HEPATIC resections are now performed widely, even in community hospitals, but during the past 10 years we have also witnessed the creation of several specialized liver surgery units; these centers serve as referral centers for patients with large hepatic tumors or for patients with end-stage liver diseases who require transplantation. [1] The routine practice of hepatic surgery in these tertiary referral centers has led to the development of highly specific approaches to the liver.

The reduction of blood loss during hepatic resection remains one of the most controversial aspects. In 1908, Pringle [2] first described the possibility of minimizing blood loss during liver surgery by clamping the hepatic pedicle. The main concern with the Pringle maneuver is that mere clamping of the pedicle does not prevent bleeding from the hepatic veins. Complete vascular control of the liver was described by Heaney et al [3] and later popularized in the clinical field by Huguet and coworkers [4-6] at the beginning of the era of hepatic surgery. Today, total vascular exclusion (TVE) of the liver seems to be the standard method for the reduction of blood loss during hepatic surgery. The attitude toward TVE differs among the various centers where major hepatectomies are routinely performed: some of them apply it widely, even when minor resections are planned, while others consider TVE a further refinement of the technique of liver surgery to be undertaken in highly selected cases. [7-9] We believe that the application of TVE during hepatic surgery has clear indications and that most liver resections can be performed with the standard techniques of pedicle clamping or by the selective ligation of the afferent vessels. Only 2.6% (19/722) of all hepatic resections performed at our Department of Surgery during the past 14 years required TVE of the liver. We retrospectively reviewed this institutional experience with hepatic surgery performed under TVE and analyzed our policy of blood inflow control of the liver during resection.

From November 1, 1981, to March 31, 1996, 722 liver resections were performed at the Second Department of Surgery of the University of Bologna, Bologna, Italy. The mean +/- SD age of patients was 55.2 +/- 13.8 years, ranging from 7 to 82 years.

Indications for surgery were hepatocellular carcinoma in 279 (38.6%) (201 of these had cirrhosis), metastasis in 213 (29.5%), hemangioma in 47 (6.5%), hydatid cyst in 37 (5.1%), focal nodular hyperplasia in 29 (4.0%), trauma in 21 (2.9%), hepatocellular adenoma in 17 (2.4%), Klatskin tumor in 16 (2.2%), and miscellaneous other diseases in the remaining 63 (8.8%).

The anatomy of the liver is described according to division into 8 segments. [10] The removal of a portion of the liver smaller than an anatomical segment is defined as a wedge resection; the removal of 1 to 3 anatomical segments, segmentectomy; the removal of segments II and III, left lobectomy; the removal of the left hemiliver, thus including segments II, III, and IV, left hepatectomy; the removal of the right hemiliver, thus including segments V, VI, VII, and VIII, right hepatectomy; and a left or a right hepatectomy that included a portion of the contralateral lobe or the caudate lobe, extended hepatectomy. The subphrenic cavity was always drained.

A total of 143 wedge resections (19.8%), 352 segmentectomies (48.8%) (1-3 anatomical segments), 172 hepatectomies (23.8%), and 55 extended hepatectomies (7.6%) were performed. A total of 375 procedures (51.9%) were performed before 1992 and the remaining 347 (48.1%) thereafter.

INDICATIONS FOR TVE

Indications for TVE during elective liver resection were the close relationship of the lesion to be resected with the inferior vena cava (IVC) or the hepatic veins at their confluence into the IVC or between the portal bifurcation and the IVC. Total vascular exclusion was also indicated in cases where the retrohepatic IVC was infiltrated by a tumor that could not be removed with a conventional hepatic resection.

ANESTHETIC AND SURGICAL MANAGEMENT

Patients were placed on a heating blanket to reduce intraoperative hypothermia. After endotracheal anesthesia was administered, standard noninvasive monitors for blood pressure, electrocardiogram, pulse oximetry, and temperature were placed. Additional invasive monitoring was used for radial arterial and blood gas values. A Swan-Ganz catheter was used during elective resections to monitor central venous pressure, pulmonary arterial pressure, pulmonary capillary wedge pressure, and cardiac output by thermodilution method during TVE.

The TVE was established according to the technique described by others. [6] The abdomen was entered through a J-shaped right subcostal incision. Intraoperative ultrasonography was always used when the resection of a liver tumor was planned, to define the relationship between the tumor and the main intrahepatic vessels and to exclude the presence of undisclosed intrahepatic spread. [11] The liver was freed from the triangular and round ligaments with the use of electrocautery. The lesser omentum was divided, but care was taken to preserve accessory left hepatic arteries arising from the left gastric artery. The IVC was completely isolated and encircled above and below the liver. The hepatic pedicle was identified and encircled as well.

