Embolization Therapy: Principles and Clinical Applications, 1 Ed.

Hepatopancreatic Disease

Ricardo Yamada • Christopher Hannegan • J. Bayne Selby • Marcelo Guimaraes

Cellular therapy has been used since late 1950s, when transplantation of hematopoietic cells was performed by E. Donnall Thomas1 based on the previous work of Jean Dousset, who identified the first human leukocyte antigen on the surface of cells, leading to a better understanding of histocompatibility and rejection. Since then, cell therapy, in the form of bone marrow transplant, has become a well-established treatment modality for some hematologic disorders. Recently, this therapeutic approach has expanded to other organs and systems, including the central nervous system, liver, pancreas, and heart. This expansion relies on several factors, such as technical improvements in cell harvesting, tracking, delivery, and engraftment.

Cell delivery to the target organ is a key step in the whole process. Bone marrow transplant is done by delivering hematopoietic cells through systemic infusion via a central venous access. Because bone marrow is located within multiple sites, systemic infusion permits widespread cell colonization. However, for a specific target organ, such as brain, heart, or liver, systemic infusion is inadequate because most of the cells will be delivered away from their desired destination, leading to poor engraftment and increased cell loss. In this context, the catheter-based technique is extremely suitable as it provides the capability of selective delivery of transplanting cells to a specific organ in a minimally invasive fashion. Therefore, selective intravascular administration of therapeutic cells, either through arteries or veins, is a growing field yet to be fully explored.


In contrast to other procedures described in this book, here, the embolic agents are live cells capable of altering the targeting organ function. The embolic effect is not the primary goal but rather a secondary result of the cell delivery process. Development of these techniques has expanded the usefulness of minimally invasive procedures beyond structural/anatomic corrections to also comprise specific cellular physiologic changes such as increased production of insulin and improved toxin clearance.

Unfortunately, the source of the “embolic agent” is a major limiting factor for widespread use of cellular therapy, as cell availability is limited by lack of donated organs. Currently, as described in the next sections, the pool of cells used for transplantation is obtained by isolating them from the donated organs of a different person (allogeneic transplant) or from the patient’s organ itself (autotransplant).

To overcome this problem, another potential source for cells is under constant investigation—the so-called stem cells. There are two different types of stem cells: embryonic and adult. The embryonic stem cells are obtained from the inner cell mass of the blastocyst and have the capability of differentiation into any type of cell (pluripotent). On the other hand, adult stem cells are obtained from the bone marrow, adipose tissue, or umbilical cord and have less differentiation capability (multipotent).2 Ethical and religious aspects limit embryonic stem cell research and clinical application, but adult stem cell use is not controversial and might become an unlimited cell source.


Type 1 Diabetes Mellitus

Clinical Application

According to the World Health Organization’s (WHO) last update on diabetes mellitus (DM), there are more than 34 million people worldwide affected by type 1 DM.3 Those patients have poor quality of life and decreased life expectancy due to inadequate glycemic control, which leads to acute and chronic complications, including severe unawareness of hypoglycemic events, retinopathy, glomerulopathy, and neuropathy. Pathogenesis is based on autoimmune destruction of β-cells within the islet of Langerhans. These cells are responsible for insulin production and are the most predominant pancreatic islet cells, representing up to 75% of islet composition.4

Currently, the mainstay therapy is intensive exogenous insulin administration through multiple daily subcutaneous injections or continuous insulin infusion via an implanted pump. Despite all different insulin regimens available, control of glucose level is not ideal yet, and high incidence of hypoglycemic events is still a major concern. So far, the only treatment that has shown successful results in controlling disease progression without hypoglycemic events is total pancreas transplantation.5,6 However, organ transplant is limited by lack of donors, in addition to associated comorbidities related to surgery and rejection.7

To overcome the unsatisfactory glucose level control with exogenous insulin therapy and surgical complications of whole pancreas transplant, a lot of effort has been put into developing an alternative therapy, which is β islet cell transplantation (ICT). This procedure gained a lot of attention after 2000, when Shapiro et al.8 reported the results of the Edmonton series in which all seven patients were still insulin-independent after a median follow-up of 11.9 months. This result was based on the use of a glucocorticoid-free immunosuppressive regimen. Until that time, only 8.2% of patients were insulin-free 1 year after ICT.9

Most likely, the population that benefits the most from ICT is the group of patients who have labile diabetes and/or severe hypoglycemic events with sustained renal function because whole pancreas transplant alone has higher morbidity and mortality compared to ICT.10 Patients with associated chronic renal insufficiency benefit more with kidney–pancreas transplant.

