Skip to main content

Detection of Salmonella Typhi nucleic acid by RT-PCR and anti-HlyE, -CdtB, -PilL, and -Vi IgM by ELISA at sites in Ghana, Madagascar and Ethiopia



We aimed to assess the prevalence of Salmonella Typhi through DNA and IgM-antibody detection methods as a prelude to extended surveillance activities at sites in Ghana, Madagascar, and Ethiopia.


We performed species-specific real-time polymerase reaction (RT-PCR) to identify bacterial nucleic acid, and enzyme-linked immunosorbent assay (ELISA) for detecting HlyE/STY1498-, CdtB/STY1886-, pilL/STY4539- and Vi-antigens in blood and biopsy specimens of febrile and non-febrile subjects. We generated antigen-specific ELISA proxy cut-offs by change-point analyses, and utilized cumulative sum as detection method coupled with 1000 repetitive bootstrap analyses. We computed prevalence rates in addition to odds ratios to assess correlations between ELISA outcomes and participant characteristics.


Definitive positive RT-PCR results were obtained from samples of febrile subjects originating from Adama Zuria/Ethiopia (1.9%, 2/104), Wolayita Sodo/Ethiopia (1.0%, 1/100), Diego/Madagascar (1.0%, 1/100), and Kintampo/Ghana (1.0%, 1/100), and from samples of non-febrile subjects from Wolayita Sodo/Ethiopia (1%, 2/201). While IgM antibodies against all antigens were identified across all sites, prevalence rates were highest at all Ethiopian sites, albeit in non-febrile populations. Significant correlations in febrile subjects aged < 15 years versus ≥ 15 years were detected for Vi (Odds Ratio (OR): 8.00, p = 0.034) in Adama Zuria/Ethiopia, STY1498 (OR: 3.21, p = 0.008), STY1886 (OR: 2.31, p = 0.054) and STY4539 (OR: 2.82, p = 0.022) in Diego/Madagascar, and STY1498 (OR: 2.45, p = 0.034) in Kintampo/Ghana. We found statistical significance in non-febrile male versus female subjects for STY1498 (OR: 1.96, p = 0.020) in Adama Zuria/Ethiopia, Vi (OR: 2.84, p = 0.048) in Diego/Madagascar, and STY4539 (OR: 0.46, p = 0.009) in Kintampo/Ghana.


Findings indicate non-discriminatory stages of acute infections, though with site-specific differences. Immune responses among non-febrile, presumably healthy participants may mask recall and/or reporting bias leading to misclassification, or asymptomatic, subclinical infection signs induced by suppression of inflammatory responses. As most Ethiopian participants were ≥ 15 years of age and not at high-risk, the true S. Typhi burden was likely missed. Change-point analyses for generating ELISA proxy cut-offs appeared robust, though misclassification is possible. Our findings provided important information that may be useful to assess sites prior to implementing surveillance for febrile illness including Salmonella disease.

Peer Review reports


Community-acquired febrile diseases remain a major cause of morbidity and mortality across sub-Saharan Africa. The management of these diseases presents critical challenges for healthcare personnel because of their non-specific clinical presentations coupled with limited diagnostic capacities [1, 2]. This frequently results in the unnecessary use of antimicrobial drugs, which promotes emerging bacterial resistance [1, 2]. Human-pathogenic febrile Salmonella disease is reported less frequently from Africa compared to Asia largely because its prevalence is assessed via cross-sectional disease investigations at settings scattered across the continent than long-term disease monitoring [3, 4]. Estimates of the World Health Organization (WHO) stated that S. Typhi, the etiologic agent of typhoid fever, is responsible for as many as 20 million cases per year worldwide leading to 128,000–161,000 deaths [5]. These findings, along with the devastating impact of recent epidemics due to Ebola virus and coronavirus/COVID-19, both pathogens presenting initially with mild to moderate infectious disease symptoms among individuals without underlying comorbidities, emphasize the importance of public health surveillance. Methods to assess fever etiologies combined with means to consolidate existing and new data points support preparedness and timely responses to disease outbreaks and emergencies [5,6,7,8].

To fill the knowledge gap of the burden of community-acquired febrile illness in Africa, and to build on findings from the multi-country, multi-center Typhoid Fever Surveillance in Africa Program (TSAP) conducted during 2010–2014, research on fever and particular on Salmonella disease continued under its successor, i.e., the Severe Typhoid Fever in Africa (SETA) program [9,10,11,12,13]. SETA was implemented during 2016–2019 to assess the burden, severity, and long-term sequelae of salmonellosis, including the identification of carriers and costs associated with this illness. Evidence generated by both programs will strengthen the decision-making on the prevention of Salmonella disease through vaccination, and will provide insights into considerations of country-specific adaptations for the management of febrile illness.

Although both programs (TSAP and SETA) implemented fever surveillance at numerous settings, new sites covering additional geographic areas were proposed in Ghana, Madagascar and Ethiopia for extended surveillance activities. We, therefore, aimed to assess the presence of S. Typhi in these sites via IgM-antibody detection directed against selected antigens by enzyme-linked immunosorbent assay (ELISA), and species-specific nucleic acid amplification by real-time polymerase chain reaction (RT-PCR) in both, blood and surgical biopsy samples, of acute febrile subjects (FSs) and non-febrile subjects (NFSs) serving as unmatched comparator cohorts.


