Antimicrobial resistance predicts death in Tanzanian children with bloodstream infections: a prospective cohort study

Background Bloodstream infection is a common cause of hospitalization, morbidity and death in children. The impact of antimicrobial resistance and HIV infection on outcome is not firmly established. Methods We assessed the incidence of bloodstream infection and risk factors for fatal outcome in a prospective cohort study of 1828 consecutive admissions of children aged zero to seven years with signs of systemic infection. Blood was obtained for culture, malaria microscopy, HIV antibody test and, when necessary, HIV PCR. We recorded data on clinical features, underlying diseases, antimicrobial drug use and patients' outcome. Results The incidence of laboratory-confirmed bloodstream infection was 13.9% (255/1828) of admissions, despite two thirds of the study population having received antimicrobial therapy prior to blood culture. The most frequent isolates were klebsiella, salmonellae, Escherichia coli, enterococci and Staphylococcus aureus. Furthermore, 21.6% had malaria and 16.8% HIV infection. One third (34.9%) of the children with laboratory-confirmed bloodstream infection died. The mortality rate from Gram-negative bloodstream infection (43.5%) was more than double that of malaria (20.2%) and Gram-positive bloodstream infection (16.7%). Significant risk factors for death by logistic regression modeling were inappropriate treatment due to antimicrobial resistance, HIV infection, other underlying infectious diseases, malnutrition and bloodstream infection caused by Enterobacteriaceae, other Gram-negatives and candida. Conclusion Bloodstream infection was less common than malaria, but caused more deaths. The frequent use of antimicrobials prior to blood culture may have hampered the detection of organisms susceptible to commonly used antimicrobials, including pneumococci, and thus the study probably underestimates the incidence of bloodstream infection. The finding that antimicrobial resistance, HIV-infection and malnutrition predict fatal outcome calls for renewed efforts to curb the further emergence of resistance, improve HIV care and nutrition for children.


Background
One in every six African children dies before the age of five years [1]. The World Health Organization (WHO) rank the major causes of mortality in African children younger than five years as neonatal causes (26%, among which the entity "sepsis or pneumonia" contributes a quarter), pneumonia (21%), malaria (18%) diarrhea (16%) and HIV-infection (6%) [2]. Bloodstream infection is a frequent cause of morbidity and associated with mortality in excess of 25% [3]. Since bloodstream infection may occur as part of localized infections with defined foci such as pneumonia and diarrhea, its importance is not reflected in the above estimates of death causes. Bloodstream infection and malaria are practically indistinguishable by clinical examination [4], and available WHO guidelines for managing childhood illnesses fail to identify up to half of the cases of bloodstream infections [5]. A recent study from Kenya [3] found that bloodstream infection caused one quarter of all deaths of children in the hospital, outnumbering malaria deaths. Antimicrobial resistance increases worldwide and does not spare developing countries [6]. However, the impact of antimicrobial resistance on the clinical outcome of infections such as bloodstream infection has been difficult to assess due to a number of factors, including confounding by underlying diseases [7,8]. We performed a prospective cohort study to gain knowledge on the etiology and antimicrobial resistance patterns of pediatric bloodstream infections and to identify microbiologic and other risk factors for fatal outcome of these infections.
Due to the young age of the study subjects (0 -7 years), the parents or other accompanying, responsible family members were asked for written consent on behalf of the patient. Information was given in writing and verbally in the national language, Kiswahili. Written informed consent was obtained before taking blood for microbiological investigations, if feasible. However, in some circumstances, in the case of critically ill patients, blood specimens were taken based on verbal consent, since these investigations are strongly recommended as routine investigations in severely ill, febrile children, and since it would be inappropriate to delay management of such patients due to paperwork. The responsible family member was then approached in retrospect for written consent to use the specimen and information in the study. The responsible family member was allowed to opt out from the HIVtesting and only consent to participation in the blood culture part of the study. As far as possible, the treatment was guided by the test results. In the following, the term "suspected systemic infection" refers to all included patients in the study, and the term "laboratory-confirmed bloodstream infection" refers to growth of one or more clinically relevant bacterial or fungal isolates from bloodculture from a patient who also confirms to the inclusion criterion of suspected systemic infection. Viral and parasitic infections were not included in the definition of bloodstream infection.
Community-acquired infection was defined as bloodstream-infection with growth of pathogenic bacteria in a blood-culture obtained within the first 48 hours after admission. For classification purposes, the time to bloodculture was calculated as the time from admission to the time of receipt of blood culture in the laboratory, allowing for overnight delay in transport for cultures obtained in the evening or at night. A neonate who was born in hospital within the last 10 days was considered as having hospital-acquired infection. If a patient was discharged and re-admitted within ten days, the episode was considered a single admission. The Muhimbili University College of Health Research Ethics Committee approved the study protocol.

