Advertisement

Safety of perioperative aspirin therapy in pancreatic operations

      Background

      Antiplatelet therapy with aspirin is prevalent among patients presenting for operative treatment of pancreatic disorders. Operative practice has called for the cessation of aspirin 7–10 days before elective procedures because of the perceived increased risk of procedure-related bleeding. Our practice at Thomas Jefferson University has been to continue aspirin therapy throughout the perioperative period in patients undergoing elective pancreatic surgery.

      Study design

      Records for patients undergoing pancreatoduodenectomy, distal pancreatectomy, or total pancreatectomy between October 2005 and February 2012 were queried for perioperative aspirin use in this institutional research board−approved retrospective study. Statistical analyses were performed with Stata software.

      Results

      During the study period, 1,017 patients underwent pancreatic resection, of whom 289 patients (28.4%) were maintained on aspirin through the morning of the operation. Patients in the aspirin group were older than those not taking aspirin (median 69 years vs 62 years, P < .0001). The estimated intraoperative blood loss was similar between the two groups, aspirin versus no aspirin (median 400 mL vs 400 mL, P = .661), as was the rate of blood transfusion anytime during the index admission (29% vs 26%, P = 0.37) and the postoperative duration of hospital stay (median 7 days vs 6 days, P = .103). The aspirin group had a slightly increased rate of cardiovascular complications (10.1% vs 7.0%, P = .107), likely reflecting their increased cardiovascular comorbidities that led to their physicians recommending aspirin therapy. Rates of pancreatic fistula (15.1% vs 13.5%, P = .490) and hospital readmissions were similar (16.9% vs 14.9%, P = .451).

      Conclusion

      This is the first study to report that aspirin therapy is not associated with increased rates of perioperative bleeding, transfusion requirement, or major procedure related complications after elective pancreatic surgery. These data suggest that continuation of aspirin is safe and that the continuation of aspirin should be considered acceptable and preferable, particularly in patients with perceived substantial medical need for treatment with antiplatelet therapy.

      Introduction

      Cardiovascular disease, including coronary artery disease (CAD), cerebrovascular disease (CVD), and peripheral vascular disease (PVD), is the leading cause of morbidity and mortality in the United States, accounting for 900,000 deaths per year.
      • Wolff T.
      • Miller T.
      • Ko S.
      Aspirin for the prevention of cardiovascular disease: U.S. Preventative Services Task Force recommendation statement.
      The risk of cardiovascular disease is increased even more in diabetic patients.
      • Pignone M.
      • Alberts M.J.
      • Colwell J.A.
      • et al.
      Aspirin for primary prevention of cardiovascular events in people with diabetes: a position statement of the American Diabetes Association, a Scientific Statement of the American Heart Association, and an Expert Consensus Document of the American College of Cardiology Foundation.
      The risk of perioperative cardiac events after major non-cardiac surgery ranges from 1.4% among the general population older than 50 years of age and increases up to 3.9% in those who are at risk of cardiac disease.
      • Devereaux P.J.
      • Goldman L.
      • Cook D.J.
      • Gilbert K.
      • Leslie K.
      • Guyatt G.H.
      Perioperative cardiac events in patients undergoing noncardiac surgery: a review of the magnitude of the problem, the pathophysiology of the events and methods to estimate and communicate risk.
      Aspirin decreases the risk of thrombotic events
      • Tohgi H.
      • Konno S.
      • Tamura K.
      • Kimura B.
      • Kawano K.
      Effects of low-to-high doses of aspirin on platelet agreeability and metabolites of thromboxane A2 and prostacyclin.
      and is the most widely prescribed antiplatelet agent in clinical practice. A randomized controlled trial showed an absolute risk reduction of 7.2% within 30 days of major noncardiac surgery when aspirin was continued.
      • Oscarsson A.
      • Gupta A.
      • Fredrikson M.
      • Jarhult J.
      • Nystrom M.
      • Pettersson E.
      • et al.
      To continue or discontinue aspirin in the perioperative period: a randomized, controlled clinical trial.
      Aspirin inhibits platelet function and, thereby, has modest effects on hemostasis. It causes irreversible inhibition of cyclooxygenase-1 (COX-1),
      • Mitchell E.A.
      • Liem T.K.
      Hemostasis and Thrombosis.
      which is necessary in the formation of prostaglandin. The prostaglandin metabolite thromboxane A2 (TXA2) causes platelet shape change and aggregation,
      • Scott-Conner C.E.H.
      • Spence R.
      • Shander A.
      • Singleton C.
      • Bennett H.
      • Rock Jr., W.A.
      Hemostasis, thrombosis, hematopoiesis, and blood transfusion.
      and the lack of TXA2 prevents platelets from having an active role in the formation of the platelet plug and vasoconstriction that leads to hemostasis. The full effect of COX inhibition on platelets is manifested just 30 minutes after the ingestion of aspirin
      • Lange R.A.
      • Hills L.D.
      Antiplatelet therapy for ischemic heart disease.
      and lasts for the lifespan of the normal human platelet, 8–10 days. Other pathways for platelet activation and aggregation remain intact, making aspirin a relatively weak anticoagulant compared with adenosine diphosphate receptor and glycoprotein IIb/IIIa inhibitors
      • Lange R.A.
      • Hills L.D.
      Antiplatelet therapy for ischemic heart disease.
      such as clopidogrel (Plavix) and eptifibatide (Integrilin), respectively.
      The fear of excessive operation-related bleeding has led to a many decade-long general recommendation that antiplatelet agents including aspirin be discontinued 7–10 days preoperatively to avoid this complication. Research has shown that interruption in aspirin therapy may lead to an increase in thrombotic events during the period that the drug is being withheld. This concept is attributed to excessive TXA2 activity and decreased fibrinolysis, the so-called “aspirin withdrawal syndrome.”
      • Beving H.
      • Zhao C.
      • Albage A.
      • Ivert T.
      Abnormally high platelet activity after discontinuation of acetylsalicylic acid treatment.
      • Harrington R.A.
      • Becker R.C.
      • Ezekowitz M.
      • Meade T.W.
      • O'Connor C.M.
      • Vorchheimer D.A.
      • et al.
      Antithrombotic therapy for coronary artery disease: The seventh ACCP conference on antithrombotic and thrombolytic therapy.
      • Chassot P.G.
      • Delabays A.
      • Spahn D.R.
      Perioperative antiplatelet therapy: the case for continuing therapy in patients at risk of myocardial infarction.
      • Biondi-Zoccai G.G.L.
      • Lotrionte M.
      • Agostoni P.
      • Abbate A.
      • Fusaro M.
      • Burzotta F.
      • et al.
      A systematic review and met-analysis on the hazards of discontinuing or not adhering to aspirin among 50,279 patients at risk for coronary artery disease.
      • Gerstein N.S.
      • Schulman P.M.
      • Gerstein W.H.
      • Petersen T.R.
      • Tawil I.
      Should more patients continue aspirin therapy perioperatively? Clinical impact of aspirin withdrawal syndrome.
      In this study we evaluated several perioperative parameters in patients undergoing major pancreatic procedures to determine whether the continuation of aspirin therapy throughout the perioperative period in this presumably greater-than-normal risk cardiac population increases the risk of procedure-related bleeding. We further analyzed differences in cardiac and noncardiac outcomes, recognizing that patients who had previously been placed on aspirin by their physicians were more likely to have an increased rate of cardiac comorbidities. Our hypothesis was that continuation of aspirin therapy would be the safest practice to minimize cardiovascular thrombotic events in patients and that it would not be associated with an increased risk of perioperative bleeding.

