^^Y M^ v^ : ^

INTERNAIIc )NAI. .I( )t 'RNA!. ol I,E-:I'I:osY Volum

(I S `

(,)N u n iVcç

1^^ r1^^1 f^,^.^^^ o^ n cA148-916X)

INTERNATIONALJOURNAL OF LEPROSY

and Other Mycobacterial Diseases

VoriJ 1E 69, NUMBER 3 SEPTEMBER 2001

Distinct Histopathological Patterns in Single LesionLeprosy Patients Treated with Single Dose Therapy

(ROM) in the Brazilian Multicentric Study'

Mauricio B. Costa, Plorencio F. Cavalcanti Neto, Celina M. T. Martelli,Mariane M. A. Stefani, Juan P. Maceira, Maria Katia Gomes,

Antonio Pedro M. Schettini, Paula F. B. Rebello, Patricia E. Pignataro,Emilia S. Ueda, Kazue Narahashi, and David M. Scollard 2

Histomorphological studies from skinbiopsies have contributed substantially tothe diagnosis and the classification of well-diagnosis

leprosy which comprises aspectrum of forms, ranging from pau-

I Received for publication on 17 April 2001. Ac-cepted for publication in revised form on 23 July 2001.

M. 13. Costa, M.D., M.Sc.: C. M. T. Martelli, M.D.,Ph.I).: M. M. A. Stefani, Ph.D., Federal University ofGoias, Rua 1141, Setor Marista, Goiana/GO, I3razil74180-080. F. F. C. Neto. M.D., Ph.D., University ofBrasilia, Brasilia, Federal District, Brazil. J. P. Madeira,M.D., Ph.D.: M. K. Goles, M.D. M.Sc.. Federal Uni

-versity of Rio de Janeiro/Hospital Clementina FragaFilho, I:io ele Janeiro, Rio de Janeiro State, Brazil. A. P.M. Schettini, M.D., M.Sc.: P. F. B. Rehello, M.D.,M.Sc., i\1Credo da Matta Foundation—MK) ReferenceLeprosy Laboratory. Rio de Janeiro, Rio de JaneiroState. 13raiiI. P. E. Pignataro, M.D., Oswald() CruzFound:ttion, Reference Leprosy Laboratory, Rio deJaneiro, Rio ele Janeiro State, Brazil. K. Narahashi,M.D., Secretariat of Health, Porto Velho, RondoniaState, I3raiil. D. NI. Scollard, M.I)., Ph.D., LaboratoryResearch Branch, GWL Hansen's Disease Center, Ba-ton Rouge, Louisiana, U.S.A.

Reprint request to Dr. Costa at the above address orFAX 55-62-212-7667: e-mail:mbareelos@ . mail.cultura.com.br

cibacillary (PB) tuberculoid to multibacil-lary (NIB) lepromatous types (. ' I ). Atthe tissue level, the morphological assess-ment of granuloma formation, the indica-tion of cutaneous nerve inflammation andthe presence of intracellular acid-fast bacilli(AFB) allow the differentiation between thediversity of host immune responses to My-cobacterium le/)me. The extent of local cel-lular immune response to M. leprae hasbeen related to protective and llllllltlllo-pat h o l og i c responses (1 I . 0)).

In the context of leprosy endemic coun-tries, histomorphology of skin biopsiesplainly has been applied for research ( 3 . 2, ).For public health purposes, leprosy case as-certainment is based on clinical descriptionand the presence of AFB in slit-skin smearsyielding the microbiological classificationof MB and PB. The clinical classificationcategorizes PB patients into those with asingle lesion or with 2 to 5 skin lesions foroperational and treatment purposes (worldHealth Org•ni/aIIon. Recommended surveill a nccst:t lords. WHO/CDS/CSR/1SR/ 99.2: 1999). Asingle skin lesion has been considered the

177

178 In ternational Journal of Leprosy 2001

earliest clinical manifestation of leprosy( 6. íO.20 ) . This clinical entity may be unsta-ble, spontaneously healing or progressing toany form in the disease spectrum (2 5. 18, 21)

Recently, the single-dose therapy of ri-fampin, ofloxacin and minocycline (ROM)trial for single skin lesion PB (SSL-PB) pa-tients without nerve involvement has raisedissues related to case definition and theirimmunopathology (`'. , ․). Few histopatho-logical studies on early leprosy have beenpublished ('. 7.'3.27 ) and, to our knowledge,there is no previous report of the morpho-logical features of single lesion patients inthe context of the Brazilian endemicity.Since 1997, a cohort of single lesion pa-tients have been monitored after ROM ther-apy. The study design and the baseline epi-demiological and immunological resultswere described in a previous report ( 14 ).

In this paper, we aim to describe the his-tomorphologic features of skin biopsies ofsingle lesion leprosy patients recruited in asingle lesion multicentric study. The fea-tures of the cellular inflammatory infiltratesand the presence of nerve involvement andAFB were used to categpri ze SSL-PB bi op-sies into different histopathological groups.Identification of homogenous groups of pa-tients according to these histologic featuresmay add some insight into the diagnosisand the prognosis in early leprosy.

