TRANSACTIONS OFTHEROYALSOCIETYOFTROPICALMEDICINEANDHYGIENE(1998)92,639~642 639

Ultrasonography of the liver and spleen in Brazilian patients with hepatosplenic

schistosomiasis and cirrhosis

Leila M. M. B. Pereiral, Ana L. C. Domingues2, Victorino Spinelliz and Ian G. McFarlanes*

IDepartment of Internal Medicine, University of Pernambuco, PE, Brazil;2Department of Internal Medicine, University Fed-
eral of Pernambuco, PE, Brazil;JInstitute of Liver Studies, Kings College Hospital, London SE5 9RS, UK

Abstract
Ultrasonography is now widely used in the diagnosis and management of patients with chronic Schisto-
soma mansoni infections. The present study was undertaken to evaluate th.e use of ultrasonography in
patients with hepatosplenic schistosomiasis (HSS) with and without cirrhosis. Ninety-seven patients (52
males; median age 38 years, range 19-68 years) with HSS, 65 with well compensated (HSSC) and 32
with decompensated (HSSD) disease and cirrhosis, were systematically examined by ultrasound. Hepatic
fibrosis was graded according to WHO recommendations. Typical atrophy of the right hepatic lobe ac-
companied by hypertrophy of the left lobe, with a rounded inferior marginal edge, was seen in 86 (88.7%)
patients. Periportal fibrosis was observed in 83 (85.6%) cases and confirmed histologically in all. In 66
patients (68.0%) thickening of rhe gallbladder wall, associated with periportal fibrosis and extending from
the branches of the porta hepatis, was noted. No evidence of biliary disease was found in these patients
and gallstones were present in only 3 cases. Fourteen (43.8%) of the HSSD patients could not be classi-
fied for grade of fibrosis because of the advanced stage of cirrhosis related to hepatitis B or C viral
infection. Of the remaining 18 HSSD patients, none had only grade I fibrosis (vs. 10.8% of HSSC,
fiO.054) and only 6 had grade II (vs. 67.7% of HSSC, PcO.O005), while the frequency of grade III was
significantly higher in the HSSD patients than in those with HSSC (37.5% vs. 21.5%, P=O.O49).These
findings indicate that although ultrasonography is a very valid technique for assessing patients with pure
HSS, and should be considered the ‘gold standard’, it is not reliable for assessing periportal fibrosis in
patients with concomitant cirrhosis due to other causes.

Keywords: schistosomiasis, Schistosoma mam~ni, ultrasonography, cirrhosis, Brazil

Introduction Nevertheless, most of these studies agree that ultra-

Brazil is one of the most important endemic areas of
schistosomiasis in the world. with an estimated 8 mil-
lion people infected with Sckstosoma mansoni, particu-
larly in the north-east region where 60 to 80% of the
population in some areas are affected (COUTINHO &
DOMINGUES. 1993). The diagnosis of chronic S. man-
soni infection’is most reliably auhieved by direct demon-
stration of S. mansoni eggs in faecal samples, together
with clinical features of schistosomiasis (intermittent di-
arrhoea, hepatomegaly and/or splenomegaly, and portal
hypertension) and a history of living in an endemic area.
Occasionally, eggs cannot be found in faeces and a rec-
tal biopsy is required for diagnosis. Infection with S.
mansoni results in lesions in a number of organs, includ-
ing the liver, the portal vein and its tributaries, the
spleen and the gallbladder.
