Tropical medicine rounds

Diagnosing leprosy: revisiting the role of the slit-skin smear

with critical analysis of the applicability of polymerase chain

reaction in diagnosis

Surajita Banerjee1, MSc, Nibir Biswas2, MBBS, Nilay Kanti Das2, MD, Amrita Sil3, MBBS,Pramit Ghosh4, MD, Abu Hena Hasanoor Raja1, MSc, Sarbani Dasgupta1, MBBS,Pijush Kanti Datta2, MD, and Basudev Bhattacharya1, MD

1Department of Biochemistry, Institute

of Postgraduate Medical Education and

Research, 2Department of Dermatology,

Medical College, Kolkata, 3Department

of Pharmacology, Burdwan Medical

College, Burdwan, and 4Department of

Community Medicine, Medical College,

Kolkata, West Bengal, India

Correspondence

Professor Basudev Bhattacharya, MD

Department of Biochemistry

Institute of Postgraduate Medical

Education and Research

244B AJC Bose Road

Kolkata 20

West Bengal

India

E-mail: bbasudev@rediffmail.com

Funding: Rameshwar Das Birla Smarak

Kosh Medical Research Center,

Bombay Hospital Trust, Mumbai, India.

Conflicts of interest: None.

Abstract

Background Diagnosing leprosy is challenging, especially in early-stage cases, and the

need for a sensitive diagnostic tool is urgent. Polymerase chain reaction (PCR) holds

promise as a simple and sensitive diagnostic tool, but its usefulness in the Indian context

requires further evaluation. Slit-skin smear (SSS) remains the conventional method of

leprosy detection. Hence, this study was undertaken to evaluate and compare the

diagnostic efficacy of PCR versus that of SSS.

Methods Punch biopsy of skin and SSS were obtained from the active margins of lesions.

Cases were clinically grouped according to whether they were multibacillary (MB) or pauci-

bacillary (PB) and classified into tuberculoid (TT), borderline tuberculoid (BT), borderline

lepromatous (BL), lepromatous (LL), histoid, and indeterminate groups after clinicopatho-

logical correlation. DNA was extracted from biopsy specimens, and multiplex PCR was car-

ried out incorporating primers intended for the amplification of a specific 372-bp fragment

of a repetitive sequence of Mycobacterium leprae DNA.

Results Among 164 patients, PCR was positive in 82.3%. The sensitivity of PCR was

significantly greater (P < 0.0001) than that of SSS in both the MB (85.9% vs. 59.8%) and

PB (75.4% vs. 1.8%) subgroups; the difference in sensitivity in the PB subgroup is remark-

able. Positivity by PCR and SSS was found in 100% of LL and histoid leprosy, but PCR

had significantly greater (P < 0.0001) positivity in BT leprosy and was of definite increased

value in indeterminate and TT leprosy.

Conclusions Polymerase chain reaction had higher sensitivity compared with SSS,

especially in diagnostically challenging and PB cases. Thus, the use of this costly but

sensitive tool should be restricted to this subgroup, because SSS is sufficiently sensitive in

the diagnosis of LL and histoid leprosy.

Introduction

Leprosy is an age-old, chronic, granulomatous infectiousdisease caused by Mycobacterium leprae, sufferers ofwhich are known to have been ostracized by society sincethe very early days of civilization. Since 1982, leprosy hasceased to be an untreatable and incurable condition as aresult of the introduction of multi-drug therapy (MDT) bythe World Health Organization (WHO). However, theambitious elimination campaign launched in the 1990s bythe WHO has been unable to render the world leprosy-free, despite a highly effective drug regime. The global

prevalence of leprosy at the beginning of 2010 was esti-mated to be 211,903 cases (the major bulk [120,456] ofwhich were represented by patients in Southeast Asia),and new cases detected in 2009 numbered 244,796, morethan half (133,717) of which were registered in India.1

These figures indicate that the disease is yet to be tamedand, given the long incubation period of leprosy, they mayrepresent only the tip of the iceberg. Another cause ofconcern is the steady rise in the number of relapsed cases(3120 relapsed cases reported from 122 countries duringthe year 2009)1 The most urgent needs are to detect newcases early and to bring all cases under the cover of MDT.1522

International Journal of Dermatology 2011, 50, 15221527 2011 The International Society of Dermatology