MATERIALS AND METHODSThe study participants were recruited at

six outpatient clinics located in four Brazil-ian states in the Northeast (Amazonas andRondonia states), Southeast (Rio de Janeirostate) and Center-West (Goiás state) betweenOctober 1997 and December 1998. Detec-tion rates ranged from 7.0 to 8.4 per 10,000inhabitants in the highly endemic regions(Center-West and North) to 1.5 per 10,000 inthe Southeast region (Brasil, Ministério da Saúde,1998) at the beginning of the study.

Each site enrolled between 45 and 80SSL-PB patients over a 1-year period. Clin-ical examination, the Mitsuda test (cutoffpoint for positivity >_5 mm) and IgM anti-PGL-1 serology (positive samples OD >_0.2)were performed at baseline. Details of thestudy design and the baseline profile ofSSL-PB patients treated with ROM therapywere described in a previous report ( 14 ).Briefly, baseline results showed predom i -

nance of adults with a median lesion size of5.1 cm (95% CI 4.6-5.6), 75.0% were Mit-suda positive (>_5 mm) and 17.3% wereseropositive for anti-PGL-I IgM antibodies.All patients had at least moderate sensoryloss assessed by monofilanients, withoutdisability. The inclusion criteria were:newly diagnosed, single lesion, untreatedleprosy patients characterized by a hy-popigmented or erythematous lesion withor without infiltration, definite sensory loss,without thickened nerves. All SSL-PBcases had negative slit-skin smears. Patientswere excluded if a subsequent histopathol-ogy reading was consistent with MB lep-rosy fornis (BB, BL or LL types) accordingto Ridley and Jopling criteria. Childrenyounger than 7 years, pregnant women andknown HIV seropositives were not enrolledin the SSL-PB cohort.

Prior to ROM therapy, all recruited pa-tients had a standard 4-mm punch skinbiopsy taken from a lesion for histopatho-logical study. The biopsies were fixed in10% buffered formal i n, processed for paraf-fin sections and stained with hematoxylinand eosin (H&E) and Fite-Faraco for AFB.In order to identify homogenous histo-pathological categories among single lesionPB patients, the features of the cellular in-flammatory infiltrates, nerve involvementand the presence of bacilli were analyzed inthe skin biopsies. Using these characteris-tics, lesions were classified into five histo-morphological groups: a) well-circum-scribed epithelioid cell granuloma (Group1); b) less-circumscribed epithelioid cellgranuloma (Group 2); c) mononuclear in-flammatory infiltrate permeated with epi-thelioid cells (Group 3); d) perivascular/periadnexal mononuclear inflammatory in-filtrate (Group 4), and e) minimal or no mor-phological alteration detected (Group 5).

The finding of AFB in single lesion skinbiopsies was considered to be a definitivediagnosis of leprosy. Leprosy was consid-ered highly probable in the presence of cu-taneous nerve involvement and/or an epi-thelioid granuloma (Groups 1 and 2) amongclinically suspected leprosy patients inhighly endemic areas. The presence ofmononuclear inflammatory infiltrates, withor without epithelioid cells, was consideredto be probable leprosy (Groups 3 and 4). Inaddition, for comparison purposes we have

69, 3 Costa, et al.: Histopathology in Single Lesions

179

TABLE 1. Recruited single-skin lesion paucibacillary leprosy patients (SSL-PB) for themulticentric Brazilian study.

Locale/exams

Findings

Regional health carters in three endemic regionsClinical and microbiology

Coordinator CenterHistomorphologic evaluation of skin biopsies

299 patients screened278 potentially eligible for ROM' therapy

with available skin biopsy

259 PB S/possible leprosy casepatients treated with ROM

7 MB` leprosy excluded12 other skin diseases excluded

'Rifampin, olloxacin and minocycline.''PB = Paucibacillary.`MB = MuItibaci11ary.

applied these criteria to the Ridley andJopling classification.

To avoid potential biases introduced bymulticenter recruitment, all skin biopsieswere processed or reviewed by one pathol-ogist with expertise in dermatopathology,without the knowledge of the patient's clin-ical classification, at the Pathology Labora-tory in the Coordinator Center (FederalUniversity of Goiás, Central Brazil). In or-der to assess inter-observer variation,14.7% (N = 38) of the biopsy samples werere-scored blindly by another experiencedpathologist (National Hansen's DiseaseCenter, U.S.A.) considering histologic fea-tures only, without knowledge of the resultsof the Fite-Faraco staining.

Written informed consent was obtainedfrom all participants, and the project wasapproved by each regional and by the Na-tional Ethical Committee Board (CONEP-Brazilian Ministry of Health).

Descriptive statistics were applied toevaluate the morphological groups stratifiedby settings. Exploratory data analysis wasused to describe Mitsuda and anti-PGL-Ivalues. The chi-squared test was applied toevaluate differences among proportions.McNemar's chi-squared test was also ap-plied to compare the frequency of discor-dant pairs between pathologists; p valueslower than or equal to 0.05 were regardedas significant.

