synergy of isoﬂurane preconditioning and propofol ... · china, and department of anesthesiology...

Clinical Science (2011) 121, 57–69 (Printed in Great Britain) doi:10.1042/CS20100435 57

Synergy of isoflurane preconditioning andpropofol postconditioning reduces myocardial

reperfusion injury in patients

Zhiyong HUANG*†‡1, Xingwu ZHONG§1, Michael G. IRWIN†, Shangyi JI*,Gordon T. WONG†, Yanan LIU†, Zhong-yuan XIA‡, Barry A. FINEGAN‖ andZhengyuan XIA†‡∗Department of Anesthesia, Sun Yat-Sen Cardiovascular Hospital, Shenzhen, People’s Republic of China, †Department ofAnesthesiology, University of Hong Kong, Hong Kong SAR, People’s Republic of China, ‡Anesthesiology Research Laboratory,Department of Anesthesiology, Renmin Hospital of Wuhan University, Wuhan, People’s Republic of China, §ZhongshanOphthalmic Center and State Key Laboratory of Ophthalmology, Sun Yat-sen University, Guangzhou, People’s Republic ofChina, and ‖Department of Anesthesiology and Pain Medicine, University of Alberta, Edmonton, AB, Canada

A B S T R A C T

Either isoflurane preconditioning or high-dose propofol treatment has been shown to attenuatemyocardial IRI (ischaemia/reperfusion injury) in patients undergoing CABG (coronary arterybypass graft) surgery. It is unknown whether isoflurane and propofol may synergisticallyattenuate myocardial injury in patients. The present study investigated the efficacy ofIsoPC (isoflurane preconditioning), propofol treatment (postconditioning) and their synergy inattenuating postischaemic myocardial injury in patients undergoing CABG surgery using CPB(cardiopulmonary bypass). Patients (n = 120) selected for CABG surgery were randomly assignedto one of four groups (n = 30 each). After induction, anaesthesia was maintained either withfentanyl and midazolam (control; group C); with propofol at 100 μg · kg− 1 of body weight · min− 1

before and during CPB followed by propofol at 60 μg · kg− 1 of body weight · min− 1 for 15 minafter aortic declamping (group P); with isoflurane 1–1.5% end tidal throughout the surgery (groupI) or with isoflurane 1–1.5% end tidal before CPB and switching to propofol at 100 μg · kg− 1

of body weight · min− 1 during CPB followed by propofol at 60 μg · kg− 1 of body weight · min− 1

for 15 min after aortic declamping (group IP, i.e. IsoPC plus propofol postconditioning). A jointisoflurane and propofol anaesthesia regimen synergistically reduced plasma levels of cTnI (cardiactroponin I) and CK-MB (creatine kinase MB) and f-FABP (heart-type fatty acid-binding protein) (allP < 0.05 compared with control, group P or group I) and facilitated postoperative myocardialfunctional recovery. During reperfusion, myocardial tissue eNOS (endothelial NO synthase)protein expression in group IP was significantly higher, whereas nitrotyrosine protein expressionwas lower than those in the control group. In conclusion, a joint isoflurane preconditioning andpropofol anaesthesia regimen synergistically attenuated myocardial reperfusion injury in patients.

Key words: coronary artery bypass graft surgery, isoflurane, myocardial reperfusion injury, postconditioning, preconditioning.Abbreviations: ACC, aortic cross-clamping; AUC, area under curve; CABG, coronary artery bypass graft; CI, cardiac index; CK-MB, creatine kinase-MB; CPB, cardiopulmonary bypass; CVP, central venous pressure; cTnI, cardiac troponin I; eNOS, endothelialNO synthase; h-FABP, heart-type fatty acid-binding protein; HR, heart rate; ICU, intensive care unit; IL-6, interleukin-6; iNOS,inducible NO synthase; IRI, ischaemia/reperfusion injury; IsoPC, isoflurane preconditioning; MAC, minimum alveolar anestheticconcentration; MAP, mean arterial blood pressure; MDA, malondialdehyde; PI3K, phosphoinositide 3-kinase; PCWP, pulmonarycapillary wedge pressure; ROS, reactive oxygen species; SOD, superoxide dismutase; TNF-α, tumour necrosis factor-α.1 These authors contributed equally to this work.Correspondence: Dr Zhengyuan Xia (email [email protected]).

