Abstract
Background
Historically, liver surgeons have withheld venous thromboembolism (VTE) chemoprophylaxis due to perceived postoperative bleeding risk and theorized protective anticoagulation effects of a hepatectomy. The relationships between extent of hepatectomy, postoperative VTE and bleeding events were evaluated using the National Surgical Quality Improvement Program (NSQIP) database.
Methods
From 2005 to 2009, all elective open hepatectomies were identified. Factors associated with 30-day rates of VTE, postoperative transfusions and returns to the operating room (ROR), were analysed.
Results
The analysis included 5651 hepatectomies with 3376 (59.7%) partial, 585 (10.4%) left, 1134 (20.1%) right, and 556 (9.8%) extended. Complications included deep vein thrombosis (DVT) (1.93%), pulmonary embolism (PE) (1.31%), venous thromboembolism (VTE) (2.88%), postoperative transfusion (0.76%) and ROR with transfusion (0.44%). VTE increased with magnitude of hepatectomy (partial 2.13%, left 2.05%, right 4.15%, extended 5.76%; P < 0.001) and outnumbered bleeding events (P < 0.001). Other factors independently associated with VTE were aspartate aminotransferase (AST) ≥27 (P= 0.022), American Society of Anesthesiologists (ASA) class ≥3 (P < 0.001), operative time >222 min (P= 0.043), organ space infection (P < 0.001) and length of hospital stay ≥7 days (P= 0.004). VTE resulted in 30-day mortality of 7.4% vs. 2.3% with no VTE (P= 0.001).
Conclusions
Contrary to the belief that transient postoperative liver insufficiency is protective, VTE increases with extent of hepatectomy. VTE exceeds major bleeding events and is strongly associated with mortality. These data support routine post-hepatectomy VTE chemoprophylaxis.
Keywords: liver, hepatectomy, venous thomboembolism, deep venous thrombosis
Introduction
With regard to chemical venous thromboembolism (VTE) prophylaxis, gastrointestinal surgeons have always needed to balance the risk of peri-operative bleeding against the risk of VTE. For the majority of these operations, peri-operative VTE chemoprophylaxis has proven effective and safe, and is now considered the standard of care, especially in high-risk cancer patients.1–6 In spite of this experience, liver surgeons have historically withheld peri-operative VTE chemoprophylaxis mainly as a result of perceived peri-operative bleeding risk. Until the most recent decade of liver surgery, in which hepatectomy mortality has dropped to 2.5% nationally7 and 1% at top centres,8,9 liver resections were considered high-risk procedures10 with frequent peri-operative bleeding events that impacted both short- and long-term outcomes.11
Further discouraging the utilization of VTE chemoprophylaxis, liver surgeons have theorized that transient liver dysfunction after a hepatectomy (related to inflow occlusion time, extent of hepatectomy and liver remnant size) produced anticoagulation effects that protected patients from thrombotic events. Using this logic, the larger the magnitude of liver resection, the larger the bleeding risk and the greater the likelihood of liver insufficiency, resulting in a presumed lesser need for chemical VTE prophylaxis.
In addition to its associated high morbidity and mortality, post-operative VTE is now also considered a preventable nosocomial complication with both cost and reimbursement implications.5,12 In this current environment, defining the true relationships between extent of hepatectomy, postoperative VTE and bleeding events, could have a significant impact on practice patterns nationally. Given the variability in clinical practice regarding this issue, analysis of a large-scale multi-institutional dataset was required to remove small dataset biases. The recent paradigm of using national patient databases such as the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) to study morbidity and mortality among general surgery operations has allowed surgeons to transcend individual beliefs and customs, looking beyond the bias of single-institution databases to study issues on a broader, national level, which can more accurately reflect general trends and behaviours in quality improvement.13 Using this paradigm, this project was designed to determine the true relationships between extent of hepatectomy, clinically significant bleeding risk and VTE risk.
The hypothesis of the present study was that, in current practice, the true postoperative VTE rate after a liver resection was significantly greater than the rate of major postoperative bleeding events. A secondary aim was to define the relationship between extent of hepatectomy and VTE rates.