Labile diabetes is defined as inconsistent glucose levels that follow no predictable pattern, interfering with a patient’s quality of life. To quantify that, a lability index (LI) was created based on glucose level measurements during a period of 4 weeks.11 A severe hypoglycemic event is defined as a hypoglycemic episode that requires outside assistance to be treated; a composite hypoglycemic score, called HYPO, was also developed to quantify the frequency, severity, and degree of unawareness of hypoglycemia.11 These measurement systems are important to promote an objective indication of the disease’s severity and thus guide treatment choice.

Similar to whole-organ transplantation, ICT for type 1 DM requires lifelong immunosuppressive therapy, which is associated with frequent and severe side effects. In addition, despite its minimal invasiveness, the procedure carries some risk of serious complications, such as infection, bleeding, and portal vein thrombosis. Therefore, the procedure should be performed only in patients who present severe complications related to poor glycemic control and/or unacceptable quality of life. The patient should understand clearly that exogenous insulin therapy would be exchanged for an immunosuppressive therapy, and so the problems associated with type 1 DM must be severe enough to justify it.

Regardless of the increasing success rate that has been reported,12 ICT is not yet considered a conventional therapy. Issues related to organ availability, cell extraction, delivery and engraftment, and also immunosuppressive regimen remain limiting factors. Hence, according to the American Diabetes Association,10 ICT should be performed only in the setting of controlled research studies in a tertiary care center capable of managing all complex medical situations associated with transplanted patients.

According to the last Collaborative Islet Transplant Registry (CITR) annual report, insulin independence can be achieved in up to 80% to 90% of patients, especially in the presence of favorable predictive factors, such as lower baseline insulin requirement and age older than 35 years.12 Unfortunately, sustained insulin independence decreases over time, and only 10% of patients remain insulin-independent 5 years after transplant.12 Graft functionality is determined by any detection of serum C-peptide by local assay or stimulated serum C-peptide level greater than or equal to 3 ng/mL. Therefore, transplanted islet cells are categorized as fully functional (detectable C-peptide level and insulin independence), partially functional (insulin dependence but detectable C-peptide level), or nonfunctional (no detectable C-peptide).

Despite the low long-term insulin-independence rate, persistent C-peptide level greater than or equal to 3 ng/mL can be achieved in around 80% of patients even 5 years after transplant,12 meaning that a significant number of patients will have at least a partially functioning graft. This has been proven clinically significant by the drop in both hypoglycemic score and lability index13 seen also among insulin-dependent patients, representing improvement of glycemic control, which is the main goal of ICT. In addition, studies have demonstrated that ICT leads to stabilization and even some improvement of the long-term type 1 DM complications, including retinopathy,14glomerulopathy,15 and neuropathy.16


The procedure is performed under moderate sedation, and preevaluation is focused on any conditions that increase sedation risk, such as airway compromise and cardiopulmonary dysfunction. If the patient has an increased risk for moderate sedation, general anesthesia should be considered. Blood workup includes serum creatinine, as the use of iodine contrast dye is necessary and these patients may have some degree of renal dysfunction. Platelets and prothrombin time are also needed because bleeding is the major procedure-related complication. For antibiotic prophylaxis, cefazolin sodium 1 mg intravenously 1 hour before the procedure and every 6 hours for 24 hours after the transplant has been typically used. Sulfamethoxazole/trimethoprim 400/80 mg is given orally twice a week for 24 weeks to prevent Pneumocystis pneumonia and valganciclovir 450 mg is prescribed orally once a day for 12 weeks to prevent cytomegalovirus infection.17

Under sterile technique and after local anesthesia, a 21-gauge Chiba needle is advanced into the liver parenchyma under fluoroscopic guidance at the level of the right midaxillary line, avoiding transpleural approach. The needle is advanced horizontally and with slight cephalic angulation (10 to 15 degrees). The abdominal midline should not be crossed. The needle is then slowly retracted while small amounts of contrast are injected, searching for portal vein branches. After finding a branch, a more forceful hand contrast injection is performed to confirm needle location within the portal system (Fig. 43.1). Use of ultrasound-guided puncture has been advocated as a safer approach, with less needle passages through the liver parenchyma and single-wall vein puncture.18 Ideally, a second- or third-order branch should be punctured to decrease the risk of bleeding, as the access to the portal system is upsized to accommodate a larger catheter. After confirming the needle location within the portal system, a 0.018-in wire is advanced into the main portal vein under fluoroscopic guidance and a 4-Fr catheter is introduced. After access is secured, heparin is given intravenously according to the patient’s weight. Baseline portal vein pressure is measured. Pressure above 20 mm Hg is a contraindication for cell infusion due to increased risk of portal vein thrombosis. Next, portography is performed through a power injector and a side-hole catheter, infusing 6 mL of iodine contrast per second and a total volume of 30 mL. Portal vein anatomy is delineated, confirming its patency and hepatopetal flow (Fig. 43.2).

