Am. J. Trop. Med. Hyg., 61(4), 1999, pp. 635–638
Copyright q 1999 by The American Society of Tropical Medicine and Hygiene
RELATIONSHIP BETWEEN CIRCULATING ANTIGEN LEVEL AND
MORBIDITY IN SCHISTOSOMA MANSONI-INFECTED CHILDREN
EVALUATED BY ULTRASONOGRAPHY
MOHSEN M. HASSAN, MOHAMED H. A. HEGAB, SAMY Z. SOLIMAN, OSAMA A. GABER, MOHSEN M. SHALABY,
AND FATEN M. M. KAMEL
Departments of Parasitology, Biochemistry, and Pediatrics, Faculty of Medicine, Zagazig University, Zagazig, Egypt; Department
of Diagnostic Radiology, Faculty of Medicine, Ain-Shams University, Cairo, Egypt
Abstract. Ninety-eight Schistosoma mansoni-infected children from an endemic area in Sharkia Governorate,
Egypt were evaluated by abdominal ultrasonography to determine liver and spleen sizes, grade of periportal fibrosis,
ands splenic vein diameter. Circulating antigen levels were measured using a double sandwich ELISA in which the
sensitivity was 91.8% and specificity was . 99%, with no evidence of cross-reactivity with other parasites. No
significant relationship was observed between antigen level and clinical stages of the disease as assessed by physical
examination (P . 0.05). When ultrasound was used to stage disease, the mean antigen level was significantly higher
among hepatosplenic cases than intestinal cases (P , 0.05). No difference in mean antigen levels were found between
the splenic and hepatic cases. Furthermore, a direct correlation (P , 0.01) was observed between antigen level and
disease severity as monitored by ultrasonography. Antigen level showed a positive correlation with the degree of
periportal fibrosis (P , 0.05). Moreover, a significant increase in the percent of children who were antigen positive
(. 80 ng/ml) was found in those with more severe periportal fibrosis (P , 0.001). The findings suggest that ultrasonography along with measurement of circulating antigen levels predict morbidity in schistosomiasis mansoni.
We have previously developed1 and refined2 an assay to
measure antigenemia in schistosomiasis, a parasitic disease
that affects more than 200 million people worldwide. Our
diagnostic approach used a monoclonal antibody designated
128C3/3/21 (kindly supplied by the late Dr. Mette Strand,
Johns Hopkins University, Baltimore, MD) in an antigencapture ELISA to detect circulating schistosome antigen and
monitor the efficacy of praziquantel therapy in children with
parasitologically proven schistosomiasis mansoni. 1 The
ELISA positivity was directly related to the fecal egg counts
obtained before treatment with praziquantel, but there was
no correlation between antigen levels and the clinical stage
of the disease.
We have now improved our assay format by increasing
the biotinylation of the marker monoclonal antibody 128C3,
which is used for both antigen capture and detection in this
assay. When this format was used to examine sera from
Schistosoma haematobium-infected children, the assay sensitivity was 100% and the specificity was . 99%, with no
evidence of cross-reactivity with other parasites.2 In these S.
haematobium-infected patients, a positive correlation was
observed between the level of circulating antigen and the
egg count (r 5 0.6, P , 0.01). The mean circulating schistosomal antigen level was significantly higher in those with
progressive urinary bladder morbidity as evaluated by ultrasonography (P , 0.001).
Studies have shown that abdominal ultrasonography can
be reliably used to diagnose hepatosplenic schistosomiasis
and assess its severity.3–5 Abdel-Wahab and others6 have reported that ultrasonographic grading accurately reflects the
hemodynamic changes and provides a good estimate of the
clinical status of patients who have periportal fibrosis as a
result of schistosomiasis mansoni. They recommended that
it should replace clinical grading based upon the liver and
spleen size detected by physical examination. We have now
compared the ELISA, egg counts, ultrasonographic evaluation, and physical examinations as indicators of clinical status in schistosomiasis mansoni.
PATIENTS, MATERIALS, AND METHODS
Patients. Ninety-eight children infected with S. mansoni
(77 males and 21 females, 8–16 years of age, mean 6 SD
5 12 6 4 years) were randomly selected from Tal-Haween
village in an area of Sharkia Governorate, Egypt where
schistosomiasis is endemic. All patients had viable S. mansoni eggs in their stool samples.
Twenty age-matched patients from the same geographic
area who had helminth infections other than Schistosoma (7
with Hymenolepis nana, 1 with Enterobius vermicularis, 5
with Fasciola, 4 with Strongyloides stercoralis, and 3 with
Ascaris lumbricoides) were included as an infected control
group. Another 10 age-matched individuals with no evidence
of parasitic infection served as a healthy control group in
this study.
