The Sera Of Healthy Blood Donors

Published Date: 02 Nov 2017

A survey of literature revealed that among infectious diseases, typhoid fever is widespread in Kenya. Widal test is still widely used in the diagnosis of this disease. Interpretation of Widal test depends on knowledge of the baseline titres among healthy population. These titres can be used as cut-off value to arrive at diagnostic titres. Since the level of normal agglutinins for these organisms varies in different countries and communities. This variation depends on the degree to which the typhoid is endemic in each area, a factor which may change over time. It has been more than a decade since the Widal titres in apparently healthy population have been assessed. Furthermore, no baseline Widal titres have been assessed in blood donors.

Objective

To determine the distribution of Salmonella enteric serovar typhi antibodies among healthy blood donors.

Study Design

Cross-sectional descriptive survey.

Methodology

Participants in this cross-sectional analysis were voluntary but apparently healthy blood donors at the KNH Blood Transfusion Unit. Consecutive sampling was applied. Consented participants’ demographic data and focused medical history was collected by use of structured pre-tested questionnaires and blood samples were taken for assessment of Widal titres. Data was presented in the form of tables, graphs and histograms. Descriptive statistics such as means, medians, proportions and standard deviation was determined where applicable.

Results

Among the 353 serum specimens which were tested, 75 (21.25%) sera were found to be positive for the Widal test and 278 were negative. The most frequently recorded titer of the reactive sera was 1:40 for both the anti-O antibodies and the anti-H antibodies.

Conclusion

This study confirms that many of our blood donors may be Salmonella carriers.

1.0 LITERATURE REVIEW

1.1 INTRODUCTION

Typhoid fever is caused by Salmonella enterica serovar typhi (S. typhi). It is a disease transmitted by the faecal-oral route. It is a public health problem in many countries in sub-Saharan Africa, with a total of 400,000 cases occurring annually in Africa. This approximates to an incidence of 50 per 100,000 persons per year. In Africa typhoid is more common in Nigeria, Mali, Ethiopia and Kenya.1

The diagnosis of typhoid fever on clinical grounds is difficult as the presenting symptoms are diverse and similar to those observed with other febrile illness.

The gold standard for diagnosis of typhoid fever is isolation of S. typhi from bone marrow, blood, stool, or urine. However, isolation of organism is often jeopardized by lack of facilities or inadequate and or improper use of antibiotics prior to culture. Furthermore doing culture is time consuming and expensive.2

For these reasons, laboratory diagnosis of enteric fever relies heavily on serological tests such as the Widal test.2 The ELISA test has also been used but the test is too expensive and also time consuming compared to the Widal test.3

In endemic region, such as Kenya, interpretation of Widal test depends on knowledge of the baseline titres among healthy population. These titres can be used as a cut off value to arrive at diagnostic titres.2 The levels of titres detectable in normal population of different areas vary considerably. This variation depends on the degree to which the typhoid is endemic in each area, a fact which may change over time.2

1.2 THE BACTERIUM

Salmonella is a Gram-negative facultative anaerobic rod-shaped bacillus in the same proteo-bacterial family as Escherichia coli, the family Enterobacteriaceae.  Salmonella live in the intestinal tracts of warm and cold blooded animals. Some species are ubiquitous. Other species are specifically adapted to a particular host. In humans, Salmonella causes two diseases, namely, enteric fever (typhoid), resulting from bacterial invasion of the bloodstream, and acute gastroenteritis, resulting from a food-borne infection or intoxication.

Salmonella species are composed of seven subspecies:

enterica I

salamae II

arizonae IIIa

adiarizonae IIIb

bhoutenae IV

bongori V

indica VI

Each subspecies contains various serovars defined by a characteristic antigenic formula.

Subspecies I strains represents more than 99.5% of the Salmonella strains isolated from humans and other warm-blooded animals.4

1.2.1 Antigenic Structure

As with all Enterobacteriaceae, the genus Salmonella has three kinds of major antigens with diagnostic or identifying applications: somatic, surface, and flagella.

1.2.1.1 Somatic (O) or Cell Wall Antigens

Somatic antigens are heat stable and alcohol resistant. They are used for serological identification. Serovars labeled with the same ‘O’ factors are closely related, although not always antigenically identical.

1.2.1.2 Surface (Envelope) Antigens

Surface antigens, commonly observed in other genera of enteric bacteria (e.g. Escherichia coli and Klebsiella), and may be found in some Salmonella serovars. One specific surface antigen is the Vi antigen. The Vi antigen occurs in only three Salmonella serovars (out of about 2,200): Typhi, Paratyphi C, and Dublin.

1.2.1.3 Flagella (H) Antigens

Flagella antigens are heat-labile proteins. A few Salmonella enterica serovars (Enteritidis, Typhi) produce flagella which always have the same antigenic specificity.  Such an H antigen is called monophasic. Other Salmonella serovars, however, can alternatively produce flagella with two different H antigenic specificities. The H antigen is then called diphasic. For example, Typhimurium can produce flagella with either antigen i or antigen 1,2.4

S. enteric serotype typhi is serologically positive for lipopolysaccharide antigen O9 and O12, protein flagella antigen Hd and polysaccharide capsular antigen Vi.5 A unique flagella type, Hj, is present in some S. Enteric serotypes isolates from Indonesia.6

1.2.2 Habitats

The principal habitat of the Salmonella is the intestinal tract of humans and animals. Salmonella serovars can be found predominantly in one particular host or can be ubiquitous. Typhi and Paratyphi A are strictly human serovars that may cause grave diseases often associated with invasion of the bloodstream. Salmonellosis in these cases is transmitted through faecal contamination of water or food.4

Ubiquitous Salmonella serovars (Typhimurium) cause very diverse clinical symptoms, ranging from asymptomatic infection to serious typhoid-like syndromes in infants. In human adults, ubiquitous Salmonella organisms are mostly responsible for food-borne toxic infections.

1.2.2.1 Salmonella in the Natural Environment

Salmonellae are disseminated in the natural environment (water, soil or plants used as food) through human or animal excretion. Humans and animals (either wild or domesticated) can excrete Salmonella either when clinically diseased or after having had salmonellosis (carriers). Salmonella organisms do not seem to multiply significantly in the natural environment, but they can survive several weeks in water and several years in soil if conditions of temperature, humidity, and pH are favorable.

1.3 PATHOGENESIS

Humans are the only source of S. typhi infection. Most organisms are shed in the stool. Urine may also be a vehicle of S. typhi; especially in those with chronic Schistosoma haematobium infection. Infection is spread via the faecal–oral route by food or water that is contaminated by a person with active disease or an asymptomatic chronic carrier.

