From the Vascular and Endovascular Surgery Society
Outcomes of arterial resection during pancreatectomy for tumor Natalia O. Glebova, MD, PhD,a,b Caitlin W. Hicks, MD, MS,b Jeffrey J. Tosoian, MD, MPH,c Kristen M. Piazza, MSPAS, PA-C,b Christopher J. Abularrage, MD,b Richard D. Schulick, MD, MBA,d Christopher L. Wolfgang, MD, PhD,e and James H. Black III, MD,b Aurora, Colo; and Baltimore, Md Objective: Arterial resection (AR) during pancreatic tumor resection is controversial. We examined the safety and efficacy of AR during pancreatectomy. Methods: We used a prospective institutional database that includes 6522 patients who underwent pancreatectomy from 1970 to 2014; 35 had AR. We performed a 2:1 propensity match for patients without and with AR on the basis of preoperative patient and tumor variables. We then compared operative and postoperative outcomes between matched groups. Results: AR included 18 hepatic, 8 celiac, 3 splenic, 3 middle colic, 2 superior mesenteric, and 1 left renal artery. There were 20 primary, 4 vein, and 2 graft reconstructions; 11 were emergent and 24 elective. Before matching, patients with AR were younger (58 6 2 vs 63 6 0.2 years old; P [ .05), more likely to be of black race (26% vs 9%; P [ .003), to have received preoperative chemotherapy (17% vs 2%; P < .001), have a later stage and larger tumor (4 6 0.8 vs 3 6 0.04 cm; P [ .05), more resections that included removal of all macroscopic disease, but microscopic residual tumor remained (31% vs 14%; P [ .02), greater blood loss (1285 6 276 vs 822 6 16 mL; P [ .02), and more frequent cardiac complications (11% vs 4%; P [ .03) compared with patients without AR. After propensity matching, baseline patient characteristics were similar between groups. For perioperative outcomes, the groups did not differ in surgical time, blood loss, length of stay, or complications including anastomotic leaks, bleeding, cardiac, infectious complications, or liver infarct or failure (all; P [ not significant). Patency was 97% at a mean follow-up of 510 6 184 days with 1 hepatic artery AR thrombosis. Long-term outcomes were significantly different: patients with AR had a lower rate of local tumor recurrence (20% vs 47%; P [ .007) but also lower 1-year (50% vs 87%; P [ .002) and median survival (22 6 18 vs 49 6 7 months; P [ .002). Conclusions: AR during pancreatectomy is safe and not associated with increased complications. Although it significantly reduces the risk of local tumor recurrence, AR is associated with worse survival compared with patients who do not undergo AR. (J Vasc Surg 2015;-:1-8.)
Pancreatic cancer is a lethal disease with an overall 5-year survival rate of 6%. The high mortality associated with the disease is, to a certain extent, secondary to frequently late diagnosis in the absence of clear early symptoms and the resultant presentation with metastatic and
From the Section of Vascular Surgery and Endovascular Therapy,a Division of GI, Tumor, and Endocrine Surgery,d Department of Surgery, University of Colorado Denver, Aurora; and the Division of Vascular Surgery and Endovascular Therapy, Department of Surgery,b Department of Urology, The James Buchanan Brady Urological Institute,c and Division of Surgical Oncology, Department of Surgery,e Johns Hopkins Hospital, Baltimore. Author conflict of interest: none. Presented at the Vascular and Endovascular Surgery Society Paper Session 2 at the 2015 Vascular Annual Meeting of the Society for Vascular Surgery, Chicago, Ill, June 17-20, 2015. Additional material for this article may be found online at www.jvascsurg.org. Correspondence: James H. Black III, MD, Division of Vascular Surgery and Endovascular Therapy, Department of Surgery, Johns Hopkins Hospital, 600 N Wolfe St, Halsted 668, Baltimore, MD 21287 (e-mail: jhblack@ jhmi.edu). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2015 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2015.09.042
