crossmark

Development of a Simple, Peripheral-Blood-Based Lateral-Flow

Dipstick Assay for Accurate Detection of Patients with Enteric Fever
Iqbal Hassan Khan,a M. Abu Sayeed,b Nishat Sultana,a Kamrul Islam,b Jakia Amin,a M. Omar Faruk,b Umama Khan,a
Farhana Khanam,b Edward T. Ryan,c,d,e Firdausi Qadrib
Incepta Pharmaceuticals Ltd., Savar, Dhaka, Bangladesha; International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladeshb; Division of Infectious
Diseases, Massachusetts General Hospital, Boston, Massachusetts, USAc; Department of Medicine, Harvard Medical School, Boston, Massachusetts, USAd; Department of
Immunology and Infectious Diseases, Harvard School of Public Health, Boston, Massachusetts, USAe

Enteric fever is a systemic infection caused by typhoidal strains of Salmonella enterica and is a significant cause of mortality and
morbidity in many parts of the world, especially in resource-limited areas. Unfortunately, currently available diagnostic tests for
enteric fever lack sensitivity and/or specificity. No true clinically practical gold standard for diagnosing patients with enteric fe-
ver exists. Unfortunately, microbiologic culturing of blood is only 30 to 70% sensitive although 100% specific. Here, we report
the development of a lateral-flow immunochromatographic dipstick assay based on the detection of Salmonella enterica serovar
Typhi (S. Typhi) lipopolysaccharide (LPS)-specific IgG in lymphocyte culture secretion. We tested the assay using samples from
142 clinically suspected enteric fever patients, 28 healthy individuals residing in a zone where enteric fever is endemic, and 35
patients with other febrile illnesses. In our analysis, the dipstick detected all blood culture-confirmed S. Typhi cases (48/48) and
5 of 6 Salmonella enterica serovar Paratyphi A blood cultured-confirmed cases. The test was negative in all 35 individuals febrile
with other illnesses and all 28 healthy controls from the zone of endemicity. The test was positive in 19 of 88 individuals with
suspected enteric fever but with negative blood cultures. Thus, the dipstick had a sensitivity of 98% compared to blood culture
results and a specificity that ranged from 78 to 100% (95% confidence interval [CI], 70 to 100%), depending on the definition of a
true negative. These results suggest that this dipstick assay can be very useful for the detection of enteric fever patients especially
in regions of endemicity.

Downloaded from https://journals.asm.org/journal/cvi on 16 March 2024 by 2401:4900:633a:8f74:a260:acbe:dfa8:e71b.

E nteric fever can be due to typhoid and paratyphoid fever and is
caused by infection with Salmonella enterica serovar Typhi (S.
Typhi) or Salmonella enterica serovar Paratyphi (S. Paratyphi) (1,
2). Approximately 22 million cases of typhoid fever and 6 million
cases of paratyphoid fever occur annually, resulting in over
100,000 deaths globally each year (3–5). The occurrence of enteric
fever is mainly associated with the lack of proper sanitation and
fecal contamination of water and food (2, 4). Rapid accurate di-
agnosis followed by early treatment with suitable antibiotics can
reduce the rates of morbidity and mortality due to enteric fever. As
the clinical features of enteric fever are nonspecific and overlap
those of other bacterial and viral febrile illnesses, rapid and accu-
rate diagnosis remains a challenge, particularly in resource-poor
settings (6–8).
Although microbiologic culturing of bone marrow culture is
considered a gold standard for diagnosing individuals with enteric
fever (4), such culturing is clinically impractical due to its invasive
nature (8–10). Therefore, microbiologic culturing of peripheral
blood is often used as an alternative (11). Unfortunately, blood
culture shows poor sensitivity, ranging from 30% to 70% depend-
ing on different factors, including blood volume and prior antibi-
otic treatment (1, 11–13); it may take 2 to 7 days to confirm the
diagnosis and requires a well-equipped laboratory and expertise
for microbiologic confirmation (11). The Widal test is the most
widely used serological test for diagnosing individuals with enteric
fever, but it lacks specificity, especially in areas where enteric fever
is endemic (1, 2, 4, 10, 11). Additional serologic tests include
Typhidot (Reszon Diagnostics, Malaysia) that detects IgM and
IgG antibodies in peripheral blood to a 50-kDa outer membrane
protein of S. Typhi and the Tubex assay (IDL Biotech AB, Sweden)
that detects IgM responses in blood to S. Typhi O9 lipopolysac-
charide (LPS). These assays have been associated with sensitivities
and specificities of 56 to 95% and with specificities of 31 to 97%
in field tests (1, 8, 14–16). Molecularly based methods, includ-
ing nucleic acid amplification tests, have been hampered in
field tests by the low organism load and presence of inhibitors
in peripheral blood, reagent and equipment expense, and lack
of technical expertise although such assays may have higher
sensitivity than blood culture (17, 18). We have previously re-
ported development of an enzyme-linked immunosorbent as-
say (ELISA)-based immunodiagnostic assay, the TPTest, for
diagnosing individuals with typhoid and paratyphoid fever.
This ELISA-based platform has a sensitivity of 100% compared
to blood culture results and a specificity of 78 to 97% (95%
confidence interval [CI], 73 to 100%), depending on the defi-
nition of a true negative (1, 10, 19).
The TPTest ELISA detects S. Typhi membrane preparation
(MP)-specific IgA responses in lymphocyte secretion prepared by
Received 6 December 2015 Returned for modification 8 January 2016
Accepted 29 February 2016
Accepted manuscript posted online 9 March 2016
Citation Khan IH, Sayeed MA, Sultana N, Islam K, Amin J, Faruk MO, Khan U,
Khanam F, Ryan ET, Qadri F. 2016. Development of a simple, peripheral-blood-
based lateral-flow dipstick assay for accurate detection of patients with enteric
fever. Clin Vaccine Immunol 23:403–409. doi:10.1128/CVI.00690-15.
Editor: C. J. Papasian
Address correspondence to Firdausi Qadri, fqadri@icddrb.org.
I.H.K. and M.A.S. contributed equally to this article.
E.T.R. and F.Q. are co-senior authors.
Copyright © 2016, American Society for Microbiology. All Rights Reserved.

