You're using an outdated browser. Please upgrade to a modern browser for the best experience.
Enteric Fever Diagnosis: Comparison
Please note this is a comparison between Version 1 by Jeongmin Song and Version 2 by Camila Xu.

Enteric fever, referring to typhoid fever and paratyphoid fever, is a common bacterial disease with high morbidity and mortality rates in low- to middle-income countries in Asia, Africa, and South America, associated with limited proper sanitation and safe drinking water supply.

  • enteric fever diagnosis
  • typhoid fever diagnosis
  • Salmonella

1. Introduction

Enteric fever, referring to typhoid fever and paratyphoid fever, is a common bacterial disease with high morbidity and mortality rates in low- to middle-income countries in Asia, Africa, and South America, associated with limited proper sanitation and safe drinking water supply [1][2]. The World Health Organization (WHO) estimates up to 21 million enteric fever cases and 161,000 deaths each year worldwide. However, the actual burden of the disease is unknown since this estimate was extrapolated from the limited number of surveillance studies using current diagnostic measures [3]. Among over 2600 closely-related

Enteric fever, referring to typhoid fever and paratyphoid fever, is a common bacterial disease with high morbidity and mortality rates in low- to middle-income countries in Asia, Africa, and South America, associated with limited proper sanitation and safe drinking water supply [1,2]. The World Health Organization (WHO) estimates up to 21 million enteric fever cases and 161,000 deaths each year worldwide. However, the actual burden of the disease is unknown since this estimate was extrapolated from the limited number of surveillance studies using current diagnostic measures [3]. Among over 2600 closely-related

Salmonella enterica

serovars, human-restricted

Salmonella enterica

serovars Typhi and Paratyphi A, B, and C (

S.

Typhi and

S.

Paratyphi A, B, and C) are the cause of enteric fever. Different

Salmonella

serovars, including

S.

Typhi and

S.

Paratyphi, are characterized by a distinct set of their surface antigens: lipopolysaccharide O (somatic), flagellar H, and virulence-capsule (Vi) antigens [4]. Based on their host-specificity and disease outcomes,

S. enterica

are grouped into typhoidal and nontyphoidal

Salmonella

serovars (NTS). The majority of NTS serovars represented by

S.

Typhimurium and

S. Enteritidis can infect humans and animals and cause a self-limiting gastrointestinal Salmonellosis in humans, with some exceptions of NTS causing invasive disease [5][6][7].

Enteritidis can infect humans and animals and cause a self-limiting gastrointestinal Salmonellosis in humans, with some exceptions of NTS causing invasive disease [5,6,7].

In addition to diagnostic challenges associated with closely related

Salmonella

serovars, the infection route and some clinical presentations are also shared among

Salmonella

serovars.

Salmonella serovars are transmitted through the fecal–oral route after the ingestion of contaminated food and water. The incubation period of enteric fever is approximately 8–14 days [8], while the duration and severity of the disease are affected by the types of bacterial strains and doses as well as host immune responses [9][10][11]. Typhoid and paratyphoid fevers are clinically indistinguishable from each other. They can present with comparable severity of the complications, although typhoid fever is more prevalent than paratyphoid fever in most endemic areas [12][13]. For instance, clinical presentations such as high fever, headache, malaise, anorexia, rapid pulse, leukopenia, thrombocytopenia, abdominal discomfort, and neurological complications are not specific to enteric fever [14][15], making a clinical-presentation-based diagnosis difficult. Viral (e.g., dengue, influenza), parasitic (e.g., malaria, typhus, leishmaniosis), and other bacterial (e.g., brucellosis, tuberculosis) infections that are also common in endemic areas may develop similar symptoms [16]. The current diagnostic tests cannot reliably distinguish enteric fever from others.

