Mol Cell Biochem DOI 10.1007/s11010-014-2122-3
Metabolic derangement and cardiac injury early after reperfusion following intermittent cross-clamp fibrillation in patients undergoing coronary artery bypass graft surgery using conventional or miniaturized cardiopulmonary bypass Bao A. V. Nguyen • M-Saadeh Suleiman • Jonathan R. Anderson • Paul C. Evans • Francesca Fiorentino • Barnaby C. Reeves • Gianni D. Angelini
Received: 25 March 2014 / Accepted: 2 June 2014 Ó Springer Science+Business Media New York 2014
Abstract Myocardial ischemic stress and early reperfusion injury in patients undergoing coronary artery bypass grafting (CABG) operated on using intermittent crossclamp fibrillation (ICCF) are not presently known. The role of mini-cardiopulmonary bypass (mCPB) versus conventional CPB (cCPB) during ICCF has not been investigated. These issues have been addressed as secondary objective of randomised controlled trial (ISRCTN30610605) comparing cCPB and mCPB. Twenty-six patients undergoing primary elective CABG using ICCF were randomised to either cCPB or mCPB. Paired left ventricular biopsies collected from 21 patients at the beginning and at the end of CPB were used to measure intracellular substrates (ATP and related compounds). Cardiac troponin T (cTnT) and CKMB levels were measured in plasma collected from all patients preoperatively and after 1, 30, 60, 120, and 300 min after institution of CPB. ICCF was associated with significant ischemic stress as seen by fall in energy-rich phosphates early after reperfusion. There was also a fall in nicotinamide adenine dinucleotide (NAD?) indicating cardiomyocyte death which was confirmed by early release of cTnT and CK-MB during CPB. Ischemic stress and early myocardial injury were similar for cCPB and mCPB. B. A. V. Nguyen J. R. Anderson P. C. Evans F. Fiorentino B. C. Reeves G. D. Angelini Department of Cardiothoracic Surgery, NHLI, Hammersmith Hospital, Imperial College, London, UK M.-S. Suleiman (&) Bristol Heart Institute, University of Bristol, Bristol, UK e-mail: [email protected] Present Address: P. C. Evans Department of Cardiovascular Science, University of Sheffield, Sheffield, UK
However, the overall cardiac injury was significantly lower in the mCPB group as measured by cTnT (mean ± SEM: 96 ± 14 vs. 59 ± 8 lg/l, p = 0.02), but not with CK-MB. ICCF is associated with significant metabolic derangement and early myocardial injury. This early outcome was not affected by the CPB technique. However, the overall cardiac injury was lower for mCPB only when measured using cTnT. Keywords Cardiac injury Ischemia ATP Cardiopulmonary bypass Cross-clamp fibrillation Cardiac surgery Abbreviations ICCF Intermittent cross-clamp fibrillation CABG Coronary artery bypass grafting cCPB Conventional cardiopulmonary bypass mCPB Mini-cardiopulmonary bypass cTnT Cardiac troponin T CK-MB Creatine kinase
Introduction Induced ventricular fibrillation in conjunction with intermittent aortic cross-clamping to arrest the heart in patients undergoing open heart surgery for coronary revascularization was widely used more than 35 years ago [1]. It is estimated that this technique is routinely used by 10–15 % of surgeons in the UK [2]. Proponents of this intervention advocate its use on the assumption that the heart can tolerate repeated short periods of aortic cross-clamping as the metabolic stress due to global myocardial ischemia can be
123
Mol Cell Biochem
addressed by a short period of reperfusion. Cardiac centres that continue to use this technique in preference to cardioplegic arrest are also encouraged by reports showing that the cardioprotective efficacy is similar to cardioplegic arrest [3]. The potential advantages of using intermittent cross-clamp fibrillation (ICCF) over cardioplegic arrest include shorter ischemic time and possibly myocardial protection from ‘‘ischemic preconditioning’’ as has been demonstrated experimentally [3, 4]. However, a potential disadvantage of this technique is the risk of thromboembolism and repeated aortic clamping. Myocardial biopsies from patients undergoing coronary artery bypass graft (CABG) surgery using cold or warm blood cardioplegia have been used to demonstrate metabolic derangement [5]. However, myocardial biopsies following ICCF have been used only to monitor the deterioration in structural integrity but not in cellular metabolites [6]. More recently, work involving measurement of myocardial pH showed that using ICCF does not result in reliable reperfusion of the myocardium [7]. Consequently, a direct measurement of cardiac cellular metabolites associated with ICCF and reperfusion is important to determine the extent of the ischemic insult and reperfusion injury. Therefore, we decided to measure myocardial metabolites and early reperfusion injury associated with ICCF. The work was done as a secondary objective in an ongoing randomised controlled trial comparing conventional (cCPB) and miniaturized cardiopulmonary bypass (mCPB) in patients undergoing CABG. Earlier studies have already shown improved myocardial protection using mCPB using cardioplegic arrest [8–10]. However, there are no reports comparing cardiac injury (or ischemic stress) between the two CPB techniques using ICCF. Therefore, in addition to our main aim to assess metabolic derangement and early reperfusion injury during arrest using ICCF, we also compared myocardial metabolic changes and injury using the two CPB techniques.
Materials and methods The study was sub-study of a larger trial approved by the hospital ethics committee (registration number for the trial is ISRCTN30610605) and informed consent was obtained from all participants who were undergoing primary elective CABG using ICCF at the Hammersmith Hospital. Patients under 18 years of age were excluded as well as those who had any of the following: ejection fraction \30 % on echocardiogram; recent cerebro-vascular accident within 3 months preoperatively or more than 75 % carotid artery obstruction on ultrasound; pre-existing renal impairment with serum creatinine in excess of 177 lmol/L; pre-existing coagulopathy; pre-existing liver dysfunction or the
123
Table 1 Characteristics of mCPB and cCPB cCPB
mCPB
Target activated clotting time (ACT)
[400 s
[400 s
Venous cannula
34 Fr 2 stage (Medtronic)
29Fr Optiflow (Sorin) 3/8 inch tubing
1/2 inch tubing Reservoir
Sorin Evo
Soft shell reservoir
Pump
Roller pump (Stockert, Germany)
Centrifugal pump (Stockert, Germany)
Heat exchanger/ oxygenator
Avant (Sorin)
Eos (Sorin)
Flows (target cardiac index)
1.8–2.4 L/min/ m2 cooled to 32 °C
1.8–2.4L/min/m2 cooled to 32 °C
Prime volume
1,400 mls Hartmann’s
300 mls Hartmann’s with retrograde autologous prime
Safety features Operative field blood
Venous air removal device (VARD) Cardiotomy suction
Cell salvage
recent (within 5 days) use of antiplatelet agents (aspirin/ clopidogrel). Patients selected for this sub-study (n = 26) were allocated to the same surgeon and were randomized to either cCPB (n = 13) or mCPB (n = 13). The number of patients recruited for this study was based on our extensive work (e.g. [11]) collecting sequential ventricular biopsies from the same patient and showing differences in metabolites during surgery.
