Campylobacter species are the most common cause of bacterial gastroenteritis in the developed world. However, comparatively few studies have determined the epidemiological features of campylobacteriosis in resource-poor settings.
A total of 1,941 faecal specimens collected from symptomatic (diarrhoeic) children and 507 specimens from asymptomatic (non-diarrhoeic) children hospitalised in Blantyre, Malawi, between 1997 and 2007, and previously tested for the presence of rotavirus and norovirus, was analysed for C. jejuni and C. coli using a real time PCR assay.
Campylobacter species were detected in 415/1,941 (21%) of diarrhoeic children, with C. jejuni accounting for 85% of all cases. The median age of children with Campylobacter infection was 11 months (range 0.1–55 months), and was significantly higher than that for children with rotavirus and norovirus (6 months and 7 months respectively; P<0.001). Co-infection with either rotavirus or norovirus was noted in 41% of all cases in the diarrhoeic group. In contrast, the detection rate of Campylobacter in the non-diarrhoeic group was 14%, with viral co-infection identified in 16% of children with Campylobacter. There was no association between Campylobacter detection rate and season over the 10 year period.
Citation: Mason J, Iturriza-Gomara M, O’Brien SJ, Ngwira BM, Dove W, Maiden MCJ, et al. (2013) Campylobacter Infection in Children in Malawi Is Common and Is Frequently Associated with Enteric Virus Co-Infections. PLoS ONE 8(3): e59663. https://doi.org/10.1371/journal.pone.0059663
Editor: Georgina L. Hold, University of Aberdeen, United Kingdom
Received: November 29, 2012; Accepted: February 17, 2013; Published: March 26, 2013
Copyright: © 2013 Mason et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The study was funded by the University of Liverpool. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
It is estimated that 3.552 million children under the age of five years die each year in Africa; diarrhoeal disease accounts for 11% of these deaths . Diarrhoea also contributes to childhood morbidity, particularly malnutrition and growth stunting , . The introduction of intervention strategies, such as provision of oral rehydration therapy, has had a positive impact on reducing childhood diarrhoea mortality, with an estimated annual reduction of diarrhoeal deaths in Africa since 2000 of 3.7% ; however diarrhoea remains a leading cause of child mortality and morbidity in this region. The importance of pathogens such as rotavirus and enterotoxigenic Escherichia coli in the aetiology of severe childhood diarrhoea in developing countries is well recognised , . However the role of Campylobacter is less well understood.
Campylobacter is a fastidious gram negative bacterium and C. jejuni and C. coli are considered the most common cause of bacterial gastroenteritis in the developed world. The clinical presentation ranges from mild watery to severe inflammatory diarrhoea which may be complicated by post infectious sequelae such as Guillain-Barré Syndrome . Using bacterial culture methodology, estimates of the prevalence of Campylobacter infection in young children with diarrhoea in Sub Saharan Africa range from 1.5% to 18% , , , , . While molecular techniques have been developed and employed for Campylobacter detection in epidemiological studies in developed countries, such methods have not been widely adopted in Sub Saharan Africa. Where molecular detection (notably PCR) was used to examine for Campylobacter species in adults and children with diarrhoea in South Africa the prevalence estimates of C. jejuni, C. coli and C. concisus were 12%, 7.5% and 2.7% respectively; however only 34 of the 255 samples analysed were from children <5 years of age .
As part of a long-term research programme investigating viral gastroenteritis in children in Malawi, we have collected stool samples since 1997 from children <5 years of age admitted to the Queen Elizabeth Central Hospital (QECH), Blantyre, Malawi with moderate to severe diarrhoea . We have now examined stored faecal specimens using a real time PCR assay to determine the prevalence and epidemiological features of Campylobacter infection in this population.
Written, informed consent was obtained from the child’s parent or guardian prior to enrolment. Ethical approval was obtained from the Malawi National Health Sciences Research Committee.
The QECH is a large government run tertiary referral hospital in the southern Malawian city of Blantyre, which has a population of approximately 1 million living in urban and peri-urban settlements.