May 2016 Volume 23 Number 5 Clinical and Vaccine Immunology cvi.asm.org 403
Khan et al.

TABLE 1 Characteristics of study participants culturing positive bottles on MacConkey agar, blood agar, and chocolate
Value for the subject group agar plates and identifying colonies using standard biochemical tests and
reaction with Salmonella-specific antiserum (1). Blood culture was carried
Suspected Other febrile Healthy out in patients with other illnesses except those with tuberculosis or kala-
enteric fever illness controls azar.
Parameter (n ⫽ 142) (n ⫽ 35) (n ⫽ 28) Preparation of the coating antigen for strips. We used membrane
Median age (25th and 75th 7 (3, 11) 25 (23, 30) 25 (25, 28) preparation (MP) and lipopolysaccharide (LPS) as coating antigens pre-
percentiles [yr]) pared from the Ty21a vaccine strain and S. Typhi wild-type strain (ST-
004), respectively, for making immunochromatographic strips. MP anti-
Gender gen was prepared according to a previously described method (1). The
No. of male subjects (%) 71 (50) 26 (74) 18 (64) bacterial strain was cultivated on horse blood agar plates, and bacteria
No. of female subjects (%) 71 (50) 9 (26) 10 (36) were harvested in buffer (5 mM MgCl2, 10 mM Tris, pH 8.0). The bacte-
rial suspension was sonicated five times at 60% amplitude and centrifuged
at 1,400 ⫻ g for 10 min. The supernatant was then transferred to fresh
tubes and centrifuged at 14,900 ⫻ g for 30 min. The pellet was dissolved in
isolating peripheral mononuclear cells (PBMCs) separated using harvest buffer, and the protein content was determined by a Bio-Rad
Ficoll Isopaque density gradient centrifugation (1, 10). We have protein assay. LPS antigen was prepared from a wild-type clinical isolate of
previously reported pilot analysis of simplified methods for cell S. Typhi isolated from a patient, using a phenol-water extraction proce-
separation, cell incubation, and dot blot analysis for the detection dure, followed by enzyme treatment with proteinase K, DNase, and RNase
of the MP-IgA response (1). We therefore undertook to further and ultracentrifugation as previously described (10).
develop an accurate and simple assay for diagnosing individuals Gold preparation and conjugation. To prepare colloidal gold, we
mixed 0.01% HAuCl4 (Sigma-Aldrich) with 0.024% sodium citrate (Sig-
with enteric fever using an immunochromatographic dipstick as-
ma-Aldrich) in water for injection (WFI) and boiled the solution until it
say to detect specific antibody responses in specimens prepared by became the color of red wine. The colloidal gold was then filtered through
erythrocyte lysis and room air (no supplemental CO2) incubation a 0.2-␮m-pore-size filter. We adjusted the pH of the gold solution to 8.0
at 37°C. (optimum pH for conjugation) and tested a range of goat anti-human IgG
(Jackson ImmunoResearch) and goat anti-human IgA (Jackson Immuno-
MATERIALS AND METHODS Research) to conjugate 1 ml of colloidal gold, eventually choosing 12 ␮g
Ethics statement. The study and all sample collections and analyses were and 16 ␮g, respectively, based on the data below. We blocked the conju-
approved by the research review and the ethical review committees of gated antibody-gold using 20% bovine serum albumin (BSA) and then
the International Centre for Diarrhoeal Disease Research, Bangladesh centrifuged at 10,000 rpm for 45 min at 4°C. We discarded the superna-

Downloaded from https://journals.asm.org/journal/cvi on 16 March 2024 by 2401:4900:633a:8f74:a260:acbe:dfa8:e71b.

