Button images credited to:

Patient Care Online

Wake Forest University



Korean Medical Library Engine

Center for Engineering in Medicine


Discovery and clinical trials Donor Selection Recipient selection The Procedure Differences between the Edmonton Protocol and previous methods

Limitations, Risks and Potential Problems

Reproducibility of the Edmonton Protocol

Current Statistics

Examining Costs



Discovery and clinical trials

The Edmonton Protocol was developed by James Shapiro and Jonathan Lakey of the University of Alberta , Edmonton. They were the first to publish results indicating that islet transplantation could produce insulin independence in humans with excellent metabolic control. The main difference between the Edmonton procedure and its predecessors, and the suspected reason for Edmonton 's success, was their novel method of using a glucocorticoid-free immunosuppression regimen and of infusing an adequate islet mass, obtained from multiple donors, into each recipient.









Fig. 1 & 2 Dr. James Shapiro (left) and Dr. Jonathan Lakey (right)


In the Edmonton series, islet transplantation was performed on seven patients with type 1 diabetes who had a history of severe hypoglycemia and metabolic instability. The immunosuppressive regimen used, unlike those in earlier procedures, omitted agents with known islet toxicity. Also unlike earlier efforts, the Edmonton series used two or three infusions of islets obtained from two or more donors to gain the necessary islet mass. This method has been an overwhelming success, as over 17 consecutive patients have become insulin independent provided that the mass of transplanted islets was more than 9,000 islet equivalents/kg of recipient body weight.



Donor Selection

Accuracy in donor selection is essential in achieving the critical mass necessary for the success of the procedure. Presently, donor selection for the Edmonton Protocol is very similar to the criteria required for whole pancreas donors. However, experiments done on the effects of age, non–heart-beating donors, and pancreatic tissue fibrosis indicate that these criteria can be expanded to include organs that are now rejected for clinical use. These include pancreases with vascular abnormalities and/or arteriosclerosis. Increased donor age (especially between ages 51 and 65 years) and a high body mass index (greater than 24) are two factors that have been found to positively impact the outcome of transplantation. In contrast, some factors that had a negative correlation with outcome include prolonged hypotensive episodes prior to or hyperglycemia at the time of death, higher frequency and longer duration of cardiac arrest, and increased duration of cold storage (ischemia time) before isolation.

The fact that older donor pancreases seem to be viable for islet transplantation can prove very helpful in increasing the pool of available donors since these organs are not acceptable for whole pancreas transplantation. Recent research also shows that non–heart-beating donors should be considered as potential sources of islets.


Recipient selection

Islet transplantations have traditionally been offered only to diabetic patients with end-stage renal complications. The Edmonton Protocol has made it feasible to successfully reverse diabetes by transplanting solitary islets into brittle type 1 diabetes patients. Due to this change, the criteria for islet transplant recipient selection are as follows:

  • Living with type 1 diabetes for a period of at least 5 years
  • Hypoglycemic unawareness
  • Metabolic instability as shown by hypoglycemia or ketoacidosis, chaotic glucose profiles, or disruption in lifestyle
  • Progressive secondary complications despite adherence to insulin regimen


The Procedure

Fig. 3 Islet Transplantation in the Edmonton Protocol

The Edmonton Protocol is not major surgery. The donor pancreas is infused with collagenase to separate the islets from exocrine and ductal tissue. The islets are then purified by density-gradient centrifugation and evaluated for purity and viability before transplantation. Some scientists disagree on whether the purification or culturing is needed, saying that it results in lost cells and lost time. When the cells have been harvested, the patient is locally anesthetized and a very thin needle and x-rays are used to locate the portal vein. The islet cells are injected into the portal vein via catheter from which they can travel to the liver, develop a blood supply and begin producing insulin, partly due to the capability of the liver to regenerate, build new blood vessels and make new supporting tissue when damaged.


Differences between the Edmonton Protocol and previous methods of Islet Transplantation

Previous attempts at islet isolation had very low success rates and patients remained insulin independent for only short amounts of time. The main difference between the Edmonton Protocol and previous methods of islet transplantation is the new combination of immunosuppressive drugs used, as well as the fact that islets from two or more donor pancreases are used in order to increase the number of islets cells transplanted.