Study sites

The following sites were selected for extended surveillance: Kintampo in Ghana, Diego and Mahajanga in Madagascar, and Addis Ketema Woreda-7/-8 in Addis Ababa, Adama Wenji Zuria, Debre Birhan, Arba Minch and Wolayita Sodo in Ethiopia (Additional file 1: Fig. S1A). The Kintampo site was selected because of its participation in the International Network for the Demographic Evaluation of Populations and Their Health (INDEPTH) network. The Malagasy sites were chosen based on their coastal location, a presumed high disease burden, and following discussions with Malagasy country partners. In Mahajanga, previously published information on typhoid fever also existed [14]. The Ethiopian sites were included in consultation with the Ethiopian Health Management Information System and the Federal Ministry of Health based on contemporary unpublished reports on the prevalence of diarrheal illness. Site boundaries were selected based on administrative units/municipalities within catchment areas of public, primary and secondary/tertiary recruitment healthcare facilities providing general healthcare; demographic data were extracted from most up-to-date census and statistical services, and were projected to 2016 using annual growth rates (Table 1) [15,16,17,18,19,20,21,22].

Table 1 Eligibility criteria for febrile and non-febrile subjects in Ghana, Madagascar and Ethiopia including recruitment healthcare facilities, site boundaries and demographic data

Study enrollment

Before the study initiation, study teams were trained on project and consent procedures according to Standard Operating Procedures including the handling and testing of participant samples. Acute FSs and NFSs of all ages and both sexes that presented at in-patient and out-patient departments and surgical units of recruitment healthcare facilities were assessed for their eligibility by trained study physicians before study inclusion. Inclusion criteria for FSs differed by country because of the assumed high numbers of febrile illness caused by infections of the respiratory tract, the urinary tract, and the gastrointestinal tract in Ethiopia, and high suspicion of malaria co-infections in Ghana. Fever was defined uniformly across all sites as fever history of ≥ 3 days (72 h) OR tympanic body temperature of ≥ 38.0 °C or axillary body temperature of ≥ 37.5 °C. See Table 1 for more details on the eligibility criteria.

Our goal was to recruit a convenience sample of 100 FSs and 200 NFSs per site during a three-month study period, and to ensure that NFSs visiting for routine health check-ups were enrolled simultaneously to FSs. Upon enrollment, data on demographics and specimen collection were recorded for FSs and NFSs. Data on physical examinations, pre-treatment with antimalarials and/or antibiotics, and clinical appraisals performed by trained study physicians were captured for FSs. All febrile subjects received treatment in accordance with country-specific national standards of care.

Sample handling and testing

Single whole blood samples ≥ 1.0 mL were collected from FSs and NFSs who were < 16 years of age; samples ≥ 2.0 mL were collected from FSs and NFSs who were ≥ 16-years-old. We examined blood from FSs for the presence of malaria parasites using microscopy or rapid diagnostic testing prior to transfer blood into sterile EDTA-collection tubes for subsequent separation into blood cells and plasma by centrifugation. Tissue biopsies were obtained from FSs undergoing surgical procedures due to disease complications such as gastrointestinal perforation. Biopsy specimens of 5–10 mm3 were taken, if feasible, without compromising the surrounding tissue and placed into vials containing 500 µl sterile phosphate-buffered saline. All samples were stored at − 80 °C.

S. Typhi IgM ELISA antigens detected by protein microarrays performed in previous investigations of acute naturally acquired infection included hemolysin E (HlyE)/STY1498, cytolethal distending toxin subunit B homolog (CdtB)/STY1886, pilL/STY4539, and Vi. Methods for amplification, gene expression and purification of antigen-containing plasmid constructs were described previously [6, 23]; similarly, ELISA testing was done as described earlier [6, 24]. Briefly, 96-well microtiter plates were coated with 7 µg/ml STY4539, STY1886 or STY1498 or 1 µg/ml Vi in 50 mM carbonate-bicarbonate buffer final concentration per well. We used 5% milk solution prepared in phosphate-buffered saline for washing and blocking throughout. Antigen-coated plates were incubated with 100 µl/well of 1:200 diluted participant plasma followed by 100 µl/well of alkaline phosphate-conjugated anti-human IgM serum. Plates were developed using the p-nitrophenyl phosphate substrate. Absorbance of optical density (OD) values was read at 405 nm and 490 nm wavelength by an automated plate reader.

S. Typhi RT-PCR was performed as described previously using freshly extracted DNA from blood cells and tissue biopsies according to Qiagen’s instructions for the QIAamp DNA Blood Midi Kit [25, 26]. Amplification reactions of 25 µl volume were prepared that contained 5 mM MgCl2, 0.2 mM each deoxynucleotide triphosphate, 1 U Taq DNA polymerase, and 5 µl template DNA. Final reaction concentrations of species-specific primers, i.e., ST-Frt 5′-CGC-GAA-GTC-AGA-GTC-GAC-ATA-G-3′, ST-Rrt 5′-AAG-ACC-TCA-ACG-CCG-ATC-AC-3′, and the probe, i.e., ST-Probe 5′-FAM-CAT-TTG-TTC-TGG-AGC-AGG-CTG-ACG-GTA-MRA-3′, were 0.4 µM and 0.15 µM, respectively. We utilized cycling conditions of 15 min at 95 °C followed by 45 cycles of 30 s at 95 °C, 30 s at 60 °C, and 30 s at 72 °C. The Escherichia coli VU1 strain harboring the gB gene from the Phocid herpes virus served as an internal control to monitor the efficacy of DNA extraction and DNA amplification. We set the following diagnostic crossing points: < 35 cycles as definitive positive, ≥ 35 to < 40 cycles as probable positive, and ≥ 40 cycles as negative.