(ICDDR,B), and the Institutional Review Board of the Massachusetts
General Hospital. Written informed consent was obtained from all
adult participants 18 to 59 years of age and from parents or guardians
of children 1 to 17 years of age. Assent was taken from children of 11 to
17 years of age.
Study participants and specimen collection. We enrolled partici-
pants presenting to the ICDDR,B with clinically suspected enteric fever
(n ⫽ 142), defined as a systemic febrile illness of ⱖ38°C for 3 to 7 days’
duration without another obvious source. The median age of the enrolled
patients was 7 years (25th and 75th percentiles, 3 and 11 years, respec-
tively). We also enrolled 35 study participants presenting to the ICDDR,B
with a febrile illness confirmed not to be enteric fever and 28 adult healthy
controls (median age, 25 years; 25th and 75th percentiles, 25 and 28 years,
respectively) residing in Dhaka (Table 1 and Table 2). We collected a
sample of venous blood from study participants.
Diagnosis of enteric fever by blood culture. Using a 3- to 5-ml sample
of peripheral blood, we performed microbiological cultures for all sus-
pected enteric fever patients using a BacT/Alert automated system, sub-
tant and resuspended the pellet in 0.02 M Tris buffer containing 1% BSA.
This solution was passed through a 0.2-␮m-pore-size filter and used as the
detection conjugate.
Aggregation testing. To assess stability of the conjugate and to define
the optimum pH and minimum concentration of antibody required for
conjugating the colloidal gold, we used an aggregation assay (20–22).
Specifically, we added 10% NaCl to the gold-protein suspension, incu-
bated it for 10 min, and then assessed stability and polydispersity by mea-
suring the absorbance at 520 nm, 580 nm, and 600 nm (20–22).
Preparation of the lateral-flow strip. We used a backing card contain-
ing a nitrocellulose membrane on which antigen and antibody were dis-
pensed to create test control lines by using a rapid test dispenser
(HM3030; Kin Biotech Co., China). The dispensed membrane was dried
for 90 min. A conjugate pad was then made by soaking glass fibers (Kin
Biotech Co., China) in the gold conjugate solution and drying the pad for
2 h. The pad was then pasted on the backing card in a way that overlapped
the nitrocellulose membrane (High Flow Plus 120 Membrane Card; Mil-
lipore). A glass fiber sample pad was placed at the bottom of the backing

TABLE 2 Characteristics of study participants with suspected enteric fever

Value by test results
Positive blood culture Positive blood culture Negative blood culture Negative blood culture
Parameter and dipstick and negative dipstick and dipstick and positive dipstick
Sample size (no.) 53 1 69 19
Median age (25th and 75th percentiles [yr]) 6 (3, 12) 45 7 (3, 12) 7 (2, 11)

Gender
No. of male subjects (%) 18 (34) 0 37 (54) 12 (63)
No. of female subjects (%) 35 (66) 1 32 (46) 7 (37)

Duration of fever (25th and 75th percentiles [days]) 5 (5, 6) 5 4 (3, 5) 5 (4, 6)
No. of subjects with prior use of antibiotics (%) 9 (17) 0 5 (7) 5 (26)

