The development of the Edmonton protocol by Shapiro et al. (1) has dramatically improved the chance of achieving metabolic stability and insulin-independence by way of intrahepatic allogeneic pancreatic islet transplantation (PIT). However, the true potential of this clinical treatment modality has not been realized because of the lack of cadaveric donor pancreatic islets and difficulty associated with islet isolation. Islet isolation is particularly challenging, with previous studies reporting only a successful islet isolation rate of 55% to 61 % (2, 3). If islet recovery rates could be increased, this would increase the number of patients that could benefit from PIT.
There are many variables that must be considered when addressing the achievement of successful islet isolations. These variables include but are not limited to: pre- and postbrain-death donor characteristics, distance of donor pancreas from islet isolation center, and islet isolation/purification conditions (2, 3). Previously, it has been demonstrated that islet isolation, when performed on pancreata procured at a distant procurement site, resulted in a decreased rate of successful islet isolation (2). Successful islet isolation was defined by Lakey et al. (2) as recovery of greater than 100,000 islet equivalents (IEQ) with a 50% purity, whereas Benhamou et al. (3) used islet viability at 48 hours in culture and islet yield as a measure of successful isolation. Although these values may be acceptable for PIT research, to be successful clinically, we must isolate a minimum of 5,000 IEQ/kg recipient weight per donor pancreas procurement.
Unfortunately, to date, little attention has been focused on the technical aspects of pancreatic procurement and their effects on achieving successful islet isolations. Recently, we initiated our clinical PIT program in collaboration with a distant islet isolation center (4). This collaborative effort has resulted in successful islet isolations (>5,000 IEQ/kg recipient weight and >50% purity) approximately 67% of the time. The purpose of this study was to describe the principles of pancreatic procurement for PIT, which we feel have allowed us to achieve a successful islet isolation rate despite the use of a distant pancreatic islet isolation center.
MATERIALS AND METHODS
Between January 2002 and June 2003, 39 pancreata were procured and shipped to a distant islet isolation center. All data for this study was collected prospectively under protocols approved by the Baylor College of Medicine and University of Miami Institutional Review Boards and the U.S. Food and Drug Administration. The conditions for shipment of pancreata to the University of Miami were in compliance with United Network of Organ Sharing policy #5.0–5.5, 5.7.
After informed consent was obtained, human pancreata were obtained from multi-organ, brain-dead, cadaveric donors. A midline incision was made from the suprasternal notch to the pubic symphysis. After a thorough laparotomy was performed, the aorta and inferior mesenteric vessels were identified and prepared for later cannulation. The lesser sac was opened by dividing the gastrohepatic and the gastrocolic ligaments, and the pancreas was visually inspected. The short gastric arteries and veins were then ligated and divided, and the stomach was completely freed from the spleen. With division of the gastrocolic ligament, the splenic flexure of the colon was taken down. The middle colic vessels were double ligated and divided, and the first and fourth portions of the duodenum were divided with a GIA linear cutting stapler (Endo GIA, U.S. Surgical, Norwalk, CN). After division of the duodenum, the spleen and body and tail of the pancreas were mobilized toward the midline by dissection from their retroperitoneal attachments. Throughout the dissection, care was taken to avoid manipulation of the pancreas in an attempt to minimize any injury to the organ or vasospasm. In addition, in an attempt to avoid arterial vasospasm or vascular injury, the splenic and superior mesenteric arteries (SMA) were left intact, and no vessel loops were placed around these vessels. The gastroduodenal artery (GDA) was also left intact in an effort to maximize the perfusion of the donor pancreas. The aorta was cannulated just proximal to its infrarenal bifurcation in the usual fashion while special care was taken to make sure the inferior mesenteric vein (IMV) cannula did not enter the splenic or portal veins to assure that the venous drainage of the pancreas was not mechanically impaired. After cannulation of the aorta and IMV, 3 and 2 L of University of Wisconsin (UW) solution (Dupont Pharmaceutical, Wilmington, DE) was perfused through the aortic and venous cannulas, respectively. Of note, we kept the arterial line 2 feet above the venous line to maintain a slight pressure differential between the arterial and venous systems, in an attempt to maintain the venous drainage of the pancreas. After the UW perfusate was started, the entire pancreas was packed both anteriorly and posteriorly with copious amounts of saline slush. Upon completion of the flush and before liver procurement, the pancreas was rapidly excised en bloc with the spleen. The spleen was subsequently removed on the back table before placing the pancreas into the perfluorocarbon (PFC, Fluoromed, LP, Round Rock, TX) and UW two-layer solution.
