TY - JOUR
T1 - A deep pancreas is a novel predictor of pancreatic fistula after pancreaticoduodenectomy in patients with a nondilated main pancreatic duct
AU - Iseki, Masahiro
AU - Noda, Hiroshi
AU - Watanabe, Fumiaki
AU - Kato, Takaharu
AU - Endo, Yuhei
AU - Aizawa, Hidetoshi
AU - Fukui, Taro
AU - Ichida, Kosuke
AU - Kakizawa, Nao
AU - Rikiyama, Toshiki
N1 - Publisher Copyright:
© 2020 Elsevier Inc.
PY - 2021/6
Y1 - 2021/6
N2 - Background: We investigated the risk factors for clinically relevant postoperative pancreatic fistula after pancreaticoduodenectomy in patients with a nondilated main pancreatic duct. Methods: We investigated a total of 354 patients who underwent pancreaticoduodenectomy. The diameter of the main pancreatic duct, the shortest distance from the body surface to the pancreas (the pancreatic depth), and the computed tomography attenuation index (the difference between the pancreatic and splenic computed tomography attenuation) were measured in preoperative computed tomography. Results: One hundred eighty-one (51.1%) patients had a nondilated main pancreatic duct, and 50 (27.6%) of the 181 patients with a nondilated main pancreatic duct developed a clinically relevant postoperative pancreatic fistula. Univariate analyses revealed that the calculated body mass index (≥21.8 kg/m2) (P =.004), deep pancreas (pancreatic depth ≥51.2 mm) (P =.001), and low computed tomography attenuation index (≤–3.8 Hounsfield units) (P =.02) were significant risk factors for clinically relevant postoperative pancreatic fistula. The multivariate logistic regression analysis revealed that deep pancreas (odds ratio 2.370; 95% confidence interval 1.0019–5.590; P =.049) was an independent risk factor for clinically relevant postoperative pancreatic fistula. Among patients with a nondilated main pancreatic duct, deep pancreas (in comparison to patients without deep pancreas) was associated with male sex (72.7% vs 54.9%; P =.016), higher body mass index (22.5 kg/m2 vs 19.6 kg/m2; P <.001), a history of diabetes mellitus (24.5% vs 8.5%; P =.006), a lower computed tomography attenuation index (–9.6 Hounsfield units vs –4.6 Hounsfield units; P =.007), a longer operative time (454 minutes vs 420 minutes; P =.007), and a higher volume of intraoperative blood loss (723 mL vs 500 mL; P <.001), respectively. Conclusion: Deep pancreas may be an important parameter associated with significant risk factors for clinically relevant postoperative pancreatic fistula after pancreaticoduodenectomy in patients with a nondilated main pancreatic duct.
AB - Background: We investigated the risk factors for clinically relevant postoperative pancreatic fistula after pancreaticoduodenectomy in patients with a nondilated main pancreatic duct. Methods: We investigated a total of 354 patients who underwent pancreaticoduodenectomy. The diameter of the main pancreatic duct, the shortest distance from the body surface to the pancreas (the pancreatic depth), and the computed tomography attenuation index (the difference between the pancreatic and splenic computed tomography attenuation) were measured in preoperative computed tomography. Results: One hundred eighty-one (51.1%) patients had a nondilated main pancreatic duct, and 50 (27.6%) of the 181 patients with a nondilated main pancreatic duct developed a clinically relevant postoperative pancreatic fistula. Univariate analyses revealed that the calculated body mass index (≥21.8 kg/m2) (P =.004), deep pancreas (pancreatic depth ≥51.2 mm) (P =.001), and low computed tomography attenuation index (≤–3.8 Hounsfield units) (P =.02) were significant risk factors for clinically relevant postoperative pancreatic fistula. The multivariate logistic regression analysis revealed that deep pancreas (odds ratio 2.370; 95% confidence interval 1.0019–5.590; P =.049) was an independent risk factor for clinically relevant postoperative pancreatic fistula. Among patients with a nondilated main pancreatic duct, deep pancreas (in comparison to patients without deep pancreas) was associated with male sex (72.7% vs 54.9%; P =.016), higher body mass index (22.5 kg/m2 vs 19.6 kg/m2; P <.001), a history of diabetes mellitus (24.5% vs 8.5%; P =.006), a lower computed tomography attenuation index (–9.6 Hounsfield units vs –4.6 Hounsfield units; P =.007), a longer operative time (454 minutes vs 420 minutes; P =.007), and a higher volume of intraoperative blood loss (723 mL vs 500 mL; P <.001), respectively. Conclusion: Deep pancreas may be an important parameter associated with significant risk factors for clinically relevant postoperative pancreatic fistula after pancreaticoduodenectomy in patients with a nondilated main pancreatic duct.
UR - http://www.scopus.com/inward/record.url?scp=85098617455&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85098617455&partnerID=8YFLogxK
U2 - 10.1016/j.surg.2020.11.033
DO - 10.1016/j.surg.2020.11.033
M3 - Article
C2 - 33390302
AN - SCOPUS:85098617455
SN - 0039-6060
VL - 169
SP - 1471
EP - 1479
JO - Surgery (United States)
JF - Surgery (United States)
IS - 6
ER -