Quality of life after total pancreatectomy with islet autotransplantation for chronic pancreatitis in Japan.

Chronic pancreatitis (CP) is an inflammatory disease of the pancreas that eventually culminates in irreversible fibrosis and loss of function. 1 Patients with CP often have severe and intractable abdominal pain that leads to decreased quality of life (QOL), inability to work or attend school, and increased health care costs due to repeated emergency room visits and hospitalizations. 2 , 3 The first-line treatment for CP includes a low-fat diet, pancreatic enzyme therapy to reduce pancreatic irritation, comprehensive pain management, and endoscopic sphincterotomy and stenting by endoscopic retrograde cholangiopancreatography (ERCP). 4 , 5 If medical or ERCP therapy is unsuccessful, surgical treatment, such as pancreatic duct drainage or parenchymal resection, may be considered, depending on the morphology of the pancreatic duct and tissue. Total pancreatectomy (TP) with islet autotransplantation (TPIAT) may also be considered in some patients, particularly those with diffuse small pancreatic duct disease, hereditary pancreatitis, or a history of failed surgery. 6–8 The aims of TP are to relieve pain and restore QOL, while islet autotransplantation (IAT) is intended to reduce the burden of postoperative diabetes. 9 According to the Collaborative Islet Transplant Registry (CITR), 827 IATs were performed from 1999 to September 2015 in North America, Europe, and Australia ( https://citregistry.org/system/files/1st_AR_Auto.pdf ), mostly with total or completion (≥95%) pancreatectomy. However, only a few IAT procedures with pancreatectomy, especially TP have been reported in Asia; there is one report of distal pancreatectomy, 10 and another of pancreatectomy, including 9 recipients who underwent partial pancreatectomy (50%–80%) and 1 who underwent TP. 11 We previously reported 5 Japanese CP patients who underwent TPIAT, identifying endogenous postoperative insulin secretion in all cases, 6 and here we report an additional follow-up concerning QOL before and after TPIAT.

Five patients (2 women and 3 men) underwent TPIAT at our hospital during the study period. Their characteristics are summarized in the previous report. 6 Briefly, their median age was 34 (range, 20–52) years and the indication for surgery was intractable abdominal pain in all cases. Two patients had alcohol-induced CP, 2 had hereditary pancreatitis with genetic mutations (one each in PRSS1 and SPINK1 ), and 1 had idiopathic CP. The median body mass index (calculated as kg/m 2 was 19.8 (range, 17.7–22.4). The median duration of CP was 12 (range, 4–31) years. Two of the 5 patients were preoperatively diagnosed with diabetes. The median islet yield was 270,967 (range, 116,068–467,042) islet equivalents (IEQ) and the median final islet yield was 5618 (range, 2267–13,010) IEQ/g of pancreas. 6 Purification was not performed in 4 of the 5 cases because there was severe atrophy of the exocrine tissue and a small tissue volume (<15 mL). The median viability was 94.8% (range, 89.5%–97.5%). Endotoxin levels were negative (<5 U/kg) in all patients. The median number of transplanted islets was 4149 (range, 2,038–10,836) IEQ/kg body weight. 6 In all cases, the prepared islets were transplanted through a catheter into the portal vein. The median portal vein pressure before transplantation was 9 (range, 7–13) mmHg and the highest median value after transplantation was 10 (range, 8–15) mmHg, indicating no significant increase in portal vein pressure. All patients had transient elevations in liver enzymes that resolved spontaneously. There were no serious adverse events associated with islet transplantation. The primary endpoint was the percentage of patients with improved pancreatic pain (a > 50% reduction in the Izbicki pain score 12 from the preoperative value) and good glycemic control without severe hypoglycemic episodes during the 12 months (365 ± 14 days) after surgery. Please refer to the Patients and Methods section for details. The primary endpoint was achieved in 4 patients. The success rate was estimated to be 80% (80% confidence interval: 0.35–1.00, p  = 0.0026). Because the lower limit of the confidence interval exceeded the threshold of 20%, significance was reached for efficacy. Four patients (80%) did not require narcotic analgesics at 12 months postoperatively. The remaining patient (#1) did not meet the pain criteria for the primary endpoint. But all patients’ Izbicki pain scores improved significantly 12 months