      Methods

      Since October 2005 patients undergoing major elective pancreatic resection (pancreatoduodenectomy, distal pancreatectomy, or total pancreatectomy) at the Thomas Jefferson University Hospital (Philadelphia, PA) who were taking daily aspirin have been advised to continue aspirin therapy through the morning of the operation. This practice involved a concerted and intentional effort to change the previous routine practice of discontinuing aspirin preoperatively for patients undergoing operative procedures.
      The senior surgeon in this study (C.J.Y.), as the Chair of the Department of Surgery, worked to alter the practices in the Patient Testing Center, educating those health care professionals on the merits of continuing aspirin as well as the potentially important risks of aspirin withdrawal syndrome. We were not able to clearly assess the reason for aspirin therapy in our patient population, and therefore we chose to universally continue aspirin independent of the clear or unclear reasons for its use.
      The patients were queried about the continuation of their aspirin on the morning of their procedure by their attending surgeon. If they had stopped their aspirin on the basis of an erroneous instructions they had received from others (this occurred rarely), then they were given 81 mg of enteric-coated aspirin orally in the preoperative area 1 hour before the induction of anesthesia. After the operation, oral aspirin therapy has been resumed on postoperative day 1 (POD 1) to eliminate interruptions in antiplatelet therapy.
      Patients undergoing pancreatoduodenectomy
      • Kennedy E.P.
      • Rosato E.L.
      • Sauter P.K.
      • Rosenberg L.M.
      • Doria C.
      • Marino I.R.
      • et al.
      Initiation of a critical pathway for pancreaticoduodenectomy at an academic institution—the first step in multi-disciplinary team building.
      and distal pancreatectomy
      • Kennedy E.P.
      • Grenda T.R.
      • Sauter P.K.
      • Rosato E.L.
      • Chojnacki K.A.
      • Rosato Jr., F.E.
      • et al.
      Implementation of a critical pathway for distal pancreatectomy at an academic institution.
      were managed routinely on our previously published critical pathways, which include the subcutaneous administration of 5,000 units of unfractionated heparin approximately 2 hours before the skin incision, and the routine use of either heparin (5,000 units subcutaneously every 8 hours) or enoxaparin (40 mg subcutaneously every day) postoperatively, starting on the day after the operation (POD 1).
      The records for patients undergoing pancreatoduodenectomy, distal pancreatectomy, or total pancreatectomy between October 2005 and February 2012 were queried for perioperative aspirin use. A total of 1,044 patients were identified and were eligible to be included in the analysis. Twenty-seven patients were excluded because of our inability to obtain accurately their perioperative medication usage data. The total number of patients included in the final analysis was 1,017 patients, with 289 patients being aspirin users and 728 patients not using aspirin. This study was approved by the Thomas Jefferson University Institutional Review Board.

