|
|
Back to 2017 Program and Abstracts
PANCREATIC DUCT SIZE AND GLAND TEXTURE DO NOT AFFECT PANCREATIC LEAK RATE AFTER DISTAL PANCREATECTOMY
Allison N. Martin*, Sowmya Narayanan, Florence E. Turrentine, Todd W. Bauer, Reid B. Adams, George J. Stukenborg, Victor M. Zaydfudim
Surgery, University of Virginia, Charlottesville, VA
INTRODUCTION: Development of a pancreatic leak and fistula remains a morbid complication after pancreatectomy. Small pancreatic duct size and soft gland texture have been associated with pancreatic fistula after pancreaticoduodenectomy. Since the mechanism of leak and fistula formation are different between pancreatic head resection and distal pancreatectomy, we hypothesized that pancreatic duct size and gland texture are not associated with pancreatic fistula after distal pancreatectomy.
METHODS: All patients ≥18 years in the 2014 ACS-NSQIP targeted pancreatectomy dataset were linked to the 2014 ACS-NSQIP PUF dataset. Clinical variables included indication for resection (benign vs. malignant), operative approach (open vs. minimally invasive), pancreatic duct size (<3 mm, 3-6 mm, vs. >6 mm) and gland texture (hard, intermediate, vs. soft). Two separate analyses were performed to evaluate associations between pancreatic duct size and gland texture after 1) distal pancreatectomy, and 2) pancreaticoduodenectomy. Additional univariable and multivariable comparisons assessed associations between clinical factors and development of pancreatic fistula after distal pancreatectomy.
RESULTS: Among 1452 patients who underwent distal pancreatectomy in 2014, 258 (17.8%) patients developed a postoperative pancreatic fistula in comparison to 534 of 2934 patients (18.2%) who developed a fistula after pancreaticoduodenectomy. Patients who developed a pancreatic fistula after distal pancreatectomy were more commonly male, had higher BMI, had longer operative time, and were more likely to receive perioperative blood transfusion (all p< 0.012) compared to patients who did not develop a fistula. Both pancreatic duct size and gland texture were significantly associated with pancreatic fistula after pancreaticoduodenectomy (both p<0.001). However, there was no association between pancreatic fistula and duct size or gland texture (both p≥0.282) after distal pancreatectomy (Table 1). Operative approach (open 17.5% leak rate vs. minimally invasive 18.5% leak rate) was not associated with pancreatic leak (p=0.612) after distal pancreatectomy. Patients with pancreatic fistula after distal pancreatectomy had increased length of stay, higher rates of readmission and reoperation compared to patients who did not have a pancreatic fistula (all p<0.005). There was no association between pancreatic fistula and mortality (p=0.464). After adjustment for demographic covariates, only operative time was significantly associated with pancreatic fistula formation (p= 0.032) after distal pancreatectomy.
CONCLUSION: Unlike among patients who had pancreaticoduodenectomy, pancreatic duct size and gland texture were not associated with development of pancreatic fistula following distal pancreatectomy. Other clinical factors should be considered in this patient population.
Characteristics among Patients with and without Pancreatic Fistula (N = 1452)
| | Pancreatic Fistula (N=259) | No Fistula (N = 1179) | p-value | | Age, Median (IQR) | 62 (51-69) | 62 (52-71) | 0.123 | | Male Sex, N (%) | 128 (49.4) | 474 (40.2) | 0.006 | | BMI, Median (IQR) | 29.4 (24.6-33.1) | 27.8 (24.1-31.8) | 0.012 | | Race/Ethnicity, N (%) | | | | White
Non-White | 192 (80.0)
48 (20.0) | 861 (80.5)
209 (19.5) | 0.776 | | Surgical Approach, N (%) | | | | Open
Minimally Invasive | 129 (50.0)
129 (50.0) | 609 (51.7)
568 (48.3) | 0.612 | | Malignant Diagnosis, N (%) | 139 (53.9) | 647 (55.0) | 0.749 | | Operative time, Median (IQR) | 243 (188-322) | 209 (158-282) | <0.001 | | Diabetes, N (%) | 61 (23.6) | 287 (24.3) | 0.788 | | Smoking, N (%) | 51 (19.7) | 189 (16.0) | 0.153 | | Reoperation, N (%) | 17 (6.6) | 35 (3.0) | 0.005 | | Readmission, N (%) | 98 (37.8) | 137 (11.7) | <0.001 | | Death, N (%) | 4 (1.5) | 12 (1.0) | 0.464 | | Length of Stay, Median (IQR) | 7 (5-10) | 6 (4-7) | <0.001 | | Duct Size, N (%) | | | | <3 mm
3-6 mm
>6 mm |
32 (53.3)
20 (33.3)
8 (13.3) |
165 (63.0)
62 (23.7)
35 (13.4) | 0.282 | | Gland Texture, N (%) | | | | Hard
Intermediate
Soft |
31 (33.3)
8 (8.6)
54 (58.1) |
118 (26.3)
35 (7.8)
295 (65.9) | 0.341 | | Surgical Drain, N (%) | 244 (94.2) | 986 (83.8) | <0.001 | | Percutaneous Drain, N (%) | 92 (37.1) | 63 (5.7) | <0.001 | | Peak Amylase POD#1, Median (IQR) | 3746.5 (977-9271) | 900.5 (177-3620) | <0.001 | | Peak Amylase POD#2-30, Median (IQR) | 3247 (1062-15874) | 91 (30-432.5) | <0.001 | | Transfusion within 72 hours, N (%) | 48 (18.6) | 141 (12.0) | 0.004 |
IQR: interquartile range; BMI: body mass index; POD: postoperative day
Back to 2017 Program and Abstracts
|