postoperative from preoperative using Wilcoxon one-tailed signed rank test ( p  = 0.03125) ( Table 1 ). Moreover, there was an improvement in the visual analog scale (VAS) pain score for all patients, including this patient (#1) significantly analyzed by Wilcoxon one-tailed signed rank test ( p  = 0.03125) ( Table 1 ). Table 1. Izbicki pain scores and visual analog scale pain scores before and 12 months after the operation. Items Mean (Before surgery) Medians and IQRs (Before surgery) Mean (12 months after surgery) Medians and IQRs (12 months after surgery) p-value Signifi- cance Izbicki pain score 79.6 83 (66–90) 18 0 (0–3) 0.03125 * Visual analog scale pain score 9 9 (8–10) 0.3 0.3 (0–2.86) 0.03125 * Note: *, p  < 0.05. IQR, Interquartile Range. Izbicki pain scores and visual analog scale pain scores before and 12 months after the operation. Note: *, p  < 0.05. IQR, Interquartile Range. Although the physical, mental, and social aspects of the component summary scores based on the Short-Form-36 Health Survey version 2 (SF-36v2) did not show any significant changes, the average of each of the 3 component summary scores improved 1 year after TPIAT ( Figure 1(a) and Table 2 ). Among the 8 multi-item dimensions of health, role-physical ( p  = 0.03125), general health perception ( p  = 0.0077) and vitality ( p  = 0.035) improved significantly. As for the other 5 items, role-emotional ( p  = 0.070) was improved or unchanged in all patients, and the other 4 items improved, with the averages scores of physical functioning, bodily pain, social functioning, and mental health increased 12 months after surgery ( Figure 1(b) and Table 2 ). Figure 1. Comparison of quality of life (QOL) between before and 12 months after the operation, using the short-form 36 health survey version 2 (SF-36v2 Standard, Japanese). (a) the physical, mental, and social aspects of the component summary scores. (b) the 8 multi-item dimensions of health. *, p  < 0.05. QOL between before and 12 months after the operation using the Short-Form 36 Health Survey version 2 (SF-36v2 Standard, Japanese) was analyzed. A p-value of less than 0.05 was considered significant. (a) The physical, mental, and role/social component summary scores of 5 patients before and 1 year after TPIAT are described; 4 of the 5 patients improved in each of component summary scores. There were no significant differences in any of the 3 scores. For each component summary, the patient who did not improve was not the same person. (b) The 8 multi-item dimensions of health scores of the 5 patients before and 1 year after TPIAT are described. The 8 multi-item dimensions of health are physical functioning, role-physical, bodily pain, general health perception, vitality, social functioning, role-emotional, and mental health. Physical functioning improved in 3 patients, remained unchanged in 1 patient, and deteriorated in 1 patient. Role-physical improved for all with significant differences (p = 0.03125). Bodily pain improved in 3 patients, remained unchanged in 1 patient, and deteriorated in 1 patient. General health perception and vitality improved for all, with significant differences (p = 0.03125 and p = 0.03125, respectively). Social functioning improved in 4 of the 5 patients, without a significant difference. Role-emotional improved in 4 of the 5 patients; the other patient scored 100 and thus could not improve further. The changes in the 4 patients that improved were not significantly different. Mental health improved in 3 patients, remained unchanged in 1 patient, and deteriorated in 1 patient. Table 2. Comparison of quality of life (QOL) between before and 12 months after the operation, using the short-form 36 health survey version 2 (SF-36v2 Standard, Japanese). Items Mean (Before surgery) Medians and IQRs (Before surgery) Mean (12 months after surgery) Medians and IQRs (12 months after surgery) p-value Signifi- cance Physical component summary 42.08 37.7 (35.2–48.5) 46.88 48.5 (44.6–53.5) 0.15625 　 Mental component summary 38.04 35.3 (35.2–41.4) 47.68 46.7 (45.7–51.5) 0.0625 　 Role/social component summary 30.94 43.3 (12.3–44.8) 46.06 43.3 (37.9–59) 0.09375 　 Physical functioning 54.72 50.6 (36.2–54.2) 72.18 85 (54.2–100) 0.125 　 Role physical 32.84 22.5 (15.8–32.5) 68.64 75 (39.1–100) 0.03125 * Bodily pain 41.42 26.9 (22.4–35.4) 65.04 74 (40.3–84) 0.125 　 General health perception 27.84 27.1 (20–32.5) 46.42 46.9 (44.2–52) 0.03125 * Vitality 35.02 30.6 (24.1–37) 58.66 62.5 (49.8–68.8) 0.03125 * Social functioning 34.76 31.2 (11.9–37.7) 69.98 87.5 (31.2–100) 0.0625 　 Role emotional 39.04 35.3 (6.1–47.7) 68.28 66.7 (43.6–100) 0.0625 　 Mental health 47.26 43.8 (38.4–49.1) 57.98 55 (46.5–75) 0.1875 　 Note: *, p  < 0.05. IQR, Interquartile Range. Comparison of quality of life (QOL) between before and 12 months after the operation, using the short-form 36 health survey version 2 (SF-36v2 Standard, Japanese). (a) the physical, mental, and social aspects of the component summary scores. (b) the 8 multi-item dimensions of health. *, p  < 0.05. Comparison of quality of life (QOL) between before and 12 months after the operation, using the short-form 36 health survey version 2 (SF-36v2 Standard, Japanese). Note: *, p  < 0.05. IQR, Interquartile Range. Although global health status/QOL did not show a significant change, its average considerably improved ( Figure 2(a) and Table 3 ). Moreover, the averages of 4 of the 5 functional scales increased without significant differences, whereas that of the other scale (physical functioning) remained unchanged ( Figure 2(b) and Table 3 ). Among 9 symptom scales, pain ( p  = 0.0625) and financial difficulties ( p  = 0.0625) decreased or was unchanged in all patients without significance ( Figure 2(c) and Table 3 ). The averages of the fatigue, insomnia, appetite loss, constipation, and diarrhea scales decreased, while the other 2 symptom scales – nausea and vomiting as well as dyspnea – generally stayed the same in the 5 patients. Figure 2. Comparison of quality of life (QOL) between before and 12 months after the operation using the European organization for research and treatment of cancer quality of life questionnaire core 30 (EORTC QLQ-C30). (a) Global health status/QOL. (b) Functional scales. (c) Symptom scales. QOL between before and 12 months after the operation using the EORTC QLQ-C30 was analyzed. A p-value of less than 0.05 was considered significant. (a) The global health status/QOL of 5 patients before and 1 year after TPIAT are described; 4 of the 5 patients improved, but there was no significant difference. (b) The 5 functional scales of the 5 patients before and 1 year after TPIAT are described. The 5 functional scales are physical functioning, role functioning, cognitive functioning, emotional functioning, and social functioning. Each of the 5 functional scales improved in 3 patients, remained unchanged in 1 patient, and deteriorated in 1 patient. (c) The 9 symptom scales of the 5 patients before and 1 year after TPIAT are described. The 9 symptom scales are fatigue, nausea and vomiting, pain, dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties. Fatigue scale decreased (improved) in 3 patients and increased (worsened) in 2 patients. Nausea and vomiting scale decreased (improved) in 1 patient, increased (worsened) in 1 patient, and remained unchanged in 3 patients; 2 of these 3 patients originally scored 0 and thus could not improve further. The pain scale decreased (improved) in 4 patients and remained unchanged in 1 patient (p = 0.0625). The patient without improvement originally scored 0 and thus could not improve further. The dyspnea scale decreased (improved) in 2 patients, increased (worsened) in 1 patient, and remained unchanged in 2 patients; these 2 patients originally scored 0 and thus could not improve further. The insomnia scale decreased (improved) in 3 patients, increased (worsened) in 1 patient, and remained unchanged in 1 patient; this patient originally scored 0 and thus could not improve further. The appetite loss scale decreased (improved) in 1 patient, increased (worsened) in 1 patient, and remained unchanged in 3 patients; 2 of these 3 patients originally scored 0 and thus could not improve further. The constipation scale decreased (improved) in 3 patients, increased (worsened) in 1 patient, and remained unchanged in 1 patient; this patient originally scored 0 and thus could not improve further. The diarrhea scale decreased (improved) in 2 patients, increased (worsened) in 2 patients, and remained unchanged in 1 patient. The financial difficulties scale decreased (improved) in 4 patients and remained unchanged in 1 patient, with no significant change overall. The patient with no improvement originally scored 0 and thus could not improve further. Table 3. Comparison of quality of life (QOL) between before and 12 months after the operation using the European organization for research and treatment of cancer quality of life questionnaire core 30 (EORTC QLQ-C30). Items Mean (Before surgery) Medians and IQRs (Before surgery) Mean (12 months after surgery) Medians