C© The Authors Journal compilation C© 2011 Biochemical Society

http://dx.doi.org/10.1042/CS20100435

58 Z. Huang and others

INTRODUCTION

Despite advancement in surgical and anaesthetictechniques, the risk of morbidity and mortality inCABG (coronary artery bypass graft) surgery remainssignificant, especially since patients are often those withincreased age and co-morbidities. Volatile anaestheticpreconditioning, in general, and IsoPC (isofluranepreconditioning) specifically, can provide moderate [1]to minor [2] cardiac protection during CABG. However,there may be limitations to its use given that itsduration of action is limited (25–40 min [3]) comparedwith the typical period of global ischaemia (longerthan 60 min) seen during CABG. In addition, aging[4] and diabetes [5] can attenuate IsoPC protection,largely because endogenous superoxide production (amechanism whereby volatile anaesthetic preconditioningconfers its protection [6]) is already increased in agedand/or diabetic patients.

During CABG surgery using CPB (cardiopulmonarybypass), ROS (reactive oxygen species) production isincreased [7]. Increased systemic oxidative stress mayadversely affect postoperative myocardial functionalrecovery by increasing lipid peroxidation [8]. The pro-inflammatory cytokine TNF-α (tumour necrosis factor-α) is also increased after CPB [9], which can furtherincrease ROS production. TNF-α plays an important rolein the inflammatory processes, which may induce andexacerbate cardiac dysfunction. Experimentally, volatileanaesthetic preconditioning was shown to reduce NF-κB(nuclear factor κB)-dependent inflammatory gene expres-sion and decrease TNF-α production and subsequentlyattenuate myocardial IRI (ischaemia/reperfusion injury)[10]. On the other hand, the intravenous anaestheticpropofol confers an antioxidant effect [11]. Therefore itis plausible that volatile anaesthetic preconditioning incombination with propofol treatment may synergisticallyattenuate myocardial IRI in clinical settings. Propofolreduced ROS-induced lipid peroxidation and attenuatedIRI in isolated rat hearts [12–16], in a dose-dependentmanner [13,16]. We have shown that propofol can preventhydrogen peroxide-mediated exacerbation of TNF-αcellular toxicity [17], which might be a mechanismwhereby propofol attenuates hydrogen peroxide-inducedcardiac dysfunction [18]. Recent experimental studiesshow that propofol postconditioning confers protectionagainst myocardial [19] as well as cerebral [20]IRI, probably through maintaining the activity ofthe prosurvival PI3K (phosphoinositide 3-kinase)/Aktpathway [20,21]. In addition, experimentally, propofolhas been shown to be cardioprotective when used at arelatively high dose solely during ischaemia [16] or whenapplied in a clinically relevant model of normothermicblood cardioplegic arrest and CPB [22].

We recently demonstrated that high-dose propofolduring CPB attenuates postoperative myocardial cellular

damage compared with isoflurane or low-dose propofolduring CABG [23], whereas the cardioprotective effectsof isoflurane did not differ from that of a low-dosepropofol regimen. Similarly, more recent clinical trialresults show that the use of isoflurane-opioid anaesthesiain comparison with propofol-opioid anaesthesia doesnot afford additional benefit in terms of postoperativetropolin-I release as well as 1-year cardiac morbidityand mortality [24]. However, it is unknown whetheror not isoflurane and propofol may have synergy inreducing postischaemic myocardial injury in patients.We hypothesized that synergy of IsoPC and propofolpostconditioning may confer superior protection againstmyocardial IRI in patients compared with an isofluraneor propofol anaesthesia regimen alone and that themechanism of the synergy is related to promotingor maintaining postischaemic myocardial levels ofeNOS (endothelial NO synthase), a critical downstreammediator of the prosurvival PI3K/Akt pathway [25] thatplays a key role in isoflurane mediated pre- and post-conditioning cardioprotection [26,27].

Part of this work was presented at the AmericanSociety of Anesthesiologists Annual Meeting, held inNew Orleans on 17–21 October, 2009 and subsequentlypublished in an abstract form [27a].

MATERIALS AND METHODS

Patient population and study designThe clinical trial was carried out in accordance with theDeclaration of Helsinki (2000) of the World MedicalAssociation. The study protocol was approved bythe institutional ethics committee. All subjects gavewritten informed consent after having been given afull explanation of the purpose, nature and risk of allprocedures used.

After obtaining written informed consent, 120 ASA(American Association of Anesthesiologists) class II toIII patients, aged 53–76 years, presenting for scheduledprimary elective CABG were assigned according to acomputer-generated random code to one of four groups:a control group receiving midazolam and fentanyl (groupC; n = 30); propofol alone (group P; n = 30); isofluranealone (group I; n = 30); and a combination of isofluraneand propofol (group IP; n = 30). Subjects were assignedtreatment numbers in ascending chronological order ofadmission in the study. The surgeons, research assistantsand medical and nursing staff in the operation roomwere blinded to the group assignments, facilitated bycovering the drug infusion pump and lines and shieldingthe isoflurane vaporizer from view.