Organisms are introduced by ingestion into the gastrointestinal tract. The infectious dose of S. enteric serotype typhi in volunteers varies between 1000 and 1 million organisms. S. enteric serotype typhi must survive the gastric acid barrier to reach the small intestine, and a low gastric pH is an important defence mechanism. Achlorhydria as a result of aging, previous gastrectomy, or treatment with histamine H2-receptor antagonists, proton-pump inhibitors, or large amounts of antacids lowers the infective dose.

In the small intestine, the bacteria adhere to mucosal cells and then invade the mucosa. The M cells, specialized epithelial cells overlying Peyer’s patches, are probably the site of the internalization of S. enteric serotype typhi. Salmonellae are then transported to lymphoid tissue including the mesenteric lymph nodes.

Organisms may gain entrance into the bloodstream directly or via the thoracic duct. The incubation period is typically 7 to 14 days (3–60 days). Salmonellae infect mononuclear cells and have the ability to survive within phagosomes. Haematogenous dissemination to the reticuloendothelial system occurs with hepatic, splenic, bone marrow, lymphoid, and gallbladder involvement. Gall bladder invasion can also occur by retrograde spread from bile. Organisms excreted in the bile either reinvade the intestinal wall or are excreted in the feces.

Histologically, aggregates of infected macrophages, called typhoid nodules, occur in the intestine, liver, spleen, and many other organs. The intestinal pathologic stages include:

hyperplasia within typhoid nodules in the Peyer’s patches and the lymphoid follicles of the caecum within the first week

necrosis of the mucosa during the second week

sloughing of the necrotic mucosa to form ulcerations that may bleed or perforate during the third week.

Perforation and bleeding typically occur on the anti-mesenteric border of the intestine near the ileocaecal valve.

Interactions between S. typhi and macrophages lead to the production of cytokines such as tumour necrosis factor α, interleukin-1, and interferon-α and β, leading in turn to fever and other constitutional symptoms.5, 7

1.4 EPIDEMIOLOGY OF TYPHOID FEVER

In the year 2000, it is estimated that typhoid fever caused approximately 21650974 cases and 216510 deaths. Paratyphoid caused 5412744 illnesses. Infants, children and adolescents in South East Asia experience the greatest burden of illness.8

The estimated global crude incidence rates in the year 2000 are as follows:

Figure : Global Crude Incidence Rates of Typhoid fever in the year 2000

TyphoidMap.jpg

The best figure available for the global burden of enteric fever suggest that Africa, with a crude incidence rate of 50 per 100000, has a far lower burden than Asia, which has a crude incidence rate of 274 per 100000.8, 9

In Africa, the burden of typhoid fever is largely unknown due to the limited studies done. In a population based study in Egypt, an incidence rate of 59 per 100000 was determined.1 However, a number of hospital based surveillances and case reports suggest that typhoid is indeed a major public health concern in African countries. Crump et al estimated a crude incidence rate in East Africa as 39 per 100000 people whereas Southern Africa had the highest incidence rate (233/100000) in Africa.8 In 2007-9, Breiman et al conducted a population-based surveillance in Kibera, an urban informal settlement in Nairobi, and in Lwak, a rural area in western Kenya. In the urban site, the overall crude incidence of Salmonella enterica serovar Typhi (S. Typhi) bacteremia was 247 cases per 100,000 person-years of observation (pyo). Crude overall incidence in Lwak was 29 cases per 100,000 pyo.35

It is estimated that 3% of patients with fever reporting for treatment at hospitals in Kenya are infected with Salmonella typhi.10

1.5 CLINICAL FEATURES

The incubation period of typhoid fever is 10 – 14 days but the duration of illness is 4 weeks. The illness begins insidiously with mounting fever, headache, vague abdominal pain and constipation during the first week. Relative bradycardia and a palpable spleen may be present at the end of first week.

During the second week, patients become dull and apathetic. The abdomen will become distended and rashes (Rose spots) may appear over the lower chest and upper abdomen between days 7 – 10.

During the third week, the patient will assume the ‘typhoid state’ which is a state of prolonged toxaemia, delirium, disorientation and/or coma. Around the fourth week diarrhea will become apparent. If left untreated there is high risk of intestinal hemorrhage and perforation. Otherwise patient recovers by the fourth week.

However, there are variations to the above description in patients with multi-drug resistant disease, in children and HIV positive patients. Some patients do not develop high fevers during the illness and remain undiagnosed (walking typhoid).

1.5.1 Carrier State

About 5% of patients excrete the organism after 3 months. This is due to persistence infection of gall bladder, resulting in S. typhii to be excreted in feces. The likelihood of becoming a chronic carrier increases with age, especially in women. In endemic areas where Salmonella and Schistosoma species co-occur, several lines of evidence suggest a synergistic bacteria-parasite interaction that results in a protracted course for the Salmonella infection that has proven difficult to diagnose and therapeutically remedy.36

1.5.2 Complications

Gastrointestinal complications are the commonest and manifest as hemorrhage and intestinal perforation (5 –10%). Ileal perforation has an overall frequency of 3% and an overall mortality of 39.6%.

Other complications include cardiovascular insufficiency, myocarditis, meningitis, hepatitis, cholecystitis, pleura-pulmonary involvement, osteitis and thyroiditis. These complications are more common in multi – drug resistant disease. This is mainly due to delayed initiation of appropriate treatment rather than the virulence of the organism.11

A study of cancer risk in chronic typhoid and paratyphoid carriers showed a large excess of cancer of gall bladder, pancreas, colorectal and lung.12

1.6 MULTI-DRUG RESISTANT TYPHOID FEVER

Towards the end of 1980s and 1990s, S. enteric serovar typhi developed resistance simultaneously to all the drugs that were then used as first line treatment (chloramphenicol, trimethoprim-sulfamethoxazole and ampicillin). These MDR-strains carried the IncH1 plasmids that encoded the resistance genes.5 Similarly in Kenya, Kariuki et al in 2000 established that in 1988 -1993 S.typhi strains were fully sensitive to all drugs used. Whereas, in 1997 – 1999; 82% of the isolates were resistant to ampicillin, chloramphenicol and tetracycline.

In the 2000s strains resistant to fluoroquinolones arose and have become a major problem especially in Asia.