Methods
Data acquisition and patients
From the ACS-NSQIP participant use file (PUF) for 2005 to 2009, all hepatectomy procedures were identified. After excluding emergency cases and wedge resections [current procedural terminology (CPT) code 47100] a total of 5651 hepatectomies were included for analysis. The pre-operative NSQIP risk factors assessed included age, gender, race, weight/body mass index, albumin, haematocrit, platelets, blood urea nitrogen, creatinine, partial thrombin time, international normalized ratio (INR), alkaline phosphatase, aspartate aminotransferase (AST), bilirubin, American Society of Anesthesiologists (ASA) class, smoking, chronic obstructive pulmonary disease, diabetes, bleeding disorder, ascites, pre-operative chemotherapy, pre-operative radiation therapy, pre-operative transfusion and previous operation within 30 days. Intra-operative factors included intra-operative transfusion, operative time, concurrent operation and extent of hepatectomy.
For thrombotic events within 30 days of a hepatectomy, the analysis focused on three NSQIP postoperative occurrence variables: deep vein thrombosis (DVT), pulmonary embolism (PE) or the combination of the two. For postoperative haemorrhagic events, the analysis focused on two postoperative occurrence variables: postoperative transfusion within 72 h of surgery and postoperative transfusion within 72 h of surgery with an unplanned return to the operating room (ROR). Additional post-operative complications included in the analysis were surgical site infections, organ space infections (abscess/biloma), wound disruption (fascial dehiscence), length of stay and mortality.
The extent of hepatectomy was classified by the primary CPT code and included the following liver resections by order of increasing magnitude: partial (CPT 47120), left (47125), right (47130), and extended (47122) hepatectomies. VTE was defined as a clinically detected DVT or PE. Patients with both DVT and PE documented were labelled only once for VTE. Major postoperative bleeding was defined by the ACS-NSQIP variable ‘bleeding transfusion,’ which included any postoperative transfusion outside of the postoperative recovery room and excluding blood finishing from the operating room. ROR was defined as a re-operation within the same hospitalization. Patients were labelled with simultaneous ROR and bleeding transfusion codes with the complication of ‘ROR with transfusion,’ reflecting presumed major bleeding requiring operative intervention.
Statistical analysis
To identify clinical factors associated with VTE, postoperative bleeding/transfusions and ROR with bleeding/transfusions, univariate analyses were performed with chi-squared test or Fisher's exact test for categorical data and Mann–Whitney U-test for continuous data. Univariate factors with P < 0.15 were entered into Cox proportional hazard models to find independent associations with each major complication type. Statistical significance was defined as a two-sided P < 0.05. Multivariate modelling was reported with odds ratio (OR) and 95% confidence interval (CI). All analyses were performed using IBM SPSS Version 19 software (IBM, Armonk, NY, USA).
Results
Patient characteristics and distribution of hepatectomy types
From the cohort of 5651 patients, the median age was 60 years (range, 17–89) with 48.7% male patients and 76.1% Caucasian race. Regarding pre-operative risk factors, 66.0% were ASA class ≥3, with 16.7% smokers and 15.6% diabetics. Patients with known bleeding disorders accounted for 3.4% of all study patients. The case distribution included 3376 (59.7%) partial; 585 (10.4%) left; 1134 (20.1%) right; and 556 (9.8%) extended hepatectomies. Other clinically relevant pre-operative, intra-operative and postoperative factors are described in Table 1.
Table 1.
