Histomorphometry of hepatic portal fibrosis in patients with surgical schistosomiasis mansoni

Histomorfometria da fibrose portal hepatica em pacientes com esquistossomose mansônica cirúrgica

Abstracts

The usual histology report of hepatic fibrosis in patients with hepatosplenic schistosomiasis mansoni presents no association with hemodynamic and clinical liver parameters. Histomorphometry is adding a new tool of investigation for measuring density of portal fibrosis in these patients. This investigation was set up for assessing a possible agreement between the well-accepted international classification and the fibrosis density grades measured by histomorphometry. Thirty-five children and equal number of adults were included in this study. All patients underwent splenectomy and ligature of the left gastric vein. Histology findings were assessed in surgical liver biopsy stained with Masson trichrome. The official histology report was used as reference. The histomorphometric studies were done by semi-automatic morphometry. The mean percentage (X) of portal fibrosis plus or minus one standard deviation (SD) was classified as grade II (7.06% up to 34.72%); grade I was up to 7.06%; and grade III above 34.72%. Although, not reaching statistical significance, there is a tendency of the fibrosis to be more intense in children than adults (X±SD - 22.02±13.46% versus 20.63%±15.33% "t" = 0.379 p>0.05). Seven out of nine (77.8%) patients classified as grade I, by morphometry, had the same result on the official report, however, two (22.2%) were described as grade III. Sixteen out of forty-four (36.4%) classified as grade II on morphometry had the same classification as the histology grade, but, twenty seven (61.4%) were classified as grade III and one (2.3%) as grade I. Fifteen (21.4%) out of 70 patients had grade III on both classifications, but, two (11.8%) out of seventeen G III on morphometry were grade II. The kappa (k) measurement of agreement between both classification was k = 0.319, showing a fair strength of association. The histomorphometric measurements of Symmers fibrosis in surgical patients with mansonic schistosomiasis partially support the report from The Department of Pathology - University Hospital, Federal University of Pernambuco - Brazil. However, there is a discrepancy in grade III. While in the official classification, the majority (62.8%) accounts for this grade, on morphometry only 38.6% represent the same grade. On the contrary, on morphometry, similar majority (62.8%) is grade II.

Portal liver fibrosis; Symmers' fibrosis; Schistosoma mansoni; Histomorphometry


O relato usual de fibrose hepática em pacientes com esquistossomose mansônica na forma hepato-esplênica não apresenta associação com os parâmetros clínico-hemodinâmicos e de reserva funcional hepática. A histomorfometria adicionou uma nova ferramenta de investigação para medir a densidade da fibrose portal nesses pacientes. Esta investigação teve como objetivo investigar uma possível concordância entre a bem aceita classificação internacional clássica e os graus de densidade da fibrose portal medidos pela histomorfometria. Trinta e cinco crianças e igual número de adultos foram incluídos nesse estudo. Todos os pacientes foram submetidos a esplenectomia e ligadura da veia gástrica esquerda. Os achados histológicos foram avaliados em biópsias hepáticas cirúrgicas coradas pelo tricômico de Masson. O relato histológico oficial (clássico) foi usado como referência. Os estudos histomorfométricos foram feitos por morfometria semi-automática. A percentagem média (X) da fibrose portal ± um desvio padrão (DP) foi classificado como grau II (7,06 até 34,72%); grau I foi até 7,06%; e grau III acima de 34,72%. Embora não alcançando significação estatística ocorreu uma tendência da fibrose ser mais intensa nas crianças que nos adultos (X ± DP - 22,02 ± 13,46% versus 20,63% ± 15,33% "t" = 0,379 p > 0,005). Sete de nove (77,8%) pacientes classificados como grau I, pela morfometria, tiveram a mesma classificação no relato oficial, contudo, dois (22,2%) foram descritos como grau III. Dezesseis de 44 (36,4%) classificados como grau II na morfometria tiveram a mesma classificação na histologia clássica, entretanto, 27 (61,4%) foram classificados como grau III e um (2,3%) como grau I. Quinze (21,4%) de 70 pacientes tiveram grau III em ambas as classificações, mas, dois (11,8%) de dezessete pacientes grau III na morfometria foram grau II na histologia. A medida de concordância kappa (k) entre ambas as classificações foi k=0,319, mostrando uma discreta força de associação. As medidas histomorfométricas da fibrose de Symmers em pacientes cirúrgicos portadores de esquitossomose mansônica dão suporte parcial ao relato oficial do Departamento de Patologia do Hospital das Clínicas - Universidade Federal de Pernambuco. Contudo, existe uma discrepância no grau III. Enquanto na classificação oficial a maioria (62,8%) é constituída por pacientes grau III, na morfometria, a maioria (62,8%) é grau II.

