As the number of liver resections in the United States has increased, operations are more commonly performed on older patients with multiple comorbidities. The advent of effective chemotherapy and techniques such as portal vein embolization, have compounded the number of increasingly complex resections taking up to 75% of healthy livers. Four potentially devastating complications of liver resection include postoperative hemorrhage, venous thromboembolism, bile leak, and post-hepatectomy liver failure. The risk factors and management of these complications are herein explored, stressing the importance of identifying preoperative factors that can decrease the risk for these potentially fatal complications.
Key points
- •
As the number of liver resections in the United States has increased, operations are more commonly performed on older patients with multiple comorbidities.
- •
The advent of effective chemotherapy, as well as techniques such as portal vein embolization, have compounded the number of increasingly complex resections taking up to 75% of healthy livers. As a result, although the operations have become safer from a mortality standpoint, the morbidity from liver resections has not decreased.
- •
Four potentially devastating complications of liver resection include postoperative hemorrhage, venous thromboembolism, bile leak, and post-hepatectomy liver failure.
Liver resection remains the most effective curative treatment of primary liver malignancies, including cholangiocarcinoma and hepatocellular carcinoma (HCC), as well metastatic disease, such as colorectal liver metastases. The number of liver resections performed in the United States nearly doubled between 1988 and 2000, with more than 7000 liver resections performed between 1996 and 2000. As the incidence of these primary and metastatic cancers to the liver increase, the number of liver resections will grow.
Recent advances in patient selection and operative technique have substantially reduced the risk of mortality, from a historical high of 20% to current risk of 1% to 5%. Despite this, morbidity rates still range from 20% to 56%, depending on the patient, the extent of resection, the disease process, and the hospital and surgeon.
Most studies on complications after hepatectomy have been from single institutions. Jarnagin and colleagues refuted the 13% to 20% operative mortality published in 1977 by Foster and Berman when they published a retrospective review from more than 1800 patients undergoing hepatic resection at Memorial Sloan Kettering between 1991 and 2001. Operative mortality decreased from 4.0% to 1.3% over the course of the review, with a 45% overall complication rate. Benzoni and colleagues evaluated 134 patients undergoing liver resection for HCC and 153 patients undergoing liver metastasectomies in a single institution and found a 4.5% mortality rate and 47.7% morbidity rate. Significant increases in complications were found in major hepatectomies, extended hepatectomy, Pringle longer than 20 minutes, and blood transfusions greater than 600 mL. Additionally, Childs B and C classification and histopathologic grading were associated with increased complications in patients with HCC. Sadamori and colleagues evaluated major morbidity following liver resection for HCC and found significantly higher rates of bile leakage (12.8% overall) and organ/space surgical-site infections (8.6% overall) in patients undergoing repeat hepatectomy and prolonged surgery. In 2007, however, Virani and colleagues used the National Surgical Quality Improvement Program–Patient Safety in Surgery (NSQIP-PSS) initiative to look at 30-day morbidity and mortality after liver resection among 14 hospitals in the United States. Overall complications occurred in 22.6% of patients, with 5.2% requiring return to the operating room for complications. Of these, sepsis, wound infection, urinary tract infection, and organ space infection were the most common. Patients who had any complications remained at significantly higher risk of death. In light of this information, identifying patients who are more likely to experience post-hepatectomy complications is the first step toward preventing them.
The steady to increasing rates of complications following liver resections are multifactorial. In general, operations are being offered to patients at increasing age with significant comorbidities. The operations are becoming increasingly complex with more extended resections and more repeat hepatectomies. Additionally, with increasingly effective chemotherapy regimens for colorectal liver metastases, previously unresectable patients are converted to resectable, but with livers that are more subject to steatosis, steatohepatitis, and sinusoidal compromise after months of chemotherapy. As these patients are increasingly offered surgical resection, the rates of morbidity remain high.
