Prevalence of positive blood cultures in pediatric emergency room

1 Introduction

Paired blood cultures (BCs) are the diagnostic method of choice in suspected cases of bacteremia (1, 2); however, consensus regarding the indications for BCs in pediatric emergency rooms (PER) is still under development.

In the past decade, infectious processes have increased by 15%−20% in PER. Infectious diseases and fever ranked among the top five reasons for PER consultation, whereas bacteremia and sepsis were associated with short-term mortality. The epidemiology of bacteremia has changed in recent years, influenced by factors such as vaccination, widespread use of broad-spectrum antibiotics, and the growing number of immunocompromised children or those with invasive devices (3–6).

Currently, there is no definitive tool to clearly distinguish between true bacteremia and blood culture contamination. Thus, interpretation of BC results relies on clinical judgment. The diagnostic yield of BCs in PERs is low; the estimated yield of a single BC is approximately 1%, whereas that of paired BCs is approximately 7.5%. Moreover, the false-positive rate can reach 50%, and the contamination rates range from 1 to 3%, posing a significant diagnostic challenge (5, 7).

In emerging economies such as Mexico, in public general hospitals, BC collection is included in the care protocols for febrile children, those with suspected bacteremia, sepsis, or septic shock, and particularly in those with suspected neuroinfection (4–12). The results are not immediately available; the earliest reports of culture positivity typically occurred after 24 h.

The primary objective of this study was to determine the prevalence of positive and contaminated BCs in patients in PER. Secondary objectives included describing the demographic, clinical, and laboratory characteristics; sepsis scores; and discharge outcomes of children with positive BCs.

2 Materials and methods

This descriptive cross-sectional study was conducted from January to June 2024 in a tertiary-care general hospital serving an open population in Western Mexico. The study was approved by the institution's local ethics and research committee (approval code: CEI 235/24).

In this study, pediatric emergency room (PER) admission refers to pediatric patients who were evaluated, stabilized, and, when necessary, temporarily admitted to a pediatric emergency bed located within the emergency department. These short-stay beds are designated for observation, advanced monitoring, or initiation of critical care procedures prior to definitive disposition. This differs from inpatient admission, which involves transfer to hospital wards or the Pediatric Intensive Care Unit (PICU) for ongoing management after the emergency phase.

2.1 Inclusion criteria

All BCs reported by the hospital's Clinical Bacteriology Laboratory and obtained from pediatric patients in the PER were included.

In this study, pediatric emergency room (PER) admission refers to pediatric patients who were evaluated, stabilized, and, when necessary, temporarily admitted to a pediatric emergency bed located within the emergency department. These short-stay beds are designated for observation, advanced monitoring, or initiation of critical care procedures prior to definitive disposition. This differs from inpatient admission, which involves transfer to hospital wards or the Pediatric Intensive Care Unit (PICU) for ongoing management after the emergency phase.

2.2 Procedure

For each positive BC report, the patient's clinical course was reviewed from admission to discharge using the medical records. The patient data were coded and entered into a database to ensure anonymity and confidentiality. Sociodemographic variables, clinical presentation at admission, hospital course, antibiotic use, need for advanced support, and outcomes were also extracted. Data collection included the clinical parameters at PER admission, hospital evolution, biomarkers, and outcomes. This study followed the STROBE guidelines for observational research.

2.3 Blood culture procedure

Blood samples for culture are usually obtained by medical residents from the Pediatric Emergency o Pediatrics specialties, and less frequently by Emergency Medicine specialists. Two peripheral blood cultures are sequentially collected. In critically ill patients, central and peripheral sample are obtained immediately after central venous catheter (CVC) insertion. The procedure includes hand hygiene, sterile gown and gloves, aseptic skin preparation, sterile draping, and venipuncture. Depending on the child age, 1–5 ml of blood was drawn and immediately inoculated into culture bottles previously disinfected with 70% alcohol by a nurse preparing the specimen.

2.4 Statistical analysis plan

Descriptive statistics were used for statistical analysis. Prevalence was reported as absolute and relative frequencies (%). Measures of central tendency for quantitative variables were expressed as medians, with minimum and maximum values used to describe dispersion due to non-parametric distributions because it corresponds to small sample.

3 Results

During the study period (January–June 2024), 12,474 pediatric emergency visits were recorded for patients aged 0–14 years old.