After removal from the body, the pancreas was stored using a two-layer cold storage method (5) and held in position between the PFC-UW interface with a sterilized precut screen. The pancreas was then shipped to the University of Miami by way of charter jet for islet isolation and processing. Total transit time ranged from 5 to 7 hours to Miami and then 5 to 7 hours back to our transplantation center.
Upon arrival at University of Miami, Diabetes Research Institute (DRI), the pancreas was digested using Liberase enzyme (Roche, Welwyn Garden City, UK), which was injected into the proximal and distal pancreatic ducts. The pancreatic islets were then separated by mechanical dissociation using the Ricordi chamber and Ficoll-diatrizoic acid gradients (Seromed-Biochron, Berlin, Germany) (6). An isolation was considered successful if more than 5,000 IEQ/kg recipient body weight was obtained. After completion of the digestion and islet isolation, the islets were flown back to The Methodist Hospital (Houston, Texas) and transplanted as previously described (4).
Donor Characteristics, Pancreatic Islet Variables, and Data Analysis
The donor characteristics from cadaveric pancreatic donors were analyzed using Microsoft Excel (Redmond, WA), and medians with ranges were calculated for all recorded variables. Cause of death, donor age, sex, body mass index (BMI), blood pressure stability, plasma glucose, serum sodium, serum amylase, creatinine, minimum hourly urine output, and time from declaration of brain death to operative cross clamp were recorded. Operative variables were recorded as pancreatic weight, pancreatic excision time (time from aortic cross clamp to placement into preservation solution), cold ischemia time (time of cross clamp to start time of islet isolation), and type of preservations solution used. Pancreatic islet variables were IEQ per kilogram of recipient weight, pancreatic islet purity, viability, and sterility.
Between January 16, 2002 and June 30, 2003, 39 pancreata were procured and processed for PIT at a distant islet isolation center. Twenty-six of the 39 (67%) pancreata resulted in successful islet isolations. The median IEQ per kilogram transplanted per patient was 6,854 (5,332–9,851), with a median purity of 65% (35–90%). The median islet viability was 100% (87–100%). All islet preparations were sterile, and there was no evidence of endotoxin contamination.
The following results, presented in Table 1, were obtained from our successful cadaveric donors and are not representative from all attempted isolations. Specifically, there were 19 male and 7 female donors with a median age of 37 (18–55) years. The median BMI was 30.75 (19.81–41.99). Cause of death was traumatic for 16 donors and cerebral vascular in origin for 10 donors. All donors were hemodynamically stable with adequate renal function. Median pancreatic weight was 107.9 (67.8–206.3) g. The pancreas was rapidly excised from the abdominal cavity, with median time to excision from the abdominal cavity being 34 (17–76) minutes. PFC/UW was used in 22 (84%) of the successful isolations. The median cold ischemia time was 6 hours and 40 minutes (3 hours, 12 minutes to 14 hours, 56 minutes).
PIT still represents a challenge for transplant centers. One of the steps involved in successful transplantation is obtaining sufficient purity and quantity of pancreatic islets. For successful transplantation, we feel a minimum of 5,000 IEQ/kg recipient weight must be transplanted, with an islet purity of 50%. Previous reports have described islet recovery rates ranging from 55% to 61% (2, 3), which demonstrates the difficulty of successful islet isolation. There are numerous factors that affect islet purity and yield. Lakey et al. (2) identified multiple variables that affect islet isolation. One factor was local versus distant procurement teams. They found that with local procurement teams, the islet isolation rate was 69%, whereas with a distant procurement team, it was 45%.
Because we use a distant pancreatic islet isolation center, we decided to analyze the results from our islet isolation experience. Our data shows that our islet isolation rate is 67%, which is higher then previously published results when a local procurement team was not used. Even though the median cold ischemia time was 6 hour and 40 minutes, we postulated that the success of our PIT isolations when using a distant islet isolation center was caused by specific attention to the method of pancreatic procurement.
