Abstract
Total or partial pancreatectomy followed by autologous islet transplantation is a therapeutic option for the treatment of refractory chronic pancreatitis (CP). Maximization of islet yields from fibrotic and inflamed organs is crucial for prevention of posttransplant diabetes. We adapted technical advancements developed for islet allotransplantation toward islet autotransplantation. Eight patients (two men, six women; ages 24–58 years) underwent total (n = 7) or partial (n = 1) pancreatectomy for the treatment of CP refractory to maximal medical management. Pancreata were preserved in UW solution (UW group) in initial three cases and the last five pancreata were preserved with pancreatic ductal injection followed by ET-Kyoto/oxygenated PFC solutions (DI+TLM group). Islets were isolated by modified Ricordi method and were purified only in one case. All islet infusions were performed under general anesthesia via direct vein injection into the portal venous system with pressure monitoring. Total islet yields (129,314 ± 51,627 vs. 572,841 ± 116,934 IEQ, p < 0.04), islet yield/pancreas weight (1,233 ± 359 vs. 6,848 ± 847 IEQ/g, p < 0.003), and islet yield/patient body weight (1,951 ± 762 vs. 7,305 ± 1,531 IEQ/kg, p < 0.05) were significantly higher in the DI+TLM group when compared to the UW group. Pellet size was also higher (5.3 ± 0.3 vs. 13.5 ± 3.4 ml) in the DI+TLM group, suggesting that this method of preservation effectively protected pancreatic tissue against autolysis. First month posttransplant basal C-peptide and the secretory unit of islet transplant objects (SUITO) index were also higher in the DI+TLM group when compared to the UW group (2.0 ± 0.3 vs. 1.4 ± 0.4 ng/ml and 42.6 ± 12.7 vs. 14.6 ± 5.6, respectively). There were no technical complications related to the infusion. Our results suggest that higher islet yields can be achieved even from chronically inflamed and fibrotic organs using DI+TLM. The techniques applied for islet isolations from normal pancreata are showing promise for fibrotic pancreata from CP patients.
Keywords
Introduction
Chronic pancreatitis (CP) is a debilitating inflammatory disease that causes permanent damage to both the endocrine and exocrine sections of the pancreas. Although precise etiology of CP is unclear, several causative factors have been identified, including genetic predisposition, alcohol use, and autoimmunity (25). Medical management of CP consists of analgesics, pancreatic enzyme replacement, and surgical intervention. The primary objective of surgery for CP is pain relief. A secondary objective is to preserve endocrine function that has not yet been lost to the disease process. It is estimated that at least 50% of CP patients will ultimately require some form of surgical intervention and roughly 30–50% of CP patients who forego surgery will eventually become diabetic (3,21). Hence, distal or total pancreatectomy followed by islet autotransplantation is a practical approach to prevent diabetes in these patients. The type of surgical procedure will be based on endocrine and exocrine function, pancreatic ductal pathology, and intensity of pain.
The first demonstration that islets isolated from resected pancreas could function as an autograft in humans was in 1977, when Sutherland et al. performed intraportal infusion of islets after near-total pancreatectomy (23). The islet transplant recipient remained insulin independent, as well as pain free, for 6 years of her life (26). Therefore, islet autotransplantation after partial or total pancreatectomy could alleviate pain in these patients and preserve endocrine function (3). Islet autotransplantation has been performed in more than 300 patients worldwide, with the longest successful islet graft survival being more than 13 years (3,20). While some patients achieve insulin independence after islet autotransplantation, the other patients with less optimal outcomes require exogenous insulin to treat hyperglycemia. The rate of insulin independence varies among different islet cell processing centers (3). However, even if insulin independence is not achieved, transplanted islet cells could maintain excellent glycemic control. Factors correlated with postoperative insulin independence included female sex, weight, body mass index (BMI) <28, and the number of islet equivalents (IEQ) infused per kilogram weight of the patient (2). Posttransplant analyses of islet autotransplant patients have shown significant improvement in health-related quality of life in most of the patients and further prevention of diabetes mellitus in a significant number of cases (21).