About 10% of patients with hepatointestinal schisto-
somiasis (HIS) develop hepatosplenic schistosomiasis
(HSS) (BINA & PRATA, 1984) and chronic liver disease
(COUTINHO, 1979). The livers of these patients reveal
a characteristic ‘pipe-stem’ pattern of fibrous tissue
deposition in portal tracts with interstitial fibrous septa
(Symmers’ fibrosis) that is distinct from a true cirrhotic
pattern (ANDFUDE, 1965). Demonstration of the path-
ological liver lesions due to S. mansoni can be achieved
by a variety of techniques but assessment of the severity
of periportal fibrosis has usually required histological
examination of wedge liver biopsy specimens. In the
past 12 years, however, state-of-the-art ultrasonography
has contributed greatly to the study of liver fibrosis and
portal hyperten$on, particularly -in patients with S.
mansoni (ABDEL-WAEIAB et al.. 1992: HATZ et al.. 1992:
DOMIN~UES et al., 1993; ~VERA et al., 1996, GER:
SPACHER-LARA et al., 1997), because it allows for the
visualization of pathology that can otherwise be assessed
only by invasive methods. Consequently, it has become
the method of choice for detecting periportal fibrosis in
schisrosomiasis (COUTINHO, 1990). The rapid advanc-
es during the continuing evolution of ultrasonographic
techniques make it difficult to compare directly studies
from different centres performed at different times.
*Author for correspondence.
sound imaging in schistosomiasis is highly sensitive and
correlates well with histological findings (CERRI, 1984;
HOMEIDA etal., ~~~~;ABDEL-WAHAB etal., 1989).A~-
DEL-WAHAB et al. (1989) reported that schistosomal fi-
brosis could be distinguished from cirrhosis but, to our
knowledge, the reliability of ultrasonography for assess-
ing fibrosis in schistosomiasis patients with concomitant
cirrhosis (due to other causes) has not been systemati-
cally investigated in large numbers of patients. The
present study was therefore undertaken to evaluate the
ultrasonographic features in schistosomiasis patients
with and without cirrhosis from a well defined area and
to assess the relationship between the degree of peripor-
tal fibrosis determined ultrasonographically and severity
of liver disease.
Patients and Methods
A total of 97 patients (52 males; median age 38 years,
range 19-68 years) with HSS were recruited from liver
out-patients’ clinics. The laboratory diagnosis of chron-
ic schistosomiasis was based on 3 consecutive stool
samples by the standard qualitative sedimentation
method (HOFFMAN et al., 1934). and the auantitative
method df -TO & MIUG (195ij as modifiid by KATZ
et al. (1972) was used for estimating the intensity of the
infection. Rectal biopsies during rigid proctoscopy us-
ing the OTTOLINA & ATENCIO(1943) technique were
performed in 30 patients who were negative for S. man-
soni eggs in stool samples. Of the 97 patients, 32 had de-
comDensated liver disease (HSSD), with at least 2 of the
following features: a serum albumin less than 30 g/L,
prothrombin time prolonged by more than 4 s above the
normal control value, total bilirubin greater than 25
PmoliL, ascites, or encephalopathy. All of the HSSD
patients had cirrhosis (with or without chronic hepati-
tis) related to concomitant hepatitis B or C virus infec-
tions. The remaining 65 patients did not have any of
these features and were considered to have compensat-
ed liver disease (HSSC).
Percutaneous (using a Menghini biopsy needle) or
surgical (at splenectomy) liver biopsies were performed
in all patients. The presence or absence of periportal fi-
brosis with variable degrees of portal tract expansion
640 LEILAM.M.B.PEREIRA ETAL.

and inflammation and other histological features of

chronic heuatic schistosomiasis, including Schistusoma
egg granuioma formation, deposition o? Schistosomw
pigment, focal nodular proliferation, periductular fibro-
sis, and portal vein obliteration (ANDRADE, 1965), were
noted. Histological features of chronic hepatitis were
noted according to DESMET et ~2. (1994).
Ultrasonography using a portable machine (Aloka
SSD-500, Japan) was performed while the patients were
fasting. The examination was done using the linear and
sector scanners to assess pathology of the liver, spleen
and abdominal vessels in the supine, right oblique and
left lateral positions during expiration and suspended
inspiration. The liver was examined for smoothness of
surface, echogenicity, and abnormalities of the biliary
tract and of the hepatic and portal arterio-venous sys-
tems. Liver span was measured both in the midclavicu-
lar line and the midline, and spleen size was noted.