404 cvi.asm.org Clinical and Vaccine Immunology May 2016 Volume 23 Number 5

card to overlap the conjugate pad to facilitate the flow of sample from a
specimen vial to the strip. To accelerate migration of the samples through
the strip, we used cellulose fiber as an absorbent pad pasted on the backing
card opposite the conjugate pad. All pads were cut to make the desired
strip shape by using a guillotine cutter (CT300 and ZQ2000; Kin Biotech
Co., China).
Separation of peripheral blood cells by erythrocyte lysis. To recover
peripheral lymphocytes, we lysed erythrocytes present in 1 to 2 ml of
venous blood using lysis buffer (0.15 M ammonium chloride, 1 mM po-
tassium bicarbonate, and 0.01 mM disodium EDTA) at a 1:5 dilution and
mixed the sample by gently inverting the tube (BD Falcon) three to five
times. The tube was then kept at room temperature for 10 min and cen-
trifuged at 953 ⫻ g for 5 min at 20°C. We decanted the supernatant and
resuspended the pellet in 150 ␮l of RPMI 1640 medium (Gibco) supple-
mented with 10% heat-inactivated fetal bovine serum (HyClone), 1%
penicillin-streptomycin (Gibco), 1% sodium pyruvate (Gibco), and 1%
L-glutamine (Gibco). We cultured the suspended cells in culture vials
(North China Pharmaceuticals Co. Ltd., China) without any antigenic
stimulation at 37°C without 5% CO2 for 48 h. We then harvested the
culture suspension and centrifuged it at 11,600 ⫻ g at 20°C for 5 min to
collect the supernatant.
Testing the strip. The strip contained two lines on the nitrocellulose
membrane: one was the test line containing MP or LPS antigen, and the
other was the control line containing rabbit anti-goat IgG (Jackson
ImmunoResearch). The conjugate pad contained goat anti-human IgG or
goat anti-human IgA conjugated to colloidal gold. We diluted 75 ␮l of the
lymphocyte culture supernatant with 0.02 M Tris–1% BSA–3% Tween at
a 1:1 dilution in a microcentrifuge tube and dipped the strip into the tube
for 15 min. The test line and/or control line would appear as a red line. The
presence of both the control line and the test line indicated that the sample
was positive for the test undertaken. The presence of only the control line
but no test line indicated a negative result for the test.
Detection of S. Typhi LPS-specific antibodies in specimens by
ELISA. We compared strip results to those obtained using an anti-S. Ty-
phi LPS ELISA format, as previously described (10). Briefly, we coated
microtiter plates (Nunc F; Nunc) with 100 ␮l of LPS (2.5 ␮g/ml) and
blocked the samples with 1% BSA in phosphate (PBS). To detect antigen-
specific responses, we added 100 ␮l of lymphocyte culture supernatant
(1:2 dilution in 0.1% BSA-PBS-Tween) to coated plates, which were then
incubated for 90 min at 37°C. After plates were washed with 0.05% PBS-
Tween, we added rabbit anti-human IgG (1:1,000 dilution; Jackson Im-
munoResearch) conjugated to horseradish peroxidase and incubated the
plates for 90 min at 37°C. We then developed the plates with ortho-phen-
ylenediamine (Sigma) in 0.1 M sodium citrate buffer and 0.1% hydrogen
peroxide after the plates were washed with 0.05% PBS-Tween. We then
read the plates kinetically at 450 nm for 5 min at 19-s intervals. The
maximal rate of optical density (OD) change was expressed as milliabsor-
bance per minute. For normalization of the data, the results were divided
by readings of in-house pooled convalescent-phase standard sera of
blood culture-confirmed typhoid patients and multiplied by 100. Re-
sults were expressed as ELISA units (EU).
Statistical analysis. We used GraphPad Prism, version 4, and Open-
Epi, version 3, for data management, analysis, and graphical presentation.
RESULTS
Characterization of colloidal gold. After making colloidal gold,
we determined the size of the gold nanoparticles by differential
light scatterings using a Zetasizer Nano ZS90 instrument
(Malvern Instrument, Ltd.). The measurements were carried out
at 25°C with a count rate of 193.7 kcps at a scattering angle of 173°.
The average diameter of the prepared gold nanoparticles was 20
nm, as determined from the dynamic light scattering (DLS) spec-
trum (Fig. 1).
Determination of optimum pH and minimum concentra-
tion of detection antibody for conjugation. We performed ag-
May 2016 Volume 23 Number 5 Clinical and Vaccine Immunology
Lateral-Flow Assay for Enteric Fever Detection
FIG 1 Representative DLS spectrum of gold nanoparticles showing an average
diameter (d) of 20 nm.
gregation testing at different pHs and with increasing amounts
of goat anti-human IgG or goat anti-human IgA to produce
conjugated gold. After the addition of NaCl, we measured the
optical density (OD) at 520 nm, 580 nm, and 600 nm. We used
the ratio of the OD at 520 nm to that at 580 nm to assess
stability and the ratio of the OD at 600 nm to that at 520 nm to
assess polydispersity (20–22). We found the highest stability
and lowest polydispersity when colloidal gold was conjugated
to both anti-human IgG and IgA at pH 8.0 (Fig. 2). We also
found that a minimum of 12 ␮g of goat anti-human IgG or 16
Downloaded from https://journals.asm.org/journal/cvi on 16 March 2024 by 2401:4900:633a:8f74:a260:acbe:dfa8:e71b.
␮g of goat anti-human IgA was required to stabilize 1 ml of
colloidal gold solution (Fig. 3).
Results of blood culture. We enrolled total 142 suspected en-
teric fever patients; among them, 54 patients were blood culture-
confirmed enteric fever patients. Among 54 bacteremic patients,
48 were S. Typhi-positive and 6 were S. Paratyphi A-positive pa-
tients.
Results of the strip test. We tested the immunochromato-
graphic strips detecting S. Typhi LPS-specific or MP-specific IgA
or IgG in lymphocyte culture supernatant of subsets of our co-
horts. The strip detecting S. Typhi LPS-specific IgG in lymphocyte
culture supernatant was positive in all (n ⫽ 48) S. Typhi bactere-
mic patients and in 5 of 6 S. Paratyphi A bacteremic patients.
We also tested the LPS-specific IgG assay using lymphocyte
culture supernatant prepared from venous blood of 88 blood
culture-negative patients (clinically suspected to have enteric
fever), as well as from 28 healthy controls and 35 patients with
other febrile illness (tuberculosis, kala-azar, dengue, and non-
Salmonella bacteremia). We found that the dipstick was positive
for 19 blood culture-negative patients and negative for all
healthy controls as well as for all the patients with other febrile
illness (Fig. 4 and Table 3). The strip that detected S. Typhi
MP-specific IgA responses showed variable intensities of test
and control lines in culture-confirmed patients, and the strip
detecting MP-specific IgG responses, although detecting all
blood culture-confirmed patients, was also positive in patients
with other febrile illnesses.
Assessing LPS-specific antibody responses in lymphocyte
culture supernatant using ELISA. We next assessed IgG antibody
responses using an ELISA format and lymphocyte culture super-
natant collected from our different categories of patients. Among
cvi.asm.org 405
Khan et al.