The quality of pancreatic tissue retrieved from CP patients undergoing total or partial pancreatectomy is inferior from the pancreas retrieved from deceased donors. Pancreas excised from CP patients is often fibrotic with calcifications and damaged by prolonged inflammation (7,17). Hence it is widely acknowledged that isolating a sizable amount of islet mass from such pancreata is problematic (17).
Previously we have shown that pancreatic ductal injection (DI) of extracellular-type trehalose-containing Kyoto (ET-Kyoto) solution immediately after pancreas procurement and pancreas preservation by the two-layer method (TLM) significantly increased islet yields (14,15). DI+TLM was originally developed for islet isolation from a non-heart-beating donor (12) and applied for living donor islet transplantation (13) in Japan. Currently this technique is also applied for islet isolation from heart-beating brain-dead donor pancreas (5,10,11).
In this study, we adapted the DI+TLM method for islet autotransplantation for the first time. We obtained significantly improved islet yields using DI+TLM and describe the improved islet autotransplant outcome in this article.
Materials and Methods
Ethical Guidelines
This study was approved by the institutional review boards.
Patients and Surgery
We have performed eight islet autotransplantations into CP patients from October 2006 to November 2008. The diagnosis of CP was based on the patient's medical history, laboratory test results, and clinical image studies including endoscopic retrograde cholangiopancreatography (ERCP) or endoscopic ultrasonography (EUS). The indication for surgery in all patients was intractable pain and all used narcotics chronically for analgesia. All operations were performed under the guidance of attending surgeons. Eight patients (two men, six women; ages 24–58 years) underwent total (n = 7) or partial (n = 1) pancreatectomy for the treatment of CP refractory to maximal medical management.
Pancreas Preservation
The first three pancreata were preserved in University of Wisconsin solution (UW group) whereas last five pancreata were preserved with pancreatic ductal injection followed by ET-Kyoto solution/oxygenated perfluorochemical (PFC) solutions (15) (DI+TLM group). For the DI+TLM group, the islet team surgeons removed the duodenum and spleen from the pancreas on the back table at the operating room. The pancreas was weighed and a cannula was immediately inserted into the main pancreatic duct from the direction of the pancreatic head. Approximately 1 ml/g pancreas of ET-Kyoto solution (Otsuka Pharm Factory Inc., Naruto, Japan) was administered intraductally before preservation in oxygenated perfluorodecalin (11,12). For the UW group, the ductal injection process was not performed and organ was preserved in cold UW solution.
Islet Isolation
Islets were isolated by modified Ricordi method (19) and purified in two cases. Liberase HI (Roche Molecular Biochemicals, Indianapolis, IN) was used for all cases of the UW group and two cases of the DI+TLM group. Collagenase NB with neutral proteases (Serva Electrophoresis GMbH, Heidelberg) was used for the other three cases of the DI+TLM group. In one case in the DI+TLM group, the islets were purified with a continuous density gradient using Biocoll in a chilled apheresis system (COBE 2991 Cell Processor, CaridianBCT, Denver, CO).
Islet Characterization
Islet yield was determined using dithizone staining (2 mg/ml; Sigma Chemical Co., St. Louis, MO) under optical reticule and converted into a standard number of IEQ (diameter standardizing to 150 μm) (14). Purity was assessed by comparing the relative quantity of dithizone-stained tissue to unstained exocrine tissue. Islet viability was evaluated using fluorescein diacetate (FDA) and propidium iodide (PI) staining to visualize living and dead cells simultaneously. All the islet preparations were tested for sterility using gram and fungal stain followed by 2-week aerobic and anaerobic cultures.
Islet Transplantation and Follow-up
All islet infusions were performed under general anesthesia via direct vein injection into the portal venous system with pressure monitoring in the operating room. Initial postoperative care was in the surgical intensive care unit where either a continuous insulin drip or intermittent insulin dose was administered by sliding scale protocol. Posttransplant islet function was assessed by daily plasma glucose levels, serum C-peptide, and daily insulin requirement. Furthermore, engraftment and function of transplanted islets was assessed by secretory unit of islet transplant object (SUITO) index, which was originally described by Matsumoto et al. (9,16). It is calculated using a simple formula: fasting C-peptide (ng/ml)/[fasting blood glucose (mg/dl) - 63] × 1500. This index is quite useful to decide whether the patient has sufficient islet mass to achieve insulin independence.
