Discussion

Although it is not the most sensitive technique, blood culture is considered the gold standard and clinically relevant microbiological diagnostic tool for BSIs, directing appropriate antibiotics prescription. However, CBCs, which cause false-positive results, remain challenging for physicians and clinical microbiologists [7]. The number of cultures sets that grow specific organisms can help differentiate contamination from true infection; if only one of a given set grows an organism, this generally represents a contaminant, and multiple sets will usually grow the same organism in true-positive bacteraemia. However, many studies have suggested that these criteria are insufficient and should not be used alone to differentiate true-positive results from contaminants [8,9]. Therefore, CBCs cause the initiation of unnecessary broadspectrum antibiotic prescription. This carries a risk of modifying and changing the intestinal microbiome, and in infancy, this has been shown to increase the risk of asthma, allergies, obesity, celiac disease, and antibiotic resistance later on [10,11].

The present study showed a high percentage of CBCs (26.5%). This rate could be higher if the TPBCs due to coagulase-negative Staphylococci are included, as this group of microorganisms is generally more responsible for contamination than real infections. However, because this study is not a clinical survey, it is difficult to distinguish between CBCs and real BSIs due to coagulasenegative Staphylococci.

This high rate of CBCs in pediatric patients can be explained by difficulties in collecting blood samples from this patient group and poor aseptic techniques. In neonates and small infants, phlebotomy is difficult sometimes and potentially harmful in certain situations [12,13]. In addition, chlorhexidine gluconate products are not approved for use in infants <2 months old, and iodine-containing antiseptic is contra indicated for use in neonates due to the potential of developing subclinical hypothyroidism. Therefore, 70% isopropyl alcohol remains the most used antiseptic, which requires a longer contact time to dry, and nurses sometimes ignore this [14]. In a recent study, Maeda et al. found that chlorhexidine gluconate-ethanol skin preparations were more effective than povidone-iodine disinfection before venepuncture in children to prevent blood culture contamination [15]. Considering these, strict adherence to sterile venepuncture techniques by well-trained pediatric phlebotomy teams should be enforced.

In the present study, the incidence of HO TPBCs was slightly higher than that of CO TPBCs, with an incidence rate of three episodes per 1000 PDs in all pediatric wards and nine episodes per 1000 PDs in the pediatric and neonatal ICUs. This predominance of HO TPBC could be explained by our hospital’s type of inpatients and activities. First, most patients were transferred from other healthcare facilities because our facility is a referral center for hematopoietic stem cell and solid organ transplants. Therefore, many patients at our facility were immunocompromised. Second, many invasive and complicated procedures and surgeries are performed in our institution. These factors contributed to the incidence of nosocomial BSIs.

With regard to age, almost half of the PBCs in this study were observed in children aged between 1 month and 1 year. In a national BSI survey conducted in the United Kingdom from January 2009 to March 2010, 8718 episodes of BSI were reported; 1734 were nosocomial BSIs, giving a rate of 4.74 episodes per 1000 admissions. In addition, the median age at infection was 1 year old, and 54% of the infections were in males [16]. In another prospective study in the Pediatrics ICU of a tertially and regional referral hospital in North India, conducted during one-year (July 2003-July 2004), 116 episodes of primary BSIs were noted in 86 (30%) patients, with an incidence rate of 31.2 episodes per 1000 PDs. Furthermore, 73.3% of patients had single episode of BSIs, and 26.7% presented multiples episodes [17]. A survey conducted in 2015, including 300 children, at the Institute of Child Health and Hospital for Children, Chennai, observed that maximum culture positivity was noted in children aged 1-3 months old (2).

