Clinical Infectious Diseases
SUPPLEMENT ARTICLE
Diagnosing Salmonella enterica Serovar Typhi Infections
by Polymerase Chain Reaction Using EDTA Blood Samples
of Febrile Patients From Burkina Faso
Hassan M. Al-Emran,1,2 Andreas Hahn,1 Jana Baum,3 Ligia Maria Cruz Espinoza,3 Jessica Deerin,3 Justin Im,3 Samuel Ibrango,4 Leon Parfait Kabore,5
Vera von Kalckreuth,3 Frank Konings,3 Florian Marks,3 Emmanuel Sampo,5 Ursula Panzner,3 Se Eun Park,3 Gi Deok Pak,3 Heidi Schütt-Gerowitt,3,6
Christof David Vinnemeier,3,7 Michelle Warren,3 and Abdramane Bassiahi Soura5
1
Bernhard Nocht Institute for Tropical Medicine, and 2German Center for Infection Research, partner site Hamburg-Borstel-Lübeck, Germany; 3International Vaccine Institute, Seoul, Republic of Korea;
Ministry of Health, Ouagadougou, and 5Institut Supérieur des Sciences de la Population, University of Ouagadougou, Burkina Faso; 6Institute of Medical Microbiology, University of Cologne, and
7
Department of Medicine, University Medical Center Hamburg-Eppendorf, Germany
4
Background. Globally, there are an estimated 22 million cases of Salmonella enterica serovar Typhi infection each year. However,
this figure is likely to be an underestimate due to the low sensitivity of blood culture in S. Typhi diagnosis. The aim of this study was
to diagnose S. Typhi by conventional polymerase chain reaction (PCR) using patient’s blood preserved with ethylenediamine tetraacetic acid (EDTA).
Methods. From April 2012 to September 2013, typhoid fever surveillance was conducted in Polesgo and Nioko, 2 dry slum areas in
Ouagadougou, Burkina Faso. Blood culture was performed for febrile patients using an automated blood culture system. Additional blood
was collected in EDTA tubes from those patients and preserved at −80°C. DNA was extracted from EDTA blood and PCR was performed
to identify presence of S. Typhi. Randomly selected PCR products were further sequenced to identify S. Typhi–specific amplicons.
Results. Of 1674 patients, S. Typhi was isolated from 18 (1.1%) individuals by blood culture. EDTA blood was collected from 1578
patients, of which 298 EDTA samples were tested by PCR. Salmonella Typhi–specific DNA was identified in 44 (14.8%) samples. The
sensitivity of S. Typhi–specific PCR from EDTA blood was 89% (74%–100%) among the blood culture–positive cases. Sixteen S. Typhi–
positive PCR products were sequenced, and 13 retrieved the sequence of a S. Typhi–specific amplicon.
Conclusions. These findings suggest that blood culture–based diagnoses of S. Typhi underestimate the burden of typhoid fever in
Burkina Faso. PCR could be considered as an alternative method for the identification and diagnosis of S. Typhi in blood samples.
Keywords. Burkina Faso; Salmonella; Typhi; PCR; sensitivity.
Globally, there were 22 million febrile cases attributed to infection by Salmonella enterica serovar Typhi that led to 217 000
deaths in 2000 [1]. This estimation was adjusted for blood culture sensitivity based on a conservative assumption of 50% [1].
The sensitivity of blood cultures for S. Typhi may be affected by
several factors including antimicrobial pretreatment, the volume of blood collected [2, 3], the duration and conditions of
sample transportation [4], laboratory facilities, and method of
pathogen identification. Few alternative diagnostics for the detection of S. Typhi are available. The Widal test has been one of
the most commonly used diagnostics to date despite its limited
sensitivity of only 39% [5, 6]. Typhidot and Tubex, both serological assays, are also available for use [7, 8], and have a reported
sensitivity of around 70% [9].