Factors associated with no VTE vs. VTE
| Clinical characteristic | All patients (n= 5651) | No VTE (DVT or PTE) | VTE (DVT or PTE) | P | |||
|---|---|---|---|---|---|---|---|
| n or median | % | n or median | % | n or median | % | ||
| n | 5651 | 100% | 5488 | 163 | |||
| Pre-operative factors | |||||||
| Median age (range), years | 60 | 60 | 62 | 0.040 | |||
| Gender, male | 2752 | 48.7% | 2648 | 48.3% | 104 | 63.8% | <0.001 |
| Race, white | 4299 | 76.1% | 4161 | 75.8% | 138 | 84.7% | 0.009 |
| Weight, kg | 77.7 | 77.7 | 81.7 | 0.059 | |||
| BMI | 27 | 27 | 27.2 | 0.696 | |||
| Albumin ≥4 | 2874 | 50.9% | 2804 | 51.1% | 70 | 42.9% | 0.047 |
| Alkaline phosphatise | 93 | 93 | 105 | 0.004 | |||
| AST | 27 | 27 | 31 | 0.002 | |||
| ASA class ≥3 | 3728 | 66.0% | 3604 | 65.7% | 124 | 76.1% | 0.006 |
| Ascites | 72 | 1.3% | 67 | 1.2% | 5 | 3.1% | 0.056 |
| Peri-operative | |||||||
| Operative time, min | 222 | 220 | 297 | <0.001 | |||
| Operative time >222 min | 2825 | 50.0% | 2698 | 49.2% | 127 | 77.9% | <0.001 |
| Any intra-operative transfusion | 1492 | 26.4% | 1413 | 25.7% | 79 | 48.5% | <0.001 |
| RBC ≥4 | 510 | 9.0% | 477 | 8.7% | 33 | 20.2% | <0.001 |
| Concurrent Operation | 223 | 3.9% | 211 | 3.8% | 12 | 7.4% | 0.037 |
| Extent of hepatectomy | <0.001 | ||||||
| Partial | 3376 | 59.7% | 3304 | 60.2% | 72 | 44.2% | |
| Left | 585 | 10.4% | 573 | 10.4% | 12 | 7.4% | |
| Right | 1134 | 20.1% | 1087 | 19.8% | 47 | 28.8% | |
| Extended | 556 | 9.8% | 524 | 9.5% | 32 | 19.6% | |
| Partial vs. left/right/extended | 3376 | 59.7% | 3304 | 60.2% | 72 | 44.2% | <0.001 |
| Right/extended vs. left/partial | <0.001 | ||||||
| Right/Extended | 1690 | 29.9% | 1611 | 29.4% | 79 | 48.5% | |
| Left/Partial | 3961 | 70.1% | 3877 | 70.6% | 84 | 51.5% | |
| Postoperative transfusion | 43 | 0.8% | 38 | 0.7% | 5 | 3.1% | 0.007 |
| Return to operating room (ROR) | 256 | 4.5% | 222 | 4.0% | 34 | 20.9% | <0.001 |
| ROR with postoperative transfusion | 25 | 0.4% | 22 | 0.4% | 3 | 1.8% | 0.034 |
| SSSI | 271 | 4.8% | 257 | 4.7% | 14 | 8.6% | 0.037 |
| DSSI | 54 | 1.0% | 49 | 0.9% | 5 | 3.1% | 0.019 |
| Organ space infection | 363 | 6.4% | 323 | 5.9% | 40 | 24.5% | <0.001 |
| Fascial dehiscence | 49 | 0.9% | 41 | 0.7% | 8 | 4.9% | 0.001 |
| Postoperative LOS | 6 | 6 | 11 | <0.001 | |||
| Postoperative LOS ≥7 | 2395 | 42.4% | 2260 | 41.2% | 135 | 82.8% | <0.001 |
| Death within 30 days | 137 | 2.4% | 125 | 2.3% | 12 | 7.4% | 0.001 |
BMI, body mass index; AST, aspartate aminotransferase; ASA, American Society of Anesthesiologists; RBC, (units) red blood cells; ROR, return to the operating room; SSSI, superficial surgical site infection; DSSI, deep surgical site infection; LOS, length of stay.
Not significant: white blood cells, platelets, blood urea nitrogen, creatinine, haematocrit, international normalized ratio, partial thrombin time, total bilirubin, alcohol use, pulmonary disease, smoking, diabetes, pre-operative radiation therapy, pre-operative transfusion, bleeding disorder, operation in preceding 30 days, high level resident involvement; pre-operative chemotherapy within 30 days; pre-operative weight loss ≥10%.
Thrombotic and haemorrhagic event rates in relation to extent of hepatectomy
Overall complication event rates included the following: DVT (1.93%), PE (1.31%), VTE (2.88%), postoperative transfusion (0.76%) and ROR with transfusion (0.44%). VTE event rates (2.1–5.8%) were more frequent than postoperative transfusion rates (0.6–1.3%) and/or ROR with transfusion rates (0.3–0.7%) across all extents of hepatectomies (3.5–8.3 times, P < 0.001). Analysis of the association between the magnitude of hepatectomy and VTE rates showed a proportional relationship with VTE rates for partial, left, right and extended hepatectomies, of 2.05%, 2.13%, 4.15%, and 5.76%, respectively (P < 0.001; Fig. 1). While intra-operative transfusions ≥1 unit packed red blood cells (PRBC) were relatively common (26.4%), massive intra-operative transfusions (≥4 PRBC) were less common (9.0%) and postoperative transfusions were rare (0.8%).
