Functional and morphological evolution of remnant pancreas after resection for pancreatic adenocarcinoma

Abstract

Functional and morphological evolution of remnant pancreas after resection for pancreatic adenocarcinoma is investigated.The medical records of 45 patients who had undergone radical resection for pancreatic adenocarcinoma from March 2010 to September 2013 were reviewed retrospectively. There were 34 patients in the pancreaticoduodenectomy (PD) group and 10 patients in the distal pancreatectomy (DP) group. One patient received total pancreatectomy. The endocrine function was measured using the glucose tolerance index (GTI), which was derived by dividing daily maximum serum glucose fluctuation by daily minimum glucose. Remnant pancreas volume (RPV) was estimated by considering pancreas body and tail as a column, and head as an ellipsoid, respectively. The pancreatic atrophic index (PAI) was defined as the ratio of pancreatic duct width to total pancreas width. Representative indices of each patient were compared before and after resection up to 2 years postoperatively.The area under receiver operating characteristic curve of GTI for diagnosing DM was 0.823 (95% confidence interval, 0.699-0.948, P < .001). Overall, GTI increased on postoperative day 1 (POD#1, mean ± standard deviation, 1.79 ± 1.40 vs preoperative, 1.02 ± 1.41; P = .001), and then decreased by day 7 (0.89 ± 1.16 vs POD#1, P < .001). In the PD group, the GTI on POD#14 became lower than preoperative (0.51 ± 0.38 vs 0.96 ± 1.37; P = .03). PAI in the PD group was significantly lower at 1 month postoperatively (0.22 ± 0.12 vs preoperative, 0.38 ± 0.18; P < .001). In the PD group, RPV was significantly lower at 1 month postoperatively (25.3 ± 18.3 cm vs preoperative, 32.4 ± 20.1 cm; P = .02), due to the resolution of pancreatic duct dilatation. RPV of the DP group showed no significant change. GTI was negatively related to RPV preoperatively (r = -0.317, P = .04), but this correlation disappeared postoperatively (r = -0.044, P = .62).Pancreatic endocrine functional deterioration in pancreatic adenocarcinoma patients may in part be due to pancreatic duct obstruction and dilatation caused by the tumor. After resection, this proportion of endocrine insufficiency is corrected.

Figure 1

Measurements of total pancreatic width (17.06 mm), pancreatic duct width (5.69 mm), antero-posterior diameter (15.64 mm), and latero-lateral length (82.47 mm) of the pancreas body and tail in a patient with cancer of the pancreatic head (A). The cranio-caudal diameter (21.66 mm) was measured in the coronal view (B).

Figure 2

Simulation of pancreatic head, body, and tail using solid figures. APD = antero-posterior diameter, CCD = cranio-caudal diameter, LLL = latero-lateral length.

Figure 3

Receiver operating characteristic curve of glucose tolerance index for the diagnosis of diabetes mellitus. AUC = area under ROC curve, ROC = receiver operating characteristic.

Figure 4

Evolution of glucose tolerance index (GTI) in diatal pancreatectomy (A) and pancreaticoduodenectomy (B) groups. Case numbers are not shown (refer to Fig. 6). DP = distal pancreatectomy, GTI = glucose tolerance index, PD = pancreaticoduodenectomy, SD = standard deviation.

Figure 5

Evolution of pancreas atrophic index (PAI) in patients after pancreaticoduodenectomy. Case numbers are not shown (refer to Fig. 6). PAI =  pancreas atrophic index, SD = standard deviation.

Figure 6

Evolution of remnant pancreatic volume (RPV) in patients after distal pancreatectomy (A) and pancreaticoduodenectomy (B). Case numbers are presented in each column. DP = distal pancreatectomy, NA = not available, PD = pancreaticoduodenectomy, RPV = remnant pancreatic volume.

Figure 7

Correlations between the preoperative glucose tolerance index (GTI) and future remnant pancreats volume (A), and between postoperative GTI and remnant pancreatic volume (B). GTI = glucose tolerance index.