Warm Blood Cardioplegia for Myocardial Protection: Concepts and Controversies

Physiologic Basis of Warm Blood Cardioplegia

Blood cardioplegic solution releases only 50% of its total oxygen content when cooled to 20 °C, and only 37% to 38% when cooled to 10 °C, because of the leftward shift of the oxyhemoglobin dissociation curve with hypothermia.¹¹ In addition, normothermic cardioplegic arrest results in an oxygen requirement of 1.1 mL oxygen/100 g/min, greater than 90% reduction of baseline values.¹² If oxygenated blood can reach the arrested myocardium, oxygen demand can be met via warm cardioplegia. Oxygen demand is reduced to less than 0.3 mL/100 g/min at 20 °C.¹² However, reduction in intracellular concentrations of metabolites and high-energy phosphates and suboptimal recovery of contractile function after intermittent hypothermic cardioplegia suggests that anaerobic metabolism does not completely meet the substantially reduced metabolic needs of the arrested hypothermic heart.¹³ In fact, researchers have found that extremely low myocardial temperatures (<10 °C) may inhibit energy-generating processes, such as glycolysis, more than they inhibit energy consumption.¹⁴ Finally, hypothermia has been shown to destabilize cell membranes, inhibit sodium-channel pumps, and cause calcium sequestration, thus resulting in edema, reperfusion injury, and impaired sarcoplasmic reticulum function.^14–16

Technical Considerations

Optimal Temperature. Most investigators have reported using warm cardioplegia at 37 °C, and others, at temperatures above 35 °C. In contrast, cold cardioplegia temperatures range between 4 and 15 °C, most below 10 °C.

Delivery Route. To induce cardiac arrest, warm blood cardioplegia can be delivered in an antegrade or retrograde fashion. In antegrade cardioplegia, the aorta is clamped and warm blood cardioplegic solution is delivered through the aortic root at a rate of 300 mL/min, along with a high-potassium cardioplegic mixture (target concentration, 20–25 mmol/L). Once the heart is arrested, a low-potassium solution (concentration, 7–10 mmol/L) is used at flow rates of 100 to 150 mL/min to maintain cardiac arrest. Warm blood cardioplegic solution containing at least 4 to 5 g of hemoglobin per 100 mL carries greater than 10 mL oxygen per 100 mL—more than enough to adequately oxygenate the arrested myocardium. In cases where myocardial activity resumes, repeat arrest can be accomplished with use of a high-potassium cardioplegic solution or lidocaine (50–100 mg). If myocardial activity continues, cardioplegic solution may be cooled to tepid or cold temperatures until the heart stops. To avoid hyperkalemia, ultrafiltration is recommended, and when the potassium level exceeds 5 mEq/L, the potassium concentration in the cardioplegic solution should be decreased.

Protecting the Right Side of the Heart. Cardiac venous drainage comprises 2 systems: the greater system drains approximately 70% of blood into the right atrium, and the lesser system drains approximately 30% of blood directly into the cardiac chambers.¹⁷ The lesser system is composed of the arteriosinusoidal and Thebesian vessels.

Retrograde cardioplegia delivery enables 70% of blood to perfuse the arrested myocardium through the coronary sinus and 30% to enter the ventricular chambers through the Thebesian and arteriosinusoidal vessels. Arrest is initiated in an antegrade fashion, followed by retrograde flow of approximately 100 to 150 mL/min. Coronary sinus pressure is usually maintained between 30 and 50 mmHg and monitored continuously. It has been argued that flow to the right ventricle (RV) may be deficient in the retrograde approach because retrograde flow to the RV has been estimated to be 20% that of antegrade flow.¹⁸ However, during arrest, the RV oxygen demand is lower than the 1.1 mL needed by the left ventricle (LV). Thus, retrograde delivery is sufficient to meet the requirements of the arrested right side of the heart.¹⁹ In addition, retrograde delivery may provide more uniform distribution of cardioplegic solution in advanced coronary artery disease because it is delivered through the venous system, thus bypassing any coronary artery stenosis.

Higher flow rates of 250 to 300 mL/min are advisable for operations involving valves or dilated and hypertrophic hearts, which have higher metabolic demands.^20,21 A left superior vena cava is not an absolute contraindication to this technique, but intermittent antegrade cardioplegia may be preferable, or cardioplegia can be delivered through completed vein grafts. As an alternative, the left superior vena cava can be cannulated. Continuous warm blood cardioplegia may mask the surgical field during crucial parts of any operation. However, continuous cardioplegia is not necessary; it can be interrupted for up to 10 min for adequate exposure.²²

Table I summarizes the technical aspects of warm cardioplegia.