During elective procedures, the hepatic pedicle and the IVC above the liver were clamped for a few minutes to test the patient's hemodynamic stability before the procedure was started. The adrenal vein was not tied if there was no need to do so. Small phrenic veins were tied and sectioned when their position was considered to potentially jeopardize the procedure. The first clamp was positioned to sagittally occlude the entire retrohepatic IVC. The hepatic pedicle and the IVC above the liver were then clamped. Care was taken to place the lower IVC clamp in close contact with the superior IVC clamp to ensure complete control of the vessel (Figure 1). The aorta was never clamped.

Figure 1. Technique of total vascular exclusion of the liver.

Resections were defined as transparenchymal when the main glissonian branches were identified and divided within the liver, after the incision of the hepatic parenchyma. Resections were defined as typical when the main glissonian branches were identified and divided before the incision of the parenchyma was started. Resections were performed with the so-called Kellyclasia. Briefly, the glissonian capsule was incised with the use of electrocautery and the hepatic parenchyma was gently crushed with the tip of a Kelly clamp to identify intrahepatic vessels. Main branches were sutured while smaller vessels were secured with clips or cauterized. An ultrasonic dissector was never used.

Data were prospectively collected and retrospectively analyzed. Results are expressed as mean +/- SD. Differences between means were evaluated with the Student t test. Differences between groups were evaluated by means of the chi squared test. Data were evaluated with the use of the SPSS software package. [12] A value of P<.05 was considered statistically significant.

Total vascular exclusion was attempted in 20 patients. In 1 patient, it could not be used; this 69-year-old woman had a metastasis from adenocarcinoma of the breast localized in the caudate lobe and infiltrating the IVC. She had received several courses of intravenous chemotherapy. This patient did not show the hemodynamic stability necessary for the procedure when the intraoperative clamping test was performed. Since other methods to safely occlude the IVC were not possible, the planned liver resection was abandoned.

The remaining 19 TVEs represent 2.6% of all liver resections performed in our department. Indications for resection were metastases in 10 cases (52.6%) (5 from colorectal cancers, 3 from leiomiosarcomas, and 2 from neuroendocrine tumors), hepatocellular carcinoma in 3 (15.8%), hepatocellular adenoma in 2 (10.5%), and trauma, malignant paraganglioma, cholangiocarcinoma, and echinococcosis in 1 case (5.3%) each.

Procedures performed were wedge resection in 1 case (5.3%), segmentectomy in 1 (5.3%), bisegmentectomy in 2 (10.5%), and major hepatectomy in 15 (78.9%); of these, 7 were extended ( Table 1) and ( Table 3). Left lobectomy was simultaneously carried out in 2 patients who received minor resections. In 3 cases, it was necessary to remove small portions of the diaphragm muscle infiltrated by the tumor. In 1 case, a simultaneous pancreatectomy was also performed. An implantable device for locoregional chemotherapy was placed in 3 patients with colorectal liver metastases.

Table 1. Clinical Details of 19 Patients Who Underwent Hepatic Surgery Under Total Vascular Exclusion (TVE) From:   Grazi: Arch Surg, Volume 132(10).October 1997.1104-1109

Table 3. Clinical Details of 19 Patients Who Underwent Hepatic Surgery Under Total Vascular Exclusion (TVE)* (continued) From:   Grazi: Arch Surg, Volume 132(10).October 1997.1104-1109

Mean duration of TVE was 44.3 +/- 12.8 minutes (range, 11-61 minutes). The use of TVE made liver resection possible without the need for blood transfusions in 5 cases (26.3%). In the 14 cases in which blood was transfused, the mean volume was 1267.8 +/- 578.3 mL (range, 500-2400 mL). Duration of the procedures was 328.5 +/- 97.3 minutes (range, 180-500 minutes).

In 6 cases, the neoplastic infiltration of the retrohepatic IVC was confirmed at laparotomy. In 4 of these cases, it was necessary to resect the infiltrated segment of the IVC and to directly suture the vessel during TVE.

In the remaining 2 patients, in whom a right hepatectomy was performed, the retrohepatic IVC was completely resected and removed with the diseased liver. In both cases, it was possible to leave a 2-cm IVC stump below the confluence of the hepatic veins. This allowed placement of a new clamp on the IVC below the confluence of the hepatic veins and removal of the previously placed IVC clamp together with the clamp on the hepatic pedicle, thus revascularizing the remnant liver. A synthetic prosthesis (Fep Ringed Vascular Graft, Medical Products, Flagstaff, Ariz) was interposed between the 2 IVC stumps and the remaining clamps were eventually removed. There were no complications related to the placement of the graft. The patients were treated postoperatively with oral anticoagulants. In both cases, routine radiological imaging carried out during the follow-up of the resected tumors confirmed the patency of the IVC grafts after 18 and 6 months.