and IQRs (12 months after surgery) p-value Signifi- cance Global health status/QOL 35.002 33.33 (16.67–41.67) 66.668 75 (50–91.67) 0.09375 　 Physical functioning 82.668 86.67 (73.33–86.67) 82.666 100 (80–100) 0.375 　 Role functioning 53.334 50 (50–66.67) 80 100 (66.67–100) 0.1875 　 Cognitive functioning 63.332 83.33 (50–83.33) 73.334 66.67 (66.67–100) 0.3125 　 Emotional functioning 54.998 66.67 (33.33–83.33) 80 91.67 (75–100) 0.125 　 Social functioning 36.666 33.33 (0–50) 66.666 100 (33.33–100) 0.1875 　 Fatigue 53.334 55.56 (22.22–88.89) 28.89 11.11 (11.11–55.56) 0.21875 　 Nausea and vomitting 16.668 16.67 (0–16.67) 16.668 0 (0–16.67) 0.75 　 Pain 80 100 (100–100) 23.334 16.67 (0–33.33) 0.0625 　 Dyspnea 13.332 0 (0–33.33) 13.334 0 (0–0) 0.625 　 Insomnia 53.332 33.33 (33.33–100) 40.002 66.67 (0–66.67) 0.5 　 Appetite loss 33.332 33.33 (0–33.33) 20 0 (0–33.33) 0.5 　 Constipation 46.666 33.33 (0–100) 33.334 33.33 (0–66.67) 0.375 　 Diarrhea 46.666 33.33 (0–100) 39.998 33.33 (33.33–33.33) 0.5 　 Financial difficulties 60 66.67 (33.33–100) 13.334 0 (0–0) 0.0625 　 Note: *, p  < 0.05. IQR, Interquartile Range. Comparison of quality of life (QOL) between before and 12 months after the operation using the European organization for research and treatment of cancer quality of life questionnaire core 30 (EORTC QLQ-C30). (a) Global health status/QOL. (b) Functional scales. (c) Symptom scales. Comparison of quality of life (QOL) between before and 12 months after the operation using the European organization for research and treatment of cancer quality of life questionnaire core 30 (EORTC QLQ-C30). Note: *, p  < 0.05. IQR, Interquartile Range. Seven of the 18 items – pancreatic pain ( p  = 0.023), bloated abdomen ( p  = 0.25), night pain ( p  = 0.25), flatulence ( p  = 0.125), decreased muscle strength ( p  = 0.25), treatment side effects ( p  = 0.25), fear for future health ( p  = 0.125), and ability to plan ahead ( p  = 0.125)—were improved or unchanged in all patients ( Figure 3 and Table 4 ). Four items improved on average: the average of jaundice, body image, and sexual functioning decreased while satisfaction with health care increased. The averages of indigestion and dry mouth remained unchanged. The other 5 QOL scores worsened compared with before the operation: digestive function scores decreased on average, and the averages of altered bowel functioning, alcohol-related guilt, taste changes, and weight loss increased. Figure 3. Comparison of quality of life (QOL) between before and 12 months after the operation using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire pancreatic modification (EORTC QLQ-PAN28). QOL between before and 12 months after the operation using the EORTC QLQ-PAN28 was analyzed. A p-value of less than 0.05 was considered significant. The scores of 19 items for 5 patients before and 1 year after TPIAT are described. The 19 items are pancreatic pain, digestive function, jaundice, altered bowel functioning, body image, alcohol-related guilt, satisfaction with health care, sexual functioning, bloated abdomen, night pain, taste changes, indigestion, flatulence, weight loss, decreased muscle strength, dry mouth, treatment side effects, fear for future health, and ability to plan ahead. Pancreatic pain scores decreased (improved) in 4 patients and remained unchanged in 1 patient (p = 0.0625). The patient without improvement originally scored 0 and thus could not improve further. Digestive function scores increased (improved) in 1 patient, decreased (worsened) in 2 patients, and remained unchanged in 2 patients. Jaundice scores decreased (improved) in 3 patients, increased (worsened) in 1 patient, and remained unchanged in 1 patient; this patient originally scored 0 and thus could not improve further. Altered bowel functioning scores increased in 2 patients, decreased in 1 patient, and remained unchanged in 2 patients. Body image scores increased in 1 patient, decreased in 3 patients, and remained unchanged in 1 patient. Alcohol-related guilt scores increased in 3 patients and remained unchanged in 2 patients. Satisfaction with health care scores increased in 2 patients, decreased in 1 patient, and remained unchanged in 2 patients. Sexual functioning scores increased in 1 patient, decreased in 2 patients, and remained unchanged in 2 patients. Bloated abdomen scores decreased (improved) in 2 patients and remained unchanged in 3 patients; 1 of these 3 patients originally scored 0 and thus could not improve further. Night pain scores decreased (improved) in 2 patients