Search tips
Search criteria 


Logo of bloodtransLink to Publisher's site
Blood Transfus. 2010 July; 8(3): 163–169.
PMCID: PMC2906183

Distribution of antibodies to Salmonella in the sera of blood donors in the south-western region of Nigeria



There are currently no reports of a multicentre assessment of the presence of antibodies to Salmonella in donated blood in Nigeria.

Materials and methods

Blood specimens from blood donors presenting at six selected public blood collection centres in the south-western region of Nigeria were analysed for the presence of Salmonella antibodies.


Out of the 200 specimen analysed, 106 (53%) were found to be Widal-positive with a minimum titre of 160. Among the positive cases, reactivity was most common to S. typhi (D) antigens (48.6%). The most frequently recorded titre of reactive specimens (i.e., the modal titre) was 160, with 71 donors (35.5%) of the total sample population from the different centres reacting at this level. Thirty-one donors had elevated titres of = 320 and nine donors had a titre of 640. The percentages of Widal-positive cases among donors with A positive and O positive blood groups were 58.7% and 64.2%, respectively; the percentage of Widal-positive cases was lowest among B positive blood donors (26.7%). The p-value for these differences (p<0.05) shows that blood group has a statistically significant impact on Widal reaction in donors.


This study confirms that salmonellosis is endemic in Nigeria and that many of our blood donors may be Salmonella carriers. This study also revealed a positive association between blood groups and typhoid fever.

Keywords: Salmonella antibodies, blood donors, blood transfusion, Widal reaction


In Nigeria, as in many other developing countries, awareness about the safety of blood and blood products is still very low. Even among the community of blood bank workers, the concern for blood product safety is still below expectation1. Indeed, although many standard blood banks in Nigeria test for human immunodeficiency virus (HIV) and hepatitis B virus (HBV) in donated blood, there are still numerous centres in which screening for HIV is not carried out2 and awareness about testing for hepatitis C virus in donor blood is not yet high in Nigeria3. In contrast to hepatitis C, Salmonella is known, albeit to the medical community, to be an endemic infectious agent in this country. Because it is blood-borne, the possibility of it being transmitted via blood transfusion is real. Nsutebu et al.4 found significant Salmonella antibody titres in more than 10% of apparently healthy blood donors in neighbouring Cameroon. It is very unlikely that there is a significant difference in the living conditions of Nigerians and Cameroonians. A study to determine the distribution of Salmonella antibodies among donors in Nigeria and the possible implications of such findings is, therefore, relevant.

Two key issues have dominated discussion and research among blood transfusion scientists to date. One is the preservation of the viability of blood constituents in order to benefit patients receiving a transfusion. The other is the safety of blood and blood products for transfusion. Having succeeded in isolating the various immunological factors likely to cause adverse transfusion reactions, attention has shifted in recent years to excluding pathogens that are either blood-borne and transmitted via the donor or that are introduced during the process of preparing the blood and its products5,6. While efforts have been made to identify and curtail pathogens that pose a risk to the safety of the blood supply in developed countries6,7, it appears that there is little enthusiasm among stakeholders for ensuring the safety of the blood supply from infectious agents in Nigeria. Apart from some concern about HIV and HBV, which are blood-borne and can compromise the safety of the blood supply, policy-makers do not seem to be interested in screening for other blood-borne pathogens. Yet, no blood or blood product is safe until it is free from all agents that can create an untoward consequence after transfusion8. This is the reason why no effort should be spared in ensuring that all types of pathogens - viral or bacterial - are eliminated from the blood supply. Given that salmonellosis is endemic in our region, there is a potential risk of Salmonella being transmitted via transfusion9. Indeed, unscreened units of blood which harbour live Salmonella organisms or endotoxin could cause severe, possibly fatal, post-transfusion reactions.

