Editorial: Advancing pediatric critical care: sepsis, immune dysregulation, and precision therapies

Abstract

Pediatric critical care stands at a pivotal intersection of developmental immunology, organ system physiology, and rapidly evolving therapeutic innovation. Sepsis remains the leading cause of morbidity and mortality in pediatric intensive care units (PICUs) worldwide, accounting for over 20% of global childhood deaths [1]. Yet, children are not merely small adults-their dynamically maturing immune systems, agespecific inflammatory thresholds, and developing organ crosstalk create pathophysiological landscapes that demand pediatric-specific investigation rather than extrapolation from adult cohorts. Despite decades of advances in supportive care, targeted therapeutic trials in pediatric sepsis have repeatedly failed, largely due to the profound clinical and biological heterogeneity that defines this syndrome [2]. The transition from the 2005 systemic inflammatory response syndrome (SIRS)-based framework to the 2024 Phoenix Sepsis Criteria underscores an evolving recognition that sepsis is fundamentally a disorder of immune dysregulation driving life-threatening organ dysfunction [3]. In this editorial, we synthesize the key findings across six interconnected themes: diagnostic innovation and risk stratification, immune mechanisms and transcriptomic insights, organ dysfunction and complications, therapeutic interventions and blood purification, special populations, and health systems and nursing care. Figure 1 provides a conceptual framework illustrating the translational continuum addressed in this collection, from pathogen-specific triggers and immune dysregulation, through organ-specific complications, to precision diagnostic and therapeutic strategies. Early and accurate recognition of sepsis remains the cornerstone of improved outcomes, yet children frequently present with atypical or non-specific signs that delay intervention. Li et al. conducted a prospective cohort study directly comparing the 2005 and 2024 international consensus criteria for pediatric sepsis and septic shock in PICU patients [4]. The 2024 criteria, anchored on the Phoenix Sepsis Score (PSS), identified fewer cases of sepsis (80 vs. 240) and septic shock (49 vs. 86), with poor diagnostic agreement between the two frameworks (Kappa = 0.161). While the 2024 criteria better reflected disease severity and improved specificity, they were associated with potential underdiagnosis of early septic shock-particularly in 20 cases meeting vasoactive criteria only-suggesting that cardiovascular thresholds in the PSS may be overly strict. This study provides timely real-world evidence for clinicians navigating the transition to new diagnostic standards and highlights the need for AI-supported early warning models to complement clinical criteria.Complementing this diagnostic comparison, Jiang et al. developed and internally validated a LASSOlogistic regression nomogram for the early diagnosis of pediatric sepsis using seven routinely available clinical and laboratory indicators: respiratory rate, temperature, immature granulocyte percentage, platelets, procalcitonin, fibrinogen, and lactic acid [5]. In a cohort of 834 children with severe infections (25.4% developing sepsis), the model demonstrated good discriminatory ability, calibration, and clinical applicability, offering a practical bedside tool for primary care pediatricians to quantitatively assess sepsis risk.Moving beyond binary diagnosis to prognostic stratification, Wang et al. identified platelet count, Ddimer, and the C-reactive protein-albumin-lymphocyte (CALLY) index as independent risk factors for mortality in pediatric sepsis, constructing a nomogram with an AUC of 0.859 [6]. The CALLY index, integrating inflammation, nutritional/hepatic status, and immune competence, represents a novel composite biomarker readily obtainable from routine laboratories. Huang et al. further advanced prognostic modeling through machine learning-based time-to-event survival analysis in 223 children with severe sepsis [7]. Among five survival analysis algorithms evaluated, RandomSurvivalForest achieved the highest performance with a time-dependent AUC of 0.97. SHAP analysis identified total calcium and red cell distribution width (RDW) as the strongest mortality predictors, with clinically actionable thresholds established for calcium (<1.10 mmol/L), RDW (>15.07%), sodium (<131.68 mmol/L), and pH (<7.32). A web-based prediction calculator was developed to facilitate clinical implementation.Metabolic heterogeneity was explored by Wang et al., who applied group-based trajectory modeling (GBTM) to 72-hour blood glucose data from 1,178 pediatric sepsis patients, identifying four distinct trajectory phenotypes [8]: slow-recovery hypoglycemia (7.3%, mortality 13.79%), normoglycemia (59.9%, mortality 5.10%), persistent mild hyperglycemia (27.7%, mortality 8.26%), and persistent severe hyperglycemia (4.9%, mortality 17.24%). Persistent severe hyperglycemia independently predicted inhospital mortality (aOR = 3.13), while in the septic shock subgroup, both slow-recovery hypoglycemia and persistent severe