Journal of Hepatology
Volume 41, Issue 2 , Pages 340-343, August 2004

Microvascular graft dysfunction

Fédérations des spécialités digestives CHU Bordeaux, GREF/INSERM E 0362, Université Bordeaux 2, Bordeaux cedex 33075, France

See Article, pages 299–306

Article Outline

 

The damage to the liver caused by ischemia and reperfusion (I/R) represents a continuum of processes that culminate in hepatocellular injury. These processes are triggered when the liver is transiently deprived of oxygen and reoxygenated, and can occur in a number of clinical settings, such as those during the organ procurement for transplantation.

Although the importance of I/R injury to the success in solid organ transplantation has been recognized for more than 40 years, the major clinical advances in organ preservation were deduced empirically, with an incomplete understanding of the molecular mechanisms involved in the injury itself.

I/R injury contributes to the acute shortage of livers available for transplantation because of the higher susceptibility of marginal livers to the ischemic insult. Indeed, minimizing the adverse effects of I/R injury could significantly increase the number of patients that may successfully undergo liver transplantation. However, at present there is no treatment available to prevent hepatic I/R injury. As intervention on more than one level is most likely needed to allow the recovery of cellular and organ function, extensive research efforts to better understand the mechanisms of hepatic I/R injury are warranted.

Back to Article Outline

1. Liver microvascular architecture 

The high capacitance with high compliance is regarded as the most basic nature of the liver that makes a liver a liver [1]. The portal system ends at the inlet venules, feeding the sinusoids. All portal blood is delivered to the lobules through these minute twigs reaching an average 7–25μm in diameter. Sinusoids are first freely branching and anastomosing capillary beds but are subject to preferential grouping into repeated units: the hepatic microcirculatory subunits (HMS). The HMS base, located at the periphery of the lobule contains constitutively more plexiform sinusoids which include corresponding irregular plates. This architecture has three functionalities. It spreads and homogenizes the inflow bed from the inlet; it helps delay the flow velocity; it dams back any backward flow transmitted retrogradely from the central vein.

The high perfusion pressure of the hepatic artery is not readily admitted by the delicate wall of sinusoids. Basic anatomy of the intrahepatic arterial microvasculature indicates that the artery does not supply the hepatocytes, hence the sinusoids as its principal target because its primary vascular beds are formed not in the parenchymal but in the stromal compartments, from which second hand blood is drained into the secondary vascular beds, the lobular sinusoids [1]. The artery supplies highly oxygenated blood to five compartments [1]: (1) peribiliary vascular plexus; (2) portal tract interstitium, including nerves; (3) portal vein vasa vasorum; (4) hepatic capsule; (5) central-sublobular-hepatic vein vasa vasorum.

Blood from arterial beds is collected via the so-called hepatic artery-derived portal system (APS) that drains into the conventional portal vein at many levels, including inlet venules; this is usually recognized as an arterioportal anastomosis, although it may not be a true anastomosis in a strict sense. The APS is weIl developed in the large portal tracts, less so in the hepatic capsule and the hepatic tract. ln the latter, the arterial beds are drained either into the surrounding lobules or directly into the sublobular hepatic vein, bypassing the sinusoids, a situation possibly called an arteriohepatic anastomosis.

Back to Article Outline

2. The consequences of the lack of arterial blood 

Hepatic artery thrombosis (HAT), or even acute ligation, is usually well tolerated in humans by virtue of abundant collateral arterial sources that preserve the native liver from ischemia. Ligation of a hepatic artery branch (right, left, or segmental) in trauma is a safe adjunct for control of hemorrhage and does not result in biliary necrosis.

The consequence of the lack of arterial blood flow is, however, different in the setting of orthotopic liver transplantation (OLT). This role has been well studied in rats. Rearterialization improved survival following 48h of storage in UW solution from less than 10–50%, and reduced early enzyme release (AST) by about 50%. These data are consistent with the hypothesis that hepatic rearterialization minimizes hypoxic injury to parenchymal cells postoperatively, most likely by increasing oxygen delivery [2]. Lack of a post-transplant arterial supply may lead to persistent microvascular perfusion failure, hepatocellular/endothelial cell swelling, and microvascular anomalies related to bile duct injury. Recovery from microcirculatory alterations induced by cold preservation/reperfusion injury appears to depend on an intact hepatic arterial blood supply [3]. Without rearterialization, marked ductular proliferation with increasing fibrosis is observed, resulting in a secondary biliary cirrhosis by the 12th week [4].

In humans, although the total hepatectomy at OLT disrupts these collaterals, the allograft may survive on portal flow only while arterial collaterals develop. However, HAT is a cause of morbidity and graft loss in approximately 7% (range, 4–25%) of adults with an orthotopic liver transplant [5].