A catheter with at least 700-mm inner diameter is recommended for infusion to avoid cell damage from possible shear forces or increased pressure; therefore, a 4-Fr catheter is more than appropriate.19 The catheter tip is positioned beyond portal vein bifurcation to allow cell distribution in both hepatic lobes (Fig. 43.2). Harvested cell administration by gravity flow is preferred over direct syringe infusion because administration by gravity allows a safety mechanism of natural flow reduction that parallels any increase of portal vein pressure, avoiding precipitous pressure rise.19 Direct syringe infusion seems to be associated with an increased risk of portal vein thrombosis. To avoid that, bag method infusion uses a closed gravity feeding bag system, consisting of a transfer bag and a rinse bag connected via sterile tubing (Fig. 43.3). During infusion, portal vein pressure is measured periodically, and an increase of more than double the baseline or above 22 mm Hg for more than 10 minutes should prompt interruption of the infusion due to increased risk of portal vein thrombosis. After infusion, a final portal vein pressure and portography are obtained.

Despite the use of low-profile systems (4-Fr), embolization of the parenchymal track is performed at the end of the procedure as it is believed to substantially decrease the risk of postprocedure bleeding.20Under fluoroscopic guidance, the catheter is slowly withdrawn from the portal vein. Proper catheter location within the parenchymal track is confirmed with gentle hand injection of a small amount of iodine contrast. Embolization of the track can then be performed with different types of embolic agents, including gelfoam, coils, and N-butyl cyanoacrylate. The ideal embolic agent should promote a complete seal of the track with accurate deployment, avoiding intravascular embolization. Embolization is recommended for at least 5 to 7 cm of hepatic parenchyma to prevent postprocedure bleeding from the liver surface.21

After the procedure is terminated, the patient should be admitted to an intensive care unit, with close monitoring of vital signs and complete blood cell count measurement every 6 hours. If bleeding is not initially suspected, to prevent portal vein thrombosis, intravenous heparin infusion is started, aiming a thromboplastin time of 50 to 60 seconds. After 48 hours, intravenous heparin is switched to subcutaneous low-molecular-weight heparin for 1 week.

Chronic Pancreatitis

Clinical Application

Chronic pancreatitis (CP) is an important disabling entity that leads to significant detriment of life quality, mainly due to severe abdominal pain and opioid abuse. CP has a reported incidence of 3.5 to 10 per 100,000 people per year, and incapacitating pain is present in nearly 90% of these patients.22 The primary treatment goal is pain alleviation, which can be associated with pancreatic duct dilation or not. When ductal dilation is present, endoscopic or surgical drainage is indicated. However, when dilation of the pancreatic duct is absent or drainage fails, total pancreatectomy (TP) should be considered.

In patients with established diabetes, the decision to proceed with pancreatic resection is more convenient because surgery is not adding a new comorbidity. On the other hand, nondiabetic patients submitted to pancreatectomy will have an incidence greater than 50% of postoperative diabetes, usually associated with more difficult glucose control and severe hypoglycemic events.23,24

To overcome this problem, islet cell autotransplantation (IAT) was first introduced in 1977 in the University of Michigan.25 Since then, other centers in the United States have been performing this type of transplant successfully. In this process, the resected organ is sent to a cell laboratory, where a collagenase-based digestion process is started by pancreatic duct cannulation and enzymatic infusion. This will separate islet cells from the exocrine pancreas and connective tissue. Once islet cells are isolated, they are placed in a bag with albumin solution and antibiotic, and they are ready to be transplanted.

In the initially described technique, the patient (under general anesthesia) and the surgical team wait in the operating room (OR) while the cells are harvested. This process usually takes around 4 hours. More recently, we started performing the cell infusion at the interventional radiology (IR) angiography suite with the goal to maximize cost and efficiency. Soon after the pancreas is resected, the cells get processed at the cell therapy center while the patient’s abdomen is closed. Alternatively, the patient may be sent to the intensive care unit for a few minutes in case there is any delay in pancreatic islet cell separation and preparation process26 (Fig. 43.4).