and remained unchanged in 3 patients; 1 of these 3 patients originally scored 0 and thus could not improve further. Taste changes increased in 2 patients and remained unchanged in 3 patients. Indigestion scores decreased (improved) in 2 patients, increased (worsened) in 2 patients, and remained unchanged in 1 patient; this patient originally scored 0 and thus could not improve further. Flatulence scores decreased (improved) in 3 patients and remained unchanged in 2 patients; these 2 patients originally scored 0 and thus could not improve further. Weight loss scores increased (worsened) in 2 patients and remained unchanged in 3 patients. Decreased muscle strength scores decreased (improved) in 2 patients and remained unchanged in 3 patients; 2 of these 3 patients originally scored 0 and thus could not improve further. Dry mouth scores decreased (improved) in 1 patient, increased (worsened) in 1 patient, and remained unchanged in 3 patients; these 3 patients originally scored 0 and thus could not improve further. Treatment side effects scores decreased (improved) in 2 patients and remained unchanged in 3 patients; 2 of these 3 patients originally scored 0 and thus could not improve further. Fear for future health decreased (improved) in 3 patients and remained unchanged in 2 patients. Ability to plan ahead scores decreased (improved) in 4 patients and remained unchanged in 1 patient, showing no significant change overall (p = 0.125). The patient without improvement originally scored 0 and thus could not improve further. Table 4. Comparison of quality of life (QOL) between before and 12 months after the operation using the European organization for research and treatment of cancer quality of life questionnaire pancreatic modification (EORTC QLQ-PAN28). Items Mean (Before surgery) Medians and IQRs (Before surgery) Mean (12 months after surgery) Medians and IQRs (12 months after surgery) p-value Signifi- cance Pancreatic pain 55.558 66.67 (66.67–66.67) 22.22 33.33 (11.11–33.33) 0.0625 　 Digestive function 46.666 33.33 (0–100) 23.332 33.33 (0–33.33) 0.875 　 Jaundice 23.332 33.33 (0–33.33) 16.666 16.67 (0–33.33) 0.375 　 Altered bowel functioning 20 16.67 (0–33.33) 26.664 33.33 (33.33–33.33) 0.375 　 Body image 30.002 16.67 (16.67–50) 20 16.67 (0–33.33) 0.3125 　 Alcohol related guilt 0 0 (0–0) 20 16.67 (0–33.33) 0.125 　 Satisfaction with health care 53.336 66.67 (66.67–66.67) 60 66.67 (33.33–100) 0.625 　 sexual 40 0 (0–100) 16.666 0 (0–33.33) 0.25 　 Bloated abdomen 33.332 33.33 (33.33–33.33) 13.332 0 (0–33.33) 0.25 　 Night pain 40 33.33 (33.33–66.67) 19.998 33.33 (0–33.33) 0.25 　 Taste changes 0 0 (0–0) 13.332 0 (0–33.33) 0.25 　 Indigestion 33.332 33.33 (0–33.33) 33.334 33.33 (0–66.67) 0.625 　 Flatulence 46.668 66.67 (0–66.67) 13.332 0 (0–33.33) 0.125 　 Weight loss 0 0 (0–0) 26.666 0 (0–33.33) 0.25 　 Decreased muscle strength 46.666 33.33 (0–100) 20 0 (0–33.33) 0.25 　 Dry mouth 20 0 (0–0) 20 0 (0–33.33) 0.75 　 Treatment side effects 46.666 33.33 (0–100) 13.332 0 (0–33.33) 0.25 　 Fear for future health 80 100 (66.67–100) 46.666 33.33 (33.33–66.67) 0.125 　 Ability to plan ahead 53.332 33.33 (33.33–100) 6.666 0 (0–0) 0.125 　 Note: *, p  < 0.05. IQR, Interquartile Range. Comparison of quality of life (QOL) between before and 12 months after the operation using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire pancreatic modification (EORTC QLQ-PAN28). Comparison of quality of life (QOL) between before and 12 months after the operation using the European organization for research and treatment of cancer quality of life questionnaire pancreatic modification (EORTC QLQ-PAN28). Note: *, p  < 0.05. IQR, Interquartile Range. One of the primary endpoints is good glycemic control defined as HbA1c < 7.4% or less than the preoperative value plus 1.0% in patients with preoperative diabetes without severe hypoglycemic episodes during the 12 months after surgery. No severe hypoglycemic episode was observed. One of the 5 patients had diabetes (A1c 7.1%) before surgery without insulin treatment and maintained A1c less than 8.1% (preoperative value plus 1.0%) after surgery with insulin therapy ( Table 5 ). After surgery, the 3 of the other 4 patients needed exogenous insulin administration, but kept A1c 7.0% or less. One patient remained insulin-free with A1c less than 6.5% after surgery. Table 5. Transition of HbA1c(%) (Daily insulin dose (units)). Patient number Before surgery 3 months after surgery 6 months after surgery 9 months after surgery 12 months after surgery #1 7.1 (-) 7.4 (25) 7.8 (15) 7.9 (38) 6.7 (28) #2 5.7 (-) 7.0 (27) 6.6 (12) 6.5 (10) 7.0 (7) #3 5.8 (-) 6.2 (19) 5.8 (20) 6.3 (21) 6.8 (21) #4 5.4 (-) 6.2 (10) 6.3 (13) 6.8 (19) 6.3 (25) #5 5.6 (-) 6.2 (-) 6.4 (-) 6.3 (-) 6.0 (-) Median 5.7 6.2 6.4 6.5 6.7 Interquartile Range 5.6–5.8 6.2–7 6.3–6.6 6.3–6.8 6.3–6.8 p-value - 0.03125 0.0625 0.03125 0.0625 Note: Daily insulin doses are shown in parentheses (units/day) along with HbA1c. A p -value is based on Wilcoxon one-tailed signed rank test compared with before surgery. Transition of HbA1c(%) (Daily insulin dose (units)). Note: Daily insulin doses are shown in parentheses (units/day) along with HbA1c. A p -value is based on Wilcoxon one-tailed signed rank test compared with before surgery.

This study was conducted based on the clinical trial “Clinical Study of Pancreatectomy with Autologous Islet Transplantation for Treatment of Chronic Pancreatitis” (UMIN000014368). The protocol for this study was approved by our institutional ethical review board (NCGM-G-001325) and the Committee for Specific Designated Regenerative Medicine (PC3160124). Patients who underwent TPIAT between August 2016 and June 2019 and had 12 months of postoperative follow-up were enrolled in this study. Written informed consent was obtained from all participants. Patients with CP and pain refractory to medical and endoscopic treatment were eligible for TPIAT. Narcotic analgesics were used for pain relief in all cases. Patients were ineligible for the procedure if they were alcohol-dependent or did not have sufficient support to manage the complex regimen needed after TPIAT. The primary endpoint was the percentage of patients with improved pancreatic pain (a > 50% reduction in the Izbicki pain score 12 from the preoperative value) and good glycemic control without severe hypoglycemic episodes during the 12 months (365 ± 14 days) after surgery. Good glycemic control was defined as an HbA1c value<7.4% (National Glycohemoglobin Standardization Program) or less than the preoperative value plus 1.0% in patients with preoperative diabetes and no severe hypoglycemic events from 30 days to 12 months after surgery. Secondary endpoints included assessment of pain, glycemic control, and QOL. Pain scores were averaged over the previous 7 days. All patients underwent TP, which was performed using the standard technique. The pancreas was often atrophic, fibrotic, hard, and adherent to the surrounding tissue. The splenic artery and/or gastroduodenal artery were preserved until just before pancreatic resection to minimize the warm ischemia time. The spleen was resected in all cases. The pancreas was transported by the two-layer method 28 after intraductal organ preservation 29 and delivered to the cell processing facility for islet isolation. The gastrointestinal tract was reconstructed by simultaneous gastrojejunostomy and choledochojejunostomy. If necessary, a jejunal tube was placed for postoperative nutritional support. The patient then remained in the operating room with an open abdomen until islet transplantation. A digestive enzyme solution (Liberase MTF C/T GMP Grade, Roche Diagnostics) was injected into the main pancreatic duct through a previously placed angiocatheter. If necessary, additional enzyme solution was manually injected into the poorly distended portion of the pancreas using a needle and syringe. The pancreas was then cut into multiple pieces, transferred to a Ricordi ® chamber, and manually shaken to promote mechanical and enzymatic digestion. If the volume of tissue after digestion exceeded 15 mL, it was purified by density gradient centrifugation. The pancreatic tissue density was measured before purification. 30 The islets were purified using the bottle method and a mixture of OptiPrep ® solution (Axis Shield) and ETK solution (Otsuka Pharmaceutical Factory). 31 The purified islets were then washed and counted under a microscope. The islet yield was expressed as IEQ. The islets were also tested for viability (using a fluorescent dye inclusion/exclusion assay), endotoxin levels (limulus amebocyte lysate assay, <5 EU/kg), and sterility (by Gram staining). The final samples were subjected to bacterial and fungal culture. After isolation of the islets, a 5-Fr angiocatheter was inserted into the inferior mesenteric vein under direct vision and the tip was placed in the main portal vein. The isolated islets were suspended in 200 mL of a solution containing 10% human albumin and heparin (70 U per kg body weight) and infused into the portal vein. The portal vein pressure was monitored at the beginning, middle, and end of the islet infusion. Intravenous or subcutaneous heparin was continued for 1 week postoperatively to prevent thromboembolism. Doppler ultrasonography of the portal vein was performed within 24 h after transplantation to confirm portal blood flow. The blood glucose level was maintained at 80–110 mg/dL by an artificial pancreas intraoperatively and for several hours postoperatively to protect the transplanted islets. 