Enrolment and data collection procedures have previously been described in detail . Briefly, children age <5 years admitted to the QECH with ≥3 loose or watery stools within a 24 hour period for <14 days, were eligible for inclusion. Children were enrolled Monday to Friday, 9 am to 5 pm, from 1st July 1997 to 30th June 2007. A second group of children <5 years of age without diarrhoea, who were admitted to the QECH with conditions such as malaria and respiratory infections, were enrolled between 1997 and 1999. A single faecal specimen was obtained from each child. Clinical data (illness severity, blood in stools etc.) were not routinely gathered. Following EIA testing for rotavirus , remaining samples were shipped to the University of Liverpool and stored at −80°C until testing for norovirus by real time PCR  and for Campylobacter (this study).
Of 2,458 faecal specimens collected from hospitalised diarrhoeic children in the primary study , 1,941 were available for testing for Campylobacter infection, together with 507 samples from children admitted to hospital without diarrhoea. DNA was extracted from all samples using an automated extractor (Qiasymphony, Qiagen). The presence of a 95 base pair fragment of the mapA gene of C. jejuni and a 103 base pair fragment of the ceuE gene of C. coli were detected using a real time PCR method .
Characteristics of the Study Population
In total 2,448 samples was analysed; 1,941 from diarrhoeic children and 507 from non-diarrhoeic children. The median age of children in the diarrhoeic group was 9 months (range 0–55 months) and in the non-diarrhoeic group was 6 months (range 1–50 months). The diarrhoeic group contained 55% males and the non-diarrhoeic group 52% males.
Over the 10 year study period (1997–2007) Campylobacter was detected in 415/1941 (21%) of diarrhoeic specimens. For the two year period (1997–1999) in which faecal specimens from non-diarrhoeic children were collected, the detection rate of Campylobacter was significantly higher in the diarrhoeic specimens than the non-diarrhoeic specimens (28% vs. 14%; p<0.001) (Table 1). Of the Campylobacter species detected between 1997 and 1999, C. coli comprised 10% and 4% of all Campylobacter from diarrhoeic and non-diarrhoeic specimens respectively (p<0.001). There was no statistical difference in the Campylobacter cycle threshold values obtained from diarrhoeic vs. non-diarrhoeic specimens (data not shown).
The median age of children with diarrhoea in whom Campylobacter was detected was 11 months (range 0.1–55 months) which was higher than the age of children with rotavirus or norovirus (median age 6 months and 7 months respectively (p<0.001)). The detection rate of Campylobacter was relatively constant across all age groups, in contrast to the detection rate of rotavirus and norovirus which decreased with age (Figure 1). The median age of non-diarrhoeic children with Campylobacter infection was 11 months (range 1–34 months); the detection rate increased from 5% of children in the 0–2 month age group to 20% of children age >18 months (Table 1).
In the diarrhoeic group 40% of children with Campylobacter had an enteric virus co-infection. These co-infections occurred predominantly in children <1 year of age with 50% of all Campylobacter cases in this age group also having either rotavirus or norovirus in the specimen. In the non-diarrhoeic group 16% of children with Campylobacter in the specimen also had a viral co-infection (Table 1). Although the overall prevalence of Campylobacter was significantly higher in the diarrhoeic than in the non-diarrhoeic group, when single infections were considered (i.e. in the absence of either rotavirus or norovirus) this difference is less pronounced (Campylobacter detection rate of 16% in the diarrhoeic vs. 11% in the non-diarrhoeic group, p = 0.02).
The detection rate of Campylobacter did not vary consistently by month or season of specimen collection during the 10 year study period. In total 49% of Campylobacter positive specimens in the diarrhoeic group occurred in the dry season (May to October) and 51% in the wet season (November to April). In the non-diarrhoeic group 41% and 59% of positive specimens occurred in the dry and wet season respectively (p = 0.89; data not shown).
In this large study of Campylobacter infection in Malawian children, using a sensitive molecular assay we documented Campylobacter in 21% of specimens obtained from children <5 years of age hospitalised with diarrhoea. There are no previous data describing the prevalence of Campylobacter infection in Malawi. Estimates of the prevalence of Campylobacter infection in hospitalised children <5 years of age with diarrhoea in Sub Saharan Africa range from 1.5% in Botswana , 1.7% in Mozambique , 9% in Uganda , 11% in The Central African Republic  to 18% in Tanzania . In community settings the detection rate of Campylobacter in diarrhoeal faecal specimens among children <5 years of age has been estimated at 15.9% in The Central African Republic , 3.3% in Djibouti  and 0.8% in Guinea-Bissau . This variation in detection rate across Sub Saharan Africa may reflect the technical difficulties of isolating Campylobacter species in resource poor settings because of its fastidious growth requirements and/or the relative insensitivity of some culture techniques.