Grading of hepatic fibrosis was determined according to
the WHO recommendations (WHO, 1991) from the
mean total thickness of 3 peripherally located portal
tracts as follows: grade I=3 to 5 mm; grade II=>5 to 7
mm; grade III=>7 mm.
Informed consent for the above investigations was
obtained from all patients and the study was conducted Fig. 1.Transverse sonogram of the liver of a patient with hepato-
splenic schistosomiasis with compensated liver disease, showing
under the supervision, and in accordance with the rules, an area of increased echogenicity which corresponds to peripor-
of the Ethics Committee of University Federal of Per- tal fibrosis (arrows).
nambuco.
Fisher’s exact test was used for statistical compari-
sons of dichotomous variables between groups of pa-
tients. A P value of CO.05 was considered a significant
difference.

Results
Ultrasound examination revealed abnormalities of
the liver, spleen and/or gallbladder compatible with a
diagnosis of chronic hepatic schistosomiasis in all 97 pa-
tients. Atrophy of the right hepatic lobe accompanied by
hypertrophy of the left lobe, with a rounded inferior
marginal edge, which is very typical of hepatic schisto-
somiasis (SILVA & CARRILHO, 1992), was seen in 86
(88.7%) patients (Table 1). Periportal fibrosis, mani-
Table 1. Ultrasonographic findings in patients
with hepatosplenic schistosomiasis
Fig. 2. Longitudinal sonogram in a patient with hepatosplenic
Number (%) of schistosomiasis with compensated liver disease showing thick-
Feature cases with feature ening of the gallbladder wall (arrows).
Liver Table 2. Portal vessel diameters in the 72 patients
Hypertrophy of left lobe and with splenomegaly
atrouhv of right lobe 86 (88.7%)
Peripbrial fib&is Portal vessels Median cm diameter (range) YZ(%)
Thickened gallbladder wall
Granulomas Portal vein
>1.3 cm 1.8(1.4-2.1) 58(81%)
Spleena cl.3 cm 1.1(0.9-1.3) 14(19%)
Splenomegaly 72(100.0%)
Nodular pattern 3 (3.1%) Splenic vein
>I.0 cm 1.4(1.1-2.0) 49 (68%)
aIn the 72 patients who had not had splenectomies. cl.0 cm 0.9 (0.8-1.0) 23 (32%)
Table 3. Relationships between grades of peripor-
fested sonographically as an area of increased echo- tal fibrosis determined ultrasonographically and
genicity (Fig. l), was observed in 83 (85.6%) patients severity of liver disease assessed on clinical, labo-
(Table 1) and confirmed histologically in all cases. In 66 ratory and histological criteria
patients (68.0%) thickening of the gallbladder wall (Fig.
2), associated with periportal fibrosis and extending Grade of HSS
from the branches of the porta hepatis, was noted. No periportal fibrosis HSSC HSSD P
evidence of biliary disease was found in these patients
and gallstones were present in only 3 cases. Granulo- Fibrosis I 7(10.8%) 0 =0.054
mas, appearing as round nodes of increased echogenic- Fibrosis II 44 (67.7%) 6 (18.8%) <0.0005
ity, were visualized in only 8 (8.2%) patients. Fibrosis III 14 (21.5%) 12(375%) =0.049
Splenectomies for relief of portal hypertension had Unclassified 14 (43.8%) <0.0005
been performed in 25 of the 97 patients. In all of the re- Total of cases 6; 32
maining 72, various grades of splenomegaly were found
(Table 1). In 3 cases (3.1%), the spleen exhibited mul- HSS=hepatosplenic schistosomiasis; HSSGHSS patients
tiple nodular images, which were thought to be charac- with compensated liver disease; HSSD=HSS patients with de-
compensated liver disease.
ULTRASONOGRAPHY IN HEI’ATOSPLENIC SCHISTOSOMIASIS
teristic of either siderotic nodules (Gamma-Gandy
bodies) or (in 1 patient) nodular lymphoma associated
with HSS. However, as a spleen biopsy was not per-
formed in this latter patient this could not be confirmed.