32 The patients were observed in the hospital for 4–6 weeks postoperatively or longer if necessary. Thereafter, they attended outpatient clinics on a regular basis. All patients took a sufficient dose of digestive enzyme medication with each meal (pancrelipase 1800 mg/day). Insulin doses were set by a diabetologist according to glycemic control. Pain scores were evaluated using a VAS and the Izbicki pain score 33 preoperatively and 12 months postoperatively. QOL was quantified based on 3 questionnaires. The first was the SF-36v2 ® 34 (SF-36v2 ® Standard, Japanese 35 , 36 , which was created in conjunction with the Medical Outcome Study (MOS), a medical evaluation study conducted in the United States in the 1980s and used in previous reports on TPIAT for CP. 7 , 9 , 16 It comprises 36 questions that measure 8 domain scores (role-physical, role-emotional, physical functioning, social functioning, mental health, vitality, bodily pain, and general health perception) and 3 summary scores (physical, mental, and role/social component summaries) aggregated from the 8 scales above. The other questionnaires were the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) 20 , 37 and its pancreatic modification (QLQ-PAN28), modified from the QLQ-PAN26 pancreatic cancer module, 21 , 22 both of which have been used to determine QOL scores in CP. 22 , 38–40 REDCap ® electronic data capture tools were used to collect and manage study data. 41 Pain scores and QOL scores before and 12 months after surgery were compared using Wilcoxon signed rank test, one-tailed. All statistical analyses were performed using OriginPro2016 (OriginLab Corporation) and SAS version 9.4 (SAS Institute Inc.). A p -value<0.05 was considered statistically significant.

Our findings indicate that significant pain relief was achieved in all 5 patients ( Table 1 ) and that 4 patients no longer needed narcotic analgesics at 12 months after surgery (80%), which is equivalent to the results of previous reports. 9 , 13 The HbA1c value was<7.4% at 12 months postoperatively in all patients ( Table 5 ). Four of the 5 patients required insulin therapy but had C-peptide present in blood without severe hypoglycemia. 6 Since IAT prevented severe hypoglycemia and stabilized blood glucose levels, it is thought that a decline in QOL was not so prominent in spite of daily insulin treatment and that QOL increase associated with the pain relief exceeded the QOL decrease derived from insulin injection. If we compare a TPIAT group with an only TP group as a control group, the effect of IAT on QOL becomes clear, however, due to the established efficacy of IAT, the no-IAT group cannot be performed for ethical reasons. Although it is unable to evaluate the effect of IAT alone on QOL, we evaluated the impact of TPIAT as a whole in this study. Thus, our results revealed that TPIAT performed in our hospital is also effective in CP patients with severe pain that persists despite medical, endoscopic, and other surgical treatments. On the other hand, most QOL scores improved but not significantly so. One reason for this may be the small sample size. However, given the findings of significant improvements in role-physical, general health perception and vitality of the SF-36 ( Figure 1 and Table 2 ), even with the small sample size, TPIAT should be further evaluated for CP patients. In most of the items other than those mentioned above, the average value improved. In the SF-36, all of the averages of the 3 component summary scores and 8 multi-item scales increased or improved ( Figure 1 and Table 2 ). It is worth noting that improvements were obtained in not only physical QOL, but also psycho-emotional and social QOL. In the EORTC QLQ-C30, global health status and 4 of the 5 functional scales increased on average, and the other item, physical functioning, remained unchanged. Moreover, all symptom scales also improved or remained flat on average. Specifically, the averages of nausea and vomiting, as well as dyspnea, were unchanged, but the others were decreased (i.e., improved). In particular, financial difficulties was