Role of Mesohepatectomy for Pediatric Centrally Located Liver Tumors

ABSTRACT

Background:

Mesohepatectomy (MH) or central hepatectomy (CH) is a recognized surgical technique for centrally located pediatric liver tumors. This technique of liver resection avoids extended liver resections and thereby helps in the preservation of liver parenchyma. In this article, we aim to analyze our experience and outcome with this technique of liver resection.

Methods:

A retrospective analysis of patients who underwent MH from January 2015 to June 2023 at our institute was performed. The variables analyzed include demographics, preoperative treatment, perioperative management, and postoperative outcome.

Results:

A total of five patients underwent CH. Four patients had hepatoblastoma, and one patient had mesenchymal hamartoma. All four patients with hepatoblastoma received neoadjuvant chemotherapy. All five patients had negative surgical margins, and one of the five developed disease recurrences necessitating resurgery and ultimately died of metastasis. one patient sustained intraoperative major bile duct injuries, and one patient had postoperative Abnormal, well-circumscribed, extra-biliary collection of bile.

Conclusion:

MH with clear margins is feasible in the selected subset of pediatric liver tumors involving segments IV, V, and VIII with outcomes equivalent to extended hepatic resections.

INTRODUCTION

Mesohepatectomy (MH) or central hepatectomy (CH) is a recognized surgical technique for centrally located pediatric liver tumors.[1] The prerequisite for cure in malignant pediatric liver tumors is complete surgical excision with clear margins. Liver resections for centrally located liver tumors (CLLTs) that straddle the Cantlie’s line (middle hepatic vein) are technically demanding. These tumors involve central Couinaud segments IV, V, and VIII and are characterized by their proximity to the inferior vena cava (IVC), hepatic artery, hepatic vein, and portal vein bifurcation. Traditionally, extended hepatectomy (EH) is considered the procedure of choice for CLLTs. However, pretreatment extent of tumors (PRETEXT)/POSTTEXT Stage III (Type D) centrally located hepatic tumors are typically confined to segments IV, V, and VIII [Figure 1a] without involvement of the right or left hepatic veins. Both right and left extended hepatic resections are suitable for this subset of tumor creating a dilemma. Excision of major portions of liver parenchyma during EH results in massive loss of hepatic parenchyma and can progress to hepatocellular failure. MH is a parenchyma-sparing liver resection technique for tumor excision and preserves segments VI and VII, which would be resected unnecessarily in extended right hepatectomy without compromising the oncological principles. In this study, we analyze the surgical feasibility and oncologic outcome of CLLT in pediatric patients who underwent MH procedure.

Figure 1.

(a) POSTTEXT Stage III (Type D) tumors, (b) Contrast-enhanced computed tomography image of tumor in Segments IV and V

METHODS

A retrospective review of patients who underwent MH for CLLTs in the pediatric population from January 2015 to June 2023 by a single pediatric surgeon experienced in complex hepatic resections was performed.

Inclusion criteria included PRETEXT/POSTTEXT Stage III (Type D) centrally located hepatic tumors involving segments IV, V, and VIII with or without involvement of the middle hepatic vein [Figure 1]. Exclusion criteria include metastatic tumors, recurrent tumors, and tumors involving IVC, right and left hepatic veins, portal vein, and extrahepatic spread. Contraindications include tumors involving the right or left hepatic vein and amenable for extended right or left hepatectomy.

Preoperative evaluation included complete blood count, liver function tests, and coagulation profile with serum alpha-fetoprotein (AFP) levels. Triple-phase contrast-enhanced computed tomography (CECT) imaging of the abdomen with the screening of the chest was the preferred imaging modality. Based on PRETEXT staging, annotation factors, and AFP levels, risk group stratification was performed. A preoperative biopsy was performed, and neoadjuvant chemotherapy was started after tumor board discussion. Cisplatin-based chemotherapy (PLADO regimen) was used in all patients with biopsy-proven hepatoblastoma. Patients underwent reevaluation with CECT abdomen after two to three cycles of neoadjuvant chemotherapy. CH was offered to patients with POSTTEXT Stage III (Type D) tumors [Figure 1b].