When discussing outcomes following hepatectomy, it is imperative to define the extent of resection. Liver resections range from a small, nonanatomic wedges to trisegmentectomies in which more than 75% of the liver parenchyma is removed. In general, “major hepatectomy” is defined as resection of 3 or more liver segments as defined by Couinaud. Some studies demonstrate similar morbidity across the board, whereas others show significant differences according to the extent of resection. Zimmitti and colleagues analyzed postoperative complications based on increasingly complex liver resections. They found that, with the exception of biliary leak, the rates of complications did not increase as complexity of the operation increased. Li and colleagues also looked at patients undergoing minor or major hepatectomy versus ablation for HCC. Rates of major complication between minor and major hepatectomy and ablation were 21.3%, 35.1%, and 9.3% ( P <.01), respectively. Overall complications were 26.9%, 41.0%, and 11.5% ( P <.01), respectively. As compared with minor hepatectomy, major hepatectomy was associated with higher rates of infectious (organ/space, superficial skin infections, pneumonia, sepsis, septic shock), pulmonary (unplanned reintubation, prolonged ventilator support), renal (progressive renal insufficiency, acute renal failure), and hematologic (bleeding within 72 hours requiring transfusions, deep venous thrombosis [DVT]) complications.
The importance of the pathology of the underlying liver is also paramount when discussing complications. In HCC, at least 80% of patients have hepatic fibrosis or frank cirrhosis. As such, the remnant liver is already damaged and perhaps more susceptible to additional insult. Whether to resect or transplant these livers remains controversial. In fact, some studies show a significantly improved overall survival rate (65.7% vs 43.8%, P = .005) and recurrence-free survival (85.3% vs 22.7%, P <.001) for patients who undergo resection within Milan Criteria, especially for patients with underlying chronic hepatitis C infection. Other studies maintain that surgical resection of solitary HCC in patients with preserved liver function remains the preferable approach. Nevertheless, when discussing complications after liver resection, it is important to note that some livers, including those with cirrhosis, fibrosis, and/or steatosis, may be more susceptible to baseline injury than an otherwise healthy liver.
Patients who undergo hepatic resection are subject to the “routine” complications, such as wound infections, sepsis, pneumonia, and other morbidity commensurate with any operation. For purposes of this article, we focus on complications particular to hepatic resection, including post-hepatectomy bleeding, venous thromboembolism (VTE), bile leak, and liver failure.
Post-hepatectomy hemorrhage
In 1977, Foster and Berman reported operative mortality for major liver resection to be 20%, with 20% of these deaths resulting from hemorrhage. Over time, however, with improved imaging, meticulous operative technique, and advances in perioperative management, the need for transfusions has decreased. Given the possible association of postoperative transfusions with adverse outcomes, this is an important focus to decrease morbidity. Recent studies document the incidence of postoperative hemorrhage from 0.6% to 8.0%.
A challenge remains to define precisely what postoperative bleeding means. Definitions range from requiring a certain number of packed red blood cells (PRBCs) to bleeding requiring reexploration to “hemorrhage from the operative site” to bleeding via a drain. In 2011, the International Study Group of Liver Surgery (ISGLS) set forth guidelines for the definition of post-hepatectomy hemorrhage. The consensus definition was a drop in hemoglobin greater than 3 g/dL postoperatively compared with postoperative baseline (immediately after surgery) and/or any postoperative transfusion of PRBCs for a falling hemoglobin and/or the need for invasive reintervention (embolization or relaparotomy) to stop the bleeding. They further categorize the definition by grade based on requirements for less than 2 units of PRBCs (grade A), more than 2 units of PRBCs (grade B), or the need for interventional radiologic intervention or reoperation (grade C) ( Box 1 ). This definition was validated in an 835-patient sample that correlated well with in-hospital mortality for grades A, B, and C of 0%, 17%, and 50%, respectively. Although it is important to have this definition now, previous studies did not have a uniform definition.
ISGLS definition of post-hepatectomy hemorrhage:
- 1.
Drop in hemoglobin >3 g/dL after establishing a postoperative baseline
- 2.
Any postoperative transfusion for a falling hemoglobin and/or the need for any reintervention (embolization or relaparotomy) to stop bleeding
- 3.
Evidence of bleeding, such as blood loss via drains or active hemorrhage by imaging.
- 1.
Grade A: Bleeding requiring transfusion of up to 2 units PRBCs
Grade B: Bleeding requiring transfusion of >2 units PRBCs but without invasive intervention
Grade C: Bleeding requiring interventional treatment or relaparotomy
Excluded: Patients requiring immediate postoperative transfusion secondary to intraoperative blood loss
Abbreviations: ISGLS, International Study Group of Liver Surgery; PRBCs, packed red blood cells.
