- Email: [email protected]
Right inferior phrenic vein indicating the right hepatic vein confluence into the inferior vena cava
The American Journal of Surgery 192 (2006) 690 – 694
How I do it
Right inferior phrenic vein indicating the right hepatic vein confluence into the inferior vena cava Guido Torzilli, M.D., Ph.D.a,b,*, Marco Montorsi, M.D.a, Angela Palmisano, M.D.a, Daniele Del Fabbro, M.D.a, Andrea Gambetti, M.D.b, Matteo Donadon, M.D.a, Natale Olivari, M.D.b, Masatoshi Makuuchi, M.D., Ph.D.c a
3rd Department of Surgery, Istituto Clinico Humanitas, IRCCS, University of Milan, Via Manzoni, 56, I-20089 Rozzano, Milano, Italy Hepatobiliary Surgery Unit, 1st Department of Surgery, Ospedale Maggiore di Lodi, Azienda Ospedaliera della Provincia di Lodi, Lodi, Italy c Hepato-Biliary-Pancreatic Surgery Division, Artificial Organ and Transplantation Division, Department of Surgery, Graduate School of Medicine, University of Tokyo, Tokyo, Japan b
Manuscript received June 27, 2005; revised manuscript October 26, 2005
Abstract Background: Limiting backflow bleeding from the hepatic veins is a priority when performing hepatectomy. However, hepatic vein encirclement is difficult, especially in re-resection. We verified the presence and trajectory of the right inferior phrenic vein (RIPV), which could be a useful anatomic landmark to guide surgeons in targeting the extrahepatic right hepatic vein (RHV) before dissection. Methods: Between May 2001 and January 2005, 100 consecutive patients with liver tumors were enrolled and underwent hepatectomy: 77 patients underwent surgery for tumors located in the right hemiliver. Results: RIPV was detected in all but 1 patient (99%), and its trajectory was always guided toward the extrahepatic RHV. The only patient in whom RIPV was not detected had undergone prior liver resection and interstitial therapies for colorectal cancer liver metastases. Conclusions: Apart from exceptional conditions, detection of the RIPV is always feasible and allows safe surgical dissection while approaching the extrahepatic RHV before hepatic resection. © 2006 Excerpta Medica Inc. All rights reserved. Keywords: Liver anatomy; Liver surgery; Liver neoplasms; Inferior vena cava; Hepatic vein; Phrenic vein
Avoiding excessive bleeding from hepatic veins is a priority in liver resections because it represents a major source of hemorrhage, which is a well-known cause of impairment in patient outcome and prognosis [1–3]. For this purpose, during right-sided liver resections such as right hepatectomy or resections involving segments 7 and 8, clamping of the right hepatic vein (RHV) at its caval has been proposed [4]. However, this approach necessitates encirclement of the RHV at its caval confluence before liver transection, which often is a difficult and risky maneuver, especially in re-resections, and some investigators prefer to perform the liver resection without it. Our experience has shown that the right inferior phrenic vein (RIPV) could be a useful anatomic landmark for guiding surgeons to detect and target the site of RHV confluence * Corresponding author. Tel.: ⫹39-02-8224-4083; fax: ⫹39-02-8224-4590. E-mail address: [email protected] or [email protected]
into the inferior vena cava (IVC). We have estimated the prevalence of this empiric observation on a prospective cohort of patients who underwent liver surgery for rightsided hepatic tumors and analyzed its usefulness in performing this surgery. Patients and Methods Between May 2001 and January 2005, 100 consecutive patients with liver tumors were enrolled. All patients underwent liver resection, and of these patients 77 underwent surgery for tumors located in the right hemiliver. These patients were considered for the study. Patient characteristics including type, size, and number of tumors are shown in Table 1. All but 1 tumor were malignant and 29 of these tumors were metastatic neoplasms. Twenty-one patients had undergone prior treatment in other centers: 7 patients had received percutaneous thermal ablation, 7 had undergone liver resection, 3 had received chemoembolization, and 4 had undergone both hepatectomy
0002-9610/06/$ – see front matter © 2006 Excerpta Medica Inc. All rights reserved. doi:10.1016/j.amjsurg.2005.10.022
G. Torzilli et al. / The American Journal of Surgery 192 (2006) 690 – 694 Table 1 Clinical characteristics of patients Character Sex M F M/F ratio Age Mean Median Range Type of tumor Hepatocellular carcinoma Colorectal liver metastases Breast cancer liver metastases Primary neuroendocrine tumor GIST liver metastases Leiomyosarcoma liver metastases Multiple hemangiomas Background liver Cirrhosis Chronic hepatitis Fatty liver Normal liver Tumor size, cm Mean Median Range Number of lesions Mean Median Range
Overall cases (77 patients) 51 26 2/1 65.3 66.5 36–79 44 26 1 1 1 1 1 38 7 28 4 4.7 4.5 1.4–10 2.2 2 1–8
GIST ⫽ gastrointestinal stromal tumor.