More specifically, the Edmonton Protocol can be thought of as having four novel approaches to islet transplantation:

  1. In order to optimize islet function, the total cold ischemia time was minimized by transplanting islets immediately rather than culturing them for days before infusion. Older methods of islet transplantation cultured the cells for several days before transplantation. While culturing cells allowed them to recover from the purification process and also gave patients more time to travel to the center where the procedure would be performed, some decreased islet function was noted as a result of culturing. The Edmonton procedure sought to fix this.
  2. Fetal calf serum was eliminated from the isolation process in order to minimize islet immunogenicity. The reasoning behind eliminating this step is the elimination of the culturing process by the Edmonton group altogether.
  3. Immunosuppressive therapy no longer consisted of glucocorticoids, but instead included a combination of sirolimus, tacrolimus, and daclizumab.
  4. Transplants from more than one donor were used in order to increase the chances of insulin independence by a quantitative increase in the number of transplanted cells.

Many comparisons have been made between the Edmonton Protocol and whole pancreas transplantation. Pancreas transplantation is performed using an intra-abdominal approach and is considered major surgery. Consequences of major surgery include delayed return of bowel function and more complicated post-operative care. Complications of pancreas transplants include vascular thrombosis, pancreatitis, as well as infection, which could lead to prolonged hospital stays and repeat operations. If bladder drainage of pancreatic exocrine secretions occurs in pancreas transplantation, patients can develop intractable metabolic acidosis, dehydration, chronic urethritis, urinary leaks, and recurrent urinary tract infections. If systemic drainage is used, there is risk of hyperinsulinemia and hypoglycemia. Another important concern of pancreas transplants is the great difficulty of detecting rejection. Studies also show that patients receiving a pancreas transplant have a higher mortality than diabetes patients still waiting for a transplant. Overall, pancreas transplantation has a high morbidity and mortality rate, and does not show significant improvement for complicated diabetes cases. Islet transplantation, despite uncertainties in long-term results, remains a less invasive and more effective procedure.

Limitations, Risks and Potential Problems

Islet cells from two donor pancreases are required per patient undergoing this procedure. There are only about 3,000 cadaveric pancreases available per year. This leaves a maximum of 1,500 patients that can be helped through this procedure per year. This figure is optimistic since it does not include pancreases that will be used for pancreas transplants, nor does it take into account cells lost due to harvesting and purification techniques.

The actual procedure does not come without risk. Gaining access to the portal vein is not an easy task, and risks include portal vein thrombosis and bleeding. More specific adverse effects due to the procedure include: bleeding episodes related to puncture of the liver, partial thrombosis of the portal vein with hepatic subscapular hemorrhage caused by anticoagulant therapy that required transfusion and surgery, elevated liver-enzyme levels, hypercholesterolemia, increased dosages of antihypertensive drugs, as well as a need for retinal laser photocoagulation. Despite all this, there have been no deaths or life-threatening infections to date.

Another cause for concern is the lack of knowledge on the long term effects of the immunosuppressive drugs used in the Edmonton Protocol due to the recent establishment of the procedure. The long-term safety of the islets themselves is not known, and animal models have shown that they do produce effects on nearby liver cells. The main problem with transplanting cells in the liver is the inability of the islet cells to release glucagon in hypoglycemia. One of the functions of normal islet cells is to secrete glucagon in cases of low blood sugar and transplanted islet cells have been shown to not perform this task. Careful monitoring of insulin levels is therefore necessary because of this imperfect glycemia control. Successive islet infusions can also result in increased portal pressure, and the long-term effects of this are also unknown. Patients with functioning grafts after three years of the procedure have been identified, but more follow-up data is needed. Exhaustion of cells over long periods of time is quite likely and follow-up studies are needed in order to confirm or refute this.


