|

 

1471-2334-10-319-12

Schwarz et al. BMC Infectious Diseases 2010, 10:319http://www.biomedcentral.com/1471-2334/10/319 Systemic bacteraemia in children presenting with clinical pneumonia and the impact of non-typhoid salmonella (NTS)Norbert G Schwarz1*, Nimako Sarpong2, Frank Hünger2, Florian Marks3, Samuel EK Acquah2, Alex Agyekum2,Bernard Nkrumah2, Wibke Loag1, Ralf M Hagen1, Jennifer A Evans4, Denise Dekker2, Julius N Fobil1,Christian G Meyer1, Jürgen May1, Yaw Adu-Sarkodie5 Background: The diagnosis and antimicrobial treatment of pneumonia in African children in the absence ofdiagnostic means such as x-ray facilities or microbiological laboratories relies primarily on clinical symptomspresented by the patients. In order to assess the spectrum of bacterial pathogens, blood cultures were performedin children fulfilling the clinical criteria of pneumonia.
Methods: In total, 1032 blood cultures were taken from children between 2 months and 5 years of age who wereadmitted to a rural hospital in Ghana between September 2007 and July 2009. Pneumonia was diagnosed clinicallyand according to WHO criteria classified as "non-severe pneumonia" and "severe pneumonia" ("severe pneumonia"includes the WHO categories "severe pneumonia" and "very severe pneumonia").
Results: The proportion of bacteriaemia with non-typhoid salmonella (NTS) was similar in children with pneumonia(16/173, 9.2%) compared to children hospitalized for other reasons (112/859, 13%). NTS were the predominantorganisms isolated from children with clinical pneumonia and significantly more frequent than Streptococcuspneumoniae (8/173, 4.6%). Nine percent (9/101) of children presenting with severe pneumonia and 10% (7/72) ofchildren with non-severe pneumonia were infected with NTS. Nineteen out of 123 NTS isolates (15%) weresusceptible to aminopenicillins (amoxycillin/ampicillin), 23/127 (18%) to chlorampenicol, and 23/98 (23%) to co-trimoxazole. All NTS isolates were sensitive to ceftriaxone and ciprofloxacin.
Conclusion: In Sub-saharan Africa, sepsis with NTS should be considered in children with symptoms of pneumoniaand aminopenicillins might often not be the adequate drugs for treatment.
the commonly suspected organisms (e.g. Streptococcus Pneumonia is one of the most common diagnoses in pneumoniae and Haemophilus influenzae) [3]. However, African children presenting at hospitals and peripheral this treatment scheme does not capture enteric bacteria health centres and is the most important cause of mor- and there is some evidence that Gram-negative bacteria tality in children under five years of age [1,2]. Due to such as Salmonella spp. occur more frequently among the lack of bacteriological laboratory facilities, antibiotic African children with septic pneumonia than expected.
treatment of pneumonia is guided by presumptions Among 166 Kenyan children with blood cultures posi- based on clinical symptoms. According to the WHO- tive for non-typhoidal salmonella (NTS), 46% fulfilled IMCI (World Health Organisation Integrated Manage- the clinical criteria of pneumonia according to WHO- ment of Childhood Illness) amoxicillin is the first line IMCI. On the other hand, bacteraemia due to NTS was drug for empirical treatment of pneumonia as it covers not associated with diarrhoea [4]. In other reports fromEast, West and Central Africa, the impact of NTS insystemic infections was recognized as well [5-16]. More- * Correspondence: [email protected] 1Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany over, NTS were among the most frequent organisms Full list of author information is available at the end of the article 2010 Schwarz et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction inany medium, provided the original work is properly cited.
Schwarz et al. BMC Infectious Diseases 2010, 10:319 isolated from children with clinical pneumonia in some definitions for "very severe pneumonia" and "severe studies that were carried out before the vaccination against H. influenzae (Hib-vaccination) was introduced[14,17-19]. In the context of an improving coverage of Hib and pneumococcal immunisations, the impact of 1-3 ml of venous blood was taken from hospitalized NTS as a cause of pneumonia will, most likely, increase children and directly injected into a Becton Dickinson considerably among African children [20].
