Abstract
Background
The fear of an early post-pancreatectomy haemorrhage (PPH) may prevent surgeons from prescribing post-operative venous thromboembolism (VTE) chemoprophylaxis. The primary hypothesis of this study was that the national post-pancreatectomy early PPH rate was lower than the rate of VTE. The secondary hypothesis was that patients at high risk for post-discharge VTE could be identified, potentially facilitating the selective use of extended chemoprophylaxis.
Patients and methods
All elective pancreatectomies were identified in the 2005 to 2010 American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) database. Factors associated with 30-day rates of (pre-versus post-discharge) VTE, early PPH (transfusions > 4 units within 72 h) and return to the operating room (ROR) with PPH were analysed.
Results
Pancreaticoduodenectomies (PD) and distal pancreatectomies (DP) numbered 9140 (66.4%) and 4631 (33.6%) out of 13 771 pancreatectomies, respectively. Event rates included: VTE (3.1%), PPH (1.1%) and ROR+PPH (0.7%). PD and DP had similar VTE rates (P > 0.05) with 31.9% of VTE occurring post-discharge. Independent risk factors for late VTE included obesity [odds ratio (OR), 1.5], age ≥ 75 years (OR, 1.8), DP (OR, 2.4) and organ space infection (OR, 2.1) (all P < 0.02).
Conclusions
Within current practice patterns, post-pancreatectomy VTE outnumber early haemorrhagic complications, which are rare. The fear of PPH should not prevent routine and timely post-pancreatectomy VTE chemoprophylaxis. Because one-third of VTE occur post-discharge, high-risk patients may benefit from post-discharge chemoprophylaxis.
Introduction
For most major abdominal operations, routine post-operative venous thromboembolism (VTE) chemoprophylaxis has been proven to be both safe and effective, and is now considered the international standard of care, especially for cancer surgery.1–7 Post-operative VTE chemoprophylaxis in the form of low-dose unfractionated heparin or low-molecular-weight heparin is a Grade 1B recommendation from the most recent American College of Chest Physicians (ACCP) guidelines, for all moderate-to high-VTE-risk patients undergoing major abdominal surgery, as long as there is no significant bleeding risk.1,2,8 In spite of this recommendation, hepatopancreatobiliary (HPB) surgeons have frequently withheld VTE chemoprophylaxis owing to the perceived risk of peri-operative haemorrhage in liver and pancreatic surgery. This traditional mentality has placed surgical HPB patients, most of whom have cancer, at risk for the significant morbidity and mortality associated with VTE.9,10
As with all major abdominal surgery, pancreatic surgeons must maintain a balance between the risk of intra-or post-operative bleeding and the risk of VTE. In addition, pancreatic cancer patients, who make up the majority of patients undergoing pancreatic surgery, have an intrinsically high risk of VTE.11,12 With regard to bleeding risk, the true incidence of an early post-pancreatectomy haemorrhage (PPH)13 has not been delineated except in institutional studies.14 Nor has it been compared with VTE event rates. In the absence of large-scale comparative data regarding the incidence of bleeding and thrombosis,14 pancreatic surgery patients have been treated based on the surgeon's discretion, with a variety of approaches to the initiation, timing and duration of post-operative VTE chemoprophylaxis. We recently reported that in liver surgery patients, the risk of VTE outweighed the risk of early post-hepatectomy bleeding events.15 In the present study, we sought to determine whether this paradigm held true for pancreatectomies as well.
A second set of issues to address is the timing of post-pancreatectomy VTE and the potential benefit of extended, post-discharge, chemoprophylaxis.16 Several previous studies of abdominal surgery patients have shown the value of extended chemoprophylaxis in high-risk patients.8,17,18 Thus, in high-VTE-risk cancer patients undergoing abdominal surgery, the ACCP currently recommends 4 weeks of extended duration chemoprophylaxis if the risk of bleeding is not high (Grade 1B recommendation).1 However, longer-term prescriptions of extended chemoprophylaxis are associated with problems related to patient quality of life, financial costs and potential bleeding risk. The most tailored strategy would, therefore, be to focus extended chemoprophylaxis on only patients who are truly high risk and thus would benefit most from the prescription. Before developing post-discharge VTE prophylaxis recommendations, there is a need to demonstrate that post-discharge VTE is a separate entity from pre-discharge VTE and that post-discharge VTE comprises a clinically relevant proportion of all post-operative VTE. Then there must be identifiable risk factors (and a large enough study cohort with sufficient events) to stratify patients into a high-risk group who would potentially benefit the most from extended chemoprophylaxis.
The primary hypothesis of the present study was that the national post-pancreatectomy VTE rate was greater than the rates of early PPH events. The secondary hypothesis was that a certain subset of pancreatectomy patients was at high risk for post-discharge VTE and that there were clinical factors that could be used to select patients for extended chemoprophylaxis. To test these hypotheses, this study was designed to evaluate the rates, timing and risk factors for pre-and post-discharge VTE after a pancreaticoduodenectomy (PD) and a distal pancreatectomy (DP).
Patients and methods
Data acquisition, patients, and definitions
From the American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) participant use file for 2005 to 2010, all pancreatectomy procedures were selected. After excluding emergency cases, enucleations, necrosectomies, pancreatic biopsies, hyperthermic intraperitoneal chemotherapy cases, islet cell transplants and ampullectomies, all remaining pancreatectomies were included for analysis. For PD, this included current procedural terminology (CPT) codes 48150, 48152, 48153 and 48154. For the purposes of the analysis, total pancreatectomies (CPT 48155) were grouped with PD. A distal pancreatectomy included CPT codes 48140, 48145 and 48146. CPT codes were analysed from the primary CPT column as well as from any of the concurrent procedure columns. CPT coding did not differentiate between open and laparoscopic cases.
For thrombotic events within 30 days of a pancreatectomy, the analysis focused on three post-operative NSQIP variables: deep vein thrombosis (DVT), pulmonary embolism (PE) or a combination of the two. VTE was defined as a clinically detected DVT or PE. When both DVT and PE occurred, the first instance of either was considered the presenting VTE. The timing of VTE was compared with discharge date and labelled as ‘early’ (pre-discharge) versus ‘late’ (post-discharge).
For post-operative haemorrhagic events, the analysis focused on two post-operative variables: post-operative ‘bleeding transfusion’ which was defined by NSQIP as a transfusion of > 4 units packed red blood cells (PRBC) within 72 h of surgery (early PPH) and unplanned return to the operating room (ROR) combined with bleeding transfusion (ROR with PPH). Risk factors for pre-and post-discharge VTE were derived from analysis of all NSQIP-collected clinical factors.
The pre-operative NSQIP risk factors that were assessed included race, age, gender, performance status, weight/body mass index, haematocrit, platelets, blood urea nitrogen, creatinine, albumin, partial thrombin time, international normalized ratio, bilirubin, white blood cell count, alkaline phosphatase, aspartate aminotransferase, American Society of Anesthesiologists (ASA) class, smoking, chronic obstructive pulmonary disease (COPD), pneumonia, ascites, sepsis, disseminated cancer, diabetes, bleeding disorder, pre-operative radiation therapy, pre-operative chemotherapy, pre-operative transfusion, pre-operative hospitalization and a previous operation within 30 days.
Intra-operative variables included operative time, intra-operative transfusion, type of pancreatectomy and concurrent major operation. Concurrent major operations were defined as adrenalectomy, gastrectomy, oesophagectomy, intestinal resection, colorectal resection, retroperitoneal tumor resection, nephrectomy and/or hepatectomy. The definition of major concurrent operations did not include splenectomy, cholecystectomy, liver wedge resection/biopsy, radiofrequency ablation, feeding tube placement and abdominal wall hernia repair.
