|
Annals of African Medicine
Annals of African Medicine Society
ISSN: 1596-3519
Vol. 2, Num. 2, 2003, pp. 64-67
|
Annals of African Medicine, Vol. 2, No. 2, 2004, pp. 64-67
RATE OF CO-INFECTION WITH MALARIA PARASITES AND SALMONELLA
TYPHI IN ZARIA, KADUNA STATE, NIGERIA
Florence A. Mbuh, Musa Galadima and *Lucy
Ogbadu
Department of Microbiology, Ahmadu Bello University, Zaria,
Nigeria
*Department of Biological
Sciences, Microbiology Unit, Benue State University, Benue State, Nigeria
Reprint request to: Florence A. Mbuh, Department
of Microbiology, Ahmadu Bello University, Zaria, Nigeria
Code Number: am03014
ABSTRACT
Background: Typhoid and malaria co-infection
is a major public health problem in many developing
countries. Most of the co-infections treated are based on methods of diagnosis
plagued with assumptions which possibly exaggerate the situation. Thus the
aim of this work was to investigate the rate of co-infection with respect to
the
use of Widal test and blood culture methods for diagnosing typhoid fever in
Zaria, Nigeria.
Method: A total of 218 blood
samples were collected from patients with a clinical suspicion of malaria and
typhoid fever and examined
for malaria parasites and S. typhi infection.
Results: Sixty samples were
positive for malaria parasites, 22 of which were positive for typhoid by the
Widal test and only one by the culture method. The rate of co-infection was
significantly high when typhoid was diagnosed by Widal (10.1%) than by blood
culture method (0.5%). A correlation analysis showed no specific relationship
between malaria parasite load and the level of Salmonella antibody titres in
malaria patients (r = 0.05 and 0.08 for somatic and flagella antigens of S.
typhi respectively).
Conclusion: The incidence
of typhoid and malaria co-infection will greatly reduce if the diagnosis of
typhoid fever
in malaria endemic areas such as Zaria is bases
on blood culture.
Key
words: Malaria, typhoid fever,
co-infection
INTRODUCTION
The treatment of malaria and typhoid
co-infection is a common phenomenon in many
parts of Africa.1 Malaria and typhoid remain a treat to many people
in Sub Saharan Africa for several reasons: the increasing poverty, deterioration
in public health services, compounded by HIV / AIDS and
increasing resistance of malaria parasites to chloroquine;2 the lack
of portable water and widespread misuse of the Widal agglutination test for diagnosing
typhoid fever, 1.3.4 increased requests for Widal tests as a means
of making quick money by private laboratories are other
factors. 1.5 Malaria and typhoid fever often present with mimicking
symptoms especially in the early stages of typhoid 1.6. Thus it is
very common to see patients undergoing both typhoid and malaria treatments even
if their diagnosis has not been confirmed. There are more typhoid cases in areas
of drug resistant malaria and a cross reaction between malaria parasites and
salmonella antigens may cause false positive Widal agglutination test.7. A
reliable diagnosis of typhoid is based on culture of blood, stool and
bone marrow.3 Bone marrow aspiration has technical difficulties
and stool cultures are positive in most patients only in the third week of infection.
This leaves blood culture as the best method for diagnosing early Salmonella
typhi infections in the absence of other alternatives. The objective of this
study was to determine the rate of co-infection with malaria parasites and S.
typhi in patients with respect to the use of a single Widal test and blood culture
methods for the diagnosis of typhoid fever in Zaria.
METHOD
Patients directed to the laboratory
for the Widal and malaria parasite tests by the attending physician in four
hospitals in Zaria were sampled for this study. Following an explanation to
and consent of the patient or parent of children, a simple
questionnaire was filled with regards to age, sex
and current medication. A total of 218 blood samples were collected from consecutive
febrile patients between February and April 1996. Patients who had started
treatment were excluded. Blood samples were also collected from 52 apparently
healthy individuals as controls. Five milliliters of blood drawn by venepuncture
from each person were tested for malaria parasites, S. typhi O and H antibodies
and also cultured for S. typhi.
Parasitological
examination
Giemsa-stained thick and thin blood
films were prepared for each sample and parasitaemia was evaluated per microliter
of blood using the thick film preparation according to
standard methods 8, assuming a leukocyte count of 5400 μl-1 of
blood established for healthy Nigerians 9. Films were examined microscopically
for the presence of malaria parasites within red blood cells in thin films. For
thick films, the ring forms, trophozoites and gametocytes were looked for. A
smear was considered negative for malaria parasites if no parasites were seen
after examining at least 100
microscopic fields. The number of parasites μl-1 of blood was expressed
as (parasite count X 5400)/No of leucocytes counted,
which was 100.