Microbiologic methods
One and five ml blood from neonates and older children, respectively, were inoculated in BACTEC Myco/F lytic blood-culturing vials (Becton Dickinson, Franklin Lakes, NJ), which supports the growth of fungi and bacteria, including mycobacteria [10]. The blood-cultures were incubated for six weeks. Positive blood-cultures were subcultured on Columbia II agar base (Oxoid Ltd, Basingstoke, UK) with five percent human blood, chocolate agar and MacConckey agar (Difco/BD Diagnostic Systems, Sparks, MI, USA). The isolates were identified by standard methods [11], including the use of API20E, API20NE and API 20 AUX systems (bioMérieux SA, Marcy l'Etoile, France). Due to economic limitations, we did not perform anaerobic culture, and, in general, only a single blood culture was taken from each patient. Thus, the study was not designed to evaluate the clinical significance of coagulasenegative staphylococci (CoNS) and bacteria of doubtful or limited pathogenicity. Consequently, CoNS and probable contaminant such as diphteroids, Bacillus species and micrococci were not considered pathogens in this study. Enterococcal isolates were included in the study as pathogens. Cultures with polymicrobial growth were considered clinically relevant if known pathogens were among the constituent isolates.
Susceptibilities against antimicrobial agents were tested by the disk diffusion method according to the Clinical and Laboratory Standards Institute guidelines [12]. Testing for minimum inhibitory concentration (MIC) of antimicrobials was not routinely performed on all isolates. However, MIC determination by E-test (AB Biodisk, Solna, Sweden) was performed for more detailed characterization of the susceptibilities for antimicrobials in Gram-negative bacteria and enterococci. Gram-negative bacteria were investigated for extended-spectrum betalactamases (ESBL) with E-test, PCR and DNA sequencing as described previously [13]. Enterococcal isolates were investigated by PCR to affirm identity (E. faecalis, E. faecium) and to detect vancomycin resistance (vanA, vanB). We verified the identity of isolates of S. aureus and detected resistance to methicillin/oxacillin with a multiplex PCR targeting the nuc gene and the mecA gene [14].
Malaria testing was performed by microscopy of Giemsastained thick and thin drop blood smears. HIV testing was performed anonymously using a rapid test for HIV1/HIV2 antibodies (ACON HIV 1/2, ACON laboratories, Inc. San Diego, California, USA). HIV antibody-positive sera from children younger than 18 months were analyzed for HIV-1 RNA by reverse transcriptase polymerase chain reaction (RT-PCR) using primers targeting the pol gene, JA17 through JA20 [15], and the vif region OG 462/502 [16].

Data analysis
Data were entered in a database based on the free-ofcharge WHONET software for surveillance of antimicrobial resistance available as a download from the World Health Organization [17,18], and further data management was done in Filemaker Database software. Statistical analysis was performed in Stata 8 (Stata Corporation, College Station, Texas, US). Univariate assessment of risk factors for intra-hospital death was done by Fisher's exact test with a two-sided P-value and odds ratios and 95% confi-dence intervals were obtained by the 'logistic' function in Stata. Significant factors to the level of P < 0.2 from the univariate analysis and a priori important factors such as sex, age and underlying diseases were included in the multivariate analysis. Multivariate analysis was performed by automated and manual backwards step-wise logistic regression where factors with P > 0.2 were removed from the model. We present four logistic regression models using different subsets of the study population analyzing cases of laboratory-confirmed bloodstream infection (n = 216) as well as cases of clinically suspected systemic infection (n = 1527), and for each category we re-analyzed the data for those who had known HIV status (n = 128 and n = 790, respectively). Comparisons of medians of time variables were done by Wilcoxon rank-sum (Mann-Whitney) test.

Patients
A total of 1787 patients with a median age of 8.5 months were admitted 1828 times. Forty-four percent (795/1787) were female. Neonates (median age 3 days, range 0-30) accounted for almost a third of the admissions (29.3%, 535/1828), and stayed a median of 6 days (range 1-53) in the hospital. Older children (median age 1 year, range 1 month -7 years) stayed a median of 7 days (range 1-78) in the hospital.

