INTERNATIONAL JOURNAL OF LEPROSY Volume 54, Number 2Printed in the U.S.A.

Hepatitis B Virus (HBV) Serum Markers inGreek Leprosy Patients'

Dimitri J. Papaioannou, Evangelia P. Kaklamani,Nikos G. Parissis, Ivoni G. Koumantaki, Dimitri Th. Karalis,

and Dimitri B. Trichopoulos2

The association between hepatitis B sur-face antigen (HBsAg) and lepromatous lep-rosy (LL), first reported by Blumberg, et al.(4), has been the subject of several contro-versial investigations. While in some sur-veys elevated rates of HBsAg were found inLL patients (3.7' 9, 14, 16

), in others the prev-

alence of HBs antigenemia in leprosy caseswas found to be in the "normal" range(II, 17, 20, 21s .) From these studies it was con-cluded that when the prevalence of HBsAgexceeds 1% in the healthy population andthe survey sample consists mainly of hos-pital patients, an association between LLand HBsAg is usually observed (I• 8 ). On theother hand, when the prevalence of HBsAgis less than 1% in the general population,high rates of HBsAg are not found in theleprosy cases (17

• 21

• 22) .

Investigators who have found an associ-ation between LL and HBsAg favor the hy-pothesis that the immune defect which pre-disposes to LL is also responsible for theincreased susceptibility to the persistence ofHBsAg (" 6 ). Alternatively, immune re-sponse genes associated with leprosy arepostulated to be correlated with a gene forpersistent carriage of the hepatitis B virus(4. 19) Some studies, however, did not con-trol for the fact that lepromatous patientsare more likely to be hospitalized for longperiods during which the opportunity for

Received for publication on 10 October 1985; ac-cepted for publication in revised form on 17 February1986.

D. J. Papaioannou, M.D.; E. P. Kaklamani, M.D.,Associate Professor; I. G. Koumantaki, M.D., Asso-ciate Professor; D. Th. Karalis, M.D., Associate Pro-fessor; D. B. Trichopoulos, M.D., Professor, Depart-ment of Hygiene and Epidemiology, University ofAthens School of Medicine, 11527 Athens (Goudi),Greece. N. G. Parissis, M.D., Director, Center for Han-sen's Disease, Hospital for Infectious Disease, Athens,Greece.

Reprint requests to Dr. Kaklamani.

infection is greater. Thus, the argument asto whether the prevalence of HBs antige-nemia is indeed higher among lepromatouspatients, as a result of some defect, is stillunresolved ( 5 ).

The purpose of this study was to inves-tigate the degree of exposure of Greek lep-rosy patients to hepatitis B virus (HBV) andthe ability of the exposed leprosy patientsto clear the HBV from the circulation. We,therefore, studied the presence of HBsAg,antibody to the hepatitis B surface antigen(anti-HBs), and antibody to the hepatitis Bcore antigen (anti-HBc) in the sera of pa-tients with the two polar types of leprosyand in the sera of hospital controls. We ana-lyzed the data for two distinct serologicalpatterns: a) effective exposure, character-ized by the presence of HBsAg and/or anti-HBs and/or anti-HBc and b) active infec-tion, characterized by the presence ofHBsAg.

MATERIALS AND METHODSThe group studied consisted of 217 lep-

rosy cases, 135 of whom had polar lepro-matous leprosy and 82 polar tuberculoidleprosy. Cases of borderline and indeter-minate forms of leprosy were excluded fromthe study. The leprosy cases were hospital-ized or followed up as outpatients at theCenter for Hansen's Disease, Hospital forInfectious Diseases, Athens, Greece. Thecontrol group consisted of 382 patients oflow socioeconomic class hospitalized innearby hospitals for diseases other than lep-rosy. All subjects were of Caucasian origin.

Serum samples were collected from allsubjects within a six-month period and werekept at —20°C. HBsAg, anti-HBs, and anti-HBc were detected in the sera of all of thepatients and the controls by radioimmu-noassay (RIA) (Austria II, Ausab; Corab-Abbott Laboratories). The tests were per-

245

246^ International Journal of Leprosy^ 1986

TABLE 1. Distribution of leprosy cases of the two polar types (LL and TT) and of hospitalcontrols according to IIB V serum markers.

