Myocardial oxygen consumption during histidine-tryptophan-ketoglutarate cardioplegia in young human hearts

Abstract

OBJECTIVES

Energy demand and supply need to be balanced to preserve myocardial function during paediatric cardiac surgery. After a latent aerobic period, cardiac cells try to maintain energy production by anaerobic metabolism and by extracting oxygen from the given cardioplegic solution. Myocardial oxygen consumption (MVO₂) changes gradually during the administration of cardioplegia.

METHODS

MVO₂ was measured during cardioplegic perfusion in patients younger than 6 months of age (group N: neonates; group I: infants), with a body weight less than 10 kg. Histidine-tryptophan-ketoglutarate crystalloid solution was used for myocardial protection and was administered during a 5-min interval. To measure pO₂ values during cardioplegic arrest, a sample of the cardioplegic fluid was taken from the inflow line before infusion. Three fluid samples were taken from the coronary venous effluent 1, 3 and 5 min after the onset of cardioplegia administration. MVO₂ was calculated using the Fick principle.

RESULTS

The mean age of group N was 0.2 ± 0.09 versus 4.5 ± 1.1 months in group I. The mean weight was 3.1 ± 0.2 versus 5.7 ± 1.6 kg, respectively. MVO₂ decreased similarly in both groups (min 1: 0.16 ± 0.07 vs 0.36 ± 0.1 ml/min; min 3: 0.08 ± 0.04 vs 0.17 ± 0.09 ml/min; min 5: 0.05 ± 0.04 vs 0.07 ± 0.05 ml/min).

CONCLUSIONS

We studied MVO₂ alterations after aortic cross-clamping and during delivery of cardioplegia in neonates and infants undergoing cardiac surgery. Extended cardioplegic perfusion significantly reduces energy turnover in hearts because the balance procedures are both volume- and above all time-dependent. A reduction in MVO₂ indicates the necessity of a prolonged cardioplegic perfusion time to achieve optimized myocardial protection.

Under aerobic conditions, cardiomyocytes obtain their energetic substrate as adenosine triphosphate (ATP), mainly from mitochondrial oxidative phosphorylation due to nicotinamide adenine dinucleotide and flavin adenine dinucleotide dehydration.

INTRODUCTION

Under aerobic conditions, cardiomyocytes obtain their energetic substrate as adenosine triphosphate (ATP), mainly from mitochondrial oxidative phosphorylation due to nicotinamide adenine dinucleotide and flavin adenine dinucleotide dehydration. However, in anaerobic conditions, oxygen requirements exceed oxygen delivery; hence ATP production is insufficient to cover the energy demand of the heart. Only 2 mol ATP/mol glucose are produced anaerobically, instead of 36 mol during aerobiosis. Thus, with progressive ischaemia, an energy deficit occurs that becomes more severe the longer the ischaemia lasts [1].

During open-heart surgery, ischaemic conditions occur directly after aortic cross-clamping, and myocardial cells switch from aerobic to anaerobic metabolism [2–5]. Administration of cardioplegic solution has made it possible to prolong the ischaemic period without irreversible damage of the myocardium. Requirements for optimized cardioplegia include both reduction of the myocardial turnover and decrease of the lactate acidosis in combination with optimal organ perfusion. Previous clinical studies focused mostly on pH and lactate values [6, 7] or enzyme release [8, 9] to evaluate the efficiency of the applied cardioplegic solution and to analyse tissue injury, whereas human myocardial oxygen consumption (MVO₂) during cardioplegic perfusion in adults has only been reported once [10].

Because the ideal amount of time needed to arrest the heart using cardioplegic perfusion is not clinically established and data obtained from experimental studies cannot be transferred directly to humans, we evaluated myocardial behaviour during cardioplegic perfusion in children younger than 6 months by measuring MVO₂ intraoperatively. Gradual changes in MVO₂ during the administration of cardioplegia were analysed accordingly.

MATERIALS AND METHODS

From December 2013 to June 2014, we prospectively evaluated 20 patients with a body weight <10 kg and aged <6 months who underwent cardiac surgery at our institution. The study was conducted with local institutional review board approval and with the consent of the patients’ families.

