Sepsis represents a series of clinical syndromes that are caused by a dysregulated host response to infection, which leads to widespread inflammation, tissue damage, and potentially life-threatening multiple organ dysfunction (Singer et al., 2016; Schlapbach et al., 2024). Sepsis poses a significant global health challenge, with approximately 50 million cases reported annually, half of which occur in children under 19 years of age (Rudd et al., 2020). In 2017, the World Health Assembly formally recognized sepsis as a critical focus area for the next decade through a global resolution (Kissoon et al., 2017; Reinhart et al., 2017). Notably, a substantial number of sepsis-related deaths occur shortly after hospital admission, primarily due to severe septic shock and rapid organ failure (Härtel et al., 2020). Furthermore, children who survive sepsis often experience long-term complications, including organ dysfunction or neurodevelopmental impairments (Watson et al., 2024). Therefore, the rapid and accurate recognition of septic shock, followed by timely and effective interventions, is important.

Conventional blood cultures have long been considered the gold standard for diagnosing bacterial sepsis. However, their long processing times and susceptibility to contamination limit their clinical utility. As a result, researchers have used clinical symptoms and laboratory biomarkers to predict or diagnose sepsis and septic shock in pediatric patients. However, the clinical presentation of pediatric sepsis is often ambiguous, and current available laboratory markers lack sufficient sensitivity and specificity, presenting a significant challenge for timely diagnosis (Chiesa et al., 2004).

Current protocols for diagnosing pediatric sepsis and septic shock rely on clinical criteria and biomarkers, such as C-reactive protein (CRP), procalcitonin (PCT), and white blood cell (WBC) counts. While these markers have demonstrated utility in identifying inflammation and infection, their diagnostic accuracy remains limited, especially in pediatric patients. CRP, for instance, is a widely used acute-phase biomarker; however, its delayed elevation—requiring 24–48 hours to peak—reduces its sensitivity in detecting early infections.

Furthermore, CRP is non-specific and may be elevated in a variety of inflammatory conditions, limiting its diagnostic precision. Similarly, PCT, which is an immunologically active protein that is released during systemic inflammatory responses, plays a role in immune regulation and vascular contraction. Its levels increase within 3–4 hours of bacterial infection and peak at 24 hours, making it a more dynamic marker than CRP (Reinhart et al., 2000). However, PCT levels can also increase in noninfectious inflammatory conditions, such as severe trauma or burns, which reduces its specificity. In addition, the high cost and limited availability of PCT testing in primary hospitals and resource-limited settings restrict its routine application.

Recently, nucleated red blood cells (NRBCs) have emerged as a potential biomarker for pediatric sepsis. NRBCs are immature precursors of red blood cells that are not present in the circulation of healthy children (older than 28 days) or adults (healthy newborns have circulating NRBCs that rapidly disappear within a few weeks after birth) (Pedersen et al., 2022). NRBCs help diagnose hematological disorders related to erythropoiesis, anemia, or hemolysis. Recent studies have shown that this economical and easily accessible analysis has potential for the diagnosis and prognosis of critically ill adults and preterm infants (Christensen et al., 2011; Desai et al., 2012; Narcı et al., 2021). However, the value of NRBCs in predicting septic shock remains largely unknown.

Given the limitations of individual conventional biomarkers, combining multiple parameters may enhance diagnostic accuracy and prognostic evaluation. Therefore, this study aimed to analyze the predictive performance of individual biomarkers, including CRP, PCT, and NRBCs, in identifying septic shock. Furthermore, we aimed to compare the predictive performance of NRBCs with that of traditional biomarkers and evaluate the combined predictive utility of CRP, PCT, and NRBCs. By exploring the complementary roles of these biomarkers, we hope to provide a more reliable approach for the early diagnosis and management of pediatric sepsis.