Table I The Ten Commandments of Warm Blood Cardioplegia

View Fullscreen

Outcomes of Warm versus Cold Cardioplegia

The Warm Heart Trial, published in 1994, was among the first randomized trials to compare the results of warm and cold cardioplegia.²³ It included 1,732 patients randomized to intermittent antegrade warm or intermittent antegrade cold cardioplegia for coronary artery bypass grafting (CABG). Investigators found that myocardial enzyme release (12.3% vs 17.3%; P <0.001) and postoperative low-output syndrome (6.1% vs 9.3%; P <0.01) were lower after cardiac surgery in the warm group. However, there was no difference in nonfatal Q-wave myocardial infarction (MI), in-hospital mortality rates, or 30-day mortality rates.²³ At 84 months, the survival rate was significantly reduced in patients who had nonfatal perioperative outcomes, such as low-output syndrome (P <0.0001).²⁴ The original trial data showed that these nonfatal perioperative outcomes were significantly more prevalent among the cold group. A substudy of this trial by Christakis and colleagues²⁵ stratified patients into high, medium, and low risk. It showed a significant reduction in morbidity and mortality rates overall as defined by death, low-output syndrome, or MI: 15.9% (warm) versus 25.2% (cold) (P <0.01). The risk reduction was similar in all 3 risk categories, indicating that warm cardioplegia has a similar effect on outcomes among high-, medium-, and low-risk patients.²⁵ A statistically insignificant reduction of in-hospital death was shown in multiple randomized trials around the same time.^26–29Figure 1 shows an overview of the effect of warm cardioplegia on in-hospital mortality rates.

View Figure

• Download as Powerpoint
• Download Figure

Mallidi and associates³⁰ conducted a prospective cohort study of 6,064 patients from 1989 through 2000 to evaluate the short- and long-term clinical outcomes of cold versus warm or tepid cardioplegia during CABG. The large sample size of this study enabled evaluation of low-frequency events. Cold cardioplegia was found to be an independent risk factor for early death or MI (adjusted odds ratio [OR]=1.70; 95% CI, 1.30–2.21; P <0.0001). They also reported significantly higher risk of postoperative intra-aortic balloon pump (IABP) insertion (OR=1.77; 95% CI, 1.31–2.39; P =0.0002) and low-output syndrome in the cold group (OR=1.71; 95% CI, 1.42–2.05; P =0.0006), but no significant difference in stroke risk. Patients were monitored for 6 years after their initial operation. Cold cardioplegia was not a significant predictor of late death alone (P =0.09) but did predict the combined endpoint of late death or MI (P =0.0001). Similar results were reported by investigators in a retrospective cohort study, which showed significant decreases in overall and in-hospital death, need for inotropic support, peak creatine kinase-myocardial band isoenzyme (CK-MB) levels, and intensive care unit stay in the warm versus cold group.³¹ Other study results have not shown such clear benefits. Pelletier and colleagues³² found no significant difference in clinical outcomes between groups, yet results of a retrospective cohort study indicated a detrimental effect of warm cardioplegia when aortic cross-clamp time was 75 min or longer.³³ Cardiac-related death was more prevalent in the warm group (P =0.015); however, this is the only study in the literature documenting a significantly inferior outcome of warm cardioplegia in the context of mortality rates.

Yau and associates³⁴ evaluated the effects of cardioplegia temperature on ventricular function. The authors concluded that although warm cardioplegia resulted in greater lactate production, recovery of oxygen consumption improved during reperfusion. This finding correlated clinically with increased preload-recruitable stroke work index and increased early diastolic relaxation. Therefore, ventricular recovery was found to be superior in warm versus cold cardioplegia. Landymore and colleagues³⁵ did not find LV stroke work index to be significantly different between warm and cold cardioplegia and concluded that both methods are suitable. Investigators in multiple studies have shown improved electrophysiologic function with the use of normothermic cardioplegia, which correlates clinically to fewer new or permanent conduction disturbances, including ventricular arrhythmia.^31,36,37