All procedures were performed under cardiopulmonary bypass (CPB) and cardioplegic arrest. Histidine-tryptophan-ketoglutarate (CUSTODIOL^® HTK Solution) solution was applied once in all patients at a dosage of 50 ml/kg body weight. CPB was established in all patients using bicaval and aortic cannulation. Once CPB started, the aorta was crossed-clamped and the cardioplegic solution was constantly delivered antegradely at 4°C, using a separate roller pump (MPS^®2 smart myocardial protection Cardioplegia Delivery System, Quest Medical Inc., Allen, TX, USA), maintaining a flow rate of 50 ml per min for 5 min at a perfusion pressure of 65–70 mmHg, via the ascending aorta. The right atrium was routinely opened, and the cardioplegic coronary venous effluent ran out into the pericardial sack. In all patients, the aortic valve was competent, and the flow rate of the cardioplegic solution was reliable.

Continuous blood ultrafiltration was performed throughout CPB.

Immediately before cardioplegia delivery, 1 fluid sample was taken from the inflow line (the arterial sample) as the baseline value. Subsequently, 3 fluid samples were taken directly from the coronary sinus representing the atrial effluent at the first, third and fifth min of cardioplegic perfusion. Samples were stored under sterile conditions at −4°C and promptly analysed with an arterial gas analyser (GEM 4000^®, Instrumentation Laboratory, Bedford, MA, USA) in the operating room. The same surgeon performed storage and sample analysis as well as processing, thus reducing potential operator-related variability. As is done routinely in our laboratory, the samples were stored in a dedicated refrigerator at −4°C. The samples were transferred from the operating room to the laboratory in special thermic boxes fixed at −4°C.

Myocardial oxygen consumption was calculated according the Fick principle: MVO₂ = CBF × (CaO₂ − CvO₂), whereby CBF represents the coronary blood flow and CaO₂ − CvO₂, the arterial-venous oxygen content difference (ml O₂/ml blood). To measure MVO₂ during cardioplegic perfusion, the formula was adapted as follows: MVO₂ = CPR × (PaO₂ − Pcs_xO₂) × α.

CPR represents the cardioplegic perfusion rate; PaO₂ − Pcs_xO₂ stands for differences in oxygen partial pressures between the base value (PaO₂ ‘arterial sample’) and the coronary venous pO₂ (Pcs_xO₂) taken at defined intervals (first, third and fifth min of cardioplegic perfusion); α represents the oxygen solubility coefficient of water at 4°C (α at 4°C is 0.026).

High-sensitivity troponin (T-hs) was chosen as a specific marker for ischaemic myocardial injury because it does not present cross-reactivity with muscular troponin and presents an independent metabolism from possible coexisting renal dysfunction. A baseline T-hs level was measured preoperatively for each patient. Blood samples for measuring T-hs were taken 30 min after coronary reperfusion and subsequently at 6, 12, 24 and 48 h postoperatively.

Statistical analyses

Statistical analyses were performed with IBM SPSS Statistics for Windows, Version 25.0 (released 2017) (IBM Corp., Armonk, NY, USA). Descriptive statistics were expressed as mean value and standard deviation, and considered variables were analysed using the Student’s t-test. The χ² test was used when the expected values in any of the cells of a contingency table were above 5; the Fisher exact test was used when the expected values were below 5. Univariate analysis was performed by the analysis of variance test when the t-test or the χ² results were statistically significant. A P-value <0.050 was considered significant.

RESULTS

Twenty patients were included in the study and separated into 2 groups: neonates (N) less than 1 month of age (n = 7) and infants (I) aged <6 months (range 1.1–6 months) (n = 13). All patients were born at term. The indication for surgery in group N was transposition of the great arteries [7/20 (35%)], whereas infants in group I had complete atrioventricular septal defect [1/20 (5%)], tetralogy of Fallot (TOF) [6/20 (30%)], atrial septal defect associated with pulmonary stenosis [1/20 (5%)] and ventricular septal defect (5/20, 25%). Mean age at surgery was 0.2 ± 0.09 vs 4.5 ± 1.1 months (P = 0.01) at a mean body weight of 3.1 ± 0.2 vs 5.8 ± 1.6 kg (P = 0.02) in neonates and infants, respectively (Tables 1 and 2).

Table 1:

Patient characteristics

Variables	Neonates	Infants	P-value
Gender male/female	5/2	6/7	NS
Weight (kg)	3.1 ± 0.2	5.8 ± 1.6	0.02
CPB time (min)	156 ± 34	108 ± 39	0.79
CC time (min)	107 ± 25	77 ± 37	0.25
Temperature (°C)	31.2 ± 0.8	32.5 ± 0.8	0.94

CC: cross-clamping; CPB: cardiopulmonary bypass.