      Data collection

      The patient data were collected in a prospectively maintained, institutional review board−approved pancreatic resection database that included demographics, intraoperative parameters, pathology, postoperative duration of hospital stay, and perioperative complications. A retrospective chart review was performed to obtain detailed perioperative medication use, including use of aspirin, other COX inhibitors, ADP inhibitors, glycoprotein IIb/IIIa inhibitors, vitamin K antagonists, antithrombin III activators, and statins. A separate retrospective analysis of all blood transfusions administered to the patients was performed by obtaining data from the Thomas Jefferson University Hospital Blood Bank and including the entire index admission for the pancreatic resection.
      The primary outcomes of this study were estimated intraoperative blood loss, intraoperative or postoperative packed red blood cell transfusion, major cardiovascular events, and 30-day all-cause mortality. Secondary outcomes assessed were anastomotic leak rate, postoperative duration of hospital stay, and readmissions. Specific complications assessed were delayed gastric emptying, pancreatic fistula, intra-abdominal abscess, arrhythmias, cardiac ischemia, Clostridium difficile diarrhea, pulmonary complication, small bowel obstruction, sepsis, peptic ulcer, urinary tract infection, superficial surgical-site infection, and anastomotic leak. These complications have been defined consistently, reported previously from our institution,
      • Assifi M.M.
      • Lindenmeyer J.
      • Leiby B.E.
      • Grunwald Z.
      • Rosato E.L.
      • Kennedy E.P.
      • et al.
      Surgical Apgar score predicts perioperative morbidity in patients undergoing pancreaticoduodenectomy at a high-volume center.
      • Lavu H.
      • Kennedy E.P.
      • Mazo R.
      • Stewart R.J.
      • Greenleaf C.
      • Grenda D.R.
      • et al.
      Preoperative mechanical bowel preparation does not offer a benefit for patients who undergo pancreaticoduodenectomy.
      • Berger A.C.
      • Howard T.J.
      • Kennedy E.P.
      • Sauter P.K.
      • Bower-Cherry M.
      • Dutkevitch S.
      • et al.
      Does type of pancreaticojejunostomy after pancreaticoduodenectomy decrease rate of pancreatic fistula? A randomized, prospective, dual-institution trial.
      and are consistent with the definitions of the International Study Group of Pancreatic Fistula and the International Study Group of Pancreatic Surgery.
      • Bassi C.
      • Dervenis C.
      • Butturini G.
      • Fingerhut A.
      • Yeo C.J.
      • Izbicki J.
      • et al.
      International Study Group on Pancreatic Fistula Definition. Postoperative pancreatic fistula: an international study group (ISGPF) definition.
      • Wente M.N.
      • Veit J.A.
      • Bassi C.
      • Dervenis C.
      • Fingerhut A.
      • Gouma D.J.
      • et al.
      Postoperative hemorrhage following pancreatic resection—An International Study Group of Pancreatic Surgery (ISGPS) definition.
      • Wente M.N.
      • Bassi C.
      • Dervenis C.
      • Fingerhut A.
      • Gouma D.J.
      • Izbicki J.R.
      • et al.
      Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS).

      Statistical analyses

      Statistical analyses were performed using Intercooled Stata version 12 software (College Station, TX). Comparisons between continuous variables were made using the Wilcoxon rank sum test, and categorical variables were compared with the χ2 test. Averages are reported as median values.

      Results

      Patient population

      The study consisted of 1,017 patients who underwent pancreatoduodenectomy, distal pancreatectomy, or total pancreatectomy. Table I presents the demographics of the study population as a whole and for the two cohorts, aspirin users and aspirin nonusers. Overall, 289 patients (28.4%) took aspirin perioperatively. Aspirin users were older (69 years vs 62 years; P < .001), were more likely to be male (59.9% vs 42.2%; P < .001) and had a lesser body mass index. The type of operation performed was similar between the aspirin use group and the aspirin nonuse group. Complete data on medical history and preoperative procedures were not available in the database. It is posited that patients on perioperative aspirin therapy were either taking it for primary or secondary prevention of cardiovascular disease and therefore were at greater risk for cardiovascular complications related to such conditions. Our database did not allow us to determine the precise reason for aspirin usage.
      Table IPatient demographics
      All patients (n = 1,017)Aspirin users (n = 289)Aspirin nonusers (n = 728)P value
      Age (y), median (range)65 (18–92)69 (40–87)62 (18–92)<.001
      Male sex, n (%)480 (47.2)173 (59.9)307 (42.2)<.001
      BMI, kg/m2, median (range)26 (17.5–56.4)27 (17.5–52.6)26 (19.4–56.4).002
      Pancreatoduodenectomy, n (%)688 (67.6)204 (70.6)484 (66.5)N/A
      Distal pancreatectomy, n (%)322 (31.7)82 (28.4)240 (33.0)N/A
      Total pancreatectomy, n (%)7 (0.7)3 (1.0)4 (0.6)N/A
      BMI, Body mass index; N/A, not applicable; SD, standard deviation.
      Of those patients taking aspirin, the usual daily dose was 81 mg. Only 56 patients (19.4%) were on a daily dose of 325 mg. Medications that were recorded as being discontinued before the procedure included warfarin (Coumadin; 3.4%), clopidogrel (Plavix; 3.3%), and cilostazole (Pletal; 0.3%). Patients taking these nonaspirin anticoagulants were told to stop these medications seven days before the day of the operation.
      The total population of 1,017 patients included 688 pancreatoduodenectomies (67.6%), 322 distal pancreatectomies with splenectomy (31.7%), and 7 total pancreatectomies (0.7%). Aspirin was used by 29.7%, 25.6%, and 42.9% of patients in each of the three groups, respectively.