Statistical Analysis
Values for the data collected represent means ± SE. Two groups were compared using unpaired t-test. Ratio between two groups was compared with Fisher's exact test. A value of p < 0.05 was considered significant.
Results
Patient Characteristics
Characteristics of patients described in this study are summarized in Table 1. There was no significant difference between UW and DI+TLM groups in the ratio of gender, average age, average body weight, and average body mass index. Total pancreatectomy was performed on the first seven patients (three in the UW group and four in the DI+TLM group). The eighth patient (DI+TLM group) had undergone choledochoduodenostomy followed by a Whipple pancreaticoduodenectomy 5 years earlier and completion pancreatectomy was performed.
Patient Characteristics
UW, University of Wisconsin solution; DI, ductal injection; TLM, two-layer method.
Pancreas and Islet Characteristics
There was no significant difference between the two groups in terms of pancreas weight (Table 2). However, due to varying degrees of fibrosis and calcifications, the excised pancreatic tissue mass without previous operation (N =3 in the UW group, N = 4 in the DI+TLM group) ranged between 69 to 120 g. Tissue mass from the partial pancreatectomy procedure (N= 1 in the DI+TLM group) was 27.2 g. The cold ischemia time for all the isolations was less than 60 min.
Pancreas and Islet Characteristics
UW, University of Wisconsin solution; DI, ductal injection; TLM, two-layer method.
Total islet yields (129,314 ± 51,627 vs. 572,841 ± 116,933 IEQ, p < 0.04), islet yield/pancreas weight (1,233 ± 359 vs. 6,848 ± 847 IEQ/g, p < 0.003), and islet yield/patient body weight (1,951 ± 762 vs. 7,305 ± 1,531 IEQ/kg, p < 0.05) were significantly higher in the DI+TLM group when compared to the UW group (Fig. 1). Pellet size was also higher (5.3 ± 0.6 vs. 13.5 ± 7.6 ml) in the DI+TLM group (Table 2), suggesting that this method of preservation effectively protected pancreatic tissue against autolysis. There was no significant difference between the two groups in viability and viabilities of both groups were quite high. All but one of the final islet preparations tested negative for microbial contamination following 14 day culture period. Islet preparation from the second patient (UW group) was negative for fungal contamination but showed presence of Staphylo-coccus epidermis.
Total islet yield (right), islet yield per gram pancreas (middle), islet yield per patient body weight (left) in the University of Wisconsin (UW) group and ductal injection (DI) plus two-layer method (TLM) groups.
Posttransplant Islet Graft Function
There were no technical complications related to the infusion. One patient developed partial portal vein thrombosis from extrinsic portal vein compression but responded well to anticoagulation.
Posttransplant islet function was followed by plasma C-peptide levels and SUITO index and the details are shown in Table 3. All the patients reached normal levels of C-peptide levels (≥ 0.9 ng/ml), indicating excellent islet function. The posttransplant peak C-peptide level in patients from the UW group (1.4 ± 0.4 ng/ml) did not significantly differ from the DI+TLM group (2.0 ± 0.3 ng/ml). However, only one out of three patients from the UW group attained insulin-independent status when compared to four out of five patients from the DI+TLM group. The islet engraftment as measured by SUITO index was approximately three times higher in the DI+TLM group (42.6 ± 12.7) when compared to the DI+TLM group (14.6 ± 5.6) but did not reach statistical significance (p = 0.16).
Posttransplant Islet Graft Function
SUITO index, Secretory unit of islet transplant objects index.
In summary, by adopting key organ preservation steps developed for allogenic islet transplantation process, we were able to significantly improve islet yield using pancreas excised from patients with chronic pancreatitis.
Discussion
In the present study, we have successfully adopted the ductal injection of ET-Kyoto solution (DI) and the two-layer pancreas preservation method (TLM), which was originally developed for nondiseased pancreata to fibrotic pancreatitis pancreas for the first time. This new method could significantly increase total islet yield, islet yield per gram pancreas weight, and islet yield per patient body weight compared to original method.
