Laparoscopic liver surgery: Shifting the management of liver tumors


  • Potential conflict of interest: Nothing to report.


Laparoscopic liver surgery has evolved rapidly over the past 5 years in a select number of centers. The growing experience with these procedures has resulted in a shift in the diagnostic and therapeutic approach to common liver tumors. The fact that resection of benign and malignant hepatic masses can now be accomplished laparoscopically with relatively low morbidity has influenced the decision-making process for physicians involved in the diagnosis and management of these lesions. For example, should a gastroenterologist or hepatologist seeing a 32-year-old woman with an asymptomatic 4 cm hepatic lesion that is radiologically indeterminate for adenoma or focal nodular hyperplasia (FNH): (1) continue to observe with annual computed tomography/magnetic resonance imaging (CT/MRI) scans, (2) subject the patient to a liver biopsy, or (3) refer for laparoscopic resection? For a solitary malignant liver tumor in the left lateral segment, should laparoscopic resection be considered the new standard of care, assuming the surgeon can perform the operation safely? We present current data and representative case studies on the use of laparoscopic liver resection at 2 major medical centers in the United States. We propose that surgical engagement defined by the managing physician's decision to proceed with a surgical intervention is increasingly affected by the availability of, and experience with, laparoscopic liver resection. (HEPATOLOGY 2006;44:1694–1700.)

Historical Perspective

The first application of laparoscopic surgery to the liver consisted of wedge liver biopsies in the staging of lymphoma,1 followed by various case reports and small series of laparoscopic resection of peripheral lesions, mostly benign.2, 3 More recently, reports of anatomic left lobectomy4 and right lobectomy5 have energized the field. The availability of laparoscopic and minimally invasive liver surgery has the potential to change the diagnostic and treatment algorithms currently followed in the management of liver tumors, both benign and malignant.6, 7 Although the technical aspects of this approach are evolving rapidly, there continues to be a need for better devices, especially for laparoscopic transection of the hepatic parenchyma. The application of robotic techniques to liver surgery will require the development of these devices. Nonetheless, several surgical teams have moved forward in implementing laparoscopic liver programs as reflected by several recent publications.8–32


ERCP, endoscopic retrograde cholangiopancreatography; FNH, focal nodular hyperplasia; HCC, hepatocellular carcinoma.

Technical Aspects

Once the decision is made to proceed laparoscopically, the procedure begins with positioning of the patient. For anterior and left lobe lesions, the patient is placed supine. For posterior lesions of the right lobe, the patient is placed in the left lateral decubitus position allowing access to the midline. A 12-mm umbilical port is used for the 30 degree laparoscope. Additional working ports are placed strategically to optimize manipulation and mobilization of the liver. We use 5-mm, 10-mm, or 12-mm ports as needed, typically in the epigastric and subcostal regions. These port sites are placed in anticipation of open conversion so that the port sites are in line with standard surgical incisions if needed.

The liver resection proceeds with laparoscopic mobilization, vascular control, parenchymal transection, cut surface inspection, and specimen extraction. Use of a hand-port for hand-assisted laparoscopic liver resection is contemplated intraoperatively in cases of technical difficulties or in the presence or anticipation of hemorrhage primarily from the parenchyma. While the hand-port can shorten the operative time, it tends to interfere with the laparoscopic trocar and instrument manipulation. There is a greater preference for hand-port use at UPMC with 50% of cases using the hand-port, while conversion to a hand-assisted procedure occurred only in 6% of the cases in the Northwestern experience. There were no conversions to open procedure in the Northwestern series and 4 cases (2.9%) required conversion to an open procedure in the UPMC series. For these cases, the hand-port incision is extended to an upper midline incision which still avoided the subcostal division of the abdominal wall musculature.

Hepatic Cysts

We have previously studied the role of cyst fluid analysis and surgical management of liver cysts in the laparoscopic era.33 In this report, we compared the management of liver cysts in the 2 eras before and after laparoscopic surgery and proposed a diagnostic and treatment algorithm (Fig. 1). In the absence of laparoscopic surgery, the clinician must agonize over the decision to proceed with surgical management. If a standard approach is used, a subcostal incision is needed with a midline extension and perhaps an extension to the left subcostal region for full mobilization and exposure. Although these are commonly used incisions for liver resection and are used routinely when needed without hesitation, the uncertainty in diagnosis in combination with the morbidity of an open liver procedure might cause the surgeon to hesitate in proceeding with a standard approach,34, 35 especially if the cyst is asymptomatic. In contrast, the availability of a laparoscopic approach renders the decision to proceed easier. Thus, the procedure becomes a laparoscopic biopsy of the cyst wall, even though in essence, the surgeon performs the identical procedure that would have been done with an open approach. Certainly, from a cosmetic perspective in a young patient, this issue is worthy of consideration. Also, the same can be said from a morbidity perspective in an older individual. Moreover, laparoscopic surgical management is likely to reduce the number of patients with biliary mucinous cystadenoma who are treated conservatively because of the concerns of incisional morbidity and cosmesis, resulting in an overall optimization of the management of these diagnostically challenging lesions. The following representative case will illustrate some of the points discussed above.