improved in all patients (the score was originally 0 in 1 patient, so it could not be improved further) ( p  = 0.0625), which suggests that TPIAT is helpful for employment and/or reducing medical expenses ( Figure 2 and Table 3 ). Most of the items in the QLQ-PAN28 had good results, but some items were a little complicated ( Figure 3 and Table 4 ). Items that remained unchanged or decreased, that is, improved, in all 5 patients were physical symptoms such as pancreatic pain, night pain, bloated abdomen, flatulence, decreased muscle strength, and treatment side effects, as well as mental symptoms such as fear for future health and ability to plan ahead. It should be noted that the outlook for the future improved. On average, in the 5 patients, 4 items improved other than the above 8 items. Specifically, scores for satisfaction with health care increased while those for jaundice, body image, and sexual functioning decreased (i.e., improved). The improvement in body image and sexual functioning is an interesting result because it seems, at first glance, to be unrelated to TPIAT. Some items appear to have worsened. For example, the decrease in digestive function may be due to TP. The increase in altered bowel functioning may reflect frequent stools, which may also be due to decreased digestive function, but may also be due to increased appetite after TP. Alcohol-related guilt increased, but this is not necessarily a bad result because it seems natural to feel guilty about drinking after TPIAT. Feeling guilty suggests that a patient may be drinking, but that the amount of drinking is suppressed. Because a change in taste refers to a change from the previous taste, the increased score of taste changes may indicate a change for the better. Weight loss reflects pessimistic feelings about low weight. It may have negative psychological consequences but may also lead to a positive willingness to gain weight. It is well-known that IAT plays a significant role in preventing brittle diabetes after TP. 14 Although postoperative insulin injection is often required after TPIAT, the result that QOL such as role-physical and vitality in patients even who required insulin after TPIAT improved significantly from before operation is interesting, but consistent with a previous report. 7 This suggests that the impact on QOL change after TPIAT exceeds that on compulsory insulin injections. From another points of view, because early surgery results in more effective pain relief and better preservation of pancreatic exocrine and endocrine function 15 and insulin independence occurs in one quarter of adults and half the children after TPIAT, 7 it is expected that transplantation at a younger age is more likely to lower pain and achieve insulin-free. We recommend that TPIAT be considered as soon as possible in patients with poor pain control after endoscopic drainage or partial resection of the pancreas and in those with diffuse CP and hereditary pancreatitis. In summary, although TPIAT often results in diminished digestive function and onset or worsening diabetes with daily insulin treatment, it improves physical and mental QOLs, and even social behaviors. The present work has some limitations. First, our study has a small sample size. Second, there could be multiple interpretations of the questionnaire, depending on the patient. Finally, questionnaires are a subjective evaluation. In spite of the limited number of sample size, the result of SF-36 questionnaire which was consistent with the previous paper 7 confirmed the validity of this questionnaire. Compared to SF-36 which is usually used for QOL after TPIAT, 7 , 9 , 16–19 EORTC QLQ-C30, which is a cancer-specific questionnaire developed in 1980 by European Organization for Research and Treatment of Cancer (EORTC), 20 and QLQ-PAN28, a validated version for CP, 21 , 22 have not been used to assess QOL after TPIAT. Another noteworthy questionnaire is a pancreatitis quality of life instrument (PANQOLI) which is the first CP-specific instrument with a unique subscale for “self-worth” 23 , 24 and has been used for evaluating QOL of CP patients including Japan. 25–27 The validity of these questionnaires for QOL evaluation in TPIAT needs further verification. This is a rare report on QOL after TPIAT performed in Asia, and its findings suggest that this procedure would be useful in East Asia, as in the United States. We plan to conduct a multicenter clinical trial that includes a larger number of patients to evaluate the efficacy of this treatment in more detail.