Participant characteristics were computed as absolute and relative frequencies. S. Typhi RT-PCR definitive as well as probable positive and negative were calculated as relative frequencies by site, for both FSs and NFSs. We generated antigen-specific ELISA cut-offs by change-point analyses as initially described by Lardeux et al. [27]. These analyses identify proxy points among ascending ordered absorbance values at points of change in statistical properties or slope of data; this analysis facilitated the separation of negative and positive values in the absence of specific control sera. The cumulative sum (CUSUM) method was used as a detection option coupled with 1000 repetitive bootstrap analyses of original OD values. We applied sampling with replacement for the repetitive analyses. The results presented in Additional file 1: Fig. S2A illustrates overall distributions of original OD values by antigen as absolute and relative cumulative frequencies; Additional file 1: Fig. S3A depicts means, and Fig. 1 presents change points of original and 1000 bootstrap samples accompanied by 95% and 99% confidence intervals (CIs), respectively. We applied antigen-specific proxy points to absorbance values, and calculated absolute positivity and negativity, and prevalence rates (PRs) by site, FSs versus NFSs, gender, and age, i.e., < 15 years of age and ≥ 15 years of age. We utilized odds ratios (ORs) with 95%CIs to assess potential correlations between ELISA findings and characteristics congruent among FSs and NFSs inclusive of gender, age, site, and study subjects (Table 2); a two-sided p-value of < 0.05 was considered statistically significant. Analyses were performed with STATA statistical package (StataCorp LLC, Version 14, Texas, USA).

Fig. 1
figure 1

Change point by antigen. Red dot = change point of original data; blue dot = change point of each 1000 bootstrap random samples; light red dashed line = upper and lower 95% confidence interval (CI) of original data; dark red dashed line = upper and lower 99% CI of original data

Table 2 Baseline characteristics of febrile and non-febrile subjects in Ghana, Madagascar and Ethiopia at study enrollment


The baseline characteristics of all study participants are summarized in Table 2. Male and female FSs and NFSs were distributed nearly equally across all sites. The FSs and NFSs enrolled in the study were primarily ≥ 15 years of age, except for FSs recruited from the Kintampo and Mahajanga sites. Most FSs had an axillary body temperature of ≥ 37.5 °C, except for those from Addis Ketema Woreda-8 and Wolayita Sodo. Likewise, most FSs were enrolled at out-patient departments as opposed to in-patient departments, except for those from the two Malagasy sites (Diego: 95.0%, 95/100; Mahajanga: 85.0%, 85/100). Antimalarial pre-treatment was reported only rarely by FSs except for Kintampo (20.0%, 20/100) and Adama Zuria (19.2%, 20/104). Soley FSs from both Malagasy sites (Diego: 18.0%, 18/100; Mahajanga: 70.0%, 70/100) and Adama Zuria (38.5%, 40/104) recalled pre-treatment with antibiotics. Plasmodium falciparum malaria was detected among FSs recruited in Ghana only. Clinical appraisals revealed typhoid fever as the most common presumptive diagnosis across all sites; other common diagnoses included acute respiratory tract infection in Diego and Adama Zuria, gastrointestinal tract infection at both Malagasy sites, and fever of unknown origin in Diego.

Definitive positive RT-PCR results were found among FSs in Adama Zuria (1.9%, 2/104), Wolayita Sodo (1.0%, 1/100), Diego (1.0%, 1/100) and Kintampo (1.0%, 1/100), and NFSs in Wolayita Sodo (1.0%, 2/201) (Fig. 2). No Salmonella DNA could be amplified from tissue biopsies of two FSs from Arba Minch that underwent surgical procedures for gastrointestinal perforation due to suspected typhoid fever.

Fig. 2
figure 2

IgM ELISA and RT-PCR by antigen, site and study subjects. W7 = Woreda 7; W8 = Woreda 8; FS = febrile subject; NFS = non-febrile subject; dRTPCR = definitive RT-PCR; pRTPCR = probable RT-PCR; +  = positivity; − = negativity

Absolute ELISA positivity generated by applying proxy cut-off points (Fig. 1) to absorbance values is shown in Fig. 2. In Kintampo, PRs among FSs (0.05%) and NFSs (0.08%) < 15 years of age were highest for the STY1498 antigen, whereas FSs (0.02%) and NFSs (0.05%) ≥ 15 years of age exhibited strongest responses to STY1886 and STY1498, respectively (Fig. 3). PRs peaked among male FSs as well as male NFSs for STY1886 (0.03%) and STY1498 (0.07%), respectively, and female FSs as well as female NFSs for STY1498 (0.04%) and STY4539 (0.08%), respectively. A 2.45 times (95%CI: 0.98–6.19, p = 0.034) higher odds of anti-STY1498 IgM was detected in FSs < 15 years of age compared to FSs ≥ 15 years of age, and a 0.46 times (95%CI: 0.25–0.86, p = 0.009) lower odds of anti-STY4539 antibodies in male NFSs than female NFSs.