None of the patients died within 1 month of the operation, and all were discharged from our department. Two patients developed biliary leakage after right hepatectomies: in the first case it resolved spontaneously, while the second required the positioning of an external drainage and eventual reoperation 15 days after surgery. This latter patient also developed pleural effusion that required thoracic drainage. Three additional patients had transient liver or kidney failure; one of them required short-term dialysis. The complication rate was thus 26.3%. Mean length of postoperative stay was 19.3 +/- 13.6 days (range, 8-63 days).

Fifteen patients were operated on for malignant neoplasms. Eight had recurrence of the tumoral disease, and 2 of them underwent a second operation: a wedge hepatic resection with concomitant removal of part of the infiltrated diaphragm after the recurrence of hepatocellular carcinoma in one and hepatic lymphoadenectomy for hilar recurrence of a colorectal metastasis in the other. These 2 patients were the only ones alive at last follow-up after the development of a recurrence, 53 and 46 months after the first operation. The remaining 7 patients were alive without disease after a mean of 24.7 +/- 15.8 months (range, 6-42 months).

OTHER TECHNIQUES FOR VASCULAR CONTROL OF THE LIVER

Techniques used to achieve the control of hepatic vascular inflow during all liver resections performed in our department are summarized in ( Table 2). Of these operations, 227 (31.4%) were major hepatectomies, including 172 right or left hepatectomies and 55 extended hepatectomies. A total of 74 major hepatectomies (32.6%) were carried out after ligation of the glissonian elements for the hemiliver to be removed, without clamping the hepatic pedicle, and an additional 36 (15.8%) were performed without any control of the hepatic vascular inflow. Total vascular exclusion was used in 17.4% of all major hepatectomies.

Table 2. Techniques Used in 722 Liver Resections During 14 Years From:   Grazi: Arch Surg, Volume 132(10).October 1997.1104-1109

The number of liver resections carried out without clamping the hepatic pedicle increased in later years: the Pringle maneuver was performed in 52.0% (195/375) of the resections performed before 1992 and in 44.7% (155/347) of those performed later (P=.04). This change was accomplished by a significant reduction in the need for intraoperative transfusion: blood was given during 74.5% (279/375) of the resections performed before 1992 and in 37.2% (129/347) of those performed in more recent years (P<.001). This was achieved despite the increase in the number of extended hepatectomies performed: 21 (5.6%) of 375 during the first years in comparison with 34 (9.8%) of 347 more recently (P=.03). There was no difference in the need for intraoperative transfusion in the 227 patients undergoing major hepatectomies with respect to the use of the Pringle maneuver (48.1% in the clamped group vs 51.9% in the nonclamped group; P=.06), and even the amount of blood transfused during the operation did not differ between these 2 groups (1057.3 +/- 864.8 mL in the clamped group vs 1012.6 +/- 586.2 mL in the nonclamped group; P=.51).

COMMENT

Total vascular exclusion is a highly effective approach to the liver in hepatic surgery. This technique is the natural development of the previously described Pringle maneuver, [2] and it allows surgeons to achieve complete control of the vascular structures of the liver. [3] Its application during hepatic resections has increased in conjunction with the widespread growth of liver transplantation during the 1980s. [1] The experience gained with the intraoperative treatment of patients undergoing major liver surgery has demonstrated that TVE is applicable in many patients, without serious complications. Reports on intraoperative hemodynamic changes induced by TVE [7,5,13,14] and on postoperative hepatic and renal functions [4,5,7,8,13-18] have confirmed its safety. The reduction in cardiac output and central venous pressure, together with the compensatory increase in heart rate and peripheral vascular resistances, are well-described side effects observed a few minutes after TVE is established. [14,16]

Only 1 patient in our series did not show the hemodynamic stability needed to allow the procedure, probably as the result of the cardiac toxic effects induced by prolonged chemotherapy. Despite vigorous anesthetic support, a few patients in whom the location of the tumor dictates TVE do not demonstrate hemodynamic tolerance during the preliminary cross-clamping. [14]

With these considerations, TVE has been widely used during liver surgery in several centers, even if a major resection was not planned. [9,16,17] In most of these instances, however, the percentage of liver resections performed under TVE is not specified; thus, the role of TVE in a surgical liver center has not been well defined.