serovars are transmitted through the fecal–oral route after the ingestion of contaminated food and water. The incubation period of enteric fever is approximately 8–14 days [8], while the duration and severity of the disease are affected by the types of bacterial strains and doses as well as host immune responses [9,10,11]. Typhoid and paratyphoid fevers are clinically indistinguishable from each other. They can present with comparable severity of the complications, although typhoid fever is more prevalent than paratyphoid fever in most endemic areas [12,13]. For instance, clinical presentations such as high fever, headache, malaise, anorexia, rapid pulse, leukopenia, thrombocytopenia, abdominal discomfort, and neurological complications are not specific to enteric fever [14,15], making a clinical-presentation-based diagnosis difficult. Viral (e.g., dengue, influenza), parasitic (e.g., malaria, typhus, leishmaniosis), and other bacterial (e.g., brucellosis, tuberculosis) infections that are also common in endemic areas may develop similar symptoms [16]. The current diagnostic tests cannot reliably distinguish enteric fever from others.

The global spread of multidrug-resistant (MDR)

Salmonella

and the emergence of extensively drug-resistant (XDR)

Salmonella

also support the need for improved diagnostic tests, as well as new treatment strategies that are alternatives to current antibiotics. Antibiotics are primary treatment options for enteric fever, but

Salmonella

is continuously evolving to acquire plasmid, prophage, transposon, or chromosomal gene mutations to attain antibiotic-resistance. A myriad of reports has indicated the global spread of

S.

Typhi and

S.

Paratyphi strains that are resistant to all of the first-line antibiotics, ampicillin, chloramphenicol, and co-trimoxazole, collectively known as multidrug-resistance (MDR)

Salmonella [17][18][19]. All of the identified MDR

[17,18,19]. All of the identified MDR

S.

Typhi and

S.

Paratyphi carry the IncHI1 plasmid, while other antibiotic-resistant related genes found in MDR

Salmonella

can vary [20]. Haplotype-58 (H58) is the most dominant MDR

S. Typhi strain identified in various parts of Asia and Africa and travel-related MDR cases in other countries [21][22][23][24][25].

Typhi strain identified in various parts of Asia and Africa and travel-related MDR cases in other countries [21,22,23,24,25].

The emergence and spread of

Salmonella strains resistant to the second line of drugs have also been reported [25][26]. Resistance to fluoroquinolones has been acquired by chromosomal mutations in the quinolone resistance gene

strains resistant to the second line of drugs have also been reported [25,26]. Resistance to fluoroquinolones has been acquired by chromosomal mutations in the quinolone resistance gene

qnrS

and/or quinolone resistance determining region (QRDR) harboring

gyrA, gyrB, parC,

and

parE genes [25][26]. Resistance to third-generation cephalosporins is associated with the acquisition of several extended-spectrum β-lactamase (ESBL) genes [27]. The XDR H58

genes [25,26]. Resistance to third-generation cephalosporins is associated with the acquisition of several extended-spectrum β-lactamase (ESBL) genes [27]. The XDR H58

S.

Typhi strain, resistant to ampicillin, chloramphenicol, co-trimoxazole, fluoroquinolones, and third-generation cephalosporins, was first identified in Pakistan, affecting over 300 cases in 2016 [27]. Since then, XDR

S.

Typhi infection remains prevalent in the region, and travel-related XDR

S. Typhi infections have been reported in many other countries [28][29][30][31], indicating the rapid global spread of XDR

Typhi infections have been reported in many other countries [28,29,30,31], indicating the rapid global spread of XDR

S.

Typhi. XDR H58 isolates harbored the IncY plasmid, carrying an ESBL-resistance gene. Azithromycin and carbapenems are “last resort” antibiotics for treating

Salmonella

infection, but the emergence of azithromycin-resistant

S. Typhi strains and carbapenem-resistant invasive NTS has also been reported [32][33][34][35]. These observations support the urgent need of improved diagnostic, prevention, and treatment strategies to better control drug-resistant

Typhi strains and carbapenem-resistant invasive NTS has also been reported [32,33,34,35]. These observations support the urgent need of improved diagnostic, prevention, and treatment strategies to better control drug-resistant

S.

Typhi and

S.

Paratyphi.