Previous reports have shown that the ability to achieve insulin independence after islet autotransplantation correlates directly with the infused islet mass represented as islet equivalents per kilogram body weight of the recipient (2,4). In a group of 43 patients who underwent complete pancreatectomy and islet autotransplantation at the University of Minnesota from 2000 to 2004, a mean transplanted islet mass of 5,118 ± 587 IEQ/kg of recipient body weight correlated significantly with insulin independent status (4). Similarly, data from 45 islet autotransplant recipients at the University of Cincinnati showed that 40% of patients achieved insulin independence and the patients had a markedly higher transplanted islet dose (6,635 IEQ/kg) than the insulin-dependent group (3,799 IEQ/kg) (2). With our new method, islet yield/patient body weight was 7,305 ± 1,531 IEQ/kg, which was above the previous published data for insulin-independent and four out of five (80%) became insulin independent. Therefore, our new method should contribute to have excellent glycemic control after islet autotransplantation.
Our new method consists of the DI+TLM method. We introduced a large amount of DI for maintaining patency of pancreatic duct for the collagenase delivery as well as protecting islets and exocrine tissue from cold ischemic injury (12). Actually, DI could significantly reduced apoptotic cell death of islets and exocrine tissues after pancreas preservation (18). However, in the case of islet autotransplantation, cold ischemic time is relatively short; therefore, the other mechanism(s) might play a role for improving islet yields. One possibility is that the DI could facilitate rapid cooling of the pancreas. When we procure a pancreas from a cadaveric donor, the whole organ can be rapidly chilled using cold UW solution directly from abdominal aorta. However, in the case of total pancreatectomy for chronic pancreatitis, it is not possible to use cold UW solution to chill the pancreas before pancreas retrieval. Therefore, there is inevitable warm ischemic injury for total pancreatectomy. Warm ischemia deteriorates the quality of the pancreas more severely than the cold ischemia for islet isolation (24). Therefore, a large amount of ductal solution can facilitate rapid cooling of the pancreas and minimize warm ischemic injury.
A recent meta-analysis has concluded that the TLM of pancreas preservation significantly improved islet yields when compared to preservation in UW solution (1). Especially, the rate of successful islet isolations from marginal donor pancreas was higher in the TLM group based on this analysis (1). However, the precise mechanism(s) by which the two-layer preservation method improves islet yield remains unclear. The TLM could provide oxygen directly to the pancreas and restore the warm ischemic injury (6,8). Therefore, inevitable warm ischemic injury during total pancreatectomy might be restored during preservation by the TLM.
The posttransplant islet function was analyzed by the basal C-peptide concentration and also by the SUITO index, which was shown to predict insulin independence among allogenic islet transplant recipients (9,16). Posttransplant C-peptide levels and SUITO indexes were higher in the DI+TLM group; however, these parameters did not reach statistical significance in this small number of patients. After allogenic islet transplantation SUITO index more than 26 was associated with insulin independence (16). In this study for islet autotransplantation the SUITO index more than 24 was associated insulin independence. This indicates that the SUITO index is also a useful marker for predicting insulin independence for islet autotransplantation. The Edmonton protocol demonstrated that 10,000 IEQ/kg was necessary to achieve insulin-free status. Actually, in order to achieve SUITO index more than 26, it was necessary to transplant more than 10,000 IEQ/kg for allogenic islet transplantation (16). In sharp contrast, 2,559 IEQ/kg was the minimum number to achieve SUITO index 24 with insulin independence in our islet autotransplantation. Recently Sutherland et al. demonstrated that islet autotransplantation is much more effective than allotransplantation (22) and our results with islet autotransplantation support their observation.
In summary, higher islet yields were achieved even from chronically inflamed and fibrotic organs using DI and TLM preservation. The techniques applied for islet isolations from normal pancreata are showing promise for fibrotic pancreata from CP patients.
Footnotes
Acknowledgments
This study was supported, in part, by the All Saints Health Foundation and Baylor Health Care System. The authors thank Greg Olsen and Yoshiko Tamura for their technical support and manuscript preparation. The authors declare no conflict of interest.