Clearly, early detection requires a sensitive diagnostic toolthat has been tested at the point of care.Conventional diagnostic techniques rely on clinical

methods, which, needless to say, demand a high degree ofclinical suspicion and awareness among clinicians. Theseinclude microscopic evaluation of tissue smears for acid-fast bacilli (AFB) and histopathology.2 However, thediagnosis of leprosy remains challenging. For reliabledetection, acid-fast staining requires 104 organisms pergram of tissue3, and its sensitivity is low at the tubercu-loid end of the disease spectrum and in early cases, inwhich bacilli are rare or absent. Histopathological find-ings may also be nonspecific at times.4 Serological testsfor leprosy5,6 have shown certain potential in detecting ahost immune response against bacilli. However, evenwhen satisfactory immunodetection is achieved, this maysimply reflect a past infection and may give no informa-tion on current bacteriological status. Therefore, thedetection of bacilli is difficult, and in India, where a hugeburden of 87,190 cases were registered in at the end offirst quarter of 2010,1 diagnostic tools that can supple-ment clinicohistological findings, even in early cases, areurgently needed. Polymerase chain reaction (PCR) holdspromise as a simple and sensitive diagnostic tool,although it has some technical limitations.A recently developed PCR technique4 for the detection

of M. leprae DNA in environmental and clinical speci-mens has allowed investigators to study natural distribu-tion as well as the technique by which a very smallamount of M. leprae DNA can be detected directly inclinical specimens such as from a nasal swab, skin, etc.The usefulness of PCR in the Indian context requires

further evaluation. In this study, we applied PCR vali-dated by a previous study4 for the diagnosis of leprosyand compared the results with clinicohistological diagno-sis and the findings of slit-skin smears (SSS). The studywas undertaken with the objectives of evaluating the effi-cacy of PCR in the diagnosis of leprosy, determining theutility of PCR in the clinical setting, and comparing thefindings of PCR with those of SSS, a conventional param-eter for the diagnosis of leprosy.

Materials and methods

This cross-sectional study was carried out in the Hansens

Disease Clinic in the Outpatient Dermatology Department at the

Institute of Postgraduate Medical Education and Research,

Kolkata, and the Medical College, Kolkata, over a period of

nine years. All procedures and protocols followed in the study

were approved by the institutional ethics committees of both

institutes, and signed informed consent was obtained from all

patients before they were included in the study. Leprosy was

diagnosed clinically according to WHO guidelines7 and further

confirmed or corroborated by histopathology before any study

subject was included for final analysis. The study population

was classified into multibacillary (MB) and paucibacillary (PB)

groups as per the clinical criteria of the WHO.7

Skin biopsy, histology and slit-skin smear

Punch biopsy was taken from the active margin of lesions or

from the anesthetic area in pure neural leprosy (where skin

lesions were not present). For the purpose of biopsy, plaques

and nodules were preferred over patches in cases of the

concomitant presence of plaque, nodule, and patch. The

tissue samples were stained with hematoxylin and eosin

(H&E) for histology and FiteFaraco stain for the

demonstration of bacilli. The cases were classified after

clinicopathological correlation into tuberculoid (TT), borderline

tuberculoid (BT), mid-borderline (BB), borderline lepromatous

(BL), lepromatous (LL), indeterminate, pure neuritic, and

histoid leprosy.8 For the diagnosis of indeterminate leprosy,

any cases in which the possibility of any diagnostic mimickers

(e.g. pityriasis alba, patch stage of mycosis fungoides) was

suspected were excluded from the study. The SSS was taken

from the lesion following a standard protocol and stained using

modified ZiehlNeelsen stain for the demonstration of AFB.9

Biopsy samples were stored at ) 20 C in Tris buffer prior toDNA extraction.

Extraction of Mycobacterium leprae DNA from clinical

specimens

Genomic DNA was extracted from the skin of study subjects

using a modified version of a technique developed by Yoon

et al.10 without using Zarconium beads.