(BD) BACTEC™ PEDS PLUS™/F bottle containing 20 ml In the present study, blood cultures were performed in of enriched broth with resin. The bottles were incubated children admitted to a rural hospital with pneumonia at 35°C in a BD BACTEC™ 9050 automated Blood Cul- symptoms according to the WHO-IMCI definition in ture system (BD Diagnostics, Sparks, Massachussets, order to quantify the impact of systemic infections with USA) [22] for 5 days or until rated positive. Broth from NTS and to assess antibiotic drug resistance patterns.
positive bottles was directly examined by Gram stainand 20 μl were cultured further on sheep blood-agar, chocolate-agar and MacConkey agar (prepared in-house from Oxoid dehydrated media). Agar plates were incu- The study hospital is located in the village Agogo in the bated at 37°C for 18-24 h. Microscopical diagnosis after Asante Akim North District of the Ashanti Region in Gram staining was immediately reported to the clinical Ghana, West Africa, and is the principal hospital of the staff. Identification of S. pneumoniae was based on mor- district. Its catchment population encompasses approxi- phology of colonies and the optochin test. Oxacillin mately 80,000 people with roughly 25,000 citizens of discs were used to determine sensitivity to penicillin.
Agogo itself. The HIV/AIDS prevalence in the area is Diphteroids and propionibacteriae were classified as around 1.9% according to the National AIDS Control contaminants. If the patient was not known to be HIV Programme in 2009. Vaccination programmes against positive, coagulase negative staphylococci, bacillus spe- H. influenzae have been established in 2001.
cies and non-fermenters were usually also classified ascontaminants. On the same day, antibiotic susceptibility testing was performed using the disc diffusion method Between September 2007 and July 2009, 948 children with the susceptibility breakpoints suggested by the between 2 months and 5 years of age who were hospita- Clinical and Laboratory Standards Institute (CLSI). The lized and from whom a blood culture was taken were following antibiotics were tested: amoxycillin/ampicillin, included into the study. In total, 1032 blood cultures amoxyclav (amoxicillin & clavulanic acid), cefuroxime, were taken during 1169 hospitalizations. The study ceftriaxone, co-trimoxazole, ciprofloxacin, gentamicin, complied with the "Ethical Principles for Medical tetracycline and chloramphenicol. The microbiological Research Involving Human Subjects" as laid out in the laboratory that was involved in the study is enrolled in a Declaration of Helsinki and was approved by the Com- quarterly external quality assurance program in bacter- mittee on Human Research Publications and Ethics iology from the National Institute for Communicable (CHRPE) of the School of Medical Sciences, Kumasi, Diseases (NICD) of South Africa. If blood cultures that Ghana. Informed consent was sought from the parents were taken from the same child presenting twice within or the guardians of the child after detailed explanation 2 weeks yielded identical results, only the first result of the study and the study procedures.
was included into further analyses.
Clinical pneumonia, case classification Data collection and statistical analysis Pneumonia was classified following established WHO At admission of children, clinical data were recorded on guidelines [21]. Children with cough or difficult breath- standardized forms. Data collection was entirely ing were classified as "pneumonia cases" if they had an embedded into the clinical routine. The admission chart increased respiratory rate of ≥ 50 per minute in children contained a 4-paged admission sheet to be filled in by of 2 months up to 11 months of age and ≥ 40 per min- the admitting doctor. Double data entry was done by ute in children of 12 months up to 5 years of age. Chil- data entry clerks using a 4th Dimension Database (4D dren with symptoms of lower chest wall indrawing, San Jose, California, United States). Data analysis was nasal flaring, grunting (in young children) and/or sub- carried out using the STATA 10 software (College Sta- costal retraction or additional signs of central cyanosis, tion, Texas, United States).
severe respiratory distress, vomiting everything or inabil- For each pathogen, the number of isolates and their ity to breastfeed/drink, or other danger signs were clas- proportion among all blood cultures performed was cal- sified as "severe pneumonia" (including the WHO culated separately for children fulfilling the case Schwarz et al. BMC Infectious Diseases 2010, 10:319 definition for pneumonia, severe pneumonia and for co-infection was found in 35% of non-cases, 28% of pneumonia cases and 22% of severe pneumonia cases.