Technical aspects of the arterial anastomosis, especially selection of donor and recipient vessels, are important. Apart from technical difficulties related to the arterial anastomosis, other risk factors include a variety of conditions such as prolonged surgical time, intimal dissection, technique of graft preservation, poor baseline recipient hepatic arterial blood flow [6]. Early HAT results in massive injury to hepatocytes and bile duct epithelial cells. In particular, ischemic damage of bile ducts may lead to dehiscence of the biliary anastomosis, secondary bile ducts strictures, and intrahepatic bilomas or abscesses. For this reason, presentation with biliary sepsis is a common event in early HAT. It is estimated that one third of arterial thrombosis episodes are asymptomatic, one third are not immediately life threatening but lead to biliary tract ischemia syndromes, and one third cause parenchymal necrosis and rapid death if not rectified promptly [7].

Back to Article Outline

3. The different steps of damage during ischemia reperfusion 

The liver can be subjected to three forms of ischemia, namely cold (or hypothermic), warm (or normothermic), and rewarming [8]. Cold ischemia occurs in the transplant setting where it is applied intentionally to reduce metabolic activities of the graft while the organ awaits implantation. Warm ischemia occurs when hepatic inflow occlusion (Pringle maneuver) or inflow and outflow (total vascular exclusion) are induced to minimize blood loss while dividing the liver parenchyma (living donor). Rewarming ischemia typically occurs during the period of implantation of the graft when the cold liver is subjected to room or body temperature while performing the vascular reconstruction. Of note, injury to the liver cells after any type of ischemia is detected mainly after reperfusion when oxygen supply and blood elements are restored. Cold ischemia specifically caused injury to the sinusoidal endothelial cell (SEC). The SEC detached, lost cytoplasmic processes, became rounded as a result of alteration of the extracellular matrix and cytoskeleton, and sloughed into the sinusoidal lumen. Despite these structural changes, most SECs remain alive during the period of ischemia but rapidly die on reperfusion. Disruption of the endothelial wall leads to leukocyte and platelet adhesion which induces microcirculatory disturbances. The degree of endothelial cell detachment has been shown to correlate with the duration of cold ischemia in numerous experimental models. The morphologic changes typically identified in the endothelial cells result from active processes involving the cytoskeleton and extracellular matrix.

Adhesion of platelets to the sinusoid lining induces SEC apoptosis on reperfusion of the cold ischemic liver. Platelets are a rich source of transforming growth factor and calpains. Nitric oxide production by platelets in combination with oxygen free radical synthesis on reoxygenation of the ischemic liver can lead to the formation of perioxynitrite, a highly reactive inducer of apoptosis in endothelial cells. Leukocytes and platelets synergistically exacerbate SEC injury by induction of apoptosis. Kupffer cells, the resident hepatic macrophages, are also involved in the mechanism of injury mediated by these cells.

In contrast to cold ischemia, warm ischemia is tolerated poorly and rapidly leads to the death of hepatocytes. This severe injury of the hepatocytes probably is preceded by massive death of endothelial cells. The impact of rewarming on the structural integrity of the liver and the mechanism of this type of injury is understood poorly. It probably reflects a combination of cold and warm injury.

The process of I/R injury to the liver combines many interrelated factors that produce a cascade of events that lead to the ultimate hepatic failure [9]. Ample evidence suggests that activation of Kupffer cells, PMN, SEC, and reactive oxygen species are all critical in the pathogenesis of I/R injury. Significant microcirculatory changes reducing organ perfusion occur, reaching maximum levels within 48h of reperfusion. In viable organs that recover from the I/R insult, these changes decline and normalization of hepatic architecture is seen within 2 weeks of reperfusion. The first consequence of I/R injury is tissue anoxia that disturbs the intracellular energy metabolism and the enzyme function, resulting in depletion of adenosine triphosphate, accumulation of intracellular sodium, and cellular edema. The energy state of the cell at reperfusion is important to determine cell recovery. Reperfusion can rescue the cell, but it also induces further injury that starts as microcirculatory flow disturbances, manifested by red blood cells, polymorphonuclear cells (PMN), and platelet adhesion to SEC and sinusoidal congestion. The response of the hepatic endothelium to I/R is a significant causative factor, with SEC the least tolerant of the nonparenchymal liver cells to I/R. Indeed, SEC become activated during I/R to express an array of surface adhesion molecules and MHC antigens, priming the endothelium for further PMN interactions. The accumulation and activation of PMN exacerbate microcirculatory disturbances and transmigration into the injured tissue. Once extravasation and transmigration have occurred, there is little doubt about the destructive capabilities of the PMN. The PMN-induced injury to the hepatocyte results from adhesion between the two cells, probably mediated by integrins and intracellular adhesion molecules. PMN elaborate toxic enzymes such as elastase, serine protease, and metalloproteinases, but it can also produce oxygen-free radicals. Another nonparenchymal cell involved in hepatic I/R injury is the Kupffer cell which appears to be relatively resistant to ischemia yet becomes activated during reperfusion. When activated, it produces an array of factors, including proinflammatory cytokines (PGs, PAF, IL-1, TNF-α, IL-6, IFN-γ) and oxygen-free radicals, that act as direct cytotoxins to SEC and hepatocytes; it induces changes in cell membrane receptors on hepatocytes, EC, and PMN; it activates other Kupffer cells and PMN; and may induce chemotactic gradients for PMN.

