Whole Blood Cardioplegia: Do We Still Need to Dilute?

I have been tasked to talk about the topic “Whole Blood Cardioplegia: Do We Still Need to Dilute?” This is a question that was raised almost 20 years ago by Dr. Philippe Menasché (1) from Paris, and it’s interesting that we are still asking this question in 2015. We will get into the data that drive this persistent question during this discussion. This presentation will cover 1) a bit of history of blood cardioplegia, 2) the attributes of both hemodiluted blood cardioplegia and all-blood (microplegia), and 3) provide some answers to the question of why do we actually need all of this water from recent data provided by preclinical trials in the research laboratory, and from clinical trials. There is not a whole lot of data gathered over the past 20 years that is relevant to this subject directly, which is a little bit surprising and disappointing. Finally, I will try to use these data to come to some sort of a resolution that you can take home. Now, as you know, cardioplegia has been the cornerstone of myocardial protection for many, many years. Its most obvious role is to achieve and maintain rapid arrest; the name cardioplegia (2) (cardio = heart; plegia = paralysis) implies this role. Not only is cardioplegia an arresting solution, but it is used as a vehicle to deliver therapeutics to prevent and treat ischemia and reperfusion injuries. Cardioplegia prevents ischemia by 1) lowering the energy demands by arresting the heart, 2) imposing myocardial hypothermia, and 3) delivering various anti-ischemic agents such as adenosine, magnesium, glutamate, and aspartate. Cardioplegia can prevent reperfusion injury not only by limiting ischemia (without ischemia there is no reperfusion injury), but also by controlling the conditions and composition of the reperfusate (3). Specifically, conditions refer to temperature (warm induction to resuscitate the energydepleted heart, hypothermia during maintenance, warm terminal perfusate, and low delivery pressure to avoid edema and microvascular injury), and composition refers to buffers to limit tissue acidosis, hyperonconicity, agents to reduce calcium, and the addition of therapeutics that directly address the pathophysiology of reperfusion injury (4). The final phase of cardioplegia just before the crossclamp is released is designed to resuscitate the heart from depolarized arrest and hypothermia (if this is used) in preparation for reanimation and resumption of contractile function. That is a formidable task for a heart that’s been somewhat confused and potentially damaged by ischemia, arrest, hyperkalemia, hypothermia, and reperfusion. So what is an ideal cardioplegia solution? Arrest is almost universally achieved by hyperkalemia which depolarizes the cell, thereby preventing repolarization and subsequent generation of action potentials. However, hyperkalemia has its darker sides which contribute to postcardioplegia dysfunction and morbidity (5). Efforts have been made to arrest the heart in a polarized state with normal membrane potential by using KATP channel openers, the combination of adenosine–lidocaine–magnesium (adenocaine), or profound hypocalcemia. Regardless of the modality used to achieve arrest, this arrest presents the surgical team with a quiet, bloodless field, so they can concentrate on the task at hand. Hence, the first role of cardioplegia is to effectively and rapidly arrest the heart. Second, cardioplegia formulations should be based on sound scientific principles developed and tested by appropriately designed and statistically powered preclinical and clinical studies that avoid design flaws such as lack of subject randomization, lack of appropriate treatment blinding, and most of all, insufficient statistical power to declare significance or demonstrate non-significance. A danger in trials claiming non-superiority or equivalence by showing no difference between two or more treatments is that lack of significant difference may represent lack of power to show differences. A Type II error is when statistical differences are missed when actual biological (or treatment) differences exist. Lack of significant difference is often due to insufficient power (e.g., insufficient numbers of subjects or large deviations around the mean) to demonstrate statistical differences, or when the end point has no biological connection or relevance to the process being tested or the actions of the therapeutic. Equivalence does not equal a lack of difference between Address correspondence to: Jakob Vinten-Johansen, PhD, Emeritus Professor, Division of Cardiothoracic Surgery, Department of Surgery, Emory University, Atlanta, GA 30322. E-mail: [email protected]