      Intraoperative parameters

      Intraoperative parameters were not different between the aspirin group and the aspirin nonuser group (Table II). The median intraoperative blood loss between the two groups was equivalent, aspirin versus no aspirin (median 400 mL vs 400 mL, P = .661). Information on the administration of fluid and blood products was available at the time of analysis for 741 (72.9%) patients. Crystalloid infusion during the pancreas resection was nearly equivalent in groups (median 6.8 L vs 6.8 L, P = .680). Most patients received no intraoperative packed red blood cell (PRBC) transfusions, such that the median number of PRBC units transfused was zero for each group. There was one outlier case that required a 36-unit PRBC transfusion intraoperatively, a patient who was undergoing multivisceral resection of a giant pheochromocytoma that included an en bloc resection of the pancreatic tail.
      Table IIIntraoperative parameters
      All patients (n = 1,017)Aspirin users (n = 289)Aspirin nonusers (n = 728)P value
      Estimated blood loss, mL, median (range)400 (0–25,000)400 (25–25,000)400 (0–8,400).661
      Intraoperative crystalloid, L, median (range)6.5 (1.8–28)6.6 (1–28)6.5 (1.8–18.2).819
      Intraoperative transfused units PRBCs, median (range)0 (0–36)0 (0–36)0 (0–5).221
      PRBCs, Packed red blood cells; SD, standard deviation.
      In our study, the three most common pathologic findings in the resection specimens were pancreatic ductal adenocarcinoma (39.5%), intraductal papillary mucinous neoplasms (16.4%), and pancreatic endocrine tumors (8.5%). A malignancy was identified in 59.8% of all specimens. The remaining 40.2% of patients had various other benign tumors of pancreatic, ampullary, biliary, and duodenal origin as well as adjacent organs with pancreatic involvement, or chronic pancreatitis which was the indication for operation in only 51 patients (5.0%).

      Detailed transfusion data (Table III, Table IV)

      The detailed PRBC transfusion data are shown in Table III and indicated no difference in the percentage of patients receiving PRBC transfusions when comparing groups (29% vs 26%; P = .37). Furthermore the percentage of patients receiving PRBC transfusions was similar when considering those patients receiving PRBCs within 24 hours of the start of the procedure or from 24 hours postoperatively until 30 days postoperatively. Moreover, the median number of units of PRBCs transfused was zero in both groups.
      Table IIIRed blood cell transfusion data, entire cohort
      Aspirin users (n = 289)Aspirin nonusers (n = 728)P value
      Patients receiving any PRBC transfusion within 30 days of index operation; n, (%)83 (29%)189 (26%).37
      Patients receiving PRBC transfusion within 24 h of the start of the pancreatectomy; n, (%)49 (17%)127 (17%).85
      Patients receiving PRBC transfusion from 24 h after the start of the pancreatectomy to 30 days postoperatively; n, (%)45 (16%)100 (14%).45
      Number of units of PRBC transfusions within 24 h of the start of the pancreatectomy; units, median (range; SD)0 (0–38; 2.9)0 (0–18; 1.3).93
      Number of units of PRBC transfusions from 24 h after the start of the pancreatectomy to 30 days postoperatively; units, median (range; SD)0 (0–10; 1.6)0 (0–18; 1.4).39
      PRBCs, Packed red blood cells; SD, standard deviation.
      Table IVRed blood cell transfusion data by type of pancreatectomy
      PancreaticoduodenectomyTransfusion parametersAspirin users (n = 207)Aspirin nonusers (n = 488)P value
      Patients receiving PRBC transfusion within 30 days of index operation; n, (%)71 (34%)137 (28%).10
      Patients receiving PRBC transfusions within 24 h of the start of the pancreatectomy; n, (%)42 (20%)84 (17%).34
      Patients receiving PRBC transfusions from 24 h after the start of the pancreatectomy to 30 days postoperatively; n, (%)37 (18%)78 (16%).54
      Distal pancreatectomyTransfusion parametersAspirin users (n = 82)Aspirin nonusers (n = 240)P value
      Patients receiving PRBC transfusions within 30 days of index operation; n, (%)12 (15%)52 (22%).17
      Patients receiving PRBC transfusions within 24 h of the start of the pancreatectomy; n, (%)7 (9%)43 (18%).04
      Patients receiving PRBC transfusions from 24 h after the start of the pancreatectomy to 30 days postoperatively; n, (%)8 (10%)22 (9%).87
      Pancreaticoduodenectomy here includes pylorus-preserving resection, classic Whipple resection, and total pancreatectomy. All were performed open. Distal pancreatectomy includes any form of a left-sided pancreatectomy. Most were performed open, and the vast majority involved en bloc splenectomy.
      PRBCs, Packed red blood cells.
      Table IV presents the detailed PRBC transfusion data by type of operation performed, differentiating between right-sided pancreaticoduodenal resections and left-sided resections (distal pancreatectomies). For pancreaticoduodenal resections the percentage of patients receiving any PRBC transfusion within 30 days of the operation was similar (34% vs 28%, P = .37). For distal pancreatectomy the percentage of patients receiving any PRBC transfusion was also similar (22% vs 15%, P = .17). Interestingly, there was a significantly lesser rate of aspirin use in patients in the distal pancreatectomy group receiving PRBC transfusions within 24 hours of the start of the operation (9% vs 18%, P = .04). The cause of this difference is unknown and may represent a type I error.
      Of note, a subgroup analysis of the small numbers of patients taking other anticoagulant medications preoperatively (warfarin 3.4%, clopidogrel 3.3%, and cilostazole 0.3%) and stopped 7 days before operation suggested no increased bleeding risk (intraoperative estimated blood loss or need for transfusions) in this minority subgroup of patients, but the numbers of patients are too small to make any statistically sound conclusions.