Figure 1.
National rates of venous thromboembolism (VTE) vs. bleeding transfusion vs. return to the operating room with bleeding transfusion. VTE rates far exceed rates of bleeding complications
Factors associated with postoperative bleeding requiring a transfusion
The rate of postoperative transfusion ranged from 0.6% for partial hepatectomies up to 1.3% for extended hepatectomies, with a 0.8% overall rate. The univariate comparison of factors between patients with and without major postoperative bleeding requiring a transfusion and coded as ‘bleeding transfusion’ is detailed in Table 2. Independent pre-operative factors associated with postoperative transfusion included the following: bleeding disorder [OR 16.95, 95% confidence interval (CI), 9.90–29.41, P < 0.001], ASA class ≥3 (OR 2.13, 95% CI, 1.24–3.64, P= 0.006), albumin <4 g/dl (OR 2.74, 95% CI, 1.33–5.65, P= 0.006), pre-operative haematocrit <39% (OR 2.56, 95% CI, 1.33–4.93, P= 0.005) postoperative deep organ space infection (OR 4.05, 95% CI, 2.36–6.94, P < 0.001). In the multivariate model, the variable major vs. partial hepatectomy trended towards statistical significance (P= 0.051).
Table 2.
Factors associated with postoperative bleeding and transfusion
| Clinical characteristic | All patients (n= 5651) | No postoperative tranfusion | Postoperative transfusion | P | |||
|---|---|---|---|---|---|---|---|
| n or median | % | n or median | % | n or median | % | ||
| n | 5651 | 100% | 5608 | 43 | |||
| Pre-operative factors | |||||||
| Median age (range), years | 60 | 60 | 59 | 0.964 | |||
| Gender, male | 2752 | 48.7% | 2730 | 48.7% | 22 | 51.2% | 0.762 |
| Race, white | 4299 | 76.1% | 4264 | 76.0% | 35 | 81.4% | 0.477 |
| Weight, kg | 77.7 | 77.7 | 78.2 | 0.949 | |||
| Albumin ≥4 g/dl | 2874 | 50.9% | 2863 | 51.1% | 11 | 25.6% | 0.001 |
| Haematocrit ≥39% | 2983 | 52.8% | 2970 | 53.0% | 13 | 30.2% | 0.003 |
| PTT ≥29 s | 1963 | 34.7% | 1940 | 34.6% | 23 | 53.5% | 0.015 |
| Alkaline phosphatase | 93 | 93 | 151 | <0.001 | |||
| AST | 27 | 27 | 46.5 | <0.001 | |||
| ASA class ≥3 | 3728 | 66.0% | 3691 | 65.8% | 37 | 86.0% | 0.005 |
| Ascites | 72 | 1.3% | 71 | 1.3% | 1 | 2.3% | 0.425 |
| Bleeding disorder | 191 | 3.4% | 186 | 3.3% | 5 | 11.6% | 0.014 |
| Peri-operative | |||||||
| Operative time, min | 222 | 222 | 350 | <0.001 | |||
| Operative time >222 min | 2825 | 50.0% | 2790 | 49.8% | 35 | 81.4% | <0.001 |
| Any intra-operative transfusion | 1492 | 26.4% | 1466 | 26.1% | 26 | 60.5% | <0.001 |
| RBC ≥4 | 510 | 9.0% | 488 | 8.7% | 22 | 51.2% | <0.001 |
| Concurrent operation | 223 | 3.9% | 218 | 3.9% | 5 | 11.6% | 0.026 |
| Extent of hepatectomy | 0.198 | ||||||
| Partial | 3376 | 59.7% | 3355 | 59.8% | 21 | 48.8% | |
| Left | 585 | 10.4% | 582 | 10.4% | 3 | 7.0% | |
| Right | 1134 | 20.1% | 1122 | 20.0% | 12 | 27.9% | |
| Extended | 556 | 9.8% | 549 | 9.8% | 7 | 16.3% | |
| Partial vs. left/right/extended | 3376 | 59.7% | 3355 | 59.8% | 21 | 48.8% | 0.161 |
| Right/extended vs. left/partial | 0.045 | ||||||