and percutaneous thermal ablation. The previous hepatectomies were right-sided atypical resections in 8 (3 of them also had percutaneous thermal ablation of local recurrences in resection sites), 2 left-sided atypical resections, and 1 left lobectomy (this patient also had undergone percutaneous thermal ablation of a lesion in segment 7). All 7 patients with previous percutaneous thermal ablation only, and the 3 patients who had undergone chemoembolization underwent these treatments for right-sided tumors. Overall, 19 patients had received previous right-sided local treatments and in 8 of these patients it consisted of liver resection. Once the round, falciform ligaments were divided and the liver was lowered manually by the surgeon, the RIPV was located and mobilization of the right liver and hepatocaval confluence was achieved. The rate of detection of the RIPV and RHV confluence into the IVC following the RIPV route was the study end point. The RIPV was never skeletonized, it was either visible on the diaphragmatic surface or it was not. Surgical procedure J-shaped or inverted-T laparotomies were performed routinely. For those patients with tumors involving segments 1, 4 superior, 7, and 8, which are close to the hepatic vein confluence into the IVC, a J-shaped thoracophrenolaparotomy was considered.
691
Liver dissection was accomplished with intermittent clamping by the Pringle maneuver, using Pean forceps and a bipolar electrocautery for vessel coagulation. Each vessel that was thicker than 2 mm was ligated with thin (2-0 or 3-0) sutures. Preconditioning and total vascular exclusion were not performed. The cut surface of the liver was secured by electrocautery, fibrillar-oxidated regenerated cellulose, and fibrin glue. In cirrhotic patients, blood loss and ascite production during surgery were regulated by infusing 10% to 20% more fresh-frozen plasma than the volume of blood lost. Intraoperative blood transfusions were given only when the hematocrit level was less than 30%. A total volume of 4 to 5 mL/kg/h of fluid was infused; equal volumes of crystalloid and 5% to 10% glucose solution were administered. The level of anesthesia was maintained by general and epidural anesthesia, thereby reducing the quantity of inhalation agents and intravenous drugs. The aforementioned fluid restriction and reduction of the respiratory tidal volume to around 60% just before starting liver dissection were the techniques adopted to decrease the thoracic and right atrial pressures, keeping the central venous pressure between 0 and 4 cm H2O, consequently limiting the backflow bleeding from the hepatic veins and/or their tributaries. Hydrocortisone (100 mg) was injected intravenously before starting vascular occlusion to protect the liver during warm ischemia. Results The surgical procedures used are shown in Table 2 and all surgical procedures were performed by the same surgeon (G.T.). Forty-one patients had tumors located in segments 1, 4 superior, 7, and 8. Hepatic veins were involved in 20 patients, and segmental or sectorial portal branches were involved in 19 patients. The RIPV was evident in all but 1 patient (99%), and in all the 76 patients in whom RIPV was visible it ran from the right to the left and from the anterior to the posterior part of the diaphragm (Figs. 1 and 2). In particular, RIPV appeared to flow into the IVC just cranially to the upper-right quarter of the RHV circumference, however, the last 1 to 3 cm of the RIPV were never visible on the diaphragmatic surface. In 22 patients the RHV was skeletonized and taped at its caval confluence (Fig. 3), and in none of the patients did the
Table 2 Type of surgical procedures adopted Type of surgery Major resection (ⱖ2 segments) Right hepatectomy Extended right hepatectomy Extended right anterior sectoriectomy Extended right posterior sectoriectomy Bisegmentectomy (7⫹8) Total Minor resection (⬍2 segments) Single Multiple Total
Overall cases (77 patients)
Cirrhotic (38 patients)
3 2 1 1 3 10
3 0 0 1 0 4
50 17 67
26 8 34
692
G. Torzilli et al. / The American Journal of Surgery 192 (2006) 690 – 694
Fig. 1. The RIPV that flows into the IVC at the level of the RHV confluence into the IVC itself is indicated by the black arrows. D, diaphragm; L, liver.