The surgical technique involves exposure by a generous bilateral subcostal incision. General anesthesia was used with intraoperative monitoring using the central venous line and arterial line in all cases. The liver was mobilized by dividing the right and left triangular ligaments and the falciform ligament. The gallbladder (GB) was mobilized from the GB fossa and retained after ligating the cystic artery. This was followed by hilar dissection. The right and left branches of the hepatic artery, portal vein, and hepatic duct were identified and preserved. The segmental portal and hepatic branches that supply the central segments were dissected intraparenchymally and ligated. The right hepatic artery was traced until its bifurcation, and the branch supplying the anterior segment was ligated and divided. The right portal vein was also traced to its bifurcation, and the branches supplying the segments V and VIII were divided separately. Segment 4 blood supply was meticulously dissected from the left hepatic artery and left portal vein and divided during the parenchymal transection. Parenchymal resection extent with 1 cm margin clearance was marked, and parenchymal dissection proceeded with bipolar coagulation. The Kelly clamp technique was used with the bipolar instrument. The parenchyma was divided using bipolar, and intervening ductal and vascular branches were suture ligated. Although devices like Cavitron ultrasonic surgical aspirator (CUSA) are preferred, they cannot be used due to nonavailability. Intersegmental vascular pedicles entering the tumor were dealt during parenchymal dissection with double ligation and division. The right-sided resection plane is the same as the extended left hepatectomy along the right hepatic vein. The left-sided resection line is the same as the extended right hepatectomy along the left hepatic vein. Surgeons experienced in right and left EH alone can identify the variation in anatomy and minimize blood loss and biliary injuries. At the end of the procedure, the two-resection planes meet at the middle hepatic vein, which was ligated at the superior aspect of the tumor. MH was then completed with the preservation of the right and left hepatic veins [Figure 2].

Figure 2.

(a-d) Steps of mesohepatectomy procedure

The biliary leak from parenchyma and ducts was checked by compressing the GB after occlusion of the common bile duct with vascular clamp and observing for bile leak. This procedure was equivalent to an intraoperative cholangiogram. After ensuring biliostasis and hemostasis, cellulose polymer was applied to the raw surfaces as a hemostatic agent. A tube drain was placed abutting the resected margins of the liver before abdominal wound closure. Postoperatively, the patient was shifted to the pediatric intensive care unit for close monitoring and observation.

All complications derived from liver resection accounted for postoperative morbidity and were graded by the Clavien–Dindo classification. Posthepatectomy liver failure was defined by the International Study Group of Liver Surgery as an increased international normalized ratio and concomitant hyperbilirubinemia on the 5^th postoperative day. Postoperative chemotherapy was continued as per the risk stratification of the patient after a discussion with the institutional tumor board. Post-MH follow-up consisted of monitoring with serum AFP, chest X-ray, ultrasound abdomen imaging, and liver function tests was done every 3 months. This was followed with a biannual surveillance protocol.

RESULTS

A total of five children with a male predominance were treated by central hepatic resection from January 2015 to June 2023 [Table 1]. Four patients had biopsy-proven hepatoblastoma, and one child was an antenatal diagnosed mesenchymal hamartoma. Mesenchymal hamartoma was antenatally diagnosed and slowly increased in size to involve segments IV, V, and VIII at the time of presentation to our institute. CH was the only available option for management as extended hepatectomies would have increased the morbidity.

Table 1.

Baseline characteristics

Age at diagnosis (years)	Sex	Type of lesion	Initial AFP (ng/mL)	Liver segments involved	Treatment before surgery	Mets	Recurrence
3	Male	Hepatoblastoma – epithelial, fetal	60,500	IVa, IVb, V	PLADO – four cycles	-	-
Antenatal	Male	Mesenchymal hamartoma	38,000	IVa, IVb, V	None	-	-
2	Female	Hepatoblastoma – epithelial, fetal	1,00,500	Iva, IVb, V, VIII	PLADO – four cycles	Yes	Yes
8/12	Male	Hepatoblastoma – epithelial, fetal	92,738	IVa, IVb, V	PLADO – four cycles	-	-
2	Female	Hepatoblastoma – epithelial, fetal	11	IVa, IVb	PLADO – four cycles	-	-

AFP: Alpha-fetoprotein

The median age of diagnosis was 24 months (0–36 months). The initial AFP was elevated in three patients. Preoperative biopsy was obtained in 80% (n = 4) of patients, and they underwent neoadjuvant chemotherapy with PLADO (cisplatin and doxorubicin) regimen [Table 1].

The mean age at the time of surgical intervention was 50.5 months (8–36 months). All cases underwent resection of segments IV and V. One case segment VIII was also included in resection. None of the cases required vascular exclusion. The right and left hepatic veins were preserved in all cases. One patient (20%) sustained an injury to the common hepatic duct which necessitated biliary diversion with a Roux-en-Y hepaticojejunostomy in the same sitting. The mean estimated blood loss was 22 mL/kg (15–30 mL/kg). In all cases (n = 5, 100%), it was possible to complete MH with disease-free margins [Table 2].

Table 2.