RESULTSTwo-hundred-seventy-eight (93.0%) out

of 299 patients had a skin biopsy available,and they were considered potentially eligi-

ble for ROM therapy according to thedermato-neurological exam and negativeslit-skin smear bacteriology test (Table 1).Among excluded cases, seven single lesionpatients were diagnosed as BL or LL lep-rosy types (MB) by the histopathologicalexams. Of these seven MB cases, most hada low bacterial index (BI 1+) and only onehad a high bacillary load (BI 4+) accordingto Ridley's logarithmic scale (II). All of theMB cases were adults, ages ranging from27 to 56 years, six were males and none hada BCG scar. Additionally, 12 cases werealso excluded from the SSL-PB cohort dueto other skin diseases such as: 2 cutaneousleishmaniasis, 2 pityriasis alba, 2 foreign-body reactions, 1 granuloma annulare, 1morphea, 1 pityriasis versicolor, 1 insectbite, I cutaneous amyloidosis, and Ieczematous dermatitis. In addition to thesecases, six other biopsies were consideredunsuitable for analysis. Therefore, 259 pa-tients had skin lesions with histomorpho-logical features compatible with PB lep-rosy.

The main characteristics of the histo-pathologic sections of skin biopsies are de-scribed below.

Group 1. Well-circumscribed epithelioidcell granulomas, with well-differentiatedepithelioid cells, encompassed by a densemantle of mononuclear inflammatory cells,mainly lymphocytes. These infiltrates werepresent as a single or multiple granulomaswith defined limits and were around neu-rovascular bundles and/or periappendageal.Langhans' giant cells occasionally werepresent. Dermal nerve bundles sometimes

1 8O International .1011171(11 of I,ep1atiy 2001

FI G . I. Group I: Well-circum s cri bed epithelioidcell granulomas, with well-differentiated epithelioidcells, l'1lcolllpassc'd by a dense mantle of i11OIlollllClearinflammatory cells, mainl y lymphocytes ( H &E x125).

were absent (completely destroyed or oblit-erated) or surrounded by dense lymphocytecuffs. Nerve involvement was observed in

1.7%. Rare AFB were found in only onecase (1.1%) (Fig. 1).

Group 2. The histopathologic sections ofskin biopsies presented less-circumscribedepithelioid cell granulomas, with well-dif-ferentiated epithelioid cells permeated bymacrophages with vacuolated cytoplasm.The mantle of mononuclear cells was notwell fori»ed. AFB were scanty but werefound in 21.4% of the cases. Nerve involve-ment was represented typically by lympho-cytes and macrophages obliterating the der-mal nerve and was observed in 62.5%c ofthe samples within this group (Fig. 2).

Group 3. The histopathologic sections ofskin biopsies showed moderate Ilono-nuclear inflammatory infiltrates permeatedwith less-differentiated epithelioid cells,mainly perivascular and periadnexal. AFBwere detected in 9.7% of the biopsies, and

FIG. 2. Group 2: F listop:tthologic sections of skinbiopsies presented less-circumscribed epithelioid cellgranulomas, with Well-di Ife rent iated epithelioid cellspermeated by macrophages with vacuolated cyto-plasm. '[he mantle of mononuclear cells was not wellformed (H&E x250).

nerve involvement was observed in 71.0%(Fig. 3).

Group 4. The histopathologic sections ofthe skin biopsies showed mild/moderatemononuclear inflammatory infiltrate, plainlyaround neurovascular bundles and periad-nexal . Nerve involvement was observed in20.5%e, and AFB were found in only onecase (Fig. 4).

G roup 5. The histopathologic sections ofthe skin biopsies showed mild mononuclearinflammatory infiltrate, represented mainlyby lymphocytes, localized around superfi-cial and deep dermal vessels recallin<g whatis sometimes observed in normal skin biop-sies. Neither nerve involvement nor AFBwas observed.

The five histopathological groups de-scribed within the studied cohort of SSI.-PB leprosy cases were categorized as fol-lows: 33.6% (N = 87) of the biopsies repre-

• ..•^ Ala

- ■ _ ^^^^` ,'. ^,.

^ •^^• . ^. ^ t y. •^ ^ ; .^ ^^ •^►

. ► ^••-.f ^i N. ^ ,ti ^ ^ • ^,,^ • ^^t^, ^.

.

^ ^^ ^ ^ ^^• ^r. s ^ . ^^^ . ♦ • "‘t

^ ^ ' . •• "4/... ^ ^ ^i ^ ^• ^lVA" ^ ^, ` . . f ^••. .. 1 ^,l ^ • ^ ^ ^^ ^ M• • -. • 4.

...^ •^ • • t ?^!^ ^ ^. - ^ ,, f ; -- .^ • .•^• r • ^ i^:K,^^'iti^"^

• ^^ '• ti4 JP •4♦. ^ ^ ^'1r^ '^.

4) ^ ^..,^•*•¡^^ • ^• . ^• ^^••^^^t►r^^► •^ ^ ^ • ^ ^ ^• ' ^ •,• • ♦•,►^ ^ •

^!^^.r.^,^,y.,^ -^. ^ •1.77.

. ^''--` .^ ^..,.^^: ; . • ^ ^ ,^^ R •:. ^^. ^^^.,r : ^•^^,^, ^. ^.

p ^ ' . ^V » ^ ^4/ ^ ^ , ^ • ••.,♦^ í^ ^, ^ , i • ! ^► ^

^^`'^ ,

•^

4.^ ^

'.