Our procurement technique involves some variations in the procurement techniques, as described previously in Methods. In our procurement technique, the pancreas is excised before liver procurement. The livers procured after pancreatic excision have demonstrated no decrease in viability or function, and we feel that this method does not have an adverse impact on livers procured after pancreatic procurement for PIT. Furthermore, we adhere to a number of other principles during pancreatic procurement for PIT, beginning with particular attention to careful dissection of the pancreas and its vasculature. This prevents any undue parenchymal injury or vascular disruption to the organ. We avoid the use of vessel loops, and the GDA, SMA, and splenic arteries are not dissected. In situ vascular flush is performed shortly after the predissection is completed and before pancreatectomy. We feel that in situ flush decreases operative time and has no detrimental impact on islet recovery results. Kneteman et al. (7) also found similar results when they compared two groups of pancreatic recovery techniques: initial pancreatectomy followed by ex vivo flush versus pancreatectomy after in situ vascular flush. Their study demonstrated that in situ pancreatic flush had equal islet recovery rates versus the use of initial pancreatectomy followed by flush, but operative time was decreased, and technical procurement was easier.
During cannulation, to maximize the arterial and venous pressure differential and prevent pancreatic congestion, the aortic cannula is always elevated 2 feet above the venous cannula. Of note, the IMV cannula is carefully placed to prevent migration into the splenic or portal veins. Careful IMV cannula placement prevents any venous outflow obstruction, thereby decreasing the possibility for pancreatic congestion. Pancreatic edema has been shown to correlate with decreased islet recovery and viability (8) and suboptimal separation of the pancreatic islets during processing.
After the in situ flush is started, rapid organ cooling is then performed using saline slush. In animal models, warm pancreatic ischemia has been identified as a factor that reduces pancreatic islet recovery (9). Lakey et al. (10) have previously described improved islet recovery in humans by maintaining the core pancreas temperature at 4° Celsius. With our procurement technique, the lesser sac is opened, and the pancreas is mobilized. Elevation of the pancreatic tail and mobilization of the spleen allows for slush to be packed onto the posterior and anterior surface of the pancreas. Decreasing the core pancreas temperature possibly prevents activation of pancreatic autolytic pathways, thereby preserving islet viability. Finally, the pancreas is rapidly removed from the abdominal cavity and placed in the UW/PFC solution and transported to DRI by private charter jet for islet isolation.
In conclusion, we have reported a number of modifications to the standard pancreatic procurement technique for PIT. These modifications have improved our pancreatic islet isolation rate to 67% when using a distant islet isolation center as well as maintenance of more than 5,000 IEQ/kg recipient weight.
1. 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.
2. Lakey JR, Warnock GL, Rajotte RV, et al. Variables in organ donors that affect the recovery of human islets of Langerhans. Transplantation 1996; 61: 1047.
3. Benhamou PY, Watt PC, Mullen Y, et al. Human islet isolation in 104 consecutive cases. Transplantation 1994; 57: 1804.
4. Goss JA, Schock AP, Brunicardi FC, et al. Achievement of insulin independence in three consecutive type-1 diabetic patients via pancreatic islet
transplantation using islets isolated at a remote isolation center. Transplantation 2002; 74: 1761.
5. Matsumoto S, Kandaswamay R, Sutherland DE, et al. Clinical application of the two-layer (University of Wisconsin solution/perfluorochemical plus O2) method of pancreas preservation before transplantation. Transplantation 2000;70: 771.
6. Ricordi C, Lacy PE, Finke EH, et al. Automated method for isolation of human pancreatic islets. Diabetes 1988; 37: 413.
7. Kneteman NM, Lakey JR, Kizilisik TA, et al. Cadaver pancreas recovery technique. Impact on islet recovery and in vitro function. Transplantation 1994; 58: 1114.
8. Mahler R, Franke FE, Hering BJ, et al. Evidence for a significant correlation of donor pancreas morphology and the yield of isolated purified human islets. J Mol Med 1999; 77: 87.
9. Corlett MP, Scharp DW. The effect of pancreatic warm ischemia on islet isolation in rats and dogs. J Surg Res 1988; 45: 531.
10. Lakey JR, Kneteman NM, Rajotte RV, et al. Effect of core pancreas temperature during cadaveric procurement on human islet isolation and functional viability. Transplantation 2002; 73: 1106.