A study by Mengo et al which looked at the trends in Salmonella enteric serovar typhi in Nairobi from 2004 – 2006 reported a total of 70 out of 100 isolate was resistant to ampicillin, cotrimoxazole and chloramphenicol. 15 isolates were resistant to all drugs which included drugs such as ciprofloxacin, cefuroxime, ceftriaxone and nalidixic acid. Resistance to nalidixic acid increased from 31% in 2004 to 35% in 2005 with 18% resistance to ciprofloxacin.13

The high level of resistance to commonly available antibiotics including fluoroquinolones is of major public health concern in the treatment of typhoid fever in Kenya.

1.7 DIAGNOSIS

Several options exist for diagnosing enteric fever: clinical signs and symptoms; serological markers; bacterial cultures; antigen detection and DNA amplification. None is entirely satisfactory.

The clinical diagnosis of typhoid fever is difficult because the manifestations of the disease are diverse and there are many causes of prolonged fever in typhoid endemic regions. Signs such as relative bradycardia or leucopoenia may be useful but give a low specificity.14

1.7.1 Microbiological cultures

The isolation of S. typhi from blood, bone marrow or other sterile sites provides the most conclusive confirmation of enteric fever. Therefore, culture should be considered as a gold standard17. The choice and time of collection of specimens are important.

Blood cultures are positive in 90% of cases during the first week, stool cultures are usually positive (45 – 65%) during the second week. Urine cultures are only positive (30%) after the 3rd week. Bone marrow aspirates give the best confirmation with 85% - 95% recovery11. A duodenal string culture is a further option. Diagnostic sensitivity can be improved by culture of multiple sites.15

Major limitations of culture for diagnosis are that many endemic countries lack adequate, quality assured microbiology laboratories. Furthermore, the time taken to generate culture results is always greater than 24 hours and often a result is not available until several days and thus may not help health workers to make decision.15

1.7.2 Antibody detection

The Widal test, which measures agglutinating antibodies against lipopolysaccharides (LPS; O9, 12) and flagella (Hd) antigens of S. typhi is still widely used.15 The Widal test ideally requires both an acute and convalescent – phase serum samples taken approximately 10 days apart, and a positive result is determined by a fourfold increase in antibody titres.

A single sample test is generally plagued by false negative and false positive results.15 It can be negative in up to 30% of culture – proven cases of typhoid fever because of prior antibiotic therapy. On the other hand S. typhi shares O and H antigens with other Salmonella serotypes and has cross – reacting epitopes with other Enterobactericiae leading to false – positive results. Such results may also occur in other clinical conditions e.g. malaria, typhus, dengue fever and cirrhosis. In areas of endemicity, there is often a low background level of antibodies in the normal population.16

ELISAs measuring anti – LPS antibodies and anti – flagella antibodies have been found to be more sensitive then Widal ‘O’ and ‘H’ tests but all are hampered by similar limitations of specificity that occur with the use of Widal test.15

1.7.3 Rapid diagnostic test

Typhidot® was developed for the detection of specific IgM and IgG against a 50kD S. typhi outer membrane protein. Typhidot – M® is a modified version, but detects IgM to the same outer membrane protein and is a more specific marker of current acute infection. TUBEX® is similar to slide latex agglutination test and detects antibody against S. typhi LPS.

When compared to blood culture – positive typhoid cases, sensitivities for TUBEX® vary between 56% and 100%, between 67% and 98% for Typhidot® and between 47% and 98% for Typhidot – M®. Specificities measured against healthy controls range from 58% – 100% for TUBEX®, 73% – 100% for Typhidot® and 65% – 93% for Typhidot – M®.

Major drawback of a rapid diagnostic and indeed of any non- culture based method is the lack of an isolated organism and antimicrobial susceptibility result. Furthermore they still lack sensitivity and specificity to be able to confidently recommend in endemic setting.15

1.7.4 Antigen detection

S. typhi antigen (Vi antigen) can be detected in the urine of some typhoid patients by co-agglutination and ELISA but specificity varies from 25% – 90%.14

1.7.5 Nucleic acid amplification

Polymerase chain reaction, either conventional or real time, can be performed on peripheral mononuclear cells using primers that amplify S. typhi flagellin gene. The test is more sensitive than blood culture alone (92% compared to 50% -70%) but requires significant technical expertise.11

Table : Sensitivity, specificity and predictive values of rapid diagnostic tests for typhoid fever as determined by comparison with blood culture results (WHO 2012)

Kit

Sensitivity %

(95% CI)

Specificity %

(95% CI)

O : semiquantitative slide agglutination

95.2 (86.5–99.0)

3.6 (0.1–18.3)

H : semiquantitative slide agglutination

80.3 (68.2–89.4)

50.0 (30.6–69.4)

O: single tube Widal

87.3 (76.5–94.4)

6.9 (0.8–22.8)

H: single tube Widal

95.2 (86.5–99.0)

13.8 (3.9–31.7)

TUBEX®

73.0 (60.3–83.4)

69.0 (49.2–84.7)

Typhidot IgM®

75.0 (61.1–86.0)

60.7 (40.6–78.5)

Typhidot IgG®

69.2 (54.9–81.3)

70.4 (49.8–86.2)

1.8 TREATMENT

Supportive measures are important in the management of typhoid fever, such as oral or intravenous hydration, the use of antipyretics, and appropriate nutrition and blood transfusions if indicated. More than 90% of patients can be managed at home with oral antibiotics. Yet, they need to be closely followed up to evaluate for development of any complication or treatment failure. However, patients with persistent vomiting, severe diarrhea and abdominal distension may require hospitalization and parenteral antibiotic therapy.18 Without appropriate treatment, the disease lasts 3 to 4 weeks with fever, septicaemia with a 10% – 30% mortality.18

Emergence of multidrug resistance and decreased ciprofloxacin susceptibility in Salmonella enterica serovar Typhi have rendered older drugs, including ampicillin, chloramphenicol, trimethoprim -sulphamethoxazole, ciprofloxacin, and ofloxacin, ineffective or suboptimal for typhoid fever. Ideally, treatment should be safe and available for adults and children in shortened courses of 5 days, cause defervescence within 1 week, render blood and stool cultures sterile, and prevent relapse. Azithromycin meets these criteria better than other drugs. Among fluoroquinolones, which are more effective than cephalosporin, gatifloxacin appears to be more effective than ciprofloxacin and ofloxacin for patients infected with bacteria showing decreased ciprofloxacin susceptibility. However, azithromycin and gatifloxacin need to be preserved for treatment of MDR –TB.37 Ceftriaxone continues to be useful as a back-up choice, and chloramphenicol, despite its toxicity for bone marrow and history of plasmid-mediated resistance, is making a comeback in developing countries that show their bacteria to be susceptible to it.19

1.9 PREVENTION

The major routes of transmission of typhoid fever are through drinking water or eating food contaminated with Salmonella typhi. Prevention is based on ensuring access to safe water and by promoting safe food handling practices. Health education is paramount to raise public awareness and induce behaviour change.