In contrast to allogeneic ICT, IAT does not require immunosuppression, and therefore the patient is spared from its side effects. The long-term outcome is also better with IAT in comparison to allogeneic transplant as nearly 50% of IAT patients who achieved insulin independence remained insulin-free after 5 years.27 As shown earlier, only 10% of patients were off insulin 5 years after allogeneic transplant.13Many reasons have been brought up to explain this discrepancy, including longer cold ischemia time, donor brain death, immunosuppression toxicity, and autoimmune deleterious effects.27


When cell preparation is finished, the patient is brought to the IR room for the infusion process, which can be performed via three different approaches: percutaneous transhepatic route,26 surgically placed catheter in the portal system through a dissected mesenteric branch during pancreatecotmy, or through temporary exposure of an omental tongue containing a tributary vein.28

Infusion through a percutaneous transhepatic route requires the same technique described earlier for allogeneic ICT in patients with type 1 DM, with similar procedure-related complications, especially bleeding (see the following discussion). Using the two other techniques, bleeding is almost excluded, although a second small abdominal incision is performed to expose the omentum in the technique described by Nath et al.28 Initially at our institution, the transhepatic route was the preferred one; however, recently, we have changed our practice by using the surgically placed catheter technique, which is believed to be safer. Percutaneous transhepatic infusion is now reserved for cases in which the surgical access is lost or not ideal.

For the surgically placed catheter technique, a mesenteric vein is cannulated and a glide wire is advanced distally into the portal vein, followed by a 5-Fr KMP (Cook Medical, Inc., Bloomington, Indiana) catheter placement. The wire is removed and the catheter is secured to the mesentery with silk sutures. The vein distal to the cannulation site is ligated and the catheter is brought out through the midline abdominal incision, which is closed with the standard surgical technique. The catheter is secured to the skin with silk sutures (Fig. 43.5).

The first step of the infusion process is to verify the catheter’s tip location within the portal system by hand contrast injection. Most of the time, the catheter needs to be repositioned to a better location (Fig. 43.6). Less commonly, the access to the portal system is completely lost and percutaneous transhepatic access is required. After confirmation of the ideal position, pressure is measured and cell infusion by gravity gets started (Fig. 43.7). It is thought that active cell aspiration from the bag and then portal vein infusion by hand injection might damage the cells. A completion portogram is then performed to verify portal vein patency and absence of intraluminal filling defects within the main portal vein and its major branches. It is expected to see wedge-shaped filling defects in the periphery of the liver and delayed contrast washout of the secondary branches (Fig. 43.8).

Once infusion is terminated and before catheter removal, homeostasis of the mesenteric vein needs to be achieved to avoid intraperitoneal bleeding. Exposing the vein through the midline incision and ligation with surgical clips is one of the approaches that can be used. In situations where the vein cannot be externalized through the abdominal incision, the branch must be embolized. For this, the catheter is slowly retracted from the main portal vein until it reaches the mesenteric branch (Fig. 43.9). At this point, embolization can be performed through the 5-Fr catheter, or a microcatheter can be advanced coaxially. This can be extremely helpful when dealing with a very short vein segment, as access stability and more accurate embolization can be achieved. Use of low-profile devices in combination with a larger catheter usually promotes adequate support. Coils are the preferred embolic devices, but others can be used in combination, especially N-butyl cyanoacrylate. A postembolization venogram is performed to confirm complete sealing of the vein before the catheter is removed.


Clinical Application

Currently, clinical use of hepatocyte transplantation (HCT) is limited to a few specialized centers worldwide and is still an evolving technique. Studies in HCT have been focused on three different clinical settings: chronic liver disease, inborn metabolic liver disease, and acute liver failure. In 1976, Matas et al.29 published their successful experiment in a rat model for Crigler-Najjar syndrome type 1, showing reduction of serum bilirubin level after intraportal infusion of genetic modified hepatocytes. Then, in 1992, Mito et al.30 reported the first human experience with hepatocyte infusion in 10 patients with chronic liver disease by direct splenic inoculation of harvested hepatocytes from patient’s left hepatic lobe (autotransplantation). Although hepatocytes could be identified within the spleen up to 6 months after transplantation, clinical benefits were not achieved.30 Subsequent studies with allotransplantation of hepatocytes from noncirrhotic liver donors showed more promising clinical results, promoting declined serum ammonia levels, improved encephalopathy, and successful “bridging” to whole-organ transplant.31,32

In 1994, Grossman et al.33 reported the first human trial of genetically modified hepatocyte transfusion for familial hypercholesterolemia in a 29-year-old patient who achieved sustained reduction in the low-density lipoprotein-to-high-density lipoprotein ratio after intraportal hepatocyte transfusion.33 Since then, reports of other human inborn metabolic diseases treated with HCT have been published with encouraging results, including Crigler-Najjar, α-1 antitrypsin deficiency, factor VII deficiency, glucose storage disease, and urea cycle defect.34 For acute liver failure, Habibullah et al.35 were the first to publish HCT in humans in 1994, and similar to chronic liver disease patients, it showed promising clinical results.

