TY - JOUR
T1 - Outcomes in patients with portal hypertension undergoing gastrointestinal surgery
T2 - A propensity score matched analysis from the NSQIP dataset
AU - De Stefano, Felice
AU - Garcia, Catherine R.
AU - Gupta, Meera
AU - Marti, Francesc
AU - Turcios, Lilia
AU - Dugan, Adam
AU - Gedaly, Roberto
N1 - Publisher Copyright:
© 2018 Elsevier Inc.
PY - 2019/4
Y1 - 2019/4
N2 - Background/Aim: We aim to study the impact of PH in patients undergoing gastrointestinal surgery (GI). Methods: We queried the ACS-NSQIP database from 2005 through 2010 for patients undergoing GI surgery with PH. Esophageal varices (EV) diagnosis was used as a surrogate of PH. Results: A total of 192,296 patients underwent GI surgery, of which 379 had PH. Regression analyses revealed that patients with PH had a 6-fold (95% CI 4.6–7.9) increase in 30-day mortality, a 3-fold (95% CI 2.5–3.7) increase in morbidity, a 3.2-fold (95% CI 2.6–3.9) increase in critical care complications (CCC), and a 6.5-day (95% CI 5.1–7.8) increase in hospital LOS. After PSM, the impact of PH on the outcomes remained. These differences were significant regardless of the emergent or elective status of the procedure. AUC analysis demonstrated that MELD and MELDNa + score greater than 10.5 was the most predictive of peri-operative mortality in elective PH cases. Conclusions: PH is associated with an increased risk of poor surgical outcomes in patients undergoing elective and emergent gastrointestinal surgery.
AB - Background/Aim: We aim to study the impact of PH in patients undergoing gastrointestinal surgery (GI). Methods: We queried the ACS-NSQIP database from 2005 through 2010 for patients undergoing GI surgery with PH. Esophageal varices (EV) diagnosis was used as a surrogate of PH. Results: A total of 192,296 patients underwent GI surgery, of which 379 had PH. Regression analyses revealed that patients with PH had a 6-fold (95% CI 4.6–7.9) increase in 30-day mortality, a 3-fold (95% CI 2.5–3.7) increase in morbidity, a 3.2-fold (95% CI 2.6–3.9) increase in critical care complications (CCC), and a 6.5-day (95% CI 5.1–7.8) increase in hospital LOS. After PSM, the impact of PH on the outcomes remained. These differences were significant regardless of the emergent or elective status of the procedure. AUC analysis demonstrated that MELD and MELDNa + score greater than 10.5 was the most predictive of peri-operative mortality in elective PH cases. Conclusions: PH is associated with an increased risk of poor surgical outcomes in patients undergoing elective and emergent gastrointestinal surgery.
UR - http://www.scopus.com/inward/record.url?scp=85058576737&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85058576737&partnerID=8YFLogxK
U2 - 10.1016/j.amjsurg.2018.12.008
DO - 10.1016/j.amjsurg.2018.12.008
M3 - Article
C2 - 30578032
AN - SCOPUS:85058576737
SN - 0002-9610
VL - 217
SP - 664
EP - 669
JO - American Journal of Surgery
JF - American Journal of Surgery
IS - 4
ER -