Ideal diagnostic tests should also detect asymptomatic carriers and distinguish the infection from others. A significant population (2–5%) of recovered patients become asymptomatic chronic carriers who can shed the bacteria intermittently in their feces for years [36][37]. Chronic carriers serve as a primary reservoir of

Ideal diagnostic tests should also detect asymptomatic carriers and distinguish the infection from others. A significant population (2–5%) of recovered patients become asymptomatic chronic carriers who can shed the bacteria intermittently in their feces for years [36,37]. Chronic carriers serve as a primary reservoir of

S.

Typhi and

S. Paratyphi that persist mainly in the gallbladder for local and global spread [37][38][39], as they are human-restricted pathogens with no other known reservoirs.

Paratyphi that persist mainly in the gallbladder for local and global spread [37,38,39], as they are human-restricted pathogens with no other known reservoirs.

2. Current Enteric Fever Diagnostics

Laboratory diagnosis is required to confirm enteric fever. Although enteric fever has been well established for more than a century, there has not been a single “ideal” laboratory diagnostic biomarker available (

Table 1

).

Table 1.

Diagnostic tests for acute enteric-fever-suspected patients (fever ≥38 °C for ≥3 days).

[58][59][60][61][62][63][64].

), and other conserved genes. Due to the lack of a standard reference method, the accuracy of the test is generally calculated based on blood culture results. Various studies have demonstrated that the sensitivity ranges from 40–100%, as shown by different studies, while the specificity can be near 100% if conducted under optimal conditions [56,57,58,59,60,61,62,63,64].
Various platforms, from conventional PCR to quantitative real-time PCR (qRT-PCR), nested PCR, multiplex PCR, and loop-mediated isothermal amplification (LAMP) PCR, have been reported to demonstrate variable sensitivity. Still, none of them are free from limitations. Removal of background human DNA from blood specimens [65] and a brief culture of blood samples before the PCR reaction (dubbed blood culture PCR [66]) showed an increased sensitivity by several folds. In summary, PCR-based methods are relatively simpler, faster, and more cost-effective than their culture-based counterparts. However, disease detection sensitivity remains an issue to serving as an optimal assay.

2.3. Serological Diagnosis

Serological identification of

S. enterica

serovars relies on Kauffman–White classification. Currently available serological tests cannot reliably diagnose enteric fever (specificity is not 100%) as many of the antigens are shared among different

Salmonella

serovars. The major antigens used to differentiate

S.

Typhi and

S.

Paratyphi are often restricted to Vi, lipopolysaccharides (LPS) O, and flagellar H antigens, yet some of the antigens are shared among different

Salmonella

serovars (

Table 2

). Therefore, unlike blood-culture-mediated diagnostic methods, positive results from serological diagnostic tests are suggestive of enteric fever (

Table 1

and

Table 2

). However, serological tests are simple and quick, which is highly valuable for managing disease in impoverished endemic areas in a timely manner.

Table 2.

Serological identification of

Salmonella

.

2.1. Bacterial-Culture-Based Diagnosis

The definitive diagnosis of enteric fever requires the isolation of bacteria from blood or bone marrow, accompanied by fever ≥38 °C for at least three days [40]. Culture remains the mainstay of diagnosis, and bacterial isolation allows us to characterize the pathogen for antibiotic resistance genes and the causation of the outbreak of disease in the particular location. Although the method has 100% specificity, it lacks sensitivity. On average, blood and bone marrow cultures have a sensitivity of ~50% and ~80%, respectively, which directly correlates with the number of viable bacteria in blood (≤1 CFU/mL) and bone marrow (~10 CFU/mL) [41][42]. Various strategies have been employed to increase the sensitivity of bacterial-culture-mediated diagnosis (

The definitive diagnosis of enteric fever requires the isolation of bacteria from blood or bone marrow, accompanied by fever ≥38 °C for at least three days [40]. Culture remains the mainstay of diagnosis, and bacterial isolation allows us to characterize the pathogen for antibiotic resistance genes and the causation of the outbreak of disease in the particular location. Although the method has 100% specificity, it lacks sensitivity. On average, blood and bone marrow cultures have a sensitivity of ~50% and ~80%, respectively, which directly correlates with the number of viable bacteria in blood (≤1 CFU/mL) and bone marrow (~10 CFU/mL) [41,42]. Various strategies have been employed to increase the sensitivity of bacterial-culture-mediated diagnosis (

Table 1

). For example, supplementation of ox bile or bile salt (sodium taurocholate) to the culture media has resulted in increased bacterial isolation frequency in a shorter time [43]. More specifically, bile contents suppress the bactericidal activity of blood and lyse blood cells to release bacteria.