      Complications (Table V)

      The overall complication rate was similar between the two groups, with complications occurring in 43.4% of the total study population, 47.0% in aspirin users, and 41.9% in aspirin nonusers (P = .141). The most common complication was pancreatic fistula, which occurred in 13.9% of all patients. The rates of pancreatic fistula (15.1% vs 13.5%, P = .490) were similar between the groups. The pancreatic fistula rate was 13.3% for those patients undergoing pancreatoduodenectomy and 15.5% for those patients undergoing distal pancreatectomy. Again, no difference in fistula rates was seen in aspirin users versus aspirin nonusers.
      Table VPostoperative outcomes
      All patients (n = 1,017)Aspirin users (n = 289)Aspirin nonusers (n = 728)P value
      Any complication, %43.447.041.9.141
      Pancreatic fistula, %13.915.113.5.490
      Delayed gastric emptying, %9.210.28.8.502
      Urinary tract infection, %9.08.89.1.881
      Cardiac complication, %
      Cardiac complications are detailed in Table VI.
      7.910.17.0.107
      Death by 30 days, %1.42.41.071
      Postoperative hospital stay, days676.111
      Readmission, %15.516.915.0.451
      DVT, Deep venous thrombosis; PE, pulmonary embolus.
      Cardiac complications are detailed in Table VI.
      Delayed gastric emptying (DGE), present in 9.2% of all patients, was the second most commonly encountered postoperative complication. Overall, no difference in DGE was seen between aspirin users and aspirin nonusers (10.2% and 8.8%, respectively, P = .502). The only statistical difference seen was after stratifying for type of procedure. In patients undergoing distal pancreatectomy, DGE occurred more commonly in aspirin users (10.0% vs 3.9%; P = .038). This difference was not encountered in patients undergoing pancreatoduodenectomy or total pancreatectomy and likely represents a type I error. Almost as common as DGE was urinary tract infection. The use of aspirin did not affect rates of urinary tract infection (8.8% vs 9.1%; P = .881).
      Cardiac complications, including dysrhythmias, ischemic events, and thrombotic/embolic events, were identified in 7.9% of all patients. The cardiac complication rate possibly tended to be greater in the aspirin group (10.1% vs 7.0%; P = .107). Details of the cardiovascular complications are presented in Table VI. Arrhythmias, new-onset heart block, myocardial infarction, sudden cardiac death caused by pulseless electrical activity, and thrombotic/embolic events were somewhat more common in the aspirin user group, although only the rates of myocardial infarction and myocardial infarction requiring percutaneous coronary intervention were observed to be significantly increased in the aspirin user group. The observed greater rate of cardiovascular complications in the aspirin group likely reflects the increased preoperative cardiac comorbidities in this cohort. Half of the deaths by 30 days (n = 14) were attributed to cardiac events.
      Table VICardiac and vascular complications
      All patients (n = 1,017)Aspirin users (n = 289)Aspirin nonusers (n = 728)P value
      Arrhythmia65 (6.4%)23 (8.0%)42 (5.8%).198
       Atrial arrhythmia50 (4.9%)18 (6.2%)32 (4.4%).223
       Ventricular arrhythmia15 (1.5%)5 (1.7%)10 (1.4%).671
      New-onset heart block6 (0.6%)2 (0.7%)4 (0.5%).207
      Myocardial Infarction8 (0.8%)5 (1.7%)3 (0.4%).032
       Requiring percutaneous intervention2 (0.2%)2 (0.7%)0 (0%).025
      Sudden cardiac death/PEA3 (0.3%)2 (0.7%)1 (0.1%).141
      Thrombotic/embolic events17 (1.7%)7 (2.4%)10 (1.4%).671
       Deep venous thrombosis9 (0.9%)4 (1.4%)5 (0.8%).284
       Pulmonary embolism6 (0.6%)3 (1%)3 (0.4%).240
       Acute peripheral embolic event (not cerebral)2 (0.2%)0 (0%)2 (0.3%).372
       Stroke/transient ischemic attack0 (0%)0 (0%)0 (0%)1.0
      PEA, Pulseless electrical activity.
      There was also no difference between the groups in superficial surgical-site infections (9.5% vs 7.7%, P = .336), with a rate of 8.2% for the entire study population. Other complications without differences in groups included pulmonary complications (5.1%), sepsis (2.8%), and C. difficile infection (2.3%).