Fibrose hepatica portal; Fibrose de Symmers; Schistosoma mansoni; Histomorfometria


HISTOMORPHOMETRY OF HEPATIC PORTAL FIBROSIS IN PATIENTS WITH SURGICAL SCHISTOSOMIASIS MANSONI 1 NOTES

Carlos Teixeira Brandt2 NOTES

Ana Lúcia Coutinho Domingues 3 NOTES

Paulo Vilela 4 NOTES

Andréa Sena 5 NOTES

Karina Marques 5 NOTES

Cecília Farias Giusti 5 NOTES

Brandt CT, Domingues ALC, Vilela P, Sena A, Marques K, Giusti CF. Histomorphometry of hepatic portal fibrosis in patients with surgical schistosomiasis mansoni. Acta Cir Bras 2001; 17 (supl. 1):7-10.

SUMMARY

The usual histology report of hepatic fibrosis in patients with hepatosplenic schistosomiasis mansoni presents no association with hemodynamic and clinical liver parameters. Histomorphometry is adding a new tool of investigation for measuring density of portal fibrosis in these patients. This investigation was set up for assessing a possible agreement between the well-accepted international classification and the fibrosis density grades measured by histomorphometry. Thirty-five children and equal number of adults were included in this study. All patients underwent splenectomy and ligature of the left gastric vein. Histology findings were assessed in surgical liver biopsy stained with Masson trichrome. The official histology report was used as reference. The histomorphometric studies were done by semi-automatic morphometry. The mean percentage (X) of portal fibrosis plus or minus one standard deviation (SD) was classified as grade II (7.06% up to 34.72%); grade I was up to 7.06%; and grade III above 34.72%. Although, not reaching statistical significance, there is a tendency of the fibrosis to be more intense in children than adults (X±SD - 22.02±13.46% versus 20.63%±15.33% "t" = 0.379 p>0.05). Seven out of nine (77.8%) patients classified as grade I, by morphometry, had the same result on the official report, however, two (22.2%) were described as grade III. Sixteen out of forty-four (36.4%) classified as grade II on morphometry had the same classification as the histology grade, but, twenty seven (61.4%) were classified as grade III and one (2.3%) as grade I. Fifteen (21.4%) out of 70 patients had grade III on both classifications, but, two (11.8%) out of seventeen G III on morphometry were grade II. The kappa (k) measurement of agreement between both classification was k = 0.319, showing a fair strength of association. The histomorphometric measurements of Symmers fibrosis in surgical patients with mansonic schistosomiasis partially support the report from The Department of Pathology - University Hospital, Federal University of Pernambuco - Brazil. However, there is a discrepancy in grade III. While in the official classification, the majority (62.8%) accounts for this grade, on morphometry only 38.6% represent the same grade. On the contrary, on morphometry, similar majority (62.8%) is grade II.

SUBJECT HEADINGS: Portal liver fibrosis. Symmers' fibrosis. Schistosoma mansoni. Histomorphometry.

Brandt CT, Domingues ALC, Vilela P, Sena A, Marques K, Giusti CF. Histomorfometria da fibrose portal hepatica em pacientes com esquistossomose mansônica cirúrgica. Acta Cir Bras 2001; 17 (supl. 1):7-10.

