Abstract
OBJECTIVE To investigate the relationship between BMI, WHR and biliary tract cancers (CBT).
METHODS A population-based case-control study was conducted in urban Shanghai from June 1, 1997 to May 31, 2001 involving interviews with 627 new cases of biliary tract cancers aged 35 to 74 years and 959 frequency-matched population controls by gender and age in five-year groups. All subjects were interviewed in person by trained interviewers using a structured questionnaire. An unconditional logistic regression was performed to calculate adjusted odds ratios (ORs) and 95% confidence intervals (CIs).
RESULTS Obesity was associated with an increased risk of gallbladder cancer across adulthood at ages 20-29 and 30-39 in females. Compared with subjects in the lowest quartile of WHR, ORs for the highest quartile and P for trend for cancers of gallbladder and extrahepatic bile duct both reached significant levels among males and females.
CONCLUSION Our observations in urban Shanghai suggested that obesity in early adult life may contribute to the risk of gallbladder cancer, and increased WHR may substantially elevated risk of cancers of the gallbladder and extrahepatic bile duct.
keywords
Cancers of the biliary tract (CBT), including tumors arising from the gallbladder (GB), extrahepatic bile duct (EBD) and ampulla of Vater (AV), are uncommon malignancies in most parts of the world. However, CBT was one of the most rapidly increasing malignancies between 1972~1974 and 1996~1999 in urban Shanghai with the age-adjusted incidence rates rising from 1.2 and 1.7 per 100,000 to 3.1 and 4.6 per 100,000 for males and females, respectively. CBT are gastrointestinal tumors with a poor prognosis and little is known about the etiology, apart from a close association with gallstone disease and a female predominance among gallbladder cancer. [1] Few studies have examined the association between BMI, WHR and CBT with inconsistent results.[4-10] Using a large, population-based case-control study conducted in Shanghai, China, we attempted to explore this issue.
Materials and Methods
Clinical data
The case subjects were defined as all patients (aged 33-74 years) who were diagnosed with biliary tract cancers (ICD-9, code 156) from June 1997 to May 2001. and were permanent residents of the urban area of Shanghai. During the above-mentioned period, 627 cases of biliary tract cancers were identified. Control subjects were selected by frequency matching in accordance with age-sex distribution of the cases of biliary tract cancers ascertained by the cancer registry in 1995. Personal identification cards from the Shanghai Resident Registry were used to select control subjects. For each resident in urban Shanghai, the cards contained information with regard to name, sex, address, date of birth and other demographic variables. Cards were selected in a random manner until persons of a particular sex and age (within 5-year category) were identified. For each control subject chosen in this manner, two alternative control subjects were also selected. When the first control subject could not be interviewed, the second or the third was enrolled in the study. Among the randomly selected 1205 controls, 1000 participated in the study. Among them 41 subjects with malignancies were excluded with the response rate of 82.4%. Subjects were interviewed by trained interviewers using a structured questionnaire to obtain information on demographic characteristics, cigarette, alcohol and tea consumption, frequency of food intake, medical and occupation history, body mass index, waist to hip ratio, physical activities, and among women information on menstrual and reproductive history, oral contraceptives (OC) and hormone use as well. Biliary tract cancer cases were confirmed by clinical, pathological as well as radiographic experts using medical records, pathological slides and radiographic data. In our study, BMI [weight (kg)/height (m2)] (23-24.9 kg/m2) was classified into the group of overweight, and BMI (≥ 25 kg/m2, established cutpoint for Asians) were considered as obese.[11] Information on WHR [waist circumference (cm)/hip circumference (cm)] was measured at the interview.
Statistical analysis
The association between the risk of biliary tract cancers and BMI, WHR was measured by the odds ratios (ORs) and 95% confidence intervals (CIs) using unconditional logistic regression model. [12] Risk factors were evaluated by subsite of tumor origin, i.e., gallbladder, extrahepatic bile duct and ampullar of Vater, separately for men and women. All P-values are two-sided.
