Association of Diabetes Mellitus with the Occurrence and Prognosis of non-Hodgkin's Lymphoma
Date Issued
2004
Date
2004
Author(s)
Lin, Hsin-Yin
DOI
en-US
Abstract
Purpose: The aim of this study was to use a retrospective cohort study in conjunction with a case-control study to elucidate the effect of pre-existing DM history on the risk of developing NHL; and to assess whether pre-existing DM history can further affect cumulative survival after NHL diagnosis.
Methods: We used a retrospective cohort to set up a case-control study. The cases consisted of 242 subjects with pathologically confirmed NHL who were referred to the Department of Radiation Oncology in Chang-Gung Memorial Hospital (CGMH) for radiotherapy between January 2000 and March 2004. The controls were derived from a population-based multiple screening program, called Keelung Community based Integrated Screening (KCIS). The subjects of both cases and controls had ever filled out a questionnaire about the past history including DM, hypertension, and some life-style factors such as alcohol drinking and smoking. Logistic regression model was employed to calculate the odds ratios of the risk factors investigated to evaluate the association between pre-existing DM and the occurrence of NHL. Basic and advanced survival analyses were used to assess whether pre-existing DM history can further affect cumulative survival after NHL diagnosis.
Results: The presence of prior diagnosis of DM was associated with an increased risk of NHL development (odds ratio, 2.144; 95% confidence interval, 1.405 ~ 3.180). After controlling for other relevant factors, preexisting DM history was found to be an independent risk factor for the occurrence of NHL (OR, 1.875; 95% CI, 1.215 ~ 2.892 [p= 0.0045]). When subgroup analyses regarding certain tumor or disease characteristics were performed, the impact of preexisting DM history was particularly manifested in some subgroups such as the tumors of T-cell origin. In univariate analysis for survival, patients with DM conferred an extra 33% of death from causes related to NHL. However, the association was not statistically significant (P= 0.479). Multi-variable analysis with the incorporation of the first interaction term between survival time and presence of DM after controlling for other available factors revealed that the effect of DM on risk for death from NHL varied with follow-up time. Such an effect modification was marginally statistically significant (P=0.054).
Conclusion: It was found that preexisting DM history was an independent risk factor for the occurrence of NHL, particularly for those with extranodal involvement, of T-cell origin, and which were not localized or low grade. Preexisting DM history was also an accelerated factor for the risk of death from causes related to NHL.
Subjects
非何杰金氏惡性淋巴瘤
糖尿病
存活率
diabetes mellitus
survival
SDGs
Type
thesis