In the 72 patients who did not have splenectomies,
enlargement of the diameter of the portal vein by more
than 1.3 cm (median 1.8 cm, range 1.4-2.1 cm) and
also the splenic vein by more than 1.0 cm (median 1.4
cm, range 1.1-2.0 cm) was found, respectively, in 58/72
(80.6%) and 49/72 (68.1%) patients (Table 2). In all 25
patients who had splenectomies, ultrasound showed
normal portal (median 1.1 cm, range 0.9-l .3 cm) and
splenic vein (median 0.8 cm, range 0.6-1.0 cm) diame-
ters, but portal vein thrombosis was observed in 3.
These were the only patients in the entire series in
whom portal vein thrombosis was seen.
Table 3 shows the findings of schistosomal periportal
fibrosis assessed ultrasonographically in relation to se-
verity of liver disease assessed on clinical, laboratory
and histological criteria. Patients with HSSD tended to
have much more severe fibrosis than those with HSSC.
Thus, there were no HSSD patients with only grade I fi-
brosis (vs. 10.8% of HSSC, eO.054) and only 6 had
grade II (vs. 67.7% of HSSC, P<O.O005), while the fre-
quency of grade III was significantly higher in the
HSSD patients than in those with HSSC (37.5% vs.
21.5%, -0.049). However, 14 (43.8%) of the HSSD
patients could not be classified for grade of fibrosis be-
cause of the advanced stage of cirrhosis related to hepa-
titis B or C viral infection (Fig. 3).
Discussion
The present study demonstrates that ultrasonography
is an important tool for evaluating severity of liver dis-
ease in patients with HSS, particularly because hepatic
Fig. 3,Transverse sonogram ofthe liver in a patient with hepato-
splenic schistosomiasis with decompensated liver disease.Note
features of cirrhosis which obscured the typical periportal fibro-
sis pattern and prevented classification of degree of fibrosis.
AND CIRRHOSIS IN BRAZIL 641
and abdominal vessel imaging allows for the visualiza-
tion of pathology that can otherwise be assessed only by
invasive methods (angiogranhv and/or liver bionsv).
Left lobe hypertrophy &&mpanied by right lobe at&-
nhv (ascribed to increased vascular flow to the left lobe
as a consequence of portal hypertension) was the most
frequently observed nathological change in the livers of
the-present patients. -This is & keeping-with the findings
of other studies based on either clinical (MACKENJEE et
al., 1984; RAVERA et al., 1996; KARDORFF et al., 1996)
or’ histopathological (&D&E, 1965) criteria. Per;
portal fibrosis, which is the most important feature in
HSS, was found to be manifest sonographically as an
area of increased echogenicity extending from the major
portal vein to small intrahepatic branches. The fibrosis
extended to the gallbladder, as denoted by the marked
thickening of the gallbladder wall seen in a high propor-
tion (68%) of our patients. This feature has been re-
ported by others in advanced schistosomiasis (CERRI et
al., 1984; ALI et al., 1990; ABDEL-WAHAB & STRICK-
LAND, 1993; BOISIER et al., 1995). In the present study,
portal vein enlargement was found in 80% of patients
and enlargement of the splenic vein in 68%, in accord-
ance with the findings of other studies (CERRI et al.,
1984; BRANDT et al., 1995; KAFCDORFFet al., 1996).
Splenomegaly due to portal hypertension was found
in all patients who had not previously had splenecto-
mies. Splenic nodules were observed in a few patients
(3%). These might have been Gamma-Gandy corpus-
cles, reflecting a haemosiderin-containing fibrosis. Al-
ternatively such nodules, which may occasionally be
quite large, might correspond to giant follicles or lym-
phomatous nodules, apparently limited to the spleen,
which have been reported in 1% of spleens removed for
HSS (ANDRADE & ABREU, 1971; PAES & MARIGO,
1981) but these were not further investigated in the
present study.