1.9.1 Vaccination

Two safe and effective vaccines are now licensed and available. One is based on defined subunit antigens, the other on whole-cell live attenuated bacteria.

The first of these vaccines, containing Vi polysaccharide, is given in a single dose subcutaneous or intramuscularly. The live oral vaccine Ty2la is available in enteric-coated capsule or liquid formulation.16 Vi antibodies showed a sero-conversion rate of 76.2% and a sero-protection rate after vaccination was 74.2%.17

A new Vi conjugate candidate vaccine bound to non-toxic recombinant Pseudomonas aeruginosa exotoxin A (rEPA) has enhanced immunogenicity in adults.16

The Ty21a and Vi vaccines are recommended for travellers to areas where typhoid is endemic, household contacts of typhoid carriers and laboratory workers likely to handle S. enterica serotype Typhi.5

2.0 THE WIDAL TEST

The Widal agglutination test was developed by F. Widal in 1896. The test demonstrates the presence of antibody in the serum of a patient, against the H (flagellar) and O (somatic) antigens of Salmonella typhi.20

2.0.1 Performance technique

The test uses the bacterial suspensions of S typhi and S paratyphi A and B which are treated to retain only the O and H antigens. These antigens are then used to detect the corresponding antibodies in the serum of the patient. The IgM somatic O antibody appears first representing the initial serologic response, while the IgG flagella H antibody develops slowly yet persists for longer.20

There are two types of agglutination techniques: the slide test and the tube test. The slide test is rapid and used as a screening procedure. A drop of the antigen is added to an equal amount of the prepared serum. Agglutinations are visualised as clumps. A positive screening test requires determination of antibody strength. This is done by adding equal amounts of antigen to diluted serum.20

The tube agglutination test serves as a confirmatory test for the results of the slide test. A mixture of antigen and antibody is incubated in a water bath at 37oC – 50oC for 2 – 4 hours. Agglutinations are visualised in the form of pellets, clumped together at the bottom of the test tube.20

2.0.2 Interpretation of the test results

Salmonella are divided into distinct serologic groups (A - E) on the basis of their somatic O antigens. While all group D organisms, such as S. typhi possess O antigen 9, about 60 of the serotypes including S. typhi also have O antigen 12. Hence, infection by any of the group D serotypes can produce a positive Widal reaction. Also, all groups A and B organisms possess O antigen 12, hence cross-reactions can occur with these organism.20

The H titer is of little or no value since it is non- specific and variable. It remains elevated for years following infection or immunization against typhoid fever. However, a rise in O antibody titres during the first 2 – 3 weeks of illness is indicative of typhoid fever.21 Widal testing done on an acute phase serum of a patient had limited diagnostic capability given its low sensitivity. Revathi et al when using a cut-off value of 1:320 for both O and H agglutinins as being diagnostic of S. typhi, only 18 (26%) of the patients had diagnostic titres; 37(53.6%) had O and H titres less than 1:40.22

2.0.3 Causes of negative Widal agglutination tests

absence of infection

carrier state

an inadequate inoculum of bacterial antigen to induce antibody production

technical difficulty or errors in the performance of the test

prior antibiotic treatment

variability of the commercial antigens

A negative Widal test result does not therefore rule out the absence of infection.

2.0.4 Causes of positive Widal agglutination tests

presence of typhoid fever

previous immunisation

Cross-reaction with non-typhoidal Salmonella.

variability and poorly standardised commercial antigen preparation

infection with malaria, other Enterobacteriaceae, dengue fever

other diseases such as liver cirrhosis

2.0.5 Limitations of the Widal test

Clinically, a single Widal test in an unvaccinated or unexposed patient has some diagnostic relevance. However, the result of such a test has no diagnostic significance in an endemic region; in part due to difficulty in establishing a steady-state or baseline titer of Widal agglutination.20

Widal test should therefore be used in situation where there is no other confirmatory test, such as cultures, available.20

2.1 WIDAL TITRES IN THE NORMAL POPULATION

The problems of diagnosis of enteric fever in African countries are numerous.21 Doing a careful single Widal test on serum during an acute illness together with compatible clinical features appear to be the best option left.

The apparent increase in occurrence of typhoid fever might be due either to a true increase in infection or an over diagnosis. Over diagnosis based on the results of the Widal test may occur because the test are poorly performed or because of poor interpretation by prescribers. The majority of the prescribers (76% of doctors and 61% of nurses) could detect patients who truly had positive Widal test and needed treatment. However an average of 48% of the doctors and 54% of the nurses would treat patients who did not require treatment based on the Widal test result. Patients may therefore be treated unnecessarily. Misdiagnosis of Typhoid fever leads to unnecessary expenditure and exposure of patients to the side effects of antibiotics. Furthermore, there is resistance to the antimicrobial due to the unnecessary drug pressure. In addition, misdiagnosis may result in delayed diagnosis and treatment of other acute febrile illness.23

Duguid et al emphasized the importance of baseline enteric antibody estimations in healthy populations.24 Interpretation of Widal test depends on knowledge of baseline titres among healthy population. These titres can be used as a cut-off value to arrive at diagnostic titres.2 Any titer that occurs in more than 5% of the normal population is not regarded as a significant indicator of an active infection.25 For a single sero-diagnosis value, only a four-fold rise to what has been found to be normal should be significant.26

A number of surveys have previously been carried out to determine the distribution of levels of normal agglutinins to the typhoid bacilli in various parts of the world. These surveys revealed that the agglutinin levels varied for different communities depending on the endemicity of typhoid.