hyperglycemia carried markedly elevated risks (5.2-fold and 8.28-fold, respectively). The authors proposed phenotype-guided interventions-early insulin for hyperglycemia and prophylactic glucose for hypoglycemia-demonstrating a precision medicine approach.Understanding the molecular underpinnings of immune dysregulation is essential for developing targeted therapies. The most mechanistically detailed contribution in this collection comes from Dunwoodie et al., who performed weighted gene co-expression network analysis (WGCNA) and differential gene expression on whole blood transcriptomes from 81 children with septic shock stratified by the PERSEVERE-II biomarker risk model [9]. High-mortality-risk patients displayed significant overexpression of genes related to neutrophil degranulation and innate immunity alongside suppressed adaptive immune responses. The predominance of developing neutrophils underscored a major role of emergency granulopoiesis, with key transcription factors including LTF, FOXM1, KLF1, and CEBPB linked to high-risk signatures. These findings reveal a pathological shift toward dysregulated neutrophildriven hyperinflammation with concurrent adaptive immune suppression, identifying transcriptional regulators as potential therapeutic targets to improve outcomes in high-risk patients. Notably, this study was published in Frontiers in Immunology as part of this Research Topic, reflecting the interdisciplinary nature of pediatric sepsis immunology.At the pathogen-specific level, Cheng et al. conducted a retrospective analysis of 5,956 children with viral pneumonia to determine the incidence and pathogen-specific risks of viral-associated septic shock [10]. The overall incidence was 1.06%, varying significantly by pathogen. Influenza A carried the highest incidence (3.70%) and was the strongest predictor of shock (OR = 7.506), followed by respiratory syncytial virus (RSV) (2.24%, OR = 4.559). Compared to bacterial shock, viral shock cases showed higher rates of respiratory failure/ARDS (88.9% vs. 52.2%) but shorter hospital stays. Influenza A cases were more likely to develop refractory shock and neurologic compromise. These findings challenge the prevailing emphasis on bacterial etiologies in pediatric sepsis guidelines and underscore the value of rapid viral PCR testing for antimicrobial stewardship and precise management.Sepsis-induced organ dysfunction drives mortality and long-term morbidity, in part due to specific organ complications and their epidemiological trajectories. Acute kidney injury (AKI) was examined by Li et al. using the United States National Inpatient Sample database, analyzing 225,902 pediatric severe sepsis hospitalizations from 2010 to 2019 [11]. The incidence of AKI increased from 2.7% to 8.0% over the decade, while in-hospital mortality declined from 32.40% to 17.90%-a paradox likely reflecting improved critical care survival alongside enhanced recognition. AKI affected 4.8% of patients overall and was associated with a sevenfold increased risk of mortality, driven by identifiable risk factors including specific comorbidities and infection sites. This large-scale epidemiological analysis underscores AKI as a potent marker of severity and the need for enhanced early identification of at-risk children.Pediatric acute respiratory distress syndrome (PARDS) was characterized by Asiri et al. in a retrospective cohort of 279 children with Phoenix-defined sepsis at a tertiary PICU in Saudi Arabia [12]. PARDS developed in 57.7% of sepsis patients, with mortality substantially higher in the PARDS cohort (36.6% vs. 7.6%). Mortality increased stepwise with worsening PARDS severity and was highest among children with both septic shock and severe PARDS. Pulmonary PARDS accounted for two-thirds of cases, while extrapulmonary PARDS demonstrated a higher inflammatory burden and more bacterial infections. In adjusted analyses, PARDS presence, higher PELOD-2 scores, and greater cumulative fluid balance were independently associated with mortality. These findings establish PARDS as a common and lethal complication of pediatric sepsis, demanding earlier recognition and phenotype-specific management.Cardiovascular dysfunction remains a central and underemphasized driver of morbidity and mortality in pediatric sepsis. Septic shock, defined by persistent hypotension or need for vasoactive support despite fluid resuscitation, reflects a complex interplay of myocardial depression, endothelial injury, microcirculatory dysfunction, and vasoplegia. The case reports in this collection illustrate the profound cardiovascular consequences of refractory septic shock: Bottari et al. reported two infants requiring extracorporeal membrane oxygenation (ECMO) for severe combined respiratory and circulatory failure [13], and Montero-Yeboles et al. described a child with escalating vasopressor requirements and severe metabolic derangement refractory to maximal conventional therapy [14]. Furthermore, thromboembolic complications-including deep vein thrombosis, as documented in the multifocal osteomyelitis case [15]highlight the prothrombotic state induced by sepsis-related endothelial activation and