A further advantage of ultrasound screening is that it
is possible to examine the texture of the whole liver, and
thereby to identify lesions which might be missed by liv-
er biopsy. There are, however, some limitations with
this technique. Ultrasonographic definition of the right
lobe was difficult because of the atrophy (although the
hypertrophy of the left lobe made study of the splenic
vein easier) and granulomas (which were found in liver
biopsies from most of the patients) are usually too small
to be visualized sonographically a-rid were seen only in-
freauentlv in this studv. The nresence of cirrhosis. in
33% of our patients, also compiicates the interpretation
of ultrasonographic findings. Distinction between cir-
rhosis per se and schistosomal periportal fibrosis can
sometimes be made. First, in cirrhosis, coarse texture,
increased attenuation, nodular surface and presence of
regeneration nodules are prominent features. These are
not normally seen in pure schistosomiasis (PRATA,
1982). Secondly, widening and thickening of the portal
vessels can be documented on the basis of sonographic
measurements of the diameters of the portal vein and its
tributaries (CALSEN & FILLY, 1976). This is reportedly
found in about 50% of cirrhotic natients (BOLONDI et
al., 1982), especially those wit& portal hypertension
(KANE &KATZ, 1982), but is usually limited to the por-
tal vein stem or occasionally to its main branches. In
contrast, in schistosomiasis increased echogenicity of
the portal vein walls (indicating thickening) is much
more accentuated, sometimes reaching 2 cm (CERRI et
al., 1984), and enlargement of the portal and splenic
veins is seen much more frequently (CERRI et aZ., 1984;
BRANDT et al., 1995; KARDORFF et al., 1996). Howev-
er, in patients with schistosomiasis and cirrhosis it is
very difficult to make this distinction. Indeed, it was im-
possible to classify the grade of fibrosis by ultrasonogra-
bhy in 44% of the present patients with cirrhosis.
Perinortal fibrosis is believed to corresuond to histo-
logically defined Symmers’ fibrosis (ANGRADE, 1965).
HOMEIDA et al. (1988) have used ultrasound to grade
642
Symmers’ fibrosis but there are some reservations about
using the grading system in this way, particularly with
respect to grade I, which represents very tine fibrotic le-
sions that are not normally equated with Symmers’ fi-
brosis. Symmers’ own original description, confirmed
by subsequent pathologists, was of an intense and wide-
spread fibrosis. The simple general denomination of
portal (or better, periportal) fibrosis, classified in 3 dif-
ferent grades as recommended by the WHO (1991),
would seem to be more suitable and would be useful in
both histopathological and ultrasound studies. We
would suggest that under this system the term Sym-
mers’ fibrosis should be reserved for the characteristic
‘pipe-stem’ pattern visualized histologically. However,
it should be noted that the present findings indicate that
ultrasonography is not reliable for grading of periportal
fibrosis in patients with mixed schistosomiasis and cir-
rhosis.
References
Abdel-Wahab, M. F., Esmat, G., Milad, M., Abdel-Razek, S.
& Strickland, G. T. (1989). Characteristic sonographic pat-
tern of schistosomal hepatic fibrosis. American Journal of
Tropical Medicine and Hygiene, 40, 72-76.
Abdel-Wahab, M. F., Esmat, G., Farrag, F., El-Boraey,Y. A. &
Strickland, G.T. (1992). Grading of hepatic schistosomiasis
by the use of ultrasonography. American Journal of Tropical
Medicine and Hygiene, 46,403-408.
Abdel-Wahab, M. F. & Strickland, G. T. (1993). Abdominal
ultrasonoaraohv for assessinn morbid& from schistosomia-
sis. Trans&kas~ of the Royal “Society of ?~ropical Medicine and
Hygiene, 87, 135-137.
