Background A fecal occult blood test (FOBT) for screening for colonic cancer is accessible to all of health plan participants but adherence of FOBT-positive participants with recommended BE or CF is poor, which lowers the final CRC detection rate. The leading cause of poor adherence seems to be low specificity of the FOBT. Method With a health plan performing 15, 000+multiphasic tests per year that include a 3-day method FOBT and sigmoidofiberscopy, a retrospective chart review was conducted and linked to switching of interpretation of FOBT data that markedly improved Colonic cancer detection rate more than 3 times. Blood density of stool specimens were measured with immunoassay and reported in terms of ng/ml. The patterns of three figures of blood density for each participant's stool specimens were examined in relation to the presence of cancers or adenomas as well as the timing of removals of these neoplasia. Results Quantitative test results represented in ng/ml tell more about suspected colonic bleeders than conventional qualitative test results. Patterns of FOBT data characteristic of colonic neoplasia are as follows; 1. Continuity of positivity; In 74% and 61% of sets of 3 stool specimens of colonic cancers and adenomas respectivly, all three were positive (figures over thresholds) . Threshold values were 100 ng/ml for cancer and 30 ng/ml for adenoma. 2. Resemblance of 3 figures for blood density; In 63% of sets of 3 stool specimens of neoplasia of the colon, figures for blood density of each set closely resembled each other. Benign bleeders such as hemorrhoids and diverticuli, however, rarely showed continuity of positivity or resemblance as was noted in neoplasia. In 94% of adenoma cases, after polypectomy, all of the three figures of blood density went down to near zero, suggesting that the previous bleedings were caused by minute collision between stools and protrusions of the colonic inner surface. Conclusion When blood densities of stool specimens are measured quantitatively in terms of ng/ml and the data thus obtained are interpreted in terms of continuity, the FOBT helps physicians distinguish colonic neoplasia from other nonprotruding colonic bleeders and raises cost-efficiency of the BEs and/or CFs they perform.
A total of 348 male workers, attending a health check-up program, were studied in order to assess the relation between hyperuricemia and both subjective symptoms and daily life styles. The subjects consisted of 207 technical workers and 141 office workers, aged from 39 to 60 years, employed in a steel factory. Examination of subjective symptoms showed that odds ratios of hyperuricemia were high for subjects with irritation (2.23; 95% C. I., 1.32-3.77) and subjects with fatigue (1.66; 95% C. I., 1.02-2.73) . Examination of lifestyles showed high odds ratios for high frequency of eating out (2.58; 95% C. I., 1.32-5.03) and high frequency of alcohol consumption (2.50; 95% C. I., 1.26-4.96) . High odds ratios of hyperuricemia were found for subjects with high serum levels of retinol (4.45; 95% C. I., 2.24-8.86) and α-tocopherol (2.22; 95% C. I., 1.09-4.63), after adjusting for age, work type, smoking status, alcohol consumption, BMI, and serum levels of total cholesterol, creatinine and GPT activity. In contrast, the odds ratio was low for subjects with high serum levels of β-carotene (0.49; 95% CI., 0.25-0.96) . These results indicate that certain subjective symptoms (fatigue and irritation), high frequencies of eating out, and alcohol consumption can be risk factors for hyperuricemia, and that high intake of foods rich in β-carotene (e. g., colored vegetables and fruits) may reduce the risk of hyperuricemia.
To estimate the progression rate to diabetes mellitus (a diabetic pattern diagnosed by fasting plasma glucose (FPG) ≥126 mg/dl and diabetes mellitus diagnosed by a diabetic pattern or/and treatment for diabetes mellitus) among non diabetics (FPG≤125 mg/dl and not treated for diabetes mellitus), we studied 3, 604 subjects (mean age; 50.9) who consulted our institute for physical checkup during the period from April 1992 to March 1995 and reconsulted our institute five years later. According to the body mass index (BMI) and FPG level at starting point the subjects were subclassified. Mean differences in BMI and FPG after five years and the incidence of diabetes mellitus five years later were calculated. The results were as follows. (1) Overall, 4.1% (147/3, 604) of subjects showed a diabetic pattern at the starting point, while 7.1% (257/3, 604) showed a diabetic pattern five years later. (2) Two percent of the subjects whose FPG level was 100-104 mg/dl at the starting point, 7.7% of those whose FPG was 105-109 mg/dl, 18.3% of those whose FPG was 110-114 mg/dl, 37.7% of those whose FPG was 115-119 mg/dl and 68.4% of whose FPG was 120-125 mg/dl showed a diabetic pattern five years later. (3) There were no significant differences in the rate of progression to diabetes mellitus among BMI levels when FPG was 100-119 mg/dl at the starting point, but when FPG was 120-125 mg/dl at the starting point, 46.2% of subjects whose BMI was less 24 showed diabetes mellitus five years later and 87.2% of whose BMI was 24 or more showed diabetes mellitus five years later. This difference was significant by Mann Whitney's U test (p≤0.001) . Therefore, it is necessary for a person whose FPG is 120-125 mg/dl with a BMI over 24 to be managed strictly to prevent progression to diabetes mellitus.
Type 2 Diabetes has enormously increased for the last few decades in Japan. It is one of the life-style related diseases and prevention is thought to be very important to reduce it. We examined health check-up data longitudinally and tried to reveal the risk factors and find the natural course of the onset of the disease. Normal fasting plasma glucose (FPG) group (n=11, 123, age: 41.9±7.1) have been followed up for more than 10 years. Among them, 8.6% became diabetes (DM group, FPG≥126 mg/dl) and 24.9% became Impaired Glucose Tolerance (IGT group, 110≤FPG<126) in 14 years. DM group and IGT group had significantly high FPG, BMI, ALT and triglyceride level during the observation period. Multiple regression tests (Stepwise method) revealed the significant relation between the last FPG value and FPG initial, BMI, ALT and triglyceride. According to Cox model, FPG level initial might be crucial, even if it was within normal range. If FPG were from 90 to 94 mg/dl, 10% would become diabetes in next two decades, but if FPG were from 105 to 109 mg/dl, 46% would become diabetes in the same period.