Informed consent was obtained from parents of all children before the study was initiated. The study was reviewed
and approved by the Schistosomiasis Research Project Ethical Review Committee for the protection of Human Subjects
in Medical Research.
Full clinical histories were obtained from all children,
with special emphasis on symptoms attributed to S. mansoni
infection or its complications. Complete physical examinations were performed, with particular attention paid to the
size and tenderness of the liver and spleen. Children with
no organomegaly were grouped as intestinal cases. Other
children with organomegaly were categorized into 3 groups
according to their abdominal examination: splenic for those
with splenomegaly, hepatomegalic for those with hepatomegaly, and hepatosplenomegalic for those with an enlarged
liver and spleen.
Urine and stool examination. Stool examinations using
the modified Kato thick smear method7 were performed on
three consecutive days to detect infection with S. mansoni
or other parasites.
Antigen-capture ELISA. The ELISAs to detect circulating schistosome antigen were carried out as previously de-
635
636
HASSAN AND OTHERS
FIGURE 1. Relationship between the mean circulating antigen
level and clinical stage of the disease during active infection with
Schistosoma mansoni. The ELISA and clinical staging of the diseases were performed as described in the Patients, Materials, and
Methods in 98 S. mansoni-infected children.
scribed.1 Serum samples treated with trichloroacetic (TCA)
were tested in triplicate. Antigen levels were calculated by
comparison with a standard curve using TCA-treated soluble
egg antigen (SEA).
Abdominal ultrasonography. Children were evaluated
by abdominal ultrasonography to determine liver and spleen
size, grade of periportal fibrosis, and splenic vein diameter.
Ultrasonographic grades described by Abdel-Wahab and others3 were used to classify the patients into four groups according to the severity of periportal fibrosis: grade 0: , 3
mm; grade 1: 3–5 mm; grade 2: 5–10 mm; and grade 3: .
10 mm.
Statistical analysis. The Student’s t-test and Pearson’s
correlation coefficient (r) were used and P values , 0.05
were considered significant.
RESULTS
Sensitivity of the ELISA. The ELISA could detect $ 1
ng/ml of SEA. The mean 6 3 SD background optical density
for sera from 10 parasitologically negative children (uninfected controls) was , 80 ng of antigen/ml of serum. This
value was therefore taken as the cut-off value for positivity.
Sera from 90 of the 98 S. mansoni-infected children exceeded the cut-off value, indicating an overall sensitivity of
91.8%.
Specificity of the ELISA. All 10 sera from the uninfected
control group had antigen concentrations , 80 ng/ml. Sera
from the 20 children infected with helminths other than S.
mansoni had antigen levels , 80 ng/ml. Therefore, the test
showed a specificity . 99%, with no evidence of cross-reactivity with other parasites. The positive predictive value
was 100% and the negative predictive value was 84.7%.
Correlation between egg count and ELISA positivity.
The subjects were classified into two groups according to
the number of eggs excreted in the stool: lightly infected (,
100 eggs per gram of feces) and heavily infected ($ 100
eggs per gram of feces). All 53 heavily infected cases
(100%) had antigen in their sera (. 80 ng/ml), whereas only
82.2% (37 of 45) of lightly infected children had antigenemia. The difference between the two groups was significant
FIGURE 2. Relationship between the mean circulating antigen
level and disease stage as determined by abdominal ultrasonographic
examination during active infection with Schistosoma mansoni. The
ELISA and the diseases staging by ultrasonography were performed
as described in the Patients, Materials, and Methods in 98 S. mansoni-infected children.
(P , 0.005). A direct increase in the mean level of circulating antigen was found with increasing egg counts (r 5
0.82, P , 0.001).
Correlation between clinical stage and ELISA positivity. Ninety-one percent of the children who presented with
intestinal manifestations had antigenemia compared with
94% of those with hepatomegaly and 100% of those with
hepatosplenomegaly (P . 0.05). There were no significant
differences in mean antigen levels among persons in the intestinal, splenic, hepatic, and hepatosplenic groups (Figure
1).
Correlation between clinical and ultrasonography
staging. A poor degree of agreement was obtained between
clinically evaluated liver and spleen sizes and the ultrasonographic findings. Among the 67 children categorized clinically as having intestinal infections without complications,
32 were found to be without organomegaly by ultrasonographic examination. The number of children clinically assessed as splenic (5) or hepatosplenic (16) cases was increased by ultrasonographic examination to 17 and 43, respectively. Conversely, the 10 cases with clinical hepatomegaly were reduced to 6 by ultrasonography.