The skin biopsy samples (of both study subjects and controls)

were incised into small pieces with a no. 10 or no. 15 disposable

scalpel blade in a Petri dish. They were then homogenized using

a hand homogenizer with 1 ml sterile distilled water. The

homogenized material was then placed into separate 1.8-ml

microcentrifuge tubes and incubated with lysis buffer containing

300 ll of 100 mM Tris-HCl, pH 8.5, containing 0.05% Tween-20

and 60 lg Proteinase K per ml, for 18 hours at 60 C. Paraffinoil (40 ll) was layered on top of the sample to prevent

evaporation. Thereafter, the samples were incubated at 97 Cfor 15 minutes to inactivate Proteinase K. After the completion

of incubation, equal volumes of phenol : chloroform : isoamyl

alcohol (25 : 24 : 1) were added to each of the tubes containing

the homogenized samples. Each of the tubes was shaken gently

for 1 minute. After centrifugation at 10,000 g for 15 minutes, the

supernatant was transferred to other microcentrifuge tubes and

again mixed with phenol : chloroform : isoamyl alcohol

(25 : 24 : 1) and the centrifugation step repeated. The

supernatant was collected in separate microcentrifuge tubes;

equal volumes of chloroform were added to each tube and the

centrifugation step repeated. Supernatant fractions were

collected in microcentifuge tubes to which isopropranolol (0.7%

2011 The International Society of Dermatology International Journal of Dermatology 2011, 50, 15221527

Banerjee et al. PCR versus slit-skin smear in diagnosis of leprosy Tropical medicine rounds 1523

v/v) was added. The samples were kept at ) 20 C overnight(12 hours) and centrifugation was repeated at 12,000 g for

30 minutes the following day. The pellet of DNA was collected

by gently draining the upper portion of supernatant from the tube

without disturbing the pellet, after which 150 ll ethanol was

added to each tube and mixed thoroughly. The DNA was then

precipitated, lyophilized, and re-suspended in 100 ll distilled

water, which was subsequently used as a template DNA for

PCR.

Polymerase chain reaction

A multiplex PCR was developed in our laboratory incorporating

two sets of primers. One component of the primer set of the

multiplex PCR was designed according to Yoon et al.10 This

pair of oligomers were intended for the amplification of a 372-bp

fragment of the repetitive sequence of M. leprae DNA. The

base pair is reported to be very specific to M. leprae and is not

present in 20 other mycobacterial species. The specificity and

sensitivity of the primers R1 and R2 have been established in

earlier studies.4

Enzyme amplification (PCR) was performed with the

following sequences of oligomer primers: 5CGG CCG GATCCT CGA TGC AC3 (primer R1), and 5GCA CGT AAGCTT GTC GGT GC3 (primer R2). The reaction mixtureconsisted of 50 ll of 10 mM Tris-HCl (pH 8.5), 50 mM KCl,

1.5 mM MgCl2, 0.01% (w/v) gelatin, 200 lM dATP, dGTP,

dCTP and dTTP, 1 U of taq DNA polymerase (Perkin-Elmer

Cetus Instruments, Norwalk, CT, USA), 0.5 lM of each primer

and DNA prepared from the biopsy samples. A total of 35

cycles of PCR were performed with a thermocycler (PTC-0150

MiniCycler, Version 4.0; MJ Research, Inc., Waltham, MA,

USA). Cycles consisted of denaturation at 94 C for oneminute followed by annealing at 60 C for two minutes withprimer extension at 72 C for three minutes and a finalextension at 72 C for ten minutes (Fig. 1).

Statistical analysis

Data were statistically described in terms of range,

mean standard deviation (SD), frequency (number of cases),

and relative frequency (percentages) when appropriate. Z-test

of proportion, chi-square test, kappa test, and analysis of

variance (ANOVA) tests were used as applicable. Medcalc

statistical software Version 9.6.4.0 (http://www.medcalc.org/)

was used for statistical analysis. A P-value

(86.0%) were PCR-positive, as were 43 (75.4%) of thePB patients (chi-squared test, P = 0.142). However, AFBwere demonstrated by SSS in 65 (39.6%) cases, amongwhich positivity was significantly greater (chi-squaredtest, P < 0.0001) in MB (59.8%) than in PB (1.8%) cases.The sensitivity of PCR was significantly greater (Z-test ofproportion, P < 0.0001) in both the MB and PB sub-groups; this difference in sensitivity was remarkable inthe PB subgroup, where PCR yielded positivity in 75.4%compared with 1.8% of cases in which AFB was found inSSS (Table 1). Polymerase chain reaction positivity wasfound in 72 of 99 (72.7%) patients who showed no AFBin SSS and in 63 of 65 (96.9%) cases in which SSS didshow AFB. Two patients were negative by PCR despitebeing positive by SSS. This difference was found to be sig-nificant (chi-square test, P = 0.0002), with poor inter-rater agreement between the two tests (kappa = 0.024).In terms of the disease spectrum, PCR was maximally