Risk Factors
Although it is difficult to identify definitive risk factors for postoperative hemorrhage, Lim and colleagues suggest that cirrhotic livers bleed more; they also report that the outflow system may be affected by central venous pressure, which changes after extubation. Yang and colleagues recently analyzed risk factors for postoperative mortality after relaparotomy for hemorrhage. Patients who underwent late relaparotomy (>6 hours) had a 25% mortality, which was significantly higher than those undergoing early relaparotomy (8.6% mortality; P = .001). Independent risk factors contributing to increased hospital mortality included early time period (1997–2004), cirrhosis, ineffective hemostasis secondary to coagulopathy, late relaparotomy, postoperative liver failure, and postoperative renal failure requiring dialysis (all P <.05).
Management
Most cases of postoperative hemorrhage occur within the first 48 hours after an operation from the liver surface or the diaphragm. Unfortunately, preexisting drains or interventional radiology–placed drains are seldom useful, owing to the propensity for a clot to form within them. Close hemodynamic monitoring for hypotension and tachycardia, correction of coagulopathy, and blood transfusions are recommended. The indications for relaparotomy are often based on multiple factors. Reasonable considerations for relaparotomy include blood loss exceeding 1 L or the ongoing need for transfusion, hemoglobin decreases by 3 to 4 points, necessitating transfusion, or hemodynamic instability requiring transfusion.
Of patients who experience post-hepatectomy hemorrhage, 1% to 8% of these will require repeat laparotomy, which carries a mortality of 17% to 83%. The potential for morbidity and subsequent mortality associated with postoperative hemorrhage is high.
Post-hepatectomy hemorrhage
In 1977, Foster and Berman reported operative mortality for major liver resection to be 20%, with 20% of these deaths resulting from hemorrhage. Over time, however, with improved imaging, meticulous operative technique, and advances in perioperative management, the need for transfusions has decreased. Given the possible association of postoperative transfusions with adverse outcomes, this is an important focus to decrease morbidity. Recent studies document the incidence of postoperative hemorrhage from 0.6% to 8.0%.
A challenge remains to define precisely what postoperative bleeding means. Definitions range from requiring a certain number of packed red blood cells (PRBCs) to bleeding requiring reexploration to “hemorrhage from the operative site” to bleeding via a drain. In 2011, the International Study Group of Liver Surgery (ISGLS) set forth guidelines for the definition of post-hepatectomy hemorrhage. The consensus definition was a drop in hemoglobin greater than 3 g/dL postoperatively compared with postoperative baseline (immediately after surgery) and/or any postoperative transfusion of PRBCs for a falling hemoglobin and/or the need for invasive reintervention (embolization or relaparotomy) to stop the bleeding. They further categorize the definition by grade based on requirements for less than 2 units of PRBCs (grade A), more than 2 units of PRBCs (grade B), or the need for interventional radiologic intervention or reoperation (grade C) ( Box 1 ). This definition was validated in an 835-patient sample that correlated well with in-hospital mortality for grades A, B, and C of 0%, 17%, and 50%, respectively. Although it is important to have this definition now, previous studies did not have a uniform definition.
ISGLS definition of post-hepatectomy hemorrhage:
- 1.
Drop in hemoglobin >3 g/dL after establishing a postoperative baseline
- 2.
Any postoperative transfusion for a falling hemoglobin and/or the need for any reintervention (embolization or relaparotomy) to stop bleeding
- 3.
Evidence of bleeding, such as blood loss via drains or active hemorrhage by imaging.
- 1.
Grade A: Bleeding requiring transfusion of up to 2 units PRBCs
Grade B: Bleeding requiring transfusion of >2 units PRBCs but without invasive intervention
Grade C: Bleeding requiring interventional treatment or relaparotomy
Excluded: Patients requiring immediate postoperative transfusion secondary to intraoperative blood loss
Abbreviations: ISGLS, International Study Group of Liver Surgery; PRBCs, packed red blood cells.
