TY - JOUR
T1 - Diagnostic Ascertainment of Suspicious Pancreatic Mass
T2 - A Threshold Analysis
AU - Sonnenberg, Amnon
AU - Rodriguez, Sarah A.
AU - Faigel, Douglas O.
PY - 2008/10/1
Y1 - 2008/10/1
N2 - Background & Aims: It is frequently difficult to differentiate between a benign and malignant pancreatic mass. The aim of this study was to assess the parameters that affect the decision to perform surgery on a suspicious pancreatic head lesion. Methods: A cost-benefit analysis, using decision tree and threshold analysis, accumulates costs and quality-adjusted life years in patients undergoing pancreaticoduodenectomy or expectant management. The threshold value is defined as the diagnostic probability for pancreatic cancer when the cost-benefit relationships of pancreaticoduodenectomy or expectant management are equal. Results: For a localized pancreatic head lesion, the threshold probability of cancer is 43%. Any higher probability of pancreatic cancer makes pancreaticoduodenectomy the preferred treatment option. Within a range of $20,000 to $80,000 spent on surgery, the threshold in favor of Whipple procedure remains relatively low at 40% to 65%. A reduced quality of life after surgery weighs against surgery and raises its threshold. Varying quality of life between 100% and 80% changes the threshold between 31% and 67%. The threshold also is increased in younger patients because of the potentially more dire consequences of unnecessary surgery in instances of long life expectancy. Conclusions: Even if diagnostic certainty cannot be achieved, it frequently is beneficial to perform surgery despite the risk of subjecting the occasional patient with benign pancreatic head lesion to an unnecessary pancreaticoduodenectomy.
AB - Background & Aims: It is frequently difficult to differentiate between a benign and malignant pancreatic mass. The aim of this study was to assess the parameters that affect the decision to perform surgery on a suspicious pancreatic head lesion. Methods: A cost-benefit analysis, using decision tree and threshold analysis, accumulates costs and quality-adjusted life years in patients undergoing pancreaticoduodenectomy or expectant management. The threshold value is defined as the diagnostic probability for pancreatic cancer when the cost-benefit relationships of pancreaticoduodenectomy or expectant management are equal. Results: For a localized pancreatic head lesion, the threshold probability of cancer is 43%. Any higher probability of pancreatic cancer makes pancreaticoduodenectomy the preferred treatment option. Within a range of $20,000 to $80,000 spent on surgery, the threshold in favor of Whipple procedure remains relatively low at 40% to 65%. A reduced quality of life after surgery weighs against surgery and raises its threshold. Varying quality of life between 100% and 80% changes the threshold between 31% and 67%. The threshold also is increased in younger patients because of the potentially more dire consequences of unnecessary surgery in instances of long life expectancy. Conclusions: Even if diagnostic certainty cannot be achieved, it frequently is beneficial to perform surgery despite the risk of subjecting the occasional patient with benign pancreatic head lesion to an unnecessary pancreaticoduodenectomy.
UR - http://www.scopus.com/inward/record.url?scp=52949093536&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=52949093536&partnerID=8YFLogxK
U2 - 10.1016/j.cgh.2008.05.015
DO - 10.1016/j.cgh.2008.05.015
M3 - Article
C2 - 18928941
AN - SCOPUS:52949093536
SN - 1542-3565
VL - 6
SP - 1162
EP - 1166
JO - Clinical Gastroenterology and Hepatology
JF - Clinical Gastroenterology and Hepatology
IS - 10
ER -