Comparing culture and PCR in the detection of C. jejuni and C. coli in diarrhoeal faecal specimens in Ireland suggested that culture alone detected only 55% of all cases . In the UK, Campylobacter detection rates in community diarrhoea cases increased from 4.0% by direct culture, to 5.0% after a faecal enrichment procedure was added, to 15% by PCR . A study in South Africa of 255 adults and children with diarrhoea and 67 children without diarrhoea, demonstrated similar overall rates of Campylobacter using PCR to those reported in the current Malawi study (19.6% in symptomatic patients and 11% in asymptomatic patients) ; however the prevalence in children <5 years of age with diarrhoea was 11% compared with 21% in the current study. Thus our data suggest that the detection rate of C. jejuni and C. coli using a sensitive and specific PCR assay in faecal specimens of children <5 years of age hospitalised with diarrhoea in Sub Saharan Africa may be higher than previously reported.
Of note, we documented a relative over-representation of C. coli compared with C. jejuni in the faecal specimens of diarrhoeic compared with non-diarrhoeic children. A similar pattern was reported in Tanzania where C. coli was present in 20% of diarrhoeic and 6% of non-diarrhoeic specimens from adults and children . In the developed world C. coli is particularly associated with chicken  and the over-representation of C. coli in children with diarrhoea raises the possibility that poultry may be a particularly important source of infection in this group. Campylobacter spp. other than C. jejuni and C. coli, which require either non-selective culture or specific molecular techniques for their optimal detection, are reported to be common in some studies. For example in South Africa C. concisus and C. upsaliensis accounted for 23%, and C. coli for 3%, of all Campylobacter and related species isolated  and in Ireland using PCR C. ureolyticus accounted for 22.3% of all Campylobacter species detected . Although both C. jejuni and C. coli were commonly detected among hospitalised Malawian children with diarrhoea, the contribution of other Campylobacter species has not yet been determined.
The prevalence of Campylobacter remained relatively constant in children up to 5 years of age in the current study suggesting that unlike viral infections, Campylobacter may make a relatively greater contribution to diarrhoeal disease in children aged >12 months. It is possible that repeated exposure to Campylobacter species from environmental sources throughout childhood may explain the relatively high prevalence in children >1 year of age; however further clinical, epidemiological, and immunological studies are required to confirm the sources of transmission, environmental reservoirs and the role of Campylobacter infection in diarrhoeal disease in this setting. Where detailed age related prevalence rates have been determined in children in Sub Saharan Africa similar patterns have been observed, for example a large cohort study in The Central African Republic reported an average Campylobacter detection rate of 9.5% across all age groups . Although deaths in children secondary to diarrhoea are decreasing worldwide, diarrhoea morbidity remains constant, particularly in older children . Given the association of childhood diarrhoea with growth stunting in this population  the contribution of Campylobacter to the burden of diarrhoeal disease children >1 year of age requires further elucidation.
A significant proportion of diarrhoeal specimens in which Campylobacter was detected also contained norovirus or rotavirus, particularly in children <12 months of age. Given the high prevalence of Campylobacter in non-diarrhoeic specimens, its detection in diarrhoeic specimens may represent “flushing out” of organisms in diarrhoeal episodes caused by a viral pathogen. We do not have data on the presence of other bacterial or parasitic organisms. We are therefore unable to ascertain which, if any, pathogen is the causative agent of the diarrhoeal episode, although we speculate that Campylobacter when detected alone in older children may be more likely to be the causative agent of disease. Furthermore, although there was a significant difference in the rate of detection of Campylobacter in isolation between the diarrhoeic and non-diarrhoeic groups (16% vs. 11% respectively), further case control studies are required to confirm this finding. Mixed infections involving Campylobacter have been described previously; one large (n = 3,038) surveillance study of diarrhoea in hospitalised children <5 years of age in Bangladesh reported that 59% of all Campylobacter infections were associated with at least one other bacterial or protozoan pathogen . Few studies have reported mixed infections of Campylobacter and viral pathogens; a cohort study in India demonstrated that rotavirus was associated with 11/18 Campylobacter infections in children <5 years of age . Since data from in-vitro cell culture systems suggest that viral co-infection increases the adhesion and invasion of pathogenic Campylobacter species , further work should explore the clinical consequences of co-infection with viral enteric pathogens on the pathogenesis of Campylobacter infection.