locally or regionally advanced disease.1 Surgical resection is the only chance for long-term survival but is possible in a minority of patients. Even among patients who receive neoadjuvant therapy and surgical resection, the overall 5-year survival rate is approximately 28%.2 Because surgical resection provides the best chance for survival for patients with pancreatic cancer, aggressive removal of locoregionally advanced disease is increasingly advocated. Locally advanced pancreatic malignancies might involve the vasculature, including the portal vein and superior mesenteric vein, or the arterial structures in the region. Venous resection in the setting of venous involvement with pancreatic cancer has been shown to result in patient survival that is similar to that without the requirement for such resection.3-5 Arterial resection (AR), however, is controversial, and doubt exists as to whether such a radical approach to pancreatic malignancy is beneficial to patient survival.6 Arterial involvement in pancreatic cancer might include the superior mesenteric artery or hepatic artery in cases of head of pancreas tumors, or celiac axis in cases of neck or body of pancreas tumors. It is unclear whether arterial involvement denotes true unresectability of pancreatic cancer. It might indicate an advanced disease stage, but in the absence of preoperatively detected metastases, many surgeons offer tumor resection with attendant AR and 1
2 Glebova et al
reconstruction to such patients. Some studies have shown poor long-term survival in patients who undergo AR during pancreatectomy for tumor, but it is unclear whether the need for AR results in worse outcomes because of advanced malignancy stage in such patients, or because of increased morbidity of AR itself.7 Thus, the question remains whether it is beneficial for patient survival to undergo AR during pancreatectomy for tumor. Comparison between outcomes in patients who undergo AR during pancreatectomy for tumor and those in patients who do not undergo AR is limited by the inherent differences in the two patient groups. Patients who require AR might have more advanced tumors or different tumor biology than those who do not have arterial involvement.8 Thus, the question of whether AR is beneficial for patient survival compared with standard pancreatectomy not involving AR is difficult to answer. In this study, we used a propensity matched analysis9 to control for patient and tumor characteristics to determine whether patients who undergo AR exhibit differences in survival and tumor recurrence compared with patients who do not require AR. METHODS Patient selection. We performed a retrospective review of a prospectively maintained institutional database of all pancreatectomies performed at the Johns Hopkins Hospital between 1970 and 2014. Informed patient consent for performance of clinical research was obtained at the time of the procedure. This work was approved by the institutional review board. Of 6522 available patients, we identified 98 patients who had AR (66 splenic artery; 18 hepatic artery; 8 celiac axis; 3 middle colic arteries; 2 superior mesenteric artery; 1 left renal artery). For analysis, we excluded patients who underwent open distal or laparoscopic pancreatectomy with splenic artery resection; patients who underwent total pancreatectomies with splenic artery and vein resection; and patients who underwent the Frey procedure. We excluded laparoscopic resections because arterial involvement was a contraindication for the laparoscopic approach, and thus laparoscopic resections were not included in the control group. We excluded pancreatectomies with splenic artery and vein resection because AR is part of the surgery in such cases and is not performed specifically for tumor involvement. Patients who underwent Frey or Peustow procedures for chronic pancreatitis were excluded because these were not done for tumor. This resulted in 35 patients who underwent AR for arterial involvement during pancreatic tumor removal: 18 hepatic artery (reconstructions included 15 primary repair, 2 interposition greater saphenous vein, and 1 resection without reconstruction); 8 celiac axis (3 resections, 2 primary repair, 1 greater saphenous vein patch, 1 interposition polytetrafluoroethylene, 1 interposition cryopreserved popliteal vein); 3 splenic artery (2 primary repairs, 1 resection in setting of splenic artery aneurysm); 3 middle colic arteries (all resected without reconstruction and without colon
JOURNAL OF VASCULAR SURGERY --- 2015
resection); 2 superior mesenteric artery (1 primary repair, 1 interposition with Dacron); and 1 left renal artery (resected; left nephrectomy was also performed because of gross involvement of the kidney hilum with pancreatic tumor). Thus, there were 20 primary, 4 vein, and 2 graft reconstructions; 11 were emergent and 24 elective. Emergent cases were those in which AR was not planned preoperatively and involved intraoperative vascular or transplant surgery consultation because of unexpected arterial involvement of the tumor or inadvertent arterial injury. Prosthetic graft use for arterial reconstruction was limited to emergent situations; vein reconstructions were preferred in because of gastrointestinal