Thirty-day post-operative outcomes included a post-operative transfusion, bleeding transfusion (early PPH), ROR, renal insufficiency or failure, respiratory failure, myocardial infarction, cardiac arrest, surgical site infection, organ space infection (OSI, including abscess or pancreatic leak), fascial dehiscence, post-operative sepsis or septic shock, length of stay and mortality. As a result of the defined limitations of NSQIP data, post-pancreatectomy mortality was defined as death within 30 postoperative days or death at a later date if the patient had a continuous hospitalization from surgery to the death date.
Statistical analysis
The association of clinical factors with VTE, bleeding complications and mortality, was analysed using the chi-squared test or Fisher's exact test for non-parametric categorical data and the Mann–Whitney U-test for non-parametric continuous data. Significant univariate risk factors were entered into a multivariate logistic regression model to determine independent associations. Analyses were performed using SPSS Statistics 19 (IBM, Armonk, NY, USA). All tests were two-sided. Statistical significance was defined as P < 0.05.
Results
Patients and pancreatectomies analysed
From 2005 to 2010, 13 771 patients met the inclusion criteria with a median age 64 years (range 16–90), 51.7% female gender and 78.6% Caucasian race. Rates of common pre-operative risk factors included 66.5% with ASA class ≥3, 19.7% with age ≥ 75 years and 26.9% with obesity [body mass index (BMI) ≥ 30 kg/m2]. The case distribution included 9140 (66.4%) PD (including 306 (2.2%) total pancreatectomies) and 4631 (33.6%) DP. Indications for surgery included malignant diagnoses in 81.7% of patients with no difference in VTE rates between malignant and benign indications (3.2% versus 2.9%, P = 0.451). Other clinically relevant pre-, intra-and post-operative factors are described in Table 1.
Table 1.
Factors associated with early post-pancreatectomy haemorrhage (PPH, > 4 units blood in first 72 h after surgery)
| Clinical Characteristic | All patients (n = 13771) |
No early PPH > 4 units |
Early PPH > 4 units |
P | |||
|---|---|---|---|---|---|---|---|
| n or median | % or range | n or median | % or range | n or median | % or range | ||
| N | 13771 | 13623 | 148 | ||||
| Pre-operative factors | |||||||
| BMI ≥ 40 kg/m2 | 537 | 3.9% | 526 | 3.9% | 11 | 7.4% | 0.026 |
| Dyspnoea | 1210 | 8.8% | 1189 | 8.7% | 21 | 14.2% | 0.020 |
| Diabetes | 3063 | 22.2% | 3019 | 22.2% | 44 | 29.7% | 0.028 |
| Lack of independent function | 380 | 2.8% | 366 | 2.7% | 14 | 9.5% | <0.001 |
| Alcohol use | 417 | 3.0% | 417 | 3.1% | 0 | 0% | 0.031 |
| Haematocrit <39% | 7189 | 53.6% | 7098 | 53.5% | 91 | 65.0% | 0.007 |
| AST > 46 IU/l | 3066 | 25.4% | 3023 | 25.3% | 43 | 34.1% | 0.023 |
| Albumin < 3.5 g/dl | 3129 | 26.5% | 3082 | 26.4% | 47 | 38.2% | 0.003 |
| ASA class ≥ 3 | 9161 | 66.5% | 9045 | 66.4% | 116 | 78.4% | 0.002 |
| Disseminated cancer | 495 | 3.6% | 479 | 3.5% | 16 | 10.8% | <0.001 |
| Bleeding disorder | 382 | 2.8% | 373 | 2.7% | 9 | 6.1% | 0.014 |
| Intra-operative | |||||||
| Operative time, min | 314 | 12–1146 | 246 | 12–892 | 313 | 12–1146 | 0.004 |
| Operative time ≥ 360 min | 5069 | 36.8% | 5000 | 36.7% | 69 | 46.6% | 0.013 |
| Any intra-operative transfusion | 3002 | 28.7% | 2910 | 28.3% | 92 | 62.6% | <0.001 |
| RBC ≥4 | 718 | 6.9% | 669 | 6.5% | 49 | 33.3% | <0.001 |
| Concurrent Major GI/GU | 990 | 7.2% | 969 | 7.1% | 21 | 14.2% | 0.001 |
| PD/total versus distal pancreatectomy | 0.628 | ||||||
| PD/total | 9140 | 66.4% | 9039 | 66.4% | 101 | 68.2% | |
| Distal | 4631 | 33.6% | 4584 | 33.6% | 47 | 31.8% | |
| Post-operative | |||||||
| Return to OR | 921 | 6.7% | 832 | 6.1% | 89 | 60.1% | <0.001 |
| Sepsis/septic shock | 1899 | 13.8% | 1838 | 13.5% | 61 | 41.2% | <0.001 |
| Acute renal insufficiency/failure | 238 | 1.7% | 217 | 1.6% | 21 | 14.2% | <0.001 |
| Failure to wean ventilator 48 h | 694 | 5.0% | 633 | 4.6% | 61 | 41.2% | <0.001 |
| Post-operative pneumonia | 706 | 5.1% | 674 | 4.9% | 32 | 21.6% | <0.001 |
| Any SSI or organ space infection | 2759 | 20.0% | 2715 | 19.9% | 44 | 29.7% | 0.003 |
| Organ space infection | 1451 | 10.5% | 1421 | 10.4% | 30 | 20.3% | <0.001 |
| Cardiac arrest | 151 | 1.1% | 137 | 1.0% | 14 | 9.5% | <0.001 |
| Severe complications | 3436 | 25.0% | 3313 | 24.3% | 123 | 83.1% | <0.001 |
| Post-operative LOS | 8 | 1–215 | 8 | 1–215 | 14 | 1–119 | <0.001 |
| Death within 30 days | 357 | 2.6% | 323 | 2.4% | 34 | 23.0% | <0.001 |
- Not significant: age, gender, race, platelets, sodium, blood urea nitrogen, creatinine, partial thrombin time, international normalized ratio, total bilirubin, alkaline phosphatase, white blood cells, steroid use; chronic obstructive pulmonary disease; previous coronary stent; previous cardiac surgery; medical hypertension; pre-operative sepsis; surgical peripheral vascular disease; smoking, pre-operative stroke, pre-operative radiation therapy, operation in the preceding 30 days; pre-operative chemotherapy within 30 days; pre-operative weight loss ≥ 10%; dehiscence; superficial SSI, deep SSI, venous thromboembolism.
- Not evaluated with < 1%: pre-operative pneumonia; ascites; varices; congestive heart failure; recent myocardial infarction; angina; rest pain, pre-operative acute renal failure or dialysis; altered mental status; pre-operative open wound; pre-operative transfusion > 4 units; post-operative MI, post-operative coma, post-operative stroke.
Severe complications include: pneumonia, reintubation, ventilator >48 h, renal insufficiency/failure, cardiac arrest, myocardial infarction, coma, stroke, sepsis, septic shock, return to OR, wound dehiscence and organ space infection.
