Surgical Endoscopy

, Volume 26, Issue 6, pp 1772–1776

Video-assisted thoracoscopic transdiaphragmatic liver resection for hepatocellular carcinoma


  • Jordan M. Cloyd
    • Department of SurgeryStanford University
    • Department of SurgeryStanford University
    • Department of SurgeryStanford University
Dynamic Manuscript

DOI: 10.1007/s00464-011-2062-x

Cite this article as:
Cloyd, J.M. & Visser, B.C. Surg Endosc (2012) 26: 1772. doi:10.1007/s00464-011-2062-x


Because of technical complexity, concern for vascular control, and uncertainty in regard to oncologic outcome, the application of minimally invasive techniques to liver surgery has been slower than in most other abdominal procedures. This is despite well-known advantages with respect to postoperative pain, length of hospitalization, and recovery time. Although laparoscopic liver surgery has recently become more common, the majority of laparoscopic liver resections comprise anterolateral wedge resections and left lateral sectorectomies. Laparoscopic resections of the posterosuperior segments are more difficult and few reports are available in the literature. Compared to laparoscopy, gaining access to tumors in the dome of the liver may be more easily obtained via thoracoscopy, thereby preserving the benefits of minimally invasive surgery. This technical report describes two cases of hepatocellular carcinoma in segments VII and VIII resected via a video-assisted thoracoscopic transdiaphragmatic approach.


ThoracoscopyHepatocellular carcinomaVATSLaparoscopyMinimally invasive surgeryHepatectomy

Since its introduction, minimally invasive laparoscopic surgery has become the gold standard approach for most abdominal procedures. Application of minimally invasive techniques to liver surgery has been slower due to technical limitations and challenges (e.g., parenchymal transection and vascular control) and concerns over oncologic outcomes. However, studies support the extension of the benefits of laparoscopy to liver surgery, including reduced postoperative pain, shorter hospitalization, less morbidity, and quicker recovery [1]. Because of these advantages, there has been a dramatic increase in the number of reported laparoscopic liver resections in recent years [2]. Most centers’ experiences have involved primarily anterolateral wedge resections (segments II–VI) and, to a lesser extent left, lateral sectorectomies [2, 3]. However, performing resections of the posterosuperior portions (segments I, VII, and VIII) of the liver have proven far more difficult [3, 4], with fewer reports in the literature [4, 5].

In the era of open liver surgery, some tumors situated on the dome of the liver may be more easily accessed through the chest than the abdomen [68]. Proponents of this approach have especially suggested its utility in patients with cirrhosis. A transthoracic transdiaphragmatic approach aids in establishing the operative field with less mobilization and compression of the right lobe, which is postulated to lead to less injury to the liver in patients who already suffer from compromised hepatic function. The transthoracic approach to the liver has also been employed to perform simultaneous pulmonary and hepatic metastectomies, resecting both lesions through one incision [7, 9]. Compared to traditional thoracotomy, video-assisted thoracic surgery (VATS) also provides a transthoracic approach while maintaining the benefits of minimally invasive surgery, such as improvements in postoperative pain, pulmonary function, and length of hospitalization [10]. Therefore, gaining access to segment VII/VIII tumors using minimally invasive techniques may be obtained more easily via a thoracoscopic as opposed to a laparoscopic approach. Indeed, several investigators have already performed radiofrequency ablation (RFA) of superficial liver dome tumors through a thoracoscopic transdiaphragmatic approach [11, 12].

To our knowledge, a true thoracoscopic transdiaphragmatic liver resection has been reported by only one Japanese group [13, 14]. The purpose of this article, therefore, is to describe the technique and outcome of a video-assisted thoracoscopic transdiaphragmatic resection in two patients with hepatocellular carcinoma.

Case 1

The patient was a 56 year old man with chronic hepatitis C and cirrhosis who presented with a 2 cm segment VII mass consistent with hepatocellular carcinoma (HCC). 2 years prior he had undergone a laparoscopic RFA of a segment VI HCC. Triphasic computed tomography (CT) of the abdomen and pelvis showed a 1.3 × 1.2 cm hypervascular focus with washout in the region of segment VII (Fig. 1). Given its small size, magnetic resonance imaging (MRI) was obtained for additional characterization, confirming a 2.1 × 1.4 cm segment VII lesion and no lesions elsewhere in the liver. His preoperative lab work was notable for a platelet count of 58, creatinine of 1.3, total bilirubin of 1.0, INR of 1.4, albumin of 3.1, and AFP of 10.5. Therefore, he was Child-Pugh class A and had a MELD score of 12. Given its subdiaphragmatic location, the patient’s obese body habitus (BMI = 34), and his history of prior abdominal surgery, we elected to pursue a transthoracic transdiaphragmatic approach using VATS. A thoracoscopic resection of the tumor was performed (described below). The patient’s hospital course was unremarkable. His chest tube was removed on postoperative day (POD) 3 and he was discharged home on POD 5. The pathology report revealed a 2.3 × 2.2 cm moderately differentiated HCC with negative margins (Fig. 2). He was seen in clinic 2 weeks later and had already resumed his normal activities of daily living. He has no evidence of recurrence at 1 year follow-up.
Fig. 1