positive in the LL and histoid leprosy subgroups (100%positivity) followed by the BL, TT, BT, and indeterminatesubgroups. Polymerase chain reaction was found to benegative in all cases of pure neuritic variety. Acid-fastbacilli positivity in SSS was found to be 100% in LL andhistoid leprosy and was of absolute negative yield in TT,pure neuritic, and indeterminate cases (Table 2).Results by PCR did not significantly vary with the pres-

ence of lepra reaction (chi-squared test, P = 0.5), durationof illness (ANOVA test, P = 0.364), or number of lesions

(ANOVA test, P = 0.05). Similarly, AFB in SSS showedno significant association with presence of lepra reaction(chi-squared test, P = 0.06) or duration of disease (ANO-VA test, P = 0.973), although unlike PCR, its positivitysignificantly increased (ANOVA test, P < 0.001) in linewith an increase in the number of lesions.

Discussion

Leprosy is diagnosed clinically based on the presence oftwo or more cardinal clinical features,7 the clinical appli-cability of which has been extensively reviewed. Despitethe high sensitivity of clinical methods, delays in diagno-sis in more than 80% of new cases of leprosy werereported at the Hospital of Tropical Disease in London,UK during 19951999.11 This fact highlights the need fora more sensitive diagnostic tool that is capable of diag-nosing very early-stage cases of leprosy because earlydiagnosis is the only way to prevent deformities anddecrease subsequent morbidities.The use of PCR in the diagnosis of leprosy has ushered

in the hope that the longstanding search for a sensitivetool is drawing to an end. Despite the prevailing enthusi-asm for the use of PCR, its reliability, utility, and applica-bility must be tested rigorously before it can be madeavailable in clinical settings. Just like every other goodthing, PCR comes with a price and it is expensive. In theresource-poor countries of Africa and Asia, where, unfor-

Table 1 Percentage positivity (sensitiv-ity) of polymerase chain reaction (PCR)and acid-fast bacilli (AFB) in slit-skinsmear in multibacillary and paucibacil-lary cases

Multibacillary

(n = 107)

Paucibacillary

(n = 57) Total (n = 164)

PCR (positive : negative) 92 : 15 (85.9%) 43 : 14 (75.4%) 135 : 29 (82.3%)

AFB in slit-skin smear

(present : absent)

64 : 43 (59.8%) 1 : 56 (1.8%) 65 : 99 (39.6%)

P-value (Z-test of proportion)

tunately, leprosy load is highest, this tool must be usedjudiciously, and advice about where it should and shouldnot be used must be precise. Thus, the present authors setout to identify the clinical situations in which the use ofPCR would provide an edge over other methods of diag-nosing leprosy.The present large-scale study demonstrated a positive

yield in 82.3% of all cases of clinicopathologically diag-nosed leprosy by using amplification of the 372-bp fragmentof the repetitive sequence ofM. lepraeDNA, which is lowerthan the 93.1% yield reported previously with the same setof primers.10 This discrepancy may reflect differences in theprofile of patients selected for analysis. The present studysample included patients with the pure neural (n = 5) andindeterminate (n = 6) varieties of leprosy, as well as patientson therapy with MDT (n = 29), which are associated with amore negative yield. Other studies, which used 530 bp12 orM. leprae-specific 16S ribosomal RNA13, found PCR posi-tivity rates of 92% and 100% in AFB-positive cases, and61% and 50% in AFB-negative cases, respectively. Thepresent study, using 372 bp, found better positivity thanthose using 530 bp and comparable results with those using16S ribosomal RNA, for both AFB-positive (96.9%) andAFB-negative (72.7%) cases. Targeting by PCR of the36 kDa gene of M. leprae showed positivity in a modest66.6% of cases, and the same study found in situ hybridiza-tion to show positivity rates of 42.8% in the early-stage(indeterminate/BT) and 46.7% in the BB/BL subgroups.14

