Trend of lipid and thyroid function tests in adults without overt thyroid diseases: A cohort from Tehran thyroid study

Optimizing cardiovascular disease mortality prediction: a super learner approach in the tehran lipid and glucose study

BACKGROUND & AIM

Cardiovascular disease (CVD) is the most important cause of death in the world and has a potential impact on health care costs, this study aimed to evaluate the performance of machine learning survival models and determine the optimum model for predicting CVD-related mortality.

METHOD

In this study, the research population was all participants in Tehran Lipid and Glucose Study (TLGS) aged over 30 years. We used the Gradient Boosting model (GBM), Support Vector Machine (SVM), Super Learner (SL), and Cox proportional hazard (Cox-PH) models to predict the CVD-related mortality using 26 features. The dataset was randomly divided into training (80%) and testing (20%). To evaluate the performance of the methods, we used the Brier Score (BS), Prediction Error (PE), Concordance Index (C-index), and time-dependent Area Under the Curve (TD-AUC) criteria. Four different clinical models were also performed to improve the performance of the methods.

RESULTS

Out of 9258 participants with a mean age of (SD; range) 43.74 (15.51; 20-91), 56.60% were female. The CVD death proportion was 2.5% (228 participants). The death proportion was significantly higher in men (67.98% M, 32.02% F). Based on predefined selection criteria, the SL method has the best performance in predicting CVD-related mortality (TD-AUC > 93.50%). Among the machine learning (ML) methods, The SVM has the worst performance (TD-AUC = 90.13%). According to the relative effect, age, fasting blood sugar, systolic blood pressure, smoking, taking aspirin, diastolic blood pressure, Type 2 diabetes mellitus, hip circumference, body mss index (BMI), and triglyceride were identified as the most influential variables in predicting CVD-related mortality.

CONCLUSION

According to the results of our study, compared to the Cox-PH model, Machine Learning models showed promising and sometimes better performance in predicting CVD-related mortality. This finding is based on the analysis of a large and diverse urban population from Tehran, Iran.

The thyroid hormone levels and glucose and lipid metabolism in children with type 1 diabetes: a correlation analysis

BACKGROUND

Type 1 diabetes mellitus (T1DM) is common in clinical setting, the relationship between thyroid hormone levels and glucose and lipid metabolism in patients with T1DM remains unclear. We attempted to analyze the correlation of thyroid hormone levels and blood glucose and lipid metabolism in children with normal thyroid function.

METHODS

Children with T1DM were selected, and 85 healthy children who underwent physical examinations in our hospital as control group. The characteristics and laboratory testing results were compared, the correlation of thyroid hormone levels with blood glucose and blood lipids was analyzed by Pearson correlation analysis.

RESULTS

A total of 167 participants were included. the low-density lipoprotein-cholesterol (LDL-C), triglyceride (TG), fasting blood glucose (FBG), and glycated hemoglobin (HbAlc) in T1DM patients were higher than those in healthy controls, while the high-density lipoprotein-cholesterol (HDL-C), thyroid stimulating hormone (TSH) and free triiodothyronine (FT3) in T1DM patients were lower than that of healthy children (all P<0.05). TSH was positively correlated with LDL-C (r=0.169, P=0.032), TC (r=0.182, P=0.017) and TG (r=0.197, P=0.008), negatively correlated with FBG (r=-0.196, P=0.023) and HbAlc (r=-0.328, P=0.002). FT3 was negatively correlated with TG (r=-0.182, P=0.011), FBG (r=-0.184, P=0.009) and HbAlc (r=-0.223, P=0.005). TG in the high TSH group and the middle TSH group is higher than that of low TSH group. However, FBG and HbAlc were lower than those in the low TSH group (all P<0.05). Compared with the low FT3 group, the TG, FBG and HbAlc decreased in the high and middle FT3 group (all P<0.05).

CONCLUSIONS

The serum TSH are closely associated with FT3 and glycolipid metabolism in children with T1DM, which may be the useful indicators to assess the severity of T1DM in clinical settings to provide insights into the management of T1DM.

Sensitivity to Thyroid Hormone Indices Are Closely Associated With NAFLD

BACKGROUND

Previous studies on the association between thyroid function and non-alcoholic fatty liver disease (NAFLD) have contradicted. Acquired resistance to thyroid hormone theory might provide a reasonable explanation for these contradictions. We aimed to analyze the association between sensitivity to thyroid hormone indices with NAFLD.

METHODS

A total of 4,610 individuals from the health medical center of the First Hospital of China Medical University were included in this study. The previously used thyroid feedback quantile-based index (TFQI_FT4) was calculated. Also, we substituted free triiodothyronine (FT₃) into the TFQI formulas to get the TFQI_FT3 index. NAFLD was defined using abdominal ultrasound.

RESULTS

Study results showed that FT₃/FT₄ and TFQI_FT3 were positively correlated with the triglyceride (TG), total cholesterol (TC), and low-density lipoprotein cholesterol (LDL-C) levels (P<0.05) and negatively correlated with high-density lipoprotein cholesterol (HDL-C) level (P<0.05). In contrast, TFQI_FT4 was positively correlated with HDL-C level (P < 0.05). After adjustment for multiple confounders, FT₃, FT₃/FT₄, and TFQI_FT3 were positively associated with the risks of dyslipidemia and NAFLD (P < 0.05). TFQI_FT3 and FT₃/FT₄ performed better than TFQI_FT4 on ROC analyses for NAFLD prediction, although the diagnostic sensitivity and specificity at the optimal cut-points were low. However, no association was observed between TFQI_FT4 with the risks of dyslipidemia and NAFLD.

CONCLUSION

TFQI_FT3 and FT₃/FT₄ can be used as new indicators for predicting dyslipidemia and NAFLD, although with low sensitivity and specificity at the optimal cut-points, while TFQI_FT4 has insufficient evidence in predicting dyslipidemia and NAFLD.

No association between subclinical hypothyroidism and dyslipidemia in children and adolescents

Background

There are controversies about the correlation between higher levels of thyroid stimulating hormone (TSH) and dyslipidemia in children. This study was designed to assess the relation between lipid profile components and TSH levels in children.

Method

This cross-sectional study was performed in a pediatric endocrinology growth assessment clinic in Shiraz, southern Iran. Children aged 2–18 years who referred to the clinic from January until April 2018 were included. TSH levels equal or above 5 mIU/L and lower than 10 mIU/L with normal free T4 (FT4) were considered as having subclinical hypothyroidism (SH).

Results

Six hundred sixty-six children were euthyroid while 181 had SH. No significant difference was found between the mean serum total cholesterol (P = 0.713), LDL-C (P = 0.369), HDL-C (P = 0.211), non-HDL-C (P = 0.929), and triglyceride (P = 0.215) levels between euthyroid children and subjects with SH. There was also no significant difference in the prevalence of dyslipidemias in any lipid profile components between the two groups. The adjusted correlation was not significant between TSH levels and any lipid profile component.

Conclusion

Based on the results of our study, we found no correlation between SH and dyslipidemia in children. The association between dyslipidemia and SH in children still seems to be inconsistent based on the results of this and previous studies. We recommend a meta-analysis or a significantly larger retrospective study on this subject.