tract contamination during resection and potential postoperative biliary or pancreatic leaks. Because this was a retrospective analysis, we note that there likely was a selection bias regarding patients who received AR. Over time, the number of patients who underwent AR increased, which suggested that surgeons were more willing to perform surgery on patients with AR in 2014 compared with 1970. Unfortunately, because this was a retrospective analysis of a large database, we were unable to determine the basis for selection of patients who underwent ARdthe database includes only patients who had surgery, and does not include patients who underwent nonsurgical management. Thus, we could not compare patients with arterial involvement who did not undergo surgery with those who did to determine the basis for selection. For the control group, we started with 6522 patients and excluded the 98 patients who underwent AR as already described herein, as well as patients who had a different vascular resection (portal vein, 164; superior mesenteric vein, 61; inferior vena cava, 9; other vascular resection, 8). We also excluded 213 patients whose surgery was before November of 1989 (the date of the first recorded ARs); 20 patients with central pancreatectomies; 157 with laparoscopic distal pancreatectomies or enucleations; 1 with pancreaticojejunostomy revision; 5 with laparoscopic central pancreatectomies; 182 with missing surgery type; and 13 with missing date of birth or date of surgery. The final control group included 5591 patients. Statistical analysis. Patient-level, tumor, and surgical characteristics were compared between the AR and nonAR groups using the Student t-test for continuous variables and Pearson c2 or Fisher exact test for categorical variables. Descriptive data are reported as mean 6 standard error of the mean or median with interquartile range when the variable was not normally distributed, or count with percent as appropriate. Propensity matching of the two groups was performed in a 2:1 ratio. Patients were matched on the basis of year of surgery, sex, race, age, preoperative comorbidities (history of acute myocardial infarction, congestive heart failure, peripheral vascular disease, cardiovascular disease, dementia, pulmonary disease, connective tissue disease, peptic ulcer disease, liver disease, diabetes mellitus, diabetes mellitus complications, renal disease, cancer, human immunodeficiency virus, smoking); presenting symptoms (abdominal
JOURNAL OF VASCULAR SURGERY Volume -, Number -
pain, jaundice, weight loss); presence of preoperative chemotherapy, radiation, or biliary stent; type of surgery; and tumor characteristics (tumor size, grade, vascular invasion, perineural invasion, number of positive nodes, and total nodes, American Joint Committee on Cancer stage, and presence of positive margins on bile duct, pancreas neck, bowel, or vascular groove). We then compared short- and long-term outcomes in the AR and matched control groups. Survival analysis was performed using the Kaplan-Meier method, and medians were compared using the log rank test. All statistical analyses were performed using JMP 9.0 (SAS Institute, Cary, NC) and Stata IC version 13.1 (College Station, TX), with statistical significance defined as P # .05. RESULTS Patient characteristics before propensity-matching. We identified 35 patients who underwent AR because of arterial involvement during pancreatic tumor removal. Compared with patients who did not have AR, those who did were younger (58 6 2.4 vs 63 6 0.2 years old; P ¼ .05), more likely to be of black race (26% vs 9%; P ¼ .003), to have history of cancer (14% vs 5%; P ¼ .009), less likely to have ever smoked (34% vs 49%; P ¼ .03), and more likely to have had preoperative chemotherapy (17% vs 2%; P < .001) or radiation (11% vs 0.34%; P # .001). Other preoperative comorbidities as well as presenting symptoms were similar between the two groups (Table I). Tumor characteristics before propensity-matching. In terms of tumor characteristics, patients who underwent AR were more likely to have a later stage and larger tumor (4 6 0.8 vs 3 6 0.04 cm; P ¼ .05). The type of surgery differed between the two groups: AR patients were more likely to undergo a classic Whipple (51% vs 24%; P < .001) compared with a pylorus-preserving pancreaticoduodenectomy (9% vs 45%; P < .001). Tumor histology and grade, as well as the presence of vascular and perineural invasion on microscopic exam were similar. AR patients had fewer complete resections with no residual microscopic tumor remaining (57% vs 