RESUMO

O relato usual de fibrose hepática em pacientes com esquistossomose mansônica na forma hepato-esplênica não apresenta associação com os parâmetros clínico-hemodinâmicos e de reserva funcional hepática. A histomorfometria adicionou uma nova ferramenta de investigação para medir a densidade da fibrose portal nesses pacientes. Esta investigação teve como objetivo investigar uma possível concordância entre a bem aceita classificação internacional clássica e os graus de densidade da fibrose portal medidos pela histomorfometria. Trinta e cinco crianças e igual número de adultos foram incluídos nesse estudo. Todos os pacientes foram submetidos a esplenectomia e ligadura da veia gástrica esquerda. Os achados histológicos foram avaliados em biópsias hepáticas cirúrgicas coradas pelo tricômico de Masson. O relato histológico oficial (clássico) foi usado como referência. Os estudos histomorfométricos foram feitos por morfometria semi-automática. A percentagem média (X) da fibrose portal ± um desvio padrão (DP) foi classificado como grau II (7,06 até 34,72%); grau I foi até 7,06%; e grau III acima de 34,72%. Embora não alcançando significação estatística ocorreu uma tendência da fibrose ser mais intensa nas crianças que nos adultos (X ± DP - 22,02 ± 13,46% versus 20,63% ± 15,33% "t" = 0,379 p > 0,005). Sete de nove (77,8%) pacientes classificados como grau I, pela morfometria, tiveram a mesma classificação no relato oficial, contudo, dois (22,2%) foram descritos como grau III. Dezesseis de 44 (36,4%) classificados como grau II na morfometria tiveram a mesma classificação na histologia clássica, entretanto, 27 (61,4%) foram classificados como grau III e um (2,3%) como grau I. Quinze (21,4%) de 70 pacientes tiveram grau III em ambas as classificações, mas, dois (11,8%) de dezessete pacientes grau III na morfometria foram grau II na histologia. A medida de concordância kappa (k) entre ambas as classificações foi k=0,319, mostrando uma discreta força de associação. As medidas histomorfométricas da fibrose de Symmers em pacientes cirúrgicos portadores de esquitossomose mansônica dão suporte parcial ao relato oficial do Departamento de Patologia do Hospital das Clínicas - Universidade Federal de Pernambuco. Contudo, existe uma discrepância no grau III. Enquanto na classificação oficial a maioria (62,8%) é constituída por pacientes grau III, na morfometria, a maioria (62,8%) é grau II.

DESCRITORES: Fibrose hepatica portal. Fibrose de Symmers. Schistosoma mansoni. Histomorfometria.

INTRODUCTION

The main finding in the advanced form of schistosomiasis mansoni is Symmers´ fibrosis in the liver. This lesion is characterized by wide fibrous bands, which follow the portal vein branches since its entry in the hepatic pedicle up to the terminal subdivisions. In the wedge of the liver, in the smallest portal spaces the main portal vein is not seen. In the central areas the portal vein is dilated. In Brazil, Symmers´ fibrosis has been the subject of several studies1-9.

As a consequence of different gradient hepatic portal inflow blockade due to S. mansoni eggs and granulomas and by the consequent fibrous bands around the wide portal branches, portal hypertension occurs. Besides the obstructive portal lesions, vascular new growth, of angiomatoide type, develops. It helps to maintain the total hepatic blood inflow, which in turn preserves the liver cells, avoiding clinical manifestations of liver failure in the balanced form of the S. mansoni disease. The secondary splenic overflow due to huge spleen is another etiologic important factor of the portal hypertension associated with the hepatosplenic form of this disease10-17.

Surgical wedge hepatic biopsy has been considered the safest method for the confirmation of Symmers´ fibrosis. It shows fibrous bands dissecting the liver parenchyma without forming true nodules, unless in the ones close to the Glisson capsule18. Histology of the needle biopsy on schistosomiasis mansoni not always represents the liver status as a whole. The hepatic findings on this disease, although diffuse, are not uniform, however they vary widely in intensity according to different areas of the liver19.

The degree of the hepatic fibrosis in hepatosplenic schistosomiasis mansoni has been reported according to the classification described by Coelho, in 1971. This classification is adopted by the Department of Pathology - University Hospital - Federal University of Pernambuco20.

In the past decade histomorphometry evaluation was introduced aiming to obtain a more precise quantification of fibrosis in cirrhotic patients21. It was also used in schistosomiasis mansoni in mice22. In 1998, a study was reported comparing the ultrasonographic grading of Symmers´ fibrosis with the classical histology classification and also with the semiautomatic morpho-metry. It was not found correlation among these three methods23. The aim of this investigation was to make a comparative study between the classical classification and the histomorphometric mean percentages of Symmers´ fibrosis, including children and adult patients.

METHODS

Thirty-five children and equal number of adults, suffering from hepatosplenic schistosomiasis mansoni with previous episodes of upper digestive bleeding were included in this study. The age of the adults ranged from 16 to 60 years, with a mean of 40.3, while the children's age ranged from 8 to 16 years, with a mean of 20.4. Forty-two patients were male and twenty-one female. These patients were cared for at the University Hospital - Federal University of Pernambuco - from 1995 to 2000.

