Pancreas Graftectomy Predicting Markers

Abstract

Background Over last 13 years 193 pancreas transplantations were performed at the Department of Gastroenterological Surgery and Transplantation at the Central Clinical Hospital of the Ministry of the Interior in Warsaw. The 5-years survival of the patients in our cohort is 84%, for the pancreas grafts 65% and for kidney graft 83%. The aim of the study was to verify which laboratory parameters and imaging markers in ultrasound and computed tomography of pancreas graft may have most specific role for transplantectomy prediction. A retrospective analysis of clinical data was performed. The parameters included in the study were: sex, type of transplantation (simultaneous pancreas and kidney- SPK, pancreas alone PTA, pancreas after kidney PAK), glycemic profile and selected laboratory results (amylase, lipase, CRP, PCT, C-peptide, INR). US and CT images were revived for following findings: abnormal graft echogenicity or density, venous thrombosis (full and partial), arterial thrombosis, anastomotic stenosis and aneurysms, peripancreatic fluid collections, abdominal fluid collection, increased echogenicity of peripancreatic fat or peripancreatic fat infiltration, lymph nodes, increased RI, low arterial flow. Data from two end-points were taken into statistical analysis. For patients without complications the data was collected 3 months after transplantation, while for others on the day preceding transplantectomy. 68% of the analyzed group underwent SPK, 29%-PTA and 2%- PAK. Total mortality for the cohort was 7%. In 67% of the cohort the function of the pancreas graft during observation was satisfying, without demand for insulin and positive C-peptide. 22% patients had severe complications that lead to transplantectomy, while 10% were diagnosed with temporary graft pancreatitis without need for surgical intervention. The most specific markers that predicted transplantectomy or pancreatitis were full or partial portal vein thrombosis and abnormal echogenicity in US (p<0.005), as well as abnormal density and venous thrombosis in CT (p<0.005). Results of laboratory test occurred statistically insignificant for risk of complications in our cohort. Abnormal graft density or echogenicity and presence of venous thrombosis are the best imaging markers for graft dysfunction. Even partial thrombosis is a risk factor for pancreas graft transplantectomy. Although CT is superior to US for detection of graft dysfunction, the use of ionized radiation and administration of intravenous contrast agent hampers its application, especially in patients with simultaneous pancreas-kidney transplantation. Imaging examination occurred prior to laboratory tests in our study and marked an important role of radiologists in a treatment process. Further studies should be done to establish imaging guidelines in ultrasound and computed tomography of pancreas graft malfunction, rejection and failure.