As noted, HCT can be a valuable tool in critical clinical scenarios in which satisfactory treatment is not yet established, such as inborn metabolic liver disease, or it is not fully available, as for acute/chronic liver disease, where the number of donated organs is not sufficient to meet the demand. In addition, the minimally invasive nature of the procedure is a huge advantage over conventional liver transplantation, which is a major open surgery.


Cell Source

The first human HCT experience was actually an autotransplant using hepatocytes from the resected left lobe of a patient’s cirrhotic liver.30 As mentioned before, clinical success was not accomplished, probably related to suboptimal function of the transplanted cells. Currently, hepatocytes come from the liver of noncirrhotic donors who are not suitable for orthotopic liver transplantation due to prolonged ischemia time, traumatic damage to the graft, capsular tear, blood group incompatibility, and/or vascular or biliary injury.36 In addition to these criteria of nonsuitability for whole-organ transplant, the donor should also be free of sepsis, neoplasia, and viral infection (hepatitis C and B, HIV, human T-lymphotropic virus, and syphilis) and should present with less than 30% of steatosis. Studies have showed that fatty liver infiltration decreases cell viability.37

Once the organ is available, cell isolation is performed through an enzymatic digestion process that starts with cannulation of the hepatic arteries and portal veins, followed by parenchymal perfusion with a collagenase-based solution. Sequentially, mechanical disintegration, filtration, and centrifugation are performed, providing isolated hepatocytes. These cells are then tested for viability because at least 60% cell viability is recommended before transplantation. They are also tested for fungal, bacterial, mycoplasma, and endotoxin contamination.

At this point, the cells are ready to be infused or, in contrast to pancreatic ICT where solely fresh cells are administered, isolated hepatocytes can undergo a cryopreservation process, in which hepatocytes are frozen and preserved for later use. This storage process has the advantage of allowing a planned cell infusion and not only emergently and also permits creation of a cell bank, where hepatocytes can be preserved and readily available for transfusion. The drawback is the loss of viability after the frozen/thawing process, which can reach up to 50% of the cells.38

Route of Infusion

When first performed in humans, cell transplantation was done by direct splenic puncture.30 Although animal lab research has shown that this route of infusion leads to a better hepatocytes engraftment,39 this approach is rarely used nowadays due to increased potential risk of intraperitoneal bleeding, especially in the setting of coagulopathy and portal hypertension. Currently, two main routes have been used: intrahepatic–transportal and intrasplenic–transarterial.

The intrasplenic–transarterial route is the preferred one in patients with chronic liver disease and portal hypertension, as the embolic effect of transplanted cells can increase portal pressure and the risk of portal vein thrombosis, not to mention the possibility of having hepatofugal flow in cirrhotic patients with portal hypertension. In addition, the spleen in cirrhotic patients can be up to 8 to 10 times bigger than a normal spleen, allowing adequate accommodation of the cell load.40 For this method, the common femoral artery (usually the right) is cannulated with Seldinger technique and a 5-Fr sheath is placed to secure the access. A 5-Fr Mickelson catheter (Boston Scientif ic, Marlborough, Massachusetts) is used to select the celiac trunk, and selective angiography is performed to delineate the vascular anatomy and confirm splenic artery patency (Fig. 43.10). Next, the splenic artery is catheterized selectively using a 0.035-in hydrophilic wire and the 5-Fr diagnostic catheter. Distal splenic angiography is performed to confirm the catheter tip location and to exclude vasospasm or dissection (Fig. 43.11). At this point, cell transfusion can begin and is carried on until all the cells are infused or flow stasis is achieved. A completion angiogram is performed, which typically will reveal multiple perfusion defects, similar to what is seen in the liver parenchyma after intraportal islet cell infusion (Fig. 43.12). The hemostasis at the femoral artery puncture site is obtained with a closure device or by 15 minutes of manual compression.

Access to the portal system for the intrahepatic–transportal infusion can be obtained through different techniques, including surgical access to a mesenteric vein, percutaneous liver puncture, and by umbilical vein catheterization in newborn patients. The surgical access technique to the portal system is beyond the scope of this book. Percutaneous transhepatic access is achieved with the same technique described earlier for pancreatic islet transplantation. A different percutaneous approach can also be used, as reported by Fox et al.41 who performed HCT in a 10-year-old patient with Crigler-Najjar syndrome through the left portal vein. Under ultrasound guidance, the vein was punctured with a 21-gauge needle and a micropuncture introducer sheath was used to upsize the access, allowing the placement of a 5-Fr KMP catheter.

In newborns, during the first week of life, access to the portal system can be obtained through the umbilical vein (UV). UV catheterization is commonly used in critically ill infants as a way to obtain central venous access because the ductus venosus is still patent until the 20th day of life (Fig. 43.13).