Antimicrobial therapy
Information on antimicrobial use was available for 85.2% (1557/1828) of the admissions. The majority of patients (93.8%) received antimicrobial therapy (Table 1) and at least two-thirds (67.2%, 1046/1557) of the patients did so before blood-culture was taken. There was no formal empirical regimen for the treatment of sepsis at the hospital, partly because of the scarcity of local studies on antimicrobial resistance of relevant bacterial isolates. However, most neonates received a regimen of ampicillin + cloxacillin + gentamicin. One-fifth of the neonates received ceftriaxone, in most cases as a second-line regimen in case of unsatisfactory response to the first regimen. In older children there was a more diverse pattern of regimens used, in most cases including one or more penicillins (ampicillin, penicillin G or cloxacillin) and either chloramphenicol or gentamicin. To some extent, accompanying localizing signs did impact on the choice of antimicrobial regimen, most notably in the case of suspected meningitis in children older than 1 month when a regimen based on chloramphenicol and a penicillin would often be used, and in the case of bloody diarrhea when erythromycin was often added. Severe malaria was treated with quinine, while less severe cases were treated with chloroquine in neonates and pyrimethamine-sulfadoxin in older children. Oral thrush was treated with nystatin mixture, but systemic antifungal drugs were rarely used. Ciprofloxacin was not used in children at the hospital.

Types of bloodstream infections
The incidence of laboratory-confirmed bloodstream infection was 13.9% (255/1828) of all admissions, 15.9% (85/ 535) among neonates and 13.1% (170/1293) among older children. A single pathogen was recovered from 224 children (12.3%), while 31 (1.7%) had polymicrobial infection with two (n = 26), three (n = 2) or four isolates (n = 3). In total, 294 pathogenic bacterial and fungal isolates were recovered (Table 2). Among all laboratory-confirmed bloodstream infections, half (128/255) were defined as potentially hospital-acquired. Salmonella and Escherichia coli were the most common isolates in community-acquired infections, and klebsiella and Staphylococcus aureus were the most common in hospital-acquired infections (Table 2). Klebsiella was, by far, the most common cause of neonatal bloodstream infection, particularly in early-onset infection (0-6 days of age) where it responsible for approximately one-third of the cases (Table 3). In children older than 1 month of age, salmonellae were the most frequently isolated pathogens (Table 3).
Among all children with suspected systemic infection, those with laboratory-confirmed bloodstream infection had three times increased risk of dying (Table 4, Figure 1), and among those who survived, the duration of hospital stay was significantly longer than for children without laboratory-verified bloodstream infection (median 8 versus 6 days, P < 0.001). Univariate and multivariate analysis of risk factors for death are shown in Table 4 and 5, respectively. The mortality rate from Gram-negative bloodstream infection (45.6%) was more than twice that of malaria (20.2%) and Gram-positive bloodstream infection (16.7%). Positive blood-culture with Enterobacteriaceae, other Gram-negative bacteria and candida were independent risk factors for fatal outcome (

Susceptibility to antimicrobial agents
Enterobacteriaceae displayed high rates of resistance to commonly used antimicrobials (Table 6). Only 20% of community-acquired Enterobacteriaceae isolates was sensitive to ampicillin, while two-thirds were sensitive to gentamicin. Two-thirds of community-acquired salmonellae were sensitive to chloramphenicol, while half are resistant to ampicillin and to co-trimoxazole. Isolates of Enterobacteriaceae were almost uniformly sensitive to ciprofloxacin. P. aeruginosa isolates were commonly sensitive to antipseudomonas drugs such as ceftazidime, ciprofloxacin and tobramicin (Table 7). ESBL phenotype was found in 18% of the Enterobacteriaceae isolates (E. coli 9/37, klebsiella 9/53, Enterobacter spp. 5/9, salmonella 1/39 and Pantoea spp 2/2) involving TEM-63, SHV-2a, SHV-12 and CTX-M-15 genotypes [13], and in 3 isolates of non-Enterobacteriaceae (one Acinetobacter spp. and the 2 Chryseobacterium spp.). ESBL-producing isolates were resistant to almost all tested antimicrobials except for ciprofloxacin and meropenem.
The majority of S. aureus isolates were sensitive to commonly used anti-staphylococcal agents, including cloxacillin and gentamicin (Table 8). Among the three isolates phenotypically resistant to oxacillin, only one was available for confirmatory detection of the mecA gene by PCR.
The child with confirmed MRSA genotype died, the two other survived. Ten of 21 E. faecium isolates showed combined resistance to ampicillin and gentamicin (highlevel), but only one patient with bloodstream infection caused by these organisms died. This combined resistance trait occurred in both community-acquired (4/12) and hospital-acquired infections (6/9), and the difference was not statistically different (P = 0.198). Six of 15 E. faecalis isolates were high-level gentamicin-resistant (but not ampicillin-resistant) and were involved in the death of two patients, among whom one had community-acquired infection. The majority of candida isolates were susceptible to fluconazole (96%) and amphotericin B (87%).
Hospital-acquisition of infection was significantly associated with resistance to amoxicillin-clavulanate and cephalosporins in E. coli and with co-trimoxazole-resistance in klebsiella. The percentage of bloodstream infection episodes caused by an ESBL-producing organism was not significantly different among community-acquired (7%, 9/ 127) and hospital-acquired infections 13% (16/128, P = 0.206). Patients with hospital-acquired laboratory-verified bloodstream infection did not receive inappropriate antimicrobial treatment more often (41%) than those with community-acquired infection (42%, P≅0.892).
Impact of causative organism on survival from laboratory-confirmed bloodstream infection Figure 2 Impact of causative organism on survival from laboratoryconfirmed bloodstream infection.  Figure 1 Survival from laboratory-confirmed bloodstream infection and clinically suspected systemic infection.