Cold hepatic ischemia followed by reperfusion leads to necrotic cell death (oncosis) which often occurs within minutes of reperfusion. Recent studies also suggest a large component of apoptosis. Part of the confusion concerning the roles of apoptosis and necrosis in ischemia/reperfusion and other forms of hepatic injury arises from the assumption that apoptotic and necrotic mechanisms are distinct and separate when, in fact, these mechanisms can be shared [10]. In liver, oncotic necrosis and apoptosis share features and mechanisms. DNA degradation after necrosis causes TUNEL labeling, which may be incorrectly interpreted as apoptotic cell death. During apoptosis in pathophysiological settings, inflammatory responses and enzyme release occur that resemble a necrotic process. Frequently, oncotic necrosis and apoptosis coexist after toxic, hypoxic, and inflammatory liver injury. The coexistence of the two patterns of cell death likely reflects shared mechanistic pathways. Experimental or clinical settings will determine whether cells die predominantly by apoptosis or oncotic necrosis. Therefore, it is important to evaluate critical cell death pathways under clinically relevant conditions to discover new therapeutic intervention strategies in hepatic ischemia/reperfusion injury.

Back to Article Outline

4. The study of the microcirculation in human: its limits 

Intravital fluorescence microscopy (IFM) represents the standard method for the assessment of microcirculatory impairment. However, IFM can not be used clinically, restricting clinical studies to non-imaging techniques. The new technology of orthogonal polarization spectral (OPS)-imaging enables the direct visualization of the individual segments of the human hepatic sinusoidal network in vivo and to quantify human hepatic microvascular perfusion. This technique has been used by Puhl et al. [11] to study the microvascular consequences of the delay of rearterialization after initial portal reperfusion in living donor liver transplantation

IFM data represent the basis of our knowledge of the role of microvascular disturbances as consequence of various injuries. The OPS-imaging technology enables microvascular analysis by direct visualization. It has been tested in standardized animal models of in vivo microscopy, including the rat liver. Comparison between OPS-imaging and IFM revealed a statistically significant agreement of the results in respect to the image quality and the dimension of most of the microcirculatory parameters. Due to the principle technical basis of epi-illumination, both methods are limited to the analysis of the subcapsular parenchyma.

The major issue of the present study [11] is the description of microvascular disturbances in the consequence of initial portal reperfusion. In addition, the study is aimed to reproduce experimental findings in a clinical setting by direct visualization of microvascular perfusion. The authors resumed that OPS-imaging enabled the intraoperative analysis of deterioration of the hepatic microcirculation and that the increased manifestation of sinusoidal perfusion failure with prolongation of the interval between portal and hepatic arterial reperfusion support the view of less pronounced adverse effects caused by the anhepatic period itself than by the acceptance of a delay in rearterialization.

This study has limitations. (a) The representativity of the results presented here for the whole liver parenchyma, can not be assess, even if the correlation between the microvascular perfusion failure in the phase of initial portal reperfusion and the postoperative transaminases release suggest the vulnerability of the liver function to inadequate initial perfusion. (b) The volumetric blood flow within sinusoids was calculated from the sinusoidal diameter and the red blood cell velocity assuming cylindric geometry of the sinusoids. As said above, sinusoids especially in the portal zone have a rather complex route, it is therefore unlikely that this assumption can be 100% correct. (c) The authors have shown functional disturbances in the hepatic perfusion but the correlation of functional and morphological data are lacking. (d) The conclusion of microcirculatory impairment as described in this study to the pathogenesis of liver graft dysfunction may seem somehow problematic. (e) Eventhough the study indicates graft microcirculatory dysfunction as a major determinant of postischemic liver injury, and that microvascular impairment was significantly influenced by the interval between portal venous and hepatic arterial reperfusion, which suggests the reinforcement of the pathomechanism of injury involving hypoxia and rapid graft rewarming due to initial portal reperfusion, we have to remember that the study was performed within very short ischemic, anhepatic and rewarming periods and very small variations between patients (cold ischemic time: 74±29min; anhepatic period: 44±15min; portal vein–hepatic artery interval: 27.7±13.3min).

