Children with congenital heart disease have significant morbidity including low cardiac output syndrome and subsequent organ dysfunction that may be prevented by optimization of circulatory function. More than half of these children receive milrinone. Clinical evaluation cannot distinguish between patients with sub-therapeutic, therapeutic, and toxic milrinone drug levels. Consequently children who require pharmacologic circulatory support may be receiving sub-optimal dosing, and children who do not need milrinone may be receiving milrinone unnecessarily. The primary objective of this study is to determine if optimizing milrinone levels with therapeutic drug monitoring in critically ill children following cardiac surgery improves clinical outcomes and reduces the duration of milrinone infusion. This study hypothesizes that optimizing milrinone levels with therapeutic drug monitoring in critically ill children following cardiac surgery will improve clinical outcomes and reduce the duration of milrinone infusion.

The proposed trial is a pilot of an open label, randomized trial of milrinone therapeutic drug monitoring in patients < 18 years treated with milrinone following open-heart surgery for congenital heart disease at the Hospital for Sick Children, Toronto. We will randomize patients to (1) receiving therapeutic drug monitoring or (2) control patients who receive standard care. Standard care involves titration of milrinone infusion based on clinical examination by the treating team. Control patients will have milrinone plasma levels drawn but not analysed until the end of the study. The intervention is (1) regular measurement of milrinone levels; and (2) physician feedback of plasma levels in experimental arm by the ICU pharmacist. After obtaining consent, eligible subjects will be allocated to a trial group by random assignment (sealed envelopes) within 3 strata in a 1:1 allocation. These strata are < 2 years, 2- 10 years and > 10 years to ensure equal distribution of these age ranges in each group for pharmacokinetic analysis. For the intervention group, sampling for milrinone levels will occur at 0 hours (upon arrival to ICU with routine admission blood collection) and approximately every 6- 8 hours (whenever the line is accessed for routine blood work). The last sample will be taken 6-8 hours after cessation of infusion, or if the patient leaves the ICU, or the maximum amount of blood sampling has been reached or after 7 days for children weighing more than 8 kgs (or 5 days for children weighing less than 8 kgs), which ever comes first. Follow-up will be exclusively during the period of hospitalization in the ICU until ICU discharge. An optional algorithm with a proposed titration for milrinone will be provided for use at the discretion of the treating team. Clinical outcomes will be measured as a composite outcome of dysrhythmia, LCOS and death.

Inclusion Criteria:

• Admitted to a Pediatric (0 - 18 years) Intensive Care Unit following cardiopulmonary bypass (CPB) and surgery for congenital heart disease.
• Clinical decision by treating team to start milrinone infusion.
• Anticipated to receive milrinone infusion for more than 24hs. This limit will increase the proportion of sicker children in the sample, increasing the power of the study.
• Has an arterial line, and a central venous line defined as radiologically confirmed line
• Informed consent obtained

Exclusion Criteria:

• Premature infants (<36 weeks post-conceptual age) or weight less than 2.0 kg.
• Failure to provide consent