Survival from laboratory-confirmed bloodstream infection and clinically suspected systemic infection
Inappropriate antimicrobial treatment was a risk factors for death irrespectively of whether controls were patients with clinically suspected infection or laboratory-confirmed bloodstream infection (Table 5, Figure 3). In those who survived, inappropriate antimicrobial treatment for the bloodstream-infection was associated with increased duration of hospital stay (median 8 days versus 6 days, P = 0.043).

Malnutrition and other underlying conditions
One sixth (15.5%, 248/1603) of the patients were malnourished. Malnutrition was a risk factor for death (Table 5, Figure 6) and, in those who survived, it was associated with prolonged hospital stay (median 10 versus 6 days, P < 0.001). Other underlying diseases were grouped into Survival from community-acquired and nosocomial labora-tory-confirmed bloodstream infection Figure 4 Survival from community-acquired and nosocomial laboratory-confirmed bloodstream infection.  CA, community-acquired infection; HA, hospital-acquired infection; "-", not done. *E-test was used for susceptibility testing of enterococci against ampicillin, ciprofloxacillin, linezolid, quinupristin-dalfopristin, vancomycin and high-level gentamicin resistance. † Statistically significant difference in susceptibility among hospital-acquired and community-acquired isolates, p < 0.05 by two-sided Fisher's exact test.
Impact of antimicrobial resistance on survival from labora-tory-confirmed bloodstream infection Figure 3 Impact of antimicrobial resistance on survival from laboratory-confirmed bloodstream infection.