Serum markersLL TT Controls

Total Total Total

Active infection.' 9 6.7 7 8.5 22 5.8Anti-1111c + anti-HBsa 86 63.7 47 57.3 135 35.3Only anti-HBca 18 13.3 12 14.7 48 12.6Only anti-1413s" 3 2.2 5 6.1 22 5.8Absence of markers 19 14.1 II 13.4 155 40.5

Total 135 100 82 100 382 100

" The sum of these four categories refers to individuals with effective exposure.

formed blind and were completed withintwo months of collection of the sera.

RESULTSTwo distinct serological patterns were op-

erationally utilized: a) effective exposure,which is characterized by the presence inthe serum of HBsAg and/or anti-HBs and/or anti-HBc, and b) active infection, whichis characterized by the presence in the serumof HBsAg with or without the presence ofantibodies.

Table 1 shows the distribution of the lep-rosy cases and of hospital controls accord-ing to HBV serum markers. The prevalenceof HBsAg was 6.7% among LL cases, 8.5%among TT cases, and 5.8% among hospitalcontrols.

Table 2 shows the distribution of 82 TTleprosy cases, 135 LL cases, and 382 hos-pital controls according to sex, age (largegroups), and serological HBV patterns. Sta-tistical analysis with the x statistic (not x 2 )was done after stratification by age and sexwith the procedure introduced by Manteland Haenszel ("). The prevalence of effec-tive exposure was compared between the

two groups of leprosy cases and the hospitalcontrols. It was found that TT as well as LLcases had a higher prevalence of effectiveexposure in comparison to the controls (x =+4.45, p < 10 -5 and x = +5.12, p < 10 -6 ,respectively). The difference remained sta-tistically significant when all leprosy caseswere combined (x = +5.21, p < 10 -6 ). Nostatistically significant difference was foundbetween the two polar leprosy groups (x =—0.19, p > 0.30) and no significant differ-ence was found between bacteriologicallypositive and bacteriologically negative LLcases (x = —0.17, p > 0.30, data not shown).

The prevalences of active infection amongthe effectively exposed subjects of the twodifferent groups of leprosy cases and of hos-pital controls were compared using as be-fore age and sex stratification and the Man-tel-Haenszel procedure ( 11). It was found thatno statistically significant difference existedbetween TT cases and controls (x = +0.49,p > 0.30); LL cases and controls (x = —0.39,p > 0.30); the two polar leprosy groupscombined and controls (x = —0.11, p >0.30); the two polar leprosy groups (x

p > 0.30); and LL cases bacterio-

TABLE 2. Distribution of 82 polar tuberculoid leprosy cases (TT), 135 polar lepromatousleprosy cases (LL) and 382 hospital controls (C), according to sex, age (large groups), andserological patterns for hepatitis B.

Males^ FemalesSerological

patterns

<49 yrs^^ 50 yrs

<49 yrs^^ 50 yrs

TT LL C^TT LL C^TT LL C TT LL C

Active infection' 0 1 4 3 4 12 1 1 6 3 3 0Effectively exposed" 10 11 20 27 59 126 11 11 57 23 35 24

Total tested 13 15 34 31 70 190 12 11 120 26 39 38

a Presence of HBsAg.b Presence of HBsAg and/or anti-HBs and/or anti-HBc.

54, 2^Papaioannou, et al.: HBV in Greek Leprosy Patients^247

TABLE 3. Distribution of 382 hospital controls according to age and serological patternsof hepatitis B.

Serological patterns'Years of age

Total<39 40-49 50-59

Active infection 4 6 5 7 22Effectively exposed 41 36 42 108 227Total tested 89 65 74 154 382Active infection as °/o

of effectively exposed 10% 17% 12% 7% 1 0%Effective exposure

as % of total 46% 55% 57% 70% 59%

See footnotes for Table 2.

logically positive and those negative (x —+0.79, p > 0.30, data not shown).

The distribution of the hospital controlsand the leprosy cases according to age, polartype, and serological patterns of hepatitis Bis shown in Tables 3, 4, and 5.

