News|Articles|May 13, 2026

Predicting Sepsis-Related Mortality in Pediatric ALL Requires Greater Precision

Listen

0:00 / 0:00

Key Takeaways

Bloodstream, pulmonary, and multisite infections predominated, with bacteria leading among identified pathogens and Gram-negative organisms most common, underscoring broad infectious exposure and rapid clinical deterioration.
Poor leukemia response strongly associated with death, as partial or nonremission status suggested persistent immunosuppression from residual disease and heightened vulnerability to sepsis-related organ failure.
Higher inflammatory biomarkers and organ injury indices, plus reduced 24-hour lactate clearance, tracked with mortality, highlighting persistent hypoperfusion and early multiorgan dysfunction as key prognostic signals.
PSOFA outperformed PEWS and PCIS for mortality prediction (AUC 0.885 vs 0.788 and 0.751), aligning with sepsis pathobiology via integrated respiratory, coagulation, hepatic, cardiovascular, CNS, and renal assessment.
Mechanical ventilation and inotrope use did not independently predict mortality, likely functioning as severity markers rather than causal contributors, reinforcing the value of earlier physiologic scoring and biomarker integration.

Pediatric Sequential Organ Failure Assessment may better predict death in ALL and sepsis, supporting earlier ICU intervention.

Close to 30% of children who underwent chemotherapy for acute lymphoblastic leukemia (ALL) and developed sepsis—a long-feared complication of the treatment in pediatric patients—died, according to investigators who evaluated the mortality predictive ability of several rival scoring tools. Findings from their new retrospective study suggest that the right clinical scoring tool, deployed early, could give clinicians a meaningful edge in identifying which patients are most at risk of dying following admittance to the pediatric intensive care unit (PICU).¹

“Improving the survival rate of children with ALL complicated by sepsis who are transferred to the PICU is the key to improving the overall prognosis of children with ALL,” the authors wrote in the American Journal of Cancer Research.

Analyzing data on 260 patients with ALL transferred to the PICU at West China Second University Hospital from October 2020 through October 2025—184 who survived (70.77%; mean [SD] age, 4.38 [2.81] years) and 76 who died (29.23%; mean age, 4.41 [2.77] years)—the researchers highlight that although their mortality rate was lower compared with previous investigations,^2,3 their data still underscore the gravity of sepsis as a treatment-related complication in this population.

A Wide Net of Infection, A Narrow Window to Act

Among the entire study population, the most common infections were bloodstream infections (29.23%), pulmonary infections (27.69%), and multisite infections (23.46%), a pattern the researchers describe as reflecting a wide infection range with rapid disease progression. Bacterial infections predominated among pathogen-positive cases, with Gram-negative bacteria accounting for 29.62% and Gram-positive bacteria for 19.23%.

Critically, nearly 90% of the cohort was classified as having intermediate-risk ALL (41.54%) or high-risk ALL (45.77%), suggesting that children with more aggressive disease subtypes face a compounded threat when infection strikes during treatment.

Unlike presentation in otherwise healthy patients, children with ALL may not mount a high fever or other classic warning signs. Instead, their symptoms can be masked or mimicked by the adverse effects of chemotherapy, leading to dangerous diagnostic delays. Sepsis is extremely difficult to detect early in immunocompromised children, the authors explained.

What Predicted Death and What Didn’t

Within 48 hours of PICU admission, which the authors classified as less than 14 days or 14 days and longer, several indicators emerged as significantly associated with mortality. Leukemia remission status proved especially telling. The children who died were far more likely to have achieved only partial or no remission compared with survivors, at 60.53% vs 32.07%, a finding tied to residual leukemic cells that continue suppressing normal immune function.

Inflammatory biomarker mean (SD) levels, including C-reactive protein (CRP), procalcitonin (PCT), and interleukin-6 (IL-6), were all significantly elevated in nonsurvivors:

CRP: 98.93 (8.57) mg/L vs 80.39 (7.16) mg/L
PCT: 8.39 (0.78) ng/mL vs 6.12 (0.73) ng/mL
IL-6: 232.87 (20.29) pg/mL vs 174.79 (13.46) pg/mL

Markers of liver, kidney, and cardiac dysfunction were also elevated in nonsurvivors (all P < .001). However, the 24-hour lactate clearance rate was significantly lower in children who died (23.78% [3.41%] vs 37.56% [4.62%]; P < .05), pointing to persistent tissue hypoperfusion as a key driver of deterioration. Somewhat counterintuitively, the use of invasive mechanical ventilation and inotropic drug support did not independently predict mortality (P < .001). The authors note this is likely because these interventions reflect disease severity rather than causing harm.

PSOFA Rises to the Top

At the heart of the study is a head-to-head comparison of 3 clinical scoring systems used to assess condition severity in critically ill children: Pediatric Critical Illness Score (PCIS), Pediatric Early Warning Score (PEWS), and Pediatric Sequential Organ Failure Assessment (PSOFA) score.

Using receiver operating characteristic curve analysis, the researchers found that PSOFA significantly outperformed its rivals. The area under the curve for PSOFA was 0.885 (95% CI, 0.841-0.929) compared with 0.788 (95% CI, 0.726-0.850) for PEWS and 0.751 for PCIS (95% CI, 0.689-0.812), differences that were statistically significant.

“The PSOFA score demonstrated superior predictive performance for mortality compared with the PEWS and PCIS scores,” the authors concluded. “The combination of multidimensional clinical indicators and scoring systems facilitates early risk stratification and helps guide PICU treatment decisions.” The PSOFA assessment covers the respiratory, coagulation, liver, cardiovascular, central nervous, and renal systems, which correlates well with the multi-organ nature of sepsis-driven deterioration, they explained.

Implications for Practice

Given the complexity and rapid progression of sepsis in this population, relying on a single-dimension tool may miss the full picture of physiological compromise, driving the authors to advocate for a multidimensional approach to build a more complete early-warning profile. Such as pairing PSOFA with biomarkers like lactate clearance, inflammatory indicators, and organ function panels.

Their findings also reinforce the urgency of PICU transfer timing. Closer coordination between hematology-oncology and critical care teams in monitoring high-risk, poorly responding patients may allow for earlier escalation of care.

The authors acknowledge the study’s limitations, including its single-center retrospective design, relatively modest sample size, and the absence of dynamic longitudinal biomarker monitoring. Multicenter prospective validation is warranted before PSOFA-based protocols are widely adopted.

References

Yin Z, Shen C, Chen Y, Qing L, Li D. Pediatric sequential organ failure assessment score predicts prognosis in children with acute lymphoblastic leukemia and sepsis: association with early multiple organ dysfunction. Am J Cancer Res. 2026;16(3):1157-1170. doi:10.62347/CHAG7660
Singer K, Subbaiah P, Hutchinson R, Odetola F, Shanley TP. Clinical course of sepsis in children with acute leukemia admitted to the pediatric intensive care unit. Pediatr Crit Care Med. 2011;12(6): 649-654. doi:10.1097/PCC.0b013e31821927f1
Wu L, Jin M, Wang R, et al. Prognostic factors of sepsis in children with acute leukemia admitted to the pediatric intensive care unit. Pediatr Blood Cancer. 2023:e30382. doi:10.1002/pbc.30382