Early National Experience with Laparoscopic Pancreaticoduodenectomy for Ductal Adenocarcinoma: A Comparison of Laparoscopic Pancreaticoduodenectomy and Open Pancreaticoduodenectomy from the National Cancer Data Base - 19/06/15

Doi : 10.1016/j.jamcollsurg.2015.04.021

Susan M. Sharpe, MD ^a, Mark S. Talamonti, MD, FACS ^c, Chihsiung E. Wang, PhD ^c, Richard A. Prinz, MD, FACS ^c, Kevin K. Roggin, MD, FACS ^a, David J. Bentrem, MD, FACS ^b, David J. Winchester, MD, FACS ^c, Robert D.W. Marsh, MD ^c, Susan J. Stocker, CCRP ^c, Marshall S. Baker, MD, MBA, FACS ^c,^⁎
^a Department of Surgery, University of Chicago Pritzker School of Medicine, Chicago, IL
^b Department of Surgery, Northwestern University Feinberg School of Medicine, Chicago, IL
^c Department of Surgery, NorthShore University Health System, Evanston, IL

^∗Correspondence address: Marshall S Baker, MD, MBA, FACS, Department of Surgery, NorthShore University HealthSystem, Evanston Hospital, 2650 Ridge Ave, Walgreen Building, Suite 2507, Evanston, IL 60201.

Abstract

Background

There is considerable debate about the safety and clinical equivalence of laparoscopic pancreaticoduodenectomy (LPD) and open pancreaticoduodenectomy (OPD) for pancreatic ductal adenocarcinoma (PDCA).

Study Design

We queried the National Cancer Data Base to identify patients undergoing LPD and OPD for PDCA between 2010 and 2011. Chi-square and Student's t-tests were used to evaluate differences between the 2 approaches. Multivariable logistic regression modeling was performed to identify patient, tumor, or facility factors associated with perioperative mortality.

Results

Four thousand and thirty-seven (91%) patients underwent OPD. Three hundred and eighty-four (9%) patients underwent LPD. There were no statistical differences between the 2 surgical cohorts with regard to age, race, Charlson score, tumor size, grade, stage, or treatment with neoadjuvant chemoradiotherapy. Laparoscopic pancreaticoduodenectomy demonstrated a shorter length of stay (10 ± 8 days vs 12 ± 9.7 days; p < 0.0001) and lower rates of unplanned readmission (5% vs 9%; p = 0.027) than OPD. In an unadjusted comparison, there was no difference in 30-day mortality between the LPD and OPD cohorts (5.2% vs 3.7%; p = 0.163). Multivariable logistic regression modeling predicting perioperative mortality controlling for age, Charlson score, tumor size, nodal positivity, stage, facility type, and pancreaticoduodenectomy volume identified age (odds ratio [OR] = 1.05; p < 0.0001), positive margins (OR = 1.45; p = 0.030), and LPD (OR = 1.89; p = 0.009) as associated with an increased probability of 30-day mortality; higher hospital volume was associated with a lower risk of 30-day mortality (OR = 0.98; p < 0.0001). In institutions that performed ≥10 LPDs, the 30-day mortality rate of the laparoscopic approach was equal to that for the open approach (0.0% vs 0.7%; p = 1.00).

Conclusions

Laparoscopic pancreaticoduodenectomy is equivalent to OPD in length of stay, margin-positive resection, lymph node count, and readmission rate. There is a higher 30-day mortality rate with LPD, but this appears driven by a surmountable learning curve for the procedure.

Le texte complet de cet article est disponible en PDF.

Abbreviations and Acronyms : LOS, LPD, NCDB, OPD, PD, PDCA