FIG 2 Optimization of pH for anti-human IgG conjugation (A) and anti-human IgA conjugation (B). Optical density (OD) ratios of values at 520 nm to 580 nm
and at 600 nm to 520 nm represent stability and polydispersity, respectively.

them, 22 patients were positive by blood culture and IgG LPS-
specific strip test, 9 patients were negative by blood culture and
positive by strip test, and 35 patients were negative for both blood
culture and strip test. We also measured LPS-IgG responses in 16
healthy individuals and 16 other febrile-illness patients. The pa-
tients who had LPS-IgG responses of ⱖ16 EU were positive by the
strip (Fig. 5).
Determination of sensitivity, specificity, PPV, and negative
predictive value for strip test detecting S. Typhi LPS-IgG. We
calculated the sensitivity, specificity, positive predictive value
reliable, and affordable diagnostic test that shows high sensitivity
and specificity. In our study, we have reported development of
such a test. The test can be performed using 1 ml of venous blood,
erythrocyte lysis buffer, a tabletop centrifuge, and a 37°C incuba-
tor without CO2. Responses are detected visually with a simple
dipstick. Although such a test cannot be used at the bedside, re-
sults with excellent precision and reliability and minimal labora-
tory capacity are available at 48 h. Our previous data suggest that a
reading at 24 h may also be informative (1).

Downloaded from https://journals.asm.org/journal/cvi on 16 March 2024 by 2401:4900:633a:8f74:a260:acbe:dfa8:e71b.

(PPV), and negative predictive value (NPV) of the IgG LPS-spe-
cific lateral-flow dipstick (LFD) using OpenEpi, version 3, an open
source calculator for the evaluation of diagnostic tests. The test
had a sensitivity of 98% compared to blood culture results and
specificity, PPV, and NPV that ranged from 78 to 100% (95% CI,
68 to 100%), 73 to 100% (95% CI, 62 to 100%), and 98 to 99%
(95% CI, 91 to 99), respectively, depending on the definition of a
true negative (Table 4).
DISCUSSION
Enteric fever remains an important public health concern in many
developing countries. There is a very real need for a low-tech,
The current immunochromatographic assay and the previ-
ously reported ELISA-based TPTest are both based on detection
of antibodies secreted ex vivo by activated lymphocytes recovered
from the peripheral circulation during acute infection (1, 10, 19).
These lymphocytes have been stimulated by the recent infection
and require no ex vivo stimulation. Removing the plasma compo-
nent of blood limits the confounding influence of preexisting cir-
culating antibodies that reflect prior exposure. These circulating
antibodies can affect assay specificity and have markedly limited
the utility of plasma antibody-based assays in areas of the world
where enteric fever and salmonellosis are endemic.
We evaluated our strips using specimens from patients clini-

FIG 3 Determination of minimum amount of anti-human IgG (A) and anti human-IgA (B) required for conjugation of 1 ml of colloidal gold solution. Optical
density (OD) ratios of values at 520 nm to 580 nm and at 600 nm to 520 nm represent stability and polydispersity, respectively.