Discussion
Considering the frequent use of antimicrobials prior to blood-culture, the study probably underestimates the incidence of bloodstream infection, particularly episodes caused by fastidious organisms such as pneumococci and other organisms susceptible to commonly used antimicrobials. Nevertheless, the observed incidence of laboratory-confirmed bloodstream infection was high (13.9%) and comparable to that of malaria (21.6%). The therapeutic dilemma is evident, since bloodstream infection and malaria are difficult to distinguish based on clinical presentation [4,5]. While mortality was similar in bloodstream infection and malaria in a study from Rwanda in the 1980s [19], bloodstream infection carried much higher mortality than malaria in our study. As reported recently from Kenya [3], deaths from bloodstream infection outnumbered malaria deaths. The three times higher mortality among malaria-parasitemic patients with concomitant Gram-negative bloodstream infection underlines the relative importance of bloodstream infections. Prompt antimicrobial treatment is imperative for the survival of patients with bacterial bloodstream infection [20]. However, the impact of antimicrobial resistance on the clinical outcome has not been firmly ascertained, particularly in developing country settings [6][7][8]21,22]. The present study confirms that inappropriate treatment of bloodstream infections due to antimicrobial resistance predicts fatal outcome independently of underlying diseases, and is associated with longer duration of hospital stay in those who survive. While morbidity and cost are more sensitive measures of the impact of antimicrobial resistance in rich countries [7], our study demonstrates that the toll of antimicrobial resistance in Sub-Saharan Africa is quantifiable in loss of human lives.
The high rates of resistance in Gram-negative bacteria confirm previous findings [23]. However, the frequent use of antimicrobials prior to blood culture may have biased the findings by lowering the detection rate for organisms susceptible to commonly used drugs. It is grave that half of the klebsiella isolates, which are inherently resistant to ampicillin, are also resistant to gentamicin, since these two drugs are the most frequently used drugs for the treatment of bloodstream infections. The high incidence and case-fatality rate of klebsiella bloodstream infection in our study supports the findings by Zaidi and colleagues that klebsiella infections may be responsible for more than 0.3 million yearly neonatal deaths globally [22]. As reported elsewhere [19,24,25], salmonella was the most frequent cause of bloodstream infections in older children. Nontyphoid salmonella cause a wide range of manifestations from self-limiting gastroenteritis to systemic infections such as bloodstream infections and sometimes meningitis associated with high mortality [26,27]. Bloodstream infection caused by Salmonella Typhimurium was associated with fatal outcome with comparable case-fatality rates as other major Enterobacteriaceae species, while bloodstream infection caused by S. Enteritidis was associ-Impact of nutritional status on survival from laboratory-con-firmed bloodstream infection Figure 6 Impact of nutritional status on survival from laboratory-confirmed bloodstream infection.
Impact of HIV co-infection on survival from laboratory-con-firmed bloodstream infection Figure 5 Impact of HIV co-infection on survival from laboratory-confirmed bloodstream infection.
ated with lower case-fatality rates in the same range as Gram-positive bloodstream infections. The relatively low case-fatality rate from bloodstream infection caused by S. aureus reflects the frequent use of cloxacillin and the near absence of methicillin-resistance. Bloodstream infection caused by E. faecium resistant to ampicillin and gentamicin is a potentially serious problem considering the widespread reliance on these drugs for treatment. The relatively low case-fatality rate from bloodstream infection caused by these organisms may reflect an inherently less virulent nature. However, it is also possible that some of the E. faecium isolates may have been skin contaminants, in which case the study may overestimate the incidence of bloodstream infection caused by this organism.
Hospital-acquired infection was associated with increased rate of resistance to cephalosporins in E. coli and co-trimoxazole in klebsiella, but it was not a risk factor for fatal outcome. In contrast to studies on adults from Tanzania and elsewhere [39], we isolated Mycobacterium tuberculosis from blood-culture from only one of 97 children with clinical tuberculosis. This finding indicates that blood-culture is a poor diagnostic tool for pediatric tuberculosis and underlines the experience from other studies that it is difficult to establish microbiological evidence to support the diagnosis of tuberculosis in children [40].
One limitation of the study is that clinical and outcome data were missing for some patients. Furthermore, we performed neither anaerobic culture nor repeat cultures to assess the significance of coagulase-negative staphylococci. The frequent use of antimicrobials before blood-culture and the small blood volume cultured, particularly from neonates (1 ml), likely precluded the detection of pathogens, particularly fastidious organisms. The bloodculture system supported growth of pneumococci during quality testing in our laboratory. While available documentation by the start of our study supported use of the blood-culture system [10,41], a later study indicated that it was not optimal for detection of pneumococci and S. aureus [42]. Sub-culturing on human blood agar plates may have interfered with the detection of pneumococci in the study. Human blood agar, prepared from expired banked blood, is widely used for bacterial isolation in developing countries, but has recently been demonstrated to be inferior to animal blood agars (sheep or horse blood) in detecting growth of common bacteria such as pneumococci, S. pyogenes and S. aureus. The use of human blood agar leads to reduced bacterial colony size, altered colony morphology and poor hemolysis [43]. The reasons for this are not fully understood, but may include remnants of antimicrobials consumed by the blood donor, antibodies or other unknown factors. While citrate in banked human blood has been considered a possible explanation for the inferiority of human blood agar for bacterial culture, Russell and colleagues found citrated sheep blood agar to be a practical and superior alternative to human blood agar [43]. While sheep and horse blood is often not readily available in many developing countries, goat or pig blood is more accessible and affordable and could be used for agar production [44]. A further reason for abandoning the use of human blood agar is the associated biohazard to the laboratory personnel due to frequent infection with HIV, hepatitis B and C among blood donors.

Conclusion
Our study shows that antimicrobial resistance predicts fatal outcome of bloodstream infection. While malaria can be treated with highly effective, inexpensive and widely available quinine, treatment of bacterial and fungal bloodstream infection is compromised by antimicrobial resistance and unavailability of expensive second-line antibacterial and systemic antifungal drugs. The finding that antimicrobial resistance, HIV-infection and malnutrition independently predicts fatal outcome not only call for renewed efforts to improve HIV care, nutrition and general living conditions for children, but also argue for increased focus on interventions to limit the selection pressure for resistant organisms such as education on rational use of antimicrobials, development of judicious guidelines for treatment strategies and adequate government regulation of drug quality and access to antimicrobial drugs [45].