In Asian countries like Nepal, antibody titres are found in 62% of the population, 27 in Malaysia 61% had an H agglutinin and 34% had O agglutinins.28

In Nigeria, one study found 35% of the normal population with antibodies, with normal titer of 1:40 and 1:80 for O and H respectively,26 while another study found 53% of the normal population with antibodies with a minimum titer of 1: 160.29 In Malawi 8% of healthy blood donors had a positive Widal test.30 Furthermore, in a Cameroonian study, the baseline titres of O antibodies were 1:100 and that of H antibodies were 1:400.31 In a similar study in Sudan, Salmonella typhi O agglutinins were found in titer of 1:320 in 10.5% of the study population.32 In a study done by Mamo et al in Ethiopia, the result indicated that among the apparently health population, almost all the blood tested showed some titer of the antibody and reactivity of agglutination slide tests. The 95% probability limit (mean + 2SD) for anti H and anti O antigen titration was 1:276.89 and 1:207.89, respectively.33

In a study done by Kariuki et al in Nairobi, 41.6% showed high H antibody titres while O antibody titres remained low and insignificant.3 However a prior study done in 1995 by Jumba et al showed that a significant percentage (66%) of healthy unvaccinated individuals had low levels reactivity to typhoid antigens of which 96% of the individuals had low level reactivity with titres <1:80 for both H and O antigens while 4% had titres of 1:160 or above and 30% had no detectable Widal titres. He, therefore, concluded that ‘in this typhoid endemic regions titres up to 1:80 are common but both H and O titres of 1:160 and above found in conjunction with the clinical picture may be taken to be suggestive of typhoid fever.34

3.0 STUDY JUSTIFICATION

Typhoid fever is a serious public health problem in developing countries including Kenya. Serological diagnosis relies classically in the demonstration of rising titre of antibody in paired samples which is not always demonstrated even after 2 weeks. For practical purpose, a treatment decision must be made on the basis of results obtained with a single acute phase sample. However, the results obtained are often misinterpreted. Most treatment for typhoid disease is based on a positive Widal reaction only, regardless of the antibody titres. Misdiagnosis of Typhoid fever leads to unnecessary expenditure and exposure of patients to the side effects of antibiotics. Furthermore, there is resistance to the antimicrobial due to the unnecessary drug pressure. In addition, misdiagnosis may result in delayed diagnosis and treatment of other acute febrile illness.26

Yet Widal test can be used as a diagnostic tool in typhoid fever endemic area, if we know the baseline titres in the normal population. The level of normal agglutinins for these organisms varies in different countries and communities. This variation depends on the degree to which the typhoid is endemic in each area, a factor which may change over time. Moreover, the diagnostic titre depends on the background antibody titres level, and the level of typhoid vaccination.

The interpretation of Widal test depends upon the baseline titres which is prevalent amongst healthy individuals in a particular geographical area. These titres can then be used as a cut – off value to arrive at diagnostic titres. Antibody titres beyond a cut – off value can be regarded as significantly elevated titres which may be used for diagnosis in an appropriate clinical setting.

Blood donors are often used as proxies for the general population especially when determining reference ranges and validating assay methods as they are apparent healthy populations drawn up from the whole community, indirectly reflecting the status of the community.

There are no local studies on the distribution of Salmonella antibodies in blood donors. This study tries to evaluate and detect the Salmonella antibody titre in apparently healthy blood donors in Nairobi.

4.0 STUDY QUESTION

What is the distribution of Salmonella antibodies in the sera of healthy blood donors?

5.0 STUDY OBJECTIVES

5.1 Primary objective

To determine the distribution of Salmonella enteric serovar typhi antibodies among healthy blood donors.

6.0 METHODOLOGY

6.1 STUDY SITE

The study site was The Kenyatta National Hospital Blood Transfusion Unit. The Blood Transfusion Unit runs throughout the week. However, most blood donation is done during the weekdays.

6.2 STUDY POPULATION

The study population were voluntary but healthy blood donors over 18 years of age at the KNH Blood Transfusion Unit.

6.3 STUDY DESIGN

Cross sectional study

6.4 SAMPLE SIZE

The sample size was calculated using the following equation:

N =z2 x p(1-p)

d2

N=minimum sample size required

z=confidence interval at 95% (standard value of 1.96)

p=estimated prevalence of reactivity to typhoid antigens from Jumba et al, 1995 study = 66%.

d=margin of error (0.05)

N= (1.96) 2 x 0.66(1-0.66)

(0.05)2

The minimum sample size for this study was 345 apparently healthy blood donors.

6.5 SAMPLING METHOD

Consecutive sampling was undertaken to recruit patients in the aforementioned study site. This was done to obtain 50 patients per week until the desired sample size was achieved.

6.6 INCLUSION AND EXCLUSION CRITERIA

6.6.1 Inclusion Criteria

Eligible blood donors over 18 years of age.

Written informed consent.

6.6.2 Exclusion Criteria

The donors positive for the following screening tests:

Malaria

HBsAg

HIV

HCV

VDRL (Syphilis)

History of typhoid vaccination.

Current febrile or diarrheal illness

Currently on antibiotic treatment

6.7 SCREENING AND RECRUITMENT

The principle investigator (PI) with the help of research assistants interviewed the donors attending the blood transfusion unit for blood donation. Health screening of the blood donors was done using the Blood Transfusion Unit’s pre-transfusion screening questionnaires. The donors were then given all the relevant information about the study by the PI and/or the research assistants, and those who gave written informed consent (appendix I and appendix II) were recruited. Individuals with any of the exclusion criteria were excluded. Those individuals who were excluded from the study but were still eligible as blood donors, still carried on with blood donation. The results of the HIV, HBsAg, HCV and VDRL serology were obtained from the registry the following day.

6.7.1 SAMPLES AND LABORATORY ANALYSES

Blood samples were collected into pilot bottles after each donor had been bled into blood bags. About 5.0 ml of blood was taken from the tubes of each bag that were not diluted by CPDA1 present inside the blood bags. A thick blood smear for malaria was prepared according to standard protocol and examined under the microscope for presence or absence of malarial parasites. Serum was then separated immediately, labelled and stored in at -20°C for further processing for serological screening of transfusion transmissible infection as per WHO safe transfusion protocol. Samples from participants found to have a positive HIV, HBsAg, HCV, VDRL and/or Malarial parasite were discarded.

The Malarial test and the Widal test was done at UNITID (University of Nairobi Institute of Tropical and Infectious Diseases) laboratory by a qualified laboratory technician. Sera was first tested by the Widal slide method, and the titre of reactive specimens was then determined. Sterile physiological saline was used to make serial dilutions of 1/20, 1/40, 1/80, 1/160, 1/320, 1/640 and 1/1280 of each serum specimen. A drop of the appropriate antigen suspension was added to an equal volume of the diluted sera in an appropriate tube. After incubation, end- stage titres were examined for agglutination.

6.7.2 QUALITY ASSURANCE

Standard operating procedures for specimen collection, preparation and storage were followed to minimize pre-analytical errors. All equipment was calibrated according to manufacturer’s specification. Commercial controls were used to validate the calibration (see appendix III). The results were only acceptable if the machines were properly calibrated using standard calibration methods and materials and tests assayed against controls. Results were recorded onto data sheets which were checked by two people to minimize post analytical transcription errors.