A ° .• ^► ^•- ^ ^ ^j h • ,^•if, ,^.^, ^^ ,^, ^ ^ r^,

A ^ aLv4Irb

^` i► 10 Ì^^ ^^e', ^ ^ I• ^ ^^r ^ ^^C.:i.......11b-t ^ ,: ^ 19 .^i t

1 '^ ^ ^^.. ^7I

11

. ,i t_____• ^ ^ .♦ _^^

*,^,,,,^^.^, .^,^.... ^

-i,r4^ ...'^

^ ^ 7. ! .^.:.L' ^^,

• .6 1^

Ib m''-^ ^,, ^•^ ^ ^ ^^ .. ,^ ^ r

C)9, 3

Casia, et f li.slnpert/tele,,,i y ^^t .5'r^t^lc' L^src»r^ 181

^ , ^^^ •^^ ^^ ^ ^^ • ^

Ape. #•^ • ^ ♦ ^ . ' • s , ^'^

1 4

•••,I^•e. 4.

^ L - — • s • ^y •,^•^4^ ' ;4Tt ^ If

• ^

^

•4 /1^*as ^..^_ á.*i^► ^ i►_

FIG. 3. Group 3: 11ist01)athOlOgic sections Of skinbiopsies showed I110der:itl', mononuclear inflammatoryinfiltrate i)ermeatecl with less-J .11ferenti;ited (Tithe-hold cells, mainly i)erivascular and periaclnexal (Il&Ex25O).

Seated well-circumscribed epithelioid cellgranuloma (Group 1): 21.6% (N = 56) less-circumscribed epithelioid cell granuloma(Group 2): 12.0% (N = 31) were describedas mononuclear inflammatory infiltrate per-meated with epithelioicf cells (Group 3),and 29.7% (N = 77) had perivascula1/peri-adnexal ilmononuclear inflammatory iilfil-trate (Group 4). Minimal/no MtOrpholC`ica1alteration in the skin was detected 1 n onlyei`_ht (3.1%) SSL-I 3 13 patients categorizedas Group 5, who were considered leprosyby clinical parameters. For simplified com-parison, Grout) 1 could correspond to TTforms, Groups 2 and 3 to the 13T fOrihs inthe Ridley aild Jopling classification andGroup 4 was considered as indeterminateleprosy.

When different histopathologlc.al sectionswere examined independently by twopathologists, 30 out of 38 specimens wereconsidered compatible with leprosy '1n(1 one

FIG. 4. Group 4: 1 list0¡)athO1011ie scction Of skinhiOl)ics hOwccl mild/moderate mononuclear inflam-matory infiltr:ite, ni;iinl^ aroundneurO^^:is^•u!: , r bundles

i)ancl e riacínc.^^cal ( I I&E rx2>O ).

was not considered leprosy, yielding 81.6%(95(X CI 65.7-92.3) of overall ow -cement."There was no statistically significant differ-ence in the comparison of the proportions inpaired analysis ( Fisher, 1) = O. 1 5 ). Therewas close a reement between pathologistsin the ascertainnlent of the SSL-PI3 leprosydiagnosis, indicating reproducible criteriaused and good duality control.

"Table 2 shows the distribution Ofspiit/nerve involvement concomitant withthe presence Of baci11i stratified into the five11istopathological groups described. TheOverall Percentage Of nerve involvemmlentwas 49.8% (N = 127) and in 17 patients(6.7%Jo) are AFB were found mainly coex-isting with nerve involvement. AFI3 mainlywere detected within lesions representingthe less-circumscribed epithelioid cell gran-uloma (Group 2). There was no statisticallysignificant difference in the frequency ofnerve iinvolveimlent amon g the groups.

The box plots in Figure 5 show the distri-

ft •

182 International Journal of Leprosy 2001

TABLE 2. Morphological characteristicsof SSL-PB leprosy patients.

Histopathological groups

I 2 3 4

Nerve involvementAFB'' positive 1

11 3 1AFB negative 43

24 19 25No nerve involvement

AFB positive 0

1 0

0AFB negative 41

20 9

49Total 85c 56 31

75`

Group I = Well-circumscribed epithelioid granu-loma; Group 2 = less-circumscribed epithelioid granu-loma; Group 3 = mononuclear inflammatory infiltratepermeated with epithelioid cells; Group 4 = perivascu-lar/periadnexal mononuclear inflammatory infiltrate;Group 5 = not shown; minimal or no morphological al-teration (N = 8).

h AFB = Acid-fast bacilli.Two patients in Group 1 and 2 in Group 4 had un-

certain nerve involvement.

bution of the values of the Mitsuda reactionin each group. All patients included inGroup 1 had Mitsuda test results >3 mm,89.3% of them strongly positive (cutoff >5mm). In contrast, 23.6% of the patients as-sembled as Group 4 had negative Mitsudareactions. There was a statistically signifi-cant association between a positive Mitsudatest and patients classified as Group 1 (x2 =16.0, DF = 3, p <0.05). Variables related toage, sex, household contact, BCG scar, timeof perceived lesion and lesion size were notstatistically significantly different amongthe groups. Anti-PGL-I IgM serology waspredominantly negative, ranging from 70%to 84%, without any association with thefive histopathological groups. In all sites,approximately one third of the SSL-PBbiopsies had skin lesions presenting well-circumscribed epithelioid cell granulomas(Group 1) (Fig. 6).