Materials and methods

Blood collection centres

This was a multicentre, observational study carried out in six blood banks in Nigeria. The centres at which the samples were collected were randomly selected from stratified sets of hospital blood banks in the western axis of Nigeria. Five of them were blood collection centres located within a government hospital, while the remaining one was a blood bank serving the blood collection centres. The collection sites were located in urban centres, with one located in a semi-urban centre and one other located in a rural area.

The collection centres were stratified into whether they were in a Federal institution or a state-owned institution. The state-owned centres were those in Ogun State University Teaching Hospital, Sagamu; the General Hospital, Lagos and the General Blood Transfusion Service, Ibadan.


The samples tested were taken from 200 apparently healthy blood donors in the aforementioned centres between November, 2004 and May, 2005. One hundred and ninety-eight of the donors were males while two were females. History taking and oral pre-donation screening was done prior to the collection of blood.

All donors were apparently healthy, but some of them refused to disclose their age. The known ages ranged between 18 and 51 years with a mean of 28.9 years. One hundred and fifty-five of the subjects were commercial donors, 33 were family donors and the remaining 12 subjects were voluntary donors. The two females were both family donors.

Individuals with a recent history of fever and hypertension were excluded. Females who were undergoing their menstrual period were excluded. Blood pressure, heart rate and weight were determined for all subjects. Anyone failing to meet predetermined criteria for these variables was excluded. Samples that were positive for HIV and/or HBV were also excluded. For inclusion, males had to have a haemoglobin = 12.5 g/dL and females a haemoglobin = 11.5g/dL.

The study protocol and objectives of the study were explained to the subjects, and each gave verbal consent before specimen collection. Written consent was not, however, obtained.

Institutional approval was obtained in each hospital with the assistance of the Medical Laboratory Scientist in charge of each laboratory. All analyses carried out and data obtained are reported in this study. The follow-up of the patients to investigate clinical outcomes was based on each patient's hospital records.

Samples and laboratory analyses

Blood samples were collected into pilot bottles after each donor had been bled into blood bags. Serum was then obtained from the samples by centrifugation after the blood had been allowed to clot overnight. The sera were used for serology while the cellular part of each sample was used for blood grouping.

Sera were first tested by the Widal slide method using the Biosystem Febrile Antigen Kit (UK) (code number: FEBNC 100; batch number: FC0410- -13; expiry date: 04/07), then the titre of reactive specimens was determined. Sterile physiological saline was used to make serial dilutions of 1/40, 1/80, 1/160, 1/320 and 1/640 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. End-stage titres were confirmed microscopically.

Blood grouping was carried out for each subject by adding a drop of anti-A, anti-B and anti-A + B and anti-D to a 5% suspension of cells in a Kahn tube and observing the reactions after 2 hours. All reagents were bought from Biotec (Ipswich, UK) and used before the expiry date.

Positive control blood specimens were obtained from ten patients with clinically confirmed typhoid on admission to one of the centres participating in this study. These were pooled into two samples. Sterile saline was used as the negative control.

Statistical analysis

Data were analysed using the statistical package SPSS 13.0. The statistical analysis included descriptive inferential statistics. Demographic details were recorded and the relationship between dependent variables was analyzed using the chi-square distribution test. The level of statistical significance was set at p = 0.05.


Demographic characteristics of the enrolled subjects

Two hundred blood donors gave specimens for this study. Figure 1 shows the number of donors investigated in each blood bank. The Central Blood Transfusion Service, Ibadan, had the highest number of donors, while the Federal Medical Centre, Ido-Ekiti, had the fewest. Table I shows the demographic characteristics of the sample population. Of the 200 blood donors tested in this study, 198 were males (99%), while only two were females (1%). The age group with the highest number of donors was the group aged 21 – 25 years old, which provided 45 donors (22.5% of the total sample population). The age ‘group’ with the lowest number of donors was that aged 51 – 55 years old, which consisted of a single donor (0.5%).