Our experience with more than 700 liver resections showed that the true need for TVE is limited. In the vast majority of cases, liver resections can be performed with different approaches to the portal pedicle, ie, Pringle maneuver or extraparenchymal ligation of the main branches of the glissonian pedicle, which allows safe resection and a substantial reduction in the use of blood derivatives. [19] In selected cases, accurate dissection of the parenchyma and meticulous search for intrahepatic blood vessels can avoid the need to control vascular inflow to the liver, without increasing intraoperative blood loss and thus the need for transfusions. Of the 227 major liver resections performed in our department, 15.8% were carried out without the need for any vascular control.

Furthermore, the only published controlled study showed no differences in terms of operative blood losses between resections performed under Pringle maneuver and TVE. [14] The same study showed a 2.5-fold increase in the postoperative complication rate in patients operated on under TVE compared with those operated on under simple pedicle clamping, resulting in a significantly longer postoperative stay in the former. [14]

On the other hand, TVE is mandatory for control of the IVC when it is infiltrated by a tumor. In 4 of our patients, it was necessary to partially resect the IVC wall and directly suture it. In 2 additional patients, the retrohepatic IVC had to be completely removed. This procedure was first described by Iwatsuki et al, [20] and few reports have been published since then. [21-23] Our choice was to reconstruct the vessels with prosthetic grafts. We did not observe any complication related to this technique, and the patency of the graft was always assessed postoperatively.

Another indication for TVE is traumatic injury of the liver. In these cases, effective control of the bleeding is needed but often cannot be obtained by conventional packing or by simple pedicle clamping. We successfully used TVE in 1 case of trauma, without major difficulties. As in all our elective procedures, the aorta was not clamped and TVE was established after hemodynamics were restored by volume resuscitation. [8] Supraceliac aortic occlusion was included in the original description of TVE [3] but met with little success in subsequent reports. [15] Cross-clamping of the aorta may be required for resuscitation in patients with severe trauma, as has been shown in the larger traumatic series, [13] but the potential hazards of the maneuver [4,15] and good anesthetic management [13,14] render this approach unnecessary in most elective cases.

We thus believe that indications for TVE should be limited to carefully selected patients [4,14,15] : those with huge masses centrally located in the liver or strictly in contact with the hepatic vein confluence, and those with malignant neoplasms closely adherent to or infiltrating the retrohepatic IVC. Despite accurate preoperative radiological workup, the decision to perform a liver resection under TVE is often taken only at laparotomy, after careful exploration of the liver and the performance of intraoperative ultrasonography. Total vascular exclusion is necessary when there is no other way of safely managing the infiltrated IVC without jeopardizing the outcome of the procedure. In fact, if the infiltrated portion of the vessel can be removed after a simple partial occlusion, we prefer to complete the operation by performing a liver resection by the standard approaches to hepatic vascular inflow rather than establishing TVE. Complete hemorrhage control is also needed to obtain safe surgical margins when resections are performed for large neoplasms. [15] In these cases, TVE gives the surgeon the feeling of full control of the liver and permits safe liver resection to be performed, even if it is extended.

We advocate the use of TVE only under particular circumstances, the most important of which concern the hospital facilities. The approach to the liver with TVE needs a well-trained anesthesiology team, highly familiar with problems related to prolonged clamping of the IVC and the portal pedicle. In fact, in most instances the necessity to completely exclude the liver from blood circulation is needed in the presence of huge tumors strictly adherent to the IVC.

The use of TVE makes it possible to extend indications for surgery to huge tumors, even if they infiltrate the IVC. Long-term results of resection under TVE for malignant neoplasms are generally not reported with the short-term results of this procedure. The experience gained by us and by others [16,17] has shown that such procedures are accomplished with low rates of morbidity and mortality when performed in specialized surgical liver units. After resections under TVE, most of our patients achieved acceptable survival, comparable to that of patients with similar tumors. In the presence of such results, a highly aggressive approach is justifiable for liver neoplasms not resectable without the use of TVE.

In conclusion, TVE is a useful tool during hepatic resection for the control of blood inflow into the liver, but its role remains limited, since most of these procedures can be performed according to the principles of anatomical surgery, with excellent results. Its performance requires a well-trained team, familiar with problems related to the surgical access to the IVC and to the prolonged occlusion of the hepatic pedicle and the IVC. The use of TVE permits an aggressive approach to tumors centrally located in the liver or in relationship with the retrohepatic IVC; in these latter cases, TVE allows partial or total resections to be performed. It can also be useful in cases of liver trauma.