DISCUSSIONOur data clearly show that SSL-PB lep-

rosy patients recruited in a multicentricstudy presented histomorphology readingscomprising the whole PB leprosy spectrumbut also a few MB cases, despite beingdiagnosed as a single clinical entity fortreatment purposes. These results indicate ahistomorphological heterogeneity amongSSL-PB patients, with predominance of

well-circumscribed and less-circumscribedepithelioid cell granulomas (Groups 1 and2) in the three Brazilian endemic regionsstudied.

A single lesion has been considered moretroublesome to be ascertained as leprosysolely on clinical grounds when comparedto PB patients with 2-5 skin lesions. In thissense, histopathology is known to increasesensitivity and specificity of case diagnosisand is particularly useful when skin smearsare negative (21.22 ). In our study, the speci-ficity of case definition was enhanced bythe exclusion of seven patients who pre-sented histopathology compatible with MBleprosy and 12 patients who were diag-nosed with other skin diseases. This findingis in agreement with other studies that re-ported MB leprosy and other diseasesamong monolesion patients ( 7 .' 1 ). Neverthe-less, although some caution is warranted,our data support the public health approachof the clinical manifestation and bacil-loscopy of single lesion patients for opera-tional classification, already discussed inthe clinical and immune baseline profile ofSSL-PB patients in a previous report ( 14).

We have attempted to develop histo-pathological criteria by which subgroupswithin the PB portion of the leprosy spec-trum may be recognized. The main premisewas that the five-part immune classification( 25 ) was conceived to categorize the wholespectrum of leprosy disease and some diffi-culties were reported when applying thissystem to PB and the early lesion ( 4). Thisbaseline histopathology evaluation, evenwhen performed among newly diagnosedSSL-PB patients, cannot determine withany certainty the duration of illness or howearly the cases are. In this sense, a time di-mension should be considered when ana-lyzing the immunopathologic spectrum ofleprosy, and changes in histopathology mayoccur during clinical evolution ( 26).

In the majority of the readings, an inflam-matory infiltrate, nerve involvement and/orthe finding of AFB were found. AFB weredetected in only a small percentage of thebiopsies (6.7%). These findings are in con-cordance with previous experience thatAFB are rarely detected in skin biopsies ofPB patients, which makes it difficult to con-firm the diagnosis of leprosy by the detec-tion of bacilli alone (6, "• 12, 23) . Therefore,

*

o *20 100%

80%

60%

N = 84

54

30 72 40%

Group I

Group 2

Group 3 Group 4

69, 3 Costa, et al.: Histopathology in Single Lesions 183

FIG. 5. Mitsuda test distribution among differentgroups of paucibacillary leprosy patients. ----cut-off(mm); Group I = Well-circumscribed epithelioid gran-uloma; Group 2 = less-circumscribed epithelioid gran-uloma; Group 3 = mononuclear inflammatory infiltratepermeated with epithelioid cells; Group 4 = perivascu-lar/periadnexal mononuclear inflammatory infiltrate.

perineural inflammation and/or the pres-ence of AFB detected by histopathology ingranulomatous and non-granulomatous skinlesions, obtained from clinically suspectedpatients, have been considered highly prob-able for or definitive of leprosy, respec-tively.

By these criteria, half of the skin biopsiesin our study were considered TT or BTleprosy. Interestingly, AFB were detectedmainly in the less-circumscribed epithelioidgranulomas or mononuclear inflammatoryinfiltrates (Groups 2 and 3) considered BTin the Ridley and Jopling ( 25) classification.A possible explanation is that effective cell-mediated immunity (CM I) leads to the for-mation of well-circumscribed granuloma(Group I) which restrain M. leprae multi-plication at the tissue level ( 17 ' 9 . 30). In thisstudy, a diagnosis of probable leprosy wasmade when histomorphological features ofgranulomatous and non-granulomatous in-flammation were associated with a clinicaldiagnosis of leprosy by experts in an en-demic region, considering that there is noindependent "gold standard" for early dis-ease ( 12 ).

Most of the recent leprosy literaturerefers to early/single-lesion leprosy as anindeterminate form of the disease whichmay progress to either pole of the spectrum(5, n) . u.21.'-`,) Our study reveals that almost70% of patients with a single lesion had

North

Southeast Center-West

Fm. 6. Morphological characteristics of singleskin lesion paucibacillary leprosy patients.Group I = Well-circumscribed epithelioid granuloma;

= Group 2 = less-circumscribed epithelioidgranuloma; Illllllllhll = Group 3 = mononuclear inflam-matory infiltrate permeated with epithelioid cells;

= Group 4 = perivascular/periadnexal mononu-clear inflammatory infiltrate; ¡ = Group 5 = mini-mal or no morphological alteration.

histopathological findings compatible withthe TT or BT forms, independent of theBrazilian region. This suggests that a sub-stantial portion of monolesion patients maybe classified as TT or BT rather than inde-terminate, although it is not possible at thispoint to identify those who might have thepotential to downgrade to other borderlinetypes. This finding is in agreement withstudies conducted in India ( 5 . 13, 5 . 27) beforethe implementation of ROM therapy. No-tably also, our results showed some correla-tion of the histologic findings with the Mit-suda response, and no correlation with theserologic status of the patients. These ob-servations are consistent with earlier studiesof PB disease, and have significance for ef-forts in developing skin tests or better sero-logic tests to aid in the early diagnosis ofleprosy.

