Over the past decade, new techniques and strategies have emerged to minimize the mortality and morbidity from liver resection. The 2 most significant risks germane to anesthesiologists are major blood loss and venous air embolism (VAE). Blood loss and transfusion requirements are independent variables for overall survival and tumor recurrence.1 Maneuvers to reduce blood loss include hepatic vascular occlusion, acute isovolemic hemodilution, and maintenance of a low central venous pressure (CVP).2–11 Of particular interest is the reduction of CVP to decrease congestion in hepatic veins and sinusoids to minimize resection site venous pressure, and thereby blood loss, during parenchymal dissection. This has popularized the practice of maintaining a CVP <5 mm Hg during parenchymal dissection by fluid restriction, pharmacologic intervention, patient positioning,12–14 or autologous blood donation.15
The risk of VAE increases as the CVP is decreased, and life-threatening VAE has been reported during hepatic resection despite electrocauterization, argon-enhanced coagulation, water jet dissection, ultrasonic dissection, microwave therapy, radiofrequency ablation, clamp crushing, and Cavitron Ultrasonic Surgical Aspiration® (CUSA®, Valleylab, Inc., Boulder, CO)12,14,16–22 Increasing the CVP increases the hemorrhagic risk, although decreasing the CVP introduces and increases the risk of VAE. This makes it imperative to optimize the CVP to balance these opposing considerations.
This study was initiated after making the observation that not only did the body habitus of our patient population appreciably vary, but so did the anterior-most to posterior-most (AP) diameter of the liver. Furthermore, we noticed that the AP liver diameter can be much larger than 7 cm, which is the approximate hydrostatic pressure corresponding to a CVP of 5 mm Hg (1 mm Hg = 1.36 cm H2O). Therefore, if the AP dimension of the liver is larger than 7 cm, where should we zero the CVP relative to the depth of the hepatic resection to balance these opposing risks? The purpose of this study was to characterize the liver AP diameter and thereby describe how this might affect the placement of the CVP transducer to balance the risks of bleeding and VAE.
The study protocol was approved by the University of Florida IRB, which determined that informed consent was not needed for this retrospective study. The first 100 consecutive adult chest computed tomograms (CTs) with IV contrast performed during September 2008 were included in the study. CTs were measured from the thoracic inlet superiorly through the liver inferiorly. All 100 images were deemed suitable for measurement. Axial images with a section thickness of 5 mm were retrospectively reviewed in digital format on a picture-archiving and communication system workstation.
Using electronic caliper picture-archiving and communication system tools, measurements were made from each set of axial images. Axial planes were selected delineating the AP points of the liver by scrolling through the images. Although this methodology resulted in the maximal measurement of the AP liver diameter, it is not the conventional method used in radiology, where the AP measurement is typically made on only one axial image. However, we believed that this was the most accurate method for measurement of this variable for reference use during CVP measurements. Distance measurements were made from the anatomic margin to the zero reference plane using the standardized calibration scale that was identical on each image, enabling reproducible, reliable frames of measurement throughout the study. From this, we determined a consistent zero point, and 4 axial images were identified. First, the axial image with the most anterior margin of the liver was determined. Three points were identified on this first image as follows:
- The anterior-most margin of the liver
- The anterior-most margin of the skin
- The posterior-most margin of the skin
Three additional points were then identified on axial images as follows:
- The posterior-most margin of the liver
- Portal triad at the first divergent branches of the portal vein
- Right atrium/inferior vena cava (RA/IVC) confluence at the center of the vascular lumen
Figure 1 is a representative image through the liver with 3 of the important metrics illustrated.
The results of our study demonstrate a remarkably large interindividual variability in AP liver dimensions and standardized anatomic landmarks. Table 1 shows the range of AP liver diameter and the relationship to intrahepatic and extrahepatic landmarks.
In a cohort of 100 adult subjects, the mean AP liver diameter was 17.9 cm (this means that the mean radius was 8.9 cm), a value larger than our corresponding maximal target CVP (AP liver diameter of 7 cm approximates a CVP of 5 mm Hg). In addition, if the observed range reflecting 2 standard deviations (12.3–23.5 cm) for the liver AP diameter is considered, the potentially dangerous consequences of maintaining a CVP between 0 and 5 mm Hg with an unknown AP liver diameter is appreciated.