Risk Factors
Although it is difficult to identify definitive risk factors for postoperative hemorrhage, Lim and colleagues suggest that cirrhotic livers bleed more; they also report that the outflow system may be affected by central venous pressure, which changes after extubation. Yang and colleagues recently analyzed risk factors for postoperative mortality after relaparotomy for hemorrhage. Patients who underwent late relaparotomy (>6 hours) had a 25% mortality, which was significantly higher than those undergoing early relaparotomy (8.6% mortality; P = .001). Independent risk factors contributing to increased hospital mortality included early time period (1997–2004), cirrhosis, ineffective hemostasis secondary to coagulopathy, late relaparotomy, postoperative liver failure, and postoperative renal failure requiring dialysis (all P <.05).
Management
Most cases of postoperative hemorrhage occur within the first 48 hours after an operation from the liver surface or the diaphragm. Unfortunately, preexisting drains or interventional radiology–placed drains are seldom useful, owing to the propensity for a clot to form within them. Close hemodynamic monitoring for hypotension and tachycardia, correction of coagulopathy, and blood transfusions are recommended. The indications for relaparotomy are often based on multiple factors. Reasonable considerations for relaparotomy include blood loss exceeding 1 L or the ongoing need for transfusion, hemoglobin decreases by 3 to 4 points, necessitating transfusion, or hemodynamic instability requiring transfusion.
Of patients who experience post-hepatectomy hemorrhage, 1% to 8% of these will require repeat laparotomy, which carries a mortality of 17% to 83%. The potential for morbidity and subsequent mortality associated with postoperative hemorrhage is high.
Post-hepatectomy venous thrombosis/thromboembolism
Studies have shown the patients at greatest risk of VTE are those undergoing abdominal or pelvic surgery for cancer. The consequences of VTE cannot be overstated. Mortality from VTE has been estimated at 8 years to be between 12% and 50%. A recent population-based study demonstrated a 30-day mortality for DVT of 3% and 31% for pulmonary embolism (PE) versus 0.4% for the comparison group without VTE. Although the most dramatic increase in mortality was in the first year, the overall 30-year mortality rate ratios also were significant, with a rate of 1.55 (95% confidence interval [CI] 1.53–1.57) for DVT and 2.77 (95% CI 2.74–2.81) for PE.
Despite this potentially deadly complication, many surgeons have been reluctant to use pharmacologic prophylaxis in the setting of liver resection, citing risk of postoperative bleeding, impaired postoperative liver function, and “auto-anticoagulation.” Recent publications, however, have demonstrated that the rate of pulmonary embolus after liver resection is 6% and the rate of VTE approaches that of other abdominal operations for malignancy. For this reason, the paradigm of pharmacologic prophylaxis may be shifting.
One major challenge in comparing across studies is the lack of standard postoperative prophylaxis. Although some studies mention the exact pharmacologic regimens used, most of them have left this up to surgeon discretion, making the precise contribution of prophylaxis unclear. Additionally, early ambulation and sequential compression devices are difficult to track in the postoperative patient, contributing to additional possible inconsistencies across studies.
Risk Factors
There are no prospective, randomized, controlled trials evaluating VTE risk in patients undergoing hepatectomy. Multiple retrospective studies and 1 prospective study have been performed to establish the incidence and risk factors of VTE ( Table 1 ). Factors such as higher body mass index, longer operative times, and major liver resection are cited in multiple studies to put patients at risk for VTE. Other potential risk factors, including previous DVT, postoperative complications, and longer length of stay also could contribute to the VTE.
| Study | Rate | Risk Factors |
|---|---|---|
| Ejaz et al (variable pharmacologic prophylaxis) | 4.7% VTE 1.8% PE 3.3% DVT | History of VTE Prolonged operative time Increased length of stay |
| Nathan et al (variable pharmacologic prophylaxis) | 2.6% VTE 1.7% PE 1.1% DVT | Advanced age Higher BMI Longer procedure time Major complication Higher postoperative INR |
| Tzeng et al (no data on prophylaxis) | 2.9% VTE 1.3% PE 1.9% DVT | Major hepatectomy Male gender Preoperative AST >27 IU/L ASA class ≥3 OR time >222 min Postoperative organ space infection Length of stay ≥7 d |
| Reddy et al (variable pharmacologic prophylaxis) | 3.6% VTE 2.9% PE 0.7% DVT | Pharmacologic thromboprophylaxis |
| Melloul et al (pharmacologic and mechanical prophylaxis) | 6% PE | BMI >25 kg/m 2 Major liver resection Normal or minimally fibrotic liver parenchyma |
| Morris-Stiff et al (prophylaxis not clear) | 2.1% VTE 1.3% PE 0.7% DVT | History of DVT or PE |
Interestingly, the rationale of “auto-anticoagulation” with elevated international normalized ratio (INR) was shown by Nathan and colleagues to actually put patients at increased risk for VTE. In their discussion, it is emphasized that INR gives an “incomplete view of a patient’s coagulation profile,” as previously reported by Chitlur. It is further cautioned that an elevated INR is more indicative of the amount of liver resected than as a protective factor against VTE, which was also confirmed by Tzeng and colleagues from M.D. Anderson.