Correlation between staging by ultrasonography and
antigen level. The mean antigen level was significantly
higher among the hepatosplenic cases than the intestinal cases when ultrasound was used for staging (P , 0.05) (Figure
2). No difference in antigen level was found between splenic
and hepatic cases. A significant relationship was observed
between severity of ultrasonographically classified disease
and the level of antigenemia (P , 0.01) (Figure 2).
Correlation between periportal fibrosis and antigen
level. Periportal fibrosis in the infected children was categorized by ultrasonography into four groups.3 Eight children
had grade 0 fibrosis; 61 had grade 1; and 29 had grade 2.
No grade 3 fibrosis was observed in this group. A significant
increase in the antigen level was found with increasing severity of periportal fibrosis (P , 0.05) (Table 1). Antigen
positivity correlated with the severity of periportal fibrosis
(P , 0.001).
637
CIRCULATING ANTIGENS AND SCHISTOSOMAL MORBIDITY
TABLE 1
Relationship between mean circulating antigen level and periportal fibrosis in 98 Schistosoma mansoni–infected children
Circulating antigen level (ng/ml)
Extent of
periportal fibrosis
Mean 6 SE
Range
Grade 0 (n 5 8)
Grade 1 (n 5 61)
Grade 2 (n 5 29)
62 6 16
716 6 65
1,960 6 101
0–100
40–2,200
760–3,240
DISCUSSION
In contrast to the clear relationship observed between egg
count and antigen level, no difference in the levels of antigenemia among children with different clinical stages of schistosomiasis mansoni was observed in this study. This lack of correlation was also observed in our previous study1 and has been
reported by Fu and Carter8 for S. japonicum infection. This
suggests that the clinical stage as traditionally assessed may not
a good indicator of the morbidity produced by the disease. In
the present study, a poor correlation between clinically evaluated liver and spleen sizes and ultrasonographic findings was
also observed. Of the 67 children clinically categorized as having intestinal infections without complications, 35 had organomegaly detected by ultrasonographic examination. This inconsistency may be reflect enlargement of the organ that is below
the detectability of physical examination and/or the experience
of the examining physician.
The subjects were classified into two groups according to
the number of eggs excreted in the stool. All 53 heavily
infected cases had antigen in their sera whereas only 82.2%
of lightly infected children had antigenemia. The level of
circulating antigen correlated with egg counts. When we had
the same monoclonal antibody on a panel of sera from individuals with parasitologically proven schistosomiasis mansoni,1 an overall sensitivity of 78% was obtained. Sensitivity
appeared to be related to the number of eggs excreted in the
stool of these patients since all of those excreting $ 100
eggs per gram of feces were antigenemic compared with
72% of those excreting , 100 eggs per gram of feces. Thus,
the results of our previous and current report are consistent
with those of other S. mansoni-related studies9–12 in that they
demonstrate a positive relationship between egg excretion
and circulating antigen. De Jonge and others9,10 made similar
observations for the presence of circulating anodic antigen.
Abdominal ultrasonography can be used to diagnose hepatosplenic schistosomiasis and assess the severity of the
disease.3–5 In the present study, disease staging by ultrasound
showed a higher mean antigen level in children with hepatosplenic disease than in those with intestinal disease (P ,
0.05). Furthermore, a correlation was observed between increasing severity of ultrasonographically classified disease
and antigenemia (P , 0.01). A similar observation was
made in our previous study of S. haematobium-infected patients,2 in which the antigen level was significantly higher in
those with a higher degree of urinary bladder morbidity than
in those with less severe bladder morbidity.
Abdel-Wahab and others6 have reported that ultrasonographic grading reflects the hemodynamic changes and provides a good estimate of the clinical status of patients who
have periportal fibrosis as a result of schistosomiasis mansoni. They recommended that ultrasonography should re-
Patients with positive antigenemia
P
,0.05
No.
3
58
29
%
37.5
95.1
100.0
P
,0.0001
place grading based upon liver and spleen size detected by
physical examination. In the present study, we also observed
a significant increase in the antigen level with increasing
periportal fibrosis (P , 0.05) and a significant relationship
between antigen positivity and progression of periportal fibrosis (P , 0.001).
Among 1,106 of four villages in Mali (where 16% of the
positive schistosomiasis cases were represented by heavy infection, i.e., . 400 eggs per gram of feces), Kardorff and
others13 reported a lack of ultrasonographic evidence for severe hepatosplenic morbidity since no case of grade 3 severity was found. This observation may be attributed to the
fact that schistosomiasis mansoni did not progress to severe
hepatosplenic disease in the examined villages. These investigators also reported a lack of correlation between periportal
fibrosis and S. mansoni egg counts.
In the present study, we found that the antigen level measured by the ELISA is directly correlated with disease stage
as defined by the abdominal ultrasonography. This antigen
level also correlated with the grade of periportal fibrosis.