Anaesthetic and surgical techniques Anaesthetic technique was standardized for all patients. Thiopentone (1–3 mg/kg) was used for induction with 3–5 mg/kg fentanyl, and volatile agents were delivered in 50 % air–O2 mixture for maintenance. Propofol (3 mg/kg/h) was given as an infusion during cardiopulmonary bypass and neuromuscular blockade was achieved by 0.1–0.15 mg/kg Pancuronium bromide. Alpha stat acid–base management was adopted. Initial anticoagulation was accomplished with 3 mg/kg body weight of heparin and was supplemented as required in order to maintain an active clotting time of 480 s or above. All operations were performed using cCPB or mCPB as previously described [12] with ascending aortic cannulation, two stage venous cannulation and moderate systemic hypothermia (32 °C). The Sorin Dideco Extracorporeal Circulation Optimized (ECCO) system was used as the mCPB circuit. This ‘‘Hammersmith’’ miniaturized CPB
Mol Cell Biochem
system which has been described in details [12] is composed of 29-French OptiFlow venous cannula, venous air removal device, centrifugal pump, heat exchanger, oxygenator module, arterial line filter and parallel soft-shell reservoir. The conventional CPB (cCPB) circuit utilised a Stockert roller pump, membrane oxygenator, venous reservoir and non-heparin bonded circuit with cardiotomy suction. Key characteristics of both mCPB and cCPB are shown in Table 1. ICCF was introduced after cross-clamping of the aorta in order to perform coronary grafting. This was followed by de-clamping the aorta and reperfusion of the heart. The grafts were done sequentially for distal anastomoses; and with the clamp off and side biting clamp on for proximals. Left internal mammary artery to left anterior descending coronary artery (LIMA-LAD) was constructed last. The approximate time frame for this was as follows: 10 min for last vein top-end; 10 min for LIMA-LAD; 1–10 min to check anastomoses, undertake haemostasis, revise; and finally 2–5 min for biopsy collection and repair of biopsy site. CPB was then discontinued, and post-operative management was standard for all participants in accordance with institution protocols. Collection of ventricular biopsy Myocardial tissue biopsy specimens were sampled from the ´ 11.4 cm Cannula apex of the left ventricle using a 14 Ga. TW Trucut needle (Baxter Healthcare Corporation, Deerfield, IL 60015 USA). Two biopsies were taken, the first biopsy just after the institution of CPB and the second immediately prior to the cessation of CPB. Each specimen was immediately snap frozen (less than 5 s) in liquid nitrogen and stored at -80 °C until processing for cardiac cellular metabolites. All the biochemical analyses were performed by an investigator blinded to the interventions used. Extraction and determination of cellular metabolites The procedure used to extract cardiac metabolites was similar to that described previously [13]. In brief, frozen biopsies were crushed in liquid nitrogen with a pestle and mortar and transferred to a tube containing 500 ll 4.8 % perchloric acid, weighed and centrifuged at 3,6459g for 10 min at 4 °C. The supernatant was then taken and neutralised with equal volume of 0.44 M K2CO3 and the solution was centrifuged for 5 min at 4 °C to remove the insoluble potassium perchlorate. The supernatant was taken and stored at -20 °C until analysed for cellular metabolites using high performance liquid chromatography (HPLC) (ATP related compounds) or an enzymatic kit (Lactate).
HPLC was used to measure cellular metabolites [14] using a Beckman System and a 3-lm octadecyl silane Hypersil column (150 9 4.6 mm2) (Thermo Scientific, UK). Eluent A contained 150 mM KH2PO4 and 150 mM KCl, set at pH 6.0 with KOH. Eluent B consisted of eluent A with 15 % (v/v) HPLC grade acetonitrile. A low-pressure gradient mixing device was used to control the composition of the mobile phase. The amount of eluent B changed linearly between the time points. The analytical column was maintained at a constant temperature in the range of 17–19 °C and absorption was measured at 254 nm. The metabolites measured using HPLC were ATP, ADP, AMP, GTP, GDP, GMP, IMP, xanthine, hypoxanthine, nicotinamide adenine dinucleotide (NAD?), inosine and adenosine. The levels of NAD? in cardiac biopsies have been used as a measure of the mitochondrial permeability transition pore opening and, therefore, cardiomyocyte death [15]. Early after reperfusion, calcium loading and oxidative stress trigger opening of the permeability transition pore which can lead to cardiomyocyte death [16]. Standards for HPLC were prepared as previously described [14]. 20 ll of the extracted sample was injected into the column. The area under the peak for each metabolite was determined and compared to the standard sample, then corrected for weight. Lactate was measured using a plasma lactate determination kit from Sigma Diagnostics (Sigma, Poole, UK). Markers of cardiac injury In this study, we also monitored markers of cardiac injury in the early phases (*1 and 30 min) after the commencement of CPB. Blood collections were made before surgery (on the morning of surgery and post intubation following anaesthesia) and immediately after starting CPB and then at 30, 60, 120 and 300 min after the initiation of CPB. In view of the known reports suggesting that different markers appear at different time points during reperfusion, we decided to measure two markers of injury; cardiac troponin T (cTnT) and Creatine Kinase (isoenzyme MB). cTnT is a cardio-specific and highly sensitive marker for myocardial damage. Although it is widely accepted that cTnT release peaks relatively late within the first day, a study using ICCF during CABG has shown that this biomarker peaks at 5–6 h post-operatively [17]. Immunoassay for the in vitro quantitative determination of both CK-MB and cTnT was used (Roche Diagnostics GmbH). Data collection and analysis Intention-to-treat statistical analyses were performed using Statview for Windows (SAS Institute Inc.). Data were expressed as the mean ± SE unless otherwise stated. Nonparametric tests for paired and unpaired groups were
123
Mol Cell Biochem
carried out using Mann–Whitney and Wilcoxon signed rank test, respectively. The myocardial tissue concentration of metabolites was used to approximate the intracellular concentration. Area under the curve to estimate total protein (cTnT and CK-MB) release during CPB was calculated using the trapezium rule (Microsoft Excel).
Results Data for clinical outcome and markers of injury were collected for all 26 patients. However, it was only possible to collect two paired myocardial biopsies from 21 patients (11 on mCPB). The rest of the patients had either one very small biopsies (n = 2) or at least one biopsy that was not collected for practical reasons (n = 3). Clinical outcome There were no post-operative deaths. No differences in preoperative characteristics or in clinical outcomes were observed between cCPB or mCPB groups, except the mCPB group had a higher proportion of non-diabetics (Tables 2, 3).
Table 3 Peri- and post-operative data Variable
mCPB (n = 13)
cCPB (n = 13)
p values
Cardiopulmonary bypass time (minutes)
72.5 (4.49)
75.07 (5.08)
ns
Total cross-clamp time (minutes)
29.54 (2.33)
32.84 (2.15)
ns
Number of coronary artery grafts
3.23 (0.18)
3.23 (0.26)
ns
Time on ITU (days)
1.00 (0.13)
1.26 (0.27)
ns
New post-operative neurological dysfunction
0
0
ns
New haemofiltration post-operatively
0
1 (8 %)
ns
Infective complications Harvest site infection
0
1 (8 %)
ns
Sternal wound infection
0
2 (15 %)
ns
Return to theatre
0
0
ns
Values are expressed as number or mean with SD in parenthesis or percentage as indicated
Changes in myocardial energy-rich phosphates associated with ICCF
Table 2 Preoperative characteristics Variable
mCPB (n = 13)
cCPB (n = 13)
p values
Male/female
8/5
10/3
ns
Age (years)
68.4 (1.97)
66.3 (2.24)
ns
Body surface area (m2)
1.87 (0.06)
1.98 (0.05)
ns
Body mass index (kg/ m2) Logistic euroSCORE (%)
29.74 (1.31)
29.31 (1.19)
ns
3.35 (0.68)
3.74 (0.91)
ns
Single vessel disease
0
1 (8 %)
ns
Two vessel disease
1 (8 %)
1 (8 %)
ns
Three vessel disease
12 (92 %)
11 (84 %)
ns
Good [50 %
11 (85 %)
12 (92 %)
ns
Fair 30–50 %
2 (15 %)
1 (8 %)
ns
Not diabetic
11 (85 %)
6 (46 %)
0.03*
Oral therapy Insulin
1 (8 %) 1 (8 %)
5 (38 %) 2 (15 %)
ns ns
Coronary vessel disease
Ejection fraction
Diabetes
Values are expressed as number or mean with SD in parenthesis or percentage as indicated * indicates statistical significance
123
There was a fall in the myocardial levels of the main energy-rich phosphates, the adenine nucleotides (ATP, ADP, AMP), in biopsies taken at the end of the CPB period compared to biopsies taken just after instituting CPB (Fig. 1a). This derangement was seen in patients operated on using cCPB or mCPB (Fig. 1b). In addition to the fall in adenine nucleotides, there was also a significant fall in guanine nucleotides GTP, GDP and GMP (Fig. 2a). Again, like adenine nucleotides, the fall and trends were similar for both cCPB and mCPB (Fig. 2b). The fall in energy-rich phosphates during anaerobic metabolism is normally associated with accumulation of ATP catabolites including hypoxanthine, inosine, xanthine as well as lactate which tend to efflux upon reperfusion [18]. Nevertheless, even at the end of CPB, which follows from the last period of ICCF (see ‘‘Materials and Methods’’ for details), there was a significant increase in hypoxanthine (p = 0.02, Wilcoxon Rank test) and a strong trend for inosine to increase (p = 0.05, Wilcoxon Rank test) associated with arrest and reperfusion (Fig. 3a). In contrast, lactate levels were similar to those for basal levels (0.027 ± 0.005 vs. 0.033 ± 0.005 nmol/mg wet weight and 0.026 ± 0.006 vs. 0.024 ± 0.004 nmol/mg wet weight for cCPB and mCPB, respectively).