Widal
test
The Widal agglutination test was
performed on all blood samples by the rapid slide
titration method10 using commercial antigen suspension (Cal-Test Diagnostic
Inc. Chino, U.S.A.) for the somatic (O) and flagella (H) antigens. The slide
titration test is the prevalent method of performing the Widal test
in Zaria and other parts of Nigeria. 4, .11 A positive
Widal test
was considered for any serum sample with antibody titre ≥1 in 160 to the O antigen
of S. typhi.
Bacteriological
blood culture
Two milliliters of each blood sample
were aseptically introduced into 18 ml of thioglycolate broth (DIFCO) and incubated
at 37°C for an initial period of 48 h and sub-cultured on MacConkey agar (Oxoid).
S. typhi organisms were identified on the basis of standard cultural, microscopic
and
biochemical characterization.12 Inoculated blood culture media was
discarded as negative if there was no growth after 7-10 days. The relationship
between malaria parasite counts μl-1 of blood and Salmonella O and
H antibody titres were determined by carrying out a correlation analysis using
the Microsoft Excel Computer Package.
RESULTS
The results of this study are based
on parasitological examination for malaria parasites and bacteriological and
serological tests for the diagnosis of typhoid fever in 218 patients attending
4 hospitals and 52 apparently healthy
individuals in Zaria. The patients comprised 100 females and 118 males aged
between 2 and 59 years (mean = 21 years) Malaria parasites were found in 60 (27%)
samples (mean parasite load = 1.2×104 parasites
μl-1 of
blood, standard deviation = 2×105 parasites),
henceforth
known as malaria
patients.
Twenty-two of
the 60 malaria patient samples were positive for typhoid by the Widal test
considering a positive Widal test for any sample showing antibody titre of
greater or equal to 1 in 160 against the
somatic (O) antigen of S. typhi.13 Thirty This gave a 10.1% rate
of co-infection with malaria parasites and S. typhi when typhoid was diagnosed
by
the single Widal agglutination test and a 0.5% rate of co-infection with malaria
parasites and S. typhi when blood culture was used.
In the control
group, all blood culture samples were negative for S. typhi. Antibodies to
the somatic antigen were detected in 16 (30.8%) of 52 samples. Twelve (23.1%)
control samples had malaria parasites
(carriers), (mean parasite load = 1.3×103 μl-1 of
blood, standard deviation = 1079). Three (5.8%) carriers of malaria parasites
were positive
for the Widal test.
Table 1 shows
the distribution of somatic and flagella antibodies of S. typhi in malaria
patients and carriers of malaria parasites. There was no significant difference
between the mean antibody titres (anti O titres) in the malaria and typhoid
patients (Z= 0.03 and α=0.05), typhoid and non typhoid patients, malaria
patients and carriers of malaria parasites, where Z values at 5% confidence
level were 0.03, 1.4 and 0.03 respectively. The mean parasite density was
however significantly higher in malaria patients than in carriers of the parasites
(Z
=
4, α = 0.05).
Table 2 shows
results of correlation analysis for different categories of malaria patients
and carriers of malaria parasites. Although 71% of malaria positive samples
had detectable levels of antibodies to the somatic antigen of S. typhi, parasite
load and antibody levels were shown not to be mutually correlated, irrespective
of age and sex of the patients. The correlation coefficient (r) of -0.01
for somatic antigen and 0.05 for flagella antigen shows that the level of Salmonella
antibodies in the malaria patients was not related to the presence of malaria
parasites.
Table 1: Distribution of salmonella
antibodies in malaria patients (n-60) and carriers of malaria parasites (n
=
12)
Antigenes
|
Reciprocal
|
antibody
|
Titre
|
|
|
|
|
|
|
|
|
Patients
|
|
|
|
|
Carriers
|
|
|
|
|
|
≤20
|
40
|
80
|
160
|
≥320
|
≤20
|
40
|
80
|
160
|
≥320
|
O
|
27
|
8
|
3
|
18
|
4
|
6
|
3
|
0
|
2
|
1
|
H
|
4
|
5
|
16
|
19
|
16
|
1
|
2
|
2
|
5
|
2
|
O = Somatic antigen of S.
typhi
H = Flagella antigen
of S. typhi
Table 2: Correlation coefficient for malaria parasite count
and S. typhi O and H antibody levels in malaria patients by sex and age groups
and in carriers of malaria parasites
Category
|
Correlation Coefficient (r)
|
|
|
O antigen
|
H antigen
|
Males (n = 30)
|
0.129
|
0.033
|
Females (n = 30)
|
0.021
|
0.125
|
Age group 1-10 years (n = 6)
|
0.104
|
0.25
|
Age group 11-20 years (n = 6)
|
0. 004
|
0.004
|
Age group 12-30 years (n = 29)
|
0.128
|
0.021
|
Age group > 30 years (n = 8)
|
0.081
|
0.082
|
All patients (= 60)
|
-0.014
|
0.054
|
Carriers of malaria parasite
|
0.009
|
0.075
|
O = Somatic antigen of S.
typhi
H = Flagella antigen
of S. typhi
DISCUSSION
In this study cultural diagnosis
of typhoid fever showed that the rate of co-infection with malaria parasites
was 0.5% against 10% by the Widal test. In addition, a correlation analysis
showed that the presence of malaria parasites had no specific relationship
with S. typhi O and H antibody levels in malaria patients and carriers of malaria
parasites.