Table 2:

Indications for surgery

Diagnosis	Neonates	Infants
TGA	7/7 (100%)	None
TOF	None	6/20 (30%)
ASD with pulmonary stenosis	None	1/20 (5%)
CAVSD	None	1/20 (5%)
VSD	None	5/20 (25%)

ASD: atrial septal defect; CAVSD: complete atrioventricular septal defect; TGA: transposition of the great arteries; TOF: tetralogy of Fallot; VSD: ventricular septal defect.

All patients underwent complete surgical repair. In group I, 5 patients had a previous percutaneous atrial balloon septostomy. No other preliminary surgical palliative interventions were performed among the remaining 15 patients. CPB principally ran under moderate hypothermia, mean temperature 31.2 ± 0.8°C (N) versus 32.5 ± 0.8°C (I) (P = 0.94), duration of CPB 156 ± 34 (N) versus 108 ± 39 min (I) (P = 0.79). Cardioplegic arrest took 107 ± 25 min (N) versus 77 ± 37 min (I) (P = 0.25). Mean ultrafiltered blood volume at the end of the surgery was 298.6 ± 82.7 ml vs 365 ± 161.2 ml (P = 0.35) in neonates and infants, respectively. The mean haematocrit value before the operation was significantly different in neonates (36 ± 5.4%) compared to that in infants (30 ± 4%) (P = 0.02), whereas after surgery no significant difference was noted [36 ± 6.8% (N) vs 34.2 ± 3.2% (I) (P = 0.23)]. Neither postoperative complications nor deaths were observed in the entire cohort.

Baseline PaO₂ values were comparable in both groups [224 ± 22.4 mmHg (N) vs 227.3 ± 18.3 mmHg (I); P = 0.69]. PcsO₂ levels during cardioplegic perfusion increased equivalently in both groups: 101.6 ± 24.6 mmHg (N) versus 111.9 ± 28.4 mmHg (I) (P = 0.36) after 1 min of cardioplegia administration and 151.7 ± 30.2 mmHg (N) versus 170.4 ± 29.1 mmHg (I) (P = 0.73) and 189.3 ± 19.2 mmHg (N) versus 192.5 ± 16.2 mmHg (I) (P = 0.97) after 3 and 5 min, respectively (Fig. 1).

Figure 1:

PO₂ trend analysis.

Myocardial oxygen consumption at the first, third and fifth min of cardioplegic perfusion was significantly different at the first min evaluation: 0.16 ± 0.07 ml/min (N) 1 vs 0.36 ± 0.1 ml/min (I) (P = 0.02). This difference was still significant at the third min [0.08 ± 0.04 ml/min (N) vs 0.17 ± 0.09 ml/min (I); P = 0.04], whereas after 5 min no significant difference was noticed [0.05 ± 0.04 ml/min (N) vs 0.07 ± 0.05 ml/min (I); P = 0.29] (Fig. 2).

Figure 2:

Myocardial oxygen consumption trends.

Troponin level trends, difference between neonates and infants

As a specific marker of myocardial cytolysis, the T-hs level was measured perioperatively. Trend analysis of T-hs showed an initial dramatic increase from the baseline value, followed by a gradual decrease in the next 48 h after cardiac surgery. In detail, a significant difference was observed between preoperative T-hs levels in neonates compared to those in infants (82.5 ± 48.6 vs 30.7 ± 15.3; P = 0.03).

Evaluation of postoperative T-hs values showed a significant reduction in neonates compared to infants at 30 min after declamping the aorta (P = 0.04) and at 6 h (P = 0.04) after the operation. Moreover, at 12 (P = 0.15), at 24 (P = 0.24) and at 48 h (P = 0.18), there was no difference between neonates and infants regarding mean troponin values (Fig. 3).

Figure 3:

Troponin trend analysis.

Intraoperative muscular resection or handling was part of the surgical procedure in 13 cases (mainly TOF repair). Compared to those patients, for whom muscular resection was not performed (such as for atrial septal defect and ventricular septal defect closure), no significant difference was shown at any of the time points already considered (Table 2).

DISCUSSION

In cardiac surgery, most of the operations are performed under CPB, and clamping of the aorta is standard procedure. By this means, the energy deficit is reinforced by ongoing ischaemia unless myocardial cells extract oxygen from the cardioplegic solution. Thus, the goal of guaranteeing basal aerobic metabolism can be established during administration of cardioplegia.

We compared the changes in MVO₂ consumption in neonates and infants undergoing cardiac surgery. Although the results showed comparable values with regards to oxygen extraction, under cardioplegic arrest, the amount of oxygen consumed by the 2 subgroups differed significantly, with the neonatal hearts consuming less oxygen than the infant hearts. In addition, early after aortic clamp removal, levels of plasmatic hs-troponin [11, 12], as markers of myocardial damage, were higher in infants than in neonates. A reduced susceptibility of neonatal hearts to the ischaemic insult might be the reason.

