Implications and management of pancreatic fistulas following pancreaticoduodenectomy

The Massachusetts General Hospital experience

Gregory Veillette, Ismael Dominguez, Cristina Ferrone, Sarah P Thayer, Deborah McGrath, Andrew L. Warshaw, Carlos Fernández-del Castillo

Research output: Contribution to journalArticle

76 Citations (Scopus)

Abstract

Objective: To describe the management and impact of pancreatic fistulas in a high-volume center. Design: Retrospective case series. Setting: Tertiary academic center. Patients: Five hundred eighty-one consecutive patients who underwent pancreaticoduodenectomy from January 2001 through June 2006. Main Outcome Measures: Development of a pancreatic fistula (defined as > 30 mL of amylase-rich fluid from drains on or after postoperative day 7, or discharge with surgical drains in place, regardless of amount); the need for additional interventions or total parenteral nutrition; other morbidity; and mortality. Results: Seventy-five patients (12.9%) developed a pancreatic fistula. Fistulas were managed with gradual withdrawal of surgical drains. This allowed for patient discharge and eventual closure at a mean of 18 days in 38.7% of cases; these were classified as low-impact fistulas. The remaining 46 patients (61.3%) had an associated abscess, required percutaneous drainage or total parenteral nutrition, or developed bleeding; these were classified as high-impact fistulas and closed a mean of 35 days after surgery. Standard 30-day in-hospital mortality was 1.9% for all pancreaticoduodenectomies and 6.7% for those who developed a pancreatic fistula. The overall fistula-related mortality was 9.3% (7 patients), all but 1 of which was related to major hemorrhage. Conclusions: More than one-third of pancreatic fistulas are clinically insignificant (low impact). The remaining 60% of fistulas have a high clinical impact and nearly an 8-fold increase in overall mortality.

Original languageEnglish (US)
Pages (from-to)476-481
Number of pages6
JournalArchives of Surgery
Volume143
Issue number5
DOIs
StatePublished - May 1 2008

Fingerprint

Pancreatic Fistula
Pancreaticoduodenectomy
General Hospitals
Fistula
Total Parenteral Nutrition
Mortality
Hemorrhage
Patient Discharge
Amylases
Hospital Mortality
Ambulatory Surgical Procedures
Abscess
Drainage
Outcome Assessment (Health Care)
Morbidity

ASJC Scopus subject areas

  • Surgery

Cite this

Implications and management of pancreatic fistulas following pancreaticoduodenectomy : The Massachusetts General Hospital experience. / Veillette, Gregory; Dominguez, Ismael; Ferrone, Cristina; Thayer, Sarah P; McGrath, Deborah; Warshaw, Andrew L.; Fernández-del Castillo, Carlos.

In: Archives of Surgery, Vol. 143, No. 5, 01.05.2008, p. 476-481.

Research output: Contribution to journalArticle

Veillette, Gregory ; Dominguez, Ismael ; Ferrone, Cristina ; Thayer, Sarah P ; McGrath, Deborah ; Warshaw, Andrew L. ; Fernández-del Castillo, Carlos. / Implications and management of pancreatic fistulas following pancreaticoduodenectomy : The Massachusetts General Hospital experience. In: Archives of Surgery. 2008 ; Vol. 143, No. 5. pp. 476-481.
@article{9ad664100f3a4337970bf2da3b8ff6ed,
title = "Implications and management of pancreatic fistulas following pancreaticoduodenectomy: The Massachusetts General Hospital experience",
abstract = "Objective: To describe the management and impact of pancreatic fistulas in a high-volume center. Design: Retrospective case series. Setting: Tertiary academic center. Patients: Five hundred eighty-one consecutive patients who underwent pancreaticoduodenectomy from January 2001 through June 2006. Main Outcome Measures: Development of a pancreatic fistula (defined as > 30 mL of amylase-rich fluid from drains on or after postoperative day 7, or discharge with surgical drains in place, regardless of amount); the need for additional interventions or total parenteral nutrition; other morbidity; and mortality. Results: Seventy-five patients (12.9{\%}) developed a pancreatic fistula. Fistulas were managed with gradual withdrawal of surgical drains. This allowed for patient discharge and eventual closure at a mean of 18 days in 38.7{\%} of cases; these were classified as low-impact fistulas. The remaining 46 patients (61.3{\%}) had an associated abscess, required percutaneous drainage or total parenteral nutrition, or developed bleeding; these were classified as high-impact fistulas and closed a mean of 35 days after surgery. Standard 30-day in-hospital mortality was 1.9{\%} for all pancreaticoduodenectomies and 6.7{\%} for those who developed a pancreatic fistula. The overall fistula-related mortality was 9.3{\%} (7 patients), all but 1 of which was related to major hemorrhage. Conclusions: More than one-third of pancreatic fistulas are clinically insignificant (low impact). The remaining 60{\%} of fistulas have a high clinical impact and nearly an 8-fold increase in overall mortality.",
author = "Gregory Veillette and Ismael Dominguez and Cristina Ferrone and Thayer, {Sarah P} and Deborah McGrath and Warshaw, {Andrew L.} and {Fern{\'a}ndez-del Castillo}, Carlos",
year = "2008",
month = "5",
day = "1",
doi = "10.1001/archsurg.143.5.476",
language = "English (US)",
volume = "143",
pages = "476--481",
journal = "JAMA Surgery",
issn = "2168-6254",
publisher = "American Medical Association",
number = "5",