RIPV flow into the RHV. The RIPV did not show a trajectory crossing over the RHV caval confluence to the RHV left side in any patient; in particular, the blunt dissection of the fossa between the middle hepatic vein and RHV caval confluence was obtained in all patients and in none of the patients was the RIPV visible at that level. The only patient in whom the RIPV was not visible had undergone a previous resection of a colorectal cancer liver metastases located in segment 7 adjacent to the RHV caval confluence. This lesion recurred and the patient was treated with percutaneous thermal ablation. The
Fig. 2. During dissection of the retrohepatic IVC and ligation of Makuuchi’s ligament [7], the RIPV is well visible (black arrows) as a constant landmark to which the dissection should be directed. D, diaphragm; L, liver; IVC, inferior vena cava; ML, Makuuchi’s ligament.
Fig. 3. Trajectory of the RIPV (black arrows) fits well with the point of confluence of the RHV into the IVC. In this patient the extrahepatic RHV was skeletonized and taped. D, diaphragm; L, liver.
patient was referred to our department because of further local recurrence and for the coexistence of other nodules and underwent a right hepatectomy. Because of diaphragmatic fibrosis, the RIPV was not visible. The tight adhesions obligated us to perform a right hepatectomy by an anterior approach with the aid of the hanging maneuver [5,6]. We did not find any significant change in visibility of the RIPV once the aforementioned techniques for controlling the central venous pressure were adopted. No hospital mortality was seen and patients returned to their normal daily life. The overall morbidity rate was 18%. No repeat surgeries were required. Morbidity consisted of 1 pneumothorax after the insertion of a central venous catheter, 1 pulmonary embolism, 1 pleural effusion that required a thoracentesis, 7 patients with transient ascites, and 6 patients with superficial wound dehiscence. All cases of transient postoperative ascites occurred in cirrhotic patients. In 61 patients J-shaped laparotomies were performed, and in the remaining 16 patients a J-shaped thoracophrenolaparotomy was performed, entering the chest cavity through the 9th intercostal space. In these 16 patients with a J-shaped thoracophrenolaparotomy the phrenotomy did not require ligation or division of the RIPV. The mean clamping time was 59 minutes (median, 60 min; range, 25–178 min). The overall mean blood loss was 310 mL (median, 250 mL; range, 20 –900 mL). Blood transfusions were administered in 6 patients (8%). The overall mean surgical time was 378 minutes (median, 360 min; range, 180 – 635 min). In 3 patients, each with colorectal cancer liver metastases located in segments 7, 7-8, and 7-8, respectively, the diaphragm was resected because of tumor infiltration. The mean hospital stay was 9.8 days (median, 10 days; range, 8 –12 days).
G. Torzilli et al. / The American Journal of Surgery 192 (2006) 690 – 694
Fig. 4. If the extrahepatic RHV is not encircled and taped then the RIPV (black arrows) indicates its confluence into the IVC and drives the surgeon’s finger positioning for performing a proper compression of the RHV itself when needed during liver dissection. D, diaphragm; L, liver.
Comments The limitation of blood loss and consequent avoidance of blood transfusion are factors commonly accepted as able to improve patients’ outcome and prognosis [1–3]. In this sense, limitation of bleeding from the hepatic veins is one of the key factors. For this purpose, Cherqui et al [4] proposed encirclement and clamping of the hepatic veins during liver dissection. During right-sided liver resections other hepatic surgeons place their left hand behind the mobilized right liver [7]: in this way, during hepatic dissection the liver can be pulled up by the surgeon’s left hand, with the aim of reducing the backflow bleeding by compression. However, encirclement of the RHV at its caval confluence could be difficult, especially in those patients who undergo a repeat resection and/or have had previous interstitial therapies. For this reason other surgeons have proposed different approaches such as the anterior approach, [5] with or without the association with the so-called hanging maneuver as described by Belghiti et al [6]. However, an important key for approaching the extrahepatic RHV safely either at first resection or during a repeat resection is the recognition of the RIPV. As we have stated, this vein was visible in all but 1 patient, and, when visible, it ran into the IVC just cranially to the RHV caval confluence, at the upper-right quarter of the RHV circumference (Figs. 3 and 4). RHV confluence into the RHV is therefore almost always detectable before starting the mobilization of the right liver. This way the surgeon knows where the extrahepatic RHV is in every phase of the liver mobilization. Furthermore, in the 71% of our patients in whom the extrahepatic RHV was not skeletonized and taped, the RIPV trajectory indicated where the RHV caval confluence was located (Fig. 4), which could be useful for guiding finger compression should accidental bleeding from RHV tributaries occur during