| Right/extended | 1690 | 29.9% | 1671 | 29.8% | 19 | 44.2% | |
| Left/partial | 3961 | 70.1% | 3937 | 70.2% | 24 | 55.8% | |
| DVT | 109 | 30.4% | 105 | 30.4% | 4 | 9.3% | 0.009 |
| PE | 74 | 29.9% | 72 | 29.8% | 2 | 4.7% | 0.108 |
| VTE | 163 | 69.6% | 158 | 69.6% | 5 | 11.6% | 0.007 |
| Return to OR | 256 | 4.5% | 231 | 4.1% | 25 | 58.1% | <0.001 |
| Organ Space Infection | 363 | 6.4% | 355 | 6.3% | 8 | 18.6% | 0.005 |
| Postoperative LOS | 6 | 6 | 13 | <0.001 | |||
| Postoperative LOS ≥7 | 2395 | 42.4% | 2361 | 42.1% | 34 | 79.1% | <0.001 |
| Death within 30 days | 137 | 2.4% | 130 | 2.3% | 7 | 16.3% | <0.001 |
PTT, partial thrombin time; AST aspartate aminotransferase; ASA, American Society of Anesthesiologists; RBC, (units) red blood cells; DVT, deep vein thrombosis; PE, pulmonary embolus; VTE, venous thromboembolism; OR, operating room; LOS, length of stay.
Not significant: white blood cells, platelets, blood urea nitrogen, creatinine, total bilirubin, alcohol use, pulmonary disease, smoking, diabetes, preoperative radiation therapy, preoperative transfusion, operation in preceding 30 days, high level resident involvement, surgical site infection, fascial dehiscence, preoperative chemotherapy within 30 days; preoperative weight loss ≥10%, body mass index, international normalized ratio.
Factors associated with return to operating room with transfusion
The combined rate of ROR with transfusion ranged from 0.3% after partial hepatectomies to 0.7% after right and extended hepatectomies, with a 0.4% overall rate. The univariate analysis of factors associated with patients with and without ROR with a transfusion is detailed in Table 3. Using multivariate analysis, independent pre-operative predictors of ROR with transfusion included the following: bleeding disorder (OR 15.15, 95% CI, 8.47–27.03, P < 0.001), albumin <4 g/dl (OR 3.19, 95% CI, 1.19–8.55, P= 0.021), pre-operative haematocrit < 39% (OR 2.99, 95% CI, 1.28–6.99, P= 0.012) and major vs. partial hepatectomy (OR 2.92, 95% CI, 1.39–6.10, P= 0.005).
Table 3.
Factors associated with return to operating room (ROR) with postoperative bleeding and transfusion
| Clinical characteristic | All patients (n= 5651) | No ROR with transfusion | ROR with transfusion | P | |||
|---|---|---|---|---|---|---|---|
| n or median | % | n or median | % | n or median | % | ||
| n | 5651 | 5626 | 25 | ||||
| Pre-operative factors | |||||||
| Median age (range), years | 60 | 60 | 59 | 0.943 | |||
| Gender, male | 2752 | 48.7% | 2737 | 48.6% | 15 | 60.0% | 0.176 |
| Race, white | 4299 | 76.1% | 4278 | 76.0% | 21 | 84.0% | 0.482 |
| Weight, kg | 77.7 | 77.7 | 74.1 | 0.140 | |||
| BMI | 27 | 27 | 25.1 | 0.103 | |||
| Albumin ≥4 g/dl | 2874 | 50.9% | 2869 | 51.0% | 5 | 20.0% | 0.002 |
| Median haematocrit | 39.3 | 39.3 | 34.5 | <0.001 | |||
| Haematocrit ≥39% | 2983 | 52.8% | 2976 | 52.9% | 7 | 28.0% | 0.015 |
| Median BUN | 14 | 14 | 17.5 | 0.101 | |||
| PTT ≥29 s | 1963 | 34.7% | 1950 | 34.7% | 13 | 52.0% | 0.090 |
| Alkaline phosphatase | 93 | 93 | 148.5 | <0.001 | |||
