Surgical Versus Percutaneous Drainage for Pancreatic Pseudocysts: Patient, Disease and Procedure Characteristics

Justin Lee, MD, Romie Mundy, MD, Neal E Seymour, MD. St. Elizabeth Medical Center, Baystate Medical Center, Tufts University School of Medicine

Introduction
The objectives of this study were to use population-based data to 1) compare recent use of percutaneous and surgical drainage procedures, 2) quantify utilization of laparoscopy for pseudocyst, and 3) define patient and disease characteristics pertaining to use of specific drainage methods.

Methods
Hospitalizations for pancreatic pseudocyst-related drainage procedures were identified in the National Inpatient Sample (NIS) database from 2001 to 2008. Patient demographics, associated diagnoses, and procedures during these hospitalizations were identified. Percutaneous and surgical drainage procedures were quantified and compared for mortality, length of stay, and total hospital charges. Frequency of use of laparoscopy relative to other treatment methods was defined for the study period. Categorical and continuous variables were analyzed using the chi square and Student’s T-test respectively.

Results
The NIS search identified 23,832 hospitalizations for pancreatic pseudocyst drainage procedures. Patients characteristics included: Age 50.7±16.4 years, 59.4% male, 43.8% private insurance, 26.4% Medicare, and 72.5% urban metropolitan area residence. 13,511 (56.7%) underwent percutaneous drainage and 10,321 (43.3%) underwent surgical drainage. Utilization of laparoscopy with surgical drainage increased near four-fold from 1.0% to 3.9% (P<0.001) during the period of study. Surgical drainage was associated with decreased inpatient mortality (1.3% versus 3.4%, P<0.001), shorter mean length of stay (15.25 days vs. 20.34 days, P<0.001) and lower mean total hospital charges ($75,265.78 vs. $104,936.73, P<0.001). Overall, 65.2% were associated with acute pancreatitis and 25.1% were associated with chronic pancreatitis. Acute pancreatitis was managed with surgical drainage in 37.1%, whereas chronic pancreatitis was managed with surgical drainage in 49.9% (P<0.001). Multivariate logistic regression analysis of gender, insurance status, urban location, and associated pancreatitis identified chronic pancreatitis as the most statistically significant factor associated with increased utilization of surgical drainage (OR 2.36, CI 2.18-2.56, P<0.001).

Conclusions
Surgical treatment of pancreatic pseudocyst was associated with significantly more favorable mortality, length of stay, and total hospital charge outcomes. Chronic pancreatitis appears to be a significant clinical factor in the selection of surgical drainage as a treatment modality. Use of laparoscopy increased during the study period, but further studies are needed to identify both barriers and specific clinical circumstances where laparoscopy might offer benefit.