~reilce bel wcell

I (;~ Hel & 5'.
sulphuric acid

925
231

97
243

-075
-O'l5

-0'25
-0,5
()'25
0
075
0
-075
-15
-125

325

()

()S

4'75

3'75
375
125
125
475

-4,5

7(j'25

8825

IYI

221

(0 III
050
<02

()

D25

1:'5

05
0

05
-{)25

()

()

-05

1
0
-025
-175
-075

()

0
()

-05
-1,25
()

-12

OJ
II

1625

(()30)

0-41

Ot).!

(J'98

<(Jl

<U'02

---.~-

f)ec(}I(}lIri~er

([lid its strellgth ill the Z-N staill

131

The averagc of all thc Bfs using 1% HCI was 231 (with seven negatives). The average HI
using :1% HCl was 2A:1 (with six negatives). The difference between the BIs wa~; not
statistically significant (p < 02 and >0'1). The average BI using 5% H 2 S0 4 was 191 (with 11
negatives). Tlw average HI using 25r;{, H 2 S0 4 was 221 (with nine ncgatives). Again. thc
difference between the Bis is not statistically significant (p < 0'1 and >OOS). When the
differences between the BIs \\'ith 1% acid alcohol and 5% sulphuric acid were examined.
there was an average fall of 041. This was signillcant at the 29c level (p < 0'02). suggesting
that there might be a real decrease in BI when using sulphuric acid.
There was generally good correlation between the four slide results from each patient. An
exception was founli in patients 15, 16 and 17. where the slides stained with 5(1, H 2 SO, were
negative, whereas the other three were moderately or strongly positive. The most likely
explanation for this was that these slides were wiped clean on the wrong sick of the slide.
hence removing all four smears from the slide. If the results from these three patients are
excluded, the average BIs arc 2'26 and 236 for acid alcohol and 206 and 2'14 for sulphuric
acid. The differences between these ligures (betwecn the two acid alcohols, the two sulphuric
aeids and between I(} acid alcohol and Y/( sulphuric acid) all fail to re,1Ch statistical
significance (Ii < ().). I' < 05 and IJ < ()l. respectively).
In 3g of 15~ sites cxamined, there was a variation of two or morc BI units (i.e. Illore tlwn (I
1O-fold diffelTnce in the number of bacilli) from the same site using differcnt dccolourizers.
No pattern was obscrved.
Thc slides which were stained using an acid alcohol uecolourizer were generally easier to
examine than those llsing sulphuric acic!. Irrespective of concentration. the bacilli were
sharper, clearer, more reel. the background had accepted more blue for differcntiation and
there was little or none of the background pink 'fuzz' that is a feature of the slides which were
decolourized llsing Slfc or 257<. H 2SO,.

Discussion
POSITIVITY

This study failed to show that a stronger decolourize. resulted in a lower gT. The BI was
slightly higher with the stronger concentration. but not reaching statistical significance (see
Figure 1). This was also noted when looking at the results from patients with a low 2) BI.
The Bl was slightly lower with sulphuric acid, this reaching significance only when the
questionable data from patients 15, 16 and 17 were included.
POSSIBLE BIAS

All patients were purposefully selected. Nu attempt was made to havc a rcpresentative sample
of leprosy patients. new or otherwise. Since the purpose of the study was to comparc results
patient by patient, analysis using an average BI for all patients per technique is open to
question. The difference in BI from the same site could also be affected by th'~ quality of the
smear and within-observer variation. Three patients had results which suggest a major
laboratory error. Analysis was also performcd excluding results from these three patients.
DOES IT REALL Y

~!ATTER'!

Most new leprosy patients can be competently diagnosed and commenced on appropriate

132

R. de So/dellllOi)' el ill.

Patients

1-.-1% HCI +3% HCI +5% H2S04 ---25% H2S04l

Figure 1. Comparison of Bl results using four decolourizers.

treatment without :1 skin smear. However, there are some patients who present with single or
few lesions, but who have early multibacillary disease. 17 There are other patients, either new
or old, who have no clearly demonstrable clinical cardinal signs, but who have a positive
smear. There are suspicions that some multi bacillary patients, who are released from fixed
duration treatment but still with a BI of >3, may have a significant relapse rate. It has been
sugg'!sted that regular clinical and bacteriological review after release from treatment is
indicated for these patients. IS Where communications are difficult, it may nut be practical to
refer all smear examinations to a I eferral laboratory. 19 Hence, there is still be a need for skin
smear examinatioll to be made available,
at least at district level.
,
The success of the WHO Elimination Strategy suggests that there will gradually be fewer
positive smears. Leprosy will not, however, be eradicated by the year 2000. It is therefore
essential that there be accessible centres with trained staff who will be able to carry out
this procedure accurately. As the world tuberculosis situation deteriorates, it becomes
increasingly necessary to have laboratories in peripheral health units capable of bacteriolo
gical examination of sputum. Tuberculosis diagnosing centres (possibly one such laboratory
per 100,000 of the population"o), with an established system of quality control, can also be
responsible for leprosy microscopy. This is easier if the staining technique for both organisms
is identicaL From this study, it does not appear likely that the choice of decolourizer or its
strength is a major factor in the general unreliability of skin smear services in some part!' of
the world,
It is possible that this is a much more robust technology than is generally thought. This
study suggests that the same technique for staining leprosy and tuberculosis bacilli can be
used, furthermore, that acid alcohol is an improvement on sulphuric acid.