VTE, bleeding, and transfusions in relation to type of pancreatectomy
Overall complication rates included the following: DVT (2.2%), PE (1.2%), VTE (3.1%), PPH (1.1%) and ROR with PPH (0.6%). There were only 5 (0.04%) patients who experienced both VTE and PPH after a pancreatectomy. VTE event rates were more frequent than PPH rates and/or ROR with PPH rates for both PD (3.0%, 1.1%, 0.7%) and DP (3.3%, 1.0%, 0.5%, P < 0.001, Fig. 1). Analysis of the association between the type of pancreatectomy and complication rates showed no differences in bleeding or VTE events, P = 0.362, Fig. 1. Intra-operative transfusions ≥ 1 unit PRBC were common (28.7%), as were intra-operative transfusions ≥ 2 units PRBC (21.6%), with PD transfusion rates (31.5%, 23.4%) higher than for DP (23.5%, 18.2%, respectively, P < 0.001).
Figure 1.
Venous thromboembolism (VTE) events outnumber post-operative bleeding transfusions (post-pancreatectomy haemorrhage > 4 units in first 72 h after surgery) and returns to the operating room (ROR) with bleeding transfusions. There is no difference in the rates of bleeding or thrombotic events between a pancreaticoduodenectomy and a distal pancreatectomy
Factors associated with early PPH
The univariate analysis of peri-operative factors associated with early PPH, or bleeding transfusion, is detailed in Table 1. On multivariate analysis, independent pre-and intra-operative factors associated with early PPH included the following: ASA class ≥ 3 [odds ratio (OR) 1.74, 95% confidence interval (CI) 1.08–2.79, P = 0.023], a lack of independent functional status (OR 2.10, 95% CI, 1.11–3.95, P = 0.023) and an intra-operative transfusion ≥ 4 units PRBC (OR 6.16, 95% CI, 4.12–9.21, P < 0.001, Table 2).
Table 2.
Independent factors associated with venous thromboembolism (VTE) and bleeding events after a pancreatectomy
| Risk factor | VTE |
Early PPH > 4 units |
ROR with early PPH |
||||||
|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | P | OR | 95% CI | p | OR | 95% CI | P | |
| Pre-operative | |||||||||
| Haematocrit < 39% | 1.46 | 1.13–1.88 | 0.003 | ||||||
| Age ≥ 70 years | 1.47 | 1.15–1.88 | 0.002 | ||||||
| BMI ≥ 30 kg/m2 | 1.47 | 1.14–1.90 | 0.003 | ||||||
| Bleeding disorder | 1.93 | 1.17–3.18 | 0.010 | 2.77 | 1.23–6.25 | 0.014 | |||
| Pre-operative sepsis | 1.98 | 1.19–3.29 | 0.008 | 2.83 | 1.25–6.44 | 0.013 | |||
| Disseminated cancer | 2.12 | 1.36–3.32 | 0.001 | ||||||
| ASA class ≥ 3 | 1.74 | 1.08–2.79 | 0.023 | ||||||
| Lack independent function | 2.10 | 1.11–3.95 | 0.023 | ||||||
| Intra-operative | |||||||||
| Operative time ≥ 360 min | 1.30 | 1.02–1.65 | 0.048 | ||||||
| Transfusion ≥ 4 units | 6.16 | 4.12–9.21 | <0.001 | 6.09 | 3.64–10.21 | <0.001 | |||
| Post-operative | |||||||||
| Organ space infection | 1.72 | 1.26–2.34 | 0.001 | ||||||
| Ventilator > 48 h | 3.55 | 2.57–4.91 | <0.001 | ||||||
ASA, American Society of Anesthesiologists; BMI, body mass index; CI, confidence interval; OR, odds ratio; PPH, post-pancreatectomy haemorrhage; ROR, return to operating room; VTE, venous thromboembolism.
Factors associated with ROR with early PPH
The univariate analysis of peri-operative factors associated with ROR with early PPH is detailed in Table 3. On multivariate analysis, independent pre-and intra-operative predictors of ROR with early PPH included the following: pre-operative sepsis (OR 2.77, 95% CI, 1.23–6.25, P = 0.014), a pre-operative bleeding disorder (OR 2.83, 95% CI, 1.25–6.44, P = 0.013) and an intra-operative transfusion ≥ 4 units PRBC (OR 6.09, 95% CI, 3.64–10.21, P < 0.001, Table 2).
Table 3.
Factors associated with return to the operating room (ROR) and a post-pancreatectomy haemorrhage (PPH) > 4 units blood in the first 72 h after surgery
| Clinical characteristic | All patients (n = 13771) |
No ROR with PPH >4 units |
ROR with PPH >4 units |
P | |||
|---|---|---|---|---|---|---|---|
| n or median | % or range | n or median | % or range | n or median | % or range | ||
| n | 13771 | 13682 | 89 | ||||
| Pre-operative factors | |||||||
| Lack of independent function | 380 | 2.8% | 373 | 2.7% | 7 | 7.9% | 0.003 |
| AST > 46 IU/l | 3066 | 25.4% | 3035 | 25.3% | 31 | 39.7% | 0.003 |
| Alkaline phosphatase ≥ 126 IU/l | 4411 | 36.6% | 4376 | 36.6% | 35 | 48.6% | 0.035 |
| Albumin < 3.5 g/dl | 3129 | 26.5% | 3097 | 26.4% | 32 | 42.7% | 0.002 |
| ASA class ≥ 3 | 9161 | 66.5% | 9091 | 66.4% | 70 | 78.7% | 0.015 |
| Disseminated cancer | 495 | 3.6% | 487 | 3.6% | 8 | 9.0% | 0.006 |
| Pre-operative sepsis | 467 | 3.4% | 459 | 3.4% | 8 | 9.0% | 0.003 |
| Bleeding disorder | 382 | 2.8% | 375 | 2.7% | 7 | 7.9% | 0.003 |
| Intra-operative | |||||||
| Operative time, min | 314 | 12–1146 | 313 | 12–1146 | 358 | 81–892 | 0.021 |
| Operative time ≥ 360 min | 5069 | 36.8% | 5026 | 36.7% | 43 | 48.3% | 0.024 |
| Any intra-operative transfusion | 3002 | 28.7% | 2949 | 28.5% | 53 | 59.6% | <0.001 |
| RBC ≥ 4 | 718 | 6.9% | 689 | 6.7% | 29 | 32.6% | <0.001 |
| Concurrent major GI/GU | 990 | 7.2% | 376 | 7.1% | 14 | 15.7% | 0.002 |
| PD/total versus distal pancreatectomy | 0.267 | ||||||
| PD/total | 9140 | 66.4% | 9076 | 66.3% | 64 | 71.9% | |
| Distal | 4631 | 33.6% | 4606 | 33.7% | 25 | 28.1% | |
| Post-operative | |||||||
| Sepsis/septic shock | 1899 | 13.8% | 1858 | 13.6% | 41 | 46.1% | <0.001 |
| Acute renal insufficiency/failure | 238 | 1.7% | 222 | 1.6% | 16 | 18.0% | <0.001 |
| Failure to wean ventilator 48 h | 694 | 5.0% | 649 | 4.7% | 45 | 50.6% | <0.001 |
| Post-operative pneumonia | 706 | 5.1% | 682 | 5.0% | 24 | 27.0% | <0.001 |
| Organ space infection | 1451 | 10.5% | 1431 | 10.5% | 20 | 22.5% | <0.001 |
| Cardiac arrest | 151 | 1.1% | 141 | 1.0% | 10 | 11.2% | <0.001 |
| Post-operative LOS | 8 | 1–215 | 8 | 1–215 | 13 | 1–73 | <0.001 |
| Death within 30 days | 357 | 2.6% | 334 | 2.4% | 23 | 25.8% | <0.001 |
- Not significant: age, gender, race, body mass index, platelets, sodium, blood urea nitrogen, creatinine, partial thrombin time, international normalized ratio, total bilirubin, white blood cells, steroid use; lung disease, dyspnoea, diabetes, previous coronary stent; previous cardiac surgery; medical hypertension, alcohol use, smoking, pre-operative stroke, pre-operative radiation therapy, operation in preceding 30 days, pre-operative chemotherapy, pre-operative weight loss ≥ 10%, dehiscence; superficial or deep surgical site infection, venous thromboembolism.
