The research team, led by Fang Chen and colleagues in the hospital’s Department of Laboratory Medicine, enrolled 126 consecutive children diagnosed with immunoglobulin A vasculitis, or IgAV, between September 2021 and April 2025. Of these, 44 developed gastrointestinal bleeding, and 20 of those bled overtly, meaning the hemorrhage was clinically visible as melena, bloody stools, vomited blood, or, in one case, an endoscopically confirmed perforation of the intestinal wall. The cohort was predominantly boys, about 67.5 percent, with a median age of 8 years, matching the disease’s typical peak between ages 4 and 7.

IgAV, formerly known as Henoch-Schönlein purpura, is driven by immune complexes dominated by immunoglobulin A1 that deposit in small blood vessels throughout the body, triggering inflammation in the skin, joints, kidneys, and gastrointestinal tract. The global annual incidence ranges from 3 to 27 cases per 100,000 people. Gastrointestinal symptoms, which appear in 28 to 60.9 percent of pediatric cases, can escalate from vomiting and abdominal pain to intussusception, ischemia, and life-threatening hemorrhage. What has been missing, the authors argue, is a reliable panel of blood markers that reflects the underlying activation of the coagulation and fibrinolysis systems, the paired machinery that forms and dissolves clots.

To fill that gap, the team measured five hemostatic molecular markers: soluble thrombomodulin, or sTM, which reflects endothelial cell injury; the thrombin-antithrombin complex, or TAT, a stable surrogate for thrombin generation; the plasmin-α2-plasmin inhibitor complex, or PIC, which forms when the clot-dissolving enzyme plasmin binds its hepatic inhibitor; the tissue plasminogen activator-inhibitor complex, or t-PAIC, a marker of sustained endothelial damage and fibrinolytic progression; and D-dimer, the classic fibrin degradation product. They also tracked complete blood counts, fibrinogen, complement C3 and C4, and serum immunoglobulins A, G, and M.

The results were striking. Children with gastrointestinal bleeding had significantly higher white blood cell counts, absolute neutrophil counts, and neutrophil-to-lymphocyte ratios than those who did not bleed. Their coagulation-fibrinolysis markers told a similar story: fibrinogen, TAT, PIC, and D-dimer were all elevated, with D-dimer showing the largest effect size at r = 0.43. Meanwhile, serum IgG was significantly lower in the bleeding group. Soluble thrombomodulin and t-PAIC, however, showed no meaningful difference between groups, and complement levels were comparable.

Multivariate logistic regression, adjusted for multicollinearity with all variance inflation factors below 2.5, revealed two independent associations. Each 1 μg/mL increase in PIC was linked to 1.89-fold higher odds of gastrointestinal bleeding (OR = 1.888, 95% CI 1.013–3.518; P = 0.045). Each 1 g/L increase in serum IgG was associated with 15.7 percent lower odds of bleeding (OR = 0.843, 95% CI 0.737–0.965; P = 0.013). D-dimer remained directionally positive but fell just short of significance in the adjusted model (OR = 1.191, 95% CI 0.998–1.422; P = 0.053), as did absolute neutrophil count and TAT.

Receiver operating characteristic analysis gave the findings clinical texture. PIC alone showed modest discriminatory ability, with an area under the curve of 0.687 (95% CI 0.590–0.785; P = 0.001). At an exploratory threshold of 1.25 μg/mL, sensitivity was 69.8 percent and specificity 66.7 percent. Serum IgG below 11.86 g/L performed slightly weaker, with an AUC of 0.633, sensitivity of 76.7 percent, and specificity of 50.6 percent. Combining PIC and IgG into a single model raised the AUC to 0.737 (95% CI 0.639–0.835; P < 0.001), with a predicted-probability threshold of 0.41 yielding 55.8 percent sensitivity and 84.0 percent specificity.