In patients without pneumonia, NTS represented 66.7% (112/168), S. aureus 10.1% (17/168), S. pneumo- Of the 1032 blood cultures included, 734 (71%) did not niae 7.1% (12/168) and S. Typhi 6.5% (11/168) of the yield bacterial growth and 90 (9%) were positive with pathogen isolates. In non-severe pneumonia cases NTS represented 53.8% (7/13), S. aureus 7.7% (1/13), S. pneu- Pathogens were isolated from the remaining 208 (20%) moniae 15.4% (2/13) and S. Typhi 7.7% (1/13) of the blood cultures. Antiobiotic pre-medication was pathogen isolates. In severe pneumonia cases NTS reported from 7% of children and was independent of represented 33.3% (9/27), S. aureus 14.8% (4/27), the positivity of blood cultures. The frequency of chil- S. pneumoniae 22.2% (6/27) and S. Typhi 11.1% (3/27) dren without pneumonia (12.2%), pneumonia cases of the pathogen isolates. When looking at all pneumonia (8.9%) and severe pneumonia cases (9.8%), was similar cases of any severity, 16 of 40 (40%) of all pathogen iso- in children with available and with missing blood cul- lates were NTS (Table 1).
tures (p = 0.5, c2 = 1.2).
In total, 173 children fulfilled the case definition of Sensitivity to antibiotics pneumonia of any severity. Blood cultures were positive All NTS and S. pneumoniae isolates from children aged in 18.1% (13/72) of children with non-severe pneumonia 2 months to 5 years were included for sensitivity testing and in 26.7% (27/101) of children with severe pneumo- against 9 antibiotics irrespective of their clinical diagno- nia. The frequency of bacteraemia was similar high in sis (Table 2). 15% of all NTS isolates were sensitive the 859 children without pneumonia (168/859, 19.6%).
to aminopenicillins (amoxycillin/ampicillin), 18% to Ten children with severe pneumonia died: 2 children chloramphenicol, 23% to co-trimoxazole, and 71% to with a systemic NTS infection (1 with aminopenicillin gentamicin. All isolates of NTS were susceptible to cef- resistant strains), 3 with S. pneumoniae bacteraemia, 1 triaxone and ciprofloxacin. The frequency of multidrug with S. Typhi, and 4 with negative blood cultures. Fatal- resistance (MDR, resistance against amoxycillin, chlor- ities were not observed among the patients with non- amphenicol, and co-trimoxazole) was 75.5% (74/98).
severe pneumonia.
Among all children with and without bacteraemia, 24% and 34% were positive for malaria parasites, respec- In the present study, non-typhoid salmonella (NTS) tively. Children with NTS bacteraemia had more often a were found in 10% of the blood cultures and were the co-infection with malaria parasites (24%) compared to predominant cause of bacteraemia in hospitalized chil- those with S. pneumoniae bacteraemia (18%). Malaria dren. This prevalence is, most likely, underestimated Table 1 Blood culture results in children between 2 months and 5 years of age Non-severe pneumonia* Shigella spp.
Enterococcus spp.
Streptococcus spp.
Klebsiella spp.
Pseudomonas spp.
Acinetobacter spp.
* Pneumonia cases according to WHO-IMCI definition. Agogo Presbyterian Hospital 09/2007-07/2009# Severe pneumonia included the WHO definitions for "very severe pneumonia" and "severe pneumonia" Schwarz et al. BMC Infectious Diseases 2010, 10:319 Table 2 Proportion susceptible to antibiotics among NTS Republic of Congo [24], Malawi [8,25], The Gambia and S. pneumoniae isolates [26,27] and Kenya [28]. A study from Tanzania has demonstrated strong association of antimicrobial resis- tance with fatal courses of children with bloodstream infections [29]. Fifteen years ago, no resistance to chlor- amphenicol and 57% resistance to ampicillin was reported in NTS collected from children in Accra, Ghana [16]. A more recent study from Accra found MDR in 87% of Salmonella serogroup B isolates [30]. In Northern Ghana, among 6 isolates of Salmonella spp.
from children with diarrhoea, only one was resistant to ampicillin and chloramphenicol [31]. This demonstrates that marked local and temporal differences of antibiotic resistance patterns of NTS exist [32].