Factors associated with 30-day mortality
The 30-day postoperative mortality rate was 2.6%. Comparisons of risk factors for 30-day mortality are provided in Table 4. Independent risk factors for 30-day mortality included age ≥ 80 years (OR 2.56, 95% CI 1.723–3.79), pre-operative dyspnoea (OR 1.57, 95% CI 1.03–2.40), a lack of independent function (OR 2.14, 95% CI 1.28–3.57), albumin <3.5 g/dl (OR 1.47, 95% CI 1.07–2.03), an intra-operative transfusion ≥ 4 units PRBC (OR 2.65, 95% CI 1.81–3.88), PD (vs. DP, OR 1.87, 95% CI 1.26–2.77), unplanned ROR (OR 4.35, 95% CI 3.00–6.29), post-operative acute renal insufficiency/failure (OR 6.33, 95% CI 3.86–10.40), post-operative pneumonia (OR 2.35, 95% CI 1.56–3.54), ventilator need/failure to wean >48 h (OR 2.13, 95% CI 1.38–3.28) and early PPH (OR 2.05, 95% CI 1.11–3.81). VTE was not an independent factor for death.
Table 4.
Factors associated with 30-day mortality after a pancreatectomy
| Clinical characteristic | All patients (n = 13771) |
No death |
Death |
P | |||
|---|---|---|---|---|---|---|---|
| n or median | % or range | n or median | % or range | n or median | % or range | ||
| N | 13771 | 13414 | 357 | ||||
| Pre-operative factors | |||||||
| Age ≥ 80 years | 1120 | 8.1% | 1058 | 7.9% | 62 | 17.4% | <0.001 |
| Male | 6650 | 48.3% | 6459 | 48.2% | 191 | 53.5% | 0.046 |
| Dyspnoea | 1210 | 8.8% | 1150 | 8.6% | 60 | 16.8% | <0.001 |
| Diabetes | 3063 | 22.2% | 2953 | 22.0% | 110 | 30.8% | <0.001 |
| COPD | 598 | 4.3% | 569 | 4.2% | 29 | 8.1% | <0.001 |
| Previous coronary stent | 852 | 6.2% | 809 | 6.0% | 43 | 12.0% | <0.001 |
| Medical hypertension | 7145 | 51.9% | 6888 | 51.3% | 257 | 72.0% | <0.001 |
| Lack of independent function | 380 | 2.8% | 342 | 2.5% | 38 | 10.6% | <0.001 |
| > 10% weight loss | 2186 | 15.9% | 2103 | 15.7% | 83 | 23.2% | <0.001 |
| Haematocrit <39% | 7189 | 53.6% | 6972 | 53.4% | 217 | 62.7% | 0.001 |
| Alkaline phosphatase ≥126 IU/l | 4411 | 36.6% | 4270 | 36.4% | 141 | 43.9% | 0.006 |
| AST > 46 IU/l | 3066 | 25.4% | 2952 | 25.1% | 114 | 35.4% | <0.001 |
| PTT > 27 s | 6806 | 67.5% | 6606 | 67.4% | 200 | 73.5% | 0.032 |
| BUN ≥ 20 mg/dl | 2426 | 18.6% | 2339 | 18.4% | 87 | 25.1% | 0.002 |
| Creatinine ≥ 1.3 mg/dl | 1129 | 8.5% | 1068 | 8.2% | 61 | 17.7% | <0.001 |
| Albumin < 3.5 g/dl | 3129 | 26.5% | 2982 | 26.0% | 147 | 46.4% | <0.001 |
| Bilirubin > 1.0 mg/dl | 3490 | 29.4% | 3377 | 29.2% | 113 | 37.3% | 0.002 |
| Steroid use | 270 | 2.0% | 257 | 1.9% | 13 | 3.6% | 0.020 |
| ASA class ≥ 3 | 9161 | 66.5% | 8848 | 66.0% | 313 | 87.7% | <0.001 |
| INR ≥ 1.1 | 3920 | 34.0% | 3776 | 33.7% | 144 | 46.9% | <0.001 |
| Platelets < 150 000/μl | 804 | 6.0% | 772 | 5.9% | 32 | 9.2% | 0.012 |
| Pre-operative sepsis | 467 | 3.4% | 436 | 3.3% | 31 | 8.7% | <0.001 |
| Disseminated cancer | 495 | 3.6% | 472 | 3.5% | 23 | 6.4% | 0.003 |
| Bleeding disorder | 382 | 2.8% | 366 | 2.7% | 16 | 4.5% | 0.046 |
| Pre-operative radiation therapy | 304 | 2.2% | 289 | 2.2% | 15 | 4.2% | 0.009 |
| Intra-operative | |||||||
| Operative time ≥ 360 min | 5069 | 36.8% | 4886 | 36.4% | 183 | 51.3% | <0.001 |
| Any intra-operative transfusion | 3002 | 28.7% | 2863 | 28.1% | 139 | 55.6% | <0.001 |
| RBC ≥ 4 | 718 | 6.9% | 645 | 6.3% | 73 | 29.2% | <0.001 |
| Concurrent major GI/GU | 990 | 7.2% | 946 | 7.1% | 44 | 12.3% | <0.001 |
| PD/total versus distal pancreatectomy | <0.001 | ||||||
| Distal | 4632 | 33.6% | 4566 | 34.0% | 66 | 18.5% | |
| Post-operative | |||||||
| Early VTE | 290 | 2.1% | 264 | 2.0% | 26 | 7.3% | <0.001 |
| Late VTE | 136 | 1.0% | 132 | 1.0% | 4 | 1.1% | 0.782 |
| Any VTE | 426 | 3.1% | 396 | 3.0% | 30 | 8.4% | <0.001 |
| Bleeding transfusion | 148 | 1.1% | 114 | 0.8% | 34 | 9.5% | <0.001 |
| Any post-operative transfusion | 916 | 6.7% | 835 | 6.2% | 81 | 22.7% | <0.001 |
| Return to OR | 921 | 6.7% | 769 | 5.7% | 152 | 42.6% | <0.001 |
| Sepsis/septic shock | 1899 | 13.8% | 1714 | 12.8% | 185 | 51.8% | <0.001 |
| Acute renal insufficiency/failure | 238 | 1.7% | 158 | 1.2% | 80 | 22.4% | <0.001 |
| Failure to wean ventilator 48 h | 694 | 5.0% | 544 | 4.1% | 150 | 42.0% | <0.001 |
| Post-operative pneumonia | 706 | 5.1% | 616 | 4.6% | 90 | 25.2% | <0.001 |
| Dehiscence | 224 | 1.6% | 211 | 1.6% | 13 | 3.6% | 0.002 |
| Organ space infection | 1451 | 10.5% | 1376 | 10.3% | 75 | 21.0% | <0.001 |
| Cardiac arrest | 151 | 1.1% | 36 | 0.3% | 115 | 32.2% | <0.001 |
| Severe complications (not VTE) | 3436 | 25.0% | 3130 | 23.3% | 306 | 85.7% | <0.001 |
- Not significant: body mass index; race, alcohol use; white blood cells, smoking, pre-operative stroke, operation in preceding 30 days; pre-operative chemotherapy within 30 days; superficial and deep surgical site infections.