Reproducibility of the Edmonton Protocol

Due to the severe shortage of cadaveric donors, an increasing number of individuals on transplant waitlists, and the requirement of multiple donors for each recipient, it is unlikely that islet transplantation will become a widespread practical treatment for diabetes. This is made even more unlikely by the fact that, in comparison, whole pancreas transplantation requires only a single donor organ. A study by Markmann et al (2003) examines the results obtained in Edmonton , not only in terms of the effectiveness of the immunosuppressive regimen, but also the necessity for multiple transplants. This study found that all seven patients who completed the treatment protocol achieved insulin independence. However, unlike the Edmonton series, five of these patients required only a single infusion of islets. This is highly desirable not only in terms of the shortage of donor organs, but also due to the fact that the second and subsequent infusions were associated with a greater risk of portal vein thrombosis. Markmann's study also found that there was a steep learning curve involved in the technique of islet isolation. Perfection of this technique could greatly contribute to maximizing the number of islets obtained from each patient and further reduce the likelihood of more than one donor being required.

Overall, the islet isolation part of the procedure seems to be the most technically challenging, thus limiting reproducibility of the stellar results of the Edmonton group's original trial. The condition of the individual donor (his/her patient history), the skill with which surgeons remove the pancreas, as well as the method used to transport the pancreas are all factors to consider.

Current Statistics

As of April 2003, 48 patients have undergone 92 islet infusions in Canada . At that time, 15 centers were performing the procedure, with over 160 patients worldwide. The one-year insulin-dependence rates are comparable to receiving pancreas transplants. In January 2003, 84% of patients were insulin free after one year of transplantation and 89% of those patients were still producing insulin three years after transplantation. Compared to previous islet cell transplants with around 8% success rates, the Edmonton Protocol was a tremendous improvement. Nevertheless, even the most successful transplants show little insulin peaks when glucose is injected intravenously compared to normal individuals.

Examining Costs

A specially designated facility complying with good manufacturing practice (GMP) standards as well as a well-trained staff and technicians are required in order to examine cost-benefit ratios. The cost of isolating human pancreas islets is $20,000 per gland. At 2-4 organs per patient, this results in over $100,000 per patient per procedure. This can be compared to the cost of a whole pancreas transplant, which averages $45,000 per procedure. Immunosuppressive regimens that are necessary after the procedure may exceed $40,000 per year in cost. A diabetes patient receiving insulin treatments via insulin pump incurs costs of $5000 for the pump and an additional $1500 per year for related supplies.



1. Ault, A. "Edmonton's islet success tough to duplicate elsewhere." The Lancet 361(9374):2054, June 14 2003.

2. Hathout, E., J. Lakey and J.Shapiro. “Islet transplant: an option for childhood diabetes?” Archives of Diseases in Childhood 88(7): 591-4, July 2003.

3. Hirshberg, B. et al. "State of the art: islet transplantation for the cure of type 1 diabetes mellitus." Reviews in Endocrine & Metabolic Disorders 4(4):381-9, December 2003.

4. Hirshberg, B., Rother, K.I. and D.M. Harlan. "Islet transplantation: where do we stand now?" Diabetes/Metabolism Research and Reviews 19(3):175-8, May-June 2003.

5. http://www.diabetes.org.uk/islets/trans/edmonton.htm

6. http://www.joslin.harvard.edu/news/islet_transplant_july.shtml

7. Markmann, James F. et al . “Insulin independence following isolated islet transplantation and single islet fusions.” Annals of Surgery 237(6): 741-750, June 2003.

8. Shapiro, A.M., J. Lakey et al . “Islet preparation in seven patients with type I diabetes mellitus using a glucocortoid-free immunosuppressive regimen.” New England Journal of Medicine 343(4):230-8, 2000.

9. Shapiro, A.M., Ricordi C. and B. Hering. "Edmonton's Islet success has indeed been replicated elsewhere." The Lancet 362(9391):1242, October 11 2003.

10. Rastellini, Cristiana. “Donor and recipient selection for islet transplantation.” Current Opinion in Organ Transplantation 7(2):196-201, June 2002.

11. Robertson, R.P. "Islet transplantation as a treatment for diabetes - a work in progress." New England Journal of Medicine 350(7):694-705, February 12 2004.

Image References:

Fig. 1: http://www.expressnews.ualberta.ca/expressnews/articles/news.cfm?p_ID=2464&s=a

Fig. 2: http://www.expressnews.ualberta.ca/expressnews/articles/news.cfm?p_ID=4951&s=a

Fig. 3: http://www.islet.com/islet.html