Preoperative EUS elastography predicts pancreatic texture in pancreaticoduodenectomy

Postoperative pancreatic fistula is a serious complication of pancreaticoduodenectomy, while established risk tools may rely on intraoperative findings. This prospective, single-centre diagnostic accuracy study assessed preoperative EUS strain ratio in 100 consecutive patients undergoing elective surgery. A prespecified strain-ratio threshold of 2.5 was compared with pancreatic texture recorded independently by two surgeons blinded to the EUS result; clinically relevant fistula was also assessed as an exploratory outcome.

Soft pancreatic texture was found in 58 patients. At the threshold of strain ratio ≤2.5, sensitivity was 87.9% (95% CI 77.1–94.0), specificity 83.3% (69.4–91.7), and overall accuracy 86.0% (77.9–91.5); the area under the curve was 0.91 (0.84–0.96). Strain ratio increased with histological fibrosis grade (Spearman ρ=0.82; p<0.001). Clinically relevant fistula occurred in 16 patients. The exploratory preoperative-variable model had an optimism-corrected AUC of 0.78.

The findings support EUS strain elastography as a potential objective measure of pancreatic consistency before surgery, but do not establish that its use improves patient outcomes or fistula prediction. The single-centre cohort and only 16 fistula events limit the exploratory risk model; the authors call for external multicentre validation.