| AST | 27 | 27 | 55 | <0.001 | |||
| ASA class ≥3 | 3728 | 66.0% | 3706 | 65.9% | 22 | 88.0% | 0.019 |
| Ascites | 72 | 1.3% | 71 | 1.3% | 1 | 4.0% | 0.275 |
| Bleeding disorder | 191 | 3.4% | 187 | 3.3% | 4 | 16.0% | 0.009 |
| Peri-operative | |||||||
| Operative time, min | 222 | 222 | 332 | <0.001 | |||
| Operative time >222 min | 2825 | 50.0% | 2805 | 49.9% | 20 | 80.0% | 0.002 |
| Any intra-operative transfusion | 1492 | 26.4% | 1477 | 26.3% | 15 | 60.0% | <0.001 |
| RBC ≥4 | 510 | 9.0% | 498 | 8.9% | 12 | 48.0% | <0.001 |
| Concurrent Operation | 223 | 3.9% | 221 | 3.9% | 2 | 8.0% | 0.259 |
| Extent of hepatectomy | 0.139 | ||||||
| Partial | 3376 | 59.7% | 3366 | 59.8% | 10 | 40.0% | |
| Left | 585 | 10.4% | 582 | 10.3% | 3 | 12.0% | |
| Right | 1134 | 20.1% | 1126 | 20.0% | 8 | 32.0% | |
| Extended | 556 | 9.8% | 552 | 9.8% | 4 | 16.0% | |
| Partial vs. left/right/extended | 3376 | 59.7% | 3366 | 59.8% | 10 | 40.0% | 0.064 |
| Right/extended vs. left/partial | 0.077 | ||||||
| Right/extended | 1690 | 29.9% | 1678 | 29.8% | 12 | 48.0% | |
| Left/partial | 3961 | 70.1% | 3948 | 70.2% | 13 | 52.0% | |
| DVT | 109 | 30.4% | 106 | 1.9% | 3 | 12.0% | 0.012 |
| PE | 74 | 29.9% | 73 | 1.3% | 1 | 4.0% | 0.281 |
| VTE | 163 | 69.6% | 160 | 2.8% | 3 | 12.0% | 0.034 |
| Organ Space Infection | 363 | 6.4% | 359 | 6.4% | 4 | 16.0% | 0.073 |
| Postoperative LOS | 6 | 6 | 17 | <0.001 | |||
| Postoperative LOS ≥7 | 2395 | 42.4% | 2373 | 42.2% | 22 | 88.0% | <0.001 |
| Death within 30 days | 137 | 2.4% | 133 | 2.4% | 4 | 16.0% | 0.003 |
BMI, body mass index; BUN, blood urea nitrogen; PTT, partial thrombin time; AST aspartate aminotransferase; ASA, American Society of Anesthesiologists; RBC (units) red blood cells; DVT, deep vein thrombosis; PE, pulmonary embolus; VTE, venous thromboembolism; LOS, length of stay.
Not significant: white blood cells, platelets, creatinine, alcohol use, total bilirubin, pulmonary disease, smoking, diabetes, pre-operative radiation therapy, pre-operative transfusion, operation in preceding 30 days, high level resident involvement, surgical site infection; fascial dehiscence; pre-operative chemotherapy within 30 days, international normalized ratio; weight loss ≥ 10%.
Factors associated with VTE
The univariate analysis of peri-operative risk factors associated with postoperative VTE is detailed in Table 1. Using multivariate analysis, the following peri-operative factors were independently associated with VTE: major vs. partial hepatectomy (OR 2.58, CI, 1.93–3.47, P < 0.001), male gender (OR 2.92, 95% CI, 2.15–3.98, P < 0.001), preoperative AST ≥27 IU/L (OR 1.41, 95% CI, 1.05–1.89, P= 0.022), ASA class ≥3 (OR 1.90, 95% CI, 1.42–2.53, P < 0.001), operative time >222 min (OR 1.37, 95% CI, 1.01–1.86, P= 0.043), postoperative organ space infection (OR 10.64, 95% CI, 8.13–13.89, P < 0.001) and length of stay ≥7 days (OR 1.58, 95% CI, 1.16–2.16, P= 0.004). Patients with VTE experienced a 30-day mortality rate of 7.4% compared with 2.3% in non-VTE patients (P= 0.001).