The statistical evaluation of a possibletrend of the HBV markers with age, afterappropriate sex adjustment, was done bythe Mantel-extension procedure (n). Theresults are as follows: a) In the control group,there was a statistically significant increasein the prevalence of effective exposure withage (x = +2.56, p < 0.05). In the two polarleprosy groups, a similar increase existedbut it was not statistically significant (x =+0.93, p > 0.30 and x = +1.02, p > 0.25).b) In the control group, there was a statis-tically significant decrease in the prevalenceof active infection among exposed individ-uals with age (x = —2.19, p < 0.05). On thecontrary, among leprosy cases of either typea significant decrease was not found (X =+ 0.60, p > 0.30 and x = —0.10, p > 0.30).

DISCUSSIONThe lack of agreement between various

studies, as far as the prevalence of HBsAgamong leprosy patients is concerned, maybe due to a variety of factors, such as thegeographical area, the sample of patients,the sample of control subjects, and themethodology for testing serological param-eters and for analyzing the results. The fac-tors which have been held responsible forincreased antigenemia (where it was found)are, on the one hand, the probable immu-nological defect in leprosy subjects (7, 16. 18)

and, on the other hand, increased risk ofexposure to the virus ( 10 . 19). However, it hasnot been proved that there is really a non-specific immunological defect in leprosysubjects, and there are no studies which es-tablish the proportion of leprosy patientseffectively exposed to HBV infection afterinvestigating all of the serological parame-ters. Thus, no indisputable interpretationcan be given to the "increased" prevalence

TABLE 4. Distribution of 82 polar tuberculoid leprosy cases (TT) according to age andserological patterns.

Serological patterns'Years of age

Total<39 40-49 50-59

Active infection 1 0 2 4 7Effectively exposed 7 14 18 32 71Total tested 9 16 22 35 82Active infection as %

of effectively exposed 14% 0% 11% 13% 10%Effective exposure

as % of total 78% 88% 82% 91% 87%

See footnotes for Table 2.

248^ International Journal of Leprosy^ 1986

TABLE 5. Distribution of 135 polar lepromatous leprosy cases (LL) according to ageand serological patterns of hepatitis B.

Serological patterns"Years of age

Total<39 40-49 50-59 L-60

Active infection 0 2 3 4 9Effectively exposed 4 18 38 56 116Total tested 7 19 45 64 135Active infection as %

of effectively exposed 0% I I% 8% 7% 8%Effective exposure

as % of total 5 7% 95% 84% 88% 86%

See footnotes for Table 2.

of HBsAg which has been found in variousstudies. Hence, it is of primary importanceto establish the population of subjects ex-posed to HBV and to investigate the prev-alence of antigenemia within this popula-tion. It should be noted that the degree ofexposure to the virus in nonleprosy indi-viduals is probably different, and that ex-posed leprosy patients should be comparedwith a population on nonleprosy cases ex-posed to HBV.

In the present study, by using a sensitiveradioimmunoassay to search for three se-rological markers of HBV (HBsAg, anti-HBc, anti-HBs), we were able to investigatetwo parameters: a) the proportion of sub-jects effectively exposed to HBV, and b) theproportion of subjects with active HBV in-fection. Exposed subjects were consideredto be those in whom one or more of theserological markers of HBV were found,while subjects with active infection wereconsidered to be those in whom the pres-ence of HBsAg in the serum was establishedwith or without the other two serologicalmarkers.

The investigation of these serological pa-rameters established that, in our material,the prevalence of effectively exposed sub-jects in patients with lepromatous leprosy,in patients with tuberculoid leprosy, and inhospital controls was approximately 86%,87%, and 59%, respectively. Statistical anal-ysis revealed that the leprosy patients of bothpolar types had a significantly greater prev-alence of effective exposure to HBV thancontrol patients. On the contrary, no statis-tically significant or biologically importantdifference was found concerning the degreeof exposure to HBV between the two polar

types of the disease and between patientswith lepromatous leprosy with a positive ornegative bacterial index (BI).

These comparisons show that the leprosypatients studied, who were a representativesample of all Greek leprosy patients treatedat the Center for Hansen's Disease in Ath-ens, have an increased exposure to HBVwhen compared to hospital controls. Theabsence of any significant difference in HBVexposure between the two polar types of thedisease and between leprosy patients with apositive or negative BI implies that the in-creased prevalence is due not to biologicalparameters but to the greater opportunityfor infection. Other workers( 7 . 10 ) also be-lieve that leprosy patients will manifest agreater probability of active exposure toHBV because of their protracted stay in ahospital environment.