Fig. 3
figure 3

IgM-ELISA by antigen, site and study subjects stratified by age group. W7 = Woreda 7; W8 = Woreda 8; FS = febrile subjects; NFS = non-febrile subjects; yrs = years; + = positivity; − = negativity

In Madagascar, positive anti-Vi responses peaked at 0.02% and 0.02% among FSs < 15 and ≥ 15 years of age in Diego, and at 0.02% and 0.01% in Mahajanga, respectively. Both male and female FSs in Diego (0.02% and 0.02%, respectively) and Mahajanga (0.01% and 0.01%, respectively) exhibited the strongest anti-Vi immune responses. In Diego, the odds were 3.21 times (95%CI: 1.24–8.63, p = 0.008), 2.31 times (95%CI: 0.91–6.06, p = 0.054), and 2.82 times (95%CI: 1.06–7.86, p = 0.022) higher for detecting anti-STY1498, anti-STY1886, and anti-STY4539 IgM antibodies in FSs < 15 years of age compared to FSs ≥ 15 years of age, respectively. At the same site, the odds for exhibiting an anti-Vi responses was 2.84 times (95%CI: 0.90–10.55, p = 0.048) higher among NFSs < 15 years of age compared to NFSs ≥ 15 years of age.

Across all Ethiopian sites, NFSs exhibited higher IgM responses to all antigens compared to FSs, except for immune responses to STY4539 in FSs recruited at Debre Birhan. FSs and NFSs from Addis Ketema Woreda-7/-8 in Addis Ababa and Adama Zuria presented generally with highest PRs; NFSs from Adama Zuria demonstrated highest PRs, with the highest values detected among male NFSs for STY1498 (0.51%), STY1886 (0.66%), and Vi (0.54%). We also found that the odds were 8.00 times higher (95%CI: 0.74–9.72, p = 0.034) for detecting anti-Vi antibodies among FSs in Adama Zuria < 15 year-olds compared to those ≥ 15 years of age. Odds were also 1.96 times higher (95%CI: 1.06–3.62, p = 0.020) for detecting a positive response to anti-STY1498 antibodies among male compared to female NFSs.


Following the aim of our investigation, we amplified S. Typhi DNA among blood specimens from FSs in Adama Zuria, Wolayita Sodo, Diego and Kintampo, and among blood samples from NFSs in Wolayita Sodo. We also detected IgM-antibodies against all S. Typhi antigens investigated among FSs and NFSs across study sites though with site-specific differences; this result is indicative of non-discriminatory stages of acute infections prior to immunoglobulin class-switching due to immunological maturation [23].

HlyE/STY1498 antigen is expressed by several enteric bacteria including S. Typhi and S. Paratyphi A. It acts as transmembrane pore-forming toxin that induces lysis of erythrocytes and mammalian cells, and promotes apoptosis in macrophages. HlyE/STY1498, therefore, impacts cytotoxicity and intracellular growth within the mononuclear phagocyte system, and is capable of modulating the progression to persistent systemic infections [28,29,30]. Circulating anti-STY1498 antibodies have been reported in both acute and convalescent/transient cases, although it is unlikely to be detected in carriers as Salmonella typically resides in niches that are inaccessible to the host immune system [28, 31]. CdtB/STY1886 is expressed by both, S. Typhi and S. Paratyphi A, as well as other bacterial pathogens, but rarely in non-typhoidal serovars [28]. CdtB as one of two A-subunits of the AB-type typhoid toxin forms a complex with PltA (pertussis-like toxin A, A-subunit) and PltB (pertussis-like toxin B, B-subunit) that mediates its delivery into host cells [32]. CdtB induces cell cycle arrest via DNA damage that ultimately leads to apoptosis or cell senescence in the short term, as well as long-term consequences that include tumorigenesis, neoplastic lesions, and pathogen persistence [30, 33, 34]. Interestingly, anti-STY1886 immunoglobulins are more likely to develop during convalescence and in response to transient exposure rather than in early acute cases [28, 33]. The PilL/STY4539 is a putative exported protein and a component of the type IV pili fimbrial adhesin that is required for bacterial twitching or gliding motility, adhesion and macrophage-mediated uptake into gut epithelial cells; it also plays an important role in host immune evasion, biofilm formation, and microcolony aggregation [6, 35, 36]. Antibodies to PilL/STY4539 are well detectable among cases suffering from acute infection. Lastly, the Vi carbohydrate capsule is elementary for extracellular survival, protection against neutrophilic oxidative burst, and diminished macrophage responses following ingestion. The antigen appears detectable at a rather late stage of infection, and is presumably associated with Salmonella carriage in endemic settings [37]. Protective anti-Vi responses are, based on data obtained through a controlled human challenge model, polyisotypic and polyfunctional [38]. The Vi-antigen enhances the virulence of S. Typhi [39] and is an important component of polysaccharide and conjugate vaccine candidates [40]. Studies have shown that anti-Vi immune responses are detectable in individuals from S. Typhi-endemic areas; an estimated 19% to 58% of S. Typhi exposed individuals have circulating anti-Vi antibodies in their bloodstream [41,42,43].