Management
In the case of VTE, prevention may be the best management, although the evidence-based literature behind this is mostly extrapolated from general surgery literature. Reddy and colleagues left the administration, timing, and type of pharmacologic prophylaxis to the discretion of the surgeon, whereas mechanical prophylaxis and early ambulation were standard across patients. The only significant variable associated with VTE was that patients treated with pharmacologic prophylaxis had lower rates of VTE when compared with those not receiving prophylaxis (6.3% vs 2.2%, P = .03). Nathan and colleagues examined the timing of prophylaxis and found no difference (immediate, early, late/never) between low molecular weight heparin and unfractionated heparin as a predictor for VTE. Similarly, Ejaz and colleagues demonstrated that even the “current best practice prophylaxis for VTE” does not necessarily prevent VTE, with 70.4% of patients who developed VTE having received postoperative VTE prophylaxis within 24 hours. Although chemoprophylaxis decreases the incidence of VTE by approximately 75% in general surgery patients, it is an accepted fact that, even with adherence to chemoprophylaxis, there will still be patients who will have VTEs.
Although a major concern for administration of prophylaxis is postoperative bleeding, this had not been collaborated in current studies. Tzeng and colleagues examined NSQIP data and found that the incidence of VTE far exceeds the risk of major bleeding events, thereby supporting the recommendation for chemoprophylaxis in patients without bleeding disorders or overt postoperative bleeding. Likewise, Reddy and colleagues found lower rates of postoperative red blood cell transfusion (16.7% vs 26.4%, P = .02) and similar rates of overall transfusions (35.0% vs 30.6%, P = .36) in patients receiving VTE prophylaxis. These 2 studies support the use of VTE prophylaxis with data refuting increased bleeding complications.
Although the role for VTE prophylaxis in liver surgery has not undergone rigorous testing with randomized studies, the most recent studies indicate that this is not an uncommon, but potentially deadly, event. These studies have, likewise, refuted the long-held belief that prophylaxis will increase bleeding complication. It therefore seems prudent to at least consider routine chemoprophylaxis in patients undergoing major hepatectomy, at least until a more effective method of prevention is discovered.
Post-hepatectomy portal vein and hepatic artery thrombosis
Vascular complications, such as thrombosis of the portal vein or hepatic artery, are rarely reported complications after hepatectomy. The vast majority of these events are best described in the transplantation literature. As more aggressive liver resections are undertaken, such as those with hepatic artery and portal vein reconstructions, however, the risk factors and frequency of these complications may become more relevant.
Hepatic artery thrombosis (HAT) is a rare complication and generally reported in the literature only when associated with arterial reconstruction in liver resection or in transplantation. Even with hepatic artery resection, Azoulay and colleagues recently reported a patency of 100% in the 5 patients in their series undergoing hepatic artery reconstruction for cholangiocarcinoma. The transplantation literature, on the other hand, reports a 3% to 9% risk of HAT that results in acute graft loss. This complication may present acutely with graft failure, sepsis, or abscess. In a delayed fashion, it may present as cholangitis, bile leak, or altered liver function tests.
Again, most of the literature regarding postoperative portal vein thrombosis (PVT) is taken from the transplantation literature, where there is a 2% to 6% risk of thrombosis. PVT after hepatectomy may go unrecognized because of the lack of specific symptoms. Patients may present with abdominal pain if it involves the superior mesenteric vessels and produces bowel congestion or ischemia. Alternatively, patients can present with nausea, vomiting, anorexia, weight loss, diarrhea, or increased abdominal distention secondary to ascites. If acute thrombosis is unrecognized, collateral vessels will develop and the patient will progress to cavernous transformation of the portal vein and portal hypertension, which may manifest as varices, splenomegaly, and hemorrhage.