Risk factors and timing of post-pancreatectomy VTE
Comparisons of risk factors for VTE are provided in Table 5. Independent risk factors for any (either pre-or post-discharge) VTE within 30 days of surgery are listed in Table 2. Major post-operative complications associated with any VTE included OSI (OR 1.72, 95% CI, 1.26–2.34, P = 0.001) and post-operative ventilator requirement >48 h (OR 3.55, 95% CI 2.57–4.91). For PD, the median date of diagnosis of OSI was post-operative day (POD) 11, whereas for DP, it was POD 14. Relative to the median discharge dates of POD 9 for PD and POD 7 for DP, the median date of any VTE was POD 11, whereas it was POD13 for DP. Late thromboembolic events accounted for 33.7%, 35.7% and 31.9% of all observed DVT, PE, and VTE, respectively (Fig. 2). Among patients with early VTE, the median date of VTE was POD 8 with a median discharge date of POD 20. Among patients with late VTE, the median date of VTE was POD 19 with a median discharge date of POD 8.
Table 5.
Factors associated with no venous thromboembolism (VTE) compared with VTE
| Clinical characteristic | All patients (n = 13771) |
No VTE (DVT or PE) |
VTE (DVT or PE) |
P | |||
|---|---|---|---|---|---|---|---|
| n or median | % or range | n or median | % or range | n or median | % or range | ||
| N | 13771 | 13345 | 426 | ||||
| Preoperative factors | |||||||
| Median age (range), years | 64 | 16–90 | 64 | 16–90 | 66 | 20–90 | 0.004 |
| Age ≥70 years | 4572 | 33.2% | 4407 | 33.0% | 165 | 38.7% | 0.014 |
| BMI ≥ 30 kg/m2 | 3703 | 26.9% | 3566 | 26.7% | 137 | 32.2% | 0.013 |
| Dyspnoea | 1210 | 8.8% | 1156 | 8.7% | 54 | 12.7% | 0.004 |
| COPD | 598 | 4.3% | 567 | 4.2% | 31 | 7.3% | 0.003 |
| Steroid use | 270 | 2.0% | 256 | 1.9% | 14 | 3.3% | 0.045 |
| Medical hypertension | 7145 | 51.9% | 6892 | 51.6% | 253 | 59.4% | 0.002 |
| Lack of independent function | 380 | 2.8% | 349 | 2.6% | 31 | 7.3% | <0.001 |
| Pre-operative sepsis | 467 | 3.4% | 441 | 3.3% | 26 | 6.1% | 0.002 |
| INR > 1.0 | 5235 | 45.5% | 5045 | 45.2 | 190 | 52.6% | 0.005 |
| Haematocrit < 39% | 7189 | 53.6% | 6928 | 53.3% | 261 | 62.7% | <0.001 |
| WBC > 11 000/μl | 1060 | 7.9% | 1005 | 7.7% | 55 | 13.2% | <0.001 |
| Albumin < 3.5 g/dl | 3129 | 26.5% | 2989 | 26.2% | 140 | 38.3% | <0.001 |
| Disseminated cancer | 495 | 3.6% | 459 | 3.4% | 36 | 8.5% | <0.001 |
| Bleeding disorder | 382 | 2.8% | 354 | 2.7% | 28 | 6.6% | <0.001 |
| Neoplasm (versus benign) | 8618 | 81.7% | 8339 | 81.7% | 279 | 83.3% | 0.451 |
| Intra-operative | |||||||
| Operative time, min | 314 | 12–1146 | 313 | 12–1146 | 346 | 72–981 | <0.001 |
| Operative time ≥ 360 min | 5069 | 36.8% | 4872 | 36.5% | 197 | 46.2% | <0.001 |
| Any intra-operative transfusion | 3002 | 28.7% | 2864 | 28.3% | 138 | 41.1% | <0.001 |
| RBC ≥ 4 | 718 | 6.9% | 671 | 6.6% | 47 | 14.0% | <0.001 |
| Concurrent major operation | 990 | 7.2% | 945 | 7.1% | 45 | 10.6% | 0.006 |
| PD/total versus distal pancreatectomy | 0.362 | ||||||
| PD/total | 9140 | 66.4% | 8866 | 66.4% | 274 | 64.3% | |
| Distal | 4631 | 33.6% | 4479 | 33.6% | 152 | 35.7% | |
| Post-operative | |||||||
| Any post-operative transfusion (<72 h) | 916 | 6.7% | 873 | 6.5% | 43 | 10.1% | 0.004 |
| Return to OR (ROR) | 921 | 6.7% | 854 | 6.4% | 67 | 15.7% | <0.001 |
| Sepsis/septic shock | 1899 | 13.8% | 1747 | 13.1% | 152 | 35.7% | <0.001 |
| Acute renal insufficiency/failure | 238 | 1.7% | 223 | 1.7% | 15 | 3.5% | 0.004 |
| Failure to wean ventilator 48 h | 694 | 5.0% | 613 | 4.6% | 81 | 19.0% | <0.001 |
| Post-operative pneumonia | 706 | 5.1% | 644 | 4.8% | 62 | 14.6% | <0.001 |
| Deep SSI | 268 | 1.9% | 253 | 1.9% | 15 | 3.5% | 0.017 |
| Any SSI or organ space infection | 2759 | 20.0% | 2606 | 19.5% | 153 | 35.9% | <0.001 |
| Organ space infection | 1451 | 10.5% | 1357 | 10.2% | 94 | 22.1% | <0.001 |
| Fascial dehiscence | 224 | 1.6% | 208 | 1.6% | 16 | 3.8% | <0.001 |
| Severe complications | 3436 | 25.0% | 3209 | 24.0% | 227 | 53.3% | <0.001 |
| Post-operative LOS | 8 | 1–215 | 8 | 1–215 | 14 | 1–98 | <0.001 |
| Death within 30 days | 357 | 2.6% | 327 | 2.5% | 30 | 7.0% | <0.001 |
- Not significant: race, gender, American Society of Anesthesiologists score, platelets, sodium, blood urea nitrogen, creatinine, partial thrombin time, total bilirubin, aspartate aminotransferase, alkaline phosphatase, previous coronary stent, previous cardiac surgery, alcohol use, smoking, diabetes, pre-operative radiation therapy, operation in preceding 30 days, pre-operative chemotherapy; pre-operative weight loss ≥ 10%; post-operative bleeding transfusion; superficial SSI.
Figure 2.
Approximately 1 in 3 post-pancreatectomy VTE events are recognized post-discharge. (PE, pulmonary embolus; DVT, deep venous thrombosis; VTE, venous thromboembolism)
Post-pancreatectomy VTE pre-discharge risk factors
Univariate factors associated with early VTE are detailed in Table 6. Independent risk factors included a lack of independent functional status (OR 1.78, 95% CI 1.05–2.99, P = 0.031), pre-operative haematocrit < 39% (OR 1.69, 95% CI 1.17–2.43, P = 0.005), chronic obstructive lung disease (OR 1.74, 95% CI 1.01–2.99, P = 0.046), pre-operative leukocytosis (OR 2.12, 95% CI 1.43–3.16, P < 0.001), disseminated cancer (OR 2.50, 95% CI 1.02–3.32, P = 0.040), bleeding disorder (OR 2.56, 95% CI 1.47–4.44, P = 0.001), an intra-operative transfusion ≥ 2 units (OR 1.53, 95% CI 1.09–2.16, P = 0.014), operative time ≥300 min (OR 1.44, 95% CI 1.02–2.03, P = 0.039), ventilator need/failure to wean > 48 h (OR 3.74, 95% CI 2.42–5.77, P < 0.001) and OSI (OR1.64, 95% CI 1.10–2.45, P = 0.016). Patients with early VTE experienced a 30-day mortality rate of 9.0% (versus 2.6% overall baseline mortality rate, P < 0.001).