Figure 1
Percentage of Widal-positive donors in each blood donation centre
Table I
Demographic characteristics of the sample population

More than 50% of the donors in each of the age groups 16 – 20 years, 26 – 30 years, 31 – 35 years and 51 – 55 years were Widal-positive (73.3%, 54.8%, 66.7% and 100%, respectively). The rate of Widal-positivity in the group of donors who did not reveal their age was 51.2%, while exactly 50% of individuals in the group aged 41 – 45 years were Widal-positive. Generally a fluctuating pattern of reaction was observed as age progressed.

Widal reaction

Altogether 106 (53%) donors were found to be Widal-positive (defined by a titre = 160) (Table II). The centre at which the rate of Widal-positive cases was highest was the OAU Teaching Hospital, Ile-Ife (60.6%), followed closely by the Ogun State Teaching Hospital, Sagamu (56.8%). The Central Blood Transfusion Service had a 56% positive rate while the University of Ilorin Teaching Hospital and the General Hospital, Lagos had 51.8% and 47.6% positive rates, respectively. The lowest rate of positivity (27.3%) was found in the Federal Medical Center, Ido-Ekiti (Figure 1).

Table II
Results of Widal testing

The most frequently recorded titre of reactive specimens was 160 (Figure 2). Thus, the modal titre was 160 and 71 donors (35.5%) of the total population provided samples reacting at this level. In four centres (CBTS, Ibadan; OAUTHC, Ile-Ife; General Hospital Lagos; and OSUTH, Sagamu) the modal titre also happened to be 160.

Figure 2
Titre of Widal reactivity in donors’ specimens, showing that the modal titre was 160.

Thirty-one donors had elevated titres of = 320 and nine had a titre of 640. Most of the donors with a titre = 320 came from two centres, the Central Blood Transfusion Service (n=12) and the OAU Teaching Hospital Complex, Ile-Ife (n=9).

Blood groups

Twenty-seven donors, representing 58.7% of the donors with blood group A were Widal-positive and among the donors with blood group B, 15 were Widal-positive (26.7%). Sixty-one of the donors with group O blood were Widal-positive (65.6%) although only one (33.3%) of the three O Rhesus-negative individuals was Widal-positive (Table III).

Table III
Effect of blood group on Widal reaction

Considering Rhesus (Rh) status, it was found that more than 50% of A Rh-positive and O Rh-positive blood donors were Widal-positive (Table III). In detail, 58.7% of the A Rh-positive individuals and 64.2% of the O Rh-positive individuals were Widal-positive. The lowest proportion of Widal-positive cases was observed in the B Rh-positive group (26.7%). The observed p-value (p<0.05) shows that blood group had a significant impact on Widal reaction in the donors.


Two pools of specimens were made from samples collected from patients with clinically confirmed Salmonella infection. The first pool had titres of 320 and 160 for ‘O’ and ‘H’, respectively, of S. typhi (D), while the second sample had titres of 320 and 320 for ‘O’ and ‘H’, respectively. The sterile saline control tested negative with the Widal antigens.


This study was aimed at assessing the prevalence of Salmonella antibodies among donors in our hospital blood banks, and the implications that this may have for transfusion safety in the nation. In this study we resorted to Salmonella serology, as this is the most feasible technique in the circumstances, considering the facilities available in Nigeria10. Several studies have highlighted the limitations of using the Widal serological test in the laboratory diagnosis of Salmonella, the worst being its non-specificity1113. Despite this, there seems to be a general consensus that the Widal test remains a valuable method for the diagnosis and control of typhoid fever in many developing countries11,12, provided a baseline antibody titre of the healthy afebrile individuals in the population being studied is known. In this study, the cut-off titre for positive cases was taken to be 1:160.