Immune responses to all antigens were detected in NFSs across all sites with the strongest PRs identified at the Ethiopian sites. This finding was somewhat surprising as IgM-antibodies have been reported as suitable markers for differentiating acutely infected from healthy individuals [23]. NFSs were presumed to be healthy, i.e., without a history of fever or any clinical indication of a febrile state or infectious pathology within the previous 28 days of recruitment. Among the most likely explanations, the onset in these individuals was asymptomatic or subclinical, and, thus, too marginal to be recognized. It is also possible that previous clinical data were incorrect due to recall and/or reporting bias. Asymptomatic subjects commonly lack fever or signs of infection despite evidence of bacterial species-specific nucleic acid in the bloodstream or fecal shedding [44]. Pathogen-associated factors may promote host survival by favoring limited tissue damage and the establishment of a persistent asymptomatic carrier stage via suppression of initial inflammatory responses [45]. Nevertheless, one weakness of our investigation is that NFSs did not undergo physical examinations and/or clinical appraisals by study physicians to confirm their health status. We also did not cross-check past patient records that may have been available at the recruiting healthcare facilities.

The TSAP and the Global Burden of Disease Study 2017 reported that the highest incidence rates of typhoid fever were detected among school-aged children. Global estimates revealed that 12.6% (range: 8.7–17.7%) of cases and 17.2% (range: 12.3–23.4%) of deaths occurred in children < 5 years of age; furthermore, 55.9% (range: 50.3–61.6%) of cases and 59.3% (range: 53.6–65.2%) of deaths were seen in children < 15 years of age [9,10,11, 46]. As most of the Ethiopian participants were ≥ 15 years of age in contrast to a more balanced age distribution among Ghanaian and Malagasy subjects, the true prevalence of S. Typhi at the Ethiopian sites could have been falsified. Moreover, most of the FSs from Ghana and Madagascar presented with axillary body temperatures of ≥ 37.5 °C and at in-patient and/or out-patient departments, in contrast to FSs that presented solely to out-patient departments at the Ethiopian sites; both are potential indicators of having enrolled subjects with more severe signs of infection. Likewise, participants from Kintampo were recruited at the secondary/tertiary healthcare level compared to the primary/secondary healthcare level at the Malagasy and Ethiopian sites. This indicates a likely different clientele visiting healthcare facilities to seek care. Eligibility criteria of FSs differed by country, which also impacts the direct comparability of our findings. For instance, FSs from Ghana had to have malaria parasites though most enrollees did not fulfill this criterion; FSs from Ethiopia had to have either clinically suspected/diagnosed Salmonella infection or febrile illness of unknown reasons. The recruitment period likely did not have any major impact our data as all subjects participated during the country-specific main wet season, i.e. April to July and September to November in Ghana, November to March/April in Madagascar, and June to August/September in Ethiopia [13, 47].

Lardeux et al. [27] did show that change-point analyses are strong in correctly discriminating samples compared to other cut-off equations, which can, thus, minimize chances of misclassification. Caution is advised when classifying absorbance values in close proximity to proxy cut-off points. False results are possible and should be re-confirmed with secondary independent assays. Another limitation of our investigation was that blood specimens were collected once only and in small volumes. This precluded re-testing or bacterial culturing, which is currently considered as the diagnostic gold standard despite varying sensitivity ranging between 45 and 70% [45]. In addition, antibiotic pre-treatment as reported by FSs in Diego, Mahajanga and Adama Zuria is known to impact diagnostic sensitivity though data collected may be incorrect due to recall bias [48]. The bacterial load present in each specimen could have a significant impact on the diagnostic sensitivity of both, antibody detection by ELISA and DNA amplification by RT-PCR [6]. While findings from a S. Typhi controlled human challenge model stated detection of bacterial DNA as early as 6–72 h post-exposure, early detection did not predict the likelihood of developing typhoid infection (OR: 0.57, 95%CI: 0.12–2.71, p = 0.69) or bacteremia (OR: 0.50, 95%CI: 0.10–2.44, p = 0.45) [44].


In summary, we determined the prevalence of S. Typhi using anti-HlyE/-CdtB/-pilL/-Vi-IgM ELISAs and species-specific RT-PCR at sites in Ghana, Madagascar and Ethiopia that have been proposed for implementing surveillance of febrile illness including salmonellosis. Host immune responses revealed the strongest prevalence rates of S. Typhi at the Ethiopian sites. This was particulary notable in subjects from the Addis Ababa and Adama Zuria sites, even though the school-aged children, who were presumably at the highest risk, were likely to have been missed. Of note is that literature available for instance in PubMed provides evidence of Salmonella in Mahajanga, Adama Zuria, Debre Birhan, Arba Minch and Wolayita Sodo, but not at any of the remaining sites.

When one considers the millions of infections caused by S. Typhi each year, together with the impact of recent epidemics due to Ebola virus and COVID-19, the importance of public health surveillance leading to preparedness for immediate responses to disease outbreaks and threats remains of critical importance throughout the African continent. Despite the limitations discussed, our investigation of the utility of both, antigen-specific antibody detection and species-specific nucleic acid amplification, provides useful information that might be used to assess the sites proposed for implementing surveillance of Salmonella-associated febrile illnesses as well as other infectious diseases.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. All data generated or analyzed during this study are included in this published article.



Febrile subjects


Non-febrile subjects


Enzyme-linked immunosorbent assay


Real-time polymerase chain reaction


Optical density

S. Typhi:

Salmonella Typhi


Cumulative sum


Confidence interval


Prevalence rate


Odds ratio


Typhoid Fever Surveillance in Africa Program


Severe Typhoid Fever in Africa Program


World Health Organization


International Network for the Demographic Evaluation of Populations and Their Health


  1. Maze MJ, Bassat Q, Feasy NA, Mandomando I, Musicha O, Crump JA. The epidemiology of febrile illness in sub-Saharan Africa: implications for diagnosis and management. Clin Microbiol Infect. 2018;24(8):808–14.