In the present study, the second main finding was the high prevalence of GNB, especially in HO TPBCs. During the last two decades in Saudi Arabia and many countries, there has been a change in the epidemiological and ecological profile of TPBCs among pediatric patients. Recent studies, including the present study, confirm the predominance of GNB at the expense of GPC with an increased rate of extended-spectrum β-lactamaseproducing bacteria (ESBL) [18-21]. Similar findings were reported in China, where a retrospective study conducted at thirteen teaching hospitals from 2016 to 2018 noted that among 9345 nonduplicate bloodstream isolates, the top ten collected bacteria included: CONS (44.4%), E. coli (10.2%), K. pneumoniae (5.9%), S. aureus (5.0%), Streptococcus pneumoniae (4.9%), P. aeruginosa (2.8%), Enterococcus faecium (2.7%), Stenotrophomonas maltophilia

(2.4%), Salmonella spp (2.3%), and Streptococcus agalactiae (2.0%) [22]. Additionally, a cross-sectional study performed to identify the predominant group of causative microorganisms at a tertiary healthcare in Peshawar, in 2018 reported that gramnegative and gram-positive bacteria were identified in 71.1% and 28.9% of isolates, respectively [23]. However, a retrospective survey of TPBCs from 2007 to 2016 in Spain, including level I and II hospitals, showed that the most frequently isolated bacteria was Streptococcus pneumoniae (29.2%), with a decrease in the number of cases between 2008 and 2015. Staphylococcus aureus was the second common bacteria (19.2%), and 21.7% of them were methicillin-resistant. Furthermore, Escherichia coli was the most frequent organisms in children aged <3 months old [4]. Similarly, a multicentre, prospective cohort at 10 pediatric hospitals in Switzerland reported that E coli (20%) was the main organism identified, followed by S aureus (15%), CONS (11%), and S pneumoniae (10%) [24].

Pneumococcal bacteraemia was very rare in our survey (only one case), although Streptococcus pneumoniae is the leading isolate of CO TPBCs [25]. In addition, after introducing the pneumococcal vaccine, many studies noted a net decrease in pneumococcal bacteraemia in children [26,27].

The antibiotic stewardship team at our facility needs to consider these findings. The epidemiology of BSIs is dynamic because they depend on microbiological, host, and healthcare system factors. For example, the types and frequencies of microorganisms responsible for PBCs vary between developing and developed countries, hospital levels (primary and tertiary), and in the same institution between medical, surgical wards, and ICUs. In addition, the infection origin (CO or HO) and underlying diseases play important roles in determining the type of causative bacteria.

Regarding antibiotic sensitivity and resistance patterns, this study showed that 43.5% of Enterobacteriaceae were resistant to third-generation cephalosporins; however, a low resistance rate of MRSA and VRE and the absence of carbapenemresistant Enterobacteriaceae were observed. Third-generation cephalosporins are frequently chosen as initial empirical antibiotics in many hospitals for community-acquired bacteraemia. However, recent studies have shown an increasing rate of resistant organisms in CO infections due to the overuse of ceftriaxone, cefotaxime, or ceftazidime, by the production of ESBL or AmpC beta-lactamases [2,22,28].

In addition, in a study conducted in USA during the period 2005 to 2011, on pediatric antimicrobial susceptibility trends, carbapenems were the most effective antibiotics against gram-negative organisms. Furthermore, the incidence of MRSA was 50%, and Enterococcus faecium showed 25% and 45% susceptibility to ampicillin and vancomycin, respectively [2].

Our study had some limitations. First, it was performed in a single tertiary pediatric hospital, and the incidence of serious infections is higher in tertiary centres. Therefore, this study’s results cannot be generalized to other hospitals. Second, our data were obtained and analysed retrospectively from medical patient records. Lastly, cost estimation due to false positive or contaminant results that might lead to unnecessary admissions and investigations could not be attained.

Conclusion

The main results of this study were (i) the predominance of GNB in BSIs among the pediatric population, especially when it occurs after hospital admission, (ii) the high rate of Enterobacteriaceae resistance to third-generation cephalosporins, and (iii) the low incidence of MRSA and VRE. Therefore, we recommend that our hospital’s antibiotic stewardship team considers these findings when preparing hospital guidelines for empirical antibiotics for BSIs. Finally, a high rate of CBCs was observed among the pediatric population. Hence, a quality improvement project should be implemented to investigate and take action to reduce this rate.