406 cvi.asm.org Clinical and Vaccine Immunology May 2016 Volume 23 Number 5

FIG 4 Dipsticks detecting S. Typhi LPS-specific IgG in different samples. Both
the control line and test line appear in the positive sample. The absence of the
test line in the presence of the control line indicates a negative sample.
cally suspected to have enteric fever, as well as specimens from
healthy individuals and patients with other febrile illnesses. We
defined participants whose blood cultures were positive for S. Ty-
phi or S. Paratyphi A as definitive cases of enteric fever. The LPS-
specific IgG lymphocyte supernatant-based dipstick had a sensi-
tivity of 98% using this definition. We defined participants with
other febrile illnesses and healthy controls as definitive negatives.
The LPS-specific dipstick test had a specificity of 100% using this
definition. Since there is no clinically practical gold standard for
enteric fever and since microbiologic culturing of blood is known
to have a sensitivity of 30 to 70% (3, 12, 13), the interpretation of
the performance of the dipstick assay in individuals with sus-
pected enteric fever but negative microbiologic culturing of blood
is problematic. Presumably, a subset of these individuals actually
do have enteric fever. However, we could estimate a lower speci-
ficity value: if we assume that all blood culture-negative patients
indeed did not have enteric fever, the LPS-specific IgG dipstick test
would still have a specificity of 87% when all negative controls are
included. These results are very promising, and considering that
some of the culture-negative patients probably did indeed have
enteric fever, they are very likely an underestimate of true speci-
ficity.
Our assay is based on detecting LPS derived from S. Typhi.
Interestingly, our test also detected patients with S. Paratyphi A
bacteremia. Paratyphoid fever caused by S. Paratyphi A accounts
for up to 1 in 5 cases of enteric fever is some areas of Asia, includ-
ing Bangladesh (4), and paratyphoid and typhoid fevers can be
TABLE 3 Results of strip test detecting LPS-specific IgG
No. of subjects
Positive Negative
Study participant group Total strip test strip test
S. Typhi bacteremia 48 48 0 cific than currently available serology assays. This dipstick as-
S. Paratyphi A bacteremia 6 5 1 say thus may be quite useful in detecting patients with enteric
Clinically suspected for enteric fever 88 19 69
but blood culture negative
Febrile illness other than enteric fever 35 0 35
Healthy controls 28 0 28
May 2016 Volume 23 Number 5 Clinical and Vaccine Immunology
Lateral-Flow Assay for Enteric Fever Detection
FIG 5 S. Typhi LPS-specific IgG responses in lymphocyte culture supernatant
prepared from different groups. The dipstick was positive for specimens with
responses of ⱖ16 ELISA units.
clinically indistinguishable (4, 23). The S. Typhi LPS serotype is
defined by the O antigen, determined by the O-specific oligonu-
cleotide and polysaccharides associated with the LPS. S. Typhi O
antigens include serotypes 9 and 12, often expressed on the same
organism. S. Paratyphi A antigens include serotypes 1, 2, and 12.
The identification of S. Paratyphi A-infected patients by the dip-
stick assay presumably rests upon the detection of circulating
lymphocytes expressing anti-serotype 12 O-antigen antibodies
Downloaded from https://journals.asm.org/journal/cvi on 16 March 2024 by 2401:4900:633a:8f74:a260:acbe:dfa8:e71b.
in these individuals.
Invasive salmonellosis is a distinct clinical entity from enteric
fever that is caused by traditionally nontyphoidal strains of Salmo-
nella enterica, especially S. Typhimurium and S. enterica serovar
Enteritidis. Such invasive nontyphoidal salmonellosis (iNTS) is a
significant cause of mortality in malnourished and immunocom-
promised children, especially HIV-infected individuals in sub-Sa-
haran Africa (24). Although we did not assess our dipstick assay in
patients with iNTS (who are rare in Dhaka, Bangladesh), we are
encouraged to note that both S. Typhimurium and S. Enteritidis
can express O antigen 12, suggesting that the current dipstick as-
say might be able to detect at least a subset of individuals with
iNTS.
Our dipstick assay has a number of limitations. It is not point
of care, it requires electricity (centrifuge and incubator), it re-
quires 24 to 48 h until results are available to the clinician, and it
does not discern antimicrobial susceptibility profiles. We could
not collect a large volume of blood for culture, which may be a
reason for the low sensitivity of the blood culture. We enrolled
adult healthy controls although suspected enteric fever patients
were largely children. However, the dipstick assay is extremely
simple, requires only a small volume of peripheral blood and
rudimentary laboratory equipment and training, can be read
with the naked eye, is as sensitive as blood culture, may be more
sensitive than blood culture, and is estimated to be more spe-
fever in resource-limited settings with limited facilities, iden-
tifying those who could most benefit from appropriate treat-
ment to prevent subsequent complications of enteric fever and
minimizing the use of inappropriate antimicrobial agents that
cvi.asm.org 407
Khan et al.
TABLE 4 Sensitivity, specificity, positive predictive value, and negative predictive value of strip test detecting LPS-specific IgGa
Characterization of participant status
Blood culture-positive patients considered positive; healthy individuals and patients
with other febrile illnesses considered negative
Blood culture-positive patients considered positive; blood culture-negative patients,
healthy individuals, and other febrile illnesses considered negative
Blood culture-positive patients considered positive; only blood culture-negative
patients considered negative
a
Values in parentheses are 95% CIs.
drive drug resistance and can cause unnecessary adverse events.
This assay could also assist in defining the burden of enteric
fever in resource-limited regions and could assist in judging the
9.
impact of control programs.
ACKNOWLEDGMENTS
This work was supported by the ICDDR,B and grants from the National
Institutes of Health, including the National Institute of Allergy and Infec- 10.
tious Diseases (AI100023; to E.T.R. and F.Q.), by a Fogarty International
Center Training Grant in Vaccine Development and Public Health
(TW005572; to M.A.S., K.I., and F.K.), by the Bill and Melinda Gates
Foundation (grant number OPP50419), and also by the SIDA fund