Similarly, sodium polyethanol sulfonate can also reduce the bactericidal activity of blood [44] and shorten the testing time required for bacteria isolation without changing the overall isolation frequency [45]. After removing serum due to its bactericidal activity, blood clot culture also exhibits increased sensitivity and rapid bacterial growth [46][47][48]. More blood volumes and additional dilutions of the specimens with media have also been used to improve bacterial detection frequency and address the sensitivity issue associated with culture-based diagnosis methods [49].

Similarly, sodium polyethanol sulfonate can also reduce the bactericidal activity of blood [44] and shorten the testing time required for bacteria isolation without changing the overall isolation frequency [45]. After removing serum due to its bactericidal activity, blood clot culture also exhibits increased sensitivity and rapid bacterial growth [46,47,48]. More blood volumes and additional dilutions of the specimens with media have also been used to improve bacterial detection frequency and address the sensitivity issue associated with culture-based diagnosis methods [49].
These studies indicate that the optimum ratio of blood to bacterial culture media (e.g., tryptic soy broth (TSB)) should be 1:10 or greater. The standard method involves an incubation at 37 °C and an inspection for bacterial growth for at least a week. In general, positive cultures are subcultured at 37 °C for 24 h on both nonselective enriched media (e.g., blood agar) that supports the growth of most bacteria and selective differential media (e.g., MacConkey agar, xylose lysine deoxycholate (XLD) agar) that allows the growth of bile-tolerant bacteria such as

S.

Typhi and

S.

Paratyphi for diagnosis. The use of nonselective media such as TSB or blood agar helps isolate bacterial pathogens in blood, which should be sterile in healthy individuals. The use of selective media such as MacConkey and XLD agars helps differentiate non-lactose-fermenting bacteria such as

S.

Typhi and

S.

Paratyphi from lactose-fermenting bacteria such as

E. coli

and bile-tolerant

S.

Typhi and

S.

Paratyphi from other pathogens such as Gram-positive bacteria and

E. coli

, respectively. Biochemical identification and agglutination with specific antisera tests are followed to diagnose infection with

S.

Typhi and/or

S.

Paratyphi. The basis of serotyping is described in the latter part of this paper (

Section 2.3

).

There are some additional challenges associated with blood-culture-based diagnostic methods. In brief, compared to the use of 5–10 mL blood samples for school-age children and adults during the first two weeks of the infection, at which the bacterial load is higher [41][50], a smaller blood volume (2–4 mL) is used for preschool children [40], which is likely associated with underdiagnosis among younger populations [51]. Prior antibiotic therapy, which remains very common in endemic areas, also hinders culture-based diagnosis [52][53]. This challenge is overcome by using bone marrow samples rather than blood for bacterial culture since bacteria in the bone marrow are unlikely cleared by antibiotic treatment [52][54]. For this reason, bone marrow culture is generally considered the gold standard for enteric fever diagnosis in endemic areas. However, this method involves an invasive procedure for sample collection and requires specialized skills and equipment to conduct.

