EUS-guided and US-guided percutaneous splenic biopsy: a systematic review and meta-analysis of separately pooled diagnostic and safety outcomes

Abstract

Ultrasound (US)-guided percutaneous splenic biopsy has been widely used to evaluate focal or diffuse splenic lesions and has shown promising efficacy and safety. Endoscopic ultrasound (EUS)-guided biopsy has emerged as an alternative approach for tissue acquisition from splenic lesions because it provides high-resolution imaging and real-time needle guidance. However, because no eligible head-to-head comparative studies were identified, this systematic review and meta-analysis summarizes separately pooled diagnostic and safety outcomes for EUS-guided splenic biopsy and US-guided percutaneous splenic biopsy rather than providing a direct comparative meta-analysis. PubMed, Scopus, the Cochrane Library, and Web of Science were searched from inception until January 2026. Studies were included if they assessed diagnostic performance and/or complication rates of EUS-guided splenic biopsy or US-guided percutaneous splenic biopsy. Heterogeneity across pooled studies was evaluated using the I² statistic, and methodological quality was evaluated using the Newcastle-Ottawa Scale (NOS). Diagnostic accuracy was defined as the proportion of correctly diagnosed cases (true positives plus true negatives divided by total assessed cases), and studies applying substantially different definitions were excluded from pooled diagnostic accuracy analyses or flagged during extraction. The protocol was registered in PROSPERO (CRD420261283689). Twenty studies involving 1,147 patients were included. Separately pooled estimates showed high diagnostic performance for both modalities. EUS-guided splenic biopsy demonstrated diagnostic accuracy of 95% (95% CI, 90% to 100%), sensitivity of 97% (95% CI, 82% to 100%), and specificity of 95% (95% CI, 75% to 100%). US-guided percutaneous splenic biopsy demonstrated diagnostic accuracy of 89% (95% CI, 87% to 92%), sensitivity of 91% (95% CI, 86% to 94%), and specificity of 98% (95% CI, 96% to 99%). Overall complication rates were 7% for EUS-guided splenic biopsy and 3% for US-guided percutaneous splenic biopsy; major complication rates were 4% and 1%, respectively. These numerically different rates should be interpreted cautiously because the comparison is indirect, EUS-guided cohorts were smaller, and selection bias may have favored inclusion of technically challenging cases in the EUS-guided studies. No study reported procedure-related mortality. A formal unadjusted indirect comparison found no evidence of a modality difference in sensitivity, specificity, or diagnostic accuracy. Higher complication odds were observed in the primary study-level comparison, but the overall-complication result was influenced by one study and the major-complication result was fragile; neither should be interpreted as a causal modality effect. EUS-guided splenic biopsy and US-guided percutaneous splenic biopsy both appear diagnostically effective and generally safe in separately pooled analyses. Because no direct comparative studies were available, the findings should not be interpreted as evidence of superiority of either technique. US-guided percutaneous splenic biopsy may be considered for accessible lesions, whereas EUS-guided splenic biopsy may be considered in selected patients when percutaneous biopsy is contraindicated, technically difficult, or previously inconclusive. The formal study-level comparison remains unadjusted and indirect and does not establish superiority or inferiority.