Acknowledgment

All authors would like to thank Saudi Pediatric Infectious Diseases Society (SPIDS) for its support to finance publication of the current study.

Funding

This work was supported by Saudi Pediatric Infectious Diseases society (SPIDS). The funding services provided by SPID has been used only to pay Editage service and publication fees of this manuscript

Conflict of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data Sharing Statement

Raw data were generated at Infection and Control Center of Excellence, Prince Sultan Military Medical City. Derived data supporting the findings of this study are available from the corresponding author Dr Naoufel Kaabia on request through his E-mail.

Authors’ Contributions

Naoufel Kaabia and Abdulrahman Alodayani are the principal investigators, they prepared the proposal, did data interpretation, and wrote the draft manuscript. Faten Basnawi did data analysis and interpretation, literature review, and wrote the draft manuscript. Nada Alharbi did data collection and data entry, prepared tables and figures. Abdullah Alfarhoud did literature review and wrote the manuscript. Rana Almaghrabi did data analysis and revised the manuscript. Nabiha Bouafia revised the manuscript. All authors have approved the final article.

Informed Consent

Not applicable in the current study, because it involved only anonymous data extracted from infection control database.

References

1. Bard JD, TeKippe EM (2016) Diagnosis of bloodstream infections in children. J Clin Microbiol 54: 1418-1424.
2. Wattal C, Goel N (2020) Pediatric blood cultures and antibiotic resistance: an overview. Indian J Pediatr 87: 125-131.
3. Dargère S, Cormier H, Verdon R (2018) Contaminants in blood cultures: importance, implications, interpretation and prevention. Clin Microbiol Infect 24: 964-969.
4. Ferreira M, Santos M, Rodrigues J, Diogo C, Resende C, et al. (2021) Epidemiology of bacteremia in a pediatric population - A 10-year study. Enferm Infecc Microbiol Clin (Engl Ed) S0213–005X.
5. Ames SG, Davis BS, Angus DC, Carcillo JA, Kahn JM (2018) Hospital variation in risk-adjusted pediatric sepsis mortality. Pediatr Crit Care Med 19: 390-396.
6. Roth A, Wiklund AE, Pålsson AS, Melander EZ, Wullt M, et al. (2010) Reducing blood culture contamination by a simple informational intervention. J Clin Microbiol 48: 4552-4558.
7. Hemeg HA, Almutairi AZ, Alharbi NL, Alenezi RF, Alturkostani MA, et al. (2020) Blood culture contamination in a tertiary care hospital of Saudi Arabia. A one-year study. Saudi Med J 41: 508-515.
8. Hall KK, Lyman JA (2006) Updated review of blood culture contamination. Clin Microbiol Rev 19: 788-802.
9. Weinstein MP, Reller LB, Murphy JR, Lichtenstein KA (1983) The clinical significance of positive blood cultures: a comprehensive analysis of 500 episodes of bacteremia and fungemia in adults. I. Laboratory and epidemiologic observations. Rev Infect Dis 5: 35-53.
10. Gibson MK, Crofts TS, Dantas G (2015) Antibiotics and the developing infant gut microbiota and resistome. Curr Opin Microbiol 27: 51-56.
11. Sander SD, Andersen AMN, Murray JA, Karlstad Ø, Husby S, et al. (2019) Association between antibiotics in the first year of life and celiac disease. Gastroenterology 156: 2217-2229.
12. Alnami AY, Aljasser AA, Almousa RM, Torchyan AA, BinSaeed AA, et al. (2015) Rate of blood culture contamination in a teaching hospital: a single center study in Saudi Arabia. J Taibah Univ Medical Sci 10: 432-436.