Reprints: Alighieri Mazziotti, MD, Istituto di Clinica Chirurgica 2 degrees, Universita degli Studi di Bologna, Policlinico S. Orsola, Via Massarenti 9, 40138 Bologna, Italy.

1. Gozzetti G, Mazziotti A, Grazi GL. HPB surgery: an independent speciality or a branch of digestive surgery? HPB Surg. 1994;8:111-113. 
2. Pringle JH. Notes on the arrest of hepatic hemorrhage due to trauma. Ann Surg. 1908;48:541-549. 
3. Heaney JP, Stanton WK, Halbert DS, Seidel J, Vice T. An improved technic for vascular isolation of the liver: experimental study and case reports. Ann Surg. 1966;163:237-241. 
4. Huguet C, Nordlinger B, Galopin JJ, Bloch P, Gallot D. Normothermic hepatic vascular exclusion for extensive hepatectomy. Surg Gynecol Obstet. 1978;147:689-693. 
5. Delva E, Barberousse JP, Nordlinger B, et al. Hemodynamic and biochemical monitoring during major liver resection with use of hepatic vascular exclusion. Surgery. 1984;95:309-318. 
6. Huguet C, Addario-Chieco P, Gavelli A, Arrigo E, Harb J, Clement RR. Technique of hepatic vascular exclusion for extensive liver resection. Am J Surg. 1992;163:602-605. 
7. Bismuth H, Castaing D, Garden OJ. Major hepatic resection under total vascular exclusion. Ann Surg. 1989;210:13-19.
8. Stephen MS, Shaeil GR, Thompson JF, Wilson T, Boland SL. Aortic occlusion and vascular isolation allowing avascular hepatic resection. Arch Surg. 1990;125:1482-1485.
9. Terblanche J, Krige JEJ, Bornman PC. Simplified hepatic resection with the use of prolonged vascular inflow occlusion. Arch Surg. 1991;126:298-301. 
10. Couinaud C. Le foie: Etudes anatomiques et chirurgicales. New York, NY: Masson; 1975. 
11. Gozzetti G, Mazziotti A, Bolondi L, Barbara L. Intraoperative Ultrasonography in Hepato-biliary and Pancreatic Surgery. Dordrecht, the Netherlands: Kluwer Academic Publishers; 1989. 
12. Norusis MJ. SPSS/PC+ User's Guide, Version 5.0. Chicago, Ill: SPSS Inc; 1992. 
13. Emond J, Wachs ME, Renz JF, et al. Total vascular exclusion for major hepatectomy in patients with abnormal liver parenchyma. Arch Surg. 1995;130:824-831. 
14. Belghiti J, Noun R, Zante E, Ballet T, Sauvanet A. Vascular exclusion for major liver resection: a controlled study. Ann Surg. 1996;224:155-161. 
15. Emre S, Schwartz ME, Katz E, Miller CM. Liver resection under total vascular isolation: variations on a theme. Ann Surg. 1993;217:15-19. 
16. Emond JC, Kelley SD, Heffron TG, Nakagawa T, Roberts JP, Lim RC. Surgical and anesthetic management of patients undergoing major hepatectomy using total vascular exclusion. Liver Transplant Surg. 1996;2:91-98.
17. Habib N, Zografos G, Dalla Serra G, Greco L, Bean A. Liver resection with total vascular exclusion for malignant tumours. Br J Surg. 1994;81:1181-1184. 
18. Huguet C, Gavelli A. Experience with total vascular isolation of the liver. Semin Liver Dis. 1994;14:115-119. 
19. Gozzetti G, Mazziotti A, Grazi GL, et al. Liver resection without blood transfusion. Br J Surg. 1995;82:1105-1110. 
20. Iwatsuki S, Todo S, Starzl TE. Right trisegmentectomy with a synthetic vena cava graft. Arch Surg. 1988;123:1021-1022. 
21. Kumada K, Shimahara Y, Fukui K, Itoh K, Morikawa S, Ozawa K. Extended right hepatic lobectomy: combined resection of inferior vena cava and its reconstruction by EPTFE graft (Gore-Tex). Acta Chir Scand. 1988;154:481-483. 
22. Miller CM, Schwartz ME, Nishizaki T. Combined hepatic and vena caval resection with autogenous graft replacement. Arch Surg. 1991;126:106-108. 
23. Habib NA, Michail NE, Boyle T, Bean A. Resection of the inferior vena cava during hepatectomy for liver tumours. Br J Surg. 1994;81:1023-1024. 

Blood Loss, Surgical; Hepatic Pedicle; Hepatectomy; Liver Diseases; Vena Cava, Inferior