Recently, polymerase chain reaction (PCR) has shown great
potential as a diagnostic tool for the detection of pathogens, as
Correspondence: H. M. Al-Emran, Bernhard Nocht Institute for Tropical Medicine, Hamburg,
Germany ([email protected]).
Clinical Infectious Diseases® 2016;62(S1):S37–41
© The Author 2016. Published by Oxford University Press for the Infectious Diseases Society
of America. All rights reserved. For permissions, e-mail [email protected].
DOI: 10.1093/cid/civ770
the method has a high degree of sensitivity and specificity [10–
12]. However, the identification of bacterial DNA directly from
blood samples has also been previously reported to show reduced sensitivity [13]. Several attempts have been made to improve the sensitivity of S. Typhi detection from blood. Use of ox
bile for blood cell lysis, followed by micrococcal nuclease to remove human DNA, improves the detection of DNA from this
intracellular bacteria [14]. An additional study has shown that
use of erythrocyte lysis buffer before DNA extraction improves
the detection rate of S. Typhi by both conventional PCR and
real-time PCR [15]. However, both studies have demonstrated
their methods on spiked blood samples rather than patient samples. Application of these methods may also face challenges in
highly endemic areas and settings where laboratory capacity is
limited. Feasibility issues are compounded by evidence that the
presence of S. Typhi in the blood of most patients is very low
[2]. As a result, the pretreatment of preserved blood samples
prior to PCR amplification to decrease human DNA may also
decrease the concentration of S. Typhi DNA. The objective of
this study was to detect S. Typhi by conventional PCR using
preserved ethylenediamine tetraacetic acid (EDTA) blood
from febrile patients in Burkina Faso and to observe the
Identification of S. Typhi by PCR From EDTA Blood
•
CID 2016:62 (Suppl 1)
•
S37
frequency of detected S. Typhi by PCR in comparison to blood
culture.
METHODS
This study was conducted as part of the Typhoid Fever Surveillance in Africa Program (TSAP), a multicenter, multicountry
surveillance network including 10 countries in sub-Saharan Africa [16]. This substudy included 2 sites in Burkina Faso, Polegso, and Nioko in the north of Ouagadougou. These sites form a
component of the International Network for the Demographic
Evaluation of Populations and Their Health in Developing
Countries (INDEPTH) [17] and are supported by a comprehensive health and demographic surveillance system that is run by
the University of Ouagadougou, Institut Supérieur des Sciences
de la Population.
Blood Collection and Bacterial Culture
Blood was collected from all enrolled patients for bacterial culture and sample preservation. The volume of blood added to the
blood culture bottle was 1–4 mL from the children and 8–10 mL
from the adults. The remainder of blood (<1 mL), when available, was preserved in an EDTA tube. The collected specimens
were transported to the Medical Center Protestant Schiphra,
Ouagadougou, Burkina Faso, for further processing. EDTA
blood was stored in a freezer at −80°C, and bacterial culture was
performed using an automated blood culture system (BacTAlert,
bioMérieux, Durham, North Carolina). Signal-positive samples
were plated on MacConkey agar, Columbia agar enriched
with 5% sheep blood, and chocolate agar (Oxoid, Hampshire,
United Kingdom). For the identification of Salmonella enterica,
API20E biochemical test (bioMérieux, Durham, North Carolina)
and the Oxoid Salmonella Latex Test (Oxoid, Hampshire, United
Kingdom) were performed in parallel. All isolates were stored at
−80°C at the study laboratories. All EDTA blood tubes and bacterial isolates were transported on dry ice to the Bernhard Nocht
Institute for Tropical Medicine, Hamburg, Germany. All S. Typhi
isolates were confirmed by PCR [18].
EDTA Blood Samples
buffer (Tris-Cl, pH 8.5). The volume of DNA hydration buffer
was used depending on the available volume of EDTA blood for
extraction (one-fifth blood volume) to keep the DNA suspension ratio the same. If the EDTA blood volume was <200 µL,
the sample was excluded from further analysis in this study.
