Background/objectives: Diabetes mellitus (DM) is recognized as a risk factor for pancreatic cancer (PaC) and expected to be a clue for early diagnosis. However, it is still obscure whether a diagnostic strategy using DM as a clue can improve the prognosis or not.
Methods: We retrospectively reviewed 540 patients with PaC, and investigated the prognosis with regard to the reasons for diagnosis. We compared patients diagnosed by imaging studies performed when DM was newly diagnosed or deteriorated, and patients diagnosed by symptoms.
Results: The prevalence of DM in PaC patients was 45% (256/540) and did not differ between disease stages. More than half of DM in PaC patients (150/256) were new-onset (<2 years in duration). One hundred sixty one patients (30%) were asymptomatic at PaC diagnosis. There were 38 patients diagnosed in association with DM (by new-onset DM, 16; by aggravation of long-standing DM, 22). Asymptomatic patients had smaller primary tumor and were diagnosed at an earlier stage. The prognosis of PaC patients complicated with DM did not differ from that of patients without DM; however, patients had better prognosis if they were diagnosed in association with DM alone (median survival time, 20.2 months), compared with patients diagnosed by symptoms (10.2 months, P < 0.01).
Conclusions: Our analysis revealed that patients diagnosed in association with DM had better survival than symptomatic patients. Given the high prevalence of DM in PaC patients, DM can be a useful diagnostic clue for screening and lead to improvement of prognosis in PaC patients.
Copyright © 2013 IAP and EPC. Published by Elsevier B.V. All rights reserved.