Ceftriaxone or ciprofloxacin may be appropriate alter- natives to treat bacterial infections causing pneumonia *Blood cultures from children aged 2 months to 5 years of age. Agogo in areas with prevalences of NTS. Notably, 50% of the Presbyterian Hospital 09/2007-07/2009 pneumococcal isolates were resistant to ciprofloxacin inthe present study compared to only 4% of all pneumo- due to the small blood volumes that were available in coccal and NTS isolates in 1997 [33]. However, the use the paediatric patients and the frequent self-medication of ceftriaxone in low-resource countries has consider- with antibiotics prior to blood sampling. It has been able disadvantages which have to be considered: (i) the reported that the sensitivity of blood cultures fell by drug has to be adminstered intravenously or intramus- almost one third if a blood volume of 1 ml was inocu- cularly; (ii) well trained staff is necessary; (iii) uncritical lated to blood culture bottles compared to a volume of use may promote the development of resistance; (iv) cef- 3 ml as recommended [23].
triaxone is approximately 3 to 10 times more expensive The frequency of NTS was similar among children than amoxicillin.
with pneumonia and those without pneumonia and itcannot be determined whether bacteraemia due to NTS is the exclusive cause of pneumonia or predisposes to Aminopenicillins may often not be adequate for the secondary infections with other microorganisms. How- treatment of children presenting with pneumonia symp- ever, from a practical point of view this may be irrele- toms in Sub-saharan Africa and NTS in children with vant: if NTS occur frequently among children with pneumonia should strongly be considered as possible severe respiratory symptoms, empirical treatment of infectious agent. The development of alternative drugs pneumonia must include coverage of NTS.
and, best, vaccines against NTS as well as appropriate In Africa, aminopenicillins or co-trimoxazole are cur- surveillance of this neglected infectious disease is rently recommended for empirical treatment of commu- nity-acquired pneumonia in children, assuming that themost frequent causal infectious organisms are S. pneu-moniae and H. influenzae. The IMCI-WHO guidelines recommend treating non-severe pneumonia with cotri- We thank the chirdren and their parents for their participation in this study.
We are also grateful for the continuous endeavours of fieldworkers and moxazole or amoxicillin given orally, severe pneumonia interviewers of the Kumasi Centre for Collaborative Research in Tropical with oral amoxicillin or injectable penicillin and very Medicine (KCCR), and to the staff of the Public Health Unit of the Agogo severe pneumonia with injectable ampicillin plus inject- Presbyterian Hospital for their collaboration. We gratefully acknowledge thefinancial support received from a Swiss Foundation. The findings and able gentamicin [3].
conclusions in this paper are those of the authors and do not necessarily Empirical treatment of childhood pneumonia with the represent the views of any of their affiliated Research Institutions.
aforementioned first line drugs may readily fail due to the high frequency of drug resistance of NTS [20]. In 1Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany. 2Kumasi the present study, only 15.5%, 23.5% and 70.9% of NTS Centre for Collaborative Research in Tropical Medicine, Kumasi, Ghana.
isolates were sensitive to aminopenicillins, co-trimoxa- International Vaccine Institute, Seoul, South Korea. 4University Hospital of Wales, Cardiff, UK. 5School of Medical Sciences, Kwame Nkrumah University zole and gentamicin, respectively. Multi-drug resistance of Science and Technology, Kumasi, Ghana.
against the three standard drugs amoxicillin, chloram- Authors' contributions phenicol and co-trimoxazole was 75.5%. High levels of Authors contribution: JM, CGM, JAE, and YAS planned and initiated the MDR have also been reported from the Democratic study. NS organised the day-to-day work as study coordinator. FH planned Schwarz et al. BMC Infectious Diseases 2010, 10:319 and established the microbiological laboratory facilities at the Agogo 18. Graham SM, Mtitimila EI, Kamanga HS, Walsh AL, Hart CA, Molyneux ME: Presbyterian Hospital. SEKA, AA and BN carried out the microbiological Clinical presentation and outcome of Pneumocystis carini pneumonia in cultures and antimicrobial resistance testing. DD substantially contributed to Malawian children. Lancet 2000, 355(9201):369-373.
microbiological work-up and the preparation of the manuscript. RMH, JF and 19. Sigauque B, Roca A, Mandomando I, Morais L, Quinto L, Sacarlal J, FM contributed to the analysis and the writing of the manuscript. WL Macete E, Nhamposa T, Machevo S, Aide P, et al: Community-acquired created Case Record Forms and was responsible for data management and bacteremia among children admitted to a rural hospital in Mozambique.
data preparation for analyses. NGS designed the analysis protocol and The Pediatric infectious disease journal 2009, 28(2):108-113.
analysed the data. JM, NGS, and CGM wrote the manuscript. All authors 20. Graham SM, English M: Non-typhoidal salmonellae: a management have read and approved to the final version of this manuscript. Parts of this chal enge for children with community-acquired invasive disease in research have been financially supported by a Swiss Foundation.
tropical African countries. Lancet 2009, 373(9659):267-269.