With a large AP liver diameter, there are 3 possible placements of the transducer relative to the surgical site, as illustrated in Figure 2. In Figure 2A, the transducer is placed below the level of the surgical site. In this case, even with the transducer placed at the midlevel of the liver, the transducer can still be 11 cm beneath the operative area, which leads to a pressure measurement that is approximately 8 mm Hg less than the actual local venous pressure value at the anterior liver surface. Therefore, in this case, assuming a measured pressure of 5 mm Hg, the actual CVP at the operative site is −3 mm Hg (5 mm Hg CVP −8 mm Hg hydrostatic liver height). Although this is certainly protective against hemorrhage, it creates optimal conditions for VAE. In Figure 2B, the transducer is placed at the level of the surgical resection. In this case, with a CVP of 5 mm Hg, the risk of bleeding and VAE are minimized and balanced because the actual pressure is within the targeted value of 0 to 5 mm Hg. In Figure 2C, the transducer is placed well above the surgical site. As in Figure 2A, the transducer is placed at the midlevel of AP liver diameter, but in this example, the resection occurs 11 cm (8 mm Hg) below the transducer. If the CVP is measured at 5 mm Hg, the actual pressure at the surgical site is 13 mm Hg (5 mm Hg CVP + 8 mm Hg hydrostatic liver height). This protects against VAE, but the risk of hemorrhage increases considerably.
Traditionally, the phlebostatic point used for a CVP zero reference point is the midaxillary line between the fourth and fifth ribs.23 More recently, an external reference level at approximately four-fifths of the AP diameter of the thorax from the back has been suggested as an accurate position.24 Although the term “CVP” has been used traditionally to describe measurement of the venous pressure in the liver, we propose “hepatic venous pressure” as a more accurate term. As we have described, the venous pressure in the liver may be highly variable, depending on the AP level of the liver resection. Referring to the hepatic venous pressure rather than CVP forces the provider to recognize and address the difference between the CVP and hepatic sinusoidal pressure, which needs to be optimized during parenchymal dissection.
Given the large variability of the AP distance between the liver and other easily accessible anatomic landmarks (including the RA/IVC as shown by our measurements), the variance in depth from the anterior-most point of the abdominal skin at the level of the liver to the portal triad limits its use as a reliable and reproducible reference point.
Numerous studies support striving to achieve a CVP between 0 and 5 mm Hg to minimize blood loss and decrease the risk of VAE.4–10,12–15,25,26 However, only 3 of these studies identify a zero point reference, to which the authors refer as either “midatrial” or “right atrium”5,12,14 without specifying exactly where this measurement was referenced. The remaining studies fail to identify any zero reference point. Our data suggest that this may be a major problem because of the small margin for error when attempting to tightly maintain a CVP between 0 and 5 mm Hg. In addition, operator error could have further compounded the anatomic considerations of this study.27 The results of this study question the utility of zeroing the CVP to the traditional phlebostatic point during hepatic surgery. In fact, the CVP could be providing a false sense of security if the surgical site itself is not considered when zeroing the transducer. This cannot be done reliably from traditional phlebostatic reference points.
An additional finding from our data is the large variability of the distance from the anterior liver to the RA/IVC junction, which measured 8.5 ± 2.2 cm with a range of 2.4–12.4 mm Hg. Figure 3 is a reconstructed image developed in a sagittal plane through the RA/IVC. This illustrates that a reference transducer zeroed at the level of the RA is well below the level of most of the liver parenchyma. If the true CVP was 5 mm Hg, >86% (1SD) of patients are at risk of VAE with anterior liver resection.
One possible approach to minimize the risks of hemorrhage and VAE is to zero the transducer relative to the surgical site, which to our knowledge has not been described previously. We suggest 2 ways this can be accomplished. First, by evaluating the liver images and collaborating with the surgeon, the resection depth can be delineated. The transducer can then be placed at the approximate level of the resection. A second method involves passing a sterile transducer tubing to the surgeon and then zeroing the transducer to the anterior-most resection site. This is analogous to what is sometimes done in cardiac surgery when a transducer is zeroed to a location in the surgical field. This would provide the operative team with a more exact hepatic venous pressure at the site of resection to more precisely manage the concerns of hemorrhage and VAE.