Discussion
The aim of the present study was to examine the true postoperative event rates of VTE, major bleeding requiring postoperative transfusions, and ROR with transfusions. Based on clear definitions and the presence of trained nurse reviewers for data collection, the ACS-NSQIP database was ideally suited for this purpose. This analysis showed that the VTE rate far exceeds the event rate of postoperative bleeding complications requiring transfusion, with or without return to the operating room. The gradient between VTE event rate and bleeding complications ranged from 3.5 to 8.3 times and increased significantly with the magnitude of hepatectomy.
Published hepatectomy bleeding event rates range from 1% to 8%,14 with most series focusing their data on intra-operative bleeding and transfusion rates. By focusing this analysis on defining the true rates of major postoperative bleeding events for each extent of hepatectomy in a large, objective dataset, the relative incidence of VTE and bleeding events were determined. This analysis indicates that VTE is far more common than previously recognized and that the rate of clinically significant post-operative bleeding has probably decreased over the last decade, which is reflected in the improved postoperative mortality rate for hepatectomies.7,9
Accurate determination of these event rates is critical to the discussion of peri-operative VTE prophylaxis in liver surgery. Previous studies on surgical VTE prophylaxis have included guidelines for general abdominal surgery without defining specific strategies for liver resections.4,15 This lack of specificity has made it difficult to extrapolate results to liver resections, which are known to have higher bleeding and transfusion rates than most general surgery operations.13 Within this data vacuum, some surgeons have criticized these general surgery consensus recommendations for peri-operative and postoperative VTE chemoprophylaxis with unfractionated heparin or low-molecular-weight heparin.1,16,17
The second focus of this analysis was a determination of the relationship between the magnitude of hepatectomy and the post-operative VTE rate. In this portion of the analysis, the traditional concept that auto-anticoagulation from physiologic postoperative liver dysfunction protects post-hepatectomy patients from VTE was tested. Contrary to that perception, the VTE rate actually increased in proportion to extent of hepatectomy, disproving this long-held belief. These data suggest that patients undergoing larger volume resections, which were traditionally thought to be the most protected by auto-anticoagulation, are actually the most at risk for VTE. This relationship could be as a result of anatomic factors (caval manipulation), biological factors (greater tumour burden and/or post-hepatectomy prothrombotic state18) and/or clinical factors (decreased compliance with the use of postoperative VTE chemoprophylaxis) with larger volume resections. The clinical importance of VTE was manifested in its strong association with higher rates of prolonged length of stay and death within 30 days in this cohort.
The relatively low rate of major postoperative bleeding events combined with the proportional association between magnitude of hepatectomy and postoperative VTE make a very strong argument for the use of VTE chemoprophylaxis in all hepatectomy patients, regardless of magnitude of resection. The only exception to this recommendation would be patients with pre-operatively known bleeding disorders (including bleeding diatheses), as this group was disproportionately at risk for postoperative bleeding events in the present analysis. For these patients, confirmation of haemostasis would be advised prior to initiating VTE chemoprophylaxis.
One other interesting observation from this analysis is that the development of a postoperative organ space infection, which is likely a biloma in this population, was most strongly associated with postoperative VTE with an OR of 10.64. This covariate relationship between postoperative fluid collection, VTE and other major adverse outcomes is well explained by the frequent disability incurred by postoperative bilomas, including activation of inflammatory cascades, additional drainage procedures, infection, immobility and longer lengths of stay. This observation suggests that, in addition to routine use of VTE chemoprophylaxis, hyper-vigilance during the operation regarding control of biliary structures may be the most important indirect measure for VTE prevention.
The two questions examined in this study have direct bearing on unique areas of VTE chemoprophylaxis. The immediate postoperative bleeding event rate is relevant to the question of whether liver surgery patients should be given pre-operative VTE chemoprophylaxis, as well as the drug and dose choices in the immediate postoperative period. In contrast, the data regarding elevated VTE rates for patients with large volume liver resection has more impact on VTE chemoprophylaxis practice during the middle and later portions of the hospitalization, when the INR would be expected to have peaked and recovered.