Mol Cell Biochem
Fig. 1 Myocardial concentration of adenine nucleotides ATP, ADP & AMP in biopsies collected at the initiation of CPB and at the end of CPB from patients undergoing CABG using ICCF (a). A comparison between conventional or mini-CPB is shown in b. Data are presented as mean ± SE. *p \ 0.05 and **p \ 0.001 versus start of CPB for corresponding metabolite
Fig. 2 Myocardial concentration of guanine nucleotides GTP, GDP & GMP in biopsies collected at the initiation of CPB and at the end of CPB from patients undergoing CABG using ICCF (a). A comparison between conventional or mini-CPB is shown in b. Data are presented as mean ± SE. *p \ 0.05 and **p \ 0.001 versus start of CPB for corresponding metabolite
Opening of mitochondrial permeability transition pore
sessions of ischemia and reperfusion) irrespective of whether the biomarker was cTnT or CK-MB (Figs. 4a, 5a). Figures 4b and 5b show the entire profile of the release of cTnT and CK-MB using cCPB or mCPB. Whilst CK-MB levels were similar for both interventions (3.5 ± 0.3 vs. 3.3 ± 0.3 lg/ml, p = 0.51) total cTnT release was significantly lower in the mCPB group (95.9 ± 14.4 vs. 58.8 ± 7.6 lg/l, p = 0.02).
Figure 3b shows a significant fall in NAD? upon reperfusion and CPB, irrespective of type of CPB, indicating that triggering of cardiomyocyte death occurred soon after starting CPB and ICCF. Therefore, it was expected that levels of markers of myocardial injury would be elevated during ICCF & CPB. Early myocardial injury associated with ICCF
Discussion Consistent with opening of the mitochondrial permeability transition pore soon after reperfusion, we found significantly elevated levels of markers of myocardial injury in blood within 30 min after institution of CPB (this period is likely to be associated with at least 2 grafts and two
In this study, we show for the first time that the use of ICCF for CABG is associated with significant ischemic stress, metabolic derangement and evidence of opening of the mitochondrial permeability transition pore. In contrast
123
Mol Cell Biochem
Fig. 3 Myocardial concentration of ATP catabolites hypoxanthine, inosine and adenosine (a) and NAD? (b) in biopsies collected at the initiation of CPB and at the end of CPB and allocated to either conventional or mini-CPB. Data are presented as mean ± SE. *p \ 0.05 and **p \ 0.001 versus start of CPB in corresponding metabolite. #p = 0.05 versus start of CPB in corresponding metabolite
to other studies, we also measured markers of myocardial injury and observed significant increases in these markers during the period of ICCF and CPB. Both metabolic derangement and early reperfusion injury were found; these findings were largely independent of whether cCPB or mCPB had been used. However, total reperfusion injury as estimated by area under the curve for cTnT was significantly less in the mCPB group. Cardiac arrest during CABG with CPB is commonly performed using a depolarising hyperkalemic cardioplegic solution with different compositions. An alternative method to induce cardiac arrest utilises ICCF. In the UK, it is surprising that a relatively small 10–15 % proportion of
123
Fig. 4 Plasma concentration of cTnT collected before, at the initiation of CPB and 30 min after starting CPB in patients undergoing CABG using ICCF (a). Plasma concentration of cTnT collected at different time points and up to 5 h after starting CPB (b). Data are presented as mean ± SE. *p \ 0.001 versus earlier collections. **p \ 0.0005 versus rest. Majority of values in increased significantly with time compared to previous levels (repeated measures ANOVA)
surgeons uses ICCF in view of the fact that several prospective randomized trials have indicated that the cardioprotective efficacy and clinical outcome of cross-clamp fibrillation are similar to cold crystalloid or to cold or warm blood cardioplegia. A comprehensive review in 2004 [19] concluded that ICCF for coronary revascularization is highly effective method for myocardial protection. In comparison to cold crystalloid cardioplegia, ICCF has been shown to have similar postoperative mortality, clinical outcome plasma lipid hydroperoxides and total antioxidant status as well as markers of cardiac injury [20]. This
Mol Cell Biochem
measuring energy-rich phosphates in myocardial biopsies in patients undergoing CABG using cold or warm blood cardioplegia (e.g. [5]). Similar studies measuring metabolic status in the myocardium following arrest using ICCF have not been reported. The metabolic issue has been highlighted by work involving measurement of myocardial pH and showing that using ICCF does not result in reliable reperfusion of the myocardium [7]. This work is consistent with our finding that upon final reperfusion following ICCF, there is significant metabolic derangement in energy-rich phosphates (Figs. 1, 2). The observation that purines are high during reperfusion is an indication that the heart is struggling to re-establish metabolic homeostasis. Lactate levels do not appear to have increased as the fast lactate/H? co-transporter is presumed to have refluxed the accumulated metabolites [18]. The fall in NAD? and the implication that acidosis has been reduced (lactate efflux) suggest an opening of the mitochondrial permeability pore. This pore opening is triggered by ischemia/reperfusion induced Ca2? loading and oxidative stress [16]. The low levels of energy-rich phosphates and the indication that the mitochondrial pore is opened strongly indicate a significant reperfusion injury at this early stage of reperfusion. ICCF is associated with early reperfusion injury
Fig. 5 Plasma concentration of CK-MB collected before, at the initiation of CPB and 30 min after starting CPB in patients undergoing CABG using ICCF (a). Plasma concentration of CKMB collected at different time points and up to 5 h after starting CPB (b). Data are presented as mean ± SE. *p \ 0.001 versus earlier collections. Majority of values in increased significantly with time compared to previous levels (repeated measures ANOVA)
favourable comparison has also been extended to blood cardioplegic solutions (e.g. [21, 22] ). One key difference between cardioplegic arrest and ICCF is the degree of anaerobic stress. Whereas cardioplegic arrest induces immediate arrest and, therefore, preservation of cardiac metabolites, ICCF is associated with significant and immediate ischemic stress. It is assumed, however, that reperfusion after short periods of ischemia is sufficient to replenish energy-rich phosphates which is not the case following the lengthy ischaemic duration associated with cardioplegic arrest [6]. Evidence for this has come from studies
Although ICCF is known to be associated with reperfusion injury, this has been largely reported in biomarkers measured over relatively long periods mostly starting at least 1 h postoperatively and focusing on peak or area under the curve (total release). The gradual release of cardiac proteins over days indicates gradual death of cardiomyocytes. This death can be by necrosis or apoptosis, the latter being a slow process that can occur near the acutely damaged areas. In this study, we measured two different biomarkers CK-MB and cTnT as they are released at different rates and there is evidence that they do not always show the same outcome [22]. Interestingly, there was a significant increase in cTnT immediately following institution of CPB but not in CK-MB (Figs. 4a, 5a). However, there was a marked and significant increase in plasma CK-MB (2 fold) and cTnT (8 fold compared to basal levels) after 30 min of CPB. During this period, it is expected that most of the hearts would have had at least 2–3 sessions of ICCF ischemia/reperfusion. The pattern of cardiac protein release changed with time where cTnT slowly and gradually increased to peak levels at 5 h after CPB. Myocardial changes and reperfusion injury associated with ICCF and the role of mCPB The use of mCPB instead of cCPB during CABG attenuates neutrophil activation and cytokine release and has been shown to have some beneficial effects on blood
123
Mol Cell Biochem
conservation [23] and a better biologic profile [10]. However, the potential efficacy/relevance of mCPB is expected to be more evident in high-risk patients or in complex cardiac surgery requiring much longer cardiopulmonary perfusion. Several studies have already reported improved myocardial protection using mCPB compared to cCPB (e.g. [8–10]). These studies have used cardioplegic arrest (warm blood or cold crystalloid) in patients subjected to elective coronary artery bypass and/or aortic valve replacement and reported reduced postoperative cardiac injury (CK-MB and troponin I) using mCPB compared to cCPB [8]. Our study is a further indication showing that mCPB appears to be also cardioprotective when using ICCF. Here, we found the total cTnT release over a 5 h period was lower in the mCPB group compared to the cCPB group, which suggests this added protection can be seen irrespective of the cardioprotective strategy. What is novel in our study is the comparison between the two CPB interventions early after going off bypass. The use of mCPB did not alter myocardial metabolic changes or early reperfusion injury (Figs. 1, 5). The difference was in the reduction in overall injury in mCPB group over the entire duration of 5 h and only for cTnT but not for CK-MB. The sensitivity of cTnT as biomarker of cardiac injury is much higher than CK-MB. Therefore, during the early phases of injury, the difference can be masked but as cardiac injury evolves, this will also become significant as seen with cardioplegic arrest. The reason for the overall improvement in cardioprotection with mCPB has been attributed to factors that include pump type, lower priming volume, avoidance of open cardiotomy suction, and lower inflammation and oxidative stress [9, 10, 24] which can also be the underlying factors in our study. Of particular interest would be the inflammatory and oxidative stress, both have been implicated in cardiac injury [25, 26]. Oxidative stress (systemic or produced by cardiomyocytes) is known to sensitise the mitochondrial permeability transition pore opening to Ca2? (elevated during ischemia/reperfusion) and thus augment cardiac injury [16]. In conclusion, this study shows that ICCF in patients undergoing coronary artery bypass graft surgery using CPB causes a significant myocardial metabolic derangement. This stress is associated with early cardiac injury and the overall extent of injury is less when using mCPB compared to cCPB. Limitations This is a relatively small proof of concept sub-study with restrictions on collecting samples. Markers of injury should have been measured for several days after surgery. A larger study is needed to compare cardioplegic arrest and ICCF and to establish the benefits of mCPB.