The effects of cardioplegic temperature on clinical outcomes in aortic valve replacement surgery remain equivocal. In a prospective randomized study by Ascione and associates³⁸ in 2002, cold cardioplegia was associated with decreased myocardial ischemia in comparison with warm cardioplegia in patients undergoing valve replacement surgery. They found increased levels of lactate and troponin I, as well as a higher alanine/glutamate ratio, in patients in the warm-blood group.³⁸ These results were mirrored by those of a recent study in which warm blood cardioplegia was found to be inferior to cold crystalloid cardioplegia in patients with severely hypertrophic LVs who underwent aortic valve surgery.³⁹ These differences in myocardial markers of ischemia, however, do not appear to translate into inferior clinical outcomes. Calafiore and co-authors⁴⁰ retrospectively compared patients who underwent intermittent antegrade warm or intermittent antegrade cold cardioplegia for valve surgery and found significantly lower cardiac-related mortality rates in the warm cardioplegia group (P <0.01). There was also significantly higher recovery of spontaneous rhythm (P <0.0001) and a lower incidence of low-output syndrome and ventricular dysrhythmia. These results have been corroborated by other investigators who have found higher rates of return of spontaneous rhythm, shorter postoperative ventilation times, and lower postoperative CK-MB levels in patients undergoing warm cardioplegia than in those undergoing cold blood or cold crystalloid cardioplegia.^41–43 The consensus is that warm cardioplegia is adequate, if not superior, in protecting hypertrophied LVs during valve procedures, provided that flow is adequate.

Caputo and colleagues¹⁴ reviewed the history of and clinical evidence for intermittent warm cardioplegia. On the basis of results from several animal and human studies, they concluded that “intermittent warm cardioplegia is safe and effective.” Fan and associates⁴⁴ conducted a meta-analysis of 41 randomized clinical trials that compared the outcomes of warm and cold cardioplegia after cardiac surgery. Collectively, the analysis included 5,879 patients—2,944 in the warm-blood group, 2,007 in the cold-blood group, and 928 in the cold crystalloid group. The study results showed no significant difference in all-cause death, MI, length of stay, postoperative atrial fibrillation, or IABP use, when comparing warm and cold cardioplegia. The authors, however, documented significant improvement in postoperative cardiac index (weighted mean difference [WMD]=0.28; 95% CI, 0.26–0.31; P <0.00001) with warm versus cold cardioplegia. The day 0 and peak cardiac troponin levels and peak CK-MB were all significantly lower in the warm cardioplegia group (WMD= −0.61; 95% CI, −1.04 to −0.17; P =0.006; WMD= −1.45; 95% CI, −2.47 to −0.42; P =0.006; and WMD= −8.03; 95% CI, −13.08 to −2.97; P =0.002, respectively). The authors concluded that both warm and cold cardioplegia were adequate for myocardial protection and that warm cardioplegia was associated with improved postoperative hemodynamic performance and less myocardial injury. The study, however, did not evaluate long-term outcomes.

Abah and associates⁴⁵ conducted a best-evidence review to answer whether warm or cold blood cardioplegia was superior for myocardial protection. They summarized the findings of 20 best-evidence studies, which included meta-analyses as well as prospective cohort, retrospective cohort, and randomized trials. The authors concluded that warm and cold cardioplegia have similar short-term mortality rates. They also found that warm cardioplegia reduces adverse postoperative events. A meta-analysis comparing blood and crystalloid cardioplegia included meta-regression analysis, which showed no relationship between the outcomes of death, MI, and low-output syndrome, delivery route, temperature, or timing of delivery.⁴⁶

Most studies comparing warm and cold cardioplegia were done more than a decade ago. Many advances have been made in surgical equipment and cardiac care since that time, so updated studies are needed to provide a more accurate evaluation of clinical outcomes. In a more contemporary study, Candilio and colleagues⁴⁷ compared perioperative myocardial injury in patients who had antegrade cold-blood cardioplegia, cross-clamp fibrillation, or antegrade retrograde warm cardioplegia. They found a significantly lower level of perioperative high-sensitivity troponin T in the warm group than in the cold group (P =0.018) or the fibrillation group (P =0.044).⁴⁷ Results of a prospective observational study in 2015 showed no difference in perioperative MI or spontaneous return of rhythm between intermittent antegrade warm and cold cardioplegia, but IABP use and postoperative CK-MB levels were significantly lower in the warm group. Aortic cross-clamp time was not significantly different between the groups.⁴⁸ In contrast, intermittent warm cardioplegia had no statistically significant effect on clinical outcomes when compared with cold cardioplegia in a study by Kuhn and co-authors.⁴⁹ It did, however, result in greater endothelial injury quantified by levels of circulating endothelial cells, von Willebrand factor, and soluble thrombomodulin. Zeriouh and coworkers⁵⁰ analyzed perioperative outcomes and long-term survival after warm versus cold cardioplegia in emergency and elective CABG situations. They found no significant difference in peri-operative results and similar long-term mortality rates between groups. The authors suggested that intermittent warm cardioplegia is more cost-effective than cold and should be used in both elective and emergency situations.⁵⁰

Tables II and III summarize clinical study results.