      Hospitalization and outcomes (Table V)

      The median postoperative duration of hospital stay among the aspirin users tended to be 1 day greater than among the aspirin non-users (7 vs 6 days; P = .111). The postoperative duration of hospital stay ranged from 4 to 62 days. The overall readmission rate was 15.5%. Readmission rates were similar between the two groups (16.9% vs 15.0%; P = .451). There were no differences noted as far as the causes of readmission.
      There were 14 deaths (1.4%) within 30 days of operation, 7 in each group (2.4% and 0.9%, respectively, P = .071). As discussed previously, cardiovascular complications accounted for 50% of the deaths, with an additional 14.3% due to pulmonary causes. The majority of deaths (92.9%) were in patients who had undergone pancreatoduodenectomy, with only one mortality reported after distal pancreatectomy and none after total pancreatectomy. Ninety-day mortality for the entire population was 2.0%, with no differences between the two groups.

      Discussion

      We embarked on this analysis to provide some insight into the question of whether it is safe to continue the anti-platelet agent aspirin in the perioperative period in patients undergoing major elective pancreatic resection. Certainly, in years past, it was standard procedure for cardiologists, surgeons, dentists, and health care professionals staffing “preoperative testing centers” to instruct patients scheduled to undergo major operative (and even some dental) procedures to discontinue aspirin use 7–10 days before the scheduled procedure. Such discontinuation was recommended because of the perceived increased risk of procedure-related bleeding. On the basis of new observations and the data provided in this study, we maintain that many patients in the past may have sustained (or been placed at risk for) postoperative cardiovascular or peripheral thrombotic events because of the unnecessary cessation of aspirin therapy. Our hypothesis was that continuation of aspirin therapy would be the safest practice for patients, avoiding cardiovascular events related to aspirin withdrawal and would not be associated with an increased perioperative bleeding risk.
      In this retrospective observational study of 1,017 patients undergoing major pancreatic resection, no detrimental effects were seen in patients who continued aspirin therapy throughout the perioperative period. A trend (P = .107) was observed toward increased cardiovascular complications in the patients taking aspirin. This trend is most likely a reflection of the increased baseline risk for such complications in the aspirin group. Importantly, our data analysis revealed no increased risk of intraoperative blood loss, wound complications, or intra-abdominal collections, and no greater rate of intraoperative or perioperative PRBC transfusions in the aspirin group.
      There are several limitations to this study. First, this was not a prospective, randomized study. Second, due to the limitations of our database, we had suboptimal data collection on preoperative medical history and specifically cardiac risk factors and the indications for aspirin use. Third, we relied on patient self-reporting of aspirin use, which may result in underreporting of aspirin use because patients occasionally fail to consider over-the-counter aspirin, or aspiring containing other medications, as a medication. Fourth, we did not measure platelet function in either group, so as to correlate self-reported aspirin use with a decrement in platelet function. Although we acknowledge these limitations, we believe that the data and analyses provided here add substantially to the literature on perioperative aspirin use in major abdominal operations and specifically pancreatectomy.
      Multiple investigations have been reported in patients undergoing cardiovascular procedures showing that at-risk patients should not discontinue aspirin therapy.
      • Chassot P.G.
      • Delabays A.
      • Spahn D.R.
      Perioperative antiplatelet therapy: the case for continuing therapy in patients at risk of myocardial infarction.
      • O'Riordan J.M.
      • Margey R.J.
      • Blake G.
      • O'Connell R.
      Antiplatelet agents in the perioperative period.
      • Cao L.
      • Young N.
      • Liu H.
      • Silvestry S.
      • Sun W.
      • Zhao N.
      • et al.
      Preoperative aspirin use and outcomes in cardiac surgery patients.
      No previous publication has specifically investigated aspirin use exclusively in patients undergoing elective pancreas resection. One might consider performing further analysis in these pancreatic surgical patients, matching those who continue to take aspirin to those with similar preoperative risk factors who discontinue aspirin preoperatively, to determine whether continuing aspirin actually decreases or suppresses the rates of cardiovascular complications. Such an analysis is beyond the scope of our study and might be perceived as unethical with the numerous studies in the literature showing increased cardiac events when aspirin is discontinued.
      On the basis of existing evidence for the “aspirin withdrawal syndrome,”