However, in the present study, the early (indeterminate/TT/BT) and BL subgroups showed positivity rates of 82.3% and90.9%, respectively. This highlights the superiority of thepresent method over in situ hybridization or PCR using the36 kDa gene. A study by Goulart et al. compared differentmethods of PCR and concluded that a 130-bp primer (posi-tivity: 73.6%) was more sensitive than that for 372 bp (pos-itivity: 52.7%).15 The same study found 130 bp to bepositive in 40% of TT, 55.5% of BT, and 100% of BB, BL,and LL cases. By contrast with Goulart et al., we found abetter sensitivity profile in our study population using the372-bp primer and found positivity to be better than with130 bp at both the TT and BT poles.Several explanations can be proposed to explain PCR

negativity. No bacilli may be present in the sample, whichis likely to be true in some bi-negative cases, or bacillimay no longer be viable, either because of a strong hostimmune response or because they have been killed bytreatment.12 By contrast, we cannot rule out the possibil-ity that other factors, such as the age of the specimens,may also have an effect on the outcome of PCR. Negativeresults among PB patient samples may be explained bytheir intrinsically higher ratios of human genomic DNAto M. leprae DNA, which probably inhibits M. lepraeDNA amplification.16 In addition, there may be some

technical limitations to the detection of M. lepraeDNA caused by PCR inhibitors or other PCR proceduresadopted in this study (for reasons that are not clearto us).This study found that none of the pure neuritic cases

showed a positive PCR result. This indicates that per-forming PCR in a skin biopsy from an anesthetic/hypoes-thetic skin area in pure neuritic leprosy represents a wasteof resources. Martinez et al. showed 83.3% positivity inpure neuritic patients after TRIzol extraction of the sam-ples.16 This discrepancy may reflect a different extractionmethod, which provides better yield of DNA. By contrast,PCR has shown 33.3% positivity in indeterminate cases,which is an optimistic sign given that indeterminate casesare the most difficult to diagnose. Clinicians will alwaysbe hesitant to start MDT (which warrants at leastsix months of therapy once started) only on the groundsof perineural and/or periadnexal infiltrate from a hypo-pigmented macule (the clinicopathological finding in inde-terminate cases). The presence of a positive PCR findingwould provide the clinician with much-needed laboratoryback-up in making a diagnosis. If PCR can help in dispel-ling the diagnostic dilemma in a third of such clinicallychallenging situations, it will be welcomed by the medicalfraternity at large.It should also be remembered that, in countries like

India, leprosy is associated with stigma; hence patients(and their family members) are reluctant, if not hostile, toaccept the fact that they are affected by leprosy. Oftenthey seek objective evidence before coming to terms withthe diagnosis. Problems arise in PB cases, in which SSS isoften negative (it was positive in only 1.8% of PB casesin the present study). In this context, PCR (which waspositive in 75.4% of PB cases in this study) can providediagnostic confirmation to help ameliorate the conflictingsentiments of the patient and his or her family (includingany resentment towards the clinician, which, althoughunfortunate, does occur in real-life situations). Thus, inearly leprosy (which poses a diagnostic dilemma andoften must be differentiated from pityriasis alba,seborrheic dermatitis, etc.), PCR is a better tool than SSS.With reference to clinical mimickers of leprosy, post-kala-azar dermal leishmaniasis (PKDL) deserves specialmention because it is often confused with the LL pole ofleprosy. Our study population included one patient whowas previously misdiagnosed with PKDL because of itssimilar clinical features and the patients past history ofkala-azar. This patient showed both AFB in SSS and apositive PCR result, although in such cases the cost : ben-efit ratio would definitely favor the use of SSS over PCR.Here, it is worth mentioning that PCR is often criticizedbecause it requires skilled personnel, whose expertise isnot always affordable in the developing countries that

International Journal of Dermatology 2011, 50, 15221527 2011 The International Society of Dermatology