A, B Preoperative CT images of 56 year old man with segment VII hepatoma demonstrating hypervascular enhancement on arterial phase
Fig. 2

Bisected specimen removed thoracoscopically. Pathology demonstrated a 2.3 × 2.2 cm moderately differentiated HCC with negative margins

Case 2

The patient was a 59 year old man with chronic hepatitis C who presented with a 2 cm lesion in the dome of the liver concerning for HCC. 18 months prior he had undergone an open right posterior sectorectomy for a large 8 cm HCC via a reverse-L type incision. Triphasic CT demonstrated a 3.0 × 2.7 cm ill-defined hypervascular lesion in segment VIII. MRI was obtained to confirm the findings, revealing a 2.4 cm hepatic dome lesion with hypervascular enhancement and washout (Fig. 3). His preoperative lab work was notable for a platelet count of 131, creatinine of 0.7, total bilirubin of 1.1, INR of 1.0, albumin of 2.8, and AFP of 17.6. A thoracoscopic transdiaphragmatic hepatic wedge resection was chosen due to the patient’s previous liver resection and the tumor’s subdiaphragmatic location. The operation was tolerated well. His hospital course was unremarkable. Specifically, his chest tube was removed on POD 1 and he was discharged on POD 3. Pathology demonstrated a 2 cm well-differentiated HCC with negative margins (Fig. 4). He was seen in clinic 2 weeks later essentially recovered from surgery. A repeat CT scan at 7 months postoperatively demonstrated no evidence of recurrence.
Fig. 3

Preoperative T2 phase MRI of a 59 year old man with a segment VIII HCC
Fig. 4

Resected specimen from case 2. Pathology showed a well-differentiated HCC with negative margins

Operative technique

After induction of general anesthesia and intubation with a double-lumen endotracheal tube, the patients were placed in the left lateral decubitus flexed position. Single-lung ventilation was initiated before trocar placement. To begin, a 12 mm trocar was inserted in order to accommodate the laparoscopic ultrasound. Conventional laparoscopic trocars were utilized instead of thoracic trocars as they permitted insufflation in the chest. For patient 1, we positioned the trocars and worked from the chest toward the back because the lesion was located very posterolaterally. For patient 2, the tumor was located quite medially, so we worked from the back toward the chest. Insufflation to 15 mm Hg was used to assist in collapsing the lung and to reduce venous bleeding during hepatic parenchymal division. We started with laparoscopic ultrasound using a 7.5 MHz linear transducer. The tumor was then identified immediately underneath the diaphragm and the diaphragm was marked with cautery. Two additional 5 mm ports were then positioned in a “cluster” fashion in order to optimize triangulation. A 5 mm LigaSure (Covidien, Mansfield, MA) device was then used to open the diaphragm directly over the tumor, wide enough to allow access to the liver without active retraction of the diaphragm. Ultrasound was used to mark the transection line. For patient 1, the Habib 4 × RFA device (AngioDynamics, Latham, NY) was chosen to precoagulate the hepatic parenchyma because the patient had a dense cirrhotic liver. For patient 2, the liver was fibrotic but not as cirrhotic, hence the 5 mm LigaSure was used to divide the parenchyma (we use this instrument routinely for dividing normal liver parenchyma during laparoscopic hepatectomy). In both cases, parenchymal division proceeded smoothly despite the limited access to the liver and lack of inflow control. The insufflation was reduced before closing the diaphragm to ensure that there was no venous bleeding. The diaphragm was closed using a running 2–0 silk suture in patient 1 and interrupted 2–0 silk sutures in patient 2. In each case, the specimen was placed in a small LapSac (Cook Medical, Bloomington, IN) and withdrawn by slightly enlarging the 12 mm trocar site. A 32 Fr angled chest tube was placed through the 12 mm port site (video, see Electronic supplementary material).


Laparoscopic hepatectomy requires great technical expertise but clearly offers all the advantages of minimally invasive surgery: less postoperative pain, less operative blood loss, and shorter hospitalization [1, 2]. These benefits come with comparable oncologic outcomes [15] and hospital costs [16]. Tumors in the dome of the liver, including segments VII and VIII, are difficult to access via traditional laparoscopy. A recent report by Cho et al. [17] concluded that while feasible, laparoscopic resections of segment VII/VIII tumors involve longer operating times, increased operative blood loss, higher conversion to open rates, and longer hospitalizations compared to other laparoscopic liver resections. A subset of these tumors may best be resected by a transdiaphragmatic approach via thoracoscopy.

