Betul Hatipoglu, MD, And R. Matthew Walsh, MD
Cleveland Clinic is a non-profit academic medical center. Advertising on our site helps support our mission. We do not endorse non-Cleveland Clinic products or services Policy
Chronic pancreatitis (CP) is a disorder characterized by progressive damage to the pancreas, with inflammation and fibrosis, resulting in exocrine and endocrine insufficiency, affecting around 50 to 70 cases per 100,000 patients per year. As part of the natural progression of endocrine insufficiency in CP, 50 to 60 percent of patients will develop type 3c, or pancreatogenic, diabetes, eventually requiring insulin. Pancreatic surgery is often used for these patients when medical therapies fail to relieve pain and improve quality of life (QOL). The process of potentially preserving β-cell mass through islet cell autologous transplantation during pancreatectomy can be beneficial in maximizing endogenous insulin secretion and C-peptide positivity. Insulin independence rates after total pancreatectomy with auto islet transplantation (TP/AIT) typically range between 46 and 64 percent at five-year follow-up.
Can Successful Metabolic Outcomes Be Predicted?
We decided to evaluate factors that could predict insulin independence in these patients. For our study, we used a prospective IRB-approved database. Data were available for 36 patients from August 2008 to February 2014.
Patients underwent extensive preoperative evaluation by our pain management team. Preoperative mood was measured using the 20-point Depression Anxiety Stress Scale and the Pain Disability Index, and pain severity was analyzed using a visual analogue scale. Our hepatobiliary surgery team reviewed all test results, including genetic profile for pancreatitis, pancreatic CT, magnetic resonance cholangiopancreatography, endoscopic ultrasound (EUS), differential pain block and endoscopic pancreatic function testing. Patients’ preoperative pancreatic disease extent was classified as normal, “minimal” change in CP or “advanced” changes in CP based on the Cambridge classificationscore on preoperative CT and/or Rosemont criteria by EUS. A diabetes nurse specialist provided education, and a preoperative metabolic assessment was completed, including HbA1c, C-peptide, antibodies, and five-hour mixed-meal measures of glucose, insulin, proinsulin and C-peptide (after administration of Ensure Plus® 6 mL/kg, max 360 mL).
Total pancreatectomy was performed in all cases, usually with splenectomy. Following resection, pancreata were flushed with UW solution and transferred in cold preservation fluid to the processing laboratory. After isolation, islets were resuspended with human serum albumin, heparin and ciprofloxacin. Islet yield was determined and then expressed as islet equivalents (IEQ).
Once the solution was returned to Cleveland Clinic, infusion was done through the splenic vein stump after heparinization to prevent portal vein thrombosis.
Postoperative pain was managed using patient-controlled analgesia. A glycemic target of 60 to 126 mg/dL was maintained after transplantation by an insulin infusion protocol, followed by a standardized sliding-scale algorithm for weaning of insulin. A follow-up endocrine evaluation was performed within three to six months, repeating metabolic tests done prior to surgery. Insulin independence was defined as no daily requirements of any form of insulin regardless of C-peptide level. Hypoglycemia was defined by a measured plasma glucose concentration of ≤ 70 mg/dL, regardless of the presence or absence of typical symptoms. Severe hypoglycemia was defined by an event requiring the assistance of a third party due to symptoms of neuroglycopenia. Patient follow-up was obtained through telephone calls, questionnaires, and clinic visits at three, six and 12 months, then yearly thereafter.
A Snapshot of our Findings
For 36 patients (18 female, 18 male) with chronic pancreatitis, the mean age was 38 years with a mean followup time of 29 months. The etiology of CP was idiopathic (N = 16), alcoholic (N = 7), biliary (N = 8) and genetic/hereditary (N = 7). Preoperative narcotic use was present in 33 of patients (92 percent). All patients were insulin independent prior to surgery, and median preoperative HbA1c was 5.7 percent. Advanced changes in CP were present in 14 patients (39 percent).
The mean IEQ/kg was 4,784 ± 2,419 (range 769-9,942). From the cohort, one-third (12/36) were insulin-independent at a mean follow-up of 29 months (range 3-66 months). Among insulin-independent patients (N = 12), three had HbA1c < 5.8 percent, seven had HbA1c 5.8 to 6.4 percent and one had HbA1c ≥ 6.5 percent (Figure). In patients who were insulin-independent at most recent follow-up, there was a statistically significant difference in IEQ/kg: 6,705 ± 2,150 (range 2,980-9,942, N = 11) vs. 3,824 ± 1,952 (range 769-9,876, N = 24, P < 0.001). Those who were insulin independent were more likely to be female (P = 0.012) and have normal disease extent on preoperative pancreatic CT (P < 0.011). Females were more likely than males to have normal/minimal disease extent (P = 0.046), so this may be responsible for the association seen between gender and insulin independence. The lowest IEQ transfused in the insulin-independent group was 354,725 (2,980 IEQ/kg), and in the insulin-dependent group, the highest IEQ transfused was 672,000 (9,876 IEQ/kg).
Patients with advanced changes on preoperative CT had lower islet yield in comparison with those with normal or minimal changes. In addition, age (P = 0.049) and preoperative HbA1c (P = 0.020) were negatively associated with islet yield. A positive association was observed between islet yield and preoperative weight (P = 0.021) and also with BMI (P = 0.055). Other preoperative predictors of islet yield included glucose peak during the mixed-meal tolerance test (MMTT) (P = 0.019) and HbA1c (P = 0.010).
Conclusions and Next Steps
We have previously demonstrated improved QOL outcomes in 20 patients not linked to insulin-dependence status, and these data extend our experience with this procedure. In this series, one-third of the patients remained insulin independent at the most recent follow-up, averaging 28 months (range 3-54 months), which is similar to other data published around the world. Those who were insulin independent were more likely to be female. Normal/minimal disease extent on preoperative imaging correlated with better islet yield and higher rates of insulin-independence. Patients receiving at least 3,000 IEQ/kg were more likely to be insulin-independent at follow-up. Preoperative HbA1c and glucose peak at MMTT correlated inversely with islet yield.
More studies are needed to better assess selection criteria and timing for TP and AIT during the CP disease process to maximize β-cell function preservation and improve outcomes.
ABOUT THE AUTHORS
Dr. Hatipoglu is an endocrinologist in the Department of Endocrinology, Diabetes and Metabolism at Cleveland Clinic and a Clinical Associate Professor of Medicine at Cleveland Clinic Lerner College of Medicine of Case Western Reserve University. She can be contacted at firstname.lastname@example.org. Dr. Walsh is Chair of the Department of General Surgery and Professor of Surgery at Lerner College of Medicine, and can be contacted at 216.445.7576 or email@example.com.