This study evaluated the individual and combined diagnostic accuracies of CRP, PCT and NRBCs for predicting the occurrence and severity of sepsis in the pediatric population. Our findings revealed that NRBC counts, when analyzed alongside CRP and PCT levels, are valuable biomarkers for distinguishing between sepsis and septic shock. The combination of these three biomarkers demonstrated superior predictive accuracy compared with CRP or PCT alone, highlighting the enhanced diagnostic potential of this panel. These results emphasize the role of NRBCs as a complementary biomarker that not only improves diagnostic precision but also reflects the physiological stress and disease severity associated with sepsis.

Sepsis is a life-threatening condition that imposes a substantial burden worldwide. According to the Global Burden of Disease (GBD) study, approximately 25 million sepsis cases occur annually in pediatric and neonatal populations, resulting in approximately 3 million deaths (Watson et al., 2024). The absence of definitive biomarkers for the timely diagnosis of sepsis exacerbates the inappropriate use of antibiotics. Therefore, identifying a potential combination of biomarkers to improve the early diagnosis of sepsis is imperative.

In this study, the positive rates of blood culture in sepsis and septic shock patients were 12.5% and 36.5%, respectively. The most common bacterial isolate was Escherichia coli. These findings differ from those of a meta-analysis by Droz et al (Droz et al., 2019), which reported a pooled blood culture positivity rate of 28% (95% CI: 13.2–45.8%) among pediatric populations in low- and middle-income Asian countries. The lower positivity rate observed in our sepsis cohort may reflect differences in patient demographics, clinical settings, or regional pathogen prevalence. Although blood culture is considered the gold standard for diagnosing sepsis, its sensitivity is influenced by multiple factors. For example, delays in sample collection, previous antibiotic administration, technical limitations of detection methods, and insufficient blood sample volume all contribute to variability in positive rates (Lambregts et al., 2019; Huber et al., 2020; Whelan et al., 2024). These challenges are particularly pronounced in pediatric patients, where obtaining adequate blood volumes for culture is often more difficult than in adults.

Our analysis revealed significantly higher CRP concentrations in patients with septic shock than in those diagnosed with sepsis (P < 0.05). However, CRP has shown limited diagnostic reliability because of its gradual elevation during the early phases of infection and its frequent elevation in noninfectious conditions. In this study, the ROC curve analysis revealed that plasma CRP level exhibited the poorest diagnostic performance, with low specificity in the early diagnosis of sepsis in children. Current evidence suggests integrating CRP with other markers to optimize diagnostic accuracy rather than employing it as a standalone tool for sepsis screening (Morad et al., 2020).

Serum PCT is widely recognized as a biomarker of severe bacterial infections and the early diagnosis of sepsis, as well as for monitoring disease progression in infectious conditions. A meta-analysis by Menon et al. on severity-related variables in children demonstrated that PCT levels are positively correlated with mortality rates, which can serve as a predictive indicator for septic shock patients (Menon et al., 2022). In the present study, PCT levels were significantly higher in the septic shock group than in the sepsis group (P < 0.05). One study reported that a procalcitonin threshold of 19.1 ng/mL was associated with septic shock in children with sepsis (Rey et al., 2007). In this study, using a cutoff value of PCT ≥7.785 μg/L determined by the ROC curve, PCT demonstrated a sensitivity of 87.6% and specificity of 83.7% ( Table 3 ). In pediatric septic shock patients, procalcitonin demonstrated superior predictive capability relative to CRP, a finding consistent with established clinical research (Rey et al., 2007; Wacker et al., 2013; Morad et al., 2020).

Early recognition and differentiation of sepsis and septic shock are critical for initiating timely and appropriate management. However, diagnostic challenges persist owing to nonspecific clinical presentations and the limitations of existing laboratory tests. These limitations highlight the urgent need for novel biomarkers that can increase diagnostic accuracy, allow for early risk stratification, and guide personalized treatment strategies. Our study addresses this need by demonstrating that NRBCs improve the diagnostic performance of traditional biomarkers.