693
parenchymal dissection. The RHV was encircled in only 29% of patients, in 53% of patients the tumors were located in segments 1, 4 superior, 7, and 8, and in 51% of patients the tumor invaded portal branches (25%) and/or hepatic veins (26%), but the rate of blood transfusion was very low. Among those patients who received blood transfusion, 2 had the extrahepatic RHV encircled, and the other 4 did not. Of these last 4 patients, it should be noted that 1 underwent surgery for peritoneal hemorrhage owing to a ruptured hepatocellular carcinoma, and 2 had surgery as salvage therapy after repeated but not radical thermal ablation. In part, the low rate of blood transfusion could be a result of our policy on the restrictive use of blood transfusions [8]. Certainly, this low rate also is a result of the low number of major hepatectomies (13%) (Table 2). However, minor hepatectomies were not related to a low complexity of the patients treated in the present series, but to a well-established and previously published policy [9] based on the extensive use of ultrasound guidance for performing parenchymal-sparing liver dissection. Indeed, as previously mentioned, the rate of patients with tumors located in segments 1, 4 superior, 7, and 8, the rate of those lesions with vascular involvement, mean tumor size, and mean tumor number are consistent with those of major series and should rule out the risk for selection bias. The patient who had no visible RIPV also had a situation that did not allow a right liver mobilization with encirclement or division of the extrahepatic RHV. This patient underwent a right hepatectomy by an anterior approach after the hanging maneuver was accomplished. The RHV was divided at the end of the parenchymal dissection: therefore, knowledge of the RIPV route would have been useless. On the other hand, this patient is the only one among the 19 patients with prior right-sided local treatments (surgery, interstitial therapies, and chemoembolization) in whom the RIPV was undetectable. This fact confirms the usefulness of this landmark in this type of patient. We did not check the exact pattern of RIPV confluence into the IVC by skeletonization of this vessel. However, the end point of the present study was to estimate the prevalence of the RIPV targeting the extrahepatic RHV and to analyze the positive consequences. We did not consider the addition of any source of morbidity for showing the RIPV at its caval confluence to be justified ethically. Conversely, a cadaveric analysis of this hopefully could be the subject of further studies. In conclusion, the present series allows us to affirm that, apart from rare exceptions, simple detection of the RIPV is always feasible and represents an important landmark for an easier approach to the extrahepatic RHV before hepatic resection. Recognition of this anatomic landmark should be considered as a fundamental step during any right-sided hepatectomy. References [1] Makuuchi M, Takayama T, Gunven P, et al. Restrictive versus liberal blood transfusion policy for hepatectomies in cirrhotic patients. World J Surg 1989;13:644 – 8. [2] Yamamoto J, Kosuge T, Takayama T, et al. Perioperative blood transfusion promotes recurrence of hepatocellular carcinoma after hepatectomy. Surgery 1994;115:303–9. [3] Kooby DA, Stockman J, Ben-Porat L, et al. Influence of transfusions on perioperative and long-term outcome in patients following hepatic resection for colorectal metastases. Ann Surg 2003;237:860 –9.
694
G. Torzilli et al. / The American Journal of Surgery 192 (2006) 690 – 694
[4] Cherqui D, Malassagne B, Colau PI, et al. Hepatic vascular exclusion with preservation of the caval flow for liver resections. Ann Surg 1999;230:24 –30. [5] Lai EC, Fan ST, Lo CM, et al. Anterior approach for difficult major right hepatectomy. World J Surg 1996;20:314 –7. [6] Belghiti J, Guevara OA, Noun R, et al. Liver hanging maneuver: a safe approach to right hepatectomy without liver mobilization. J Am Coll Surg 2001;193:109 –11.
[7] Makuuchi M, Yamamoto J, Takayama T, et al. Extrahepatic division of the right hepatic vein in hepatectomy. Hepatogastroenterology 1991;38:176–9. [8] Torzilli G, Gambetti A, Del Fabbro D, et al. Techniques for hepatectomies without blood transfusion focusing on interpretation of postoperative anemia. Arch Surg 2004;139:1061–5. [9] Torzilli G, Montorsi M, Donadon M, et al. “Radical but conservative” the main goal for ultrasound guided liver resection: a prospective analysis of our experience. J Am Coll Surg 2005;201:517–28.