Table 6.
Factors associated with no venous thromboembolism (VTE), early VTE, and late VTE, after a pancreatectomy
| Clinical characteristic | All patients (n = 13 771) |
No VTE (DVT or PE) |
Early (pre-discharge) VTE |
Late (post-discharge) VTE |
Early versus none |
Late versus none |
Early versus late |
||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| n or median | % or range | n or median | % or range | n or median | % or range | n or median | % or range | P | P | P | |
| n | 13771 | 100% | 13345 | 96.9% | 290 | 2.1 | 136 | 1.0% | |||
| Pre-operative factors | |||||||||||
| Median age (range), years | 64 | 16–90 | 64 | 16–90 | 66 | 25–90 | 66 | 20–90 | 0.033 | 0.041 | 0.586 |
| Age ≥ 75 years | 2715 | 19.7% | 2618 | 19.6% | 60 | 20.7% | 37 | 27.2% | 0.649 | 0.027 | 0.135 |
| BMI ≥ 30 kg/m2 | 3703 | 26.9% | 3566 | 26.7% | 86 | 29.7% | 51 | 37.5% | 0.264 | 0.005 | 0.106 |
| Dyspnoea | 1210 | 8.8% | 1156 | 8.7% | 34 | 11.7% | 20 | 14.7% | 0.068 | 0.013 | 0.389 |
| COPD | 598 | 4.3% | 567 | 4.2% | 25 | 8.6% | 6 | 4.4% | <0.001 | 0.925 | 0.119 |
| Medical hypertension | 7145 | 51.9% | 6892 | 51.6% | 169 | 58.3% | 84 | 61.8% | 0.025 | 0.019 | 0.494 |
| Lack of independent function | 380 | 2.8% | 349 | 2.6% | 26 | 9.0% | 5 | 3.7% | <0.001 | 0.410 | 0.050 |
| Pre-operative sepsis | 467 | 3.4% | 441 | 3.3% | 19 | 6.6% | 7 | 5.1% | 0.002 | 0.233 | 0.572 |
| INR > 1.0 | 5235 | 45.5% | 5045 | 45.2% | 133 | 52.0% | 57 | 54.3% | 0.032 | 0.063 | 0.687 |
| Haematocrit < 39% | 7189 | 53.6% | 6928 | 53.3% | 193 | 68.7% | 68 | 50.4% | <0.001 | 0.494 | <0.001 |
| WBC > 11 000/μl | 1060 | 7.9% | 1005 | 7.7% | 46 | 16.4% | 9 | 6.7% | <0.001 | 0.641 | 0.006 |
| Albumin < 3.5 g/dl | 3129 | 26.5% | 2989 | 26.2% | 117 | 46.8% | 23 | 19.8% | <0.001 | 0.122 | <0.001 |
| ASA class ≥ 3 | 9161 | 66.5% | 8859 | 66.4% | 212 | 73.1% | 90 | 66.2% | 0.016 | 0.959 | 0.142 |
| Disseminated cancer | 495 | 3.6% | 459 | 3.4% | 27 | 9.3% | 9 | 6.6% | <0.001 | 0.044 | 0.352 |
| Bleeding disorder | 382 | 2.8% | 354 | 2.7% | 23 | 7.9% | 5 | 3.7% | <0.001 | 0.415 | 0.099 |
| Malignant (versus benign) | 8618 | 81.7% | 8339 | 81.7% | 191 | 82.0% | 88 | 86.3% | 0.905 | 0.231 | 0.332 |
| Intra-operative | |||||||||||
| Operative time, min | 314 | 12–1146 | 313 | 12–1146 | 362 | 81–981 | 291 | 72–872 | <0.001 | 0.160 | <0.001 |
| Operative time ≥ 300 min | 7492 | 54.4% | 7224 | 54.1% | 206 | 71.0% | 62 | 45.6% | <0.001 | 0.047 | <0.001 |
| Any intra-operative transfusion | 3002 | 28.7% | 2864 | 28.3% | 115 | 48.9% | 23 | 22.8% | <0.001 | 0.217 | <0.001 |
| RBC ≥ 2 | 2256 | 21.6% | 2138 | 21.2% | 101 | 43.0% | 17 | 16.8% | <0.001 | 0.290 | <0.001 |
| RBC ≥ 4 | 718 | 6.9% | 671 | 6.6% | 41 | 17.4% | 6 | 5.9% | <0.001 | 0.779 | 0.005 |
| Concurrent major operation | 990 | 7.2% | 945 | 7.1% | 30 | 10.3% | 15 | 11.0% | 0.033 | 0.075 | 0.830 |
| PD/total versus distal pancreatectomy | 0.017 | <0.001 | <0.001 | ||||||||
| PD/total | 9140 | 66.4% | 8866 | 66.4% | 212 | 73.1% | 62 | 45.6% | |||
| Distal | 4631 | 33.6% | 4479 | 33.6% | 78 | 26.9% | 74 | 54.4% | |||
| Post-operative | |||||||||||
| Any post-operative transfusion (72 h) | 916 | 6.7% | 873 | 6.5% | 36 | 12.4% | 7 | 5.1% | <0.001 | 0.512 | 0.020 |
| Return to OR (ROR) | 921 | 6.7% | 854 | 6.4% | 55 | 19.0% | 12 | 8.8% | <0.001 | 0.251 | 0.007 |
| Sepsis/septic shock | 1899 | 13.8% | 1747 | 13.1% | 117 | 40.3% | 35 | 25.7% | <0.001 | <0.001 | 0.003 |
| Acute renal insufficiency/failure | 238 | 1.7% | 223 | 1.7% | 15 | 5.2% | 0 | 0% | <0.001 | 0.178 | 0.007 |
| Failure to wean ventilator 48 h | 694 | 5.0% | 613 | 4.6% | 78 | 26.9% | 3 | 2.2% | <0.001 | 0.296 | <0.001 |
| Post-operative pneumonia | 706 | 5.1% | 644 | 4.8% | 52 | 17.9% | 10 | 7.4% | <0.001 | 0.172 | 0.004 |
| Superficial SSI | 1141 | 8.3% | 1095 | 8.2% | 35 | 12.1% | 11 | 8.1% | 0.018 | 0.961 | 0.217 |
| Deep SSI | 268 | 1.9% | 253 | 1.9% | 12 | 4.1% | 3 | 2.2% | 0.006 | 0.746 | 0.406 |
| Any SSI or OSI | 2759 | 20.0% | 2606 | 19.5% | 111 | 38.3% | 42 | 30.9% | <0.001 | 0.001 | 0.138 |
| OSI | 1451 | 10.5% | 1357 | 10.2% | 67 | 23.1% | 27 | 19.9% | <0.001 | <0.001 | 0.451 |
| Fascial dehiscence | 224 | 1.6% | 208 | 1.6% | 13 | 4.5% | 3 | 2.2% | <0.001 | 0.474 | 0.290 |
| Severe complications | 3436 | 25.0% | 3209 | 24.0% | 174 | 60.0% | 53 | 39.0% | <0.001 | <0.001 | <0.001 |
| Post-operative LOS | 8 | 1–215 | 8 | 1–215 | 20 | 7–98 | 8 | 3–24 | 0.001 | 0.017 | <0.001 |
| Death within 30 days | 357 | 2.6% | 327 | 2.5% | 26 | 9.0% | 4 | 2.9% | <0.001 | 0.579 | 0.025 |
- Not significant: race, gender, platelets, sodium, blood urea nitrogen, creatinine, partial thrombin time, total bilirubin, aspartate aminotransferase, alkaline phosphatase, previous coronary stent; previous cardiac surgery; surgical peripheral vascular disease; alcohol use, smoking, diabetes, steroid use, pre-operative radiation therapy, operation in preceding 30 days, pre-operative chemotherapy; pre-operative weight loss ≥ 10%; post-operative bleeding transfusion.