Figure 1.

Diagnostic and treatment algorithm for IBC. Abbreviations: CF, cyst fluid; LS, laparoscopic surgery; MS, mesenchymal stroma; IM, intestinal metaplasia.48

Case 1:

A 35-year-old female presents with a symptomatic 23 cm cystic lesion occupying the right lobe. There is no history to suggest an infectious etiology. Radiologic findings demonstrate a single septum within the cyst (Fig. 2). The presumptive diagnosis of benign hepatic cystadenoma is made.

Figure 2.

CT shows hepatic cyst occupying right lobe with a single septum.

A standard surgical approach would necessitate, at a minimum, a right subcostal incision with a midline extension for adequate exposure. Instead, the decision is made to proceed with a laparoscopic unroofing and marsupialization. The patient is discharged to their home hours after the procedure with three 1-cm incisions covered with small bandages. The pathology report is consistent with benign, serous cystadenoma. She makes a full recovery. Three years later, a scan shows no evidence of recurrence.

Hepatic Adenoma

Radiologic imaging of the liver has improved dramatically in terms of sensitivity and specificity (Reviewed in the references36, 37). As a result, serendipitous liver lesions are increasingly noted and the management of these lesions requires an organized approach to asymptomatic benign tumors.38 The differential diagnosis of benign solid liver tumors includes among others, hepatic adenoma and focal nodular hyperplasia. The clinical indecision between these two relatively common lesions continues to affect the clinician, despite recent advances in diagnostic tools.39–41 Although sulfur colloid scans can be useful, they are not pathognonomic of either condition. Even biopsies can be difficult in interpret unless Kupffer cells or scar are clearly evident consistent with focal nodular hyperplasia (FNH). The finding of “normal hepatocytes” could mean that the biopsy missed the lesion altogether. Alternatively, this finding could represent sampling error in the case of FNH and provide the erroneous diagnosis of adenoma. Thus biopsy is not necessarily definitive. The accepted approach to asymptomatic FNH is conservative (nonsurgical) and consists primarily of serial observation with imaging modalities, whereas the recommended treatment for a 5 cm adenoma in a young woman is typically resection.42, 43 Both lesions are more prevalent in young women and therefore the issue of cosmesis is often brought up in the process of management discussion and informed consent. Standard incisions used for liver resection can be disfiguring. This consideration may not be as valid in cases where resection is clearly indicated, such as in malignancy or even in the setting of diagnostic certainty about adenoma. However, it can become a significant consideration when the indication for resection may be in question. If the lesion is small and peripheral, a laparoscopic approach could easily be considered a superior alternative to a variety of diagnostic options that may lead to diagnostic uncertainty. The nascent literature on laparoscopic liver resections appears to focus on peripheral lesions.44–46 We and others have demonstrated similar results with lobar hepatic resection.47–50 As the expertise and experience in laparoscopic liver surgery increases, it is likely that laparoscopic resection will be considered for these cases when diagnostic uncertainty exists even for deeper and larger tumors necessitating formal hepatic lobectomy. A second consideration in this discussion is the necessity for repeated serial imaging of lesions when conservative treatment is chosen. Typically, the lesions are imaged with either CT or MRI at 6 to 12 month intervals in order to determine whether the lesion has grown in size or changed in appearance. Therefore, any discussion of resection versus conservative management must include an evaluation of the logistics and costs of conservative management. Thus, as in our discussion with hepatic cysts, a laparoscopic resection of a peripheral benign solid lesion essentially replaces both biopsy and serial imaging in the management of the lesion. In cases of central lesions, the same can only be said if the expertise of the specific surgical team is such that the morbidity associated with a formal lobar hepatectomy is low. The following representative case will illustrate some of the issues raised in the foregoing discussion.

Case 2:

A 27-year-old female presents with an asymptomatic 5-cm lesion in the right lobe of the liver. The lesion is radiologically consistent with hepatic adenoma, although FNH is certainly in the differential diagnosis (Fig. 3A). Sulfur colloid nuclear scan shows no uptake favoring the diagnosis of adenoma. Several diagnostic and therapeutic options are considered. The patient is an aerobics instructor and is particularly concerned about the cosmetic and physical effects of surgery.

Figure 3.

(A) CT shows solid 5-cm lesion in right lobe consistent with either adenoma or FNH. (B) Laparoscopic view following right lobe resection. Umbilical, epigastric, and one of two right subcostal laparoscopic ports remain visible. The videoscope is placed through the umbilical port.