(grants 54100020 and 51060029). The funders had no role in study design,
data collection and analysis, decision to publish, or preparation of the 11.
manuscript.
This test was developed in collaboration with a commercial company.
FUNDING INFORMATION
This work, including the efforts of Firdausi Qadri, was funded by SIDA
12.
fund (54100020 and 51060029). This work, including the efforts of Ed-
ward T. Ryan and Firdausi Qadri, was funded by HHS | National Institutes
of Health (NIH) (AI100023). This work, including the efforts of Edward
T. Ryan and Firdausi Qadri, was funded by HHS | NIH | Fogarty Interna- 13.
tional Center (FIC) (TW005572). This work, including the efforts of
Firdausi Qadri, was funded by Bill and Melinda Gates Foundation
(OPP50419).
14.
REFERENCES
1. Khanam F, Sheikh A, Sayeed MA, Bhuiyan MS, Choudhury FK, Salma
U, Pervin S, Sultana T, Ahmed D, Goswami D, Hossain ML, Mamun
KZ, Charles RC, Brooks WA, Calderwood SB, Cravioto A, Ryan ET, 15.
Qadri F. 2013. Evaluation of a typhoid/paratyphoid diagnostic assay
(TPTest) detecting anti-Salmonella IgA in secretions of peripheral blood
lymphocytes in patients in Dhaka, Bangladesh. PLoS Negl Trop Dis
7:e2316. http://dx.doi.org/10.1371/journal.pntd.0002316.
2. Bhutta ZA. 2006. Current concepts in the diagnosis and treatment 16.
of typhoid fever. BMJ 333:78 – 82. http://dx.doi.org/10.1136/bmj.333
.7558.78.
3. Crump JA, Luby SP, Mintz ED. 2004. The global burden of typhoid fever.
Bull World Health Organ 82:346 –353.
4. Crump JA, Mintz ED. 2010. Global trends in typhoid and paratyphoid 17.
fever. Clin Infect Dis 50:241–246. http://dx.doi.org/10.1086/649541.
5. Newton AE, Mintz ED. 2013. Typhoid and paratyphoid fever. In CDC health
information for international travel 2013. Centers for Disease Control and
Prevention, Atlanta, GA. http://wwwnc.cdc.gov/travel/yellowbook/2016 18.
/infectious-diseases-related-to-travel/typhoid-paratyphoid-fever.
6. Crump JA. 2012. Typhoid fever and the challenge of nonmalaria febrile
illness in sub-Saharan Africa. Clin Infect Dis 54:1107–1109. http://dx.doi 19.
.org/10.1093/cid/cis024.
7. Thriemer K, Ley B, Ame SS, Deen JL, Pak GD, Chang NY, Hashim R,
Schmied WH, Busch CJ, Nixon S, Morrissey A, Puri MK, Ochiai RL,
Wierzba T, Clemens JD, Ali M, Jiddawi MS, von Seidlein L, Ali SM.
2012. Clinical and epidemiological features of typhoid fever in Pemba,
Zanzibar: assessment of the performance of the WHO case definitions. 20.
PLoS One 7:e51823. http://dx.doi.org/10.1371/journal.pone.0051823.
8. Thriemer K, Ley B, Menten J, Jacobs J, van den Ende J. 2013. A
408 cvi.asm.org Clinical and Vaccine Immunology
Sensitivity (%) Specificity (%) PPV (%) NPV (%)
98 (90–99) 100 (94–100) 100 (93–100) 98 (91–99)
98 (90–99) 87 (81–91) 73 (62–82) 99 (95–99)
98 (90–99) 78 (68–85) 73 (62–82) 98 (92–99)
systematic review and meta-analysis of the performance of two point of
care typhoid fever tests, Tubex TF and Typhidot, in endemic countries.
PLoS One 8:e81263. http://dx.doi.org/10.1371/journal.pone.0081263.
Keddy KH, Sooka A, Letsoalo ME, Hoyland G, Chaignat CL, Morrissey
AB, Crump JA. 2011. Sensitivity and specificity of typhoid fever rapid
antibody tests for laboratory diagnosis at two sub-Saharan African sites.
Bull World Health Organ 89:640 – 647. http://dx.doi.org/10.2471/BLT.11
.087627.
Sheikh A, Bhuiyan MS, Khanam F, Chowdhury F, Saha A, Ahmed D,
Jamil KM, LaRocque RC, Harris JB, Ahmad MM, Charles R, Brooks
WA, Calderwood SB, Cravioto A, Ryan ET, Qadri F. 2009. Salmonella
enterica serovar Typhi-specific immunoglobulin A antibody responses in
plasma and antibody in lymphocyte supernatant specimens in Bangla-
deshi patients with suspected typhoid fever. Clin Vaccine Immunol 16:
1587–1594. http://dx.doi.org/10.1128/CVI.00311-09.
Ley B, Thriemer K, Ame SM, Mtove GM, von Seidlein L, Amos B,
Hendriksen IC, Mwambuli A, Shoo A, Kim DR, Ochiai LR, Favorov M,
Clemens JD, Wilfing H, Deen JL, Ali SM. 2011. Assessment and com-
parative analysis of a rapid diagnostic test (Tubex) for the diagnosis of
typhoid fever among hospitalized children in rural Tanzania. BMC Infect
Dis 11:147. http://dx.doi.org/10.1186/1471-2334-11-147.
Downloaded from https://journals.asm.org/journal/cvi on 16 March 2024 by 2401:4900:633a:8f74:a260:acbe:dfa8:e71b.
Farooqui BJ, Khurshid M, Ashfaq MK, Khan MA. 1991. Comparative
yield of Salmonella typhi from blood and bone marrow cultures in patients
with fever of unknown origin. J Clin Pathol 44:258 –259. http://dx.doi.org
/10.1136/jcp.44.3.258.
Gilman RH, Terminel M, Levine MM, Hernandez-Mendoza P, Hornick
RB. 1975. Relative efficacy of blood, urine, rectal swab, bone-marrow, and
rose-spot cultures for recovery of Salmonella typhi in typhoid fever. Lancet
i:1211–1213.
Gasem MH, Smits HL, Goris MG, Dolmans WM. 2002. Evaluation of a
simple and rapid dipstick assay for the diagnosis of typhoid fever in Indo-
nesia. J Med Microbiol 51:173–177. http://dx.doi.org/10.1099/0022-1317
-51-2-173.
Naheed A, Ram PK, Brooks WA, Mintz ED, Hossain MA, Parsons MM,
Luby SP, Breiman RF. 2008. Clinical value of Tubex and Typhidot rapid
diagnostic tests for typhoid fever in an urban community clinic in Bangla-
desh. Diagn Microbiol Infect Dis 61:381–386. http://dx.doi.org/10.1016/j
.diagmicrobio.2008.03.018.
Siba V, Horwood PF, Vanuga K, Wapling J, Sehuko R, Siba PM,
Greenhill AR. 2012. Evaluation of serological diagnostic tests for ty-
phoid fever in Papua New Guinea using a composite reference stan-
dard. Clin Vaccine Immunol 19:1833–1837. http://dx.doi.org/10.1128
/CVI.00380-12.
Abdoel TH, Pastoor R, Smits HL, Hatta M. 2007. Laboratory evaluation
of a simple and rapid latex agglutination assay for the serodiagnosis of
typhoid fever. Trans R Soc Trop Med Hyg 101:1032–1038. http://dx.doi
.org/10.1016/j.trstmh.2007.05.020.
Song JH, Cho H, Park MY, Na DS, Moon HB, Pai CH. 1993. Detection
of Salmonella typhi in the blood of patients with typhoid fever by polymer-
ase chain reaction. J Clin Microbiol 31:1439 –1443.
Khanam F, Sayeed MA, Choudhury FK, Sheikh A, Ahmed D, Goswami
D, Hossain ML, Brooks A, Calderwood SB, Charles RC, Cravioto A,
Ryan ET, Qadri F. 2015. Typhoid fever in young children in Bangladesh:
clinical findings, antibiotic susceptibility pattern and immune responses.
PLoS Negl Trop Dis 9:e0003619. http://dx.doi.org/10.1371/journal.pntd
.0003619.
Nara S, Tripathi V, Singh H, Shrivastav TG. 2010. Colloidal gold probe
based rapid immunochromatographic strip assay for cortisol. Anal Chim
Acta 682:66 –71. http://dx.doi.org/10.1016/j.aca.2010.09.041.
May 2016 Volume 23 Number 5