3.1 Frequency of positive and contaminated blood cultures

A total of 531 blood cultures (BCs) were collected from hospitalized children in the PER, of which 49 (9%) showed, bacterial growth, corresponding to 25 patients; of these, three were classified as contaminated. Therefore, 46 blood cultures were found to be positive (8%; Figure 1).

3.2 Clinical characteristics of patients with positive blood cultures

The median age of the patients was 7 years (range, 3–10 years), with a predominance of male patients (Table 1). A total of 76.9% (n = 20) had received antibiotics within 48 h prior to admission, and 24% (n = 6) had been treated with broad-spectrum antibiotics (vancomycin, meropenem, ceftriaxone, and linezolid), typically initiated in external medical facilities. Antibiotic polypharmacy was observed in 40% (n = 10) patients.

Table 1. Frequency distribution of children with positive blood culture in pediatric emergency room.

February had the highest number of positive BCs (n = 9; 36%), whereas only one case (4%) was reported in March.

3.3 Admission diagnoses

The main admission diagnoses among children with positive BCs were septic shock in 10 cases (38%), sepsis in seven cases (28%), bacteremia without sepsis in four cases (16%), pneumonia in one case (4%), and neuroinfection in four cases (16%; Table 1).

Most patients (66%) were from metropolitan areas, and 1% were referred from neighboring states.

3.4 Clinical condition at admission

All patients with positive BCs presented with abnormal vital signs upon admission, with tachypnea being the most frequent finding (92%, n = 23). Vasoactive agents were initiated within the first hour in 36% (n = 9) of the cases, and 44% (n = 11) required invasive mechanical ventilation.

3.5 Results by type of blood culture

Among the 25 patients with positive BCs, 11 (44%) had positive results from blood samples collected via a central venous catheter (CVC), obtained exclusively at the time of catheter insertion in the PER, always paired with a peripheral BC. BCs collected from CVCs reported bacterial growth within less than 24 h, in contrast to peripheral samples (Table 2). Among them, 50% had septic shock, and 14.3% had sepsis.

Table 2. Frequency distribution by isolated agent.

^*Rothia dentocariosa positive in cerebrospinal fluid culture in child.

3.6 Clinical outcomes

Five patients with positive BCs died: two (8%) within the first 24 h while still in the PER and three (12%) in the following days while hospitalized in the Pediatric Intensive Care Unit (PICU). This represents a 50% mortality rate among patients admitted for septic shock and positive BC.

4 Discussion

Detection of bacteremia in the pediatric emergency room remains a critical clinical challenge owing to its strong association with high morbidity and mortality. In recent years, the epidemiology of bacteremia has shifted, influenced by the introduction of new vaccines, a growing population of immunocompromised patients, increased use of invasive devices, and widespread administration of broad-spectrum antibiotics, among other factors (5, 7, 13).

In this study, microorganisms typically classified as commensal flora or contaminants, such as Staphylococcus haemolyticus and Staphylococcus epidermidis, were isolated from patients presenting with severe infectious syndromes including sepsis and septic shock. For example, a 10-year-old patient with septic shock had a central blood culture positive for S. haemolyticus within the first 24 h, followed by a paired peripheral culture that yielded the same pathogen more than 90 h later, effectively ruling out contamination. In another case, S. haemolyticus was isolated from a child with pneumonia, parapneumonic effusion, or septic shock. Several immunocompromised children had showed similar isolates.

This finding is consistent with previous reports highlighting the difficulty in distinguishing true bacteremia from contamination when organisms such as coagulase-negative staphylococci, viridans group streptococci, Corynebacterium spp., Propionibacterium acnes, Bacillus spp., and Clostridium species have been identified (14–17). In this study, three children with positive blood cultures for these organisms were admitted to a critical condition and later died, underscoring the importance of interpreting BC results within each patient's clinical context (Table 1).

Notably, the hospital where this study was conducted is a referral center for pediatric trauma and complex medical and surgical emergencies. As such, a significant proportion of patients are transferred from other hospitals, often with pre-inserted central venous catheters (CVCs) and prior exposure to multiple antibiotics including broad-spectrum agents. The detection of microorganisms traditionally regarded as contaminants in critically ill children reinforces the importance of clinical context for accurate interpretation.