The patient undergoes laparoscopic resection of the right hepatic lobe (Fig. 3B). She is discharged on the first postoperative day and makes a full recovery. Pathologic examination confirms the diagnosis of adenoma.

Hepatic Malignancy

The assessment of resectability of malignant liver tumors, both primary and metastatic has improved through better imaging quality. Moreover, increasing success with chemotherapeutic agents51 and the advent of local and regional ablative treatments, either as substitutes for resection or as adjuvant therapies have changed the surgical management and approach to common malignant tumors (Reviewed in the references52, 53–57).

Standard open liver resection, in the setting of cirrhosis, especially in an older individual is associated with significant morbidity and mortality. The use of laparoscopic techniques for malignant tumors has been espoused by some but not by others.58–60 The rationale behind the latter sentiment includes the risk of tumor dissemination.61, 62 Also, one of the concerns previously expressed focuses on resection margins. It is felt by some that laparoscopic resection may compromise resection margins.63 We do not feel that this consideration should contraindicate laparoscopic resection as long as careful attention is paid to the margins during resection either visually, or with the use of intraoperative ultrasonography.64, 65 The use of laparoscopic surgery for the management of hepatic malignancies extends beyond resection and includes its use for biopsy and staging purposes, as well as for laparoscopic ablative therapies, such as radiofrequency ablation. We propose that laparoscopic approaches in these patients are associated with acceptable rates of morbidity and mortality. Certainly, the rate of both perioperative and postoperative complications and the lengths of stay compare favorably with open procedures. The representative case selected for this discussion will highlight some of the important issues related to the treatment of HCC in the setting of cirrhosis.

Case 3:

A 74-year-old male with chronic hepatitis C and cirrhosis presents with a 5 cm pedunculated lesion in the left lateral segment that has grown over the past year, radiologically consistent with hepatocellular carcinoma (HCC) (Fig. 4A). The alpha-fetoprotein level is elevated. His past medical history is significant for a cardiac transplant 10 years ago. The metastatic work-up is negative. His underlying liver disease is well compensated, and he has Childs-Pugh-Turcotte A liver disease (score = 5). There is no evidence of portal hypertension. The synthetic function is normal. He refuses to consider liver transplantation citing his age, previous medical history, and social circumstances. Laparoscopic resection is selected as the preferred therapeutic option.

Figure 4.

(A) CT shows lesion in left lobe consistent with HCC. (B) Left lateral segment including underlying HCC has been removed laparoscopically through umbilical port site using heavy retrieval bag. Epigastric and left subcostal laparoscopic ports remain visible.

The patient undergoes laparoscopic resection (Fig. 4B). He is discharged on the first postoperative day and makes a full recovery. Pathologic examination reveals a low-grade HCC in the setting of cirrhosis. The resection margins are clear and at least 2 cm away from the tumor.

Current Use of Laparoscopic Liver Resection

Northwestern Experience.

Since 1993, the Division of Transplantation at Northwestern Memorial Hospital has been involved in the care of patients with hepatobiliary pathology, such as benign and malignant liver tumors. We initiated a program in laparoscopic liver surgery in the late 1990s. Patients in our clinic are managed by a multidisciplinary process and as the laparoscopic program was initiated by one of the authors (A.K.), a database was instituted that recorded the details of each case performed laparoscopically. As the program has evolved, we have observed a shift in our decision-making paradigm that has progressed over time and that remains in evolution.

From July 2001 to July 2006, 197 hepatic resections were performed at Northwestern Memorial Hospital using the laparoscopic approach. All 197 cases in this particular series were planned initially as pure laparoscopic cases. This experience includes both benign and malignant tumors and various types of resections (Table 1).

Table 1. Both Benign and Malignant Liver Tumors and Various Resection Types Are Included
Resection Type*Northwestern All (197)Cystic LesionsHemangiomaFocal Nodular HyperplasiaAdenomaPrimary Liver CaMetastasis
  • *

    Strasberg's nomenclature for hepatic resections is used to describe the types of resections.68

Lt. lateral sectionectomy22542524
Rt. post. sectionectomy7412
Lt. hemihepatectomy2194323
Rt. hemihepatectomy4711931347
Rt. trisectionectomy44
Caudate segmentectomy4121
Resection Type*UPMC      
Resection Type*UPMC All (138)Cystic LesionsHemangiomaFocal Nodular HyperplasiaAdenomaPrimary Liver CaMetastasis
Lt. lateral sectionectomy324932311
Rt. post. sectionectomy1142221
Lt. hemihepatectomy14124232
Rt. hemihepatectomy3111
Rt. trisectionectomy
Caudate segmentectomy11