21. World Health Organisation: Pocket book of hospital care for children: Competing interests guidelines for the management of common il nesses with limited The authors declare that they have no competing interests.
ressources. WHO, Geneva; 2005.
22. Eisenach K, Dyke J, Boehme M, Johnson B, Cook MB: Pediatric blood Received: 10 May 2010 Accepted: 4 November 2010 culture evaluation of the BACTEC PEDS Plus and the DuPont Isolator 1.5 Published: 4 November 2010 systems. Diagn Microbiol Infect Dis 1992, 15(3):225-231.
23. Berkley JA, Lowe BS, Mwangi I, Wil iams T, Bauni E, Mwarumba S, Ngetsa C, Slack MP, Njenga S, Hart CA, et al: Bacteremia among children admitted World Health Organisation: The global burden of disease: 2004 update to a rural hospital in Kenya. N Engl J Med 2005, 352(1):39-47.
(2008). WHO, Geneva; 2008 [http://www.who.int/healthinfo/ 24. Bahwere P, Levy J, Hennart P, Donnen P, Lomoyo W, Dramaix-Wilmet M, Butzler JP, De Mol P: Community-acquired bacteremia among hospitalized children in rural central Africa. Int J Infect Dis 2001, You D, Wardlaw T, Salama P, Jones G: Levels and trends in under-5 mortality, 1990-2008. Lancet 2010, 375(9709):100-103.
25. Gordon MA, Graham SM, Walsh AL, Wilson L, Phiri A, Molyneux E, Zijlstra EE, World Health Organisation: Technical update of the guidelines on the Heyderman RS, Hart CA, Molyneux ME: Epidemics of invasive Salmonella Integrated Managment of Childhood Illness (IMCI): evidence and enterica serovar enteritidis and S. enterica Serovar typhimurium recommendations for further adaptations. WHO, Geneva; 2005.
infection associated with multidrug resistance among adults and Brent AJ, Oundo JO, Mwangi I, Ochola L, Lowe B, Berkley JA: Salmonella children in Malawi. Clin Infect Dis 2008, 46(7):963-969.
bacteremia in Kenyan children. Pediatr Infect Dis J 2006, 25(3):230-236.
26. Enwere G, Biney E, Cheung YB, Zaman SM, Okoko B, Oluwalana C, Mankhambo LA, Chiwaya KW, Phiri A, Graham SM: Lobar pneumonia Vaughan A, Greenwood B, Adegbola R, Cutts FT: Epidemiologic and caused by nontyphoidal Salmonella in a Malawian child. Pediatr Infect Dis clinical characteristics of community-acquired invasive bacterial J 2006, 25(12):1190-1192.
infections in children aged 2-29 months in The Gambia. Pediatr Infect Dis Cheesbrough JS, Taxman BC, Green SD, Mewa FI, Numbi A: Clinical J 2006, 25(8):700-705.
definition for invasive Salmonel a infection in African children. The 27. Ikumapayi UN, Antonio M, Sonne-Hansen J, Biney E, Enwere G, Okoko B, Pediatric infectious disease journal 1997, 16(3):277-283.
Oluwalana C, Vaughan A, Zaman SM, Greenwood BM, et al: Molecular Evans JA, Adusei A, Timmann C, May J, Mack D, Agbenyega T, epidemiology of community-acquired invasive non-typhoidal Salmonella Horstmann RD, Frimpong E: High mortality of infant bacteraemia among children aged 2 29 months in rural Gambia and discovery of a clinically indistinguishable from severe malaria. QJM 2004, 97(9):591-597.