70%; P ¼ .02) and more resections with removal of all macroscopic disease, but presence of microscopic residual tumor (31% vs 14%; P ¼ .02), as well as a greater number of harvested lymph nodes (20 6 2 vs 16 6 0.1; P ¼ .04), with a trend toward a higher number of positive lymph nodes (3 6 1 vs 2 6 0.04; P ¼ .06; Table II). Surgical characteristics before propensity matching. Before accounting for the patient and tumor differences described, patients who underwent AR had a significantly greater amount of intraoperative blood loss (1285 6 1633 vs 822 6 1196 mL; P ¼ .02) and transfusions of packed red blood cell units (2.2 6 3.5 vs 0.8 6 2.2; P < .001) compared with patients without AR. There were no significant differences in length of stay and surgical time (P ¼ not significant). In terms of perioperative complications, AR patients had more frequent cardiac complications (11% vs 4%; P ¼ .03), but did not differ
Glebova et al 3
Table I. Baseline patient characteristics Variable Age, years, mean 6 SEM Sex, male Race Black White Other Preoperative comorbidities MI CHF PVD CVD Dementia Pulmonary disease CTD PUD Liver disease Diabetes Diabetes complications Renal disease Cancer HIV Cigarette smoking Never Current Former Presenting symptoms Abdominal pain Jaundice Weight loss Preoperative chemotherapy Preoperative radiation Postoperative chemotherapy Postoperative radiation Preoperative endostent Preoperative PTC
AR (n ¼ 35)
No AR (n ¼ 5591)
58 6 2.4 21 (60)
63 6 0.2 2881 (52)
9 (26) 21 (60) 5 (14)
473 (9) 4720 (84) 398 (7)
1 0 1 0 0 1 0 0 1 5 0 1 5 0
(3) (0) (3) (0) (0) (3) (0) (0) (3) (14) (0) (3) (14) (0)
241 22 183 4 4 232 5 11 31 902 1 71 267 5
(4) (0.4) (3) (0.1) (0.1) (4.1) (0.1) (0.2) (0.6) (16) (0.02) (1) (5) (0.1)
12 (34) 3 (9) 4 (11)
2743 (49) 715 (13) 653 (12)
12 12 9 6 4 3 0 9 3
1733 2042 1570 124 19 637 8 1466 848
(34) (34) (26) (17) (11) (9) (0) (26) (9)
(31) (37) (28) (2) (0.3) (11) (0.1) (26) (15)
P .05 .32 .003
1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 .18 .77 1.00 .36 .009 1.00 .03
.68 .78 .75
Outcomes of arterial resection during pancreatectomy for tumor Natalia O. Glebova, MD, PhD,a,b Caitlin W. Hicks, MD, MS,b Jeffrey J. Tosoian, MD, MPH,c Kristen M. Piazza, MSPAS, PA-C,b Christopher J. Abularrage, MD,b Richard D. Schulick, MD, MBA,d Christopher L. Wolfgang, MD, PhD,e and James H. Black III, MD,b Aurora, Colo; and Baltimore, Md Objective: Arterial resection (AR) during pancreatic tumor resection is controversial. We examined the safety and efficacy of AR during pancreatectomy. Methods: We used a prospective institutional database that includes 6522 patients who underwent pancreatectomy from 1970 to 2014; 35 had AR. We performed a 2:1 propensity match for patients without and with AR on the basis of preoperative patient and tumor variables. We then compared operative and postoperative outcomes between matched groups. Results: AR included 18 hepatic, 8 celiac, 3 splenic, 3 middle colic, 2 superior mesenteric, and 1 left renal artery. There were 20 primary, 4 vein, and 2 graft reconstructions; 11 were emergent and 24 elective. Before matching, patients with AR were younger (58 6 2 vs 63 6 0.2 years old; P [ .05), more likely to be of black race (26% vs 9%; P [ .003), to have received preoperative chemotherapy (17% vs 2%; P < .001), have a later stage and larger tumor (4 6 0.8 vs 3 6 0.04 cm; P [ .05), more resections that included removal of all macroscopic disease, but microscopic residual tumor remained (31% vs 14%; P [ .02), greater blood loss (1285 6 276 vs 822 6 16 mL; P [ .02), and more frequent cardiac complications (11% vs 4%; P [ .03) compared with patients without AR. After propensity matching, baseline patient characteristics were similar between groups. For perioperative outcomes, the groups did not differ in surgical time, blood loss, length of stay, or complications including anastomotic leaks, bleeding, cardiac, infectious complications, or liver infarct or failure (all; P [ not significant). Patency was 97% at a mean follow-up of 510 6 184 days with 1 hepatic artery AR thrombosis. Long-term outcomes were significantly different: patients with AR had a lower rate of local tumor recurrence (20% vs 47%; P [ .007) but also lower 1-year (50% vs 87%; P [ .002) and median survival (22 6 18 vs 49 6 7 months; P [ .002). Conclusions: AR during pancreatectomy is safe and not associated with increased complications. Although it significantly reduces the risk of local tumor recurrence, AR is associated with worse survival compared with patients who do not undergo AR. (J Vasc Surg 2015;-:1-8.)