Empirical antibiotic therapy in pediatric patients with suspected sepsis, bacteremia, or septic shock, particularly in those with CVCs or severe comorbidities, is both clinically indicated and essential. Our findings do not question the appropriateness of such management but rather highlight how prior antibiotic exposure may influence blood culture yield and interpretation, occasionally resulting in false-negative or delayed identification of pathogens. This complexity reflects the diagnostic challenges faced in tertiary emergency settings, where multidrug-resistant organisms and pre-hospital antimicrobial use are common. Recent consensus statements reinforce that empirical antibiotics should be initiated based on clinical presentation and validated biomarkers, whereas blood cultures remain critical for pathogen identification and antimicrobial adjustment (5, 18).

Although a single positive BC result from multiple extractions may suggest contamination (5), this is not an absolute rule. In this study, a patient with pustular melanosis and sepsis with a positive S. haemolyticus culture was admitted to the PICU, reinforcing the necessity of integrating the clinical context for diagnostic interpretation (19).

The proportion of PER consultations for suspected infections has increased to 15%−20% of daily visits, along with an increase in clinical severity and short-term in-hospital mortality (20). During the COVID-19 pandemic, increased strains in PERs and the complexity of bacterial infections have intensified this trend (21) (Table 1).

Blood cultures are routinely requested as part of the initial evaluation of febrile children at a risk of serious bacterial infections. However, their results are not immediately available and typically require more than 24 h for confirmation. This delay justifies their use in conjunction with prompt antibiotic administration as treatment delays are associated with increased sepsis-related mortality (5, 11, 22–24).

The clinical utility of BCs in immunocompetent adults has been questioned (1). In pediatrics, however, evidence remains limited, and few studies have thoroughly characterized patients with positive BCs in PERs (25, 26). The overall yield of BCs in PERs remains low, with positivity rates of 1% for single cultures and up to 7.5% for paired cultures (24, 27). Furthermore, up to 50% of positive results may be false positives, resulting in significant clinical and economic consequences including unnecessary antibiotic use, prolonged hospitalization, and increased healthcare costs (7, 24).

At the study hospital, BCs were routinely obtained in pairs from two separate sites. Nonetheless, controversy exists over the utility of cultures obtained from CVCs, even when collected at the time of insertion (28, 29).

Differentiating between viral and bacterial infections remains a major clinical dilemma in PERs (30). Although most febrile cases are attributed to viral infections, a subset of them may follow a potentially fatal course. Accurate differentiation is essential for decisions regarding hospitalization, antibiotic use, and further testing (31). At the study site, BCs were reserved for children with specific clinical criteria requiring hospitalization; and were not performed in an outpatient setting.

Although blood culture remains the gold standard for diagnosing bacteremia, its sensitivity and specificity are limited by both operational and clinical factors. Therefore, biomarkers such as age-adjusted white blood cell count, neutrophil percentage, C-reactive protein, and procalcitonin have been proposed (6, 11, 32–34). Although clinical scoring systems have been suggested, they lack sufficient sensitivity for early detection (26, 34). Tools such as the Phoenix Sepsis Score have important limitations, particularly in viral infections (e.g., adenovirus, RSV, and dengue) and neonates, where applicability is reduced (5, 35–37).

The hospital's microbiology laboratory employs automated incubation and detection systems, which have shown greater efficiency and lower contamination rates than conventional methods. The inclusion of resins and charcoal in culture bottles enhances bacterial recovery by neutralizing residual antibiotics (7, 38).

The low rate of positive BCs may be partially explained by intermittent bacteremia, which may go undetected even with appropriate sample volumes. Intermittent bacteremia caused by Staphylococcus aureus has been estimated in up to 13% of cases (27). Thus, both the number of cultures and volume extracted per episode are critical for increasing diagnostic sensitivity (39). International guidelines recommend collecting 2–4 BC sets from different venipuncture sites, as a single culture has low sensitivity and may lead to diagnostic uncertainty. In pediatrics, available studies are limited and show heterogeneous results (27).

Regarding contamination, many PERs report rates above the national standard (2%−3%), with some series reaching up to 11% (7, 17, 26). The Spanish Society for Pediatric Emergency Medicine and Spanish Society for Pediatric Infectious Diseases have issued guidelines for BC collection, although there is no consensus (32).

