Objective:To dynamic monitor the change of serum CA19-9levels before and after surgery in pancreatic cancer patients, discuss the clinical value of CA19-9in preoperative diagnosis, determination of resectability and early detection of postoperative recurrence and (or) metastasis.Methods:There was a retrospective analysis for CA19-9expression levels in141patients with pancreatic cancer, before and after surgery (1week,1month,3months,6months,12months). Meanwhile we used CT and MRI (1month,3months,6months,12months) to monitor its postoperative recurrence and (or) metastasis. By receiver operating characteristic curve (ROC), we evaluated the value of CA19-9monitoring in postoperative recurrence and metastasis, with Youden index (sensitivity+specificity-1, Youden) maximum to set the (cutoff) threshold. Furthermore, we analyzed the correlation between CA19-9expression level and postoperative recurrence and (or) metastasis for pancreatic cancer.Results:In141cases of preoperative pancreatic cancer patients, the CA19-9’s staging in TNM (P=0.133), tumor site (P=0.909) and resectability (P=0.104) had no statistically significant differences, but there was obvious differences (P=0.023) between the levels of tumor differentiation. When the preoperative CA19-9expression level was more thanã€equal or less than1000u/ml, the resectability rate had statistically differences (P=0.036). The preoperative CA19-9expression of Pancreatic cancer patients in each group showed no significant differences (P>0.05). The CA19-9expression level in palliative group had been an upward trend, while in stable and recurrent (postoperative) groups were significantly decreased after one week, and after three months it began to rise and higher than normal, but expression levels in recurrence group at each time point (after one week,1month,3months,6months,12months) were significantly higher than stable group. The CA19-9expression level in recurrence group was909.50(414.55-1739.55) U/ml, and stable group was19.9(8.7-156.33) U/ml, there was significant statistical differences (P<0.001). Using ROC curve analysis, when setting threshold value to211.35U/ml, for CA19-9monitoring to recurrence and (or) metastasis, its sensitivity was91.10%and specificity was85.50%, with better monitoring value. When setting threshold to take221.35U/ml, for CA19-9monitoring to recurrence and (or) metastasis, the average time was4.05±3.18months, but the imaging (CT or MRI) was7.32±5.82months, the differences were more significant (P<0.05). Conclusions:The tumor marker CA19-9played an irreplaceable role and value in the diagnosis and treatment of pancreatic cancer. Especially in the early monitoring of postoperative recurrence and (or) metastasis, its clinical value was more obvious. Elevated expression levels of CA19-9prompted more early than imaging for postoperative recurrence and (or) metastasis in pancreatic cancer. |