Pancreatic cancer is a lethal disease with an overall 5-year survival rate of 6%. The high mortality associated with the disease is, to a certain extent, secondary to frequently late diagnosis in the absence of clear early symptoms and the resultant presentation with metastatic and
From the Section of Vascular Surgery and Endovascular Therapy,a Division of GI, Tumor, and Endocrine Surgery,d Department of Surgery, University of Colorado Denver, Aurora; and the Division of Vascular Surgery and Endovascular Therapy, Department of Surgery,b Department of Urology, The James Buchanan Brady Urological Institute,c and Division of Surgical Oncology, Department of Surgery,e Johns Hopkins Hospital, Baltimore. Author conflict of interest: none. Presented at the Vascular and Endovascular Surgery Society Paper Session 2 at the 2015 Vascular Annual Meeting of the Society for Vascular Surgery, Chicago, Ill, June 17-20, 2015. Additional material for this article may be found online at www.jvascsurg.org. Correspondence: James H. Black III, MD, Division of Vascular Surgery and Endovascular Therapy, Department of Surgery, Johns Hopkins Hospital, 600 N Wolfe St, Halsted 668, Baltimore, MD 21287 (e-mail: jhblack@ jhmi.edu). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2015 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2015.09.042
locally or regionally advanced disease.1 Surgical resection is the only chance for long-term survival but is possible in a minority of patients. Even among patients who receive neoadjuvant therapy and surgical resection, the overall 5-year survival rate is approximately 28%.2 Because surgical resection provides the best chance for survival for patients with pancreatic cancer, aggressive removal of locoregionally advanced disease is increasingly advocated. Locally advanced pancreatic malignancies might involve the vasculature, including the portal vein and superior mesenteric vein, or the arterial structures in the region. Venous resection in the setting of venous involvement with pancreatic cancer has been shown to result in patient survival that is similar to that without the requirement for such resection.3-5 Arterial resection (AR), however, is controversial, and doubt exists as to whether such a radical approach to pancreatic malignancy is beneficial to patient survival.6 Arterial involvement in pancreatic cancer might include the superior mesenteric artery or hepatic artery in cases of head of pancreas tumors, or celiac axis in cases of neck or body of pancreas tumors. It is unclear whether arterial involvement denotes true unresectability of pancreatic cancer. It might indicate an advanced disease stage, but in the absence of preoperatively detected metastases, many surgeons offer tumor resection with attendant AR and 1
2 Glebova et al
reconstruction to such patients. Some studies have shown poor long-term survival in patients who undergo AR during pancreatectomy for tumor, but it is unclear whether the need for AR results in worse outcomes because of advanced malignancy stage in such patients, or because of increased morbidity of AR itself.7 Thus, the question remains whether it is beneficial for patient survival to undergo AR during pancreatectomy for tumor. Comparison between outcomes in patients who undergo AR during pancreatectomy for tumor and those in patients who do not undergo AR is limited by the inherent differences in the two patient groups. Patients who require AR might have more advanced tumors or different tumor biology than those who do not have arterial involvement.8 Thus, the question of whether AR is beneficial for patient survival compared with standard pancreatectomy not involving AR is difficult to answer. In this study, we used a propensity matched analysis9 to control for patient and tumor characteristics to determine whether patients who undergo AR exhibit differences in survival and tumor recurrence compared with patients who do not require AR. METHODS Patient selection. We performed a retrospective review of a prospectively maintained institutional database of all pancreatectomies performed at the Johns Hopkins Hospital between 1970 and 2014. Informed patient consent for performance of clinical research was obtained at the time of the procedure. This work was approved by the institutional review board. Of 6522 available patients, we identified 98 patients who had AR (66 splenic artery; 18 