Campylobacter was detected in 14% of non-diarrhoeic children who were recruited over a two-year period. Since no preceding clinical or microbiological data are available for these children, we are unable to determine whether this finding represents asymptomatic colonisation or extended excretion after resolution of a Campylobacter diarrhoeal episode. Asymptomatic colonisation appears to occur frequently in children in Sub Saharan Africa; one longitudinal study reported that 41.7% of all children in a community birth cohort were asymptomatically colonised with Campylobacter within the first 6 months of life . This phenomenon does not appear to occur in the developed world with the exception of workers occupationally exposed to Campylobacter . It is thought that regular exposure to Campylobacter within an abattoir or farm environment results in immune tolerance towards Campylobacter species and hence facilitates colonisation . It is feasible that a similar response occurs in children in Sub Saharan Africa explaining the observed high rates of asymptomatic infection. Given the high sensitivity and specificity of the assay used in this study it is also possible that the detection rates seen in the control cohort could represent post excretion following a diarrhoeal episode as Campylobacter is known to be excreted for up to 12 weeks post infection .
This study highlights specific differences in the epidemiology of Campylobacter in developed compared to developing countries. Firstly, in the UK, the detection rate of Campylobacter in children with and without diarrhoea is less than that seen in children in Sub Saharan Africa (6.8% and 2% in diarrhoeal and non-diarrhoeal children in the UK respectively) ; secondly, there are age specific differences in detection rates in children in the UK with prevalence increasing throughout childhood ; and lastly there is a strong seasonal association in temperate climates with a peak in incidence occurring during the warmer months whereas no seasonal association was seen in Malawi . These differences may be explained by a number of factors including differences in the routes of transmission between the two settings; in the developed world the majority of Campylobacter strains causing human infections can be epidemiologically linked to strains that colonise poultry, with the main route of transmission of the pathogen thought to be through the handling and consumption of contaminated meat . Although there are no comparative epidemiological studies linking poultry and human strains in Sub Saharan Africa, up to 40% of commercial chickens in Senegal and 60% in South Africa are colonised with Campylobacter , . Furthermore, strains colonising chickens in Senegal are genetically similar to strains colonising chickens in the UK, The Netherlands and United States suggesting that the host association of Campylobacter genotypes transcends geographical boundaries . Chicken is a widely consumed meat in Malawi and many families keep chicken flocks which are routinely housed within the human dwelling including in food preparation areas . Thus in contrast to the developed world, there is increased potential for acquisition and spread of zoonotic and foodborne disease such as Campylobacter which may account for the higher prevalence rates reported in this and other studies.
In conclusion, this large study of children hospitalised with diarrhoea in Malawi suggests that the burden of Campylobacter is higher than previously appreciated, and is frequently identified in association with concomitant rotavirus and/or norovirus infection. Given the recent introduction of rotavirus vaccines into childhood immunisation programmes in Malawi and other parts of Sub Saharan Africa it is predicted that bacterial pathogens including Campylobacter could play a more prominent role in the aetiology of diarrhoeal disease in young children in this region in the future. An improved understanding of the clinical features, epidemiology, and pathogenesis of Campylobacter infection in Sub Saharan Africa will inform future prevention strategies against this foodborne zoonotic pathogen.
Dr. Jenifer Mason is a National Institute for Health Research (NIHR) Academic Clinical Fellow in Medical Microbiology.
Conceived and designed the experiments: NC MIG BN JM. Performed the experiments: WD JM. Analyzed the data: JM MIG WD NC SOB. Contributed reagents/materials/analysis tools: BN NC MIG. Wrote the paper: JM MIG SOB MM NC.