Because of the standardized practice of positioning the arms above the head during radiographic imaging for a chest CT, we were unable to obtain the phlebostatic points located in the midaxillary line between the fourth and fifth ribs. This arm positioning altered the anatomy by displacing the phlebostatic point in an anterior and cephalad direction. Surgeons and anesthesiologists vary in their preference for arm positioning during hepatic surgery: tucked, out approximately 90°, or flexed and wedged (surrender position with elbows anterior to the ears) are all possible arm positions. This increases the variability in external landmarks based on arm position and internal landmarks based on RA/IVC position relative to a reference point.
Liver measurements performed with an intact abdomen may differ during laparotomy or laparoscopy. Surgical retraction during open hepatectomy may also slightly change the AP diameter, but in an organ with our reported mean AP diameter of 18 cm, we believe this would have minimal effect. The most appropriate zero reference level may not be constant over the course of the surgical procedure but may be dynamically affected by the use of retractors, intravascular volume status changes (relative to hepatic compliance), and the progress of the surgical procedure itself, which may require occasional transducer height readjustment, as the actual surgical site may progress along the AP axis during major resections. Kaspersen et al.28 reported a mean accuracy of 13 mm comparing perioperative CTs for guiding intraabdominal surgical procedures to intraoperative ultrasound measurements. In addition, blood volume and disease states (e.g., cirrhosis) may change the shape of the liver and slightly affect size measurements. In a laparoscopic porcine model, liver deformation associated with peritoneal insufflation in the AP plane is quite small.29 All of these various effects are on a small scale compared with the much larger variable, i.e., the AP liver diameter itself.
The significant variability in AP liver diameter coupled with a large disparity in the liver surgical site suggests that we rethink the zero reference point for CVP monitoring. By considering the actual hepatic venous pressure itself rather than a poor proxy for it (i.e., the CVP), we can minimize the risks of VAE and massive hemorrhage. This may be done by either incorporating information from the imaging studies of the liver along with collaboration with the surgeon, or by zero-referencing the transducer on the field at the anticipated apex of the surgical site.
The authors thank Amanda Smith for drawing Figure 2 and Galey Gravenstein for reconstructing Figure 3.
1. Laurent C, Blanc JF, Nobili S, Sa Cunha A, Le Bail B, Bioulac-Sage P, Balabaud C, Capdepont M, Saric J. Prognostic factors and long term survival after hepatic resection for hepatocellular carcinoma originating from noncirrhotic liver. J Am Coll Surg. 2005;201:656–62
2. Chen H, Sitzmann JV, Marcucci G, Choti MA. Acute isovolemic hemodilution during major hepatic resection—an initial report. Does it safely reduce the blood transfusion requirement? J Gastrointest Surg 1997;1:461–6
3. Arnoletti JP, Brodsky J. Reduction of transfusion requirements during major hepatic resections for metastatic disease. Surgery 1999;125:166–71
4. Melendez JA, Arslan V, Fisher ME, Wuest D, Jarnagin WR, Fong Y, Blumgart LH. Perioperative outcomes of major hepatic resections under low central venous pressure anesthesia: blood loss, blood transfusion, and the risk of postoperative renal dysfunction. J Am Coll Surg 1998;81:1181–4
5. Jones RM, Moulton CE, Hardy KJ. Central venous pressure and its effect on blood loss during liver resection. Br J Surg 1998;85:1058–60
6. Johnson M, Manner R, Wu AV. Correlation between blood loss and inferior vena caval pressure during liver resection. Br J Surg 1998;85:188–90
7. Wang WD, Liang LJ, Huang XQ, Yin XY. Low central venous pressure reduces blood loss in hepatectomy. World J Gastroenterol 2006;12:935–9