The exclusion criteria were (i) emergency revasculariz-ation for unstable angina, (ii) previous coronary or valvu-lar heart surgery, (iii) combined operations (simultaneousvalve repair, carotid endarterectomy or left ventricular


Isoflurane and propofol synergy in cardioprotection 59

Figure 1 Anaesthesia protocol used in experimental groupsAfter induction, anaesthesia was maintained either with fentanyl (F) and midazolam (group C); with propofol (P) at 100 μg · kg− 1 of body weight · min− 1 beforeand during CPB followed by propofol at 60 μg · kg− 1 of body weight · min− 1 15 min after aortic declamping (group P); with isoflurane (Isol) 1–1.5 MAC end-tidalthroughout the surgery (group I); or with isoflurane 1–1.5 MAC end-tidal before CPB and switching to propofol at 100 μg · kg− 1 of body weight · min− 1 duringCPB followed by propofol 60 μg · kg− 1 of body weight · min− 1 15 min after aortic declamping (group IP). CPB–ACC indicates the onset of CPB to the whole periodof ACC. Reperfusion starts at the time of aortic declamping.

aneurysm repair), (iv) preoperative myocardial infarctionwithin the last 4 weeks or ongoing myocardial infarction,(v) poor ventricular function (ejection fraction�0.30), (vi)preoperative haemodynamic instability with the need formedical or mechanical support, (vii) severe hepatic disease(alanine aminotransferase or aspartate aminotransferase>150 units/l), (viii) renal insufficiency (creatinine con-centration >1.5 mg/dl), (ix) severe chronic obstructivepulmonary disease [FEV1 (forced expiratory volume in1 s) >0.8 litres], (x) severe coagulation abnormalities, and(xi) history of neurological disturbances.

Anaesthetic protocols and surgeryAll patients received standard premedication of sco-polamine at 0.006 mg/kg of body weight and morphineat 0.1 mg/kg of body weight intramuscularly 60 minbefore surgery. In all groups, anaesthesia was inducedwith etomidate at 0.3 mg/kg of body weight, fentanylat 8 μg/kg of body weight and pancuronium bromide at0.1 mg/kg of body weight given intravenously. Afterinduction, all patients received continuous infusionsof fentanyl at 0.6 μg · kg− 1 of body weight · min− 1

and pancuronium bromide at 15 μg · kg− 1 of bodyweight · min− 1 during surgery. The anaesthesia protocolin various groups is illustrated in Figure 1. In brief,anaesthesia was maintained either with fentanyl andmidazolam (group C); with propofol at 100 μg · kg− 1 ofbody weight · min− 1 before and during CPB followedby propofol at 60 μg · kg− 1 of body weight · min− 1 for15 min after aortic declamping (group P); with isoflurane1–1.5 MAC (minimum alveolar anesthetic concentration)end-tidal throughout the surgery (group I) or withisoflurane 1–1.5 MAC end-tidal before CPB and switchedto propofol at 100 μg · kg− 1 of body weight · min− 1

during CPB followed by propofol 60 μg · kg− 1 of bodyweight · min− 1 15 min after aortic declamping (groupIP). During anaesthesia, patients were monitored withfive-lead ECG, pulse oximetry, capnography, invasivearterial pressure and pulmonary artery pressure duringthe operation.

Systemic arterial blood pressure was measured viaradial artery catheterization. A Swan–Ganz catheter(Edwards Life Sciences) was inserted for central venouspressure, pulmonary artery pressures and cardiac outputdeterminations. Haemodynamics was continuouslymonitored for 24 h after CPB.

Porcine heparin was administrated (300 units/kgof body weight) and supplemented when needed tomaintain an activated coagulation time above 450 sduring CPB. Heparin was neutralized with 1 mg ofprotamine/100 units of heparin administrated afterseparation from CPB.

Surgery was conducted under conditions ofnormothermic or tepid CPB (patient oesophagealtemperature of 34–37 ◦C) utilizing a non-pulsatile flowrate of 2 l · min− 1 · m− 2 and a membrane oxygenator.MAP (mean arterial blood pressure) was maintainedbetween 60 and 75 mmHg during CPB. All patients weretreated with intermittent antegrade infusion of warmhigh-potassium blood cardioplegia during continuousaortic cross-clamping, and the content of the cardioplegicsolution was the same for all patients.