13. Sallam L, Kaabia N, Al-Aidaroos A, Al-Darsony S, Al-Odayani A (2021) Predominant bacterial pathogens and resistance profile of blood isolates: blood cultures surveillance from tertiary teaching hospital. Acta Sci Microbiol 4: 46-53.
14. Murphy C, Andrus M, Barnes S, Garcia R, Juraja M, et al. (2009) Guide to the elimination of catheter-related bloodstream infections. Washington, DC.
15. Maeda N, Mori N, Shinjoh M, Komiyama O, Takahashi T (2021) Comparison of 0.5% chlorhexidine gluconate alcohol with 10% povidone-iodine for skin disinfection in children to prevent blood culture contamination. J Infect Chemother 27: 1027-1032.
16. Blackburn RM, Henderson KL, Minaji M, Muller-Pebody B, Johnson AP, et al. (2012) Exploring the epidemiology of hospital-acquired bloodstream infections in children in England (January 2009-March 2010) by linkage of national hospital admissions and microbiological databases. J Pediatric Infect Dis Soc 1: 284-292.
17. Lakshmi KS, Jayashree M, Singhi S, Ray P (2007) Study of nosocomial primary bloodstream infections in a pediatric intensive care unit. J Trop Pediatr 53: 87-92.
18. Al-Zamil FA (2008) Bacteremia in children at the University Hospital in Riyadh, Saudi Arabia. World J Pediatr 4: 118-122.
19. Al-Abdely HM, Alshehri AD, Rosenthal VD, Banjar W, Mohammed YK, et al. (2017) Device-associated infection rates in Intensive Care Units of five cities of the Kingdom of Saudi Arabia: International nosocomial infection control consortium (INCC) findings. Can J Infect Control 32: 31-36.
20. Bandy A, Almaeen AH (2020) Pathogenic spectrum of blood stream infections and resistance pattern in gram-negative bacteria from Aljouf region of Saudi Arabia. PLoS One 15: e0233704.
21. Babay HA, Twum-Danso K, Kambal AM, Al-Otaibi FE (2005) Bloodstream infections in pediatric patients. Saudi Med J 26: 15551561.
22. Qiu Y, Yang J, Chen Y, Yang J, Zhu Q, Zhu C, et al. (2021) Microbiological profiles and antimicrobial resistance patterns of pediatric bloodstream pathogens in China, 2016-2018. Eur J Clin Microbiol Infect Dis 40: 739-749.
23. Muhammad A, Noor Khan S, Jamal T, Ul Haq I, Jabbar A, et al. (2020) Bacterial spectrum and antimicrobial profile of pediatric blood stream infection at a tertiary care hospital in Pakistan. Chin Med Sci J 35: 315-322.
24. Agyeman PKA, Schlapbach LJ, Giannoni E, Stocker M, Posfay-Barbe KM, et al. (2017) Epidemiology of blood culture-proven bacterial sepsis in children in Switzerland: a population-based cohort study. Lancet Child Adolesc Health 1: 124-133.
25. Gomez B, Hernandez-Bou S, Garcia-Garcia JJ, Mintegi S (2015) Bacteremia in previously healthy children in emergency departments: clinical and microbiological characteristics and outcome. Eur J Clin Microbiol Infect Dis 34: 453-460.
26. Greenhow TL, Hung YY, Herz A (2017) Bacteremia in children 3 to 36 months old after introduction of conjugated Pneumococcal vaccines. Pediatrics 139: e20162098.
27. Zaitoon H, Bamberger E, Yaniv L, Mendelson B, Srugo I, et al. (2020) A retrospective study of the impact of Pneumococcal conjugate vaccine-13 immunization in a Northern Israel hospital. Isr Med Assoc J 22: 71-74.
28. Mao S, Ge Z, Zhao H, Cao J, Xia Z (2019) Analysis on distribution and drug resistance of pathogen caused community-onset bloodstream infection. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 31: 67-72.