The suspended DNA from all samples was then stored at
−20°C until PCR amplification was performed.
DNA Amplification
The DNA extracts from EDTA blood were amplified with
S. Typhi–specific primers and targeted the sty1599 gene, as
described by Park et al [19]. QIAGEN multiplex PCR mastermix (Qiagen, Hilden, Germany) containing 3 mM of magnesium chloride and Q-solution was used to reduce nonspecific
bands from highly concentrated human DNA. The final concentration of the primer pair was 0.10 μmol L−1 in PCR reactions.
Ten microliters of sample was added to each PCR mixture, with
a total volume of 40 µL per reaction. The PCR conditions were
denaturing (94°C for 30 seconds), annealing (58°C for 90 seconds), and extension (72°C for 45 seconds) for 35 cycles, followed by a final extension (72°C for 10 minutes).
Sensitivity of the PCR
All blood culture–positive S. Typhi cases were considered as
true positives in this study. We considered only the blood
culture–positive S. Typhi cases as reference, as the sensitivity
of blood culture is low [1]. Therefore, the sensitivity of PCR
was calculated among the cases where S. Typhi was isolated
by blood culture.
Sequencing of the PCR Amplicons
Randomly selected PCR-positive products were sent to Eurofins
Scientific (Hamburg, Germany) for purification and bidirectional
Table 1. Study Profile and Diagnosis of Salmonella Typhi, Burkina Faso,
April 2012–September 2013
Ouagadougou, Burkina Faso
Characteristic
From all collected EDTA blood samples, 300 were selected randomly using a random number generator in Microsoft Excel
(Microsoft Corporation, Redmond, Washington). Additionally,
EDTA blood samples from all blood culture–positive S. Typhi
cases were included for investigation in this study.
Patients recruited
Children aged <15 y, No. (%)
Median age, y (25th, 75th percentile)
Sex distribution, female, No. (%)
Inpatients, No. (% of total recruited)
CID 2016:62 (Suppl 1)
918
756
710 (77)
517 (68)
4 (1, 12)
467 (51)
7 (3, 21)
404 (53)
66 (7)
NA
Blood culture results
A QIAGEN FlexiGene DNA kit (Qiagen, Hilden, Germany)
was used for the scalable extraction of DNA from 200–500 µL
of EDTA blood. DNA was extracted following the manufacturer’s specifications with a minor modification to the DNA suspension step. In brief, 500 µL of EDTA blood was lysed with
lysis buffer, and the proteins were digested with proteolytic enzyme. DNA was precipitated using isopropanol, washed in 70%
ethanol, dried, and then resuspended in 100 µL of hydration
•
Polesgoa
Patient demographics
DNA Extraction
S38
Nioko
•
Al-Emran et al
Total BC performed, No.
BC positive for Salmonella Typhi,
No. (% of total BCs performed)
918
5 (0.5)
756
13 (1.7)
EDTA-PCR results
Total PCR assays performed, No.
Total positive for Salmonella Typhi,
No. (% of total PCR assays performed)
154
28 (18)
144
16 (11)
Abbreviations: BC, blood culture; EDTA, ethylenediamine tetraacetic acid; NA, not
applicable; PCR, polymerase chain reaction.
a
This site had no inpatient facility; recruitment was performed only in outpatient department.
sequencing. Forward and reverse sequences were assembled
using Seqscape software version 2.1.1 (Applied Biosystems, Waltham, Massachusetts). Recovered sequences were screened for
S. Typhi–specific amplicons (accession number NC_003198.1).
Ethical Approval
This substudy was embedded in the TSAP study [16]. The
TSAP study protocol was approved by the institutional review
board of the International Vaccine Institute and the national
ethical review board in Burkina Faso. Ethical approval included
the secondary analysis of patient samples.
RESULTS
A total of 1674 febrile patients were enrolled from Polegso
and Nioko between April 2012 and September 2013. Among
these, 1227 (73%) were <15 years of age and 871 (52%) were
female. The median age of the enrolled patients was 7 years
(interquartile range [IQR], 3–21 years) in Polesgo and 4 years
(IQR, 1–12 years) in Nioko (Table 1).