Horslen et al.42 reported multiple hepatocyte transfusions in a patient with a urea cycle disorder through the UV access. Transfusions were performed during the first 51 days of life, with the first transfusion done 10 hours after birth. Under fluoroscopic guidance, the originally placed UV catheter was manipulated into the portal vein. As the ductus venosus was still patent, placement of an occlusion balloon was performed to isolate the portal system from the systemic circulation.42 Between transfusion sessions, the catheter was retracted into the distal UV to avoid portal vein thrombosis and secure the access for the next infusion.

Because multiple transfusions may be required in the same patient, a technique for long-term portal vein access has been advocated by Darwish et al.43 The authors described surgical placement of an implantable port device in three patients. Through a small transverse left upper abdominal incision, the transverse colon is explored, allowing dissection of an appropriate colonic vein, which is cannulated with a 7-Fr catheter. The device is pulled through the left mesocolon, passed via abdominal muscles, and connected to the metallic chamber positioned in the subcutaneous tissue of the left upper quadrant along the anterior axillary line. The longest period of implantation was 5 months, and no complications were reported, especially portal vein thrombosis. One patient had catheter displacement after 30 days of implantation and required a second laparotomy to correct its location. According to the authors, special attention must be paid to correctly secure the catheter to avoid catheter migration outside the vein and potential risk of intraperitoneal bleeding.


As these procedures are considered new techniques under constant development, complication rates should be analyzed with caution. For example, the complication profile of ICT for type 1 DM has changed because operators have gained more experience over time. Use of a more refined technique with smaller devices has helped decrease the incidence of adverse events. This has been confirmed by the CITR Seventh Annual Report, which demonstrated that life-threatening events occurred in 26% of cases performed during 1999 to 2003, whereas during 2007 to 2009, those events involved 11% of patients.12

Therefore, it is important to consider the most recent set of data when analyzing the complication rate. It is also relevant to differentiate between complications related to the infusion process itself or to the immunosuppressive therapy. For instance, ICT for type 1 DM during 2007 to 2010 presented 9.6% of serious adverse events related to the infusion process and 13.3% related to the immunosuppressive regimen during the first 30 days after transplant. Of note, 90.7% of those patients who experienced a serious adverse event recovered completely or remained with minimal sequelae.12 Table 43.1 depicts some common complications after ICT in type 1 DM patients.

As noticed, bleeding is the most common complication related to the percutaneous transhepatic infusion process. Villiger et al.21 on their analysis of 132 islet cell transplants found that a cumulative number of transplantation procedures and intraprocedure heparin dose of greater than or equal to 45 units per kilogram were independent risk factors for bleeding complication. To eliminate this risk, IAT in postpancreatectomy patients has been performed through a surgically placed catheter, as mentioned earlier. Portal vein thrombosis, another procedure-related complication, occurred in 2.1% of the patients, with complete recovery in 83.3% of the cases.12

Percutaneous intraportal hepatocyte infusion has the same potential complications of ICT, including portal vein thrombosis and bleeding. So far, human experience has not showed any major adverse event, although mild complications such as transient elevation of aspartate aminotransferase (AST)/alanine aminotransferase (ALT) and hypoxemia have been reported.44 Transarterial intrasplenic cell infusion carries the same complication risks of any arterial puncture, including hematoma, pseudoaneurysm, dissection, and thrombosis, because the procedure is performed through a femoral artery access.


• Multiple sticks in the liver, while attempting to get access to the portal vein, is associated to significant bleeding complication. Consider ultrasound guidance, especially useful for the left hepatic lobe.

• Consider using micropuncture kit (0.021-in wire) that may be converted to a 0.035-in system (AccuStick system; Boston Scientific Corporation, Natick, Massachusetts).

• Any cell therapy should be infused by gravity. Avoid hand injection, which may damage the cells.

• Portal vein pressure measurement preinfusion, at one-half of the infusion, and postinfusion should be performed. The few cases of portal vein thrombosis occurred when the portal vein pressure increased above 30 mm Hg. If the initial portal pressure is above 25 mm Hg, consider performing very slow cell infusion.

• To decrease the chance of bleeding through the transparietohepatic access, access track embolization with gelfoam torpedo/coils should be always considered.

• The arterial hepatocyte infusion into the spleen can be challenging due to the increased tortuosity of the splenic artery. The combination of a soft diagnostic catheter and a glide wire can facilitate selective distal catheterization of the splenic artery.

• In coagulopathic patients, the radial artery puncture should be considered as an alternative as it has been demonstrated to reduce bleeding-related complications in comparison to femoral approach.

• Percutaneous access to the portal system can be associated with significant bleeding complication, and the use of a micropuncture can decrease this risk. Multiple passages of the 21-gauge needle through the liver parenchyma can be performed without much concern.