In Zaria, it is
common to find patients receiving typhoid malaria treatment simultaneously
since medical practitioners usually rely on a single Widal test result for
the diagnosis of typhoid fever. Moreover clinicians are often compelled by
patients behaviour to prescribe anti-typhoid drugs even when Widal test results
are not suggestive of typhoid. In this study, the 22 malaria patients who
had a positive Widal test result could have been cases of cross reactivity
between S. typhi and malaria parasite antigen14 but 31 other Widal
positive samples were negative for both malaria parasites and S. typhi. Many
patients often take anti-malaria drugs before presenting at the hospital
but would not admit it when questioned. Such patients can only be identified in
a survey through testing for residual malarial drugs in their
blood. These patients who tested negative for both typhoid and malaria but were
treated for typhoid and malaria eventually got well after the treatment. They
could have been cases of drug resistant malaria or patients suffering from self
limiting infections such as transient viraemia.
Although 36.7%
of malaria patient samples were positive for typhoid by the Widal test, blood
culture results suggest that only 1.6% of malaria patients would be infected
with S. typhi. It seems that the outcome of the Widal reaction for patients
with a clinical suspicion of typhoid and malaria depends on individual host
immune responses, which become stimulated in febrile conditions associates
with malaria fever. This memory response could cause positive Widal reactions
in previously sensitized patients. A false positive Widal test reaction occurs
in about 35% of malaria patients6 and a similar result (36.7%)
was recorded in this work. This can be accounted for by the demonstrated
high prevalence
of Salmonella antibodies in the local healthy population and the fact that
50% of the patients had detectable levels
of antibodies to the somatic antigen.
The 6.3 prevalence
of typhoid in Zaria was significantly higher that the earlier report of 3.1%.
This could have been due to the period of the year during which sampling
was done (February-April) when places are dry, water is scarce and cultural
practices
favour high transmission of S. typhi in this environment. The high rate of
typhoid and malaria co-infection associated with the Widal test (10.1%) may
be responsible for the
frequent treatment of mixed infections in Zaria. However, blood cultural results
showed that this
rate of co-infection could be reduced to only 0.5%.
In view of this
significant difference and in order to role out any case of malaria with
mimicking symptoms, or the influence of anamnestic response the practical use
of cultural
methods for the diagnosis of typhoid fever should be emphasized in our clinical
laboratories. This will also improve patient management by cutting down cost
of treatment and eliminate
other risks associated with misuse of antibiotics.
REFERENCES
-
Nsutebu E, Ndumbe P. The
Widal
test for typhoid fever: is it useful? Africa Health. 2001; 23: 5-9.
-
Alnwich D. Meeting the malaria
challenge. Africa Health. 2001; 23:18-19.
-
Edelman R, Levine M.M. Summary
of
an international workshop on typhoid fever. Rev Infect Dis 1986; 8: 329-349.
-
Edet O. Another opinion
on the
Widal test. Africa Health. 2002; 24:2.
-
Usman A. Typhoid fever-
is the
Widal test useful? Africa Health. 2002; 24: 3.
-
Petit P. L. C, Wamola I.
A.
Typhoid fever: a review of its impact and diagnostic problems. East Afr Med
J 1994; 11: 183-188.
-
Frimpong E. H, Feglo P,
Essel-Ahun M, Addy P. A. K Determination of diagnostic widal tetres in Kumasi,
Ghana. West
Afr J Med 2000; 19: 34-38.
-
World Health Organisation.
A
practical handbook on severe and complicated malaria, Geneva, 1991.
-
Akinyanju O, Grossman S.
A. Leucocyte counts in healthy Nigerians. Nigerian Medical Journal 1973;
3: 95-99.
-
Raphael S. S, Spencer F. Immunology and serology.
Lynchs medical laboratory technology. Saunders, Philadelphia.
1983.
-
Mbuh F. A. Reaction of a Widal reagent in
sera from suspected typhoid and non typhoid patients in Zaria. M.Sc.
Thesis, Ahmadu Bello University, Zaria. 1997.
-
Cowan S.T. Cowan and Steels manual for
the identification of Medical bacteria. 1981.
-
Olubuyide T. O Factors that may contribute
to death from typhoid infection among Nigerians. West Afr J Med
1992; 11: 112-115.
-
Alaribe A. A, Ejezie G. C. E, Ezedinachie
E. N. U. The prevalence of salmonella antibody among malaria patients
in Calabar. Nigerian Journal of
Parasitology (abstract) 1995;
Copyright 2003 - Annals of African Medicine
|