Risk Factors
A recent review and meta-analysis of vascular resection in the treatment of hilar cholangiocarcinoma not surprisingly found that vascular complications, including PVT or HAT, stenosis, or pseudoaneurysms were more common in patients undergoing vascular resections (odds ratio [OR] 8.8, 95% CI 3.5–22; P <.0001). Additionally, there was a significantly higher mortality in patients undergoing vascular resection (OR 2.07, 95% CI 1.21–3.57; P = .008), but an even higher mortality rate in patients undergoing hepatic artery resection (4.48, 95% CI 1.97–10.16; P = .0003). Extrapolating from the transplantation literature, relevant risk factors for hepatic artery complications include small artery diameter, older donor age (>60), prolonged ischemia time, blood transfusions, prolonged operative time, bile leak, and cholangitis. Other studies have demonstrated that extra-anatomic anastomosis was the only multivariate factor independently associated with hepatic artery complications in the living donor transplant recipient.
HAT in the transplantation literature usually is discovered on serial Doppler ultrasounds, which are an integral component of postoperative care for the patient after transplantation; however, computed tomography (CT) scans also may provide information on HAT. As there are no large series of hepatic artery resection in liver resection, it is difficult to definitively identify relevant risk factors in this population of patients.
Like HAT, PVT can be identified by Doppler ultrasound, CT, or MRI. Color Doppler is especially helpful to confirm the diagnosis of PVT and cavernous transformation of the portal vein and is superior to CT for diagnostic purposes, but this modality is extremely user dependent. Magnetic resonance angiography can provide dynamic images to look at flow and anatomy, but is time-consuming and expensive. Finally, portal venography is not only diagnostic, but can also be therapeutic; obviously, however, this is an invasive procedure with the potential for complications.
Management
The severity of HAT can dictate the management. In patients with a weak signal, but normal liver function, initial management can be conservative, including heparin and volume administration. Interventional radiology–based procedures, such as angioplasty, can be considered; however, urgent surgical thrombectomy may be required to reestablish flow.
PVT can be managed in various ways, as reviewed by Thomas and Ahmad in 2010. Anticoagulation with intravenous heparin followed by long-term warfarin has reduced the risk of thrombotic events and promoted the recanalization of the vein. There are some data that this should be instituted as quickly as possible for the highest likelihood of recanalization. Again, looking toward the transplantation literature for guidance, there are reports of successful treatment with thrombolysis, angioplasty, and stent repair. Thrombolytic therapy yields excellent results, especially with early intervention, and may provide superior results to systemic anticoagulation. Finally, open surgical thrombectomy has been used to reestablish flow in the portal system. This intervention is usually reserved for patients suffering from ischemic bowel secondary to porto-mesenteric vein thrombosis. The advantages of this approach are that it allows for immediate resolution of thrombosis, as well as operative revision of any previous vascular anastomosis that may have contributed to the thrombosis. It obviously carries with it, however, the associated risks of a major operation.
Bile leak
Bile leak remains a considerable complication after liver resection. The incidence is a substantial cause of associated morbidity, ranging from 2.6% to 33.0%. Zimmitti and colleagues recently examined liver resections at M.D. Anderson Cancer Center and found a 9.8% rate of liver-related complications, including 4.8% rate of bile leak. Although intraoperative blood loss and rates of transfusion fell over time, there was a rising incidence of bile leak from 3.7% in the early period of the study to 5.9% in the later time period. Despite decreasing mortality and improvements in other complication rates, biliary complications remain a significant concern.
The term “bile leak” has many definitions, from drainage of bile from the abdominal wound or drain, intra-abdominal collection identified at drainage or reoperation, and cholangiographic evidence of biliary leakage or stricture. A definition for bile leak was standardized by the ISGLS as a bilirubin level in a drain 3 times the serum concentration on or after pod 3 or the need for radiologic or operative intervention from a biliary collection or bile peritonitis.
Bile leaks may result in prolonged hospitalization, need for additional imaging, and increased interventions. Major bile leaks are associated with intra-abdominal sepsis, with a morality as high as 40% to 50%. Additionally, bile leak can increase hospitalization of 8 days versus 12 days ( P <.001).