The patients underwent splenectomy and ligature of the left gastric vein. And for the children autoimplantation of spleen tissue in the major omentum was added.

Liver wedge biopsy was done in the left lobe and sent for histology in the Department of Pathology. The tissue specimens were fixed in 10% buffered formalin. The sections were embedded in paraffin wax and sectioned at 4 mm slices. These sections were stained by hematoxylin-eosin and Masson´s trichrome. Systematic evaluation of Symmer´s fibrosis, portal inflammatory infiltration, portal granuloma and schistosomotic pigment.

The histology assessment of Symmer´s fibrosis was done according to the classification stated by Coelho in 19717. Two different pathologists made this investigation independently.

The histomorphometry was done at "Laboratório de Histopatologia - Centro de Pesquisas Gonçalo Muniz", Fiocruz, Bahia and at "Laboratório de Imuno Patologia Keiso Asami (LIKA)". Two slides of each liver fragment was stained by red picrosirius24 and Masson`s trichrome. These sections were assessed through semiautomatic morphometry using the processing and analyzing Leica Q500 MC system (LeicaR, Cambridge Ltda, Cambridge, England). Ten fields were selected at random each one measuring 1,679.360 mm2. The Symmers´ fibrosis area was measured directly, divided by the field area and expressed as percentages. The mean percentage or density was then obtained.

The histomorphometry classification was established using the mean percentages. Degree II = the mean plus or minus 1 SD (X +/- 1 SD). Degree I = percentages smaller than the mean minus 1 SD (< X - 1 SD). Degree III = greater than the mean plus 1 SD (> X + 1 SD).

The Student "t" test was used for assessing difference between means. The agreement measurement was done by the kappa (K) test. p < 0.05 rejected the null hypothesis.

RESULTS

From the 70 patients included in the present investigation 44 (62.8%) were classified as grade III by the routine histology; 18 patients (25.7%) as grade II; and eight (11.4%) as grade I.

Using the histomorphometric classification, established for the present study the patients could be graded as follows: nine patients (12.9%) were classified as grade I; 44 patients (62.8%) as grade II; and seventeen (24.3%) as grade III.

The distribution of the findings in grades I, II e III, from the adults, children and the whole group is expressed, respectively in tables 1, 2 e 3.

DISCUSSION

Different mechanisms can produce portal fibrosis in human beings. It has even been described in normal subjects without previous liver diseases. Domingues, 1998, for example, found, by ultrasonography, in healthy subjects used as a control group, a mean percentage of portal fibrosis of 4.60 ± 1.16%. However, this percentage was significantly smaller than the mean percentage found in patients with Symmer`s fibrosis (15.48 ± 8.81%)23. The child mean percentage of hepatic fibrosis however greater than of the adult patients did not reach statistical significance (X±SD - 22.02±13.46% versus 20.84±13.83% "t" = 0.379 p>0.05).

The finding of an important density of Symmer´s fibrosis in young patients, representing a S. mansoni severe disease, makes evident an important failure in its public control23, 25,26.

The basic considerations for the usual Symmer`s fibrosis classification are based on subjective interpretation of the amount and the extension of the fibrous tissue in the portal liver spaces and in summary is as follows: grade I - the portal spaces shows a greater number of young connective cells, discrete colagenous production and the presence of variable inflammatory infiltration, however, the periportal lamina and the reticular tissue are normal; grade II - the expansion of the connective tissue with collagen radial septa giving, at the same time, a star resemblance; and grade III- the connective septa produce bridges with other portal spaces or with the central vein making the new formed angiomatous tissue very conspicuous. However, it must be emphasized that the portal fibrosis evaluation is based on the greater founded spaces. This classical classification could, however, over estimate the cases considered as degree III, because there is just the need of a single portal space with the features of fibrosis degree III for the whole liver specimen be considered as degree III. The pathologist does not take into account other spaces with features of lesser intense fibrosis.