The major potential limitation of the present study is a lack of complete data regarding the utilization of VTE prophylaxis in this population.19,20 Given the general fear of bleeding events, it is reasonable to assume that a larger proportion of minor hepatectomy NSQIP patients received VTE chemoprophylaxis in the immediate postoperative period, compared with patients treated with major hepatectomies. However, the exact magnitude of this difference is unknown. If this assumption is at least in part correct, then the results of the present study, which indicate that major hepatectomy patients are the most likely to benefit from VTE chemoprophylaxis, have the opportunity to significantly improve outcomes through a seminal change in practice. Unfortunately, without the specific data on drug choice, dosing and timing, no comments on the risks or benefits of either pre-operative dosing (in the operating room before incision) or post-discharge dosing could be made.16,17 These two areas remain open to further clarification and study.
Conclusions
In contrast to the common belief among liver surgeons that transient postoperative liver insufficiency and auto-anticoagulation protects against postoperative VTE, this analysis confirms that VTE risk actually increases with extent of resection. Post-operative VTE incidence far exceeds major bleeding events among all extents of hepatectomy and is highly associated with post-operative mortality. With the exception of patients with bleeding disorders and those with overt ongoing bleeding, these results support a recommendation for routine postoperative VTE chemoprophylaxis in post-hepatectomy patients, including those undergoing major resections requiring longer operative times.
Conflicts of interest
None declared.
References
- 1.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) Suppl.:381S–453S. doi: 10.1378/chest.08-0656. [DOI] [PubMed] [Google Scholar]
- 2.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]
- 3.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]
- 4.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]
- 5.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]
- 6.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]
- 7.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]
- 8.Andres A, Toso C, Moldovan B, Schiffer E, Rubbia-Brandt L, Terraz S, et al. Complications of elective liver resections in a center with low mortality: a simple score to predict morbidity. Arch Surg. 2011;146:1246–1252. doi: 10.1001/archsurg.2011.175. [DOI] [PubMed] [Google Scholar]
- 9.Jarnagin WR, Gonen M, Fong Y, DeMatteo RP, Ben-Porat L, Little S, et al. Improvement in perioperative outcome after hepatic resection: analysis of 1,803 consecutive cases over the past decade. Ann Surg. Oct. 2002;236:397–406. doi: 10.1097/01.SLA.0000029003.66466.B3. discussion 406–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Belghiti J, Hiramatsu K, Benoist S, Massault P, Sauvanet A, Farges O. Seven hundred forty-seven hepatectomies in the 1990s: an update to evaluate the actual risk of liver resection. J Am Coll Surg. 2000;191:38–46. doi: 10.1016/s1072-7515(00)00261-1. [DOI] [PubMed] [Google Scholar]
- 11.Ito H, Are C, Gonen M, D'Angelica M, Dematteo RP, Kemeny NE, et al. Effect of postoperative morbidity on long-term survival after hepatic resection for metastatic colorectal cancer. Ann Surg. 2008;247:994–1002. doi: 10.1097/SLA.0b013e31816c405f. [DOI] [PubMed] [Google Scholar]
- 12.Mukherjee D, Lidor AO, Chu KM, Gearhart SL, Haut ER, Chang DC. Postoperative venous thromboembolism rates vary significantly after different types of major abdominal operations. J Gastrointest Surg. 2008;12:2015–2022. doi: 10.1007/s11605-008-0600-1. [DOI] [PubMed] [Google Scholar]
- 13.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]
- 14.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]
- 15.Stanley A, Young A. Primary prevention of venous thromboembolism in medical and surgical oncology patients. Br J Cancer. 2010;102(Suppl. 1):S10–S16. doi: 10.1038/sj.bjc.6605600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.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]
- 17.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]
- 18.Cerutti E, Stratta C, Romagnoli R, Schellino MM, Skurzak S, Rizzetto M, et al. Thromboelastogram monitoring in the perioperative period of hepatectomy for adult living liver donation. Liver Transpl. 2004;10:289–294. doi: 10.1002/lt.20078. [DOI] [PubMed] [Google Scholar]
- 19.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]
- 20.Reddy SK, Turley RS, Barbas AS, Steel JL, Tsung A, Marsh JW, et al. Postoperative pharmacologic thromboprophylaxis after major hepatectomy: does peripheral venous thromboembolism prevention outweigh bleeding risks? J Gastrointest Surg. 2011;15:1602–1610. doi: 10.1007/s11605-011-1591-x. [DOI] [PubMed] [Google Scholar]