      • Biondi-Zoccai G.G.L.
      • Lotrionte M.
      • Agostoni P.
      • Abbate A.
      • Fusaro M.
      • Burzotta F.
      • et al.
      A systematic review and met-analysis on the hazards of discontinuing or not adhering to aspirin among 50,279 patients at risk for coronary artery disease.
      • Gerstein N.S.
      • Schulman P.M.
      • Gerstein W.H.
      • Petersen T.R.
      • Tawil I.
      Should more patients continue aspirin therapy perioperatively? Clinical impact of aspirin withdrawal syndrome.
      those patients who discontinue aspirin are believed to be at increased risk of thrombosis. In this study, we have shown it is safe to continue aspirin throughout the perioperative period, thereby avoiding the increased thrombosis that may occur with a rebound escalation of thromboxane production. Evidence shows that the mean interval between aspirin discontinuation and acute cardiac events is 8.5 ± 3.6 days.
      • Gerstein N.S.
      • Schulman P.M.
      • Gerstein W.H.
      • Petersen T.R.
      • Tawil I.
      Should more patients continue aspirin therapy perioperatively? Clinical impact of aspirin withdrawal syndrome.
      • Burger W.
      • Chemnitius J.M.
      • Kneissl G.D.
      • Rucker G.
      Low-dose aspirin for secondary cardiovascular prevention–cardiovascular risks after its perioperative withdrawal versus bleeding risks with its continuation–review and meta-analysis.
      Following the past standard guidelines of having patients discontinue their aspirin 7–10 days before invasive procedures is potentially harmful to patients because their discontinuation places the rebound increase in TXA2 activity and therefore thrombosis exactly at the time of exposure of platelets to defects in the endothelium, which are a necessary sequelae of major surgery. Perhaps, also important, major abdominal procedures often are associated with increased catecholamine production due to pain, stress, and inflammatory mediators, placing increased demand on the myocardium over baseline.
      Despite many surgeons' fears that preoperative aspirin use increases bleeding, this study (among others in nonpancreatic surgery) shows that this apprehension appears unfounded. Our data reveal equivalent estimated intraoperative blood loss, crystalloid infusion, transfusion requirements, and complication rates for patients on persistent aspirin therapy undergoing major elective pancreas resection. Further, we have not observed an increased amount of bleeding from tissues or problems achieving hemostasis at the time of operation in aspirin users. Although there are anecdotal reports of such difficulties, this possibility has been tested in a prior randomized study, with no differences observed.
      • Oscarsson A.
      • Gupta A.
      • Fredrikson M.
      • Jarhult J.
      • Nystrom M.
      • Pettersson E.
      • et al.
      To continue or discontinue aspirin in the perioperative period: a randomized, controlled clinical trial.
      On the basis of these data, we believe that the continuation of aspirin is safe, and that the continuation of aspirin should be considered acceptable and preferable, particularly in patients with perceived substantial medical need for treatment with antiplatelet therapy. We continue aspirin until the day of the elective operation and we restart enteric coated aspirin orally on the first day after the procedure. Furthermore, it is our practice to administer aspirin per rectum (as a suppository) to those patients who are aspirin users if they are to be NPO (ie, nothing per mouth) for prolonged periods of time. This recommendation can be applied to most operative procedures, except perhaps those in which the risks of bleeding outweigh the risk of cardiovascular events.
      • Biondi-Zoccai G.G.L.
      • Lotrionte M.
      • Agostoni P.
      • Abbate A.
      • Fusaro M.
      • Burzotta F.
      • et al.
      A systematic review and met-analysis on the hazards of discontinuing or not adhering to aspirin among 50,279 patients at risk for coronary artery disease.
      • Gerstein N.S.
      • Schulman P.M.
      • Gerstein W.H.
      • Petersen T.R.
      • Tawil I.
      Should more patients continue aspirin therapy perioperatively? Clinical impact of aspirin withdrawal syndrome.
      Such procedures may include intracranial operations and major cutaneous cosmetic procedures, in which small amounts of hematoma may lead to poor outcomes.
      The authors thank their colleagues in the Department of Anesthesiology at the Thomas Jefferson University Hospital (TJUH) for their assistance with the preoperative assessment and intraoperative management of these patients. We also wish to acknowledge the superb clinical care provided by dozens of surgical residents, scores of nurses, and our surgical critical care attendings. Additionally, they thank colleagues in the TJUH Blood Bank for their assistance with providing the blood transfusion data. Finally, the first two authors were advanced GI/HPB surgery fellows at TJUH (A.M.W., 2011-2012, and M.J.P., 2012-2013), and their fellowship training was supported by an educational grant from the Foundation for Surgical Fellowships.

