Thyroid autoimmunity is associated with higher urinary iodine concentrations in an iodine-deficient area of Northwestern Greece.

Northwestern Greece was identified in the 1960s for its high prevalence of endemic goiter and iodine deficiency. Although iodized salt has been commercially available since then, a recent epidemiological survey of 3916 schoolchildren found that low-grade goiter is still prevalent in endemic proportions (21%). The aim of this study was to further assess the cause of goiter and the severity of iodine deficiency in children from this endemic area of Greece. Of the 800 children with clinically detectable goiter, 97 children (60 girls and 37 boys, 8-15 years) were recruited for determination of urinary iodine excretion, as well as assessment of thyroid volume and function and detection of antithyroid antibodies. The median urinary iodine concentration was 8.4 microg/dL, indicative of a mild iodine deficiency. Thyroid function was normal in all but 11 children who had subclinical hypothyroidism. Sixteen children (16.5%), including all those with subclinical hypothyroidism, were positive for antithyroid antibodies. Their median urinary iodine concentration (20.6 microg/dL) was higher compared to children who were negative for antibodies (7.4 microg/dL; p<0.001). The mean thyroid volume by ultrasonography (12.2+/-4.1 mL) was above the upper limit of normal for this age group. Thyroid volume was inversely related to the urinary iodine content in the children with negative antithyroid antibodies. Iodine deficiency is still prevalent in northwestern Greece although of mild severity and constitutes the primary cause of goiter among schoolchildren. However, it appears that autoimmune thyroiditis is emerging as a frequent cause of goiter in those children with sufficient iodine intake.

Prognostic variable in patients with advanced colorectal cancer treated with fluorouracil and leucovorin-based chemotherapy.

Possible prognostic variables for tumor response, time to progression (TTP), and survival in 141 patients with advanced colorectal cancer treated with fluorouracil and leucovorin-based chemotherapy were analyzed. None of the variables examined for their possible influence on tumor response attained significance in the stepwise logistic regression. In the univariate analysis, variables found to be strongly associated with TTP were performance status (PS) (P = 0.0301), liver involvement (P = 0.030), and the initial values of WBC (P = 0.0319), lactic dehydrogenase (LDH; P = 0.0053), gamma-glutamyl-transpeptidase (gamma-GT; P = 0.0013), alkaline phosphatase (ALP; P = 0.0186), albumin (P = 0.0004), and carcinoembryonic antigen (CEA; P = 0.0014). In the Cox analysis, liver involvement (P = 0.0553), albumin (P = 0.0181), PS (P = 0.484), and ALP (P = 0.0553) were retained as independently significant variables. When only patients with liver metastases were included in the analysis, then only albumin (P < 0.001) demonstrated a prognostic significance. Also, in the univariate analysis, variables predicting survival were PS (P = 0.0230), grade (P = 0.00600), liver involvement (P = 0.0002), LDH (P = 0.0001), gamma-GT (P < 0.001), ALP (P = 0.0006), albumin (P = 0.0309), and CEA (P = 0.005). With the multivariate analysis, gamma-GT (P = 0.0004), albumin (P = 0.0634), and CEA (P = 0.0804) were selected as significant. In those patients who presented with liver involvement, variables predicted survival were gamma-GT (P = 0.0041), albumin (P = 0.0442), and the percentage of involved liver parenchyma (P = 0.0690). These results could be helpful for the stratification of future trials in advanced colorectal cancer.