Ali, Q. M.. Abdel-Rahim, I. M.. Doehring-Schwerdtfeaer, E.,
Eranke, b., Kardorff, I?., El She&h, M & Ehrich, JYH-. HI
(1990). Ultrasonographic evaluation of gallbladder function
in patients with Schistosoma mansoni infection. Tropical Doc-
tor, 20, 113-115.
Andrade, 2. A. (1965). Hepatic schistosomiasis: morphologi-
cal aspects. In: Progress in Liver Diseases, Popper, H. &
Schaffner, F. (editors). New York: Grune & Stratton, pp.
228-242.
Andrade, Z. A. & Abreu, W. A. (197 1). Follicular lymphoma of
the spleen in patients with hepatosplenic schistosomiasis
mansoni. American Journal of Tropical Medicine and Hygiene,
20,237-242.
Bina, J. C. & Prata, A. (1984). A evolucao natural da esquis-
tossomose mansoni em uma area endemica. In: Aspectos Pe-
culiares da Infeqao por S. mansoni. Salvador: Centro
Editorial e Didatico da UFBA, pp. 13-33.
Boisier, I?, Serieye, J., Ravaoalimalala, V E., Roux, J, & Es-
terre, l? (1995). Ultrasonographical assessment of morbidity
in schistosomiasis mansoni in Madagascar: a community-
based study in a rural population. Transactions of the Royal
Society of Tropical Medicine and Hygiene, 89,208-212.
Bolondi, L., Gandolfi, L., Arienti, V., Caletti, G. C., Corcioni,
E., Gasbarrini, G. & Labo, G. (1982). Ultrasonography in
the diagnosis of portal hypertension: diminished response of
portal vessels to respiration. Radiology, 142, 167-172.
Brandt, C.T., Caneca, 0. A. F.,Tavares, D. J. de S. & Avila, L.,
jr (1995). Surgical hepatosplenic schistosomiasis mansoni in
children: a Doppler Duplex study of the portal vein and the
hepatic artery. Transactions of the Royal Society of Tropical
Medicine and Hygiene, 89,70-7 1.
Calsen, E. N. & Filly, R. A. (1976). Newer ultrasonographic
anatomy in the upper abdomen: I.The portal hepatic venous
anatomy. Journal of Clinical Ultrasound, 4, 85-90.
Cerri, G. G. (1984). A ultra-sonografia no diagnostic0 da for-
ma hepatoesplenica da esquistossomose mansonica. Revistu
de Imagem, 6, 119-134.
Cerri, G. G., Alves, V. A. F. & Magalhaes, A. (1984). Hepato-
splenic schistosomiasis mansoni: ultrasound manifestations.
Radiology 153, 777-780.
Coutinho, A. D. (1979). Fatores relacionados corn o desen-
volvimento das formas clinicas de esquistossomose mansoni.
Revista da Associagao Midica Brasileira, 25, 185-189.
LEILA M. M. B. PERBIRA ETAL.
Coutinho, A. D. (1990). A new dynamic approach to the diag-
nosis of Symmers’ fibrosis in schistosomiasis by ultrasound.
Revista do Institute de Medic&a Tropical de SZo Paula, 32,
73-77.
Coutinho, A. D. & Domingues, A. L. C. (1993). Esquisto-
ssomose mansoni. In: Gastroenterolokz Clinica. III edicao.
Dani, R. & Castro, L. I? (editors). Rio de Janeiro: Guana~
bara, pp. 1697-1728.
Desmet, V. J., Gerber, M., Hoomagle, J. H., Manns, M. &
Scheuer, I? 1. (1994). Classification of chronic henatitis: di-
agnosis,gra&ing and staging. Hepatology, 19, 151311520.
Domingues, A. L. C., Lima, A. R. F., Dias, H. S., Leao, G. C.
& Coutinho, A. (1993). An ultrasonographic study of liver fi-
brosis in patients infected with Schistosoma mansoni in north-
east Brazil. Transactions of the Royal Society of Tropical Medi-
cine and Hygiene, 87, 555-558.