Bold type highlights univariate comparisons with P < 0.05.
Post-pancreatectomy post-discharge VTE risk factors
Univariate factors associated with late VTE are detailed in Table 6. Multivariate analysis identified the following independent risk factors associated with post-discharge VTE: obesity (OR 1.53, 95% CI, 1.08–2.19, P = 0.018) and age ≥ 75 years (OR 1.76, 95% CI 1.20–2.58, P = 0.004), distal pancreatectomy (OR 2.41, 95% CI 1.71–3.39) and OSI (OR 2.14, 95% CI 1.39–3.28, P < 0.001). Of the 136 patients with late VTE, 89 (65.4%) had at least one of these 4 risk factors. Patients with late VTE experienced a 30-day mortality rate of 2.9% (versus 2.6% overall baseline mortality rate, P = 0.579).
Discussion
The primary aims of this study were to compare the post-operative rates of thromboembolic and bleeding events after pancreatectomy and to identify risk factors for VTE, early PPH and ROR with PPH, in patients undergoing PD and DP. Based on standardized definitions and the presence of trained reviewers for data collection, the ACS-NSQIP database was well-suited to address each of these aims.15,19,20 This analysis determined that, within the context of current national practice patterns, the risk of VTE outweighs the risk of haemorrhagic events for both PD and DP. In addition, the study identified unique risk factors for both pre-and post-discharge VTE after pancreatic surgery. These data suggest that late VTE is a clinically relevant complication after a pancreatectomy with one-third of post-operative VTE occurring post-discharge. Finally, by isolating the risk factors for late VTE, including obesity, age ≥75 years, DP and/or OSI, the results of this study provide a basis for identifying high-risk patients for selective use of extended chemoprophylaxis.
There is general consensus that post-operative VTE chemoprophylaxis should be prescribed for all patients undergoing major abdominal surgery if the patient's acute bleeding risk is acceptably lower than the thrombotic risk.1,2 The transferability of these recommendations to complex HPB surgery, in which intra-operative bleeding and intra-operative transfusions are common, is less clear. To disprove or to corroborate historical fears of early PPH, an accurate comparison of VTE and bleeding event rates after a pancreatectomy is needed to determine the safety and practicality of recommending routine post-operative VTE chemoprophylaxis.
It is estimated that as many as 30% of pancreatic resections in the United States are currently being performed at NSQIP member institutions. By analysing a large national sampling of patients, this study showed that early post-operative haemorrhagic events, although traditionally feared, are quite rare (<1% event rate). Unlike early PPH which is likely related to intra-operative technical issues,14 ongoing surgical bleeding, or uncorrected coagulopathy (all of which are contraindications for immediately starting post-operative VTE chemoprophylaxis), late PPH is generally related to abdominal sepsis, undrained pancreatic leaks and vascular pseudoaneurysms. Therefore, these unpredictable late bleeding events may not represent a barrier to the routine use of chemoprophylaxis. The number of patients who experienced both VTE and PPH was extremely low (5 of 13 771), to the point that detailed analysis of risk factors for this difficult clinical scenario was not feasible.
The comparative rarity of early haemorrhagic events combined with the higher risk of post-operative VTE in both PD and DP make a very strong argument for routine and timely post-operative chemoprophylaxis in all pancreatectomy patients. Although early PPH is an independent risk factor for 30-day mortality, its very low event rate is associated with the defined risk factor of a major intra-operative transfusion (≥4 units pRBC). As the most significant independent risk factor for early PPH or ROR with PPH was an intra-operative transfusion ≥ 4 units PRBC, the main exception to the regimented delivery of immediate post-operative VTE chemoprophylaxis would be patients with ongoing signs of active bleeding in the immediate post-operative period. For these rare patients, haemostasis should be confirmed before initiating VTE chemoprophylaxis.
There is ample evidence that institutional protocols for post-operative (and even pre-induction dosing)21 have led to reduced rates of post-operative VTE. However, there remains no consensus on the length of post-operative therapy, even although extended chemoprophylaxis is recommended for high-VTE-risk cancer patients.1,2,8 Thus, many, if not most, surgeons subjectively discontinue VTE chemoprophylaxis when patients meet discharge criteria and leave the hospital. In order to comment on this practice, an additional aim of this study was to examine the rates of and risk factors for post-discharge VTE. The data from this analysis provide a rationale for recommending extended chemoprophylaxis at least to obese and elderly patients, and those who experience significant post-operative complications such as OSI. In addition, the analysis suggests that DP patients are more susceptible to post-discharge VTE. There are a number of explanations for this observation, but the most plausible is that pancreatic fistulae are often identified later in the post-operative period after DP compared with PD, and with the national trend of earlier discharge of DP patients, more of these patients may develop this important risk factor after leaving the hospital.22 Finally, malignant diagnoses are typically associated with more post-operative VTE compared with benign diseases, leading the ACCP to recommend extended chemoprophylaxis for 30 days after major abdominal cancer surgery in patients with a high VTE risk. However, our data showed no difference among surgical patients within 30 post-operative days, indicating that even patients undergoing a pancreatectomy for non-malignant diagnoses may benefit from extended VTE chemoprophylaxis.
A limitation of this study was that VTE chemoprophylaxis usage was not a recorded variable in NSQIP.15,23 An assumption can be made that many, but not all, patients received some amount of inpatient post-operative chemoprophylaxis. The timing of initiation would have been inconsistent. In contrast, it is likely that post-discharge chemoprophylaxis was uncommonly utilized, as it is not yet a current standard quality measure for hospitalized or surgical patients, in spite of the ACCP recommendations. NSQIP does not proactively screen asymptomatic patients for VTE, and thus only clinically apparent VTE are recorded. Another potential weakness is the limitation of the study period to 30 days after surgery, which does not capture the full post-operative risk of patients, especially cancer patients requiring adjuvant chemotherapy. This specific risk would argue for extended chemoprophylaxis for all cancer patients regardless of the aforementioned risk factors, as a VTE can derail planned adjuvant therapy. In spite of the weaknesses of NSQIP analyses, the size of the cohort provided an event rate that powered a multivariate analysis of VTE risk factors, which would be difficult to accomplish with a smaller institutional sample or dataset.19,20 As this is the first study to describe a subset of high-risk pancreatectomy patients who might benefit from chemoprophylaxis beyond discharge, further prospective studies are required to study the clinical effectiveness and cost effectiveness of extended chemoprophylaxis after a pancreatectomy.
In conclusion, within current national practice patterns, post-pancreatectomy VTE events outnumber major bleeding complications, which rarely occur. With the exception of patients with overt ongoing bleeding, the fear of early PPH should not prevent the timely administration of routine post-operative VTE for all pancreatectomy patients. One-third of post-pancreatectomy VTE occur post-discharge with clearly identifiable risk factors, potentially providing a rationale for selective use of post-discharge VTE chemoprophylaxis in high-risk patients.
Conflicts of interest
None declared.