The basic purpose of this study was tofind the proportion of HBsAg carriers amongleprosy cases effectively exposed to HBV,and to compare the findings with those froma corresponding sample of effectively ex-posed nonleprosy cases in order to drawconclusions concerning the biological fac-tors which determine the course of HBVinfection in leprosy patients.

The results of our study indicate that theproportion of subjects with active infectionamong those effectively exposed to HBVdoes not differ among groups of patientswith tuberculoid leprosy, lepromatous lep-rosy with a positive or negative BI, or hos-pital controls. These findings do not impli-cate special biological factors responsible forthe development of HBV antigenemiaamong leprosy patients, although these fac-tors may play a role in other situations or

54, 2^Papaioannou, et al.: IIBT/ in Greek Leprosy Patients^249

under different conditions ( 2 ). It is highlylikely that the establishment of HBs anti-genemia is of multifactorial origin ( 23 ), butour findings indicate that exogenous envi-ronmental factors are the critical determi-nant of HBs antigenemia among leprosycases in Greece. Our results are in accor-dance with several previous studies whichdid not find an increased prevalence of an-tigenemia in leprosy patients (' 5 ' 17 ' 20 ' 21 ) andchallenge the results of those which foundincreased HBV antigenemia in leprosypatients (particularly in those with the lep-romatous type of the disease) because ofalleged disturbances of the nonspecific cell-mediated immunity among these patients(7 . 9 ). The disagreement may be due to dif-ferences in the samples of leprosy casesstudied, differences in the control groupsused, or to methodological differences inboth the tracing of serological parametersand the evaluation of results. Our leprosycases of both types were under continuousmedical surveillance and lived under betterconditions than many cases used in pre-vious studies, which may have a beneficialeffect on the immunological status. Never-theless, the lack of a correlation between theBI and active HBV infection does not sup-port the hypothesis that reduced nonspecificcellular immunity is responsible for the per-sistent antigenemia (15.20).

In our study the only indication of aninability of leprosy cases to clear HBsAgfrom the circulation was the lack of a down-ward trend with age in the prevalence ofactive HBV infection among patients effec-tively exposed; whereas among controls,such a trend was both obvious and statis-tically significant. This indication is, how-ever, rather weak and the main conclusionof our study is that leprosy patients of anypolar type do not have an inherently in-creased susceptibility to the infection byhepatitis B virus or to persistent HBs an-tigenemia following an infection by this vi-rus.

SUMMARYThe prevalence of HBsAg, anti-HBs, and

anti-HBc in the sera of 217 patients withthe two polar types of leprosy and 382 hos-pital controls was studied in order to in-vestigate the degree of exposure of Greek

leprosy patients to HBV and the ability ofthese patients to clear HBV from the blood.Two distinct serological patterns were ana-lyzed: a) effective exposure, characterizedby the presence of one or more of the threeserological markers, and b) active infection,characterized by the presence of HBsAg.

From the statistical analysis it was foundthat TT as well as LL cases had a higherprevalence of effective exposure in com-parison to controls (p < 10 -5 and p < 10 -6 ).No significant difference was found betweenthe two polar leprosy types (p > 0.30) orbetween bacteriologically positive and neg-ative LL cases (p > 0.30).

As far as the prevalence of active infec-tion is concerned among the effectively ex-posed subjects of all groups, no significantdifference existed between TT cases andcontrols, LL cases and controls, the two po-lar leprosy cases combined and controls, thetwo polar leprosy groups, and LL cases pos-itive and negative for Mycobacterium leprae(p for all comparisons >0.30).

It is concluded that leprosy cases are at ahigh risk of HBV infection, but the preva-lence of active infection among those effec-tively exposed does not differ between lep-rosy cases and hospital controls.

RESUMENPara poder investigar el grado de exposiciOn de los

pacientes Griegos con lepra al virus de la hepatitis B(HBV) y su capacidad para climinarlo de circulaciOn,se estudiO la prevalencia del antigeno HBsAg y de losanticuerpos anti-HBs y anti-HBc en el suero de 217pacientes con alguno de los dos tipos polares de la lepray en el suero de 382 controles de hospital. Se analizaron2 patrones serolOgicos distintos: a) la exposici6n efec-tiva caracterizada por la presencia de uno o mas de los3 marcadores serolOgicos y b) la infecciOn activa ca-racterizada por la presencia de HBsAg.