- 1. Liu L, Johnson HL, Cousens S, Perin J, Scott S, et al. (2012) Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000. Lancet 379: 2151–2161.
- 2. Lopez AD, Mathers CD, Ezzati M, Jamison DT, Murray CJL (2006) Global and regional burden of disease and risk factors, 2001: systematic analysis of population health data. Lancet 367: 1747–1757.
- 3. Checkley W, Buckley G, Gilman RH, Assis AM, Guerrant RL, et al. (2008) Multi-country analysis of the effects of diarrhoea on childhood stunting. Int J Epidemiol 37: 816–830.
- 4. Gupta SK, Keck J, Ram PK, Crump JA, Miller MA, et al. (2008) Part III. Analysis of data gaps pertaining to enterotoxigenic Escherichia coli infections in low and medium human development index countries, 1984–2005. Epidemiol Infect 136: 721–738.
- 5. Cunliffe NA, Ngwira BM, Dove W, Thindwa BD, Turner AM, et al. (2010) Epidemiology of rotavirus infection in children in Blantyre, Malawi, 1997–2007. J Infect Dis 202 Suppl: S168–174
- 6. Hughes RAC, Cornblath DR (2005) Guillain-Barré syndrome. Lancet 366: 1653–1666.
- 7. Mshana SE, Joloba M, Kakooza A, Kaddu-Mulindwa D (2009) Campylobacter spp. among children with acute diarrhea attending Mulago hospital in Kampala - Uganda. Afr Health Sci 9: 201–205.
- 8. Mandomando IM, Macete EV, Ruiz J, Sanz S, Abacassamo F, et al. (2007) Etiology of diarrhea in children younger than 5 years of age admitted in a rural hospital of southern Mozambique. Am J Trop Med Hyg 76: 522–527.
- 9. Georges MC, Wachsmuth IK, Meunier DM, Nebout N, Didier F, et al. (1984) Parasitic, bacterial, and viral enteric pathogens associated with diarrhea in the Central African Republic. J Clin Microbiol 19: 571–575.
- 10. Lindblom GB, Ahren C, Changalucha J, Gabone R, Kaijser B, et al. (1995) Campylobacter jejuni/coli and enterotoxigenic Escherichia coli (ETEC) in faeces from children and adults in Tanzania. Scand J Infect Dis 27: 589–593.
- 11. Rowe JS, Shah SS, Motlhagodi S, Bafana M, Tawanana E, et al. (2010) An epidemiologic review of enteropathogens in Gaborone, Botswana: shifting patterns of resistance in an HIV endemic region. Plos One 5: e10924.
- 12. Samie A, Obi CL, Barrett LJ, Powell SM, Guerrant RL (2007) Prevalence of Campylobacter species, Helicobacter pylori and Arcobacter species in stool samples from the Venda region, Limpopo, South Africa: studies using molecular diagnostic methods. J Infect 54: 558–566.
- 13. Trainor E, Lopman B, Iturriza-Gomara M, Dove W, Ngwira BM, et al.. (2013) Detection and molecular characterisation of noroviruses in hospitalised children in Malawi, 1997–2007. J Med Virol: In press.
- 14. Amar CF, East CL, Gray J, Iturriza-Gomara M, Maclure EA, et al. (2007) Detection by PCR of eight groups of enteric pathogens in 4,627 faecal samples: re-examination of the English case-control Infectious Intestinal Disease Study (1993–1996). Eur J Clin Microbiol Infect Dis 26: 311–323.
- 15. Georges-Courbot MC, Beraud-Cassel AM, Gouandjika I, Georges AJ (1987) Prospective study of enteric Campylobacter infections in children from birth to 6 months in the Central African Republic. J Clin Microbiol 25: 836–839.
- 16. Mikhall IA, Fox E, Haberberger Jr RL, Ahmed MH, Abatte EA (1990) Epidemiology of bacterial pathogens associated with infectious diarrhea in Djibouti. J Clin Microbiol 28: 956–961.
- 17. Molbak K, Wested N, Hojlyng N, Scheutz F, Gottschau A, et al. (1994) The etiology of early childhood diarrhea: a community study from Guinea-Bissau. J Infect Dis 169: 581–587.