Results
During the period in question, 627 incident cases of CBT including 368 cases of gallbladder cancer(GBC), 191 cases of EBDC and 68 cases of AVC were identified in urban Shanghai. As shown in Table 1, no statistical difference was observed in age at interview and marital status. Education level was lower only among females with GBC. Patients with diabetes were more frequent among females with GBC, patients with hypertension were less among males with EBDC and among females with AVC. Less consumption of allium vegetables among females with GBC and EBDC and more consumption of preserved foods among AVC and females with EBDC were observed (detailed comparisons between cases and controls by gender were not presented in Table 1). Notably, gallstone disease substantially increased the risk of GBC, EBDC as well as AVC with crude odds ratios (ORs) of 19.91 (95%CI: 14. 56-27.21), 6.51 (95%CI: 4.65-9.11) and 3.69 (95%CI: 2.24-6.08), respectively. All the confounding factors and total energy intake were adjusted in multivariate analyses.
Distribution of selected characteristics among biliary tract cancer cases, by anatomic subsites, and controls
After adjustment for corresponding confounding factors, an increased risk was observed for being obese (BMI ≥ 25kg/m2) at ages of 20-29 years with OR of 2.92 (95%CI: 0.95-9.05) among females in GBC. The corresponding OR for being obese at ages of 30-39 years was 1.69 (95% CI: 0.89-3.21), P for trend = 0.048 (Table 2). Risks of GBC and EBDC were substantially elevated with increasing WHR among both males and females, and all the P for trend for these two cancer sites reached a statistically significant level (Table 3).
Odds ratios(ORs) for biliary tract cancers (CBT) in relation to body mass index (BMI)*
Discussion
Our findings indicate that obesity, especially in early adult life may increase the risk of GBC. Some previous studies have shown a link between obesity and gallbladder cancer.[4,5] Our findings of no statistical association with risk of EBDC are contrary to an earlier report of increased risk with obesity. [6] In addition, reduced risk of EBDC with BMI (5 years ago) was observed, which may be due to the effect of sub-clinical cancer on subjects’ weight. Knowledge of the mechanism between obesity and CBT is limited; however, the risks associated with obesity may reflect the tendency for obese subjects to have high levels of endogenous estrogens as well as bile that is supersaturated with cholesterol. [1]
In our study, a close association with abdominal adiposity was found. Central adiposity, assessed by waist to hip ratio (WHR), has been positively associated with GBC as well as EBDC risks. So far, few studies have examined the association of WHR with CBT and the etiology mechanism is unclear. [1-10] It is postulated that endogenous estrogens may play a role in the pathway between WHR and CBT risk. [1] Waist to hip ratio (WHR) was obtained at the interview. Information on subjects’ waist or hip could not be obtained by measurement due to the bad medical condition of cancer patients and most of the patients had lost their usual weight. As a result, waist and hip circumference differed from what they used to be and the results may have some bias towards underestimation.
Odds ratios(ORs) for biliary tract cancers (CBT) in relation to waist to hip ratio (WHR)*
Among the limitations of this study, there are problems of reliability of information concerning the distant past, which may be subject to recall bias. The strength of our study is also worth mentioning. We included a large population-based sample, used standardized objective rather than solely self-reported measures and undertook a complete assessment of confounding factors. We had more data available on weight changes across lifetime, therefore permitting analysis of sequential weight changes beyond what has been previously reported.
In conclusion, this study indicates that obesity and higher WHR may increase the risk of CBT. However, few studies have examined the association between BMI, WHR and CBT risk, so further clarification is needed in regard to the association and underlying biologic mechanisms that might be involved in the chain between obesity and CBT risk. Our results corroborate and strengthen the evidence from previous research that avoiding obesity through life, in particular, in younger adult life is one of the means of reducing CBT risk.
- Received December 21, 2004.
- Accepted February 18, 2005.
- Copyright © 2005 by Tianjin Medical University Cancer Institute & Hospital and Springer
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.