new serovar, Salmonella enterica Dingiri. J Med Microbiol 2007, 56(Pt Graham SM, Walsh AL, Molyneux EM, Phiri AJ, Molyneux ME: Clinical presentation of non-typhoidal Salmonel a bacteraemia in Malawian 28. Bejon P, Mwangi I, Ngetsa C, Mwarumba S, Berkley JA, Lowe BS, Maitland K, children. Transactions of the Royal Society of Tropical Medicine and Hygiene Marsh K, English M, Scott JA: Invasive Gram-negative bacilli are frequently resistant to standard antibiotics for children admitted to hospital in Green SD, Cheesbrough JS: Salmonella bacteraemia among young Kilifi, Kenya. J Antimicrob Chemother 2005, 56(1):232-235.
children at a rural hospital in western Zaire. Ann Trop Paediatr 1993, 29. Blomberg B, Manji KP, Urassa WK, Tamim BS, Mwakagile DS, Jureen R, Msangi V, Tellevik MG, Holberg-Petersen M, Harthug S, et al: Antimicrobial 10. Lepage P, Bogaerts J, Van Goethem C, Ntahorutaba M, Nsengumuremyi F, resistance predicts death in Tanzanian children with bloodstream Hitimana DG, Vandepitte J, Butzler JP, Levy J: Community-acquired infections: a prospective cohort study. BMC Infect Dis 2007, 7:43.
bacteraemia in African children. Lancet 1987, 1(8548):1458-1461.
30. Mil s-Robertson F, Crupper SS, Addy ME, Mensah P: Antibiotic resistance 11. Mandomando I, Macete E, Sigauque B, Morais L, Quinto L, Sacarlal J, and genotyping of clinical group B Salmonella isolated in Accra, Ghana.
Espasa M, Valles X, Bassat Q, Aide P, et al: Invasive non-typhoidal J Appl Microbiol 2003, 94(2):289-294.
Salmonel a in Mozambican children. Trop Med Int Health 2009, 31. Djie-Maletz A, Reither K, Danour S, Anyidoho L, Saad E, Danikuu F, Ziniel P, Weitzel T, Wagner J, Bienzle U, et al: High rate of resistance to locally 12. Morpeth SC, Ramadhani HO, Crump JA: Invasive non-Typhi Salmonella used antibiotics among enteric bacteria from children in Northern disease in Africa. Clin Infect Dis 2009, 49(4):606-611.
Ghana. J Antimicrob Chemother 2008, 61(6):1315-1318.
13. Nesbitt A, Mirza NB: Salmonella septicaemias in Kenyan children. J Trop 32. Weinberger M, Keller N: Recent trends in the epidemiology of non- Pediatr 1989, 35(1):35-39.
typhoid Salmonel a and antimicrobial resistance: the Israeli experience 14. O'Dempsey TJ, McArdle TF, Lloyd-Evans N, Baldeh I, Laurence BE, Secka O, and worldwide review. Curr Opin Infect Dis 2005, 18(6):513-521.
Greenwood BM: Importance of enteric bacteria as a cause of pneumonia, 33. Holliman RE, Liddy H, Johnson JD, Adjei O: Epidemiology of invasive meningitis and septicemia among children in a rural community in The pneumococcal disease in Kumasi, Ghana. Transactions of the Royal Society Gambia, West Africa. The Pediatric infectious disease journal 1994, of Tropical Medicine and Hygiene 2007, 101(4):405-413.
15. Walsh AL, Phiri AJ, Graham SM, Molyneux EM, Molyneux ME: Bacteremia in Pre-publication history febrile Malawian children: clinical and microbiologic features. The The pre-publication history for this paper can be accessed here: Pediatric infectious disease journal 2000, 19(4):312-318.
16. Wilkens J, Newman MJ, Commey JO, Seifert H: Salmonella bloodstream infection in Ghanaian children. Clin Microbiol Infect 1997, 3(6):616-620.
Cite this article as: Schwarz et al.: Systemic bacteraemia in children 17. Berkley JA, Maitland K, Mwangi I, Ngetsa C, Mwarumba S, Lowe BS, presenting with clinical pneumonia and the impact of non-typhoid Newton CR, Marsh K, Scott JA, English M: Use of clinical syndromes to salmonella (NTS). BMC Infectious Diseases 2010 10:319.
target antibiotic prescribing in seriously ill children in malaria endemicarea: observational study. BMJ (Clinical research ed 2005, 330(7498):995.

Source: http://sph.ug.edu.gh/lecturer/fobil/10_Schwarz_BMCInfDis2010.pdf