    Article  CAS  Google Scholar 

  2. World Health Organization. WHO informal consultation on fever management in peripheral health care settings: a global review of evidence and practice. 2013; Available from: Accessed on 14072021.

  3. Mweu E, English M. Typhoid fever in children in Africa. Trop Med Int Health. 2008;13(4):532–40.

    Article  Google Scholar 

  4. Muyembe-Tamfum JJ, Veyi J, Kaswa M, Lunguya O, Verhaegen J, Boelaert M. An outbreak of peritonitis caused by multidrug-resistant Salmonella Typhi in Kinshasa, Democratic Republic of Congo. Travel Med Infect Dis. 2009;7(1):40–3.

    Article  CAS  Google Scholar 

  5. World Health Organization. Typhoid. 2020; Available from: Accessed on 13072021.

  6. Tran Vu Thieu N, Van Trinh T, Tran Tuan A, Klemm EJ, Nguyen Ngoc Minh C, Voong Vinh P, et al. An evaluation of purified Salmonella Typhi protein antigens for the serological diagnosis of acute typhoid fever. J Infect. 2017;75(2):104–14.

    Article  Google Scholar 

  7. World Health Organization. Coronavirus. 2020; Available from: Accessed on 13072021. Accessed on 13072021.

  8. World Health Organization. Ebola virus disease. 2020; Available from: Accesed on 12062021.

  9. Marks F, von Kalckreuth V, Aaby P, Adu-Sarkodie Y, El Tayeb MA, Ali M, et al. Incidence of invasive Salmonella disease in sub-Saharan Africa: a multicentre population-based surveillance study. Lancet Glob Health. 2017;5(3):e310–23.

    Article  Google Scholar 

  10. Panzner U, Pak GD, Aaby P, Adu-Sarkodie Y, Ali M, Aseffa A, et al. Utilization of healthcare in the typhoid fever surveillance in Africa program. Clin Infect Dis. 2016;62(Suppl 1):S56-68.

    Article  Google Scholar 

  11. von Kalckreuth V, Konings F, Aaby P, Adu-Sarkodie Y, Ali M, Aseffa A, et al. The typhoid fever surveillance in Africa program (TSAP): clinical, diagnostic, and epidemiological methodologies. Clin Infect Dis. 2016;62(Suppl 1):S9–16.

    Article  Google Scholar 

  12. Park SE, Toy T, Cruz Espinoza LM, Panzner U, Mogeni DO, Im J, et al. The severe typhoid fever in africa program: study design and methodology to assess disease severity, host immunity, and carriage associated with invasive salmonellosis. Clin Infect Dis. 2019;69(Suppl 6):S422–34.

    Article  Google Scholar 

  13. Pak GD, Haselbeck AH, Seo HW, Osei I, Amuasi J, Breiman RF, et al. The HPAfrica protocol: assessment of health behaviour and population-based socioeconomic, hygiene behavioural factors—a standardised repeated cross-sectional study in multiple cohorts in sub-Saharan Africa. BMJ Open. 2018;8(12): e021438.

    Article  Google Scholar 

  14. Kannan VC, Andriamalala CN, Reynolds TA. The burden of acute disease in Mahajanga, Madagascar—a 21 month study. PLoS ONE. 2015;10(3): e0119029.

    Article  CAS  Google Scholar 

  15. Ghana Statistical Service. 2010 Population & Housing Census, District Analytical Report, Kintampo South District 2014; Available from: Accessed on 11072021.

  16. Ghana Statistical Service. 2010 Population & Housing Census, District Analytical Report, Kintampo North District. 2014; Available from: Accessed on 11072021.

  17. openAFRICA. Population growth (annual %)—Ghana. 2020; Available from: Accessed on 11072021.

  18. openAFRICA. Ethiopia data portal. Available from: Accessed on 11072021.

  19. The World Bank. Population growth (annual %)—Ethiopia 2020. Avaialble from: Accessed on 11072021.

  20. Federal Demokratic Republic of Ethiopia Population Census Commission. Summary and Statistical Report of the 2007 Population and Housing Census. 2008; Available from: Accessed on 11072021.

  21. openAFRICA. Madagascar data portal. Available from: Accessed on 11072021.

  22. Populationpyramid. Madagascar 2016. 2019; Available from: Accessed on 12072021.

  23. Liang L, Juarez S, Nga TV, Dunstan S, Nakajima-sasaki R, Davies DH, et al. Immune profiling with a Salmonella Typhi antigen microarray identifies new diagnostic biomarkers of human typhoid. Sci Rep. 2013;3:1043.

    Article  CAS  Google Scholar 

  24. Watson CH, Baker S, Lau CL, Rawalai K, Taufa M, Coriakula J, et al. A cross-sectional seroepidemiological survey of typhoid fever in Fiji. PLoS Negl Trop Dis. 2017;11(7): e0005786.

    Article  Google Scholar 

  25. Moore CE, Pan-Ngum W, Wijedoru LPM, Sona S, Nga TVT, Duy PT, et al. Evaluation of the diagnostic accuracy of a typhoid IgM flow assay for the diagnosis of typhoid fever in Cambodian children using a Bayesian latent class model assuming an imperfect gold standard. Am J Trop Med Hyg. 2014;90(1):114–20.