How I do it
Right inferior phrenic vein indicating the right hepatic vein confluence into the inferior vena cava Guido Torzilli, M.D., Ph.D.a,b,*, Marco Montorsi, M.D.a, Angela Palmisano, M.D.a, Daniele Del Fabbro, M.D.a, Andrea Gambetti, M.D.b, Matteo Donadon, M.D.a, Natale Olivari, M.D.b, Masatoshi Makuuchi, M.D., Ph.D.c a
3rd Department of Surgery, Istituto Clinico Humanitas, IRCCS, University of Milan, Via Manzoni, 56, I-20089 Rozzano, Milano, Italy Hepatobiliary Surgery Unit, 1st Department of Surgery, Ospedale Maggiore di Lodi, Azienda Ospedaliera della Provincia di Lodi, Lodi, Italy c Hepato-Biliary-Pancreatic Surgery Division, Artificial Organ and Transplantation Division, Department of Surgery, Graduate School of Medicine, University of Tokyo, Tokyo, Japan b
Manuscript received June 27, 2005; revised manuscript October 26, 2005
Abstract Background: Limiting backflow bleeding from the hepatic veins is a priority when performing hepatectomy. However, hepatic vein encirclement is difficult, especially in re-resection. We verified the presence and trajectory of the right inferior phrenic vein (RIPV), which could be a useful anatomic landmark to guide surgeons in targeting the extrahepatic right hepatic vein (RHV) before dissection. Methods: Between May 2001 and January 2005, 100 consecutive patients with liver tumors were enrolled and underwent hepatectomy: 77 patients underwent surgery for tumors located in the right hemiliver. Results: RIPV was detected in all but 1 patient (99%), and its trajectory was always guided toward the extrahepatic RHV. The only patient in whom RIPV was not detected had undergone prior liver resection and interstitial therapies for colorectal cancer liver metastases. Conclusions: Apart from exceptional conditions, detection of the RIPV is always feasible and allows safe surgical dissection while approaching the extrahepatic RHV before hepatic resection. © 2006 Excerpta Medica Inc. All rights reserved. Keywords: Liver anatomy; Liver surgery; Liver neoplasms; Inferior vena cava; Hepatic vein; Phrenic vein
Avoiding excessive bleeding from hepatic veins is a priority in liver resections because it represents a major source of hemorrhage, which is a well-known cause of impairment in patient outcome and prognosis [1–3]. For this purpose, during right-sided liver resections such as right hepatectomy or resections involving segments 7 and 8, clamping of the right hepatic vein (RHV) at its caval has been proposed [4]. However, this approach necessitates encirclement of the RHV at its caval confluence before liver transection, which often is a difficult and risky maneuver, especially in re-resections, and some investigators prefer to perform the liver resection without it. Our experience has shown that the right inferior phrenic vein (RIPV) could be a useful anatomic landmark for guiding surgeons to detect and target the site of RHV confluence * Corresponding author. Tel.: ⫹39-02-8224-4083; fax: ⫹39-02-8224-4590. E-mail address: [email protected] or [email protected]
into the inferior vena cava (IVC). We have estimated the prevalence of this empiric observation on a prospective cohort of patients who underwent liver surgery for rightsided hepatic tumors and analyzed its usefulness in performing this surgery. Patients and Methods Between May 2001 and January 2005, 100 consecutive patients with liver tumors were enrolled. All patients underwent liver resection, and of these patients 77 underwent surgery for tumors located in the right hemiliver. These patients were considered for the study. Patient characteristics including type, size, and number of tumors are shown in Table 1. All but 1 tumor were malignant and 29 of these tumors were metastatic neoplasms. Twenty-one patients had undergone prior treatment in other centers: 7 patients had received percutaneous thermal ablation, 7 had undergone liver resection, 3 had received chemoembolization, and 4 had undergone both hepatectomy
0002-9610/06/$ – see front matter © 2006 Excerpta Medica Inc. All rights reserved. doi:10.1016/j.amjsurg.2005.10.022
G. Torzilli et al. / The American Journal of Surgery 192 (2006) 690 – 694 Table 1 Clinical characteristics of patients Character Sex M F M/F ratio Age Mean Median Range Type of tumor Hepatocellular carcinoma Colorectal liver metastases Breast cancer liver metastases Primary neuroendocrine tumor GIST liver metastases Leiomyosarcoma liver metastases Multiple hemangiomas Background liver Cirrhosis Chronic hepatitis Fatty liver Normal liver Tumor size, cm Mean Median Range Number of lesions Mean Median Range
Overall cases (77 patients) 51 26 2/1 65.3 66.5 36–79 44 26 1 1 1 1 1 38 7 28 4 4.7 4.5 1.4–10 2.2 2 1–8
GIST ⫽ gastrointestinal stromal tumor.