• A second method to decrease the risk of bleeding is accessing the portal system through the left lobe under ultrasound guidance. To decrease the chances of bleeding through the transparietohepatic access, access track embolization should be always considered.

• Hemostasis postpancreatectomy ICT can be achieved by embolization of the mesenteric vein branch in which the catheter was surgically inserted. A microcatheter can be used coaxially to deploy the embolic agent precisely. This technique is helpful when working with a very short vein segment because adequate support and stability can be obtained (Fig. 43.9).

• The arterial hepatocyte infusion into the spleen can be challenging due to the increased tortuosity of the splenic artery. In this situation, combination of a soft diagnostic catheter and a glide wire can facilitate selective distal catheterization of the splenic artery.

• These patients very often have at least some degree of coagulation dysfunction, which increases arterial access risk. Therefore, common femoral artery puncture can be performed under ultrasound guidance with a micropuncture kit to guarantee a single-wall puncture above the femoral bifurcation.


 1. Thomas ED, Lochte HL Jr, Lu WC, et al. Intravenous infusion of bone marrow in patients receiving radiation and chemotherapy. N Engl J Med. 1957;257(11):491–496.

 2. Nikolic B, Faintuch S, Goldberg SN, et al. Stem cell therapy: a primer for interventionalists and imagers. J Vasc Interv Radiol. 2009;20(8):999–1012.

 3. World Health Organization. Diabetes. Diabetes: fact sheet no. 312, August 2011. http://www.who.int/mediacentre/factsheets/fs312/en/. Accessed May 16, 2013.

 4. Brissova M, Fowler MJ, Nicholson WE, et al. Assessment of human pancreatic islet architecture and composition by laser scanning confocal microscopy. J Histochem Cytochem. 2009;53(9):1087–1097.

 5. Morel P, Chau C, Brayman K, et al. Quality of metabolic control at 2 to 12 years after a pancreas transplant. Transplant Proc. 1992;24:835–838.

 6. Sutherland DER, Gruessner RWG, Dunn DL, et al. Lessons learned from more than 1,000 pancreas transplants at a single institution. Ann Surg. 2001;233(4):463.

 7. Gruessner AC, Sutherland DE. Pancreas transplant outcomes for United States (US) and non-US cases as reported to the United Network for Organ Sharing (UNOS) and the International Pancreas Transplant Registry (IPTR) as of June 2004. Clin Transplant. 2005;19:433–455.

 8. Shapiro AM, Lakey JR, Ryan EA, et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med. 2000;343:230–238.

 9. Brendel MD, Hering B, Schultz AO, et al. International Islet Transplant Registry: Newsletter 8. Giessen, Germany: Justus-Liebig-University of Giessen; 2001.

10. Ryan EA, Bigam D, Shapiro AM. Current indications for pancreas or islet transplant. Diabetes Obes Metab. 2006;8(1):1–7.

11. Ryan EA, Shandro T, Green K, et al. Assessment of the severity of hypoglycemia and glycemic lability in type 1 diabetic subjects undergoing islet transplantation. Diabetes. 2004;53(4):955–962.

12. Collaborative Islet Transplant Registry. CITR seventh annual report. https://web.emmes.com/study/isl/reports/01062012_7thAnnualReport.pdf. Accessed June 4, 2013.

13. Ryan EA, Paty BW, Senior PA, et al. Five-year follow-up after clinical islet transplantation. Diabetes. 2005;54(7):2060–2069.

14. Thompson DM, Begg IS, Harris C, et al. Reduced progression of diabetic retinopathy after islet cell transplantation compared with intensive medical therapy. Transplantation. 2008;85(10):1400–1405.

15. Fiorina P, Folli F, Zerbini G, et al. Islet transplantation is associated with improvement of renal function among uremic patients with type I diabetes mellitus and kidney transplants. J Am Soc Nephrol. 2003;14(8):2150–2158.

16. Lee TC, Barshes NR, O’Mahony CA, et al. The effect of pancreatic islet transplantation on progression of diabetic retinopathy and neuropathy. Transplant Proc. 2005;37(5):2263–2265.

17. Gaba RC, Garcia-Roca R, Oberholzer J. Pancreatic islet cell transplantation: an update for interventional radiologists. J Vasc Interv Radiol. 2012;23(5):583–594.

18. Venturini M, Angeli E, Maffi P, et al. Technique, complications, and therapeutic efficacy of percutaneous transplantation of human pancreatic islet cells in type 1 diabetes: the role of US. Radiology. 2005;234(2):617–624.

19. Daly B, O’Kelly K, Klassen D. Interventional procedures in whole organ and islet cell pancreas transplantation. Semin Intervent Radiol. 2004;21(4):335–343.