NRBCs represent immature erythroid precursors that enter circulation during severe physiological stressors such as hypoxia, systemic inflammation, or bone marrow activation (Pedersen et al., 2022). Their presence in circulation beyond the neonatal period is considered pathological and has been associated with poor outcomes in critically ill patients. In this study, NRBC counts were significantly higher in children with septic shock than in those with sepsis (P < 0.05). Our findings confirm the prognostic utility of NRBC counts in pediatric septic shock. These results confirm the prognostic value of NRBC quantification for mortality risk stratification in pediatric patients with septic shock, confirming previous the results of previous studies (Pedersen et al., 2022; Noor et al., 2023; Kirsch et al., 2024).

Previous work from our group identified NRBC positivity as an independent risk factor for disease severity and mortality in pediatric sepsis (Li et al., 2023). In the adult population, Desai et al. observed that surgical sepsis patients with detectable NRBCs exhibited significantly elevated mortality risks in both ICU and general ward settings. Their threshold analysis revealed mortality rates exceeding 50% at NRBC peaks >500/μL, with survival rates dropping below 12% for counts >2000/μL (Desai et al., 2012). Narcı et al.’s retrospective analysis also established NRBC counts as independent predictors of mortality across trauma, sepsis, and critically ill populations (Narcı et al., 2021). Collectively, these findings substantiate NRBC elevation as a biomarker of pathophysiological dysregulation proportional to disease severity.

In this study, when a cutoff value of NRBC count ≥ 3 was used, as determined by ROC curve analysis, the NRBC count demonstrated a sensitivity of 90.7% and a specificity of 88.4% ( Table 3 ). When combined with CRP and PCT, the NRBC count improved the risk prediction accuracy (AUC increased from 0.746 to 0.956), enabling more precise clinical assessment in pediatric sepsis management. This combined biomarker strategy facilitates earlier recognition of sepsis and severity stratification, thereby supporting time-sensitive clinical decisions regarding therapeutic escalation.

The pathophysiological mechanisms underlying elevated NRBC counts in pediatric sepsis and septic shock patients require further investigation. Under conditions of hypoxia, systemic inflammation, or tissue damage, signaling molecules such as erythropoietin and interleukin proteins (e.g., interleukin-3 and interleukin-6 [IL-6]) are activated (Stachon et al., 2005; Kirsch et al., 2024). Experimental evidence has identified IL-6 as a key regulator that promotes erythropoietin synthesis during systemic inflammatory response syndrome following septic shock in pediatric populations (Krafte-Jacobs and Bock, 1996). These biochemical mediators stimulate the maturation of immature red blood cells in the bone marrow, leading to the appearance of NRBCs in the circulation. Our findings suggest that this phenomenon is likely related to tissue hypoxia caused by systemic hypoperfusion.

In our study, children with septic shock required more mechanical ventilation and had higher lactate levels than those with sepsis alone did, supporting the role of oxygen deprivation in NRBC production. Additionally, elevated CRP and PCT levels in children with septic shock further indicate the involvement of inflammatory mediators in the synthesis and release of NRBCs. Further research is needed to elucidate the mechanisms underlying this association.

This study has several limitations. First, the study population was drawn from a single center, resulting in a relatively small sample size, which may limit the generalizability of our findings. In addition, the retrospective design of the study may have introduced bias in data recording and collection. Further research with larger sample sizes and multicenter designs is needed to validate these findings. At last, the future prospective studies are required to validate the mechanism of NRBC sources, due to the effects of anemia or hypoxia.

In conclusion, this study demonstrated that the NRBC count combined with the CRP and PCT levels have value for the diagnosis and prognosis of pediatric sepsis and septic shock. This biomarker combination improves the ability to reliably distinguish between sepsis and the progression of septic shock, representing a substantial advancement in sepsis care. By enabling earlier detection, more precise risk stratification, and more effective therapeutic interventions, the findings of the present study help address critical challenges in the management of pediatric sepsis. Given the global burden of pediatric sepsis and the limitations of existing diagnostic tools, incorporating NRBC measurements into clinical practice has the potential to transform sepsis management and improve outcomes for children worldwide. Future research should aim to validate these findings in larger, multicenter studies, explore their applicability across diverse populations, and investigate the broader implications of NRBC measurement for sepsis management and public health strategies.