HTK versus multidose cardioplegias for myocardial protection in adult cardiac surgery: A meta‐analysis

Histidine–tryptophan–ketoglutarate (HTK) cardioplegia for myocardial protection obviates the need for maintenance cardioplegia doses, and thus allows for greater focus on procedure accuracy. The aim of this meta‐analysis is to evaluate the safety and efficacy of HTK versus multidose cardioplegias during cardiac surgery in an adult population.


| INTRODUCTION
Myocardial protection is of concern to the cardiac surgical team during cardiopulmonary bypass (CPB). CPB was first demonstrated by Gibbon in 1954 as a mechanism to prevent major organ ischemia during asystole.
Protection was not, however, sustainable for the duration of complex cardiac correction procedures, with an upper range of postoperative inhospital mortality of 65%. 1 Cardiac examination post-mortem revealed "stone heart"-irreversible myocardial contraction from ATP-mediated actin-myosin coupling, with failure of actin-myosin uncoupling due to ATP depletion. 2 This low level of success was thought to be associated with the high concentration of citrate leading to chelation of magnesium and calcium. 3 However, the idea of using moderately elevated potassium re-emerged in the 1970s. 3 One of the foremost proponents of crystalloid hyperkalemic cardioplegia was the Bretschneider group in Germany. In 1980, they devised the histidine-tryptophan-ketoglutarate (HTK) cardioplegia, which is the subject of this meta-analysis. 3 HTK (branded as Custodiol® by Essential Pharmaceuticals LLC), has been presented by industry to surgeons as an alternative solution that exceeds other cardioplegias in myocardial protection. 4 This claim relies on the single-dose administration of HTK compared with other multidose cardioplegias (MDC), sparing time in the adjustment of equipment during cardioplegia re-administration, allowing greater time to operate and thus a decreased CPB duration. 5 Other benefits include a lower concentration of sodium, calcium, and potassium compared with other cardioplegias with cardiac arrest arising from the deprivation of sodium. 5 Finally, histidine is thought to aid buffering, mannitol and tryptophan to improve membrane stability, and ketoglutarate to help ATP production during reperfusion. 6 This is the first meta-analysis to solely focus on HTK versus MDC during cardiac surgery in the adult-acquired disease population. A metaanalysis by Gambardella et al. 7 in 2019 claimed to investigate single dose versus MDC in the adult population; however, the inclusion of Careaga et al. 8 which contained both adult and pediatric populations reduces the applicability of their findings to the clinical landscape of adult cardiac surgery. Our meta-analysis also includes the findings of Vivacqua et al. 9 Vivacqua et al. 9 was submitted for publication 5 days after Gambardella et al., 7 and so was not included in their analysis. In addition, our meta-analysis solely focuses on randomized controlled trials (RCTs; Gambardella'a included both RCTs and propensitymatched studies). Most importantly, we supply new data on more outcomes and include subgroup analysis according to surgery type.
It is, therefore, the aim of this meta-analysis to define and quantify the safety and efficacy of HTK to other cardioplegias during adult cardiac surgery.

| METHODS
This is a meta-analysis. A meta-analysis involves the pooling of data from already published trials. As such, therefore, no ethical approval and no informed consent is required.
As per many of our previous surgical data analyses, 10,11 this systematic review and meta-analysis was performed in accordance with the PRISMA (preferred reporting items for systematic reviews and meta-analyses) guidelines (see Supporting Information files).

| Search strategy
To identify potential studies, systematic searches were carried out using PubMed, Science Direct, and Google Scholar. The search was supplemented by scanning the reference lists of eligible studies. The search strategy included the key concepts of "Custodiol versus Blood," "HTK solution versus blood," "Bretschneider's solution versus blood," "Custodiol versus crystalloid," "HTK versus crystalloid," "Bretschneider's solution versus crystalloid," AND "Cardiac Surgery" AND "Cardioplegia." (see Supporting Information file for the search strategy). All identified papers were assessed independently by two reviewers. A third reviewer was consulted to resolve disputes. Searches of published papers were conducted up until October 1, 2020. Data were extracted using predesigned extraction tables.