If one makes an analysis of the agreement study between the classical classification and the one done by histomorphometry it can be observed that: seven out of nine (77.8%) patients classified as grade I, by morphometry, had the same result on the official report, however, two (22.2%) were described as grade III. Sixteen out of forty-four (36.4%) classified as grade II on morphometry had the same classification as the histology grade, but, twenty-seven (61.4%) were classified as grade III and one (2.3%) as grade I. Fifteen (21.4%) out of 70 patients had grade III on both classifications, but, two (11.8%) out of seventeen G III on morphometry were grade II on routine histology.

The greater discrepancy between the two classifications occurred in grade III where there was only an agreement in 38.6% of the patients. The routine histology classification used by the Department of Pathology - University Hospital - Health Sciences Center- Federal University of Pernambuco, Brazil, overemphasis the grade III, 44 out of 70 patients (62.8%) included in this investigation, while on the histomorphometry classification, just seventeen (24.3%) were grade III.

On the basis of the morphometric findings, it is possible to suggest a new examination of the histology classical classification, perhaps increasing the number of degrees and reserving the highest degree only for those specimens with more than 50% of the portal spaces with the features of more advanced fibrosis. The greatest classification mismatch mistake occurred in two patients, one adult and one child, who had the histomorphometric classification as grade I and they were III on the classical classification. On the other hand, twenty-seven (ten adults and seventeen children), classified as grade II by histomorphometry were grade III by histology. It is possible that changing the classical classification may produce comparable results.

Address for correspondence:

Carlos Teixeira Brandt

Univ. Federal Pernambuco -

Centro de Ciências SaúdeAv. Moraes Rego, s/n HC - Bloco A

Recife-PE - 50670-420

Tel/Fax: (81) 3271-8519

E-mail: carlosbrandt@bol.com.br

1. From the Department of Surgery - Health Sciences Center, Federal University of Pernambuco.

2. Full Professor - Pediatric Surgery, Federal University of Pernambuco

3. Associated Professor - Department of Internal Medicine

4. Assistant Professor - Pediatric Surgery, Federal University of Pernambuco

5. Brazilian National Research Council (CNPq) Scholarship, UFPE

  • 1. Magalhăes Filho A. Morfogęnese da fibrose hepática na esquistossomose mansônica humana. Tese de Livre Docęncia - Faculdade de Medicina da Universidade do Recife 1952; 98p.
  • 2. Bogliolo L. Sobre o quadro anatômico do fígado na forma hepato-esplęnica da esquistossomose mansônica. O Hospital 1954; 45:284-306.
  • 3. Bogliolo L. The anatomic picture of the liver in hepato-splenic schistosomiasis mansoni, Ann Trop Méd Parasit 1957; 51:1-14.
  • 4. Coutinho A. A hipertensăo porta na síndrome hepato-esplęnica esquistossomótica. Tese de Professor Titular - Faculdade de Medicina da Universidade do Recife 1960; 175p.
  • 5. Coelho RB, Magalhăes Filho A, Carvalho ARL, Bandeira V. Patologia da esquistossomose mansônica. Monografia - Imprensa Universitária - Universidade do Recife 1965; 110p.
  • 6. Andrade ZA. Hepatic schistosomiasis. Morphological aspects. In: Popper H, Schaffner F. Progress in liver diseases. Grune & Stratton, New York 1965; 228-42.
  • 7. Coelho RB. Anatomia Patológica das afecçőes hepáticas. Monografia - Imprensa Universitária - Universidade do Recife 1971; 174-90.
  • 8. Andrade ZA, Cheever AW. Clinical and pathological aspects of schistosomiasis in Brazil. In: Mostofi FK. Bilharziasis. Springer-Velarg, Berlin 1967; 157-66.
  • 9. Andrade ZA, Bina JC. A patologia da forma hepato-esplęnica da esquistossomose mansônica em sua forma avançada (Estudo de 232 necrópsias completas). Mem Inst Oswaldo Cruz 1983; 78:285-305.
  • 10. Mies S, Larsson E, Mori T, Rosa P, Raia S. O sistema porta e a artéria hepática, esplęnica e mesentérica superior na esquistossomose hepatoesplęnica. Estudo Angiográfico. Rev Hosp Clin Fac Med S Paulo 1980; 35:121-31.
  • 11. Raia S, Mies S, Macędo AL. Portal hypertension in schistosomiasis. Clin Gastroent 1985; 14: 57-82.
  • 12. Almeida SCT, Domingues ALC, Coutinho A, Almeida CA, Azevedo RA. Study of the portal and arterial hepatic flow in schistosomotic patients. Congresso Mundial de Gastroenterologia. Săo Paulo 1986; 31: 135.
  • 13. Mies S. Hemodinâmica sistęmica e hepática na esquistossomose mansoni: Açăo do propranalol. Tese de Livre Docęncia - Faculdade de Medicina da Universidade de Săo Paulo 1992.
  • 14. Vezozzo DCP, Cerri GG. Duplex hemodinamic evaluation of hepatosplenic mansoni schistosomiasis. Mem Inst Oswaldo Cruz 1992; 87: 149-51.
  • 15. Lacerda CM, Ramos H, Raia S. Fisiopatologia da hipertensăo porta esquistossomótica e efeitos da esplenectomia com ligadura de varizes de esôfago. Acta Cir Bras 1993; 8: 113-17.
  • 16. Brandt CT, Caneca AOF, Tavares DJS, Ávila JL. Surgical hepato-splenic schistosomiasis mansoni in children: a doppler duplex study of the portal vein and the hepatic artery. Trans R Soc Trop Med Hyg 1995; 89:70-7.
  • 17. Arruda SMB. Doppler fluxometria em esquistossomóticos com e sem hemorragia digestiva. Tese - Mestrado - Faculdade de Medicina da Universidade Federal de Pernambuco 1996.
  • 18. Barreto VST, Domingues ALC. Doença hepática na esquistossomose. In. Coelho J. Aparelho Digestivo: Clínica e Cirurgia. 2 ed. Medsi, Rio de Janeiro 1996; 1071-84.
  • 19. Polak M, Montenegro MR, Meira JA, Conte VP, Espejo H, Franchini F, Pontes JF. Shistosomiais mansoni pathological changes in the liver in different stages of the disease studied by means of laparoscopy and needle biopsy. Rev Inst Med Trop 1959; 1: 18-40.
  • 20. Coelho RB 1971. Lesőes hepáticas secundárias. In: Coelho RB, Anatomia patológica das afecçőes hepáticas. Editora Universitária da UFPE, Recife 1971; 59-77.
  • 21. Ryood JW, Buschmann RJ. Morphometry of liver parenchyma in needle biopsy specimens from patients with alcoholic liver disease: preliminary variable for the diagnosis and prognosis of cirrhosis. Modern Pathology 1989; 2: 382-89.
  • 22. Coutinho EM, Souza MM, Silva LM, Cavalcante CL, Araújo RE, Barbosa JAA, Cheever AW, Andrade ZA. Pathogenesis of schistosomal pipestem fibrosis: a low-protein diet inhibits the development of pipestem fibrosis in mice. Int J Exp Path 1997; 78: 337-42.
  • 23. Domingues ALC. Ultra sonografia na esquistossomose mansônica hepato-esplęnica: Avaliaçăo da intensidade da fibrose periportal e da hipertensăo porta. Tese de Doutorado em Medicina - Centro de Cięncias da Saúde da Universidade Federal de Pernambuco 1998; 99p.
  • 24. Junqueira LCV, Bignola G, Brentani RR. Picrosirius staining plus polarization microscopy: a specific method for collagen detection in tissue sections. Histochemical Journal 1979; 11:447-55.
  • 25. Carvalho EMF, Acioli MD. Pesquisa da avaliaçăo do impacto das açőes IEC/MC no controle das doenças endęmicas em cinco estados brasileiros. Versăo Pernambuco. Relatório de Pesquisa, Recife.Banco Mundial/MS/FNS/NESC/CPqAM 1996.
  • 26. Brandt CT, Domingues ALC, Cordeiro F, Figueredo Silva J, Jucá N, Aguiar JAA, Domingues LAW. Esquistossomose hepatoesplęnica cirúrgica: Histopatologia hepática e endoscopia digestiva alta em crianças comparadas ŕ adultos. Rev Col Bras Cir 2000; 27: 13-8.

  • NOTES

Publication Dates

  • Publication in this collection
    24 Oct 2002
  • Date of issue
    2002
Sociedade Brasileira para o Desenvolvimento da Pesquisa em Cirurgia Al. Rio Claro, 179/141, 01332-010 São Paulo SP Brazil, Tel./Fax: +55 11 3287-8814 - São Paulo - SP - Brazil
E-mail: sgolden@terra.com.br