Oral Triiodothyronine for Infants Less Than 5 Months & Malnourished Children Undergoing Cardiopulmonary Bypass in Indonesia

Overview

Infants and children undergoing cardiac surgery require mechanical support for circulation called cardiopulmonary bypass. The artificial pump ensures that blood flow to the body is maintained during surgical procedure on the heart. However, the pump circulation disrupts regulation of thyroid hormone. This hormone is a critical regulator of heart and circulatory function. Deficiencies in thyroid hormone adversely affect recovery of the heart after cardiac surgery. A recent study by the principal investigator showed that supplementing patients younger than 5 months old shortened the time spent on the ventilator (breathing machine) after surgery. Populations undergoing heart surgery in developing countries such as Indonesia are different than those in the U.S. and other industrial nations. For many heart defects children undergo surgery at an older age. They are then sicker for longer periods, and are often malnourished. These older malnourished children have lower thyroid hormone levels even prior to surgery, and show dramatic further drops after surgery. Shortening time on the ventilator should improve their recovery, and also improve overall availability of resources in developing countries. The formulation of thyroid hormone provided in the U.S and given intravenous is relatively expensive. However, a much more inexpensive formulation is available in Indonesia, which can be given through a tube extending from the nose to the stomach. We propose testing this formulation made in Korea, known as Tetronine, in two vulnerable populations in Indonesia: children younger than 5 months of age, and those older than 5 months who are malnourished. All these patients will be undergoing cardiopulmonary bypass for congenital heart defects. This project will be an important collaboration between U.S and Indonesian investigators. If successful, it will have a general positive impact on care of the entire population of young children undergoing cardiac surgery in developing countries by reducing time on the ventilator and improving access to care.