Tropical medicine rounds PCR versus slit-skin smear in diagnosis of leprosy Banerjee et al.1526

bear the major brunt of leprosy. Although PCR initiallyappears to be expensive, in the long run PCR analysis inbulk is economically tolerable and cost-effective.The present study indicates that PCR always shows a

higher degree of sensitivity than the detection of AFB inSSS, which has been the tool of convention in the diagno-sis of leprosy. However, overall false negative results of17.7% (14.1% in MB cases and 24.8% in PB cases) areof major concern. It is thus evident that almost one in fiveleprosy cases will be missed if we rely solely on PCR asapplied in the study. Hence, PCR should be used as asupplementary tool in the diagnosis of leprosy and neveras a replacement for clinical acumen.It should be emphasized that PCR may not be useful as

a diagnostic tool in MB cases (in which SSS has a highyield), especially in the lepromatous pole and histoid vari-eties of leprosy, in all cases of which SSS was found to bepositive. Hence, it is imperative that the use of this costlybut sensitive tool be restricted to diagnostically challeng-ing situations and PB cases (excluding pure neuritic lep-rosy), in which it can usefully supplement existingdiagnostic methods.

References

1 World Health Organization. Global leprosy situation,2010. Wkly Epidemiol Rec 2010; 85: 33748. http://www.who.int/wer/2010/wer8535.pdf [Accessed April 12,2011.]

2 Lucas SB, Ridley DS. The use of histopathology inleprosy diagnosis and research. Lepr Rev 1989; 60: 257262.

3 Shepard CC, McRae DH. A method for counting acid-fast bacteria. Int J Lepr Other Mycobact Dis 1968; 36:7882.

4 Banerjee S, Ray D, Bandopadhyay D, et al. Developmentand application of a new efficient and sensitive multiplexpolymerase chain reaction (PCR) in a diagnosis ofleprosy. J Indian Med Assoc 2008; 106: 436440.

5 Buhrer-Sekula S, Smits HL, Gussenhoven GC, et al.Simple and fast lateral flow test for classification ofleprosy patients and identification of contacts with high

risk of developing leprosy. J Clin Microbiol 2003; 41:19911995.

6 Fine PEM, Ponnighaus JM, Burgess P, et al.Seroepidemiological studies of leprosy in northernMalawi based on an enzyme-linked immunosorbent assayusing synthetic glycoconjugate antigen. Int J Lepr 1988;56: 243254.

7 World Health Organization. Expert Committee onLeprosy, 7th Report, 1998. http://www.searo.who.int/LinkFiles/Reports_2-Leprosy_7th_Geneva1998.pdf[Accessed April 12, 2011.]

8 Job CK. Pathology of leprosy. In: Hastings RC,Opromolla DVA, eds. Leprosy, 2nd edn. London:Churchill Livingstone, 1994; 193224.

9 Rees RJW, Young DB. The microbiology of leprosy. In:Hastings RC, Opromolla DVA, eds. Leprosy, 2nd edn.London: Churchill Livingstone, 1994; 4983.

10 Yoon KH, Cho SN, Lee MK. Evaluation of polymerasechain reaction amplification of Mycobacterium lepraespecific repetitive sequence in biopsy specimens fromleprosy patients. J Clin Microbiol 1993; 31: 895899.

11 Lockwood DN, Reid AJ. The diagnosis of leprosy isdelayed in the United Kingdom. QJM 2001; 94: 207212.

12 de Wit MY, Faber WR, Krieg SR, et al. Application of apolymerase chain reaction for the detection ofMycobacterium leprae in skin tissues. J Clin Microbiol1991; 29: 906910.

13 Bang PD, Suzuki K, Phuong le T, et al. Evaluation ofpolymerase chain reaction-based detection ofMycobacterium leprae for the diagnosis of leprosy.J Dermatol 2009; 36: 269276.

14 Dayal R, Agarwal M, Natrajan M, et al. PCR and in situhybridization for diagnosis of leprosy. Indian J Pediatr2007; 74: 645648.

15 Goulart IM, Cardoso AM, Santos MS, et al. Detection ofMycobacterium leprae DNA in skin lesions of leprosypatients by PCR may be affected by amplicon size. ArchDermatol Res 2007; 299: 267271.

16 Martinez AN, Britto CFPC, Nery JAC, et al. Evaluationof real-time and conventional PCR targeting complex 85genes for detection of Mycobacterium leprae DNA inskin biopsy samples from patients diagnosed withleprosy. J Clin Microbiol 2006; 44: 31543159.

2011 The International Society of Dermatology International Journal of Dermatology 2011, 50, 15221527

Banerjee et al. PCR versus slit-skin smear in diagnosis of leprosy Tropical medicine rounds 1527

This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy.Users should refer to the original published version of the material.