Risk Factors
Several studies have explored risk factors for bile leak ( Table 2 ). Lo and colleagues found advanced age, preoperative leukocytosis, left-sided hepatectomy, and prolonged operating times as significant risk factors. Nagano and colleagues demonstrated advanced age, a large incisional surface area, and high-risk operations to be risk factors for leak. Benzoni and colleagues found additional factors of Pringle greater than 20 minutes, transfusion greater than 600 mL, pleural effusion, and extended hepatectomy to be associated with postoperative bile leak. Advanced age may or may not be associated with increased risk depending on the study.
| Study Author | % Leak | Significant Factors | |
|---|---|---|---|
| Lo et al | 8.1 | Age Non-HCC Mean hemoglobin Estimated blood loss Platelets Noncirrhotic livers | Childs B and C Major hepatectomy Caudate resection T-tube drainage Concomitant bilioenteric or bowel anastomosis Mean operative time |
| Left hepatectomy Prolonged operative time | ||
| Sadamori et al (all HCC) | 12.9 | Trisegmentectomy Repeat hepatectomy | OR time >300 min Blood transfusion |
| MVA OR time >300 min | |||
| Zimmitti et al | 4.8 | Preoperative jaundice Portal vein embolization Biliary tumors Repeat hepatectomy Two-stage resection Extended resection Caudate resection En bloc diaphragm resection | Bile duct resection/reconstruction Liver-associated procedures OR time >180 min EBL >1000 mL Tumor diameter >30 mm Portal node dissection Intraoperative transfusion |
| MVA Repeat hepatectomy Bile duct resection Intraoperative transfusion | En bloc diaphragm resection Extended hepatectomy | ||
| Benzoni et al | 6 | Major hepatectomy Left hepatectomy Trisegmentectomy Bisegmentectomy/left lobectomy | Segmentectomy/wedge Pringle >20 min Transfusion >600 mL Abscess Liver dysfunction Pleural effusion |
| Sadamori et al | 12.8 | Repeat hepatectomy OR time >300 min EBL >2000 mL | Blood transfusion Duration of parenchymal transection |
| MVA Repeat hepatectomy OR time >300 min | |||
| Nagano et al | 5.4 | Age Cut surface area High-risk operation | |
| Okumura et al | 6.5 | Fibrosis or cirrhosis OR time >5 h Major hepatic resection | Hepatectomy including Couinaud segment 4 or segment 5 |
| MVA OR time Resection of segment 4 | |||
Even when looking at risk factors for bile leak in patients with HCC, there are many of the same risk factors, including cirrhosis, major resection, and operative time, and leak rate does not appear to significantly differ from non-HCC cases ( Table 3 ). Sadamori and colleagues evaluated an HCC resection population and found univariate factors of repeat hepatectomy, operating room time of at least 300 minutes, blood loss greater than 2 L, blood transfusion, and duration of liver transection as significant factors in postoperative bile leak; repeat hepatectomy and duration of operation greater than 300 minutes persisted on multivariate analysis. Although there are numerous studies examining risk factors for bile leak, the lack of standard definition and the variation in study design make comparison difficult.
| Study Author | Pathology (number) | Leak Rate Number (%) |
|---|---|---|
| Tanaka et al, 2002 | HCC (316) | 23 (7.3%) |
| CCC (9) | 3 (33.3%) | |
| Metastatic (33) | 0 (0%) | |
| Other (5) | 0 (0%) | |
| Nagano et al, 2003 | HCC (126) | 9 (7.1%) |
| Metastatic (187) | 17 (5.4%) | |
| Sadamori et al, 2013 | HCC (359) | 46 (12.8%) |
Related posts:
Hepatocellular Cancer, Cholangiocarcinoma, and Metastatic Tumors of the Liver
Hepatocellular Cancer, Cholangiocarcinoma, and Metastatic Tumors of the Liver
Modern Technical Approaches in Resectional Hepatic Surgery
Transarterial Chemoembolization for Primary Liver Malignancies and Colorectal Liver Metastasis
Y90 Selective Internal Radiation Therapy
Imaging Approach to Hepatocellular Carcinoma, Cholangiocarcinoma, and Metastatic Colorectal Cancer
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
