Efficacy of early blood purification on serum inflammatory mediators in severe acute pancreatitis: a systematic review and meta-analysis

Abstract

Severe acute pancreatitis (SAP) is the most critical form of acute pancreatitis (AP), characterized by a dysregulated systemic inflammatory response that drives early organ failure and mortality. Current management remains mainly supportive. Early blood purification (EBP) has been proposed as an adjunctive strategy to reduce circulating inflammatory mediators, yet prior syntheses have been limited by small trial sizes, heterogeneous modalities, and inconsistent reporting. This study aims to evaluate the efficacy and safety of early EBP in adults with SAP, with outcomes stratified by treatment modality and timing of initiation. A comprehensive search was conducted across six databases (PubMed, Scopus, Web of Science, Ovid, Embase, and the Cochrane Library) for studies published until June, 2026. Eligible studies enrolled adults with SAP and compared early EBP (continuous veno-venous hemofiltration [CVVH], high-volume hemofiltration [HVHF], combined continuous blood purification, hemoperfusion, plasma exchange, or hemadsorption) against standard medical management. Pooled effects were expressed as standardized mean differences (SMDs), mean differences (MDs), or risk ratios (RRs) using random-effects models, stratified a priori by treatment modality and, where feasible, by timing of initiation. Certainty of evidence was assessed using the GRADE framework. Twenty-one studies comprising 1,959 adults were included. Early EBP was associated with a significant overall reduction in IL-6 (SMD = − 1.55; 95% CI: −2.05 to − 1.05; 12 studies, n = 1,045) and CRP (SMD = − 1.14; 95% CI: −1.93 to − 0.34; 11 studies, n = 840), with effect sizes differing significantly across modalities for both markers. APACHE II score also improved significantly (MD = − 3.40; 95% CI: −4.64 to − 2.15; 12 studies, n = 938), as did serum amylase, creatinine, and ALT. All-cause mortality was significantly reduced (RR = 0.55; 95% CI: 0.41 to 0.74; 13 studies, n = 837), with no significant heterogeneity and consistent effects across modality subgroups. TNF-α reduction was significant overall (SMD = − 1.38; 95% CI: −2.37 to − 0.38) but did not differ significantly by modality. SOFA score, and hospital and ICU length of stay, showed no consistent significant improvement. Procedure-related adverse events were inconsistently reported across studies. Certainty of evidence was rated very low (GRADE) for all pooled outcomes, reflecting risk of bias, inconsistency, and imprecision. Early EBP in SAP is associated with significant reductions in mortality and key inflammatory markers, alongside improvements in disease severity scores and biochemical parameters, though the magnitude of benefit varies by treatment modality. The geographically concentrated evidence base, mechanistic heterogeneity across modalities, inconsistent adverse-event reporting, and very low certainty of evidence preclude routine recommendation. Adequately powered multicenter RCTs with standardized protocols, modality comparisons, and safety reporting are needed before EBP can be incorporated into international SAP guidelines.