8. Niemann CU, Feiner J, Behrends M, Eilers H, Ascher N, Roberts JP. Central venous pressure monitoring during living right donor hepatectomy. Liver Transpl 2007;13:266–71
9. Liu Y, Cai M, Duan S, Peng X, Lai Y, Li Y. Effect of controlled low central venous pressure on renal function in major liver resection. Chin Ger J Clin Oncol 2008;7;1:7–9
10. Chen H, Merchant N, Didolkar M. Hepatic resection using intermittent vascular inflow occlusion and low central venous pressure anesthesia improves morbidity and mortality. J Gastrointest Surg 2000;4:162–7
11. Smyrniotis VE, Kostopanagiotou GG, Gamaletsos EL, Vassiliou JG, Voros DC, Fltopoulos AC, Contis JC. Total versus selective vascular exclusion in major liver resection. Am J Surg 2002;183:173–8
12. Moulton CA, Chui AK, Mann D, Lai PB, Chui PT, Lau WY. Does patient position during liver surgery influence the risk of venous air embolism? Am J Surg 2001;181:366–7
13. Chui AK, Moultan CE, Lau WY. Trendelenburg patient positioning: a reevaluation. J Am Coll Surg 2000;190:760–1
14. Soonawalla ZF, Stratopoulos C, Stoneham M, Wilkinson D, Britton BJ, Friend PJ. Role of the reverse-Trendelenburg patient position in maintaining low-CVP anaesthesia during liver resections. Langenbecks Arch Surg 2008;393:195–8
15. Jawan B, Cheng YF, Tseng CC, Chen YS, Wang CC, Huang TL, Engl HL, Liu PP, Chiu KW, Wang SH, Lin CC, Lin TS, Liu YW, Chen CL. Effect of autologous blood donation on the central venous pressure, blood loss and blood transfusion during living donor left hepatectomy. World J Gastroenterol 2005;11:4233–6
16. Hatano Y, Murakawa M, Segawa H, Nishida Y, Mori K. Venous air embolism during hepatic resection. Anesthesiology 1990;73:1282–5
17. Veychemans F, Michel I. Venous air embolism from an argon coagulator. Anesthesiology 1996;85:443–4
18. Palmer M, Miller CW, van Way CW III, Orton EC. Venous gas embolism associated with argon-enhanced coagulation of the liver. J Invest Surg 1993;6:391–9
19. Smith JA. Possible venous air embolism with a new water jet dissector. Br J Anaesth 1993;70:446–7
20. Lee SY, Choi BIW, Kim JS, Park KS. Paradoxical air embolism during hepatic resection. Br J Anaesth 2002;88:136–8
21. Nakayama R, Yano T, Mizutamari E, Ushijima K, Terasaki H. Possible pulmonary gas embolism associated with localized thermal therapy of the liver. Anesthesiology 2003;99:227–8
22. Koo BN, Kil HK, Choi JS, Kim JY, Chun DH, Hong YW. Hepatic resection by the Cavitron Ultrasonic Surgical Aspirator increases the incidence and severity of venous air embolism. Anesth Analg 2005;101:966–70
23. Winsor T, Burch GE. Phlebostatic axis and phlebostatic level, reference levels for venous pressure measurements in man. Proc Soc Exp Biol Med 1945;58:165–9
24. Seo JH, Jung CW, Bahk JH. Uppermost blood levels of the right and left atria in the supine position. Anesthesiology 2007;107:260–3
25. Massicotte L, Lenis S, Thibeault L, Sassine MP, Seal RF, Roy A. Effect of low central venous pressure and phlebotomy on blood product transfusion requirements during liver transplantations. Liver Transpl 2006;12:117–23
26. Smyrniotis V, Kostopanagiotou G, Theodoraki K, Tsantoulas D, Contis JC. The role of central venous pressure and type of vascular control in blood loss during major liver resections. Am J Surg 2004;187:398–402
27. Figg KK, Nemergut EC. Error in central venous pressure measurement. Anesth Analg 2009;108:1209–11
28. Kaspersen JH, Sjølie E, Wesche J, Asland J, Lundbom J, Odegård A, Lindseth F, Nagelhus Hernes TA. Three-dimensional ultrasound-based navigation combined with preoperative CT during abdominal interventions: a feasibility study. Cardiovasc Intervent Radiol 2003;26:347–56
29. Rauth RP, Bao PQ, Galloway RL, Bieszczad J, Friets EM, Knaus DA, Kynor DB, Herline AJ. Laparoscopic surface scanning and subsurface targeting: implications for image-guided laparoscopic liver surgery. Surgery 2007;142:207–14