hepatic artery; 8 celiac axis; 3 middle colic arteries; 2 superior mesenteric artery; 1 left renal artery). For analysis, we excluded patients who underwent open distal or laparoscopic pancreatectomy with splenic artery resection; patients who underwent total pancreatectomies with splenic artery and vein resection; and patients who underwent the Frey procedure. We excluded laparoscopic resections because arterial involvement was a contraindication for the laparoscopic approach, and thus laparoscopic resections were not included in the control group. We excluded pancreatectomies with splenic artery and vein resection because AR is part of the surgery in such cases and is not performed specifically for tumor involvement. Patients who underwent Frey or Peustow procedures for chronic pancreatitis were excluded because these were not done for tumor. This resulted in 35 patients who underwent AR for arterial involvement during pancreatic tumor removal: 18 hepatic artery (reconstructions included 15 primary repair, 2 interposition greater saphenous vein, and 1 resection without reconstruction); 8 celiac axis (3 resections, 2 primary repair, 1 greater saphenous vein patch, 1 interposition polytetrafluoroethylene, 1 interposition cryopreserved popliteal vein); 3 splenic artery (2 primary repairs, 1 resection in setting of splenic artery aneurysm); 3 middle colic arteries (all resected without reconstruction and without colon
JOURNAL OF VASCULAR SURGERY --- 2015
resection); 2 superior mesenteric artery (1 primary repair, 1 interposition with Dacron); and 1 left renal artery (resected; left nephrectomy was also performed because of gross involvement of the kidney hilum with pancreatic tumor). Thus, there were 20 primary, 4 vein, and 2 graft reconstructions; 11 were emergent and 24 elective. Emergent cases were those in which AR was not planned preoperatively and involved intraoperative vascular or transplant surgery consultation because of unexpected arterial involvement of the tumor or inadvertent arterial injury. Prosthetic graft use for arterial reconstruction was limited to emergent situations; vein reconstructions were preferred in because of gastrointestinal tract contamination during resection and potential postoperative biliary or pancreatic leaks. Because this was a retrospective analysis, we note that there likely was a selection bias regarding patients who received AR. Over time, the number of patients who underwent AR increased, which suggested that surgeons were more willing to perform surgery on patients with AR in 2014 compared with 1970. Unfortunately, because this was a retrospective analysis of a large database, we were unable to determine the basis for selection of patients who underwent ARdthe database includes only patients who had surgery, and does not include patients who underwent nonsurgical management. Thus, we could not compare patients with arterial involvement who did not undergo surgery with those who did to determine the basis for selection. For the control group, we started with 6522 patients and excluded the 98 patients who underwent AR as already described herein, as well as patients who had a different vascular resection (portal vein, 164; superior mesenteric vein, 61; inferior vena cava, 9; other vascular resection, 8). We also excluded 213 patients whose surgery was before November of 1989 (the date of the first recorded ARs); 20 patients with central pancreatectomies; 157 with laparoscopic distal pancreatectomies or enucleations; 1 with pancreaticojejunostomy revision; 5 with laparoscopic central pancreatectomies; 182 with missing surgery type; and 13 with missing date of birth or date of surgery. The final control group included 5591 patients. Statistical analysis. Patient-level, tumor, and surgical characteristics were compared between the AR and nonAR groups using the Student t-test for continuous variables and Pearson c2 or Fisher exact test for categorical variables. Descriptive data are reported as mean 6 standard error of the mean or median with interquartile range when the variable was not normally distributed, or count with percent as appropriate. Propensity matching of the two groups was performed in a 2:1 ratio. Patients were matched on the basis of year of surgery, sex, race, age, preoperative comorbidities (history of acute myocardial infarction, congestive heart failure, peripheral vascular disease, cardiovascular disease, dementia, pulmonary disease, connective tissue disease, peptic ulcer disease, liver disease, diabetes mellitus, diabetes mellitus complications, renal disease, cancer, human immunodeficiency virus, smoking); presenting symptoms (abdominal