123
Acknowledgments We would like to thank Hua Lin at the Bristol Heart Institute for processing blood and tissue samples and for measuring cellular metabolites. We would like to thank clinical staff members at Hammersmith Hospital, Imperial College for their valuable help and support in carrying out this study. This work was supported by the National Institute for Health Research Bristol Biomedical Research Unit in Cardiovascular Disease and the charity Heart Research UK. Disclaimer This article/paper/report presents independent research funded by the National Institute for Health Research (NIHR). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. Conflict of interest
None declared.
References 1. Koster JK Jr, Cohn LH, Collins JJ Jr, Sanders JH, Muller JE, Young E (1977) Continuous hypothermic arrest versus intermittent ischemia for myocardial protection during coronary revascularization. Ann Thorac Surg 24:330–336 2. Karthik S, Grayson AD, Oo AY, Fabri BM (2004) A survey of current myocardial protection practices during coronary artery bypass grafting. Ann R Coll Surg Engl 86:413–415. doi:10.1308/ 147870804669 3. Scarci M, Fallouh HB, Young CP, Chambers DJ (2009) Does intermittent cross-clamp fibrillation provide equivalent myocardial protection compared to cardioplegia in patients undergoing bypass graft revascularisation? Interact Cardiovasc Thorac Surg 9:872–878. doi:10.1510/icvts.2009.209437 4. Fujii M, Chambers DJ (2005) Myocardial protection with intermittent cross-clamp fibrillation: does preconditioning play a role? Eur J Cardiothorac Surg 28:821–831. doi:10.1016/j.ejcts.2005.06.048 5. Ascione R, Gomes WJ, Angelini GD, Bryan AJ, Suleiman MS (1998) Warm blood cardioplegia reduces the fall in the intracellular concentration of taurine in the ischaemic/reperfused heart of patients undergoing aortic valve surgery. Amino Acids 15:339–350 6. Pepper JR, Lockey E, Cankovic-Darracott S, Braimbridge MV (1982) Cardioplegia versus intermittent ischaemic arrest in coronary bypass surgery. Thorax 37:887–892 7. Dunning J, Hunter S, Kendall SW, Wallis J, Owens WA (2006) Coronary bypass grafting using crossclamp fibrillation does not result in reliable reperfusion of the myocardium when the crossclamp is intermittently released: a prospective cohort study. J Cardiothorac Surg 1:45. doi:10.1186/1749-8090-1-45 8. El-Essawi A, Hajek T, Skorpil J, Boning A, Sabol F, Hausmann H, Ostrovsky Y, Harringer W (2010) A prospective randomised multicentre clinical comparison of a minimised perfusion circuit versus conventional cardiopulmonary bypass. Eur J Cardiothorac Surg 38:91–97. doi:10.1016/j.ejcts.2010.01.035 9. Skrabal CA, Steinhoff G, Liebold A (2007) Minimizing cardiopulmonary bypass attenuates myocardial damage after cardiac surgery. ASAIO J 53:32–35. doi:10.1097/01.mat.0000249868.96923.1e 10. Remadi JP, Rakotoarivelo Z, Marticho P, Benamar A (2006) Prospective randomized study comparing coronary artery bypass grafting with the new mini-extracorporeal circulation Jostra system or with a standard cardiopulmonary bypass. Am Heart J 151:198. doi:10.1016/j.ahj.2005.03.067 11. Caputo M, Bryan AJ, Calafiore AM, Suleiman MS, Angelini GD (1998) Intermittent antegrade hyperkalaemic warm blood cardioplegia supplemented with magnesium prevents myocardial substrate derangement in patients undergoing coronary artery bypass surgery. Eur J Cardiothorac Surg 14:596–601
Mol Cell Biochem 12. Momin AU, Sharabiani MT, Kidher E, Najefi A, Mulholland JW, Reeves BC, Angelini GD, Anderson JR (2013) Feasibility and safety of minimized cardiopulmonary bypass in major aortic surgery. Interact Cardiovasc Thorac Surg 17:659–663. doi:10. 1093/icvts/ivt285 13. Imura H, Caputo M, Parry A, Pawade A, Angelini GD, Suleiman MS (2001) Age-dependent and hypoxia-related differences in myocardial protection during pediatric open heart surgery. Circulation 103:1551–1556 14. Ally A, Park G (1992) Rapid determination of creatine, phosphocreatine, purine bases and nucleotides (ATP, ADP, AMP, GTP, GDP) in heart biopsies by gradient ion-pair reversed-phase liquid chromatography. J Chromatogr 575:19–27 15. Di Lisa F, Menabo R, Canton M, Barile M, Bernardi P (2001) Opening of the mitochondrial permeability transition pore causes depletion of mitochondrial and cytosolic NAD? and is a causative event in the death of myocytes in postischemic reperfusion of the heart. J Biol Chem 276:2571–2575 16. Halestrap AP, Pasdois P (2009) The role of the mitochondrial permeability transition pore in heart disease. Biochim Biophys Acta 1787:1402–1415 17. Ghosh S, Galinanes M (2003) Protection of the human heart with ischemic preconditioning during cardiac surgery: role of cardiopulmonary bypass. J Thorac Cardiovasc Surg 126:133–142 18. Lin H, Suleiman MS (2003) Cariporide enhances lactate clearance upon reperfusion but does not alter lactate accumulation during global ischaemia. Pflugers Arch 447:8–13 19. Korbmacher B, Simic O, Schulte HD, Sons H, Schipke JD (2004) Intermittent aortic cross-clamping for coronary artery bypass grafting: a review of a safe, fast, simple, and successful technique. J Cardiovasc Surg (Torino) 45:535–543
20. Cohen AS, Hadjinikolaou L, McColl A, Richmond W, Sapsford RA, Glenville BE (1997) Lipid peroxidation, antioxidant status and troponin-T following cardiopulmonary bypass. A comparison between intermittent crossclamp with fibrillation and crystalloid cardioplegia. Eur J Cardiothorac Surg 12:248–253 21. Musumeci F, Feccia M, MacCarthy PA, Ellis GR, Mammana L, Brinn F, Penny WJ (1998) Prospective randomized trial of single clamp technique versus intermittent ischaemic arrest: myocardial and neurological outcome. Eur J Cardiothorac Surg 13:702–709 22. Anderson JR, Hossein-Nia M, Kallis P, Pye M, Holt DW, Murday AJ, Treasure T (1994) Comparison of two strategies for myocardial management during coronary artery operations. Ann Thorac Surg 58:768–772 discussion 772–763 23. Ohata T, Mitsuno M, Yamamura M, Tanaka H, Kobayashi Y, Ryomoto M, Yoshioka Y, Tsujiya N, Miyamoto Y (2008) Beneficial effects of mini-cardiopulmonary bypass on hemostasis in coronary artery bypass grafting: analysis of inflammatory response and hemodilution. ASAIO J 54:207–209. doi:10.1097/ MAT.0b013e3181648dbc 24. van Boven WJ, Gerritsen WB, Waanders FG, Haas FJ, Aarts LP (2004) Mini extracorporeal circuit for coronary artery bypass grafting: initial clinical and biochemical results: a comparison with conventional and off-pump coronary artery bypass grafts concerning global oxidative stress and alveolar function. Perfusion 19:239–246 25. Suleiman MS, Zacharowski K, Angelini GD (2008) Inflammatory response and cardioprotection during open-heart surgery: the importance of anaesthetics. Br J Pharmacol 153:21–33 26. Suleiman MS, Hancock M, Shukla R, Rajakaruna C, Angelini GD (2011) Cardioplegic strategies to protect the hypertrophic heart during cardiac surgery. Perfusion 26(Suppl 1):48–56
123
