Can Oral T3 Normalize Thyroid Hormone Levels Following Cardiopulmonary Bypass in Children?

The Research Ethics Board at the National Cardiovascular Center Harapan Kita approved this study and written, informed consent was obtained from the parents or legal guardians before randomization. Randomization by block permutation was performed to determine treatment group assignment. Randomization occured on the day before surgery by a nurse investigator. A pharmacist who was not involved in the study prepared the study medication. Investigators and participants were blinded to the assigned group until after the end of the study.

Thyroid hormonal levels were analyzed by standard 3rd generation thyrotropic-stimulating hormone (TSH), serum free T4 (FT4), free T3 (FT3), and total T3 (TT3) Micro particle Enzyme Immunoassays (Abbott Laboratories, Abbott Park, USA). The serum total T4 (TT4) assay used a Fluorescence Polarization Immunoassay (Abbott Laboratories, Abbott Park, USA). Hormone levels were measured on induction of anesthesia, before the study drug was given (T0) and at 1, 6, 18, 36 and 72 hours after removal of the aortic-cross-clamp.

Baseline clinical data collected included age, gender, birth weight, type of operation, and Aristotle score. Diagnosis and operative procedures were classified as high or low risk with an Aristotle score cut off of ≥ 9 as high risk. As modifying factors, we measured duration of surgery, cardiopulmonary bypass (CPB) time, cross-clamp time, ultrafiltration during CPB and degree of hypothermia during CPB, and the use of amiodarone. Non-pulsatile perfusion technique was used during CPB. Steroid (methyl prednisolone 35-50 mg/kg) was given before CPB. We used povidone-iodine for skin disinfection in all subjects. Although this study was not powered to detect clinical differences between the treatment groups, clinical outcome parameters were measured as a potential guide to subsequent adequately powered larger treatment studies. Serum lactate was measured at 1 hour, 4 hours and day 1 post surgery. Hemodynamic monitoring included heart rate, heart rhythm, and blood pressure which were recorded hourly for the first 6 hours then every 6 hours until 72 hours after surgery. Overt symptoms of hyperthyroidism were grounds for immediate removal of the subjects from the study. Time to extubation and length of stay in the intensive care unit and hospital were recorded.

Statistical analysis and sample size: The primary efficacy analysis assessed the difference between the treatment (high-dose, low-dose) and control groups with regard to the effect of T3 supplementation on the measured TT3 and FT3 serum levels. We anticipated a difference of 2.0 pg/ml in FT3 with a standard deviation of 0.8 pg/ml between groups. For a statistical power of 80% to identify a treatment effect and at a level of significance of 0.05 ( 2-sided), the target total sample size was 45 subjects, with 15 in each treatment group. Demographic data, safety and clinical outcomes were compared using the X2 test. Continuous variables for characteristics and outcomes were analyzed using one way ANOVA for data with normal distribution or the Kruskal Wallis test for not normally distributed data. Repeated measures ANOVA was used to analyze all thyroid hormone levels and clinical outcomes for those variables that were measured repeatedly over time. Paired Student's t-test for parametric or Wilcoxon signed rank test for non-parametric tests were used to evaluate the mean difference of hormone levels and clinical outcomes over time in each treatment group. Statistical significance was defined by p-values less than 0.05. Descriptive statistics are reported as mean ± standard error of the mean.