Blood culture was performed for all 1674 recruited patients,
and S. Typhi was isolated in 18 samples (1.1%). EDTA blood samples were collected from 1578 (94%) patients. From the randomly
selected 300 EDTA blood samples, 298 (99%) had a sufficient volume (≥200 µL) to extract DNA for PCR amplification. Forty-four
of the 298 (14.8%) DNA samples were positive for S. Typhi by
PCR (Figure 1) of which 4 samples were additionally blood culture positive. One blood culture–positive EDTA sample was
found to be negative by PCR amplification. The remaining 253
samples were negative by both PCR and blood culture.
From 13 blood culture–positive S. Typhi cases, which were
tested additionally, 12 were positive by PCR. In total, PCR positively detected S. Typhi in 16 of 18 blood culture–positive cases
(Table 2). Of those 18 blood culture–positive cases, PCR sensitivity was 89% (95% confidence interval, 74%–100%).
Figure 1. Gel electrophoresis of the Salmonella Typhi–specific polymerase chain reaction (PCR) assay from ethylenediamine tetraacetic acid (EDTA) blood of febrile patients.
Columns 1–16 represent samples from blood culture–negative patients. In lane A, columns 1–3, 8, and 11–16 were negative by PCR. Columns 4–7, 9, and 10 were positive for
S. Typhi. In lane B, columns 1–11 and 14–16 were negative, and columns 12 and 13 were positive for S. Typhi. Abbreviations: L, ladder (100 bp); NC, reagent negative control);
PC, S. Typhi–positive control (203 bp).
Identification of S. Typhi by PCR From EDTA Blood
•
CID 2016:62 (Suppl 1)
•
S39
Table 2. Comparison Between Blood Culture and Polymerase Chain
Reaction Results for the Diagnosis of Salmonella Typhi
Randomly Selected Samples
PCR Results
Blood Culture
Negative
PCR negative
253
1
1
PCR positive
40
4
12
293
5
13
Total
Blood Culture
Positive
Additional
Samples
Blood
CulturePositive
Abbreviation: PCR, polymerase chain reaction.
Sixteen of 44 (36.4%) samples that tested positive for S. Typhi
through PCR (4/16 blood culture positive) were sent for DNA
sequencing. DNA sequences were recovered from 13 of 16
(81.2%) PCR-positive cases. All 13 retrieved DNA sequences
were identified as S. Typhi–specific amplicons. The remaining
3 cases did not provide a sufficient signal to read the sequence
data. Sequencing data were not used for diagnosis or for the calculation of disease prevalence or sensitivity.
DISCUSSION
This study demonstrated that the use of conventional PCR assays to diagnose S. Typhi infections uncovered a number of positive cases in Burkina Faso that were not identified by blood
culture diagnoses. This molecular diagnostic method increased
the frequency of identified S. Typhi among the febrile patients
from 1.1% to 14.8%. The data confirm the low sensitivity of
blood culture and suggest that blood culture for typhoid fever
surveillance underestimates the disease burden, although it
has been widely accepted as the standard diagnostic technique
[1]. Accounting for potential false-negative results from blood
culture, the total number of true S. Typhi–positive cases was unknown. Thus, the specificity of PCR could not be calculated in
this study. From the known blood culture–positive cases, only
the sensitivity of PCR (89%) was calculated. The sensitivity of
blood culture was not compared with PCR as the gold-standard
method for the diagnosis of S. Typhi, bone marrow culture, was
not performed here. However, previous studies reported a sensitivity of 85%–99% for the detection of S. Typhi in blood samples by nested PCR, compared with 30%–62% sensitivity for
blood cultures [6, 20].
The low sensitivity of blood culture has been shown to be associated with prior treatment with antimicrobials and the long
duration between the onset of disease and sample collection [2].