Metabolic derangement and cardiac injury early after reperfusion following intermittent cross-clamp fibrillation in patients undergoing coronary artery bypass graft surgery using conventional or miniaturized cardiopulmonary bypass Bao A. V. Nguyen • M-Saadeh Suleiman • Jonathan R. Anderson • Paul C. Evans • Francesca Fiorentino • Barnaby C. Reeves • Gianni D. Angelini
Received: 25 March 2014 / Accepted: 2 June 2014 Ó Springer Science+Business Media New York 2014
Abstract Myocardial ischemic stress and early reperfusion injury in patients undergoing coronary artery bypass grafting (CABG) operated on using intermittent crossclamp fibrillation (ICCF) are not presently known. The role of mini-cardiopulmonary bypass (mCPB) versus conventional CPB (cCPB) during ICCF has not been investigated. These issues have been addressed as secondary objective of randomised controlled trial (ISRCTN30610605) comparing cCPB and mCPB. Twenty-six patients undergoing primary elective CABG using ICCF were randomised to either cCPB or mCPB. Paired left ventricular biopsies collected from 21 patients at the beginning and at the end of CPB were used to measure intracellular substrates (ATP and related compounds). Cardiac troponin T (cTnT) and CKMB levels were measured in plasma collected from all patients preoperatively and after 1, 30, 60, 120, and 300 min after institution of CPB. ICCF was associated with significant ischemic stress as seen by fall in energy-rich phosphates early after reperfusion. There was also a fall in nicotinamide adenine dinucleotide (NAD?) indicating cardiomyocyte death which was confirmed by early release of cTnT and CK-MB during CPB. Ischemic stress and early myocardial injury were similar for cCPB and mCPB. B. A. V. Nguyen J. R. Anderson P. C. Evans F. Fiorentino B. C. Reeves G. D. Angelini Department of Cardiothoracic Surgery, NHLI, Hammersmith Hospital, Imperial College, London, UK M.-S. Suleiman (&) Bristol Heart Institute, University of Bristol, Bristol, UK e-mail: [email protected] Present Address: P. C. Evans Department of Cardiovascular Science, University of Sheffield, Sheffield, UK
However, the overall cardiac injury was significantly lower in the mCPB group as measured by cTnT (mean ± SEM: 96 ± 14 vs. 59 ± 8 lg/l, p = 0.02), but not with CK-MB. ICCF is associated with significant metabolic derangement and early myocardial injury. This early outcome was not affected by the CPB technique. However, the overall cardiac injury was lower for mCPB only when measured using cTnT. Keywords Cardiac injury Ischemia ATP Cardiopulmonary bypass Cross-clamp fibrillation Cardiac surgery Abbreviations ICCF Intermittent cross-clamp fibrillation CABG Coronary artery bypass grafting cCPB Conventional cardiopulmonary bypass mCPB Mini-cardiopulmonary bypass cTnT Cardiac troponin T CK-MB Creatine kinase
Introduction Induced ventricular fibrillation in conjunction with intermittent aortic cross-clamping to arrest the heart in patients undergoing open heart surgery for coronary revascularization was widely used more than 35 years ago [1]. It is estimated that this technique is routinely used by 10–15 % of surgeons in the UK [2]. Proponents of this intervention advocate its use on the assumption that the heart can tolerate repeated short periods of aortic cross-clamping as the metabolic stress due to global myocardial ischemia can be
123
Mol Cell Biochem
addressed by a short period of reperfusion. Cardiac centres that continue to use this technique in preference to cardioplegic arrest are also encouraged by reports showing that the cardioprotective efficacy is similar to cardioplegic arrest [3]. The potential advantages of using intermittent cross-clamp fibrillation (ICCF) over cardioplegic arrest include shorter ischemic time and possibly myocardial protection from ‘‘ischemic preconditioning’’ as has been demonstrated experimentally [3, 4]. However, a potential disadvantage of this technique is the risk of thromboembolism and repeated aortic clamping. Myocardial biopsies from patients undergoing coronary artery bypass graft (CABG) surgery using cold or warm blood cardioplegia have been used to demonstrate metabolic derangement [5]. However, myocardial biopsies following ICCF have been used only to monitor the deterioration in structural integrity but not in cellular metabolites [6]. More recently, work involving measurement of myocardial pH showed that using ICCF does not result in reliable reperfusion of the myocardium [7]. Consequently, a direct measurement of cardiac cellular metabolites associated with ICCF and reperfusion is important to determine the extent of the ischemic insult and reperfusion injury. Therefore, we decided to measure myocardial metabolites and early reperfusion injury associated with ICCF. The work was done as a secondary objective in an ongoing randomised controlled trial comparing conventional (cCPB) and miniaturized cardiopulmonary bypass (mCPB) in patients undergoing CABG. Earlier studies have already shown improved myocardial protection using mCPB using cardioplegic arrest [8–10]. However, there are no reports comparing cardiac injury (or ischemic stress) between the two CPB techniques using ICCF. Therefore, in addition to our main aim to assess metabolic derangement and early reperfusion injury during arrest using ICCF, we also compared myocardial metabolic changes and injury using the two CPB techniques.
Materials and methods The study was sub-study of a larger trial approved by the hospital ethics committee (registration number for the trial is ISRCTN30610605) and informed consent was obtained from all participants who were undergoing primary elective CABG using ICCF at the Hammersmith Hospital. Patients under 18 years of age were excluded as well as those who had any of the following: ejection fraction \30 % on echocardiogram; recent cerebro-vascular accident within 3 months preoperatively or more than 75 % carotid artery obstruction on ultrasound; pre-existing renal impairment with serum creatinine in excess of 177 lmol/L; pre-existing coagulopathy; pre-existing liver dysfunction or the
123
Table 1 Characteristics of mCPB and cCPB cCPB
mCPB
Target activated clotting time (ACT)
[400 s
[400 s
Venous cannula
34 Fr 2 stage (Medtronic)
29Fr Optiflow (Sorin) 3/8 inch tubing
1/2 inch tubing Reservoir
Sorin Evo
Soft shell reservoir
Pump
Roller pump (Stockert, Germany)
Centrifugal pump (Stockert, Germany)
Heat exchanger/ oxygenator
Avant (Sorin)
Eos (Sorin)
Flows (target cardiac index)
1.8–2.4 L/min/ m2 cooled to 32 °C
1.8–2.4L/min/m2 cooled to 32 °C
Prime volume
1,400 mls Hartmann’s
300 mls Hartmann’s with retrograde autologous prime
Safety features Operative field blood
Venous air removal device (VARD) Cardiotomy suction
Cell salvage
recent (within 5 days) use of antiplatelet agents (aspirin/ clopidogrel). Patients selected for this sub-study (n = 26) were allocated to the same surgeon and were randomized to either cCPB (n = 13) or mCPB (n = 13). The number of patients recruited for this study was based on our extensive work (e.g. [11]) collecting sequential ventricular biopsies from the same patient and showing differences in metabolites during surgery.