Gerspacher-Lara, R., Pinto-Silva, R. A., Rayes, A. A. M.,
Drummond, S. C. & Lambertucci, J. R. (1997). Ultrasono-
graphy of periportal fibrosis in schistosomiasis mansoni in
Brazil. Transactions of the Roval Societv of Troaical Medicine and
Hygiene, 91,307-369. A d ’ 1
Hatz, C., Jenkins, J. M., Ali, Q. M., Abdel-Wahab, M. F., Cer-
ri. G. G. &Tanner, M. (1992). A review of the literature on
the use of ultrasonograbhy in schistosomiasis with special
reference to its use in field studies-schistosomiasis manso-
ni. Acta Tropica, 51, 15-28.
Hoffman, W. A., Pans, 5. A. & Janer, I. L. (1934).The sedimen-
tation concentration method in-schistosomiasis mansoni.
Puerto Rico Journal of Public Health, 9,283-29 1.
Homeida, M., Abdel-Gadir, A. F., Cheever, A. W., Bennett, J.
L., Arbab, B. M. O., Ibrahium, S. Z., Abdel-Salam, I. M.,
Dafalla, A. A. & Nash, T. E. (1988). Diagnosis of patholog-
ically confirmed Symmers’ periportal fibrosis by ultrasono-
graphy: a prospective blinded study. American Journal of
Tropical Medicine and Hygiene, 38, 8 6-9 1.
Kane, R. A. & Katz, S. G. (1982). The spectrum of sonograph-
ic findings in portal hypertension: a subject review and new
observations. Radiology, 136, 717-723.
Kardorff, R., Stelma, F. F., Vocke, A. K., Yazdanpanah, Y.,
Thomas, A. I<., Mbaye, A., Talla, I., Niang, M., Ehrich, J.
H. H., Doehring, E. & Gryseels, B. (1996). Ultrasonogra-
phy in a Senegalese community recently exposed to Schisto-
soma mansoni infection. American Journal ofTropical Medicine
and Hygiene, 54, 586-590.
Kato, K. & Miura, M. (1954). Comparative examinations.
Japanese Journal of Parasitology, 3, 35.
Katz, N., Chaves, A. & Pellegrino, J. (1972). A simple device
for quantitative stool thick smear technique in schistosomia-
sis mansoni infection in inhabitants from endemic area in
Minas Gerais State. Brasilia Midica, 11, 55-59.
Mackenjee, M. I<. R., Coovadia, H. M. & Chutte, C. H. J.
(1984). Clinical recognition of mild hepatic schistosomiasis
in an endemic area. Transactions of the Royal Society of Tropical
Medicine and Hygiene, 78, 13-l 5.
Ottolina, C. & Atencio, H. M. (1943). Nuevos caminos para el
diagnostic0 clinico precise de la schistosomiasis mansoni.
Revista Policlinica de1Caracas, 12, 348-380.
Paes, R. A. I? & Marigo, C. (1981). Linfoma folicular gigante
e esquistossomose mansoni. Revista do Institute de Medicina
Tropical de Scio Paula, 23,287-292.
Prata, A. (1982). Infection with Schistosoma mansoni. In: Schis-
tosomiasis, Epidemiology, Treatment and Control, Jordan, l? &
Webbe, G. (editors). London: William Heinemann, pp.
105-127.
Ravera, M., Reggiori, A., Cocozza, E., Ciantia, F. & Riccioni,
G. (1996). Clinical and endoscopic aspects of hepatosplenic
schistosomiasis in Uganda. European Journal of Gastroenter-
ology and Hepatology, 8, 693-697.
Silva, L. C. & Carrilho, F. J. (1992). Hepatosplenic schisto-
somiasis: pathophysiology and treatment. Gastroenterology
Clinics ofNorthAmerica, 21 (l), 163-178.
WHO (1991).The control of schistosomiasis: report of aWH0
expert committee. WHO Teaching Representative Service, 728,
l-1 13.
Received 3 August 1998; revised 7 September 1998; accept-
ed for publication 8 September 1998