The ECG was recorded with a standard 12-leadelectrocardiography preoperatively and then daily post-operatively. The ECG diagnosis criteria for perioperativeacute myocardial ischaemia were 1 or more of thefollowing: new Q waves of more than 0.04 s in territoriesother than those identified before surgery; significant ST



segment deviation (elevation at the J point in two ormore contiguous leads with cut off point no less than0.2 mV in men or no less than 0.15 mV in women in leadsV2 or V3, and no less than 0.1 mV in other leads withprominent R wave or R/S >1). Postoperative myocardialinfarction was considered when new pathological Qwave was accompanied with elevated biomarkers [cTnI(cardiac troponin I) and CK-MB (creatine kinase-MB)] asdescribed in the current expert consensus document re-garding the universal definition of myocardial infarction[28].

Postoperative inotropic support was defined as theuse of dopamine (�5 μg · kg− 1 of body weight · min− 1)with or without adrenaline (0.1 μg · kg− 1 of bodyweight · min− 1) for a duration of 30 min or longer duringthe first 12 h postoperatively. Inotropic therapy wasstandardized to a treatment algorithm to treat a meanradial arterial blood pressure less than 60 mmHg despiteoptimization of preload, afterload and HR (heart rate).Nitroglycerin (0.3–0.5 μg · kg− 1 of body weight · min− 1)was administrated intravenously when the values ofCVP (central venous pressure) or PCWP (pulmonarycapillary wedge pressure) were more than 12 mmHgwith or without concomitant hypotension. Patients wereextubated when they were able to sustain adequatespontaneous respiration and required minimal oxygensupport to maintain arterial blood gas values at normalrange. The patients were then discharged from ICU(intensive care unit) when they were haemodynamicallystable with normal blood gas parameters without the needof inotropic or oxygen support.

Sample collectionCentral venous blood samples were obtained prior toCPB (pre-CPB), 5 min, 1 h, 4 h, 12 h and 24 h afteraortic declamping for the measurements of plasmalevels of cTn I and CK-MB (enzyme immunoassay),h-FABP (heart-type fatty acid-binding protein), the lipidperoxidation product MDA (malondialdehyde), theantioxidant enzyme SOD (superoxide dismutase) activityand levels of pro-inflammatory cytokines TNF-α andIL-6 (interleukin-6). Samples were immediately cooled to4 ◦C and centrifuged at 1000 g for 10 min at 4 ◦C. Plasmawas collected and stored at − 70 ◦C until analysed.

Samples of right atrial tissue were harvested from 33patients (n = 8, 7, 9 and 9, respectively, in groups C, P,I and IP) who gave consent to this specific procedure.Samples were harvested with cold sharp dissectionand handled in a non-traumatic fashion. Double 3-0polypropylene purse-string sutures were placed in theatrial appendage. During placement of the venouscannula, the first sample of atrial appendage was harvested(pre-CPB). The superior suture was tightened to securethe venous cannula. The inferior suture remained looseto allow this portion of the atrium to be perfusedwith blood, exposed to CPB and blood cardioplegia,

and reperfused after removal of the aortic cross-clamp.The second sample of atrial appendage (post-CPB) washarvested after CPB during removal of the venouscannula. Part of the tissue samples were frozen and storedat − 70 ◦C until analysed for protein content by Westernblotting, and part of the tissue was paraffin-embeddedand blocked according to standard procedures to be usedfor immunochemical analysis.

Nitrotyrosine histological immunostainingand assessment of eNOS and iNOS(inducible NO synthase) byimmunostaining and Western blottingAtrial tissue was processed with standard methodsfor immunohistochemical and Western blot analysisas described [29]. In brief, paraffin-embedded atrialtissue blocks were sectioned at 5 μm. The sections weredeparaffinized, rehydrated, treated with target retrievalbuffer (S1699; Dako), blocked with 3 % hydrogenperoxide, washed with PBS and blocked with 5 % normalgoat serum in PBS for 30 min. The slides were thenincubated with primary antibodies in PBS containing 1 %normal goat serum overnight at 4 ◦C.

eNOS and iNOS were stained with monoclonal mouseanti-eNOS and mouse anti-iNOS antibodies (1:300–400 dilutions; Santa Cruz Biotechnology). Secondaryantibody used was goat anti-mouse antibody (1:400dilution; Dako). After incubation with secondaryantibody, the sections were dipped in lithium carbonate(to blue for 90 s), washed in running water, dehydratedin increasing grades of alcohol and cleared in xylenebefore being mounted in resinous mounting mediumwith Permount (Fischer) coverslips. Some sections wereincubated with non-specific mouse IgGs and served asnegative controls.