    Article  CAS  Google Scholar 

  26. Nga TV, Karkey A, Dongol S, Thuy HN, Dunstan S, Holt K, et al. The sensitivity of real-time PCR amplification targeting invasive Salmonella serovars in biological specimens. BMC Infect Dis. 2010;10:125.

    Article  CAS  Google Scholar 

  27. Lardeux F, Torrico G, Aliaga C. Calculation of the ELISA’s cut-off based on the change-point analysis method for detection of Trypanosoma cruzi infection in Bolivian dogs in the absence of controls. Mem Inst Oswaldo Cruz. 2016;111(8):501–4.

    Article  Google Scholar 

  28. Charles RC, Sheikh A, Krastins B, Harris JB, Bhuiyan MS, LaRoque RC, et al. Characterization of anti-Salmonella enterica serotype Typhi antibody responses in bacteremic Bangladeshi patients by an immunoaffinity proteomics-based technology. Clin Vaccine Immunol. 2010;17(8):1188–95.

    Article  CAS  Google Scholar 

  29. Hunt S, Green J, Artymiuk PG. Hemolysin E (HlyE, ClyA, SheA) and related toxins. Adv Exp Med Biol. 2010;677:116–26.

    Article  CAS  Google Scholar 

  30. Thakur R, Pathania P, Kaur N, Joshi V, Kondepudi KK, Suri CR, et al. Prophylactic potential of cytolethal distending toxin B (CdtB) subunit of typhoid toxin against Typhoid fever. Sci Rep. 2019;9(1):18404.

    Article  CAS  Google Scholar 

  31. von Rhein C, Hunfeld KP, Ludwig A. Serologic evidence for effective production of cytolysin A in Salmonella enterica serovars typhi and paratyphi A during human infection. Infect Immun. 2006;74(11):6505–8.

    Article  CAS  Google Scholar 

  32. Charles RC, Liang L, Khanam F, Sayeed MA, Hung C, Leung DT, et al. Immunoproteomic analysis of antibody in lymphocyte supernatant in patients with typhoid fever in Bangladesh. Clin Vaccine Immunol. 2014;21(3):280–5.

    Article  CAS  Google Scholar 

  33. Song J, Gao X, Galán JE. Structure and function of the Salmonella Typhi chimaeric A^sub 2^B^sub 5^ typhoid toxin. Nature. 2013;499(7458):350–4.

    Article  CAS  Google Scholar 

  34. Miller RA, Wiedmann M. The cytolethal distending toxin produced by nontyphoidal Salmonella serotypes javiana, montevideo, oranienburg, and mississippi induces DNA damage in a manner similar to that of serotype typhi. mBio. 2016; 7(6).

  35. Rehman T, Yin L, Latif MB, Chen J, Wang K, Gen Y, et al. Adhesive mechanism of different Salmonella fimbrial adhesins. Microb Pathog. 2019;137: 103748.

    Article  CAS  Google Scholar 

  36. Wagner C, Hensel M. Adhesive mechanisms of Salmonella enterica. Adv Exp Med Biol. 2011;715:17–34.

    Article  CAS  Google Scholar 

  37. House D, Ho VA, Diep TS, Chinh NT, Bay PV, Vinh H, et al. Antibodies to the Vi capsule of Salmonella Typhi in the serum of typhoid patients and healthy control subjects from a typhoid endemic region. J Infect Dev Ctries. 2008;2(4):308–12.

    PubMed  Google Scholar 

  38. Jin C, Hill J, Gunn BM, Yu WH, Dahora LC, Jones E, et al. Vi-specific serological correlates of protection for typhoid fever. J Exp Med, 2021; 218(2).

  39. Hornick RB, Greisman SE, Woodward TE, DuPont HL, Dawkins AT, Snyder MJ. Typhoid fever: pathogenesis and immunologic control. N Engl J Med. 1970;283(13):686–91.

    Article  CAS  Google Scholar 

  40. Seyfer AE, Mologne LA, Morris RL, Clark JR. Endometriosis causing acute small bowel obstruction: report of a case and review of the literature. Am Surg. 1975;41(3):168–71.

    PubMed  CAS  Google Scholar 

  41. Keddy KH, Klugman KP, Hansford CF, Blondeau G, Bouveret le Cam NN. Persistence of antibodies to the Salmonella typhi Vi capsular polysaccharide vaccine in South African school children ten years after immunization. Vaccine. 1999;17(2):110–3.

    Article  CAS  Google Scholar 

  42. Panchanathan V, Kumar S, Yeap W, Devi S, Ismail R, Sarijan S, et al. Comparison of safety and immunogenicity of a Vi polysaccharide typhoid vaccine with a whole-cell killed vaccine in Malaysian Air Force recruits. Bull World Health Organ. 2001;79(9):811–7.

    PubMed  PubMed Central  CAS  Google Scholar 

  43. Klugman KP, Koomhof HJ, Robbins JB, Le Cam NN. Immunogenicity, efficacy and serological correlate of protection of Salmonella typhi Vi capsular polysaccharide vaccine three years after immunization. Vaccine. 1996;14(5):435–8.

    Article  CAS  Google Scholar 

  44. Darton TC, Zhou L, Blohmke CJ, Jones C, Waddington CS, Baker S, et al. Blood culture-PCR to optimise typhoid fever diagnosis after controlled human infection identifies frequent asymptomatic cases and evidence of primary bacteraemia. J Infect. 2017;74(4):358–66.