Intraoperative and postoperative details

Patient	Age at surgery (years)	Operative time (min)	Blood transfused (mL/kg)	Biliary injury and management
1	3	210	25	None
2	9/12	240	15	Bilioma – Image-guided drainage
3	2	210	30	Common hepatic duct – Roux-en-Y hepaticojejunostomy
4	8/12	200	20	None
5	2	180	20	Bilioma – Image-guided drainage

Postoperatively, two patients developed bilioma (Clavien–Dindo Type 3), which necessitated image-guided pigtail insertion. Bile leak settled without resurgery within 4–6 weeks. The remainder of the patients had an unremarkable postoperative period with a mean duration of hospital stay being 7.6 days (7–14 days). None of the children developed other complications requiring reoperation. There was no occurrence of delayed jaundice or liver cell failure during follow-up. Long-term biliary complications like stricture did not occur in any patient.

The gross and microscopic resection margins were negative for disease in 100% (n = 5). The median follow-up duration was 33.8 months (10–96 months) [Table 3]. One patient developed recurrence 1 year after surgery and required a second surgical resection for the recurrent left lateral segment of the liver away from the resection line [Table 1]. Pathology was consistent with recurrent hepatoblastoma. The same patient developed lung metastasis during follow-up and succumbed [Table 1]. The overall 2-year survival (OS) was 75% in children with hepatoblastoma. Two children have crossed 4-year disease-free survival.

Table 3.

Follow-up details of hepatoblastoma (n=4)

Parameter	n (%)
Follow-up duration (months)	33.8 (10–96)
Clear resection margins	4 (100)
Overall 2-year survival (OS) rate	3 (75)
DFS rate	3 (75)
Intrahepatic recurrence rate	1 (25)
Metastasis	1 (25)
Resurgery rate	1 (20)

DFS: Disease-free survival, OS: Overall survival

DISCUSSION

Pretext Stage III (Type d) tumors involving central segments that include left medial (IV) and right anterior (V and VIII) sections are traditionally treated by extended hepatic resections. The technique of CH or MH was first described in 1972 by McBride and Wallace.[1] In this procedure, the caudate lobe forms the posterior resection margin. After preserving the right and left hepatic veins, the parenchymal transection plane meets in front of IVC posteriorly.[2] Although much more commonly performed and reported in adults, the literature regarding pediatric CH is sparse.

A concern regarding MH is the need for surgical dissection close to the liver hilum and thus increased intraoperative blood loss. Although technically challenging, none of our cases had major vascular injury or intraoperative hemorrhage. The mean estimated blood loss was 22 mL/kg (15–30 mL/kg).

Our case series reports a 20% chance of intraoperative major bile duct injury and a 40% chance of postoperative bile leak. Guérin et al. reported a 22% incidence of biliary injuries due to the proximity of resection planes and biliary structures.[3] Conventional hepatectomy has reported a 12.5%–17% incidence of biliary injuries.[4,5] None in our patient group developed late biliary complications like ischemic biliary stricture as reported by Chen et al.[6]

MH has an increased risk of major bile duct injury due to the proximity of the tumor to the liver hilum and both hepatic ducts. Meticulous and careful dissection is required to avoid major bile duct injury. Most of the major bile duct injuries can be identified intraoperatively and managed according to the nature of the injury. Minor bile leaks in the postoperative period usually settle with image-guided drainage. In our series, biliary leaks (two cases) occurred from the preserved right posterior segments and were treated conservatively with pigtail catheter insertion. The postoperative recovery was excellent. The role of intraoperative use of indocyanine green (ICG) administration in the bloodstream and detecting ICG leakage from the hepatic dissection plane using an ICG camera for reliable detection of biliary leak and consequent reduction of biliary fistula has been studied.[7]

There was a 100% negative margin status in our study group. The overall and disease-free survival rates are 75% in hepatoblastoma and are comparable to 88% reported by Guérin et al.[3] This shows that there is no compromise of surgical margin with MH, and it is oncologically equivalent to extended hepatic resection. One recurrence that was reported was in the left lateral segment and would have occurred even if the patient had undergone an extended right hepatectomy. Since the right posterior and left lateral segments were preserved in MH, recurrences can be managed with additional hepatic resection if diagnosed early. In this case, the patient underwent left lateral segmentectomy after diagnosis of recurrence. Unlike adults, although there is no preexisting liver cirrhosis in pediatric liver malignancy, postresection remnant liver remains vulnerable due to neoadjuvant chemotherapy, and the occurrence of postoperative liver failure is not uncommon. In our case series, there is no delayed jaundice or hepatocellular failure, as CH has 35% less parenchymal loss compared to EH and thus increased functional liver reserve.[8] Scudamore et al. also reported a lower morbidity with CH compared with EH.[9] However, long-term follow-up is required to confirm these benefits of MH. The limitations of the study include a small number of cases and the need for long-term follow-up.

CONCLUSION

CH can be considered an oncologically equivalent procedure to EH for the subset of pediatric liver tumors involving segments IV, V, and VIII. With the distinct benefit of hepatic parenchyma preservation, it is a viable alternative to extended hepatic resection which fulfills the oncologic principles. Meticulous surgical principles have to be adhered to accomplish MH without complications.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.