For nitrotyrosine staining, a monoclonal mouse anti-nitrotyrosine antibody (1:400 dilution; Calbiochem) wasused. Intensity of tissue nitrotyrosine staining was quan-tified by an observer blinded to the experimentalconditions. A series of three randomly chosen imagesof equal surface of nitrotyrosine-stained atrial tissuewere delineated with an operator-controlled cursor.The sections were analysed with an image analyser(IBAS-2000; Kontron) as described [30]. The densityof nitrotyrosine protein expression was expressed inarbitrary units.

In addition, atrial tissue eNOS and iNOS proteincontents were quantified by standard Western blotting asdescribed previously [29], with the same antibodies (SantaCruz Biotechnology) as used for immunostaining. Thesame blots were stripped and reblotted with antibody toβ-actin as internal controls. The images were analysed byan imaging densitometer and quantified by normalizingthem with that from β-actin.



Table 1 Patient characteristicsValues are means+− S.D. or the number of patients. No significant differences were observed betweenthe control group and the mixed group; LVEF, left ventricular ejection fraction; COPD, chronic obstructivepulmonary disease; ACE, angiotensin-converting enzyme.

Group

Variable C P I IP

Gender (males/female) (n) 23/6 25/5 25/5 23/7Age (years) 63.3 +− 8.2 63.1+− 6.4 59.8+− 10.3 62.5+− 7.6Weight (kg) 68.4+− 10.3 66.8+− 11.1 69.4+− 9.3 67.5+− 8.7LVEF (%) 0.53+− 0.11 0.55+− 0.08 0.53+− 0.07 0.54+− 0.09Diabetes (n) 6 5 7 6Hypertension (n) 10 9 8 9COPD (n) 4 3 4 3Preoperative medication

β-Blockers (n) 18 17 17 18Calcium channel blockers (n) 5 6 5 7ACE inhibitors (n) 4 5 4 5Nitrates (n) 17 18 19 18Diuretics (n) 8 9 8 9Insulin (n) 4 3 5 4Oral anticoagulants (n) 3 2 3 3

Biochemical analysish-FABP, a marker of perioperative myocardial damage,has been shown to peak earlier than CK-MB and cTnIduring coronary bypass surgery [31]. Plasma h-FABPwas measured using ELISA kits (Hycult Biotechnology),and CK-MB and cTnI were measured by using aphotometric method (Diasys Diagnostic System) andELISA (Beckman) respectively.

Plasma TNF-α and IL-6 were measured usingELISA kits (R&D Systems). Plasma SOD activity wasdetermined according to the method described by Sunet al. [32]. Plasma levels of MDA were measured bychemical analysis (commercial kits; Nanjing JiangzhengBiological Engine Institute) as described previously [33].

Plasma samples used for biochemical assays werecoded, and the laboratory investigator was blinded inregard to treatment regimen. All samples were assayedin duplicate. Similarly, all haemodynamic data werecollected by trained observers who did not take anauthorship in this study and who were blinded to theanaesthetic regimen used.

Sample size estimationGroup sample size was estimated based on differencesin cTn I concentration measured at 4 h post-CPB ina pilot study of patients who received propofol(5.21 +− 0.95 ng/ml) and midazolam (4.50 +− 0.82 ng/ml)anaesthesia. The formula [34]: n = 15.7/ES2 + 1, whereES = effect size = (difference between groups)/(mean ofthe S.D. between groups), with α = 0.05 and power = 0.8,

was used to determine that the study would be adequatelypowered with n = 26 per group.

Statistical analysisAll continuous data are expressed as means +− S.D.Statistical evaluation of patients’ file and perioperativedata was performed by unpaired Student’s t test or χ2 testwhen appropriate. Between-groups and within-groupdifferences of bioassay data were analysed using two-way ANOVA with repeated measures and Bonferroni’scorrections (GraphPad Prism). Values were consideredstatistically significant when P < 0.05.

RESULTS

Preoperative and intraoperative dataPatient characteristics, preoperative haemodynamic data,preoperative medication, dosage of fentanyl used,duration of CPB and ACC (aortic cross-clamping) weresimilar between groups (Tables 1 and 2). Surgery wascompleted in all patients. In one patient (male, 69 years,weight 73 kg, from group C), a severe right coronarythrombosis was detected during the operation, andconsequently, a right coronary artery endarterectomywas performed in addition to conventional CABG. Thispatient suffered from acute perioperative myocardialinfarction and died of severe right heart failure on day2 after surgery. This patient’s data were excluded fromstatistical analysis.



Table 2 Intra-operative and postoperative procedure characteristics of the patientsValues are means+− S.D. or the number of patients. P values represent the difference among the groups. *Significant difference between the control group and theIP group; CPB, cardiopulmonary bypass; MI, myocardial infarction; NTG, nitroglycerin; SNP, sodium nitroprusside; IABP, intra-aortic balloon pump; NS, not significant.