- 18. Bessede E, Delcamp A, Sifre E, Buissonniere A, Megraud F (2011) New methods for detection of campylobacters in stool samples in comparison to culture. J Clin Microbiol 49: 941–944.
- 19. Tam CC, O’Brien SJ, Tompkins DS, Bolton FJ, Berry L, et al. (2012) Changes in causes of acute gastroenteritis in the United Kingdom over 15 years: microbiologic findings from 2 prospective, population-based studies of infectious intestinal disease. Clin Infect Dis 54: 1275–1286.
- 20. Mughini Gras L, Smid JH, Wagenaar JA, de Boer AG, Havelaar AH, et al. (2012) Risk factors for campylobacteriosis of chicken, ruminant, and environmental origin: a combined case-control and source attribution analysis. PLoS ONE 7: e42599.
- 21. Lastovica AJ, le Roux E (2000) Efficient isolation of campylobacteria from stools. J Clin Microbiol 38: 2798–2799.
- 22. Bullman S, Corcoran D, O’Leary J, O’Hare D, Lucey B, et al. (2011) Emerging dynamics of human campylobacteriosis in Southern Ireland. FEMS Immunol Med Microbiol 63: 248–253.
- 23. Walker CL, Black RE (2010) Diarrhoea morbidity and mortality in older children, adolescents, and adults. Epidemiol Infect 138: 1215–1226.
- 24. Weisz A, Meuli G, Thakwalakwa C, Trehan I, Maleta K, et al. (2011) The duration of diarrhea and fever is associated with growth faltering in rural Malawian children aged 6–18 months. Nutrition J 10: 25.
- 25. Glass RI, Stoll BJ, Huq MI, Struelens MJ, Blaser M, et al. (1983) Epidemiologic and clinical features of endemic Campylobacter jejuni infection in Bangladesh. J Infect Dis 148: 292–296.
- 26. Rajendran P, Babji S, George AT, Rajan DP, Kang G, et al. (2012) Detection and species identification of Campylobacter in stool samples of children and animals from Vellore, south India. Indian J Med Microbiol 30: 85–88.
- 27. Konkel ME, Joens LA (1990) Effect of enteroviruses on adherence to and invasion of HEp-2 cells by Campylobacter isolates. Infect Immun 58: 1101–1105.
- 28. Cawthraw SA, Lind L, Kaijser B, Newell DG (2000) Antibodies, directed towards Campylobacter jejuni antigens, in sera from poultry abattoir workers. Clin Exp Immunol 122: 55–60.
- 29. Havelaar AH, van Pelt W, Ang CW, Wagenaar JA, van Putten JP, et al. (2009) Immunity to Campylobacter: its role in risk assessment and epidemiology. Crit Rev Microbiol 35: 1–22.
- 30. Richardson N, Koornhof H, Bokkenheuser V, Maynet Z, Rosen E (1983) Age related susceptibility to Campylobacter jejuni infection in a high prevalence population. Arch Dis Child 58: 616–619.
- 31. Lal A, Hales S, French N, Baker MG (2012) Seasonality in human zoonotic enteric diseases: A systematic review. PLoS ONE 7: e31883.
- 32. Humphrey T, O’Brien S, Madsen M (2007) Campylobacters as zoonotic pathogens: A food production perspective. Int J Food Microbiol 117: 237–257.
- 33. Nierop W, Duse AG, Marais E, Aihara M, Thothobolo N, et al. (2004) Contamination of chicken carcasses in Gauteng, South Africa, by Salmonella, Listeria monocytogenes and Campylobacter. Int J Food Microbiol 99: 1–6.
- 34. Kinana AD, Cardinale E, Tall F, Bahsoun I, Sire JM, et al. (2006) Genetic diversity and quinolone resistance in Campylobacter jejuni isolates from poultry in Senegal. Appl Environ Microbiol 72: 3309–3313.
- 35. Sheppard SK, Colles F, Richardson J, Cody AJ, Elson R, et al. (2010) Host association of Campylobacter genotypes transcends geographic variation. Appl Environ Microbiol 76: 5269–5277.
- 36. Gondwe TN, Wollny CB (2007) Local chicken production system in Malawi: household flock structure, dynamics, management and health. Trop Anim Health Prod 39: 103–113.