The goal of this study was to provide further information about myocardial behaviour in children undergoing cardiac surgery and cardioplegic arrest by measuring oxygen consumption (MVO₂) intraoperatively in a direct, non-invasive fashion. Our data support the data in the currently available literature regarding myocardial maturation in the early stages of life and may have clinical implications in assessing the best myocardial protection in young children undergoing cardiac surgery.

Measurements of MVO₂ during cardiac arrest under experimental conditions have already been published [6, 13, 14]. Previous attempts to measure MVO₂ intraoperative as an index of myocardial metabolic activity were performed primarily in animal models with the goal of evaluating normal adult myocardial behaviour under cardioplegic arrest and hypothermia [5, 14–16]. Matherne et al. [17] analysed mature and immature myocardium and described age-related differences in intracellular metabolism. Mahle et al. [18] elaborated on an extensive inflammatory response to cardiac surgery early in life and saw a correlation between high levels of inflammatory markers and the occurrence of the low-output syndrome. Certainly, sequential stages of human heart maturation, either healthy or congenitally diseased, are influenced differently by the cardioplegic method applied [5, 17–20].

In 1996, Jessen et al. [5] first measured MVO₂ in explanted and blood-perfused neonatal hearts from piglets, showing that under ventricular fibrillation the myocardium consumes more oxygen than does the empty beating heart. The authors concluded that the addition of an arresting concentration of KCl did not lower oxygen consumption below the level observed under pure hypothermic conditions. Based on their findings, the authors hypothesized that MVO₂ measurements in neonates might have some intrinsic differences compared to those observed in studies focusing on adult hearts.

The concept that certain cardiac events begin at birth and change the characteristics of the neonatal heart into those of a mature heart was presented in early studies from Romero et al. [21] in 1972, but the time frame of many of these events is still unknown [22]. Generally, normal neonatal and infant hearts are both believed to be more resistant to ischaemic and reperfusion damage than the normal adult heart [2, 6, 17, 23–28]. The immature heart of a neonate has a denser structure than that of the adult heart and is generally believed to be more resistant to calcium damage than are myocytes from mature hearts. Biochemical and physiological mechanisms, recognized as factors associated with improved tolerance to ischaemia, include greater glycogen stores and improved anaerobic metabolism, better maintenance of ischaemic calcium exchange, higher levels of ATP substrates and increased utilization of amino acid substrates [2, 17, 24].

Based on our preliminary findings, it is reasonable to consider that the architectural immaturity of the myocardium reflects its reduced oxygen need, thus its reduced oxygen utilization. It has been previously demonstrated that, within the first month of life, myocardiocytes present a degree of maturation comparable to that of immature myocardiocytes [2, 5]. Our data support the thesis that functionally the immature heart is not yet fully developed. A higher level of oxygen consumption and consequently greater susceptibility to the ischaemic insult under cardioplegic arrest are more evident in infants than in neonates.

Limitations

Our study cohort was small (20 patients) but served the purpose of gathering preliminary results to encourage further investigation. Two discrete subgroups of congenital heart diseases (transposition of the great vessels belonging to the neonatal group and TOF belonging to the infant group) could represent a confounding variable, although cyanosis and hypertrophy did not differ significantly in our cohort. In addition, cardiac mass calculation, as it is routinely used in experimental settings, was not possible retrospectively. Finally, the extent and quality of the available tissue samples (atrial vs ventricular tissue) also depended on the type of surgery performed. For instance, in TOF repair (infant group), it was possible to gather a much more representative ventricular component because the myectomy was normally part of the surgical procedure. Moreover, the impact of preoperative cyanosis on the myocardial protection strategy used during cardiac surgery is still controversial and was not addressed specifically in this study [29, 30].

A larger prospective study population is mandatory to achieve supplementary results, to confirm or refute our hypothesis and to support future investigation in this field.

CONCLUSIONS

During cardiac surgery procedures, although neonatal and infant hearts are equally capable of extracting oxygen from the cardioplegic solution, intraoperatively measured MVO₂ levels are lower in neonates than in infants, suggesting their reduced need of oxygen to maintain their intracellular metabolism.

ABBREVIATIONS

ATP

Adenosine triphosphate

CBF

Coronary blood flow

CPB

Cardiopulmonary bypass

CPR

Cardioplegic perfusion rate

MVO₂

Myocardial oxygen consumption

T-hs

High-sensitivity troponin

TOF

Tetralogy of Fallot