There are some additional challenges associated with blood-culture-based diagnostic methods. In brief, compared to the use of 5–10 mL blood samples for school-age children and adults during the first two weeks of the infection, at which the bacterial load is higher [41,50], a smaller blood volume (2–4 mL) is used for preschool children [40], which is likely associated with underdiagnosis among younger populations [51]. Prior antibiotic therapy, which remains very common in endemic areas, also hinders culture-based diagnosis [52,53]. This challenge is overcome by using bone marrow samples rather than blood for bacterial culture since bacteria in the bone marrow are unlikely cleared by antibiotic treatment [52,54]. For this reason, bone marrow culture is generally considered the gold standard for enteric fever diagnosis in endemic areas. However, this method involves an invasive procedure for sample collection and requires specialized skills and equipment to conduct.
Besides bacterial isolation from blood and bone marrow samples, in some cases, other biological samples such as rose spot, duodenal bile, stool, and urine are used for

Salmonella

isolations via culture. A rose spot culture gives ~60% sensitivity, which is a noninvasive procedure, but the occurrence of these spots is relatively rare among enteric fever patients (1–30%) [42]. Duodenal aspirate culture can provide a better diagnostic value than stool culture, but the test’s tolerance, particularly among children, hampers its use [55]. The positive results from these other biological samples are only suggestive of active disease due to chronic carriage prevalence in endemic areas. Therefore, a positive result should be interpreted in combination with other assays.
As described above, a bacterial-culture-based diagnosis is the gold-standard for enteric fever diagnosis, also allowing for antibiotic-susceptibility testing that is essential for determining a proper antibiotic treatment strategy. The primary challenges of this method include a slower turnaround time (≥48 h) required for bacterial growth and identification and the need for appropriate laboratory infrastructure, which is not necessarily common in endemic areas.

2.2. Bacterial Nucleic Acid Detection-Based Diagnosis

Nucleic acid detection involves polymerase chain reaction (PCR) that amplifies

Salmonella

serovar-specific DNA for diagnosis. The primary advantage of this method is the rapid turnaround time. PCR methods are advantageous because they can detect

Salmonella

-specific DNA extracted from live or dead bacteria or both. Dead bacteria in blood can result from antibiotic treatments, which is also common in some endemic areas and/or outcomes of host immune responses. The disadvantage of PCR methods includes the need for trained personnel and special equipment to conduct the PCR.
This method involves DNA extraction from patient samples followed by amplification of

Salmonella

-specific DNA sequences. The most commonly used target genes for enteric fever diagnosis include flagellin (

fliC

), Vi polysaccharide (

viaB

), 16s rRNA, heat-shock protein (

groEL

), cytotoxin (

clyA), and other conserved genes. Due to the lack of a standard reference method, the accuracy of the test is generally calculated based on blood culture results. Various studies have demonstrated that the sensitivity ranges from 40–100%, as shown by different studies, while the specificity can be near 100% if conducted under optimal conditions [56][57]

Vi antigen is a linear polymer of α-1,4-2-deoxy-2-N-acetylgalacturonic acid [68]. The genes involved in the expression regulation (

tviA

), synthesis (

tviBCDE

), and transport and localization (

vexABCDE

) of Vi polysaccharide are located in the

viaB

locus as part of

Salmonella

Pathogenicity Island 7 (SPI 7) [69]. The synthesis of Vi polysaccharides is also regulated by a global regulator system

rcsABC

located in the

viaA

locus [70]. Vi polysaccharides are exclusively present in

S.

Typhi and

S.

Paratyphi C (as well as

S.

Dublin and

Citrobacter freundii

) while absent from

S.

Paratyphi A and B and NTS [71]. Vi-antigen-based agglutinations have two major limitations, associated with the recent emergence of Vi-negative strains of

S.

Typhi [72] and the requirement of certain environmental cues for Vi antigens to be expressed by Vi-positive

S. Typhi strains (e.g., higher osmolality) [73][74].

Typhi strains (e.g., higher osmolality) [73,74].

Somatic O-antigen, a portion of LPS, is present on the outer surface of Gram-negative bacteria.

Salmonella

strains fall into 46 O serogroups that differ in types of sugars, their arrangements, and the linkage within and between repeated O-antigen units, contributing to one of the most variable cell constituents, encoded by highly polymorphic

rfb genes [75], thus providing the basis for serotyping schemes [76][77]. Vi antigen expressed on the bacterial cell surface can interfere with O-antigen-mediated agglutination, which can be overcome by boiling the bacteria culture for 10 min, a procedure that removes heat-labile Vi but not heat-stable O-antigen [78].

genes [75], thus providing the basis for serotyping schemes [76,77]. Vi antigen expressed on the bacterial cell surface can interfere with O-antigen-mediated agglutination, which can be overcome by boiling the bacteria culture for 10 min, a procedure that removes heat-labile Vi but not heat-stable O-antigen [78].