| Types of trials to be included and excluded
Included trials were those that directly compared the use of HTK versus any other type of MDC in open-heart surgery. There were no language restrictions. Animal studies and review papers were excluded. Studies that did not have any of the desired outcome measures were excluded.
Incomplete data, or data from an already included study, were excluded.
Other treatment modalities and interventions for coronary artery disease such as percutaneous coronary intervention and valvular disease such as transcatheter aortic valve intervention were excluded.

| Participants/population
This meta-analysis analyzed RCTs of both male and female adult (≥8 years) patients with either coronary artery disease or valvular disease who were undergoing cardiac surgery using either HTK or other types of MDC.

| Intervention(s) and exposure(s)
This meta-analysis considered all RCTs where HTK or another type of MDC were used in either coronary artery bypass graft (CABG) or in patients being treated for valvular disease. This specifically included all RCTs where the intention was to carry out cardiac surgery using HTK cardioplegia.

| Search results
Our initial search found 361 articles of which 330 were excluded on the basis of title and abstract and a further 14 duplicate studies were excluded. A further six studies were excluded as they were investigating REYNOLDS ET AL. | 1335 pediatric cases; one study was excluded as it had a mixed population; and, three studies were excluded because they did not use HTK (- Figure S1). Seven studies were included in our analysis. 12-18

| Risk of bias (quality) assessment
The risk of bias was assessed using a modified Jadad scale for randomized controlled trials. 19 Publication bias was investigated using funnel plots. 20

| Aortic cross-clamp time
Seven studies involving 668 patients measured the aortic crossclamp duration ( Figure 2

| Spontaneous defibrillation (spontaneous return of heartbeat)
Two studies involving 286 patients were measured for the incidence of spontaneous defibrillation ( Figure 3).

| Mortality
Three studies involving 285 patients reported the incidence of mortality ( Figure 5). The OR was 1.237, with 95% CI of 0.385, 3.978; p = .721; I 2 = 0% and p = .605. There was, therefore, no significant difference in the incidence of mortality between HTK and MDC. and p = .19. There was, therefore, no significant difference in the plasma CK-MB release between HTK and MDC in the valvular population.
F I G U R E 1 CPB time. On the left is a table displaying the mean difference, lower limit, upper limit, z value, and p value for each of the pooled studies with the final line presenting the overall statistics. On the right is a forest plot, which displays the mean difference (squares) and 95% CIs (whiskers) for each of the individual studies. Also shown is the overall mean difference (diamond) where the width of the diamond represents the 95% CIs. For the sake of clarity, the overall results with heterogeneity scores have been written below the forest plot. CABG, coronary artery bypass graft; CPB, cardiopulmonary bypass; CI, confidence interval; HTK, histidine-tryptophan-ketoglutarate cardioplegia; MD, multidose cardioplegia F I G U R E 2 Cross-clamp time. All details as in Figure 1 F I G U R E 3 Spontaneous return of heartbeat. All details as in Figure 1, except that the effect size shown is the odds ratio REYNOLDS ET AL.
| 1339 no significant difference in the serum TnI release between HTK and MDC.

| Study quality
The modified Jadad scale was used to assess study quality.