In spite of these limitations inherent to this type of study, the authors have been able to demonstrate, for the first time in human, using the OPS-imaging technology that: (1) hepatic nutritive perfusion, as given by the functional sinusoidal density and the volumetric blood flow, was found significantly decreased during portal reperfusion when compared to baseline, (2) rearterialization resulted in hyperperfusion of individual sinusoids. Interestingly, the time interval between portal venous and hepatic arterial reperfusion significantly correlated with the changes of the liver grafts' microcirculation.

Further studies are needed to see if this new technique, if applied in routine conditions of liver transplantation, can be useful to correlate the changes of the liver grafts' microcirculation with ischemic time, rewarming time, quality of the graft, and the outcome. (i.e. primary dysfunction and non function and biliary diseases).

There are presently few strategies in clinical use to protect against IR injuy. Ischemic preconditioning is one of them [12]. It consists of a brief period of ischemia followed by a short interval of reperfusion prior to cold preservation. Our current understanding of the underlying biologic principle is that cells primed by various kinds of subinjurious stress trigger defense mechanisms against subsequent lethal injury of the same or different type. The protective effects of ischemic preconditioning on hepatic microcirculation have been studied recently in animal models [13], [14]. Improved microcirculation either might be the result of or the reason for the preserved parenchymal architecture. It would be interesting to know in humans, using the OPS-imaging technology, if the protective strategies against ischemic reperfusion injury improve the microcirculation.

Back to Article Outline

References 

  1. Ekataksin W, Kaneda K. Liver microvascular architecture: an insight into the pathophysiology of portal hypertension. Semin Liver Dis. 1999;19:359–382
  2. Gao W, Lemasters JJ, Thurman RG. Development of a new method for hepatic rearterialization in rat orthotopic liver transplantation. Reduction of liver injury and improvement of surgical outcome by arterialization. Transplantation. 1993;56:19–24
  3. Tawadrous MN, Zimmermann A, Zhang XY, Wheatley AM. Persistence of impaired hepatic microcirculation after nonarterialized liver transplantation in the rat. Microcirculation. 2002;9:363–375
  4. Imamura H, Rocheleau B, Cote J, Huet PM. Long-term consequence of rat orthotopic liver transplantation with and without hepatic arterial reconstruction: a clinical, pathological, and hemodynamic study. Hepatology. 1997;26:198–205
  5. Langnas AN, Marujo W, Stratta RJ, Wood RP, Shaw BW. Vascular complications after orthotopic liver transplantation. Am J Surg. 1991;161:76–82
  6. Pastacaldi S, Teixeira R, Montalto P, Rolles K, Burroughs AK. Hepatic artery thrombosis after orthotopic liver transplantation: a review of nonsurgical causes. Liver Transpl. 2001;7:75–81
  7. Pinna AD, Smith CV, Furukawa H, Starzl TE, Fung JJ. Urgent revascularization of liver allografts after early hepatic artery thrombosis. Transplantation. 1996;15:1584–1587
  8. Selzner N, Rudiger H, Graf R, Clavien PA. Protective strategies against ischemic injury of the liver. Gastroenterology. 2003;125:917–936
  9. Fondevila C, Busuttil RW, Kupiec-Weglinski JW. Hepatic ischemia/reperfusion injury—a fresh look. Exp Mol Pathol. 2003;74:86–93
  10. Jaeschke H, Lemasters JJ. Apoptosis versus oncotic necrosis in hepatic ischemia/reperfusion injury. Gastroenterology. 2003;125:1246–1257
  11. Puhl G, Schaser K-D, Pust D, Köhler K, Vollmar B, Menger MD, et al. The delay of rearterialization after initial portal reperfusion in living donor liver transplantation significantly determines the development of microvascular graft dysfunction. J Hepatol. 2004;41:299–306
  12. Carini R, Albano E. Recent insights on the mechanisms of liver preconditioning. Gastroenterology. 2003;125:1480–1491
  13. Zapletal C, Mehrabi A, Scharf J, Hess T, Mihm D, Jahnke C, et al. Experimental evaluation of dynamic MRI for quantification of liver perfusion. Transplant Proc. 1999;31:421–422
  14. Sawaya DE, Brown M, Minardi A, Bilton B, Burney D, Granger DN, et al. The role of ischemic preconditioning in the recruitment of rolling and adherent leukocytes in hepatic venules after ischemia/reperfusion. J Surg Res. 1999;85:163–170

PII: S0168-8278(04)00285-5

doi:10.1016/j.jhep.2004.07.001

Journal of Hepatology
Volume 41, Issue 2 , Pages 340-343, August 2004

Article Tools