Anaesthetic and surgical techniques Anaesthetic technique was standardized for all patients. Thiopentone (1–3 mg/kg) was used for induction with 3–5 mg/kg fentanyl, and volatile agents were delivered in 50 % air–O2 mixture for maintenance. Propofol (3 mg/kg/h) was given as an infusion during cardiopulmonary bypass and neuromuscular blockade was achieved by 0.1–0.15 mg/kg Pancuronium bromide. Alpha stat acid–base management was adopted. Initial anticoagulation was accomplished with 3 mg/kg body weight of heparin and was supplemented as required in order to maintain an active clotting time of 480 s or above. All operations were performed using cCPB or mCPB as previously described [12] with ascending aortic cannulation, two stage venous cannulation and moderate systemic hypothermia (32 °C). The Sorin Dideco Extracorporeal Circulation Optimized (ECCO) system was used as the mCPB circuit. This ‘‘Hammersmith’’ miniaturized CPB
Mol Cell Biochem
system which has been described in details [12] is composed of 29-French OptiFlow venous cannula, venous air removal device, centrifugal pump, heat exchanger, oxygenator module, arterial line filter and parallel soft-shell reservoir. The conventional CPB (cCPB) circuit utilised a Stockert roller pump, membrane oxygenator, venous reservoir and non-heparin bonded circuit with cardiotomy suction. Key characteristics of both mCPB and cCPB are shown in Table 1. ICCF was introduced after cross-clamping of the aorta in order to perform coronary grafting. This was followed by de-clamping the aorta and reperfusion of the heart. The grafts were done sequentially for distal anastomoses; and with the clamp off and side biting clamp on for proximals. Left internal mammary artery to left anterior descending coronary artery (LIMA-LAD) was constructed last. The approximate time frame for this was as follows: 10 min for last vein top-end; 10 min for LIMA-LAD; 1–10 min to check anastomoses, undertake haemostasis, revise; and finally 2–5 min for biopsy collection and repair of biopsy site. CPB was then discontinued, and post-operative management was standard for all participants in accordance with institution protocols. Collection of ventricular biopsy Myocardial tissue biopsy specimens were sampled from the ´ 11.4 cm Cannula apex of the left ventricle using a 14 Ga. TW Trucut needle (Baxter Healthcare Corporation, Deerfield, IL 60015 USA). Two biopsies were taken, the first biopsy just after the institution of CPB and the second immediately prior to the cessation of CPB. Each specimen was immediately snap frozen (less than 5 s) in liquid nitrogen and stored at -80 °C until processing for cardiac cellular metabolites. All the biochemical analyses were performed by an investigator blinded to the interventions used. Extraction and determination of cellular metabolites The procedure used to extract cardiac metabolites was similar to that described previously [13]. In brief, frozen biopsies were crushed in liquid nitrogen with a pestle and mortar and transferred to a tube containing 500 ll 4.8 % perchloric acid, weighed and centrifuged at 3,6459g for 10 min at 4 °C. The supernatant was then taken and neutralised with equal volume of 0.44 M K2CO3 and the solution was centrifuged for 5 min at 4 °C to remove the insoluble potassium perchlorate. The supernatant was taken and stored at -20 °C until analysed for cellular metabolites using high performance liquid chromatography (HPLC) (ATP related compounds) or an enzymatic kit (Lactate).
HPLC was used to measure cellular metabolites [14] using a Beckman System and a 3-lm octadecyl silane Hypersil column (150 9 4.6 mm2) (Thermo Scientific, UK). Eluent A contained 150 mM KH2PO4 and 150 mM KCl, set at pH 6.0 with KOH. Eluent B consisted of eluent A with 15 % (v/v) HPLC grade acetonitrile. A low-pressure gradient mixing device was used to control the composition of the mobile phase. The amount of eluent B changed linearly between the time points. The analytical column was maintained at a constant temperature in the range of 17–19 °C and absorption was measured at 254 nm. The metabolites measured using HPLC were ATP, ADP, AMP, GTP, GDP, GMP, IMP, xanthine, hypoxanthine, nicotinamide adenine dinucleotide (NAD?), inosine and adenosine. The levels of NAD? in cardiac biopsies have been used as a measure of the mitochondrial permeability transition pore opening and, therefore, cardiomyocyte death [15]. Early after reperfusion, calcium loading and oxidative stress trigger opening of the permeability transition pore which can lead to cardiomyocyte death [16]. Standards for HPLC were prepared as previously described [14]. 20 ll of the extracted sample was injected into the column. The area under the peak for each metabolite was determined and compared to the standard sample, then corrected for weight. Lactate was measured using a plasma lactate determination kit from Sigma Diagnostics (Sigma, Poole, UK). Markers of cardiac injury In this study, we also monitored markers of cardiac injury in the early phases (*1 and 30 min) after the commencement of CPB. Blood collections were made before surgery (on the morning of surgery and post intubation following anaesthesia) and immediately after starting CPB and then at 30, 60, 120 and 300 min after the initiation of CPB. In view of the known reports suggesting that different markers appear at different time points during reperfusion, we decided to measure two markers of injury; cardiac troponin T (cTnT) and Creatine Kinase (isoenzyme MB). cTnT is a cardio-specific and highly sensitive marker for myocardial damage. Although it is widely accepted that cTnT release peaks relatively late within the first day, a study using ICCF during CABG has shown that this biomarker peaks at 5–6 h post-operatively [17]. Immunoassay for the in vitro quantitative determination of both CK-MB and cTnT was used (Roche Diagnostics GmbH). Data collection and analysis Intention-to-treat statistical analyses were performed using Statview for Windows (SAS Institute Inc.). Data were expressed as the mean ± SE unless otherwise stated. Nonparametric tests for paired and unpaired groups were
123
Mol Cell Biochem
carried out using Mann–Whitney and Wilcoxon signed rank test, respectively. The myocardial tissue concentration of metabolites was used to approximate the intracellular concentration. Area under the curve to estimate total protein (cTnT and CK-MB) release during CPB was calculated using the trapezium rule (Microsoft Excel).
Results Data for clinical outcome and markers of injury were collected for all 26 patients. However, it was only possible to collect two paired myocardial biopsies from 21 patients (11 on mCPB). The rest of the patients had either one very small biopsies (n = 2) or at least one biopsy that was not collected for practical reasons (n = 3). Clinical outcome There were no post-operative deaths. No differences in preoperative characteristics or in clinical outcomes were observed between cCPB or mCPB groups, except the mCPB group had a higher proportion of non-diabetics (Tables 2, 3).
Table 3 Peri- and post-operative data Variable
mCPB (n = 13)
cCPB (n = 13)
p values
Cardiopulmonary bypass time (minutes)
72.5 (4.49)
75.07 (5.08)
ns
Total cross-clamp time (minutes)
29.54 (2.33)
32.84 (2.15)
ns
Number of coronary artery grafts
3.23 (0.18)
3.23 (0.26)
ns
Time on ITU (days)
1.00 (0.13)
1.26 (0.27)
ns
New post-operative neurological dysfunction
0
0
ns
New haemofiltration post-operatively
0
1 (8 %)
ns
Infective complications Harvest site infection
0
1 (8 %)
ns
Sternal wound infection
0
2 (15 %)
ns
Return to theatre
0
0
ns
Values are expressed as number or mean with SD in parenthesis or percentage as indicated
Changes in myocardial energy-rich phosphates associated with ICCF
Table 2 Preoperative characteristics Variable
mCPB (n = 13)
cCPB (n = 13)
p values
Male/female
8/5
10/3
ns
Age (years)
68.4 (1.97)
66.3 (2.24)
ns
Body surface area (m2)
1.87 (0.06)
1.98 (0.05)
ns
Body mass index (kg/ m2) Logistic euroSCORE (%)
29.74 (1.31)
29.31 (1.19)
ns
3.35 (0.68)
3.74 (0.91)
ns
Single vessel disease
0
1 (8 %)
ns
Two vessel disease
1 (8 %)
1 (8 %)
ns
Three vessel disease
12 (92 %)
11 (84 %)
ns
Good [50 %
11 (85 %)
12 (92 %)
ns
Fair 30–50 %
2 (15 %)
1 (8 %)
ns
Not diabetic
11 (85 %)
6 (46 %)
0.03*
Oral therapy Insulin
1 (8 %) 1 (8 %)
5 (38 %) 2 (15 %)
ns ns
Coronary vessel disease
Ejection fraction
Diabetes
Values are expressed as number or mean with SD in parenthesis or percentage as indicated * indicates statistical significance
123
There was a fall in the myocardial levels of the main energy-rich phosphates, the adenine nucleotides (ATP, ADP, AMP), in biopsies taken at the end of the CPB period compared to biopsies taken just after instituting CPB (Fig. 1a). This derangement was seen in patients operated on using cCPB or mCPB (Fig. 1b). In addition to the fall in adenine nucleotides, there was also a significant fall in guanine nucleotides GTP, GDP and GMP (Fig. 2a). Again, like adenine nucleotides, the fall and trends were similar for both cCPB and mCPB (Fig. 2b). The fall in energy-rich phosphates during anaerobic metabolism is normally associated with accumulation of ATP catabolites including hypoxanthine, inosine, xanthine as well as lactate which tend to efflux upon reperfusion [18]. Nevertheless, even at the end of CPB, which follows from the last period of ICCF (see ‘‘Materials and Methods’’ for details), there was a significant increase in hypoxanthine (p = 0.02, Wilcoxon Rank test) and a strong trend for inosine to increase (p = 0.05, Wilcoxon Rank test) associated with arrest and reperfusion (Fig. 3a). In contrast, lactate levels were similar to those for basal levels (0.027 ± 0.005 vs. 0.033 ± 0.005 nmol/mg wet weight and 0.026 ± 0.006 vs. 0.024 ± 0.004 nmol/mg wet weight for cCPB and mCPB, respectively).