and percutaneous thermal ablation. The previous hepatectomies were right-sided atypical resections in 8 (3 of them also had percutaneous thermal ablation of local recurrences in resection sites), 2 left-sided atypical resections, and 1 left lobectomy (this patient also had undergone percutaneous thermal ablation of a lesion in segment 7). All 7 patients with previous percutaneous thermal ablation only, and the 3 patients who had undergone chemoembolization underwent these treatments for right-sided tumors. Overall, 19 patients had received previous right-sided local treatments and in 8 of these patients it consisted of liver resection. Once the round, falciform ligaments were divided and the liver was lowered manually by the surgeon, the RIPV was located and mobilization of the right liver and hepatocaval confluence was achieved. The rate of detection of the RIPV and RHV confluence into the IVC following the RIPV route was the study end point. The RIPV was never skeletonized, it was either visible on the diaphragmatic surface or it was not. Surgical procedure J-shaped or inverted-T laparotomies were performed routinely. For those patients with tumors involving segments 1, 4 superior, 7, and 8, which are close to the hepatic vein confluence into the IVC, a J-shaped thoracophrenolaparotomy was considered.
691
Liver dissection was accomplished with intermittent clamping by the Pringle maneuver, using Pean forceps and a bipolar electrocautery for vessel coagulation. Each vessel that was thicker than 2 mm was ligated with thin (2-0 or 3-0) sutures. Preconditioning and total vascular exclusion were not performed. The cut surface of the liver was secured by electrocautery, fibrillar-oxidated regenerated cellulose, and fibrin glue. In cirrhotic patients, blood loss and ascite production during surgery were regulated by infusing 10% to 20% more fresh-frozen plasma than the volume of blood lost. Intraoperative blood transfusions were given only when the hematocrit level was less than 30%. A total volume of 4 to 5 mL/kg/h of fluid was infused; equal volumes of crystalloid and 5% to 10% glucose solution were administered. The level of anesthesia was maintained by general and epidural anesthesia, thereby reducing the quantity of inhalation agents and intravenous drugs. The aforementioned fluid restriction and reduction of the respiratory tidal volume to around 60% just before starting liver dissection were the techniques adopted to decrease the thoracic and right atrial pressures, keeping the central venous pressure between 0 and 4 cm H2O, consequently limiting the backflow bleeding from the hepatic veins and/or their tributaries. Hydrocortisone (100 mg) was injected intravenously before starting vascular occlusion to protect the liver during warm ischemia. Results The surgical procedures used are shown in Table 2 and all surgical procedures were performed by the same surgeon (G.T.). Forty-one patients had tumors located in segments 1, 4 superior, 7, and 8. Hepatic veins were involved in 20 patients, and segmental or sectorial portal branches were involved in 19 patients. The RIPV was evident in all but 1 patient (99%), and in all the 76 patients in whom RIPV was visible it ran from the right to the left and from the anterior to the posterior part of the diaphragm (Figs. 1 and 2). In particular, RIPV appeared to flow into the IVC just cranially to the upper-right quarter of the RHV circumference, however, the last 1 to 3 cm of the RIPV were never visible on the diaphragmatic surface. In 22 patients the RHV was skeletonized and taped at its caval confluence (Fig. 3), and in none of the patients did the
Table 2 Type of surgical procedures adopted Type of surgery Major resection (ⱖ2 segments) Right hepatectomy Extended right hepatectomy Extended right anterior sectoriectomy Extended right posterior sectoriectomy Bisegmentectomy (7⫹8) Total Minor resection (⬍2 segments) Single Multiple Total
Overall cases (77 patients)
Cirrhotic (38 patients)
3 2 1 1 3 10
3 0 0 1 0 4
50 17 67
26 8 34
692
G. Torzilli et al. / The American Journal of Surgery 192 (2006) 690 – 694
Fig. 1. The RIPV that flows into the IVC at the level of the RHV confluence into the IVC itself is indicated by the black arrows. D, diaphragm; L, liver.