coagulopathy. These findings underscore the need for cardiovascular phenotyping in pediatric sepsis, including echocardiographic assessment of myocardial function, microcirculatory monitoring, and early identification of children at risk for refractory shock who may benefit from advanced circulatory support.A particularly illuminating case report by Yang et al. described a 12-year-old boy with septic shock who developed progressive limb necrosis that preceded disseminated multifocal osteomyelitis with pathological fracture, recurrent pneumothorax, and deep vein thrombosis, complicated by multidrugresistant infections including MRSA, non-typeable Haemophilus influenzae, and Candida albicans [15]. After over 90 days of intensive care involving ultrasound-guided drainage, pathogen-directed antibiotics, and multidisciplinary conservative management prioritizing medical treatment over high-risk surgery, the patient stabilized. This case illustrates that limb necrosis in septic shock is not merely a consequence of microcirculatory failure but may serve as a persistent source of deep infection and hematogenous dissemination, advocating for systematic screening of adjacent joints, bones, and deep soft tissues when limb necrosis develops.Beyond antibiotics and supportive care, the search for effective adjunctive therapies in pediatric sepsis remains urgent. Fu et al. conducted a randomized, single-blind, single-center trial of vitamin A supplementation (VAS) in 156 children with sepsis [16]. While VAS did not significantly reduce ICU length of stay or 28-day mortality, post hoc analyses revealed beneficial effects on systemic inflammation and lactate metabolism: VAS showed a significant downward trend in lactate levels, higher 24-hour lactate clearance (25%), and significantly lower lactate on day 3 post-intervention. Procalcitonin levels were significantly lower in the VAS group on days 3 and 7, with a greater reduction within 24 hours (73% vs. 48%). No adverse reactions were observed. Although the primary endpoints were not met, these findings suggest that VAS may modulate inflammatory and metabolic pathways in pediatric sepsis, warranting further investigation into the relationship between vitamin A, albumin, and outcomes.Extracorporeal blood purification has emerged as a potential adjunctive therapy for refractory septic shock, and two case reports in this collection explore novel devices in pediatric populations. Bottari et al. reported the first case series of hemadsorption using HA60 cartridges-devices with low priming volumes designed for low-weight patients-in four newborns and small infants (2.8-10 kg) with septic shock [13]. All patients received CRRT with HA60 added in series; two required ECMO for severe respiratory and circulatory failure. Three of four patients survived, with no major adverse events observed. The authors emphasized that these findings demonstrate feasibility and safety rather than efficacy, given the descriptive nature and multiple concomitant therapies. Montero-Yeboles et al. reported the compassionate use of the Seraph® 100 Microbind® affinity blood filter-a pathogen-binding hemoperfusion device-in a 7-yearold girl with refractory Streptococcus pyogenes septic shock and streptococcal toxic shock syndrome [14]. Despite maximal conventional therapy, the patient developed severe multiorgan dysfunction; hemoperfusion was initiated 42 hours after PICU admission as a stand-alone 4-hour session due to circuit instability. Following treatment, progressive reduction in vasoactive support and improvement in lactate clearance were observed within 24-48 hours, and the patient stabilized without requiring VA-ECMO, although early severe peripheral ischemia led to bilateral lower-limb amputations. Both reports highlight the potential role of extracorporeal blood purification as adjunctive rescue therapy while underscoring the need for prospective studies to determine safety, optimal timing, and clinical efficacy.Within the precision therapy framework, fluid management represents a critical yet incompletely optimized intervention. The traditional paradigm of aggressive initial fluid resuscitation has increasingly been challenged by evidence linking fluid overload to adverse outcomes, particularly in pediatric ARDS and acute kidney injury. Asiri et al. demonstrated that greater cumulative fluid balance was independently associated with mortality in sepsis-associated PARDS [12], while the AKI epidemiological analysis highlighted renal dysfunction as a key determinant of fluid tolerance [11]. A precision approach to fluid therapy should incorporate phenotype-guided strategies: children with hypovolemic shock and preserved cardiac function may benefit from targeted initial resuscitation guided by dynamic markers of fluid responsiveness, whereas those with myocardial depression, PARDS, or established AKI require a more conservative approach with early initiation of vasoactive support and judicious fluid administration. The metabolic phenotypes identified by Wang et al.-particularly persistent severe hyperglycemia and slowrecovery hypoglycemia-may further inform fluid composition and dextrose content [8]. Future