Acknowledgements
We thank Dr A. Colin McDougall of Oxford, UK, for his advice and encouragement. We also

Dcc%llri:er {llId ifS .I'trellgrh ill rile Z-N

StU;1I

l:n

thank the leprosy patients and staff of the B.P. Kulit Skin Clinic and R.S, Dayu Leprosy
Hospital, for their assistance. The leprosy control programme of the Indonesian Ministry of
Health in South Sulawesi is supported by the Netherlands Leprosy Relief Association (NSL).

References
WHO. II gllide ttl elilllin{/ting lepros.' {IS a pII/;lic heallh problelil. WHOILEP/lJS.I. 1st Edn. 1995.
Leiker DL. ~kDollgall AC. Tee/llliutl 8l1id" Jilt SI/IC{l1' /'X(/I/l;lIlIIiOIl jor lepros\'. 2nd revised edition. Talmi1cp.
Wurzburl!, I '1X7.
1 DavidsOl; AR. Deeolorizing of Mrco}wcleriulil lepme. lilt J [cpr, 1943; 2: 49-51.
Davidson AR. The tec1mique of staining leprosy hacilli in smears. Lepl' ReI', 19fiO; 31: 305-307.

5 Padm:J Mr-;. A standard technique of acid fast staining for M. leprae in sme~rs. I.qn lllliil!. 1'163; 35: 62-64.

(, Ridky l\'lJ & Ridley OS. Stain techniqucs and the morphology of M. lepme. L"pr Rev, 1'371; 42: RR-95.

Ridley 1\0\1, Thc cellular eXlldatc-MrcobacleriulIl lepme relationship and the critic,ll reading of skin Slilears. l.epr
11"1', 1%9; 60: ~~'J-24().
x International Union Against Tuberculosis and Lung Diseases (!UATLDI. Technical Guide For Sputum
Examination For Tuberculosis By Direct Micro,,opy. 1978. In: Tu/J('fculosis gllid" jill' 1m,' illcol1le cOlllllri".I,
4th eun. IUATLD. Paris; 1996.
'J WHO. l_olior(l/on 1('('lilliques }ill' leprosy. WHO/CDS/LEP/86.4, I 'Hn.
l\I Nilsen T, Spare II G. Skill .lIlIears}C)r leprosy. ALERT, Addis Ababa, DAHW/Talmilep. 1990.
II SLR&TC. Skill .I/Ilear <'.\wllil/fllioll. Technical Manual Sr. No. 3.1, Schicifelin Leprosy Rehabilitation and
Training Cenlre. Karigiri, India, 19R7.
J~ Dharllll'lldra (cd). '-<'prosy. Vol. I, Kothari r\'1edical Publ. Hou~(:. Bombay. I f.J7K
u VcttOI11 L, PrilZe S. Reliahility of skill smear rcsuits: ~xpcriences with quality control of skin smears in different
I.' routine services in ieplOs)' control programmes. Lepr ReI', 1'189; 60: 1g7 -1'16.
Bllku I'cuolll<In Pemhcrant<Isan Pcnyakil Kusta (Mallllal/i,,' workers ill 111e lepros.\' l',ogIYtJIIJlle). Edll. IX,
Department of Health, Jakarta, 1995.
15 Petunjuk Teknis Pcnatalak!.)anaan Lahoratoriunl Di Pu:-.kesmas (Technical guidelines/or ht'a/lh C(,111re /ubOrCl(Ory
procedurl's). Dep<li'lmcnt of Health, Jak<lrta, 191)0.
16 Kirkwood BR. Esset1tials (~lmedi('(ll statistics, Blackwell Scientific Publications, Oxford, 1992.
17 POflnighalls ltv!. Diagnosis and management of single lesion leprosy. Lepr ReI', 1996; 67: 89-94.
18 Desikan KV. The risk of relapse after multidrug therapy in leprosy. upr Rev, 1997; 6X: 114-116.
19 Georgie\' GD, ~'lcDollgall AC. Skin smears and the bacterial index (Bi) in Illultiple drug therapy leI""')' contrl)1
programmes: an unsatisfactory and potentially hazardous state of affairs. 1111 J I.ep!'. 1988; 56: I() 1-1 03.
I

~o International Union Against Tuberculosis and Lung Diseases (IUATLD). Tuher('ulosis guide for
cowllrie.l, 4th cdn. IUATLD, Paris, 1996.

/OH'

incoIJJe