Finally. the different histomorphologicalprofiles among SSL-PB patients demon-strated the heterogeneity of the local cellu-lar immune response. The identification of

20%

0%

I I

I 8 International Journal of Leprosy 200 I

more "homogeneous" subgroups of patientsmay be very valuable in research to evalu-ate the different outcomes to one-doseRUM therapy during clinical follow up.

SUMMARYThis paper aims to describe the h i stomor-

phologic features of skin biopsies of singlelesion leprosy patients recruited at outpa-tient clinics in four Brazilian states in theNortheast (Amazonas and Rondonia),Southeast (Rio de Janeiro) and Center-West(Goiás) between October 1997 and Decem-ber 1998. Patients clinically diagnosed assingle skin lesion paucibacillary (SSL-PB)leprosy had a standard 4-mm punch biopsytaken from the lesion before rifampin,ofloxacin, minocycline (ROM) therapy. Thefeatures of the cellular inflammatory infil-trates, the presence of nerve involvementand acid-fast bacilli (AFB) were used tocategorize SSL-PB biopsies into differenthistopathological groups. Two-hundred-seventy-eight (93.0%) out of 299 patientshad a skin biopsy available. Seven singlelesion patients were diagnosed as BL or LLleprosy types (MB) by the histopathologicalexams and 12 cases were excluded due toother skin diseases. Therefore, 259 patientshad skin lesions with histomorphologicalfeatures compatible with PB leprosy cate-gorized as follows: 33.6% (N = 87) of thebiopsies represented well-circumscribedepithelioid cell granuloma (Group 1);21.6% (N = 56) less-circumscribed epithe-lioid cell granuloma (Group 2); 12.0% (N =31) were described as mononuclear inflam-matory infiltrate permeated with epithelioidcells (Group 3), and 29.7% (N = 77) hadperivascular/periadnexal mononuclear in-flammatory infiltrate (Group 4). Minimal/no morphological alteration in the skin wasdetected in only 8 (3.1%) SSL-PB patientscategorized as Group 5, who were consid-ered to have leprosy by clinical parameters.SSL-PB leprosy patients recruited in a mul-ticentric study presented histomorphologyreadings comprising the whole PB leprosyspectrum but also a few MB cases. Theseresults indicate heterogeneity among SSL-PB patients, with a predominance of well-circumscribed and less-circumscribed ep-ithelioid cell granulomas (Groups I and 2)in the sites studied and the heterogeneity oflocal cellular immune response.

RESUMENEste estudio describe los hallazgos histomorlulógi-

cos de las biopsias de piei de lesiones únicas en los pa-cientes con lepra. recltitados en clínicas de 4 estadoshrasilenos del Noreste (Amazonas y Rondonia),Sureste (Rio de Janeiro), y Centro -Este (Goias) deBrasil, entre octobre de 1997 y diciembre de 1998. Lasbiopsias de estos pacientes se tmaron antes y despuésdel tratamiento con rifampina, ofloxacina y minoci-clina. Las biopsias se clasifìcaron en diferentes gruposhistológicos sobre la base de los infiltrados celulares,la presencia de afección nerviosa y el número de baci-los ido-resistentes.

Se contó con 278 biopsias de 299 pacientes (93%).Siete lesiones únicas se diagnosticaron como lepra Bl.o LL nlultibacilar (MB) y 12 casos se descartaron de-hido a otras afecciones de la piei. Las restantes 259biopsias correspondieron a lesiones compatibles conlepra paucibacilar: 33.6% (N = 87) de las biopsiasmostraron granulomas de células epitelioides bien cir-cunscritos (Grupo 1); 21.6% (N = 56) tuvieron granu-lomas epitelioides menos circunscritos (Grupo 2):12.0% (N = 31) mostraron infiltrados inflamatoriosmononucleares con algunas células epitelioides intil-trantes (Grupo 3), y 29.7% (N = 77) tuvieron infiltradomononuclear Inflamatorio perivasculai/perianexal(Grupo 4). Sólo 8 pacientes (3.1%) con lesiones únicasde la lepra mostraron ausente o mínima alteración dela piei. Los pacientes paucibacilares (PB) con lesionesúnicas, incluidos en el estudio nlulticéntrico, presen-taron características que abarcaron todo el espectro dela lepra paucibacilar, aunque también hubieron al-gunos casos multibacilares. Los resultados indican het-erogeneidad histomorfológica entre los pacientes PBcon lesiones únicas, con predominio de granulomas decélulas epitelioides más o menos circunscritos (GruposI y 2), y heterogeneidad de la respuesta celular local.