    Article  Google Scholar 

  45. Del Bel BL, Guidi R, Pateras IS, Levi L, Mihaljevic B, Rouf SF, et al. The typhoid toxin promotes host survival and the establishment of a persistent asymptomatic infection. PLoS Pathog. 2016;12(4): e1005528.

    Article  CAS  Google Scholar 

  46. Global Burden Typhoid and Paratyphoid collaborators. The global burden of typhoid and paratyphoid fevers: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Infect Dis. 2019;19(4):369–81.

    Article  Google Scholar 

  47. Teferi M, Desta M, Yeshitela B, Beyene T, Cruz Espinoza LM, Im J, et al. Acure febrile illness among children in Butajira, South-Central Ethiopia during the typhoid fever surveillance in Africa Program. CID, 2019.

  48. Chin KL, Redhuan NE, Balaram P, Phua KK, Ong EB. Detection of salivary IgA antibodies against the HlyE antigen as a diagnosis of typhoid fever. J Clin Diagn Res. 2016;10(6):DM01-3.

    PubMed  PubMed Central  CAS  Google Scholar 

Download references


We are very grateful to residents in Kintampo, Ghana, Diego and Mahajanga, Madagascar, and Addis Ababa, Adama-Wenji, Debre Birhan, Arba Minch and Wolayita Sodo, Ethiopia for their participation in this study. Likewise, we are grateful to the healthcare personnel at the healthcare centers and hospitals in each participating site for their exceptional efforts and contributions, which made this research possible. We thank Melaku Yedenekachew, Yemesrach Seralegn, Gebeyehu Aseffa, Emawayish Andargie, Oumer Ali and Tigist Beyene from the Armauer Hansen Research Institute/All Africa Leprosy, Tuberculosis and Rehabilitation Training Center, Ethiopia for their dedicated contribution to this study. Our appreciation goes to Seth Owusu-Agyei, Kwaku Poku Asante, Seyram Kaali, David Dosoo, Rupert Delimini, and Mieks Twumasi from the Kintampo Health Research Center, Gavin Apio from the Kintampo Municipal Hospital, and Stephen Apanga from the Yezura Hospital for their immense contributions to this study.


This research was funded by the Bill & Melinda Gates Foundation (OPP1127988). The International Vaccine Institute acknowledges its donors including the Republic of Korea and the Swedish International Development Cooperation Agency.

Author information

Authors and Affiliations



FM, TT, SEP, UP, ODM, HJJ, YAS, EOD, YE, AA, MT, BY, RR conducted this SETA-adjunct study in the three countries; SB, TVT developed the assays and performed laboratory analyses; GDP performed the data management; UP performed the scientific analyses. UP, ODM wrote the initial manuscript. All authors provided critical feedback leading to the final manuscript version. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Florian Marks.

Ethics declarations

Ethics approval and consent to participate

We received ethical clearance from committees of the International Vaccine Institute (approval number: 2015-004), South Korea, the Kintampo Health Research Centre (approval number: 2016-1), Ghana, the Ministère de la Santé du République (approval number: 109MSANP/CE), Madagascar, and the Armauer Hansen Research Institute/All Africa Leprosy, Tuberculosis, and Rehabilitation Training Center (approval number: PO38/15) and the Ethiopian National Research Ethics Review Committee (approval number: 3.10/116/2016), Ethiopia. Written informed consent or fingerprint with witness signature in case of illiteracy was obtained voluntarily from each eligible participant or parent/guardian/legal representative in case of minors after the purpose of our investigation was explained. In addition, informed assent was completed for the age ranges ≥ 10 to ≤ 17 years in Ghana, ≥ 12 to ≤ 18 years in Madagascar, and ≥ 12 to ≤ 17 years in Ethiopia as adviced by site principal investigators. The study protocol was performed in accordance with the relevant gudielines including the declaration of Helsinki. We processed all information on enrollees anonymously. Records were maintained confidential with only restricted access.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1.

Location of sites in Ethiopia [1A], Ghana [1B] and Madagascar [1C]. Notes: The location of each site is indicated as a black dot and the site’s name in red font; Ethiopia (Figure 1A): nine regional states (black italic, capital letters) and two chartered cities (black italic, lower letters) are shown;SNNPR=Southern Nations, Nationalities and People’s Region; Ghana (Figure 1B): ten regions (black italic, capital letters) are shown; Madagascar (Figure 1C): six provinces (black italic, capital letters) are shown.

Additional file 2.

Distribution of ELISA OD values by antigen.

Additional file 3.

Mean ELISA ODs by antigen. Notes: OD=optical density; red dot=mean of original data; blue dot=mean of each 1,000 bootstrap random samples; light red dashed line=upper and lower 95% confidence interval (CI) of original data; dark red dashed line=upper and lower 99% confidence interval (CI) of original data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Panzner, U., Mogeni, O.D., Adu-Sarkodie, Y. et al. Detection of Salmonella Typhi nucleic acid by RT-PCR and anti-HlyE, -CdtB, -PilL, and -Vi IgM by ELISA at sites in Ghana, Madagascar and Ethiopia. BMC Infect Dis 22, 766 (2022).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


  • HlyE-/CdtB-/PilL-/Vi-antigen
  • Salmonella
  • Febrile illness
  • Surveillance