References
- 1.Guyatt GH, Akl EA, Crowther M, Gutterman DD, Schuunemann HJ. Executive summary: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2):7S–47S. doi: 10.1378/chest.1412S3. Suppl. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM, Lassen MR, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition) Chest. 2008;133(6):381S–453S. doi: 10.1378/chest.08-0656. Suppl. [DOI] [PubMed] [Google Scholar]
- 3.Mismetti P, Laporte S, Darmon JY, Buchmuller A, Decousus H. Meta-analysis of low molecular weight heparin in the prevention of venous thromboembolism in general surgery. Br J Surg. 2001;88:913–930. doi: 10.1046/j.0007-1323.2001.01800.x. [DOI] [PubMed] [Google Scholar]
- 4.Collins R, Scrimgeour A, Yusuf S, Peto R. Reduction in fatal pulmonary embolism and venous thrombosis by perioperative administration of subcutaneous heparin. Overview of results of randomized trials in general, orthopedic, and urologic surgery. N Engl J Med. 1988;318:1162–1173. doi: 10.1056/NEJM198805053181805. [DOI] [PubMed] [Google Scholar]
- 5.Gould MK, Garcia DA, Wren SM, Karanicolas PJ, Arcelus JI, Heit JA, et al. Prevention of VTE in nonorthopedic surgical patients: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2 Suppl):e227S–e277S. doi: 10.1378/chest.11-2297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Leonardi MJ, McGory ML, Ko CY. A systematic review of deep venous thrombosis prophylaxis in cancer patients: implications for improving quality. Ann Surg Oncol. 2007;14:929–936. doi: 10.1245/s10434-006-9183-9. [DOI] [PubMed] [Google Scholar]
- 7.Blom JW, Vanderschoot JP, Oostindier MJ, Osanto S, van der Meer FJ, Rosendaal FR. Incidence of venous thrombosis in a large cohort of 66,329 cancer patients: results of a record linkage study. J Thromb Haemost. 2006;4:529–535. doi: 10.1111/j.1538-7836.2006.01804.x. [DOI] [PubMed] [Google Scholar]
- 8.Farge D, Debourdeau P, Beckers M, Baglin C, Bauersachs RM, Brenner B, et al. International clinical practice guidelines for the treatment and prophylaxis of venous thromboembolism in patients with cancer. J Thromb Haemost. 2013;11:56–70. doi: 10.1111/jth.12070. [DOI] [PubMed] [Google Scholar]
- 9.Agnelli G, Bolis G, Capussotti L, Scarpa RM, Tonelli F, Bonizzoni E, et al. A clinical outcome-based prospective study on venous thromboembolism after cancer surgery: the @RISTOS project. Ann Surg. 2006;243:89–95. doi: 10.1097/01.sla.0000193959.44677.48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lyman GH. Venous thromboembolism in the patient with cancer: focus on burden of disease and benefits of thromboprophylaxis. Cancer. 2011;117:1334–1349. doi: 10.1002/cncr.25714. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Horsted F, West J, Grainge MJ. Risk of venous thromboembolism in patients with cancer: a systematic review and meta-analysis. PLoS Med. 2012;9:e1001275. doi: 10.1371/journal.pmed.1001275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yhim HY, Jang MJ, Kwak JY, Yim CY, Choi WI, Lee YC, et al. The incidence, risk factors, and prognosis of recurrent venous thromboembolism (VTE) in patients with advanced solid cancers receiving anticoagulation therapy after the diagnosis of index VTE. Thromb Res. 2013;131:e133–e140. doi: 10.1016/j.thromres.2013.01.025. [DOI] [PubMed] [Google Scholar]
- 13.Rahbari NN, Garden OJ, Padbury R, Maddern G, Koch M, Hugh TJ, et al. Post-hepatectomy haemorrhage: a definition and grading by the International Study Group of Liver Surgery (ISGLS) HPB. 2011;13:528–535. doi: 10.1111/j.1477-2574.2011.00319.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Correa-Gallego C, Brennan MF, D'Angelica MI, DeMatteo RP, Fong Y, Kingham TP, et al. Contemporary experience with postpancreatectomy hemorrhage: results of 1,122 patients resected between 2006 and 2011. J Am Coll Surg. 2012;215:616–621. doi: 10.1016/j.jamcollsurg.2012.07.010. [DOI] [PubMed] [Google Scholar]
- 15.Tzeng CW, Katz MH, Fleming JB, Pisters PW, Lee JE, Abdalla EK, et al. Risk of venous thromboembolism outweighs post-hepatectomy bleeding complications: analysis of 5651 National Surgical Quality Improvement Program patients. HPB. 2012;14:506–513. doi: 10.1111/j.1477-2574.2012.00479.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Merkow RP, Bilimoria KY, McCarter MD, Cohen ME, Barnett CC, Raval MV, et al. Post-discharge venous thromboembolism after cancer surgery: extending the case for extended prophylaxis. Ann Surg. 2011;254:131–137. doi: 10.1097/SLA.0b013e31821b98da. [DOI] [PubMed] [Google Scholar]
- 17.Bergqvist D, Agnelli G, Cohen AT, Eldor A, Nilsson PE, Le Moigne-Amrani A, et al. Duration of prophylaxis against venous thromboembolism with enoxaparin after surgery for cancer. N Engl J Med. 2002;346:975–980. doi: 10.1056/NEJMoa012385. [DOI] [PubMed] [Google Scholar]
- 18.Rasmussen MS, Jorgensen LN, Wille-Jorgensen P, Nielsen JD, Horn A, Mohn AC, et al. Prolonged prophylaxis with dalteparin to prevent late thromboembolic complications in patients undergoing major abdominal surgery: a multicenter randomized open-label study. J Thromb Haemost. 2006;4:2384–2390. doi: 10.1111/j.1538-7836.2006.02153.x. [DOI] [PubMed] [Google Scholar]
- 19.Aloia TA, Fahy BN, Fischer CP, Jones SL, Duchini A, Galati J, et al. Predicting poor outcome following hepatectomy: analysis of 2313 hepatectomies in the NSQIP database. HPB. 2009;11:510–515. doi: 10.1111/j.1477-2574.2009.00095.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Strasberg SM, Hall BL. Postoperative morbidity index: a quantitative measure of severity of postoperative complications. J Am Coll Surg. 2011;213:616–626. doi: 10.1016/j.jamcollsurg.2011.07.019. [DOI] [PubMed] [Google Scholar]
- 21.Reinke CE, Drebin JA, Kreider S, Kean C, Resnick A, Raper S, et al. Timing of preoperative pharmacoprophylaxis for pancreatic surgery patients: a venous thromboembolism reduction initiative. Ann Surg Oncol. 2012;19:19–25. doi: 10.1245/s10434-011-1858-1. [DOI] [PubMed] [Google Scholar]
- 22.Baker MS, Bentrem DJ, Ujiki MB, Stocker S, Talamonti MS. Adding days spent in readmission to the initial postoperative length of stay limits the perceived benefit of laparoscopic distal pancreatectomy when compared with open distal pancreatectomy. Am J Surg. 2011;201:295–299. doi: 10.1016/j.amjsurg.2010.09.014. discussion 299-300. [DOI] [PubMed] [Google Scholar]
- 23.Pitt HA, Kilbane M, Strasberg SM, Pawlik TM, Dixon E, Zyromski NJ, et al. ACS-NSQIP has the potential to create an HPB-NSQIP option. HPB. 2009;11:405–413. doi: 10.1111/j.1477-2574.2009.00074.x. [DOI] [PMC free article] [PubMed] [Google Scholar]