6.8 STUDY VARIABLES

6.8.1 Dependent Variables

Baseline Salmonella antibody titres

6.8.2 Independent Variables

Sociodemographics factors such as:

Gender

Age

6.9 DATA MANAGEMENT AND ANALYSIS

To ensure quality (reliability and credibility) of the data, each study proforma was provided with a unique study serial number to prevent duplication of data collection. All data collected was entered into a password protected computer database using Microsoft access computer software. Statistical analysis was done using statistical package for social scientists (SPSS) version 17 after cleaning and verification. Data is presented in the form of tables, graphs and histograms as illustrated below. Descriptive statistics such as means, medians, proportions and standard deviation were determined where applicable.

C:\Users\mohammed ezzi\Desktop\thesis\methodology\1476-0711-8-1-1.jpg C:\Users\mohammed ezzi\Desktop\thesis\methodology\1476-0711-8-1-2.jpg

C:\Users\mohammed ezzi\Desktop\thesis\methodology\1476-0711-8-1-3.jpg C:\Users\mohammed ezzi\Desktop\thesis\methodology\image003.jpg

C:\Users\mohammed ezzi\Desktop\thesis\methodology\image002.jpg C:\Users\mohammed ezzi\Desktop\thesis\methodology\image004.jpg

6.11 ETHICAL CONSIDERATIONS

The study was undertaken after approval by the Department of Clinical Medicine & Therapeutics, University of Nairobi and the KNH/UoN Scientific and Ethical Review Committee.

The objectives and purposes of the study were clearly explained to eligible participants in a language suitable to them prior to inclusion into the study. Only blood donors who gave informed consent were enrolled. Blood donors were free to withdraw during the study period without discrimination. Information gathered from the study participants were kept confidential. All those donors who had a positive screening test, were duly informed and advised on further management.

6.12 STUDY FEASIBILITY

The KNH Blood Tranfusion Unit runs almost daily with most blood donation done over the weekdays. There were approximately 15 blood donors recruited daily. This translates to a collection period of approximately 8 weeks.

7.0 RESULTS

A total of 396 subjects were screened between October 15th and November 20th 2012. Of these 43 subjects were excluded for various reasons: 1 refused consent, 27 had either received typhoid vaccination or taken an antibiotic two weeks prior, 3 had failed peripheral vein access, 12 had transfusion transmissible infections (TTIs). A total of 353 subjects were enrolled into the study. All study subjects had their Widal titres analysis done.

Table : Prevalence of Transfusion Transmissible Infections

Transfusion Transmissible Infections

% (n)

HIV

0.2% (1)

Hepatitis B

2.5% (9)

Hepatitis C

0.5% (2)

Syphilis

0.0% (0)

Malaria

0.0% (0)

396 donors screened

1 refused consent

27 had prior typhoid vaccination or taken an antibiotic two weeks prior

3 had failed peripheral vein access

365 donors recruited

1 HIV +ve

9 HbSAg +ve

2 HCV +ve

353 donors eligible

Figure 2: Flow-chart illustrating Screening, Recruitment and Eligibility of Blood Donors

7.1 DEMOGRAPHICS

Of the 353 subjects who were eligible, 278 (79%) were male and 75 (21%) were female. Majority, 171 (48.4%) subjects, were aged between 20 years to 29 years. The mean age was 30.22 years with a standard deviation of 8.47 years, while the median was 28 years. The youngest donor was 18 years old and the oldest 56 years old.

Figure 3: Sex distribution of Eligible Blood Donors

Figure 4: Age Categories among Eligible Blood Donors

MEDIAN AGE: 28 YEARS

Of the 43 subjects who were excluded, 34 (79%) were male and 9 (21%) were female. Majority, 20 (46.51%) subjects, were aged between 30 years to 39 years. The mean age was 32.84 years with a standard deviation of 8.05 years, while the median was 32 years. The youngest donor was 20 years old and the oldest 52 years old.

Figure 5: Sex Distribution among Excluded Blood Donors

Figure 6: Age Categories of Exluded Blood Donors

MEDIAN AGE: 32 YEARS

7.2 WIDAL REACTION

Among the total 353 samples tested, 75 samples (21.25%) showed agglutination (Widal Positive) for the O and H antibodies. The distribution of the individual antibodies is shown in Table 3 below.

Figure 7: Proportion of Reactive Widal Titres

Table : Distribution of Individual Antibodies

Antibody type

Frequency

Percentage (%)

Anti –O antigen

27

36.0

Anti –H antigen

22

29.3

Anti –O and Anti –H antigen

26

34.7

The distribution of 53 samples with a positive anti-O titer showed that 5% of the sample had a titre of 1:160. About half (49% of all samples) had a titer of 1:40 (Figure 8). Of note, a significant proportion of blood samples (12 individuals) had an anti-O titer of 1:80. Similarly, among the 48 samples showing a positive anti-H titer, 12 of the samples were positive at a titer of 1:80 and 11 of the samples had a titer of 1:160 (Figure 9). Figure 10 depicts the percentages of sera with end titres in 353 apparently normal blood donors. It was observed that titres of 1: 40 occurred in a significant proportion of the sample (5.95 – 7.67%) in the O and H antigen respectively. Titres ≥ 1: 80 occurred in more than 4% of O titres and more than 6% of H titres.

Figure 8: Distribution of positive anti - O titres

Figure 9: Distribution of positive anti - 0 titres

Figure 10: Percentages of Sera with End Titres in 353 Blood Donors

8.0 DISCUSSION

This was the first study which was done on the blood donors in Nairobi, to estimate the baseline antibody titre in the healthy population against various serotypes of S. enteric serovar typhi by using the Widal tube agglutination test. This study was done at the Blood Transfusion Unit in Kenyatta National Hospital, Nairobi. Blood donors are often used as proxies for the general population especially when determining reference ranges and validating assay methods as they are apparent healthy populations drawn up from the whole community. Furthermore, this group of people is not routinely screened for typhoid fever and portends a risk of transmitting Salmonellosis with transfusion of blood. This study was aimed at assessing the prevalence of Salmonella antibodies among donors in our hospital blood bank.