}

TY - JOUR

T1 - Implications and management of pancreatic fistulas following pancreaticoduodenectomy

T2 - The Massachusetts General Hospital experience

AU - Veillette, Gregory

AU - Dominguez, Ismael

AU - Ferrone, Cristina

AU - Thayer, Sarah P

AU - McGrath, Deborah

AU - Warshaw, Andrew L.

AU - Fernández-del Castillo, Carlos

PY - 2008/5/1

Y1 - 2008/5/1

N2 - Objective: To describe the management and impact of pancreatic fistulas in a high-volume center. Design: Retrospective case series. Setting: Tertiary academic center. Patients: Five hundred eighty-one consecutive patients who underwent pancreaticoduodenectomy from January 2001 through June 2006. Main Outcome Measures: Development of a pancreatic fistula (defined as > 30 mL of amylase-rich fluid from drains on or after postoperative day 7, or discharge with surgical drains in place, regardless of amount); the need for additional interventions or total parenteral nutrition; other morbidity; and mortality. Results: Seventy-five patients (12.9%) developed a pancreatic fistula. Fistulas were managed with gradual withdrawal of surgical drains. This allowed for patient discharge and eventual closure at a mean of 18 days in 38.7% of cases; these were classified as low-impact fistulas. The remaining 46 patients (61.3%) had an associated abscess, required percutaneous drainage or total parenteral nutrition, or developed bleeding; these were classified as high-impact fistulas and closed a mean of 35 days after surgery. Standard 30-day in-hospital mortality was 1.9% for all pancreaticoduodenectomies and 6.7% for those who developed a pancreatic fistula. The overall fistula-related mortality was 9.3% (7 patients), all but 1 of which was related to major hemorrhage. Conclusions: More than one-third of pancreatic fistulas are clinically insignificant (low impact). The remaining 60% of fistulas have a high clinical impact and nearly an 8-fold increase in overall mortality.

AB - Objective: To describe the management and impact of pancreatic fistulas in a high-volume center. Design: Retrospective case series. Setting: Tertiary academic center. Patients: Five hundred eighty-one consecutive patients who underwent pancreaticoduodenectomy from January 2001 through June 2006. Main Outcome Measures: Development of a pancreatic fistula (defined as > 30 mL of amylase-rich fluid from drains on or after postoperative day 7, or discharge with surgical drains in place, regardless of amount); the need for additional interventions or total parenteral nutrition; other morbidity; and mortality. Results: Seventy-five patients (12.9%) developed a pancreatic fistula. Fistulas were managed with gradual withdrawal of surgical drains. This allowed for patient discharge and eventual closure at a mean of 18 days in 38.7% of cases; these were classified as low-impact fistulas. The remaining 46 patients (61.3%) had an associated abscess, required percutaneous drainage or total parenteral nutrition, or developed bleeding; these were classified as high-impact fistulas and closed a mean of 35 days after surgery. Standard 30-day in-hospital mortality was 1.9% for all pancreaticoduodenectomies and 6.7% for those who developed a pancreatic fistula. The overall fistula-related mortality was 9.3% (7 patients), all but 1 of which was related to major hemorrhage. Conclusions: More than one-third of pancreatic fistulas are clinically insignificant (low impact). The remaining 60% of fistulas have a high clinical impact and nearly an 8-fold increase in overall mortality.

UR - http://www.scopus.com/inward/record.url?scp=44149114299&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=44149114299&partnerID=8YFLogxK

U2 - 10.1001/archsurg.143.5.476

DO - 10.1001/archsurg.143.5.476

M3 - Article

VL - 143

SP - 476

EP - 481

JO - JAMA Surgery

JF - JAMA Surgery

SN - 2168-6254

IS - 5

ER -