Group

Variable C (n = 29) P (n = 30) I (n = 30) IP (n = 30) P value

Intra-operative dataCPB time (min) 123.6+− 40.5 128.9+− 38.5 119.6+− 46.6 125.4+− 33.2 NSAortic cross-clamp time (min) 65.4+− 20.3 67.7+− 19.6 68.9+− 24.6 66.3+− 22.5 NSNumber of grafts 3.2+− 0.5 3.1+− 0.7 3.1+− 0.9 3.2+− 0.6 NSRectal temperature ( ◦C) 31.6+− 1.9 32.1+− 2.2 31.8+− 1.5 31.4+− 1.7 NSPatients requiring cardioversion to restore

rhythm after CPB (n)8 5 5 1* <0.05*

Postoperative dataInotropic drug requirement (n)

None (NTG only) 2 5 4 10 <0.05Mild (dopamine �5 μg · kg− 1 of

body weight · min− 1)16 17 19 16 NS

Moderate (dopamine or dobutamine<5 μg · kg− 1 of body weight · min− 1 oradrenaline <0.1 μg · kg− 1 of bodyweight · min− 1)

11 8 7 4 <0.05

Requirement for IABP (n) 2 0 1 0 NSST segment deviation (n) 8 5 6 3* <0.05New pathological Q waves (n) 4 2 2 1 <0.05Duration of ventilation (h) 15.5+− 4.3 12.8+− 5.4 13.1+− 6.5 9.7+− 3.8* <0.05ICU stay (h) 36.7+− 6.2 32.5+− 7.3 34.8+− 5.8 22.7+− 5.1* <0.05*Hospital stay (day) 20.3+− 8.7 17.6+− 9.2 18.5+− 7.5 15.8+− 5.2 0.058Death in hospital (n) 0 0 0 0 NS

Postoperative outcome dataClinical outcome parameters are presented (Tables 2 and3). On day 1, fewer patients in group IP (n = 4) hadsigns of myocardial ischaemia evidenced as ST segmentdeviation and/or new pathological Q waves comparedwith group C (n = 12), group P (n = 7) and group I (n = 8)(P < 0.05). Significantly fewer patients in the IP grouprequired postoperative inotropic support compared withall other groups.

MAP, HR, CVP and PCWP did not differ amonggroups throughout the observation interval (Table 3)except that all the values were transiently lower thanthe corresponding baseline values at the first 5 minof reperfusion. No significant changes of systemicvascular resistance index were observed throughout theobservation interval. CI (cardiac index) was significantlyincreased 1 h after reperfusion and onward in groups P,I and IP compared with baseline or pre-CPB values. Bycontrast, significant improvement in CI did not occurin group C until after 12 h of reperfusion. During earlyreperfusion [1 and 4 h post-ACC; Table 3], the values ofCI in group IP, but not in groups P or I, were significantlyhigher than that in group C. Durations of postoperative

ventilation and ICU stay in group IP, but not in groupsP or I, were significantly shorter than those in group C.All patients (except the one excluded as mentioned above)were discharged from hospital uneventfully. The durationof hospital stay in group IP tended to be shorter than thatin group C, but overall, the difference among groups didnot reach statistical significance (P = 0.058; Table 2).

Postischaemic myocardial cellular injuryand eNOS and nitrotyrosine expressionBaseline and pre-CPB plasma levels of h-FABP, cTnI andCK-MB did not differ among groups, but all significantlyincreased upon reperfusion at 5 min post-ACC (P < 0.01compared with baseline, Figure 2). Plasma levels ofh-FABP (Figure 2A), cTnI (Figure 2C) and CK-MB(Figure 2B) peaked at 1, 4 and 12 h post-ACC respectivelyin all groups. At 1 h post-ACC, h-FABP concentrationin group IP, but not in groups P or I, was significantlylower than that in group C. The h-FABP level in groupIP was also significantly lower than that in groups Pand I respectively at 1 h post-ACC. Similarly, at 4 hpost-ACC, plasma cTnI in group IP was lower thanthose in all other groups. The change of CK-MB largely



Table 3 Perioperative haemodynamic data of the patientsValues are means+− S.D., SVRI, systemic vascular resistances index. n = 29 in group C, n = 30 in other groups. *P < 0.05 or P < 0.01 compared with baseline;�P < 0.05 compared with group C.