The study consisted of a total of 396 donors, which constituted 79% males and 21% females with an average age of 30.5 years. A total of 365 blood donors were screened for transmissible transfusion infections, and out of these 3.2% were found to have transmissible transfusion infections. Of note, none of the blood donors screened for transmissible transfusion infection had a positive blood smear for malaria or VDRL positivity. These figures are in contrast to the last quarter 2012 figures of Nairobi Regional Blood Transfusion Services. Their donors were younger and constituted of 68% males and 32% females with an average age of 25 years. This discrepancy is due to the fact that National Blood Transfusion Services core donor population consists of College students. However they had a higher prevalence of transmissible transfusion infections (4.1%).38

The results showed that the sera of a significant proportion of healthy individuals contained antibodies which were capable of reacting to the variable titres in the Widal test. Among the 353 samples of the healthy volunteers who were eligible, 21.25% had a positive reaction on the slide agglutination test. Henceforth, 1 in 5th individual may be treated for "typhoid" just based on a positive screening test. Therefore, it is of paramount importance to follow up a positive slide agglutination test with a semi-quantitative tube titration test to get the exact reactive titre.

The possible explanations for the high prevalence of Salmonella among the blood donors may be that Salmonella is persistent in the human host at sub-clinical levels (carriers). Some donors may have developed tolerance to repeated exposure of small inocula of Salmonella.29

However, this is in contrast to a study done by Kariuki et al who found 41.6% of his study population had high H antibody titres while O antibody titres remained low and insignificant. He attributed this to typhoid vaccine inoculation as his study population consisted of apparently healthy adults who had only come to the hospital for medical exam for the issuance of Food Handlers Health Certificate and were mostly from the hotel industry. Also H antigens are non-specific and tend to remain elevated for many years, especially in typhoid vaccinated individuals.3

The most frequently recorded O and H agglutinin titres of the reactive samples were 1:40. This titre constituted approximately 49% and 43% of positive O and H agglutinins respectively. Hence, this should be considered as the baseline Widal titre for this population.

However, also to note that there was a significant proportion of the sample size that reacted to titres greater than 1:80. About 28% of the samples reacted to titres ≥1:80 and 6% of samples showed agglutination to titres of 1:160 for O agglutinins. According to Corales and Schmitt, high 'O' antibody agglutinins indicate acute infection. The implication of this is that this donor population is capable of transmitting Salmonella bacteria to their blood receipients.39 This subgroup should be followed up by doing stool or urine cultures and serological test like Vi antigen, so as to confirm whether they are asymptomatic carriers or not. As mentioned earlier, the prevalence of Widal positive donations was 21%. If screening of donors for Salmonella were to be carried out routinely, this could lead to a high loss of blood inventory. Such a loss of inventory would not augur well in which voluntary blood donors are hard to come by. However, not all blood units which test positive to Widal antigen carry live Salmonellae bacteria.

This study reported a slightly lower baseline titre of Widal reactivity in the healthy population than similar studies done across the African continent.26, 29, 30, 31, 32, 33, 34 A similar study done by Jumba et al in 1995, showed that titres of up to 1:80 are common for both O and H titres. His sample consisted of healthy unvaccinated individuals representing a number of housing estates, learning institutions and workplaces in Nairobi. Furthermore, he had a second subgroup from Naivasha who were workers and their families who were staying at an Animal Husbandry Research Centre. High residual agglutinin levels were found to be more abundant in the sera of the healthy population from Naivasha as compared to the sera from the Nairobi population. The difference between the sera from the Naivasha population and sera from the Nairobi population with respect to antibody titre distribution was significant. He attributed this to the use of untreated borehole water as the source of drinking water in the Naivasha population. Hence, this could have contributed to the overall high reactive titres of 1:80 found in his study.34 In addition; persons with possible Transfusion Transmissible Infection, any other chronic illnesses or risk factors for both were not excluded; this could have led to the differences in titres obtained.

Other factors may have contributed to this discrepancy. The differences in the antibody response may be due to the fact that the Widal test procedure is subject to inter- and intra-observer variability inherent to the Widal test and the sharing of the antigen determinants with other Salmonellae. A widespread antibiotic abuse can dampen the antibody response, giving a low titre in the Widal test and a previous immunization with the TAB vaccine and technical differences may be the other contributory factors.2 With respect to the last point, the Widal test which was performed on the same serum specimen in four laboratories gave widely different results.40 Furthermore in areas where fever due to infectious causes is a common occurrence, the possibility exists that false positive reactions may occur as a result of non-typhoidal fever.

8.1 STUDY LIMITATIONS

Several studies have highlighted the limitations of using the Widal serological test in the laboratory diagnosis of Salmonella, the worst being its non-specificity. Furthermore, some donors might not remember recent infections or usage of medications (especially over the counter medications) and can be inadvertently included in the study. This study may not be generalized to the whole population, as the study population is selective of blood donors who are thoroughly screened for various illness before being allowed to donate blood. Blood donors, who had titres of 1:160, were not able to be followed up for either rising titres or for later development of febrile illness.

8.2 RECOMMENDATIONS

Studies should be carried out among non-typhoidal febrile patients and confirmed typhoid cases, and compared with the gold standard of blood cultures. Furthermore similar studies should be carried out in the pediatric age group and among the various population of different economic background to determine the significant antibody level with certainty. Furthermore, additional studies should be carried out to look at stool or urine cultures and Vi antigens in stool of those blood donors who had elevated titres of 1:160. I also recommend that other rapid identification tests such as TUBEX©, Typhidot M©, IgM dipstick and Salmonella Antigen tests be evaluated for use in Kenya and compared with Widal test and blood cultures as these tests have been found to be more sensitive than the Widal test in several studies in Asia.

8.3 CONCLUSION

In conclusion, it was found out that about 20% of healthy blood donors had a positive Widal screening test. About 50% of this population had a baseline titre of 1:40 for both the O and H antigen with a significant proportion of blood donors having titres ≥ 1:80.

9.0 REFERENCES

Samuel Kariuki, Typhoid fever in sub-Saharan Africa: Challenges of diagnosis and management of infections, J Infect Developing Countries 2008; 2(6):443-447.

Anand M. Patil, M.L. Kulkarni and Akhil M. Kulkarni, Baseline Widal Titres in Healthy Children, Indian J Pediatr 2007; 74(12) : 1081-1083.

Samuel Kariuki, Joyce Mwituria, Agnes Munyalo et al, Typhoid is over-reported in Embu and Nairobi, Kenya, Afr J Health Sci 2004; 11: 103-110

Kenneth Todar, Salmonella and Salmonellosis, Todar’s Online Textbook of Microbiology, www.textbookofbacteriology.net

Christopher M. Parry, Tran Tinh Hien, Gordon Dougan, Nicholas J. White and Jeremy J. Farrar, Typhoid Fever, N Engl J Med, Vol. 347, No. 22.