21. Tripathi V, Nara S, Singh K, Singh H, Shrivastav TG. 2012. A compet-
itive immunochromatographic strip assay for 17-alpha-hydroxy proges-
terone using colloidal gold nanoparticles. Clin Chim Acta 413:262–268.
http://dx.doi.org/10.1016/j.cca.2011.10.016.
22. Ye Y, Zhou Y, Mo Z, Cheng W, Yang S, Wang X, Chen F. 2010. Rapid
detection of aflatoxin B1 on membrane by dot-immunogold filtration assay.
Talanta 81:792–798. http://dx.doi.org/10.1016/j.talanta.2010.01.017.
May 2016 Volume 23 Number 5 Clinical and Vaccine Immunology
Lateral-Flow Assay for Enteric Fever Detection
23. Fangtham M, Wilde H. 2008. Emergence of Salmonella paratyphi A as a
major cause of enteric fever: need for early detection, preventive measures,
and effective vaccines. J Travel Med 15:344 –350. http://dx.doi.org/10
.1111/j.1708-8305.2008.00237.x.
24. Morpeth SC, Ramadhani HO, Crump JA. 2009. Invasive non-Typhi
Salmonella disease in Africa. Clin Infect Dis 49:606 – 611. http://dx.doi.org
/10.1086/603553.
Downloaded from https://journals.asm.org/journal/cvi on 16 March 2024 by 2401:4900:633a:8f74:a260:acbe:dfa8:e71b.
cvi.asm.org 409