Time post-ACC

Variable Group Baseline Pre-CPB 5 min 1 h 4 h 12 h 24 h

HR (beats/min) C 72+− 13 75+− 10 57+− 13 * 85+− 12 87+− 12 83+− 12 78+− 9P 73+− 14 77+− 11 60+− 11 * 75+− 11 78+− 13 81+− 11 75+− 11I 75+− 14 76+− 13 63+− 9 * 81+− 13 80+− 13 79+− 13 74+− 11IP 74+− 15 75+− 10 63+− 8 * 71+− 10� 75+− 9� 78+− 12 77+− 11

MAP (mmHg) C 77.3+− 10.0 75.6+− 11.6 54.7+− 10.2* 73.8 +− 13.7 78.6 +− 13.7 80.2 +− 11.6 73.7 +− 12.8P 75.2+− 9.7 77.3+− 10.2 52.5+− 11.6* 68.6 +− 11.9 74.5 +− 10.8 73.6 +− 16.9 76.5 +− 12.3I 76.6+− 11.3 79.4+− 12.7 56.2+− 8.4* 77.2+− 11.6 76.7 +− 17.6 75.3 +− 13.5 72.8 +− 14.1IP 74.7+− 16.0 77.0+− 14.1 54.7+− 9.6* 73.5+− 10.7 78.2 +− 19.1 75.8 +− 14.6 76.6 +− 10.7

CVP (mmHg) C 7.8 +− 1.2 8.0+− 1.2 4.4 +− 0.7* 6.9+− 0.7 11.3+− 3.1 10.7+− 2.6 11.5+− 3.6P 8.1 +− 2.1 8.5+− 1.8 5.5 +− 0.9* 7.1+− 1.1 10.5+− 2.1 11.8+− 3.8 12.0+− 3.2I 8.2 +− 1.7 8.8+− 1.2 4.8 +− 0.8* 6.8+− 0.9 9.8+− 2.7 10.9+− 3.4 10.7+− 2.3IP 7.9 +− 1.5 8.6+− 1.7 5.1 +− 1.1* 7.0+− 0.6 10.6+− 3.5 11.2+− 2.4 11.4+− 2.8

PCWP (mmHg) C 7.4+− 1.7 8.0+− 2.1 4.9 +− 0.5* 6.5+− 0.4 10.8+− 2.1 11.2+− 3.5 11.6 +− 2.6P 7.7 +− 2.0 7.9+− 1.2 5.0 +− 0.2* 6.8+− 0.3 9.5+− 3.0 10.4+− 3.7 11.3+− 2.4I 7.6 +− 1.4 8.1+− 1.5 4.8 +− 0.7* 6.9+− 0.5 9.4+− 2.8 10.2+− 2.4 10.7+− 3.7IP 7.9 +− 1.8 8.2+− 1.8 5.1 +− 0.8* 6.7+− 0.7 10.0+− 1.3 10.5+− 3.5 10.9+− 2.2

CI (l · min− 1 · m− 2) C 2.1 +− 0.5 2.2+− 0.4 2.3 +− 0.4 2.3+− 0.5 2.3+− 0.5 2.7+− 0.7* 2.5+− 0.4P 1.9 +− 0.3 2.1+− 0.3 2.2 +− 0.3 2.5+− 0.4* 2.5+− 0.4* 2.8+− 0.5* 2.6+− 0.5*I 2.2 +− 0.4 2.1+− 0.5 2.2 +− 0.5 2.6+− 0.3* 2.7+− 0.3* 2.8+− 0.4* 2.7+− 0.2*IP 1.9 +− 0.5 2.2+− 0.4 2.3 +− 0.2 2.8+− 0.3*� 3.1+− 0.3*� 3.0+− 0.3* 2.8+− 0.3*

SVRI (dyn s · cm− 5 · m− 2) C 2216+− 472 1857+− 274 1922+− 302 2172+− 226 2371+− 319 1973+− 227 1852+− 218P 2526+− 383 20365+− 412 2072+− 278 2246+− 307 2138+− 285 1882+− 268 1971+− 273I 2361+− 436 2138+− 263 1875+− 318 2302+− 232 1974+− 318 2051+− 287 1791+− 317IP 2672+− 328 2217+− 305 2093+− 274 2148+− 186 1635+− 206 1923+− 281 1882+− 263