20. Baidal DA, Froud T, Ferreira JV, et al. The bag method for islet cell infusion. Cell Transplant. 2003;12(7):809–813.

21. Villiger P, Ryan EA, Owen R, et al. Prevention of bleeding after islet transplantation: lessons learned from a multivariate analysis of 132 cases at a single institution. Am J Transplant. 2005;5(12):2992–2998.

22. Thuluvath PJ, Imperio D, Nair S, et al. Chronic pancreatitis. Long-term pain relief with or without surgery, cancer risk, and mortality. J Clin Gastroenterol. 2003;36(2):159–165.

23. Park JW, Jang JY, Kim EJ, et al. Effects of pancreatectomy on nutritional state, pancreatic function and quality of life. Br J Surg. 2013;100(8):1064–1070.

24. Barbier L, Jamal W, Dokmak S, et al. Impact of total pancreatectomy: short- and long-term assessment [published online ahead of print January 29, 2013]. HPB (Oxford).

25. Sutherland DE, Matas AJ, Najarian JS. Pancreatic islet cell transplantation. Surg Clin North Am. 1978;58(2):365–382.

26. Morgan KA, Nishimura M, Uflacker R, et al. Percutaneous transhepatic islet cell autotransplantation after pancreatectomy for chronic pancreatitis: a novel approach. HPB (Oxford). 2011;13(7):511–516.

27. Sutherland DE, Gruessner AC, Carlson AM, et al. Islet autotransplant outcomes after total pancreatectomy: a contrast to islet allograft outcomes. Transplantation. 2008;86(12):1799–1802.

28. Nath DS, Kellogg TA, Sutherland DE. Total pancreatectomy with intraportal auto-islet transplantation using a temporarily exteriorized omental vein. J Am Coll Surg. 2004;199(6):994–995.

29. Matas AJ, Sutherland DER, Steffes MW, et al. Hepatocellular transplantation for metabolic deficiencies: decrease plasma bilirubin in Gunn rats. Science. 1976;1992:897.

30. Mito M, Kusano M, Kawaura Y. Hepatocyte transplantation in man. Transplant Proc. 1992;24(6):3052–3053.

31. Soriano HE, Wood RP, Kang DC, et al. Hepatocellular transplantation (HCT) in children with fulminant liver failure. Hepatology. 1997;30:239A.

32. Strom SC, Fisher RA, Thompson MT, et al. Hepatocyte transplantation as a bridge to orthotopic liver transplantation in terminal liver failure. Transplantation. 1997;63(4):559–569.

33. Grossman M, Raper SE, Kozarsky K, et al. Successful ex vivo gene therapy directed to liver in a patient with familial hypercholesterolaemia. Nat Genet. 1994;6(4):335–341.

34. Fisher RA, Strom SC. Human hepatocyte transplantation: worldwide results. Transplantation. 2006;82(4):441–449.

35. Habibullah CM, Syed IH, Qamar A, et al. Human fetal hepatocyte transplantation in patients with fulminant hepatic failure. Transplantation. 1994;58(8):951–952.

36. Pareja E, Martinez A, Cortes M, et al. Hepatic cell transplantation. Technical and methodological aspects. Cir Esp. 2010;87(3):139–147.

37. Donato MT, Lahoz A, Jimenez N, et al. Potential impact of steatosis on cytochrome P450 enzymes of human hepatocytes isolated from fatty liver grafts. Drug Metab Dispos. 2006;34(9):1556–1562.

38. Terry C, Dhawan A, Mitry RR, et al. Cryopreservation of isolated human hepatocytes for transplantation: state of the art. Cryobiology. 2006;53(2):149–159.

39. Nagata H, Ito M, Shirota C, et al. Route of hepatocyte delivery affects hepatocyte engraftment in the spleen. Transplantation. 2003;76(4):732–734.

40. Sterling RK, Fisher RA. Liver transplantation. Living donor, hepatocyte, and xenotransplantation. Clin Liver Dis. 2001;5(2):431–460.

41. Fox IJ, Chowdhury JR, Kaufman SS, et al. Treatment of the Crigler-Najjar syndrome type I with hepatocyte transplantation. N Engl J Med. 1998;338(20):1422–1426.

42. Horslen SP, McCowan TC, Goertzen TC, et al. Isolated hepatocyte transplantation in an infant with a severe urea cycle disorder. Pediatrics. 2003;111:1262–1267.

43. Darwish AA, Sokal E, Stephenne X, et al. Permanent access to the portal system for cellular transplantation using an implantable port device. Liver Transpl. 2004;10:1213–1215.

44. Smets F, Najimi M, Sokal EM. Cell transplantation in the treatment of liver diseases. Pediatr Transplant. 2008;12(1):6–13.