RIPV flow into the RHV. The RIPV did not show a trajectory crossing over the RHV caval confluence to the RHV left side in any patient; in particular, the blunt dissection of the fossa between the middle hepatic vein and RHV caval confluence was obtained in all patients and in none of the patients was the RIPV visible at that level. The only patient in whom the RIPV was not visible had undergone a previous resection of a colorectal cancer liver metastases located in segment 7 adjacent to the RHV caval confluence. This lesion recurred and the patient was treated with percutaneous thermal ablation. The
Fig. 2. During dissection of the retrohepatic IVC and ligation of Makuuchi’s ligament [7], the RIPV is well visible (black arrows) as a constant landmark to which the dissection should be directed. D, diaphragm; L, liver; IVC, inferior vena cava; ML, Makuuchi’s ligament.
Fig. 3. Trajectory of the RIPV (black arrows) fits well with the point of confluence of the RHV into the IVC. In this patient the extrahepatic RHV was skeletonized and taped. D, diaphragm; L, liver.
patient was referred to our department because of further local recurrence and for the coexistence of other nodules and underwent a right hepatectomy. Because of diaphragmatic fibrosis, the RIPV was not visible. The tight adhesions obligated us to perform a right hepatectomy by an anterior approach with the aid of the hanging maneuver [5,6]. We did not find any significant change in visibility of the RIPV once the aforementioned techniques for controlling the central venous pressure were adopted. No hospital mortality was seen and patients returned to their normal daily life. The overall morbidity rate was 18%. No repeat surgeries were required. Morbidity consisted of 1 pneumothorax after the insertion of a central venous catheter, 1 pulmonary embolism, 1 pleural effusion that required a thoracentesis, 7 patients with transient ascites, and 6 patients with superficial wound dehiscence. All cases of transient postoperative ascites occurred in cirrhotic patients. In 61 patients J-shaped laparotomies were performed, and in the remaining 16 patients a J-shaped thoracophrenolaparotomy was performed, entering the chest cavity through the 9th intercostal space. In these 16 patients with a J-shaped thoracophrenolaparotomy the phrenotomy did not require ligation or division of the RIPV. The mean clamping time was 59 minutes (median, 60 min; range, 25–178 min). The overall mean blood loss was 310 mL (median, 250 mL; range, 20 –900 mL). Blood transfusions were administered in 6 patients (8%). The overall mean surgical time was 378 minutes (median, 360 min; range, 180 – 635 min). In 3 patients, each with colorectal cancer liver metastases located in segments 7, 7-8, and 7-8, respectively, the diaphragm was resected because of tumor infiltration. The mean hospital stay was 9.8 days (median, 10 days; range, 8 –12 days).
G. Torzilli et al. / The American Journal of Surgery 192 (2006) 690 – 694
Fig. 4. If the extrahepatic RHV is not encircled and taped then the RIPV (black arrows) indicates its confluence into the IVC and drives the surgeon’s finger positioning for performing a proper compression of the RHV itself when needed during liver dissection. D, diaphragm; L, liver.
Comments The limitation of blood loss and consequent avoidance of blood transfusion are factors commonly accepted as able to improve patients’ outcome and prognosis [1–3]. In this sense, limitation of bleeding from the hepatic veins is one of the key factors. For this purpose, Cherqui et al [4] proposed encirclement and clamping of the hepatic veins during liver dissection. During right-sided liver resections other hepatic surgeons place their left hand behind the mobilized right liver [7]: in this way, during hepatic dissection the liver can be pulled up by the surgeon’s left hand, with the aim of reducing the backflow bleeding by compression. However, encirclement of the RHV at its caval confluence could be difficult, especially in those patients who undergo a repeat resection and/or have had previous interstitial therapies. For this reason other surgeons have proposed different approaches such as the anterior approach, [5] with or without the association with the so-called hanging maneuver as described by Belghiti et al [6]. However, an important key for approaching the extrahepatic RHV safely either at first resection or during a repeat resection is the recognition of the RIPV. As we have stated, this vein was visible in all but 1 patient, and, when visible, it ran into the IVC just cranially to the RHV caval confluence, at the upper-right quarter of the RHV circumference (Figs. 3 and 4). RHV confluence into the RHV is therefore almost always detectable before starting the mobilization of the right liver. This way the surgeon knows where the extrahepatic RHV is in every phase of the liver mobilization. Furthermore, in the 71% of our patients in whom the extrahepatic RHV was not skeletonized and taped, the RIPV trajectory indicated where the RHV caval confluence was located (Fig. 4), which could be useful for guiding finger compression should accidental bleeding from RHV tributaries occur during