Mol Cell Biochem
Fig. 1 Myocardial concentration of adenine nucleotides ATP, ADP & AMP in biopsies collected at the initiation of CPB and at the end of CPB from patients undergoing CABG using ICCF (a). A comparison between conventional or mini-CPB is shown in b. Data are presented as mean ± SE. *p \ 0.05 and **p \ 0.001 versus start of CPB for corresponding metabolite
Fig. 2 Myocardial concentration of guanine nucleotides GTP, GDP & GMP in biopsies collected at the initiation of CPB and at the end of CPB from patients undergoing CABG using ICCF (a). A comparison between conventional or mini-CPB is shown in b. Data are presented as mean ± SE. *p \ 0.05 and **p \ 0.001 versus start of CPB for corresponding metabolite
Opening of mitochondrial permeability transition pore
sessions of ischemia and reperfusion) irrespective of whether the biomarker was cTnT or CK-MB (Figs. 4a, 5a). Figures 4b and 5b show the entire profile of the release of cTnT and CK-MB using cCPB or mCPB. Whilst CK-MB levels were similar for both interventions (3.5 ± 0.3 vs. 3.3 ± 0.3 lg/ml, p = 0.51) total cTnT release was significantly lower in the mCPB group (95.9 ± 14.4 vs. 58.8 ± 7.6 lg/l, p = 0.02).
Figure 3b shows a significant fall in NAD? upon reperfusion and CPB, irrespective of type of CPB, indicating that triggering of cardiomyocyte death occurred soon after starting CPB and ICCF. Therefore, it was expected that levels of markers of myocardial injury would be elevated during ICCF & CPB. Early myocardial injury associated with ICCF
Discussion Consistent with opening of the mitochondrial permeability transition pore soon after reperfusion, we found significantly elevated levels of markers of myocardial injury in blood within 30 min after institution of CPB (this period is likely to be associated with at least 2 grafts and two
In this study, we show for the first time that the use of ICCF for CABG is associated with significant ischemic stress, metabolic derangement and evidence of opening of the mitochondrial permeability transition pore. In contrast
123
Mol Cell Biochem
Fig. 3 Myocardial concentration of ATP catabolites hypoxanthine, inosine and adenosine (a) and NAD? (b) in biopsies collected at the initiation of CPB and at the end of CPB and allocated to either conventional or mini-CPB. Data are presented as mean ± SE. *p \ 0.05 and **p \ 0.001 versus start of CPB in corresponding metabolite. #p = 0.05 versus start of CPB in corresponding metabolite
to other studies, we also measured markers of myocardial injury and observed significant increases in these markers during the period of ICCF and CPB. Both metabolic derangement and early reperfusion injury were found; these findings were largely independent of whether cCPB or mCPB had been used. However, total reperfusion injury as estimated by area under the curve for cTnT was significantly less in the mCPB group. Cardiac arrest during CABG with CPB is commonly performed using a depolarising hyperkalemic cardioplegic solution with different compositions. An alternative method to induce cardiac arrest utilises ICCF. In the UK, it is surprising that a relatively small 10–15 % proportion of
123
Fig. 4 Plasma concentration of cTnT collected before, at the initiation of CPB and 30 min after starting CPB in patients undergoing CABG using ICCF (a). Plasma concentration of cTnT collected at different time points and up to 5 h after starting CPB (b). Data are presented as mean ± SE. *p \ 0.001 versus earlier collections. **p \ 0.0005 versus rest. Majority of values in increased significantly with time compared to previous levels (repeated measures ANOVA)
surgeons uses ICCF in view of the fact that several prospective randomized trials have indicated that the cardioprotective efficacy and clinical outcome of cross-clamp fibrillation are similar to cold crystalloid or to cold or warm blood cardioplegia. A comprehensive review in 2004 [19] concluded that ICCF for coronary revascularization is highly effective method for myocardial protection. In comparison to cold crystalloid cardioplegia, ICCF has been shown to have similar postoperative mortality, clinical outcome plasma lipid hydroperoxides and total antioxidant status as well as markers of cardiac injury [20]. This
Mol Cell Biochem
measuring energy-rich phosphates in myocardial biopsies in patients undergoing CABG using cold or warm blood cardioplegia (e.g. [5]). Similar studies measuring metabolic status in the myocardium following arrest using ICCF have not been reported. The metabolic issue has been highlighted by work involving measurement of myocardial pH and showing that using ICCF does not result in reliable reperfusion of the myocardium [7]. This work is consistent with our finding that upon final reperfusion following ICCF, there is significant metabolic derangement in energy-rich phosphates (Figs. 1, 2). The observation that purines are high during reperfusion is an indication that the heart is struggling to re-establish metabolic homeostasis. Lactate levels do not appear to have increased as the fast lactate/H? co-transporter is presumed to have refluxed the accumulated metabolites [18]. The fall in NAD? and the implication that acidosis has been reduced (lactate efflux) suggest an opening of the mitochondrial permeability pore. This pore opening is triggered by ischemia/reperfusion induced Ca2? loading and oxidative stress [16]. The low levels of energy-rich phosphates and the indication that the mitochondrial pore is opened strongly indicate a significant reperfusion injury at this early stage of reperfusion. ICCF is associated with early reperfusion injury
Fig. 5 Plasma concentration of CK-MB collected before, at the initiation of CPB and 30 min after starting CPB in patients undergoing CABG using ICCF (a). Plasma concentration of CKMB collected at different time points and up to 5 h after starting CPB (b). Data are presented as mean ± SE. *p \ 0.001 versus earlier collections. Majority of values in increased significantly with time compared to previous levels (repeated measures ANOVA)
favourable comparison has also been extended to blood cardioplegic solutions (e.g. [21, 22] ). One key difference between cardioplegic arrest and ICCF is the degree of anaerobic stress. Whereas cardioplegic arrest induces immediate arrest and, therefore, preservation of cardiac metabolites, ICCF is associated with significant and immediate ischemic stress. It is assumed, however, that reperfusion after short periods of ischemia is sufficient to replenish energy-rich phosphates which is not the case following the lengthy ischaemic duration associated with cardioplegic arrest [6]. Evidence for this has come from studies
Although ICCF is known to be associated with reperfusion injury, this has been largely reported in biomarkers measured over relatively long periods mostly starting at least 1 h postoperatively and focusing on peak or area under the curve (total release). The gradual release of cardiac proteins over days indicates gradual death of cardiomyocytes. This death can be by necrosis or apoptosis, the latter being a slow process that can occur near the acutely damaged areas. In this study, we measured two different biomarkers CK-MB and cTnT as they are released at different rates and there is evidence that they do not always show the same outcome [22]. Interestingly, there was a significant increase in cTnT immediately following institution of CPB but not in CK-MB (Figs. 4a, 5a). However, there was a marked and significant increase in plasma CK-MB (2 fold) and cTnT (8 fold compared to basal levels) after 30 min of CPB. During this period, it is expected that most of the hearts would have had at least 2–3 sessions of ICCF ischemia/reperfusion. The pattern of cardiac protein release changed with time where cTnT slowly and gradually increased to peak levels at 5 h after CPB. Myocardial changes and reperfusion injury associated with ICCF and the role of mCPB The use of mCPB instead of cCPB during CABG attenuates neutrophil activation and cytokine release and has been shown to have some beneficial effects on blood
123
Mol Cell Biochem
conservation [23] and a better biologic profile [10]. However, the potential efficacy/relevance of mCPB is expected to be more evident in high-risk patients or in complex cardiac surgery requiring much longer cardiopulmonary perfusion. Several studies have already reported improved myocardial protection using mCPB compared to cCPB (e.g. [8–10]). These studies have used cardioplegic arrest (warm blood or cold crystalloid) in patients subjected to elective coronary artery bypass and/or aortic valve replacement and reported reduced postoperative cardiac injury (CK-MB and troponin I) using mCPB compared to cCPB [8]. Our study is a further indication showing that mCPB appears to be also cardioprotective when using ICCF. Here, we found the total cTnT release over a 5 h period was lower in the mCPB group compared to the cCPB group, which suggests this added protection can be seen irrespective of the cardioprotective strategy. What is novel in our study is the comparison between the two CPB interventions early after going off bypass. The use of mCPB did not alter myocardial metabolic changes or early reperfusion injury (Figs. 1, 5). The difference was in the reduction in overall injury in mCPB group over the entire duration of 5 h and only for cTnT but not for CK-MB. The sensitivity of cTnT as biomarker of cardiac injury is much higher than CK-MB. Therefore, during the early phases of injury, the difference can be masked but as cardiac injury evolves, this will also become significant as seen with cardioplegic arrest. The reason for the overall improvement in cardioprotection with mCPB has been attributed to factors that include pump type, lower priming volume, avoidance of open cardiotomy suction, and lower inflammation and oxidative stress [9, 10, 24] which can also be the underlying factors in our study. Of particular interest would be the inflammatory and oxidative stress, both have been implicated in cardiac injury [25, 26]. Oxidative stress (systemic or produced by cardiomyocytes) is known to sensitise the mitochondrial permeability transition pore opening to Ca2? (elevated during ischemia/reperfusion) and thus augment cardiac injury [16]. In conclusion, this study shows that ICCF in patients undergoing coronary artery bypass graft surgery using CPB causes a significant myocardial metabolic derangement. This stress is associated with early cardiac injury and the overall extent of injury is less when using mCPB compared to cCPB. Limitations This is a relatively small proof of concept sub-study with restrictions on collecting samples. Markers of injury should have been measured for several days after surgery. A larger study is needed to compare cardioplegic arrest and ICCF and to establish the benefits of mCPB.
