Blood versus Crystalloid Cardioplegia
Blood cardioplegia resembles normal physiology and causes an improved postoperative cardiac outcome(23).  It allows for rapid cardiac arrest in an oxygenated environment, intermittent reoxygenation of the myocardium when cardioplegic solution is delivered so anaerobic substrates like glucose and insulin are no longer needed.  In a meta-analysis by Guru et al., blood cardioplegia was associated with decreased low-output syndrome (LOS) and creatine kinase-myocardial band (CK-MB) release therefore, a decreased mortality rate(23). Additionally, blood cardioplegia preserves ventricular performance and systolic function. A meta-analysis by Zeng et al. found that cold blood cardioplegia also caused a significantly lower rate of perioperative MI compared to cold crystalloid cardioplegia(24). Blood cardioplegia’s oxygen carrying capacity is advantageous(23). Hypothermia during cardiac surgery may offset this advantage by causing a left shift in the oxyhaemoglobin dissociation curve- so less oxygen is available for myocardial tissue(24). At 20⁰C only 50% of the total oxygen is released from blood cardioplegia, falling to 37% at 10⁰C. This is a stark contrast to crystalloid cardioplegia which releases all of its oxygen at all temperatures.
There is, however, myocardial uptake of haemoglobin bound oxygen from the blood cardioplegic solution during hypothermia. This is due to the acidotic environment developing during cardiac arrest which shifts the oxyhaemoglobin dissociation curve to the right, along with a higher tissue affinity for oxygen in hypothermia.  Delivering oxygen to the ischaemic myocardial tissue in blood cardioplegia may produce oxygen free radicals which cause ischaemia and reperfusion injury. However, blood also contains endogenous oxygen radical scavengers which protect against this(24).
It is important to note that haemoglobin has six times higher buffering capacity than plasma proteins. Whole blood also contains physiological oncotic pressure constituents. This minimises myocardial oedema, although this is not a significant issue with crystalloid cardioplegia either.  Red blood cells present in blood cardioplegia are also critical for perfusion of the capillary bed due to improved rheologic effects which leads to better oxygen delivery.
With blood cardioplegia there is also avoidance of the cost and need to prepare a complex pharmacological mixture as in crystalloid. However, apparatus to deliver blood cardioplegia is more complicated and expensive than crystalloid. Due to these higher costs and impaired visualisation of the operating field some surgeons choose not to use blood cardioplegia(23). A study by Gundry et al. concluded that blood cardioplegia was also associated with a higher rate of perioperative and postoperative conduction disturbances such as right bundle-branch block, than crystalloid(25).  On the contrary, crystalloid cardioplegia is widely used due to its simplicity, lower costs and good efficacy in the majority of patients(26). It allows for good visualisation of the operating field due to being clear.
A strong argument against the use of crystalloid cardioplegia is it contains only one-fourth as much oxygen as blood cardioplegia(19). It also decreases the oncotic pressure, increasing risk of oedema(26). Some studies have found crystalloid cardioplegia causes intracellular oedema, depletion of glycogen stores and higher release of CK-MB. It also causes a higher degree of intra-operative haemodilution which is associated with increased blood transfusion requirement, significantly greater intensive care requirements, longer hospital stays, higher operative costs and higher mortality rates and postoperative organ failure e.g. renal failure(26). Mullen et al. concluded in a study that although crystalloid cardioplegia was associated with higher incidence of MI and higher CK-MB release postoperatively, it lead to better right ventricular systolic function compared to blood cardioplegia(27). The advantages and disadvantages of blood and crystalloid cardioplegia are summarised in Table 3.