693
parenchymal dissection. The RHV was encircled in only 29% of patients, in 53% of patients the tumors were located in segments 1, 4 superior, 7, and 8, and in 51% of patients the tumor invaded portal branches (25%) and/or hepatic veins (26%), but the rate of blood transfusion was very low. Among those patients who received blood transfusion, 2 had the extrahepatic RHV encircled, and the other 4 did not. Of these last 4 patients, it should be noted that 1 underwent surgery for peritoneal hemorrhage owing to a ruptured hepatocellular carcinoma, and 2 had surgery as salvage therapy after repeated but not radical thermal ablation. In part, the low rate of blood transfusion could be a result of our policy on the restrictive use of blood transfusions [8]. Certainly, this low rate also is a result of the low number of major hepatectomies (13%) (Table 2). However, minor hepatectomies were not related to a low complexity of the patients treated in the present series, but to a well-established and previously published policy [9] based on the extensive use of ultrasound guidance for performing parenchymal-sparing liver dissection. Indeed, as previously mentioned, the rate of patients with tumors located in segments 1, 4 superior, 7, and 8, the rate of those lesions with vascular involvement, mean tumor size, and mean tumor number are consistent with those of major series and should rule out the risk for selection bias. The patient who had no visible RIPV also had a situation that did not allow a right liver mobilization with encirclement or division of the extrahepatic RHV. This patient underwent a right hepatectomy by an anterior approach after the hanging maneuver was accomplished. The RHV was divided at the end of the parenchymal dissection: therefore, knowledge of the RIPV route would have been useless. On the other hand, this patient is the only one among the 19 patients with prior right-sided local treatments (surgery, interstitial therapies, and chemoembolization) in whom the RIPV was undetectable. This fact confirms the usefulness of this landmark in this type of patient. We did not check the exact pattern of RIPV confluence into the IVC by skeletonization of this vessel. However, the end point of the present study was to estimate the prevalence of the RIPV targeting the extrahepatic RHV and to analyze the positive consequences. We did not consider the addition of any source of morbidity for showing the RIPV at its caval confluence to be justified ethically. Conversely, a cadaveric analysis of this hopefully could be the subject of further studies. In conclusion, the present series allows us to affirm that, apart from rare exceptions, simple detection of the RIPV is always feasible and represents an important landmark for an easier approach to the extrahepatic RHV before hepatic resection. Recognition of this anatomic landmark should be considered as a fundamental step during any right-sided hepatectomy. References [1] Makuuchi M, Takayama T, Gunven P, et al. Restrictive versus liberal blood transfusion policy for hepatectomies in cirrhotic patients. World J Surg 1989;13:644 – 8. [2] Yamamoto J, Kosuge T, Takayama T, et al. Perioperative blood transfusion promotes recurrence of hepatocellular carcinoma after hepatectomy. Surgery 1994;115:303–9. [3] Kooby DA, Stockman J, Ben-Porat L, et al. Influence of transfusions on perioperative and long-term outcome in patients following hepatic resection for colorectal metastases. Ann Surg 2003;237:860 –9.
694
G. Torzilli et al. / The American Journal of Surgery 192 (2006) 690 – 694
[4] Cherqui D, Malassagne B, Colau PI, et al. Hepatic vascular exclusion with preservation of the caval flow for liver resections. Ann Surg 1999;230:24 –30. [5] Lai EC, Fan ST, Lo CM, et al. Anterior approach for difficult major right hepatectomy. World J Surg 1996;20:314 –7. [6] Belghiti J, Guevara OA, Noun R, et al. Liver hanging maneuver: a safe approach to right hepatectomy without liver mobilization. J Am Coll Surg 2001;193:109 –11.
[7] Makuuchi M, Yamamoto J, Takayama T, et al. Extrahepatic division of the right hepatic vein in hepatectomy. Hepatogastroenterology 1991;38:176–9. [8] Torzilli G, Gambetti A, Del Fabbro D, et al. Techniques for hepatectomies without blood transfusion focusing on interpretation of postoperative anemia. Arch Surg 2004;139:1061–5. [9] Torzilli G, Montorsi M, Donadon M, et al. “Radical but conservative” the main goal for ultrasound guided liver resection: a prospective analysis of our experience. J Am Coll Surg 2005;201:517–28.