The 53% of the total donor population found to be Widal-positive in our study contrasts sharply with the 10% recorded among donors in Yaounde, Cameroon by Nsutebu et al.4, who used the same diagnostic technique. The high rate of Widal-positive cases among donors in this study is not totally unrelated to the indiscriminate resort to the use of commercial donors in our blood banks. In this study, 155 (77.5%) of the subjects studied were commercial donors, of whom 93 (60%) were Widal-positive. Commercial donors are known to constitute a high-risk group in the blood supply chain. Wallace14 postulated that individuals who have to sell their blood to survive are those most at risk of serious communicable diseases. Since it was not demonstrated that titres were rising among the donors, only those with high antibody titres can be assumed to have had active infection. According to Corales and Schmitt,9 high ‘O’ antibody agglutinins indicate acute infection. In other words, 92 individuals (46%) with ‘O’ agglutinin titre = 160 had active Salmonella infection at the time of blood donation. The implication of this is that almost half of the donor population was capable of transmitting Salmonella bacteria to the recipients of their blood. There are two possible explanations for this high prevalence of Salmonella among the blood donors: Salmonella is persistent in the human host15, a feature common to intracellular organisms. Its persistence at sub-clinical levels may be responsible for asymptomatic individuals (carriers) with relatively high antibody titres. Some affected donors might also have developed tolerance to repeated exposure of small inocula of Salmonella13 as a result of the organism being endemic in the communities in which the subjects live. Simply put, a lot of our donors are Salmonella carriers, and this portends danger. Apart from possible adverse transfusion reactions, the transmission of live Salmonella through blood could result in fatal bacteraemia or enteritis9.

The marked imbalance in the male to female ratio of donors makes any analysis of gender associations inappropriate. There were 198 male and only two female donors. Traditionally, females do not feature prominently as donors because of the demand placed on them by menstruation. According to Gordenk et al.17, women are at risk of iron deficiency if they donate more than one unit in a year.

The age of the donor did not affect the Widal reaction of the participants. Unlike for other infections associated with age-related behaviour, such as HIV, it appears that age does not need to be considered in selecting a donor for screening with regard to Salmonella. Although the age of the donor is usually an important factor in the choice of blood donors, infection with Salmonella cuts across all age groups. Mvere opines that the younger the donor, the safer he or she is likely to be1. It is believed that exposure to sexually transmitted diseases would be minimal in the teenage years; hence, the recruitment of high school students as voluntary donors is preferable to the recruitment of older subjects17.

Most donors reacted to the Salmonella D group of both the ‘O’ and ‘H’ antigens. The D group represents the S. typhi strain known to be the most lethal of the Salmonella genera. S. typhi, the causative agent of typhoid, is a debilitating and fatal disease that claims many lives in underdeveloped countries9. Thus, in the bid to ameliorate one disease condition, the transfusion of Salmonella-infected blood is capable of inducing another deadly one.

While the safety of the blood supply is a primary concern, this concern needs to be matched by adequacy of the blood supply in each community18. The prevalence of Salmonella antibodies in this study is, therefore, quite revealing. The prevalence of Widal-positive donations was above 25% in all of the centres. The lowest prevalence recorded was 27% (FMC, Ido-Ekiti), while the prevalence in one centre was as high as 62.1% (OSUTH, Sagamu). If screening of donors for Salmonella were to be carried out routinely, there would be an unacceptably high loss of blood inventory, which could lead to a discard rate as high as 25%. Such a colossal loss of inventory would not augur well for a nation in which voluntary blood donors are hard to come by. Krishnan and Brecher19 suggested that an acceptable loss of inventory should be less than 10%. On the basis of the our findings, the high level of discards that would be necessary to eliminate the risk of transfusion-related transmission of Salmonella would trigger a vicious cycle of discarding many donor blood units only to recruit more donors whose blood could result in further inventory loss. The consequences are a huge economic waste and demoralisation of the blood banking community.