BCs remain the diagnostic tool of choice in immunocompromised children. However, false positives in automated systems can lead to diagnostic confusion and therapeutic error. Factors such as a high patient volume, emergency resuscitation procedures, and frequent staff rotation contribute to elevated contamination rates (26, 40).

There are clinical scenarios in which the diagnostic value of BCs is very limited, such as febrile, immunocompetent children with a good general condition, full vaccination status, no medical devices, and no relevant medical history (1, 4, 6, 7, 24, 41). This subgroup represents an opportunity to rationalize BC use and improve diagnostic precision (1, 3, 25, 33, 42).

Microorganisms traditionally considered contaminants include coagulase-negative staphylococci, Propionibacterium spp., Bacillus spp., Corynebacterium spp., viridans group streptococci, Clostridium spp., and Micrococcus spp. (43). Automated systems may also yield false positives owing to slow-growing organisms, anaerobic conditions, L-form bacterial transformation from antibiotics, or genetic material contamination in bottles (26, 40). Several interrelated factors may contribute to elevated blood culture contamination rates in pediatric emergency departments. These include time pressure during critical resuscitations, limited availability of dedicated phlebotomy staff, and sampling performed by personnel during training, which may increase the variability in aseptic technique. Environmental crowding, suboptimal workspace organization, and interruptions during procedures have also been reported to compromise sterility. Moreover, difficulty in obtaining venous access in infants and small children, the use of small-gauge needles, and emergency sampling under unstable hemodynamic conditions further increase the risk of skin flora contamination. Institutional policies promoting rapid collection before antibiotic administration may inadvertently reduce adherence to aseptic standards.

Together, these factors illustrate the complex operational challenges in public tertiary pediatric emergency care in a developing country and underscore the importance of continuous staff training, procedural supervision, and adherence to standardized aseptic protocols to minimize contamination.

In such cases, clinical interpretation is indispensable. Although molecular techniques such as PCR offer high sensitivity, their limitations include high cost and the restricted range of pathogens included in the commercial panels (40). Notably, 98.2% of BCs in hospitals were collected from PERs (27).

This study highlights both the utility and limitations of paired BCs as diagnostic tool for pediatric emergency room. This study has several limitations. This study was conducted at a single tertiary pediatric center, which may limit the generalizability of the findings to other settings. The sample size was relatively small, and patient heterogeneity, including differences in underlying conditions, prior antibiotic exposure, and illness severity at admission, may have influenced the blood culture results. Additionally, variability in the training and experience of the healthcare providers involved in sample collection could have introduced procedural bias.

Despite these limitations, this study provides relevant evidence to support quality improvement in pediatric emergency diagnostics and underscores the need for multicenter research with standardized sampling protocols. Although the positivity rate aligns with international standards, the contamination rate exceeded acceptable thresholds, indicating room for improvement in the collection techniques and indication criteria. Some patients continue to present with diagnostic uncertainty, especially when prior antibiotic use, interhospital referral, and high clinical severity coincide.

5 Conclusions

This study underscores the diagnostic and clinical complexity of severe infections in pediatric emergency settings and highlights the need for strict protocols to distinguish between truly positive and potentially contaminated blood cultures. Paired BCs demonstrated a positivity rate in line with international standards; however, the contamination rate exceeded acceptable margins, revealing areas for improvement in the sampling techniques and indication criteria.

The presence of microorganisms traditionally considered contaminants, such as coagulase-negative staphylococci, in critically ill patients reinforces the need for contextual interpretation of microbiological findings, particularly in children with risk factors such as immunosuppression, invasive devices, or prior antibiotic therapy. In addition to BCs, biomarkers and clinical scoring systems are indispensable for the timely diagnosis of sepsis and bacteremia. Clinical decisions must be guided by standardized protocols that account for comprehensive patient evaluation as well as the indications and techniques of BC collection.

Finally, multicenter studies with larger sample sizes and ongoing training for medical and resident staff are needed to improve the diagnostic quality, reduce contamination rates, and optimize the rational use of blood cultures in pediatric emergency care.

Generative AI statement

The author(s) declare that no Gen AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Richard Hamilton, Drexel University, United States

Reviewed by: Ozgur Karcioglu, University of Health Sciences, Türkiye

Luis Gamboa, St. Christopher's Hospital for Children, United States