      References

        • Wolff T.
        • Miller T.
        • Ko S.
        Aspirin for the prevention of cardiovascular disease: U.S. Preventative Services Task Force recommendation statement.
        Ann Intern Med. 2009; 150: 396-404
        • Pignone M.
        • Alberts M.J.
        • Colwell J.A.
        • et al.
        Aspirin for primary prevention of cardiovascular events in people with diabetes: a position statement of the American Diabetes Association, a Scientific Statement of the American Heart Association, and an Expert Consensus Document of the American College of Cardiology Foundation.
        Circulation. 2010; 121: 2694-2701
        • Devereaux P.J.
        • Goldman L.
        • Cook D.J.
        • Gilbert K.
        • Leslie K.
        • Guyatt G.H.
        Perioperative cardiac events in patients undergoing noncardiac surgery: a review of the magnitude of the problem, the pathophysiology of the events and methods to estimate and communicate risk.
        Can Med Assoc J. 2005; 173: 627-634
        • Tohgi H.
        • Konno S.
        • Tamura K.
        • Kimura B.
        • Kawano K.
        Effects of low-to-high doses of aspirin on platelet agreeability and metabolites of thromboxane A2 and prostacyclin.
        Stroke. 1992; 23: 1400-1403
        • Oscarsson A.
        • Gupta A.
        • Fredrikson M.
        • Jarhult J.
        • Nystrom M.
        • Pettersson E.
        • et al.
        To continue or discontinue aspirin in the perioperative period: a randomized, controlled clinical trial.
        Br J Anaesth. 2010; 104: 305-312
        • Mitchell E.A.
        • Liem T.K.
        Hemostasis and Thrombosis.
        in: Moore W.S. Vascular and endovascular surgery: A comprehensive review. 7th ed. Elsevier, Philadelphia2005: 71-72
        • Scott-Conner C.E.H.
        • Spence R.
        • Shander A.
        • Singleton C.
        • Bennett H.
        • Rock Jr., W.A.
        Hemostasis, thrombosis, hematopoiesis, and blood transfusion.
        in: O'Leary J.P. Capote L.R. The physiologic basis of surgery. Lippincott, Williams and Wilkins, Baltimore2002: 535
        • Lange R.A.
        • Hills L.D.
        Antiplatelet therapy for ischemic heart disease.
        N Engl J Med. 2004; 350: 277-280
        • Beving H.
        • Zhao C.
        • Albage A.
        • Ivert T.
        Abnormally high platelet activity after discontinuation of acetylsalicylic acid treatment.
        Blood Coag Fibrinolysis. 1996; 7: 5-99
        • Harrington R.A.
        • Becker R.C.
        • Ezekowitz M.
        • Meade T.W.
        • O'Connor C.M.
        • Vorchheimer D.A.
        • et al.
        Antithrombotic therapy for coronary artery disease: The seventh ACCP conference on antithrombotic and thrombolytic therapy.
        Chest. 2004; 126: 513S-548S
        • Chassot P.G.
        • Delabays A.
        • Spahn D.R.
        Perioperative antiplatelet therapy: the case for continuing therapy in patients at risk of myocardial infarction.
        Br J Anaesth. 2007; 99: 316-328
        • Biondi-Zoccai G.G.L.
        • Lotrionte M.
        • Agostoni P.
        • Abbate A.
        • Fusaro M.
        • Burzotta F.
        • et al.
        A systematic review and met-analysis on the hazards of discontinuing or not adhering to aspirin among 50,279 patients at risk for coronary artery disease.
        Eur Heart J. 2006; 27: 2667-2674
        • Gerstein N.S.
        • Schulman P.M.
        • Gerstein W.H.
        • Petersen T.R.
        • Tawil I.
        Should more patients continue aspirin therapy perioperatively? Clinical impact of aspirin withdrawal syndrome.
        Ann Surg. 2012; 255: 811-819
        • Kennedy E.P.
        • Rosato E.L.
        • Sauter P.K.
        • Rosenberg L.M.
        • Doria C.
        • Marino I.R.
        • et al.
        Initiation of a critical pathway for pancreaticoduodenectomy at an academic institution—the first step in multi-disciplinary team building.
        J Am Coll Surg. 2007; 204: 917-924
        • Kennedy E.P.
        • Grenda T.R.
        • Sauter P.K.
        • Rosato E.L.
        • Chojnacki K.A.
        • Rosato Jr., F.E.
        • et al.
        Implementation of a critical pathway for distal pancreatectomy at an academic institution.
        J Gastrointest Surg. 2009; 13: 938-944
        • Assifi M.M.
        • Lindenmeyer J.
        • Leiby B.E.
        • Grunwald Z.
        • Rosato E.L.
        • Kennedy E.P.
        • et al.
        Surgical Apgar score predicts perioperative morbidity in patients undergoing pancreaticoduodenectomy at a high-volume center.
        J Gastrointest Surg. 2012; 16: 275-281
        • Lavu H.
        • Kennedy E.P.
        • Mazo R.
        • Stewart R.J.
        • Greenleaf C.
        • Grenda D.R.
        • et al.
        Preoperative mechanical bowel preparation does not offer a benefit for patients who undergo pancreaticoduodenectomy.
        Surgery. 2010; 148: 278-284
        • Berger A.C.
        • Howard T.J.
        • Kennedy E.P.
        • Sauter P.K.
        • Bower-Cherry M.
        • Dutkevitch S.
        • et al.
        Does type of pancreaticojejunostomy after pancreaticoduodenectomy decrease rate of pancreatic fistula? A randomized, prospective, dual-institution trial.
        J Am Coll Surg. 2009; 208: 738-749
        • Bassi C.
        • Dervenis C.
        • Butturini G.
        • Fingerhut A.
        • Yeo C.J.
        • Izbicki J.
        • et al.
        International Study Group on Pancreatic Fistula Definition. Postoperative pancreatic fistula: an international study group (ISGPF) definition.
        Surgery. 2005; 138: 8-13
        • Wente M.N.
        • Veit J.A.
        • Bassi C.
        • Dervenis C.
        • Fingerhut A.
        • Gouma D.J.
        • et al.
        Postoperative hemorrhage following pancreatic resection—An International Study Group of Pancreatic Surgery (ISGPS) definition.
        Surgery. 2007; 142: 20-25
        • Wente M.N.
        • Bassi C.
        • Dervenis C.
        • Fingerhut A.
        • Gouma D.J.
        • Izbicki J.R.
        • et al.
        Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS).
        Surgery. 2007; 142: 761-768
        • O'Riordan J.M.
        • Margey R.J.
        • Blake G.
        • O'Connell R.
        Antiplatelet agents in the perioperative period.
        Arch Surg. 2009; 144: 69-76
        • Cao L.
        • Young N.
        • Liu H.
        • Silvestry S.
        • Sun W.
        • Zhao N.
        • et al.
        Preoperative aspirin use and outcomes in cardiac surgery patients.
        Ann Surg. 2012; 255: 399-404
        • Burger W.
        • Chemnitius J.M.
        • Kneissl G.D.
        • Rucker G.
        Low-dose aspirin for secondary cardiovascular prevention–cardiovascular risks after its perioperative withdrawal versus bleeding risks with its continuation–review and meta-analysis.
        J Intern Med. 2005; 257: 399-414