These data suggest that antigen level is a good indicator of
clinical status. We are now designing studies to compare the
effectiveness of circulating antigen levels, abdominal ultrasound findings, and egg counts to determine the most effective predictor of morbidity and disease status.
Financial support: This work was supported by a Schistosomiasis
Research Project agreement 05/02/53 to Mohsen M. Hassan; this
agreement was funded by the United Agency for International Development and the Egyptian Ministry of Health.
Authors’ addresses: Mohsen M. Hassan, Mohamed H. A. Hegab and
Samy Z. Soliman, Parasitology Department, Faculty of Medicine,
Zagazig University, Zagazig, Egypt. Osama A. Gaber, Biochemistry
Department, Faculty of Medicine, Zagazig University, Zagazig,
Egypt. Mohsen M. Shalaby, Pediatrics Department, Benha Faculty
of Medicine, Zagazig University, Zagazig, Egypt. Faten M. M. Kamel, Diagnostic Radiology Department, Faculty of Medicine, AinShams University, Cairo, Egypt.
REFERENCES
1. Hassan MM, Badawi MA, Strand M, 1992. Circulating schistosomal antigens in diagnosis and assessment of cure in children infected with Schistosoma mansoni. Am J Trop Med Hyg
46: 737–744.
2. Hassan MM, Medhat A, Makhlouf MM, Shata T, Nafeh MA,
Osman OA, Fouad YM, 1998. Detection of circulating antigen in-patients with active Schistosoma haematobium infection. Am J Trop Med Hyg 58: 295–301.
3. Abdel-Wahab MF, Esmat G, Milad M, Abdel-Razek S, Strickland GT, 1989. Characteristic sonographic pattern of schistosomal hepatic fibrosis. Am J Trop Med Hyg 40: 72–76.
4. Doehring-Schwerdtfeger E, Mohammed-Ali G, Abdel-Rahim
IM, Kardorff R, Franke D, Kaiser C, El-Sheikh M, Ehrich JH,
1989. Sonomorphological abnormalities in Sudanese children
638
5.
6.
7.
8.
9.
10.
HASSAN AND OTHERS
with Schistosoma mansoni infection: a proposed staging system for field diagnosis of periportal fibrosis Am J Trop Med
Hyg 41: 63–69.
Babello AT, Da Silva RAP, Rocha RS, Katz N, 1994. Abdominal
ultrasonography in acute clinical schistosomiasis mansoni. Am
J Trop Med Hyg 50: 748–752.
Abdel-Wahab MF, Esmat G, Farrag A, El-Boraey YA, Strickland GT, 1992. Grading of hepatic schistosomiasis by the use
of ultrasonography. Am J Trop Med Hyg 46: 403–408.
Martin LK, Beaver PC, 1968. Evaluation of Kato thick-smear
technique for quantitative diagnosis of helminth infection. Am
J Trop Med Hyg 17: 382–388.
Fu C, Carter CE, 1990. Detection of a circulating antigen in
human schistosomiasis japonica using a monoclonal antibody.
Am J Trop Med Hyg 42: 347–351.
de Jonge N, Gryseels B, Hilbert GW, Deelder AM, 1988. Detection of circulating anodic antigen by ELISA for sero-epidemiology of schistosomiasis mansoni. Trans R Soc Trop
Med Hyg 82: 591–594.
de Jonge N, Fillie YE, Hilberath GW, Krijger FW, Lengeler C,
De Savigny DH, van Vliet NG, Deelder AM, 1989. Presence
of schistosome circulating anodic antigen (CAA) in urine of
patients with Schistosoma mansoni or S. haematobium infections. Am J Trop Med Hyg 41: 563–569.
11. van Lieshout L, de Jonge N, Bassily S, Mansour MM, Deelder
AM, 1991. Assessment of cure in schistosomiasis patients after chemotherapy with praziquantel by quantitation of circulating anodic antigen (CAA) in urine. Am J Trop Med Hyg
44: 323–328.
12. Polman K, Stelma FF, Gryseels B, Van Dam GJ, Talla I, Niang
M, van Lieshout L, Deelder AM, 1995. Epidemiological application of circulating antigen detection in a recent Schistosoma mansoni focus in northern Senegal. Am J Trop Med Hyg
43: 152–157.
13. Kardorff R, Traore M, Diarra A, Sacko M, Maiga M, Franki D,
Vester U, Hansen U, Traori HA, Fongoro S, Gorgen H, Korte
R, Gryseels B, Doehring-Schwerdtfeger E, Ehrich JHH, 1994.
Lack of ultrasonographic evidence for severe hepatosplenic
morbidity in Schistosoma mansoni in Mali. Am J Trop Med
Hyg 51: 190–197.