Both procedures decrease the likelihood of growing S. Typhi in
bacterial cultures. Use of low volume of blood for bacterial culture also affects sensitivity [3]. Bone marrow culture for the diagnosis of typhoid fever has been reported to have a higher
sensitivity than blood culture [21]; however, the sample collection procedure is highly invasive and requires skilled medical
personnel. Any culture-based diagnoses require a minimum
S40
•
CID 2016:62 (Suppl 1)
•
Al-Emran et al
of 2 days for isolation and identification of the pathogen. In
contrast, PCR diagnosis can be completed in 7–8 hours. In addition, the volume of blood that is required for PCR is much less
than what is recommended for blood culture. This is particularly important for low-income countries, where many children
are often malnourished or underweight and reluctant to provide
blood. In this study, 200–500 µL of blood was used for the DNA
extraction and 3000–5000 µL of blood was drawn for blood culture in most cases. For each PCR reaction, 10 µL of sample
DNA extracted from 50 µL of blood was used.
The study only analyzed 19% of EDTA samples collected in
Burkina Faso. PCR of all EDTA samples may improve the
precision of site-specific disease frequency estimates. The
DNA sequence analysis was performed on only 36% of
the PCR-positive amplicons. Salmonella Typhi–specific DNA
was identified in all 13 sequence-recovered cases, indicating
the presence of S. Typhi in PCR-positive samples.
The frequency of typhoid fever measured by conventional
PCR was high among febrile patients (18% in Nioko and 11%
in Polesgo), and low for blood cultures (0.5% in Nioko and 1.7%
in Polesgo). Based on the blood culture results, annual incidence of S. Typhi calculated in the TSAP study (Marks et al, unpublished data) was 107 and 402 per 100 000 in Nioko and
Polesgo, respectively. Despite this underestimation of incidence
due to low sensitivity of blood cultures, the World Health Organization has classified Burkina Faso as a high-incidence region for S. Typhi [1]. The findings from this study suggest
that the true burden of S. Typhi cannot be estimated by blood
culture alone. We recommend using PCR in addition to standard blood culture for the routine diagnosis of suspected
typhoid fever cases in hospitals, as well as extending this method to S. Typhi surveillance programs in other low-income
countries.
Notes
Disclaimer. The funders had no role in study design, data collection
and analysis, decision to publish, or preparation of the manuscript. The
findings and conclusions contained within are those of the authors and
do not necessarily reflect the positions or policies of the Bill & Melinda
Gates Foundation.
Financial support. This research was funded by the Bill & Melinda
Gates Foundation (OPPGH5231). The International Vaccine Institute
acknowledges its donors, including the Republic of Korea and the Swedish
International Development Cooperation Agency. This publication was
made possible through a grant from the Bill & Melinda Gates Foundation
(OPP1129380).
Supplement sponsorship. This article appears as part of the supplement
“Typhoid Fever Surveillance in Africa Program (TSAP),” sponsored by the
International Vaccine Institute.
Potential conflicts of interest. All authors: No potential conflicts of
interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the
content of the manuscript have been disclosed.
References
1. Crump JA, Luby SP, Mintz ED. The global burden of typhoid fever. Bull World
Health Organ 2004; 82:346–53.
2. Wain J, Bay PVB, Vinh H, et al. Quantitation of bacteria in bone marrow from
patients with typhoid fever: relationship between counts and clinical features.
J Clin Microbiol 2001; 39:1571–6.
3. Hoffman SL, Edman DC, Punjabi NH, et al. Bone marrow aspirate culture superior to streptokinase clot culture and 8 ML 1:10 blood-to-broth ratio blood culture
for diagnosis of typhoid fever. Am J Trop Med Hyg 1986; 35:836–9.
4. Frickmann H, Dekker D, Boahen K, et al. Increased detection of invasive enteropathogenic bacteria in pre-incubated blood culture materials by real-time PCR in
comparison with automated incubation in sub-Saharan Africa. Scand J Infect Dis
2013; 45:616–22.