RÉSUMÉ

Cet article vise à décrire les caractères histo-mor-phologiques de biopsies cutanées provenant de pa-tients non-hospitalisés souffrant de lésion unique delèpre, recrutés it partir de cliniques situées dans quatreétats bréziliens du nord-est (Amazonas et Rondonia),sud-est (Rio de Janeiro) et centre-ouest (Goias) entreoctobre 1997 et décembre 1998. Préalablement autraitement associant la rifampine, l'orofloxacine et laminocycline (ROM), les patients diagnostiqués clin-iquement comme lésion cutanée unique paucibacillaire(SSL-PB) eurant leur lésion prélevée à l'aide d'un tro-card punch' de 4 mm. Les caractères de l'infiltrat cel-lulaire inflammatoire, la présence de nerfs lésés et debactéries acido-alcoolo-résistantes (AAR) furent util-isés pour regrouper les biopsies SSL -PB dans plusieursensembles histopathologiques.

Deux cent soixante dix-huit (93,0%) parmi 299 pa-tients présentaient une hiopsie exploitable. Sept 16-sions uniques de patients furent diagnostiquées commede type BL et LL (multihacillaire; MB) par l'examenhistopathologique et 12 cas furent exclus it cause de

69, 3

Costa, et al.: Histopat hology in Single Lesions 185

maladies intercurrentes. Ainsi 259 patients eurent deslésions cutanées avec des caracteres Instomor-phométriques compatibles avec une lèpre paucibacil-laire (PB), qui furent regroupées comme suit: 33,6%(N = 87) des biopsies présentaient des granulomes ép-ithélioïdes bier circonscrits (groupe I); 21,6% (N =56) des granulomes épithélioïdes moms bien cir-conscrits (groupe 2); 12,0 (N = 31) furent décritscomme des infiltrat de cellules mononuclééesparsemés de macrophages épithélioïdes (groupe 3), et29,7% (N = 77) avaient un inliltrat périvasculaire et/oupériannexiel de cellules mononucléúes (groupe 4). Desaltérations morphologiques minimes ou absentes de lapeau ne furem observées que chez 8 (3,1%) des pa-tients SSL-P13, qui furent classés dans le groupe 5, carconsidérés comme hanséniens sur des criteres etparamètres cliniques.

Les patients hanséniens SSL-PB recrutés pour cetteétude multicentrique ont présenté des aspectshistopathologiques comprenant non seulement('ensemble du spectre de la leprc PB, mais aussi derares cas MB. Ces résultats indiquent un aspect histo-morphologique hétérogène des patients SSL-PB, avecune prédominance de granulomes épithélioïdes Bien etmoms bien circonscrits (groupes 1 et 2), accompagnéd'une hétérogénéité e la réponse locale immune à mé-diation cellulaire.

Acknowledgment. We thank the members of theLeprosy Study Group (Dr. Sylvana Castro Sacchetim,Dr. Silmara Navarro Pennini, Dr. Ana Lúcia OsórioMaroclo Souza, Dr. lima M. Silva, Dr. José Augusto C.Nery, Dr. Flávia Pretti) for their assistance. We thankDr. Thomas P. Gillis and Dr. Euzenir Sarno for helpfulsuggestions on the manuscript. This study was par-tially sponsored by PAHO, the Brazilian Ministry ofHealth and FUNAPE-Federal University of Goiás in1998. This investigation received financial supportfrom the UNDP/World Bank/WHO Special Pro-gramme for Research and Training in Tropical Dis-eases (TDR) ID number 981007.

REFERENCESl . BINFORD, C. H. The histologic recognition of the

early lesions of leprosy. Int. J. Lepr. 39 (1971)225-230.

2. CLEMENTS, B. R. and SCOLLARD, D. M. Leprosy.In: Atlas of Infections Disease. Vol. 8. Mandell, G.L. and Fekety, R., eds. Philadelphia: Current Med-icine, 1996, pp. 9.1-9.28.

3. FINE, P. E. M., JOB, C. K., LUCAS, S. B., MEYiats,W. M., PONNIGHAUS, J.M. and STERNE, J. A. C.Extent, origin, and implications of observer varia-tion in the histopathological diagnosis of sus-pected leprosy. Int. J. Lepr. 61 (1993) 270-282.

4. FLEURY, R. N. Dificuldades no emprego cla classi-ficação de Ridley e .1oplinb uma análise mor-fológica. Hansen. Int. 14 (1989) 101-106.

5. JACOBsoN, R. R. and KRAI tENBUHL, J. L. Leprosy.Lancet 353 (1999) 655-659.

6. JOB, C. K., BASKARAN, B., JAYAKUMAR, J. andASCHHOFF, M. Histopathologic evidence to showthat indeterminate leprosy may be a primary le-sion of the disease. Int. J. Lepr. 65 (1997)443-449.

7. JOB, C. K., KAHKONEN, M. E., JACOBSON, R. R.and HASTINGS, R. C. Single lesion subpolar lepro-matous leprosy and its possible mode of origin.Int. J. Lepr. 57 (1989) 12-19.

8. KATOCtt, K., NA7RAJAN, M., YADAV, V. S. andBHATIA, A. S. Response of leprosy patients withsingle lesions to MDT. Acta Leprol. 9 (1995)133-137.

9. LocKwooD, D. N. J. Rifampicin/minocycline andotloxacin (ROM) for single lesions-what is theevidence? Lepr. Rev. 68 (1997) 299-300.