Not all blood units which test positive for Widal antigens carry live Salmonella bacteria. The survival rate of Salmonella in blood stored in blood bank conditions (4°C–8°C) should be very low since these bacteria grow optimally at 37°C, and usually do not thrive below 8°C. Transfused fresh blood may, however, transmit live Salmonella bacteria. In addition, live Salmonella organisms at the time of phlebotomy may release pyrogen-inducing endotoxin into the donor unit even after refrigeration has decimated the bacterial population. Endotoxin from Salmonella is capable of inducing immunosuppression by causing neutropenia and, consequently, leucopenia in the host20,21. According to Bhutta et al.22, patients on cytotoxic therapy often develop neutropenia and fever. Incidentally, such patients are precisely those usually requiring transfusion of blood and blood products. Donor blood infected with Salmonella could be fatal in such cases. Furthermore, blood transfusion in itself is known to cause immunomodulation23. Thus, transfusion of Salmonella-infected blood has multiple negative effects on the recipient.

The rate of Widal-positive cases was highest among donors with an O blood group (64.2%). The rate of reactive donors was also highest (95.6%) in donors with this blood group. It can, therefore, be assumed that individuals with blood group O are more susceptible to Salmonella infections than individuals with other ABO blood groups. Mollison24 had already pointed out that individuals with blood group O are more vulnerable to certain diseases, particularly intestinal ulcer, than individuals with other ABO blood groups. He suggested that this susceptibility is due to the permeability of the intestinal wall in O blood group individuals. Incidentally, this aligns with the fact that Salmonella easily enters the intestinal wall cells of the human host25.

This study was not done with the aim of serotyping the subtypes of Salmonella spp., but rather it was intended to assess the prevalence of Salmonella antibodies in blood donated for transfusion and the associated risk. Most of the articles that criticise the use of the Widal test for determining the presence of Salmonella have been written by scientists working in developed countries where sophisticated procedures have replaced the use of serological diagnosis13,26,27. It is noteworthy that while the publication by Olopoenia and King13 was a review, it was based on original work done in a region of Africa where salmonellosis is endemic. The study included negative and positive samples and all other factors responsible for false positive and false negative reactions were properly taken care of. It is also pertinent to note that, in the absence of molecular procedures, in most centres in Nigeria, the routine diagnosis of Salmonella infection will continue to depend on the determination of antibody titres in the serum of patients suspected to have this infection. A study by Asuquo et al., carried out to compare the efficacy of serological diagnosis with that of bacteriological cultures of stool and blood, proved that the Widal test is still a more useful diagnosis than bacteriological cultures28. In their study, the prevalence of Salmonella antibodies was as high as 52% for S. typhi O and 47.5% for S. typhi H; thus, the 53% prevalence recorded in our study is consistent with data from other regions of Africa where salmonellosis is endemic. This could be a result of many healthy carriers who are capable of transmitting the infection to others, as discussed earlier.