Flagella are present on the cell surface of some bacteria, which facilitate bacterial locomotion. Flagellin protein is the main component of the extracellular flagellar filament that is expressed by one of two genes, H1 (

fliC

) and H2 (

fljB

), one at a time, in

Salmonella

, known as phase variation [79]. The major types of flagellar H-antigens present in typhoidal

Salmonella

are shown in

Table 2

.

S. Typhi primarily consists of monophasic H:d antigen; however other variants, H:j or H:z66, have also been reported [80][81].

Typhi primarily consists of monophasic H:d antigen; however other variants, H:j or H:z66, have also been reported [80,81].

The most commonly used serological assay in the endemic setting is the Widal test, which measures the agglutination of bacterial O and H antigens with antisera specific for these antigens [82][83]. This test should be performed twice to improve test accuracy: once during the acute phase and the other during the convalescent phase of the infection, which can be approximately 10 days apart. The test result is considered positive if there is a four-fold increase in antibody titers between the two tests [84]. However, due to the unique circumstances posed in endemic areas, a single Widal test is widely used in the field, especially during the early phase of acute infection [85][86]. The interpretation of a single Widal test is complicated by various background antibodies in people of different endemic areas, necessitating determining the cutoff values of antibodies level for determining a positive result [87]. Therefore, when optimum cutoff values tailored for the particular endemic regions are implemented, the specificity and sensitivity of the Widal test can be significantly improved and is better than most of the available rapid diagnostic tests (RDTs) such as Tubex and Typhidot [88]. However, caution should be taken as some other bacteria, as listed above, also express O- and/or H-antigens, which can result in false-positive results [89].

The most commonly used serological assay in the endemic setting is the Widal test, which measures the agglutination of bacterial O and H antigens with antisera specific for these antigens [82,83]. This test should be performed twice to improve test accuracy: once during the acute phase and the other during the convalescent phase of the infection, which can be approximately 10 days apart. The test result is considered positive if there is a four-fold increase in antibody titers between the two tests [84]. However, due to the unique circumstances posed in endemic areas, a single Widal test is widely used in the field, especially during the early phase of acute infection [85,86]. The interpretation of a single Widal test is complicated by various background antibodies in people of different endemic areas, necessitating determining the cutoff values of antibodies level for determining a positive result [87]. Therefore, when optimum cutoff values tailored for the particular endemic regions are implemented, the specificity and sensitivity of the Widal test can be significantly improved and is better than most of the available rapid diagnostic tests (RDTs) such as Tubex and Typhidot [88]. However, caution should be taken as some other bacteria, as listed above, also express O- and/or H-antigens, which can result in false-positive results [89].

Several RDTs evaluating the presence of enteric-fever-specific immunodominant antigens have been developed to meet the speedy diagnosis requirement in endemic areas [90]. The most commonly used RDTs are Tubex and Typhidot. Tubex detects anti-O9 IgM antibodies in

S.

Typhi [91] and anti-O2 antibodies in

S.

Paratyphi [92]. Typhidot detects IgM and IgG antibodies against the 50-kDa outer membrane protein of

S. Typhi [93][94]. IgM detection is the most suitable marker for diagnosing acute infection among people who have not previously been infected with

Typhi [93,94]. IgM detection is the most suitable marker for diagnosing acute infection among people who have not previously been infected with

S.

Typhi and

S.

Paratyphi and have not been vaccinated with the Ty21a live attenuated vaccine. IgG detection suggests reinfection in convalescent patients, infection in vaccinated people, or asymptomatic carriers. These tests showed 80–90% specificity and 70–80% sensitivity [90], supporting a possibility of more extensive use of these tests in clinical diagnostic laboratories in endemic regions where a rapid point-of-care (POC)-compatible test is desired for timely management of the disease.

Academic Video Service