5. Levine MM, Grados O, Gilman RH, Woodward WE, Solis-Plaza R, Waldman W.
Diagnostic value of the Widal test in areas endemic for typhoid fever. Am J Trop
Med Hyg 1978; 27:795–800.
6. Hatta M, Smits HL. Detection of Salmonella Typhi by nested polymerase chain
reaction in blood, urine, and stool samples. Am J Trop Med Hyg 2007; 76:139–43.
7. Lim PL, Tam FCH, Cheong YM, Jegathesan M. One-step 2-minute test to detect
typhoid-specific antibodies based on particle separation in tubes. J Clin Microbiol
1998; 36:2271–8.
8. Prakash P, Sen MR, Mishra OP, Gulati AK, Shukla BN, Nath G. Dot enzyme immunoassay (Typhidot) in diagnosis of typhoid fever in children. J Trop Pediatr
2007; 53:216–7.
9. Dutta S, Sur D, Manna B, et al. Evaluation of new-generation serologic tests for the
diagnosis of typhoid fever: data from a community-based surveillance in Calcutta,
India. Diagn Microbiol Infect Dis 2006; 56:359–65.
10. Abdissa A, Asrat D, Kronvall G, et al. High diversity of group A streptococcal emm
types among healthy school children in Ethiopia. Clin Infect Dis 2006; 42:1362–7.
11. Berkley JA, Lowe BS, Mwangi I, et al. Bacteremia among children admitted to a
rural hospital in Kenya. N Engl J Med 2005; 352:39–47.
12. Moyo SJ, Maselle SY, Matee MI, Langeland N, Mylvaganam H. Identification of
diarrheagenic Escherichia coli isolated from infants and children in Dar es Salaam,
Tanzania. BMC Infect Dis 2007; 7:92.
13. Bhutta ZA. Current concepts in the diagnosis and treatment of typhoid fever. BMJ
2006; 333:78–82.
14. Zhou L, Pollard AJ. A fast and highly sensitive blood culture PCR method for clinical detection of Salmonella enterica serovar Typhi. Ann Clin Microbiol Antimicrob 2010; 9:14.
15. Boyd MA, Tennant SM, Melendez JH, et al. Adaptation of red blood cell lysis represents a fundamental breakthrough that improves the sensitivity of Salmonella
detection in blood. J Appl Microbiol 2015; 118:1199–209.
16. Von Kalckreuth V, Konings F, Aaby P, et al. The Typhoid Fever Surveillance in
Africa Program (TSAP): clinical, diagnostic, and epidemiological methodologies.
Clin Infect Dis 2016; 62(suppl 1):S9–16.
17. Sankoh O, Byass P. The INDEPTH network: filling vital gaps in global epidemiology. Int J Epidemiol 2012; 41:579–88.
18. Al-Emran HM, Krumkamp R, Dekker DM, et al. Validation and identification of
invasive Salmonella serotypes in sub-Saharan Africa by multiplex polymerase
chain reaction. Clin Infect Dis 2016; 62(suppl 1):S81–3.
19. Park SH, Kim HJ, Cho WH, et al. Identification of Salmonella enterica subspecies
I, Salmonella enterica serovars Typhimurium, Enteritidis and Typhi using multiplex PCR. FEMS Microbiol Lett 2009; 301:137–46.
20. Prakash P, Mishra OP, Singh AK, Gulati AK, Nath G. Evaluation of nested PCR in
diagnosis of typhoid fever. J Clin Microbiol 2005; 43:431–2.
21. Gasem MH, Dolmans WM, Isbandrio BB, Wahyono H, Keuter M, Djokomoeljanto R. Culture of Salmonella Typhi and Salmonella Paratyphi from
blood and bone marrow in suspected typhoid fever. Trop Geogr Med 1995;
47:164–7.
Identification of S. Typhi by PCR From EDTA Blood
•
CID 2016:62 (Suppl 1)
•
S41