Se encontrO que tanto los pacientes TT como los LLtuvieron una mayor prevalencia de exposiciOn efectivaque los individuos del grupo control (p < 10 -3 y p <10 -6). No se encontr6 una diferencia significativa entrelos dos tipos polares de lepra (p > 0.3) ni entre loscasos LL bacteriolOgicamente positivos y los negativos(p > 0.3).

En cuanto a la prevalencia de la infecciOn activa entrelos sujetos efectivamente expuestos, no se encontrarondiferencias significativas entre los casos TT y los con-troles, entre los casos LL y los controles, entre los casosLL y TT, ni entre los casos LL bacteriolOgicamentepositivos y los negativos (en todos los casos los valoresde p fueron >0.3).

250^ International Journal of Leprosy^ 1986

Se concluye que los pacientes con lepra tienen unalto riesgo de infecciOn por el HI3V pero que la pre-valencia de infecciOn activa en los individuos efecti-vamente expucstos es igual a la encontrada en los con-troles de hospital.

RESUMEEn vue d' explorer Ic degre d'exposition de malades

grecs atteints de lepre a HBV et la capacité de cesmalades a eliminer HBV de leur sang, on a etudie laprevalence de HBsAg, d'anti-HBs et d'anti-HI3c dansIc serum de 217 malades atteints de l'une ou l'autredes formes polaires de lepre, et de 382 temoins choisisparmi les malades hospitalises.

Deux profils serologiques distincts ont etc analyses:a) une exposition reelle caracterisee par la presenced'un ou plusieurs des trois marqueurs serologiques; b)une infection active, caracterisee par la presence deHBsAg. L'analyse statistique a permis de montrer queles cas TT aussi bien que les cas LL presentaient unefrequence plus &levee d'exposition reelle, par compa-raison aux temoins (p < 10 -5 et p < 10 -6 ). Aucunedifference significative n'a etc trouvee en fonction desdeux types polaires de la lepre (p > 0.30), ou du Ca-ractere bacteriologique positif ou negatif des cas LL(p > 0.30).

En ce qui concerne la prevalence d'infection reciteparmi les sujets reellement exposés dans chacun desgrouper, aucune difference significative n'a etc mise enevidence entre les cas TT et les temoins, les cas LL etles temoins, les cas des deux types polaires de la leprecombines, d'une part, et les temoins d'autre part, lesdeux types polaires de la lepre, et les cas LL respective-ment positifs ou negatifs pour Mycobacterium leprae(la probabilite pour toutes ces comparaisons etait su-perieure a 0.30).

On en conclut que les cas de lepre presenient unrisque Cleve d'infection HBV mais que la prevalenced'une infection active parmi ceux qui ont etC reelle-ment exposés ne dillere pas entre les cas de lepre et lestemoins hospitaliers.

Acknowledgments. We thank Ms. P. Drongari andMs. E. Toussi for the technical assistance. This workwas supported by a grant from the World Health Or-ganization.

REFERENCES1. ANATHAKRISHNAN, R., ARNDT-HANSER, A. and

WALTER, H. Association between Australia anti-gen and leprosy. Humangenetik 16 (1972) 235-239.

2. BLUMBERG, B. S., FRIEDLANDER, J. S., WOODSIDE,

A., SUTNICK, A. I. and LONDON, W. T. Hepatitisand Australia antigen: autosomal recessive inher-itence of susceptibility to infection in humans. Proc.Natl. Acad. Sci. U.S.A. 62 (1969) 1108-1115.

3. BLUMBERG, B. S. and MELARTIN, L. Australia an-tigen and lepromatous leprosy studies in South

India and elsewhere. Int. J. Lepr. 38 (1970) 60-67.

4. BLUMBERG, B. S., MELARTIN, L., Li.cilA , M. andGUINTO, R. S. Association between lepromatousleprosy and Australia antigen. Lancet 2 (1967) 173-176.