Comparing occult with overt bleeding added a second layer of insight. Among the 44 children with confirmed bleeding, 24 had occult hemorrhage detected only by fecal occult blood testing, while 20 bled overtly. Children with overt bleeding stayed in the hospital significantly longer (median 14.5 days versus 9.0 days; P < 0.001), received higher doses of methylprednisolone (median 40 mg/day versus 30 mg/day; P = 0.012), and had elevated serum IgM (P = 0.025), suggesting greater clinical burden and humoral immune activation in visible hemorrhage. The abdominal phenotype was more frequent in the overt group, 70.0 percent versus 41.7 percent, though this difference did not reach statistical significance (P = 0.060).

Mechanistically, the pattern is telling. The absence of significant differences in sTM and t-PAIC, both markers of endothelial injury, suggests the coagulofibrinolytic activation in these children is driven not by damaged blood vessel linings but by neutrophils. This aligns with histological findings from Louie and colleagues, who documented lamina propria hemorrhage, fibrin deposition, and leukocytoclastic vasculitis in gastrointestinal biopsies of IgAV patients without obvious endothelial necrosis. It also fits with emerging work showing that IgA1 immune complexes bind polymorphonuclear neutrophils via Fcα receptor I, prompting the release of neutrophil extracellular traps, or NETs, which have been detected at elevated levels in the serum, renal, gastric, and duodenal tissues of pediatric IgAV patients during the acute phase. Recent research by Qiu and colleagues, published in Nature, further demonstrated that neutrophils contribute to early clot resolution by releasing elastase within inflammatory microthrombi, a process that could explain the hyperfibrinolytic state captured by elevated PIC.

The decreased serum IgG in bleeding children, consistent with prior reports from Yang and colleagues and Zhang and colleagues, points to disturbed humoral immunity as a co-contributor. The authors note that serum IgG below 11.86 g/L and PIC above 1.85 μg/mL were identified as factors associated with gastrointestinal bleeding, echoing Yang’s finding that reduced IgG served as an independent predictor in their own cohort.

Sensitivity analyses explored how incomplete outcome ascertainment might have shaped the estimates. When all children with undetermined bleeding status were classified as non-bleeders, higher absolute neutrophil count and lower serum IgG emerged as significant; when they were classified as bleeders, higher PIC and D-dimer were significant. The variables retained in the multivariable models shifted across scenarios, leading the authors to frame their findings as exploratory rather than definitive.

Several limitations temper the conclusions. The study was retrospective and single-center, with a relatively small sample size for subgroup analyses, particularly the single case of gastrointestinal perforation. The IgG association became borderline non-significant under one sensitivity scenario. All biomarker thresholds were exploratory, selected by maximizing the Youden index, and the authors explicitly call for large-scale, prospective, multicenter studies to validate the findings externally.

Still, the clinical implications are immediate. PIC and serum IgG are already measurable in routine laboratory panels, and a combined model incorporating both could help pediatricians stratify risk at the bedside, identifying which children with IgAV need closer monitoring for gastrointestinal complications before bleeding becomes overt. The study also reinforces a broader shift in how scientists understand IgAV: not simply as a vasculitic disorder of blood vessel walls, but as a thromboinflammatory condition in which neutrophils, immune complexes, and the clotting-fibrinolysis axis conspire to produce disease. For the roughly 27.2 percent of IgAV children in this cohort who developed gastrointestinal bleeding, that reframing could translate into earlier detection, tighter surveillance, and, ultimately, fewer emergencies.

Subject of Research: Coagulation and fibrinolysis biomarkers associated with gastrointestinal bleeding in children with immunoglobulin A vasculitis: a retrospective cohort study

Article Title: Coagulation and fibrinolysis biomarkers associated with gastrointestinal bleeding in children with immunoglobulin A vasculitis: a retrospective cohort study

Article References: Chen, F., Shen, H., Shao, X., & Wei, J. (2026). Coagulation and fibrinolysis biomarkers associated with gastrointestinal bleeding in children with immunoglobulin A vasculitis: a retrospective cohort study. Frontiers in Pediatrics, 14, Article 1886693. https://doi.org/10.3389/fped.2026.1886693

Image Credits: AI Generated

DOI: 10.3389/fped.2026.1886693

Keywords: immunoglobulin A vasculitis, gastrointestinal bleeding, children, biomarkers, coagulation, fibrinolysis, PIC, serum IgG, D-dimer, neutrophils, retrospective cohort study, pediatric rheumatology