10. LOMBARDI, C., COHEN, S., LEIKER, D. L., SouzA, J.M. P., CUNHA, P. R., MARTELLI, C. M. T.. AN-

DRADE, A. L. S. S. and ZICKER, F. Agreement be-tween histopathological results in clinically diag-nosed cases of indeterminate leprosy in São Paulo,Brazil. Acta Leprol. 9 (1994) 83-88.

11. LUCAS, S. Bacterial disease-leprosy. In: LeversHistopathologv of the Skin. 8th edn. Elder, D.,Elenitsas, R., Kaworsky, C. and Johnson, B., Jr.,eds. Philadelphia: Lippincott-Raven Publishers,1997, pp. 477-502. Chapter 21.

12. LUCAS, S. B. and RIDLEY, D. S. The use of histo-pathology in leprosy diagnosis and research. Lepr.Rev. 60 (1989) 257-262.

13. MAJUMDER, V., SAHA, B., HAJRA, S. K., BISwAS,

S. K. and SAHA, K. Efficacy of single-dose ROMtherapy plus low-dose Convit vaccine as an adju-vant for treatment of paucibacillary leprosy pa-tients with a single skin lesion. Int. J. Lepr. 68(2000) 283-290.

14. MARTELLI, C. M. T., STEFANI, M. M. A.. GOMES,

M. K., REBELLO, P. F. B., PENINNI, S., NARAHASHI,

K., MAROCLO, A. L. O., CosrA, M. B., SILVA, S.

A., SACCHETIM, S. C., NERY, J. A. C., SALLES, A.M., GILLIS, T. P., KRAHENBUHL, J. and ANDRADE,

A. L. S. S. Single lesion paucibacillary leprosy:baseline profile or the Brazilian multicenter cohortstudy. Int. J. Lepr. 68 (2000) 247-257.

15. MATHAI, R., GEORGE, S. and JACOB, M. Fixed du-ration MDT in paucibacillary leprosy. Int. J. Lepr.59 (1991) 237-241.

16. MCDOUGALL, A. C., PONNIGHAUS, J. M. and FINE,

P. E. M. Histopathological examination of skinbiopsies from an epidemiological study of leprosyin northern Malawi. Int. J. Lepr. 55 (1987) 88-98.

17. MODLIN, R. L., MELANCON-KAPLAN, J., YOUNG, S.

M. M., PIRMEZ, C., KINO, H. , CoNVrr, J., REA, T.H. and BLOOM, B. R, Learning from lesions: pat-terns of tissue inflammation in leprosy. Proc. Natl.Acad. Sci. U.S.A. 85 (1988) 1213-1217.

18. OPRomoi.LA, D. V. A. Hanseníase com lesão única.Hansen. lot. 21 (1996) 1-2.

19. Orr1:NHoFF, T. H. M., KUMARARATNE, D. and CASA-

NOVA, J.-L. Novel human immunodeficiencies re-

1 K6 International Journal of Leprosy 2001

veal the essential role of type- I cytokines in im-munity to intracellular bacteria. Innnnno'. Today19 (1998) 491-494.

20. PANNIKAR, V. K. Defining a case of leprosy. Lepr.Rev. 63 (1992) 61s-65s.

2 I . PONNIGHAUS, J. M. Diagnosis and management ofsingle lesion in leprosy. Lepr. Rev. 67 (1996)89-94.

22. PONNIGHAUS, J. M. and FINE, P. E. M. Sensitivityand specificity of the diagnosis and the search forrisk factors for leprosy. Trans. R. Soc. Trop. Med.Hyg. 82 (1988) 803-809.

23. RIDLEY, D. S. The pathogenesis of the early skinlesion in leprosy. J. Pathol. 111 (1973) 191-206.

24. RIDLEY, D. S. and HILSON, G. R. F. A logarithmicindex of bacilli in biopsies. Int. J. Lepr. 35 (1967)184-186.

25. RIDLEY, D. S. and .IOPLING, W. H. Classificationof leprosy according to immunity—a five-groupsystem. Int..1. Lepr. 34 (1966) 255-273.

26. SCOLLARD, D. M. Time and change: new dimen-

sions in the imniunopathologic spectrum of lep-rosy. Ann. Soc. Belg. Méd. Trop. 73 ( 1993) 5-11.

27. SIIINDE, A., KIIOPKAR, U., PAI, V. V. and GANAPATI.R. Single-dose treatment for single lesion lep-rosy; histopathologieal observations. hit. J. Lepr.68 (2000) 328-330.

28. SINGLE- LESION MUI:rICEN'I RIC TRIAL GROUP. Effi-cacy of single dose multidrug therapy for the treat-ment of single lesion paucibacillary leprosy. In-dian .1. Lepr. 69 (1997) 121-129.

29. TAKAHASHI, M. D., ANDRADE, H. F., WAKAMATSU,A., SIQUEIRA, S. and BRITO, T. Indeterminate lep-rosy: histopathology and histocheniical predictiveparameters involved in its possible change to pan-cihacillary or multibacillary leprosy. hit. J. Lepr.59 (1991) 12-19.

30. YAMAMURA, M., UYEMURA, K., DEANS, R. .I.,WEINBERG, K., REA, T. H., BLOOM, B. R. andMODLIN, R. L. Defining protective responses topathogens: cytokines profiles in leprosy lesions.Science 254 (1991) 277-279.