| DISCUSSION
Cardioplegia has evolved since the inception of cardiac surgery itself.
The current state of the industry sees several cardioplegia products presented to surgeons; with data often focussed only on areas of F I G U R E 4 Inotropic support. All details as in Figure 3 F I G U R E 5 Mortality. All details as in Figure 3 F I G U R E 6 Incidence of atrial fibrillation. All details as in Figure 3 clear statistical advantage, with the present data of HTK in cardiac transplantation exemplifying this. 5 The lack of data presented by industry on the performance of HTK in the broader cardiothoracic arena, however, suggests that a review of the current literature is required to establish a fair consensus on the value of HTK in all cardiac surgery, with particular emphasis in this paper on the treatment of acquired pathology in adults. By producing this metaanalysis, we met our objective to determine whether superiority exists in HTK compared with other marketed cardioplegias in the adult population, with pooled data for several outcomes in both perioperative and postoperative phases, clinical and basic sciences; we have produced several parameters by which we can contribute to the discussion of HTK's role in cardiac surgery.
While major adverse cardiac and cerebrovascular events were rare (with stroke being a never event in our population), mortality was reported in both investigation groups. The statistical insignificance between cardioplegia groups for mortality suggest the cause of death in each case was unrelated to the chosen cardioplegia, with authors attributing preoperative ventricular impairment, 15 postoperative low cardiac output syndrome, 16 graft thrombosis, 14 and subsequent continued infarction 13 as the cause of death in their respective populations. Previous research suggests that neither cardioplegia type nor technique are significantly influential on the incidence of myocardial ischemia following surgery, 23 and this would, therefore, support our population authors' death attributions and the statistical findings of our pooled data.
The incidence of spontaneous defibrillation was shown to significantly favor the HTK group in our population. This analysis was derived from a moderate population size, and may suggest that HTK has greater protective qualities to the myocardium during aortic cross-clamp phase.
The incidence of postoperative administration of inotropes and postoperative AF also suggests the protective effect of cardioplegia on myocardial physiology and cardiomyocyte biochemistry. Our findings found insignificant differences between HTK and MDC for both parameters, suggesting that adverse events are due to other factors (such as perioperative myocardial handling, or cardioplegia technique) 24 rather than cardioplegia type. This is supported by our pooling of biochemical data, which found insignificant differences in TnI and CK-MB, suggesting that, while both were observed to be elevated in each study following surgery, there was no significantly greater concentration in either the HTK or multidose populations.
The combination of pooled clinical and biochemical findings shows that, while myocardial physiology and biochemistry are disrupted perioperatively, these cases are sporadic and cannot be attributed to cardioplegia type in our population.
Duration of perioperative phases of surgery has been widely demonstrated to influence convalescence following surgery, for F I G U R E 7 Creatine kinase myocardial band. All details as in Figure 1, except that the effect size shown is the standardized mean difference F I G U R E 8 Troponin I. All details as in Figure 1 REYNOLDS ET AL.
example, the incidence of stroke. 25 CPB duration, during which patient blood is exposed to foreign surfaces via the CPB machinery, has correlated with several coagulopathic and pro-inflammatory responses within the body. 26  We originally hypothesized that statistically significant differences in our clinical endpoints may be found by stratifying our data according to aortic cross-clamp duration, that is, ischemic time of >90 min versus ischemic time of <90 min. However, screening of participating literature revealed, at present, only one RCT 18 (n = 107) with mean cross-clamp duration of >90 min, which alone would render statistically meaningless data. Surgical data analysis relies on regular publication from individual centers; greater clarity will be derived on our cross-clamp duration hypothesis following the publication of further RCT data.
In a recent meta-analysis, Gambardella et al. 7 reported that del Nido cardioplegia was significantly favored in several of the outcomes investigated including spontaneous defibrillation. In contrast, our meta-analysis was solely focused on HTK cardioplegia in comparison to MDC and has shown that spontaneous defibrillation is higher in the HTK group. This and other limitations in Gambardella et al., 7 as described in Section 1, suggest that it is too soon to be referring to superiority.

| Limitations
There was insufficient data to stratify clinical parameters according to cross-clamp duration time. It was hypothesized that statistical significance may exist in our clinical endpoints, should the data be stratified according to studies with a mean cross-clamp duration >90 min. However, only one study in our population met this criterion, and therefore, a meaningful analysis could not be performed.

| Conclusions
HTK and MDC are safe and effective in the myocardial protection of adult patients undergoing cardiac surgery. HTK was shown to be as effective as multidose in most clinical parameters measured.
Spontaneous defibrillation occurred significantly more often in the HTK group following aortic cross-clamp removal. More data measuring this parameter are needed to determine the external validity of this finding.
Single-dose administration of HTK is, therefore, the only significant difference in our population. Surgeon preference, anticipated procedure duration, and case complexity should therefore determine the selection of HTK when planning procedures.
More data are currently needed for procedures with a crossclamp duration >90 min, to ascertain whether statistical significance exists between HTK and MDC in procedures over 90 min.