123
Acknowledgments We would like to thank Hua Lin at the Bristol Heart Institute for processing blood and tissue samples and for measuring cellular metabolites. We would like to thank clinical staff members at Hammersmith Hospital, Imperial College for their valuable help and support in carrying out this study. This work was supported by the National Institute for Health Research Bristol Biomedical Research Unit in Cardiovascular Disease and the charity Heart Research UK. Disclaimer This article/paper/report presents independent research funded by the National Institute for Health Research (NIHR). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. Conflict of interest
None declared.
References 1. Koster JK Jr, Cohn LH, Collins JJ Jr, Sanders JH, Muller JE, Young E (1977) Continuous hypothermic arrest versus intermittent ischemia for myocardial protection during coronary revascularization. Ann Thorac Surg 24:330–336 2. Karthik S, Grayson AD, Oo AY, Fabri BM (2004) A survey of current myocardial protection practices during coronary artery bypass grafting. Ann R Coll Surg Engl 86:413–415. doi:10.1308/ 147870804669 3. Scarci M, Fallouh HB, Young CP, Chambers DJ (2009) Does intermittent cross-clamp fibrillation provide equivalent myocardial protection compared to cardioplegia in patients undergoing bypass graft revascularisation? Interact Cardiovasc Thorac Surg 9:872–878. doi:10.1510/icvts.2009.209437 4. Fujii M, Chambers DJ (2005) Myocardial protection with intermittent cross-clamp fibrillation: does preconditioning play a role? Eur J Cardiothorac Surg 28:821–831. doi:10.1016/j.ejcts.2005.06.048 5. Ascione R, Gomes WJ, Angelini GD, Bryan AJ, Suleiman MS (1998) Warm blood cardioplegia reduces the fall in the intracellular concentration of taurine in the ischaemic/reperfused heart of patients undergoing aortic valve surgery. Amino Acids 15:339–350 6. Pepper JR, Lockey E, Cankovic-Darracott S, Braimbridge MV (1982) Cardioplegia versus intermittent ischaemic arrest in coronary bypass surgery. Thorax 37:887–892 7. Dunning J, Hunter S, Kendall SW, Wallis J, Owens WA (2006) Coronary bypass grafting using crossclamp fibrillation does not result in reliable reperfusion of the myocardium when the crossclamp is intermittently released: a prospective cohort study. J Cardiothorac Surg 1:45. doi:10.1186/1749-8090-1-45 8. El-Essawi A, Hajek T, Skorpil J, Boning A, Sabol F, Hausmann H, Ostrovsky Y, Harringer W (2010) A prospective randomised multicentre clinical comparison of a minimised perfusion circuit versus conventional cardiopulmonary bypass. Eur J Cardiothorac Surg 38:91–97. doi:10.1016/j.ejcts.2010.01.035 9. Skrabal CA, Steinhoff G, Liebold A (2007) Minimizing cardiopulmonary bypass attenuates myocardial damage after cardiac surgery. ASAIO J 53:32–35. doi:10.1097/01.mat.0000249868.96923.1e 10. Remadi JP, Rakotoarivelo Z, Marticho P, Benamar A (2006) Prospective randomized study comparing coronary artery bypass grafting with the new mini-extracorporeal circulation Jostra system or with a standard cardiopulmonary bypass. Am Heart J 151:198. doi:10.1016/j.ahj.2005.03.067 11. Caputo M, Bryan AJ, Calafiore AM, Suleiman MS, Angelini GD (1998) Intermittent antegrade hyperkalaemic warm blood cardioplegia supplemented with magnesium prevents myocardial substrate derangement in patients undergoing coronary artery bypass surgery. Eur J Cardiothorac Surg 14:596–601
Mol Cell Biochem 12. Momin AU, Sharabiani MT, Kidher E, Najefi A, Mulholland JW, Reeves BC, Angelini GD, Anderson JR (2013) Feasibility and safety of minimized cardiopulmonary bypass in major aortic surgery. Interact Cardiovasc Thorac Surg 17:659–663. doi:10. 1093/icvts/ivt285 13. Imura H, Caputo M, Parry A, Pawade A, Angelini GD, Suleiman MS (2001) Age-dependent and hypoxia-related differences in myocardial protection during pediatric open heart surgery. Circulation 103:1551–1556 14. Ally A, Park G (1992) Rapid determination of creatine, phosphocreatine, purine bases and nucleotides (ATP, ADP, AMP, GTP, GDP) in heart biopsies by gradient ion-pair reversed-phase liquid chromatography. J Chromatogr 575:19–27 15. Di Lisa F, Menabo R, Canton M, Barile M, Bernardi P (2001) Opening of the mitochondrial permeability transition pore causes depletion of mitochondrial and cytosolic NAD? and is a causative event in the death of myocytes in postischemic reperfusion of the heart. J Biol Chem 276:2571–2575 16. Halestrap AP, Pasdois P (2009) The role of the mitochondrial permeability transition pore in heart disease. Biochim Biophys Acta 1787:1402–1415 17. Ghosh S, Galinanes M (2003) Protection of the human heart with ischemic preconditioning during cardiac surgery: role of cardiopulmonary bypass. J Thorac Cardiovasc Surg 126:133–142 18. Lin H, Suleiman MS (2003) Cariporide enhances lactate clearance upon reperfusion but does not alter lactate accumulation during global ischaemia. Pflugers Arch 447:8–13 19. Korbmacher B, Simic O, Schulte HD, Sons H, Schipke JD (2004) Intermittent aortic cross-clamping for coronary artery bypass grafting: a review of a safe, fast, simple, and successful technique. J Cardiovasc Surg (Torino) 45:535–543
20. Cohen AS, Hadjinikolaou L, McColl A, Richmond W, Sapsford RA, Glenville BE (1997) Lipid peroxidation, antioxidant status and troponin-T following cardiopulmonary bypass. A comparison between intermittent crossclamp with fibrillation and crystalloid cardioplegia. Eur J Cardiothorac Surg 12:248–253 21. Musumeci F, Feccia M, MacCarthy PA, Ellis GR, Mammana L, Brinn F, Penny WJ (1998) Prospective randomized trial of single clamp technique versus intermittent ischaemic arrest: myocardial and neurological outcome. Eur J Cardiothorac Surg 13:702–709 22. Anderson JR, Hossein-Nia M, Kallis P, Pye M, Holt DW, Murday AJ, Treasure T (1994) Comparison of two strategies for myocardial management during coronary artery operations. Ann Thorac Surg 58:768–772 discussion 772–763 23. Ohata T, Mitsuno M, Yamamura M, Tanaka H, Kobayashi Y, Ryomoto M, Yoshioka Y, Tsujiya N, Miyamoto Y (2008) Beneficial effects of mini-cardiopulmonary bypass on hemostasis in coronary artery bypass grafting: analysis of inflammatory response and hemodilution. ASAIO J 54:207–209. doi:10.1097/ MAT.0b013e3181648dbc 24. van Boven WJ, Gerritsen WB, Waanders FG, Haas FJ, Aarts LP (2004) Mini extracorporeal circuit for coronary artery bypass grafting: initial clinical and biochemical results: a comparison with conventional and off-pump coronary artery bypass grafts concerning global oxidative stress and alveolar function. Perfusion 19:239–246 25. Suleiman MS, Zacharowski K, Angelini GD (2008) Inflammatory response and cardioprotection during open-heart surgery: the importance of anaesthetics. Br J Pharmacol 153:21–33 26. Suleiman MS, Hancock M, Shukla R, Rajakaruna C, Angelini GD (2011) Cardioplegic strategies to protect the hypertrophic heart during cardiac surgery. Perfusion 26(Suppl 1):48–56
123