JOURNAL OF VASCULAR SURGERY Volume -, Number -
pain, jaundice, weight loss); presence of preoperative chemotherapy, radiation, or biliary stent; type of surgery; and tumor characteristics (tumor size, grade, vascular invasion, perineural invasion, number of positive nodes, and total nodes, American Joint Committee on Cancer stage, and presence of positive margins on bile duct, pancreas neck, bowel, or vascular groove). We then compared short- and long-term outcomes in the AR and matched control groups. Survival analysis was performed using the Kaplan-Meier method, and medians were compared using the log rank test. All statistical analyses were performed using JMP 9.0 (SAS Institute, Cary, NC) and Stata IC version 13.1 (College Station, TX), with statistical significance defined as P # .05. RESULTS Patient characteristics before propensity-matching. We identified 35 patients who underwent AR because of arterial involvement during pancreatic tumor removal. Compared with patients who did not have AR, those who did were younger (58 6 2.4 vs 63 6 0.2 years old; P ¼ .05), more likely to be of black race (26% vs 9%; P ¼ .003), to have history of cancer (14% vs 5%; P ¼ .009), less likely to have ever smoked (34% vs 49%; P ¼ .03), and more likely to have had preoperative chemotherapy (17% vs 2%; P < .001) or radiation (11% vs 0.34%; P # .001). Other preoperative comorbidities as well as presenting symptoms were similar between the two groups (Table I). Tumor characteristics before propensity-matching. In terms of tumor characteristics, patients who underwent AR were more likely to have a later stage and larger tumor (4 6 0.8 vs 3 6 0.04 cm; P ¼ .05). The type of surgery differed between the two groups: AR patients were more likely to undergo a classic Whipple (51% vs 24%; P < .001) compared with a pylorus-preserving pancreaticoduodenectomy (9% vs 45%; P < .001). Tumor histology and grade, as well as the presence of vascular and perineural invasion on microscopic exam were similar. AR patients had fewer complete resections with no residual microscopic tumor remaining (57% vs 70%; P ¼ .02) and more resections with removal of all macroscopic disease, but presence of microscopic residual tumor (31% vs 14%; P ¼ .02), as well as a greater number of harvested lymph nodes (20 6 2 vs 16 6 0.1; P ¼ .04), with a trend toward a higher number of positive lymph nodes (3 6 1 vs 2 6 0.04; P ¼ .06; Table II). Surgical characteristics before propensity matching. Before accounting for the patient and tumor differences described, patients who underwent AR had a significantly greater amount of intraoperative blood loss (1285 6 1633 vs 822 6 1196 mL; P ¼ .02) and transfusions of packed red blood cell units (2.2 6 3.5 vs 0.8 6 2.2; P < .001) compared with patients without AR. There were no significant differences in length of stay and surgical time (P ¼ not significant). In terms of perioperative complications, AR patients had more frequent cardiac complications (11% vs 4%; P ¼ .03), but did not differ
Glebova et al 3
Table I. Baseline patient characteristics Variable Age, years, mean 6 SEM Sex, male Race Black White Other Preoperative comorbidities MI CHF PVD CVD Dementia Pulmonary disease CTD PUD Liver disease Diabetes Diabetes complications Renal disease Cancer HIV Cigarette smoking Never Current Former Presenting symptoms Abdominal pain Jaundice Weight loss Preoperative chemotherapy Preoperative radiation Postoperative chemotherapy Postoperative radiation Preoperative endostent Preoperative PTC
AR (n ¼ 35)
No AR (n ¼ 5591)
58 6 2.4 21 (60)
63 6 0.2 2881 (52)
9 (26) 21 (60) 5 (14)
473 (9) 4720 (84) 398 (7)
1 0 1 0 0 1 0 0 1 5 0 1 5 0
(3) (0) (3) (0) (0) (3) (0) (0) (3) (14) (0) (3) (14) (0)
241 22 183 4 4 232 5 11 31 902 1 71 267 5
(4) (0.4) (3) (0.1) (0.1) (4.1) (0.1) (0.2) (0.6) (16) (0.02) (1) (5) (0.1)
12 (34) 3 (9) 4 (11)
2743 (49) 715 (13) 653 (12)
12 12 9 6 4 3 0 9 3
1733 2042 1570 124 19 637 8 1466 848
(34) (34) (26) (17) (11) (9) (0) (26) (9)
(31) (37) (28) (2) (0.3) (11) (0.1) (26) (15)
P .05 .32 .003
1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 .18 .77 1.00 .36 .009 1.00 .03
.68 .78 .75