5. BULLOCK, W. E. Leprosy: a model of immunolog-ical perturbation in chronic infection. J. Infect.Dis. 137 (1978) 341-354.

6. CIIIRON, J. P., DI:NIS, F., MAUPAS, P. andLANGUILLON, J. Infection par le virus de l'hépatiteB chez les hanseniens. I.- Detection de l'antigeneet de l'anticorps de surface du virus de l'hepatiteB par radio-immunologic. Bull. Soc. Med. Afr.Noir Lang. Fr. 23 (1978) 174-182.

7. FAKUNLE, Y. M. and Will Fru:, H. C. Hepatitis-Bvirus infection in patients with leprosy: a serolog-ical study in a leprosarium in Northern Nigeria.Trans. R. Soc. Trop. Med. Hyg. 75 (1981) 623-625.

8. FRANCIS, T. I. and SMITH, J. A. Australia (Aul)antigen in Nigerian patients with leprosy. Int. J.Lepr. 40 (1972) 68-72.

9. KITKAR, S. S., NIPHADKAR, K. B., KIIARE, P. M.and GlIARPURAY, M. B. Environment and carriageof hepatitis B antigen in leprosy. Indian J. Med.Res. 62 (1974) 1791-1799.

10. KELKAR, S. S., NIPHADKAR, K. B., KHARE, P. M.and JUNNARKAR, R. V. Hepatitis B antigen in aleprosy hospital. Bull. WHO 48 (1973) 555-558.

11. LECHAT, M., MELARTIN PREHN, L., BLUMBERG, B.S. and MORN R. Australia antigen in Zaire. Stud-ies on leprosy. Ann. Soc. Belg. Med. Trop. 53 (1973)173-178.

12. MANTEL, N. Chi-square tests with one degree offreedom, extensions of the Mantel-Haenszel pro-cedure. J. Am. Statist. Assoc. 58 (1963) 690-700.

13. MANTEL, N. and HAENSZEI., W. Statistical aspectsof the analysis of data from retrospective studiesof disease. J. Natl. Cancer Inst. 22 (1959) 719-748.

14. PAPAEVANGELOU, G. L., PAPASTAVROPOULOS, J. andKOUREA, T. Hepatitis-associated antigen (HAA)in leprosy. Lepr. Rev. 42 (1972) 273-276.

15. PAPAGEORGIOV, P. S., VERNACE, S. and GLADE, P.R. Hepatitis-associated antigen and cell-mediatedimmunity. Lancet 1 (1972) 1118.

16. SAHA, K. and DUTTA, R. Subtypes of Australiaantigen in persistent Australia antigenemia andsporadic hepatitis among patients with leproma-tous leprosy and their segregated children with noapparent clinical illness. Int. J. Lepr. 45 (1977)38-48.

17. SALZANO, F. M. and BLUMBERG, B. S. The Aus-tralia antigen in Brazilian healthy persons and inleprosy and leukaemic patients. J. Clin. Pathol. 23(1970) 39-42.

18. SAOH, A. M. and MENE, A. R. Persistence of Aus-tralia antigen in leprosy-a frustrating puzzle inimmunology. Lepr. India 50 (1978) 7-10.

54, 2^Papaioannou, et al.: IIBV in Greek Leprosy Patients^251

19. SERJEANTSON, S. and W000E -JELL), D. G. Immune

response of leprosy patients to hepatitis B virus.

Am. J. Epidem. 107 (1978) 321-327.

20. SHER, R., MACKAY, E. M., MACNAII, G. M., KOK,

S. H. and KooRNEE0E, H. J. Hepatitis B antigen,

hepatitis B antibody, and subtypes in leprosy. In-

fect. Immun. 17 (1977) 1-3.21. SIIwE, T. and ZUCKERNIAN, A. J. Australia antigen

and antibody in British patients with leprosy. J.Clin. Pathol. 25 (1972) 401-402.

22. SWANEPOEL, R. and CRUICKSHANK, J. G. Australiaantigen in Rhodesia. Lancet 2 (1972) 446.

23. YAP, E. H., ONG, Y. N., SIMONS, M. J., OKOCIII,

K., MAYUMI, M. and NISHIOKA, K. Australia an-tigen in Singapore. II. Differential frequency in

Chinese, Malays and Indians. Vox Sang. 22 (1972)371-375.