In terms of perioperative complications, these compared favorably with those of open procedures, with the exception of a single case of CO2 embolism causing transient hemodynamic instability. Also, although there were no conversions to open procedures, conversion from laparoscopy to hand-assisted laparoscopy occurred in 12 of 197 patients (6%) primarily as a result of either bleeding or concern about potential bleeding. The average surgical time compared favorably with that of open procedures (134 ± 47 minutes; range: 20-265) although given the heterogeneity of resection types, these data are difficult to interpret. There were 3 bile leaks (1.5%); 2 sealed spontaneously and 1 required endoscopic retrograde cholangiopancreatography (ERCP). Two patients (1%) required blood transfusions (2 units of pRBC each). No patients required return to the operating room. All surgical resection margins for malignant lesions were at least 1 cm. No cases of either incision or port site recurrences were noted. Postoperative complications included 4 cases of urinary retention and 2 cases of minor wound complications. The average hospital length of stay was 1.4 days (range 0.5-3) again comparing favorably with open resections. There were no cases of hepatic decompensation and no 30-day mortalities.

We noted a gradual increase in the ratio of laparoscopic to open resections over the study period from 70% in 2005, including hand-assisted procedures.

UPMC Experience.

The laparoscopic liver resection program at the UPMC Liver Cancer Center began in 2001 as an extension of the liver tumor service of the Starzl Transplant Institute. From August 2001 to August 2006, 138 laparoscopic hepatic resections have been performed (Table 1). Fifty percent of cases used a hand-port placed at the beginning of the case based on diagnosis or size/location of the lesion. Four of 138 cases were converted to open due to size (1 case) or bleeding (3 cases). Blood transfusion was given in 6 of 138 patients (4.3%) consisting of 1, 2, 2, 2, 2, and 2 units of pRBC. Three bile leaks were observed early in the series (2.2%); 2 sealed spontaneously and 1 required ERCP. Other morbidities included incisional hernia in 4 patients (1 trocar and 3 hand-port sites), subcutaneous emphysema in 1 patient, symptomatic right pleural effusion in 1 patient, and mild hepatic decompensation in a patient with Child's B cirrhosis. There were no operative or 30-day mortalities.

Concerns and Limitations

We need to address the concern that the advent of minimally invasive liver surgery could result in overuse of these procedures. Some authors have cautioned that laparoscopic procedures could lead to their use in cases where surgery is not indicated and therefore that laparoscopic procedures should only be used when an open procedure is clearly indicated. Morino et al. state that the availability of minimally invasive liver surgery should not affect the surgical management of benign liver tumors.44 We disagree with this perspective. Although we are in agreement with the fact that these procedures should be used cautiously and with appropriate indications, we argue that in some cases, especially when faced with diagnostic and therapeutic dilemmas such as those highlighted by the first 2 case studies, laparoscopic procedures might be considered instead of conservative nonsurgical management, although we concede that the risk of overuse exists. We have emphasized that the availability of laparoscopic techniques may alter decision-making beyond considerations of wound morbidity and cosmesis. In the management of solid benign tumors, we propose that a laparoscopic resection may in fact replace the need for exhaustive diagnostic tests designed to ascertain the nature of the lesion. For malignant lesions, oncologic principles must be adhered to without deviation in order to achieve optimal cancer-free survival. In general, the primary cause for altering clinical decision-making in these cases may result from a shift in the risk-to-benefit ratio associated with laparoscopic versus open procedures in the hands of experienced laparoscopic surgeons. The assessment of risk must include a full evaluation of the surgeons' and center's experience and expertise with laparoscopic liver surgery in determining any safety concerns. Similarly, any evaluation of benefit should incorporate alternative diagnostic and treatment modalities that must include, but not be limited to, open surgical resection. Certainly, the rates of postoperative complications observed in both the Northwestern and UPMC series are similar to those published in other reviews.6, 9, 10, 12, 14

No randomized clinical trials have been performed comparing laparoscopic to open hepatic resection. Although there are a few case-controlled trials comparing laparoscopic liver surgery to open resection, the numbers are small and the endpoints may not be relevant.66, 67 Innovative investigative strategies will need to be used in developing the appropriate study designs needed to compare open procedures to their laparoscopic counterparts in the context of a seemingly rapid adaptation of these procedures. Let us recall that although laparoscopic cholecystectomy has replaced open cholecystectomy as the standard of care for removal of the gallbladder, the merits of the laparoscopic procedure were never tested in a randomized controlled trial comparing it to the open approach.

Given the rapid adoption in specialty centers and dramatic acceptance by patients, it is unlikely that a randomized trial will be performed for benign hepatic lesions. The more relevant question is the need for a randomized trial in the setting of cancer to assess the impact of minimally invasive hepatic resection on local recurrence and survival.

We reviewed the use of this rapidly evolving technology in the surgical management of common liver tumors at 2 large U.S. medical centers. We demonstrated that the growing availability of these procedures can affect clinical decision-making in the management of liver tumors.