Grossman DA, Witham ND, Burr DH, Lesmana M, Rubin FA, Schoolnik GK, Parsonnet J, Flagellar serotypes of Salmonella typhi in Indonesia: relationships among motility, invasiveness, and clinical illness, J Infect Dis. 1995 Jan;171(1):212-6

Nancy F. Crum, Current Trends in Typhoid Fever, Current Gastroenterology Reports 2003, 5:279–286.

John A. Crump, Stephen P. Luby, Eric D. Mintz, The Global Burden of Typhoid Fever, Bulletin of the World Health Organisation 2004; 82: 346-353.

Atul Kothari, Amit Pruthi, Tulsi D. Chugh, The Burden of Enteric Fever, J Infect Developing Countries 2008; 2(4): 253-259.

Okoth et al, The use and interpretation of Widal test in the diagnosis of Typhoid fever by health care workers in Misori District, Kenya, Dissertation 2008.

I. Merican, Typhoid Fever: Present and Future, Med J Malaysia Vol 52 No 3 Sept 1997.

Caygill CP, Hill MJ, Braddick M, Sharp JC, Cancer mortality in chronic typhoid and paratyphoid carriers, Lancet. 1994 Jan 8; 343(8889):83-4.

Doris M. Mengo, Sam Kariuki, Ann W. T. Muigai, Gunturu N. Revathi, Trends in Salmonella enteric serovar Typhi in Nairobi, Kenya from 2004 to 2006, J Infect Dev Ctries 2010; 4(6):393-396.

John Wain, Salih Hosoglu, The laboratory diagnosis of enteric fever, J Infect Developing Countries 2008; 2(6):421-425.

Christopher M Parry, Lalith Wijedoru, Amit Arjyal and Stephen Baker, The utility of diagnostic tests for enteric fever in endemic locations, Expert Rev. Anti Infect. Ther. 9 (6), 711–725 (2011).

WHO/V&B/03.07, Background document: The diagnosis, treatment and prevention of typhoid fever, www.who.int/vaccines-documents/

Mirza NB, Wamola IA, Estambale BA, Mbithi E, Poillet M, Typhim Vi vaccine against typhoid fever: a clinical trial in Kenya, East Afr Med J. 1995 Mar;72(3):162-4.

Effa EE, Lassi ZS, Critchley JA, Garner P, Sinclair D, Olliaro PL, Bhutta ZA, Fluoroquinolones for treating typhoid and paratyphoid fever (enteric fever) (Review), 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

T. Butler, Treatment of typhoid fever in the 21st century: promises and shortcomings, Clin Microbiol Infect 2011; 17: 959–963

Lateef A Olopoenia, Aprileona L King, Widal agglutination test 100 years later: still plagued by controversy, Postgrad Med J 2000;76:80–84.

N. B. Mirza, Diagnosis and overdiagnosis of typhoid fever, East Afr Med J. 1995 Dec;72(12):753-4.

Geoffrey Omuse, Ruchika Kohli, Gunturu Revathi, Diagnostic utility of a single Widal test in the diagnosis of typhoid fever at Aga Khan University Hospital (AKUH), Nairobi, Kenya, TROPICAL DOCTOR 2010; 40: 43 – 44

Nsutebu EF, Ndumbe PM, Koulla S, The increase in occurrence of typhoid fever in Cameroon: overdiagnosis due to misuse of the Widal test?, Trans R Soc Trop Med Hyg. 2002 Jan-Feb;96(1):64-7.

Duguid J. P, Marmiun B. P, Salmonella 1: typhoid and paratyphoid fevers, Macartney Medical Microbiology (13th Ed) - Churchill 1978.

Collard P, Sen R, Montefiore D, The distribution of Salmonella agglutinins in sera of healthy adults at Ibadan, J Hyg (Lond). 1959 Dec;57:427-34.

Tanyigna KB, Ayeni JA, Okeke EN, Onah JA, Bello CS, Antibody levels to Salmonella typhi and paratyphi in Nigerians. East Afr Med J. 1999 Nov;76(11):623-5.

Bharat M Pokhrel, Rajendra Karmacharya, Shyam K Mishra and Janak Koirala, Distribution of antibody titer against Salmonella enterica among healthy individuals in Nepal, Annals of Clinical Microbiology and Antimicrobials 2009, 8:1

T Pang and S D Puthucheary, Significance and value of the Widal test in the diagnosis of typhoid fever in an endemic area, J Clin Pathol 1983 36: 471-475.

Teddy C. Adias, Zaccheaus A. Jeremiah, Ayo O. Ilesanmi, Distribution of antibodies to Salmonella in the sera of blood donors in the south-western region of Nigeria, Blood Transfus 2010;8:163-9

Harries AD, Kamoto O, Maher D, Mukibii J, Khoromana C, Specificity of Widal test in healthy blood donors and patients with meningitis, J Infect. 1995 Sep;31(2):149-50.

Nsutebu EF, Ndumbe PM, Adiogo D, The distribution of anti-Salmonella antibodies in the sera of blood donors in Yaoundé, Cameroon, Trans R Soc Trop Med Hyg. 2002 Jan-Feb;96(1):68-9.

el-Shafie S, The Widal test in a normal healthy population in the Sudan, East Afr Med J. 1991 Apr;68(4):266-9.

Mamo Y, Belachew T, Abebe W, Gebre-Selassie S, Jira C, Pattern of widal agglutination reaction in apparently healthy population of Jimma town, southwest Ethiopia, Ethiop Med J. 2007 Jan;45(1):69-77.

Jumba MM, Mirza NB, Mwaura FB, Salmonella typhi and Salmonella paratyphi antibodies in Kenya, East Afr Med J. 1995 Dec;72(12):755-7.

Breiman RF, Cosmas L, Njuguna H, Audi A, Olack B, et al. (2012) Population-Based Incidence of Typhoid Fever in an Urban Informal Settlement and a Rural Area in Kenya: Implications for Typhoid Vaccine Use in Africa. PLoS ONE 7(1): e29119. doi:10.1371/journal.pone.0029119

Melhem RF, LoVerde PT. Mechanism of interaction of Salmonella and Schistosoma species. Infect Immun. 1984 May;44(2):274-81.

Adhvaryu M, Vakharia B. Drug-resistant tuberculosis: emerging treatment options. Clin Pharmacol. 2011;3:51-67. Epub 2011 Dec 14.

Regional Blood Tranfusiom Service Quarterly Report for the month October 2012 – December 2012.

Corales R, Schmitt, SK (2002). Typhoid fever. Available from URL:http://www.emedicine/MED/topic2331.htm.

Pang T, Puthucherry SD. The significance and the value of the Widal test in the diagnosis of typhoid fever in an endemic area. J Clin Pathol 1983; 36:471-75.