1. Mvere DA. A strategy for blood safety in the African Region. Africa Health. 2002;26:9–14.
2. Esumeh FI, Ugbomoiko D, Isibor JO. Seroprevalence of HIV and hepatitis B surface antigen among blood donors in Central Hospital, Benin city. J Med Lab Sc. 2003;12:52–5.
3. Chukwura EF, Ogbodo SO, Obi GO. Seroprevalence of hepatitis C virus infection among blood donors in a South Eastern State of Nigeria. Biomed Res. 2005;16:17–20.
4. Nsutebu EF, Ndumbe PM, Adiogo D. The distribution of anti-Salmonella antibodies in the sera of blood donors in Yaounde, Cameroon. Trans R Soc Trop Med Hyg. 2002;96:68–9. [PubMed]
5. Bove JR. Testing in the years ahead: new pressures and new concerns. Transfusion. 1990;30:63–7. [PubMed]
6. Hill B. Current controversies in blood transfusion. The Biomed Sc. 2005;49:30–1.
7. Brown MR, Fristma MG, Margues MB. Transfusion safety: what has been done, what is still needed. Med Lab Observ. 2005;37:20–6. [PubMed]
8. Barr A, Muir W. Quality monitoring of blood and its components. Med Lab Sc. 1993;40:253–6. [PubMed]
9. Corales R, Schmitt SK. Typhoid fever. 2002. Available from URL: http://www.emedicine/MED/topic2331.htm.
10. Igbani MA. Technological disadvantage: an impediment to health-care delivery for developing African nations. Lab Med. 20031:310–1.
11. Hamze M, Naboulsi M, Vincent P. Evaluation of the Widal test for diagnosing of typhoid fever in Lebanon. Pathologie Biologie. 1998;46:613–6. [PubMed]
12. Parry CM, Hien TT, Dougan G, et al. Typhoid fever. New Engl J Med. 2000;347:1770–82. [PubMed]
13. Olopoenia LA, King AI. Widal agglutination test - 100 years later: still plagued by controversy. Postgrad Med J. 2000;76:80–4. [PMC free article] [PubMed]
14. Wallace B. Life-saver or bloodsucker? AIDS Action. 1991;14:1.
15. Sukupolvi S, Edeltein A, Rhen M, et al. Development of a murine model of chronic Salmonella infection. Infect Immun. 1997;65:838–42. [PMC free article] [PubMed]
16. Gordenk VR, Brittenham GM, Bravo J, et al. Prevalence of iron deficiency with carbonyl iron in female blood donors. Transfusion. 1990;30:239–43. [PubMed]
17. Parrish EC, Summers SH. Incidence of HIV seropositivity in high school and college age blood donors (1990–93) Clin Lab Sc. 1995;8:350–4.
18. Marwick C. Donor pool may rise as screening test falls. JAMA. 1995;273:366. [PubMed]
19. Krishnan AG, Brecher ME. Transfusion-transmitted bacterial infection. Hematol Oncol Clin North Am. 1995;9:167–83. [PubMed]
20. Hornick RB, Greisman SE, Woodward TE, et al. Typhoid fever: pathogenesis and immunologic control. New Engl J Med. 1970;283:686–91. 739–46. [PubMed]
21. Dick HM, Wilkinson PC, Powis S. Topley and Wilson's Principles of Bacteriology, Virology and immunity. 8th Edition. Vol. 360 Hodder and Stoughton Ltd; 1990.
22. Bhutta ZA, Mansurali N. Rapid serologic diagnosis of paediatric typhoid fever in an endemic area: a progressive evaluation of Typhidot enzyme immunoassays and the Widal test. Am J Trop Med Hyg. 1999;61:23–9. [PubMed]
23. Bloomberg N, Heal JM. Transfusion and recipient immune function. Arch Pathol Lab Med. 1989;113:246–53. [PubMed]
24. Mollison PL. Clinical Medicine. 7th Edition. Oxford: Blackwell Scientific Press; 1983. Blood transfusion; pp. 276–8.
25. Sansonetti PJ, Phalipon AM. Cells as portals of entry for enteroinvasive pathogens: mechanism of interaction and consequences for the disease process. Semin Immunol. 1999;11:193–203. [PubMed]
26. Bhutta ZA. Current concepts in the diagnosis and treatment of typhoid fever. BMJ. 2006;333:78–82. [PMC free article] [PubMed]
27. Narayanappa D, Sripathi R, Jagdishkumar K, Rajani HS. Comparative study of dot enzyme immunoassay (Typhidot-M) and Widal test in the diagnosis of typhoid fever. Indian Pediatrics. 2009. Apr 15, (E-pub ahead of print) [PubMed]
28. Asuquo HE, Ekpo MS, Ekpoke J, Abia-Bassey L. Determination of antibody titres and isolation of Salmonella species in suspected enteric fever patients. J Med Lab Sci. 2004;13:38–44.

Articles from Blood Transfusion are provided here courtesy of SIMTI Servizi