Methods Study Overview and Objectives We will perform a randomized double blinded placebo controlled trial in Indonesian children under 3 years of age, who are undergoing cardiopulmonary bypass procedures for correction or palliation of congenital heart disease. The study is approved by the Ethic Committee for Medical Research National Cardiovascular Center Harapan Kita (NCCHK). Age 3 years was chosen as the cutoff, as the population \< 3 years has the highest mortality rate in Indonesia. Also, this cutoff reduces age-based population heterogeneity, which could confound the analyses. We will study two independent vulnerable cohorts within this population as previously noted. 1. Age \< 5 months 2. Age \> 5 months and \< 3 years with moderate to severe malnutrition Moderate to severe malnutrition is defined as weight more than 15% below the predicted value for height. The predicted value will be determined using WHO clinical growth charts based on the patient gender and baseline length and weight measurements. There is no overlap in study cohorts (patients ≤ 5 months of age and patients \> 5 months of age with moderate to severe malnutrition form mutually exclusive groups). Thus, this study can be viewed as two separate, independent trials, each with an overall type I error rate of 0.05 (2-sided). The primary endpoint within each study cohort is a comparison of time to the first of death, 7 days, or extubation based on the log-rank statistic from separate Cox Proportional Hazards Models. These models will include Aristotle Score and other baseline covariates imbalanced across treatment groups. Patients will be monitored until the first of death, withdrawal of consent, or hospital discharge. Safety Endpoints Safety endpoints for this study include: 1. Death from any cause prior to hospital discharge 2. Death or use of mechanical circulatory life support 3. Incidence of serious adverse events 4. Incidence of unexpected adverse events 5. Incidence of any arrhythmia and specific arrhythmias Objective 1. To establish the efficacy of Tetronine®, Dalim Biotech Korea (oral T3) in improving clinical outcome for Indonesian children age 0 to 5 months after congenital heart disease surgery associated with cardiopulmonary bypass. Eligible population: The age distribution of children undergoing cardiac surgery in Indonesia is substantially different than in the U.S. Very few patients at NCCHK less than 6 weeks age undergo cardiopulmonary bypass procedures (7/400 sequential patients 2011-12). As the The Triiodothyronine during Cardiopulmonary bypass in Infants and Children (TRICC) trial indicated benefit for T3 supplementation in U.S children under age 5 months, we will study this age group within the scope of specific aim 1 using Tetronine. This will be a single center study performed in Indonesia at NCCHK. Historical records show that approximately 170 NCCHK patients will be eligible for the specific aim 1 portion of the study within the 3 years of enrollment. Objective 2. Patients older than 5 months will be enrolled if at screening their weight is equal or less than -2 Z Score for height on WHO DATA Table of Infant Weight for Length Charts by gender. Historical records at NCCHK show that 160 patients will be eligible for specific aim 2 over a 3 year period. Studies required to achieve these objectives will be run concurrently. Power Analysis: We determined a requirement of 120 patients for objective 1 and 100 patients for specific aim 2 to achieve reasonable power. Safety Adverse Event (AE) Reporting Although this study will be performed in Indonesia and does not come under regulation by the U.S Food and Drug Administration, we will use FDA standards to report adverse events. Reporting of adverse events in cardiac surgical trials in infants is complicated as there is no regulatory guidance specific to conducting clinical research in critically ill children. We will use The Pediatric Heart Network reported method with predefined "sentinel events" in place of serious adverse events (SAEs). Data Analysis Two populations will be considered. ITT Population The primary efficacy analysis population will be based on the principle of Intention-to-Treat (ITT), and will include all randomized patients who receive any double-blind study medication. Patients will be grouped according to randomization assignment. As-Treated Population The As-Treated population will only be used if there are patients who mistakenly receive the wrong treatment or have other issues with dosing. In the As-Treated analyses, patients will be grouped according to the treatment actually received. All safety analyses will consider only the As-Treated population. Analysis of the primary efficacy outcome The primary analysis will be performed using Cox Proportional Hazards, including terms for stratification factors (age, Aristotle score and nutrition status). Patients will be included in analyses if they were randomized and received at least one dose of study drug according to the principal of Intention-to-Treat (ITT). Occasionally patients are randomized and then surgery is cancelled or rescheduled due to scheduling conflicts, and not due to patient related issues. This can rarely result in the patient missing the study medication when the operation is rescheduled at the last minute. However, since this study is double-blind, the ITT principal is preserved because the treatment assignment is blinded and so unknown to persons responsible for these decisions. Handling of missing values/ censoring/ discontinuations In the event that the parents of a patient withdraw consent prior to extubation, the patient will be censored at the time of discontinuation. Patients who die prior to extubation will be censored at the time of death. For the primary efficacy analysis, patients who remain intubated at 7 days post cross-clamp removal (168 hours) will be censored at 7 days. It is imperative that the date and time of extubation be recorded accurately. If for some reason the date and time of extubation is not recorded, patients will be considered intubated through 7 days and censored at 7 days. Sensitivity analyses The two planned sensitivity analyses will use an identical model to the primary analysis, but time to extubation may be different for some patients (if extubation occurs after 7 days and before 30 days, for the first sensitivity analysis, or if the successful extubation does not coincide with clinical readiness for extubation for the second sensitivity analysis). Subgroup Analyses For each subgroup variable, a treatment-by-subgroup term will be added to the primary analysis model to test for heterogeneity of treatment effect across subgroups. The primary analysis model will be repeated within each subgroup (removing terms for stratification factors that are related to the subgroup of interest for individual models). Planned subgroups for analysis include: 1. Nutritional state (normal, mild, moderate and severe malnutrition as determined by percent of expected weight for length) 2. Age \< 6 weeks versus ≥ 6 weeks 3. Age \< 24 months versus ≥ 24 months 4. Aristotle score category (6.0-7.9, 8.0-9.9, 10.0-15.0 4) 5. High risk versus Lower risk Aristotle score (\< 10 versus ≥ 10) Analysis of secondary efficacy objectives Time on inotropic agents will be analyzed based on the percent of time in the first 48 hours that inotropic agents were used. Patients who do not require inotropic agents will be included in the analysis with 0% of the first 48 hours of use. The distribution of the percent of 48 hours on inotropic agents will be right-skewed, so the Wilcoxon Rank-Sum Test will be used to assess treatment differences. Non-parametric regression may also be used to account for stratification factors. Inotropic score is likely to also be right-skewed, so similar analysis techniques will be used. Mean heart rate and mean systemic mean blood pressure at baseline, 1, 6, 12 and 24 hours post cross-clamp removal will be assessed using ANCOVA, controlling for stratification factors. Total and free T3 levels at baseline, 1, 12, 24 and 48 hours post cross-clamp removal will be right-skewed. Lactate levels will also be drawn at these times points as an indicator of metabolic status, and often used as a clinical indicator in pediatric cardiac intensive care units. Differences across treatment groups will be assessed using the Wilcoxon Rank-Sum Test will be used to assess treatment differences. Non-parametric regression may also be used to account for stratification factors. Longitudinal analyses may also be performed. Post-operative ICU and Hospital stay durations will be right-skewed. Differences across treatment groups will be assessed using the Wilcoxon Rank-Sum Test will be used to assess treatment differences. Non-parametric regression may also be used to account for stratification factors. Analysis of safety endpoints Safety endpoints will be reported individually, as incidence rates by treatment. Fisher's Exact Test or the Chi-square test will be used to assess treatment differences. Odds ratios adjusted for stratification factors may also be produced.