Teramoto et al. [13] previously described their experience with thoracoscopic transdiaphragmatic liver resections. While performing 219 open liver resections for HCC between 1997 and 2003, they performed 10 laparoscopic and 5 thoracoscopic liver resections. Of the five thoracoscopic resections, four hepatomas were located in segment VIII and one was located in segment VII. The technique was similar to ours described above. Specifically, they used five ports at varying intercostal levels, utilized thoracoscopic ultrasound, incised the diaphragm directly above the tumor, and performed a partial hepatectomy with use of a microwave tissue coagulator. There was no mention of how the diaphragm was closed. A Penrose drain was left over the resection bed exiting the abdominal wall. They reported no major postoperative complications and no local recurrences with a mean follow-up of 23 months.

We agree with these authors that appropriate patient selection remains paramount in considering this technique. A VATS transdiaphragmatic approach may be appropriate in patients with small tumors located immediately below the diaphragm and away from the hepatic veins. Thoracoscopic ablation may be considered for deeper tumors. Although patients must have adequately preserved liver function, it is likely that a thoracoscopic approach, like the laparoscopic approach, is preferable to open surgery in cirrhotic patients [18]. Less hepatic compression and manipulation is required to access the dome of the liver compared to a thoracoabdominal or transabdominal approach. In patients with cirrhosis and portal hypertension, this may preserve hepatic blood flow, thus preventing hepatic and systemic circulation compromise [6, 19, 20]. Patients with prior abdominal procedures, as was the case for both of our patients, may also be better served by a thoracoscopic approach, avoiding intra-abdominal adhesions. Certainly in case 2, a laparoscopic resection would have been impossible after his previous open right posterior sectorectomy. Finally, another indication for this technique may be patients with synchronous pulmonary and hepatic metastases from colorectal cancer that could be simultaneously resected via VATS metastectomy similar to that reported with open thoracotomy [7, 9].

Several factors should lead the liver surgeon to exercise caution with a VATS approach. Proximity of the tumor to nearby major vessels is the most important exclusion factor. Although it has not been studied, size is an important factor to consider. Larger hepatomas are likely associated with greater bleeding risk, a larger diaphragmatic defect to close, and more difficulty extracting from the chest. Teramoto et al. [13] recommended size greater than 3 cm as a relative contraindication, although this may be considered a loose recommendation. Because of limited exposure, an anatomic resection would be difficult if not impossible with this approach. Other contraindications include the inability to tolerate single-lung ventilation and previous right-sided chest surgery.

It is important to note that the transthoracic approach prohibits any inflow control of the liver (i.e., Pringle maneuver). Therefore, taking preventative steps and having plans in place for major bleeding are imperative. Insufflation of 15 mmHg CO2 may decrease venous bleeding by reducing venous back pressure. There should be careful thought about the method of transection of the liver parenchyma, with perhaps greater consideration of precoagulation prior to parenchymal division (particularly in patients with fibrosis or cirrhosis). If significant bleeding is encountered, immediate conversion to thoracotomy rather than a laparotomy would be necessary. The time required to reposition the patient and then gain access into the abdomen, especially if the patient has had previous abdominal surgery, would be prohibitive for laparotomy. Importantly, a close relationship of the tumor to major vessels on preoperative imaging should be viewed as a relative contraindication to this approach.

Although our experience was aided by previously performing thoracoscopic RFA, we encountered several challenges in instituting this novel technique. We found it challenging to plan port placement that maximized access to the lesion. For example, in case 1, our initial placement of the 12 mm trocar was too posterior, making our working angle far too steep. We found that clustering our 5 mm ports at varying intercostal levels allowed us to approach the tumor with improved angulation. One of the most technically challenging aspects of the procedure is closing the diaphragm because intracorporeal suturing in the chest is limited by the presence of the ribs.

We acknowledge that this article serves primarily as a technical report and long-term outcomes are not included. Specifically, data on oncologic outcomes and diaphragmatic hernia formation will be important to follow. As long as appropriate margins and routine meticulous oncologic techniques are ensured, there should be no greater risk of recurrence in the chest than there is of peritoneal or wound recurrence after an abdominal approach. In addition, the liver effectively provides protection for the diaphragmatic repair, and diaphragmatic hernia is not expected to be a major complication. As with all patients with HCC, we follow with frequent serial CT imaging. Routine surveillance liver scans include the base of the chest and should be adequate to assess for both intrathoracic recurrence and problems with the diaphragm repair.

In summary, we describe two cases of segment VII/VIII hepatomas resected via a thoracoscopic transdiaphragmatic approach. Only five cases of thoracoscopic hepatic resection have been previously described, all by the Japanese group led by Teramoto et al. [13]. Although definite conclusions cannot be drawn from this case study, a completely thoracoscopic approach may allow minimally invasive hepatic resections to be done safely in select patients with subdiaphragmatic tumors.


J. Cloyd and B. Visser have no conflicts of interest or financial ties to disclose.

Supplementary material

View video

The online version of this article (doi: 10.1007/s00464-011-2062-x) contains supplementary material, which is available to authorized users. (WMV 70358 kb)

Copyright information

© Springer Science+Business Media, LLC 2011