Table II Summary of Findings in Randomized Studies

View Fullscreen

Table III Summary of Findings in Cohort and Observational Studies

View Fullscreen

Delivery Route and Temperature

The ideal delivery technique and temperature for normothermic cardioplegia are still debated. Hayashida and colleagues⁵¹ compared warm, tepid, and cold cardioplegia and antegrade versus retrograde delivery of each. They concluded that warm retrograde cardioplegia results in greater oxygen consumption and lactic acid release, as well as delayed ventricular recovery. Warm antegrade cardioplegia similarly increased oxygen consumption and lactic acid release, but it promoted the immediate restoration of ventricular function in high-risk patients. The LV stroke work index was significantly greater after warm antegrade and tepid antegrade cardioplegia than after cold antegrade cardioplegia (P <0.05) after 12 hours. Although cold cardioplegia reduced metabolic demand, functional delays in ventricular recovery limited its usefulness in resuscitation, according to the authors.⁵¹ Tepid retrograde cardioplegia resulted in decreased oxygen consumption and metabolite release, yet ventricular recovery was slow. A tepid antegrade technique was suggested to provide the best myocardial protection, not only in ensuring cardioplegic delivery but also in reducing metabolite demands.

Neurologic Complications

Engelman and associates⁵² analyzed postoperative neurologic function and fibrinolytic potential among patients randomized to cold, tepid, and warm perfusate and cardioplegia during coronary revascularization surgery. The Mathew scale uniformly decreased postoperatively, but no significant distinction occurred between groups (P >0.5). The cold group had significantly more patients with clinically suspected stroke, but there was no significant difference in computed tomography–confirmed stroke incidence between groups (P =0.1). Another trial, by the Emory University group,⁵³ confirmed the similar effectiveness of normothermic and hypothermic cardioplegia for myocardial preservation but raised the possibility of an increased incidence of postoperative neurologic events. However, systemic normothermia is not necessary to administer warm blood cardioplegia. Moreover, the use of a partially occluding clamp technique for the construction of the proximal anastomoses and the use of hyperglycemic crystalloid solutions in the Emory trial may have exacerbated intraoperative neurologic injury. In a study where cardioplegic temperature was altered but cardiopulmonary bypass was performed at normothermia, there was no difference in stroke rates between cold, tepid, or warm cardioplegia groups.⁵⁴ Similarly, in studies where both warm and cold cardioplegia were performed with systemic hypothermia, stroke rates were low and did not differ significantly.⁵⁵ Thus, we recommend that, with warm cardioplegia, systemic perfusion be maintained at 32 to 33 °C to protect the brain.

Duration of Ischemic Intervals

Many surgeons opt not to use warm delivery because of their misconception that warm cardioplegia should not be interrupted. Initially, interrupting cardioplegic flow had been shown experimentally to be deleterious in porcine heart models.^56,57 However, many results of animal studies have since shown that warm cardioplegia may be safely interrupted for about 10 min without detrimental effects.^10,58,59 In one animal study, no detrimental effects were seen after ischemic periods of up to 15 min.⁶⁰ These findings correlated with those in human studies. Rousou and co-authors²² studied the correlation between intermittent warm cardioplegia and negative clinical outcomes and found that repeatedly interrupting cardioplegic flow for up to 10 min per interruption and keeping the cumulative interruption time ≤45% of the arrest period has no adverse clinical or metabolic effects on the myocardium. Ali and Kinley⁶¹ concluded that intermittent antegrade warm cardioplegia was safe when cross-clamp time was less than 90 min. Furthermore, Lichtenstein and associates⁶² studied the effects of the longest single interruption as well as cumulative ischemic time during interrupted warm cardioplegia and found that the former has the largest effect on outcomes. They concluded that a reasonable margin of safety exists with intermittent antegrade warm blood cardioplegia, and that repeated interruptions of warm blood cardioplegia are unlikely to lead to adverse clinical results if single interruptions are shorter than 13 min.⁶²

Right Ventricular Protection

Preserving the RV is a challenge during cardioplegic arrest, especially in patients who have an occluded right coronary artery that blocks antegrade flow. Retrograde flow provides a less robust distribution to the RV but may mitigate this problem. Honkonen and associates⁶³ found that recovery of RV function after coronary surgery is improved with warm continuous cardioplegia compared with intermittent cold cardioplegia. Kulshrestha and colleagues¹⁹ studied RV ejection fraction (RVEF) pre- and postoperatively in patients undergoing retrograde warm cardioplegia and found no difference in postoperative RVEF but a significantly improved RV systolic work index (P <0.0001). Furthermore, results of a similar study showed improved RVEF in patients who had intermittent warm cardioplegia compared with those who had cold cardioplegia.⁶⁴ In the CABG Patch Trial, the rate of RV dysfunction after normothermic cardioplegia was lower than that after cold, suggesting that interrupted warm cardioplegia adequately protects the right side of the heart.⁶⁵