mirrored that of cTnI during early reperfusion up to 12 hpost-ACC. After 24 h post-ACC, plasma levels ofh-FABP reduced to baseline values in all groups, whereasplasma cTnI and CK-MB levels remained significantlyincreased in groups C and P relative to their baselinevalues and were significantly higher than that in groupIP. Plasma values of h-FABP in group P at 12 h post-ACC and values of cTnI in groups P and I as well asvalues of CK-MB in group I at 24 h post-ACC wererespectively lower than that in group C, indicative ofmarginal cellular protective effects of the propofol orisoflurane anaesthesia regimen used in the current study.At 24 h post-ACC, values of cTnI and CK-MB in groupIP were significantly lower than that in group P. Thecalculated values of AUC (area under curve) for plasmacTnI (Figure 2D) revealed that AUC values in group P(14.1 +− 0.7) and group I (13.7 +− 0.5) were all marginallybut significantly lower than that in group C (15.7 +− 1.1)(P < 0.05 or P < 0.01). Of note, the AUC value in groupIP (9.2 +− 0.5) was much lower than those in all the othergroups (P < 0.001 compared with groups C, P or I).AUC values of cTnI did not differ significantly between

groups P and I (P>0.2). Similarly, AUC for CK-MBin group IP (229.1 +− 10.2) was much lower than thosein group C (292.9 +− 20.2), group P (277.5 +− 18.9) andgroup I (265.7 +− 8.7) (all P < 0.01), while the differencesin CK-MB among groups P, I and C were marginal(P < 0.05, group I compared with group C; P>0.05 groupP compared with C).

Atrial tissue eNOS protein expression did not differamong groups pre-CPB (results not shown). After CPB,eNOS expression in blood vessels could hardly be seenin group C (Figure 3A), but it was visible in group I(Figure 3B) and group P (Figures 3C). Strongest eNOSexpressions in vessels were seen in group IP (Figure 3D).Myocardial tissue eNOS expression could be seen inall groups, but it was weakest in group C and wasstrongest in group IP. The total myocardial eNOS proteincontent, as measured by Western blotting in group IP,was significantly higher than that in both groups C and P(Figure 3E). Myocardial eNOS protein content in groupsI and P were relatively higher than that in group C, andthe difference reached statistical significance in group Ibut not group P. Myocardial nitrotyrosine expression was



Figure 2 Perioperative variations of cardiac-specific biomarkers h-FABP (A), CK-MB (B) and cTnI (C) in plasma as well ascalculated values of AUC of cTnI (D)Note that plasma levels of h-FABP, cTnI and CK-MB peaked at 1, 4 and 12 h after reperfusion respectively. Values are means+− S.D., n = 29 in group C, n = 30in other groups. *P < 0.05 or P < 0.01 compared with baseline; �P < 0.05 and ��P < 0.01 compared with group C; #P < 0.05 and ##P < 0.01 comparedwith group P; $P < 0.05 and $$P < 0.01 compared with group I.

marginally detectable before CPB but with no apparentdifference among groups (results not shown). However,after CPB, protein expression of nitrotyrosine, an indexof nitrative stress, was highest in group C (Figure 4A)and was relatively lower in group P (Figure 4B), groupI (Figure 4C), and was further lower in group IP(Figures 4D and 4E). The myocardial protein expressionof iNOS was not detectable both pre- and post-CPB,when measured by immunostaining or Western blotting.

Pro-inflammatory cytokines TNF, IL-6 andantioxidant enzyme SOD and MDABaseline levels of TNF-α, IL-6, SOD and MDA (Table 4)did not differ among groups. Plasma levels of TNF-αand IL-6 significantly increased immediately uponreperfusion at 5 min post-ACC in all groups, which

preceded the significant decrease in SOD activity andincrease in MDA content that occurred relatively laterduring reperfusion at 1 h post-ACC and onward.TNF-α levels in group IP were significantly lower thanthose in all other groups at 1, 4 and 12 h post-ACC(Table 4). At 24 h post-ACC, values of TNF-α in groupIP but not in group C or groups P and I returned tobaseline values. The combined isoflurane and propofolregimen also significantly reduced plasma levels of IL-6 at 4 and 12 h post-ACC compared with group C orgroups P and I. Plasma MDA content in group IP wassignificantly lower than all other groups at 1 and 4 h post-ACC. During early reperfusion, plasma SOD activitiesin groups IP, P and I tended to be higher than that inthe control group, but the difference did reach statisticalsignificance.



Figure 3 Myocardial (atrium) eNOS expression status afterCPB in patientsRepresentative staining from the control (group C) (A), isoflurane (group I) (B),propofol (group P) (C) and isoflurane plus propofol (group IP) (D) groups. Solidarrows point to vascular vessels, and dashed arrows point to cardiac tissue. eNOSexpression in the vessels was hardly detected in the control group but was easilyseen in the isoflurane and propofol group. Strongest eNOS expression in vesselswas seen in the isoflurane plus propofol group. Tissue eNOS expression could beseen in all groups, but it was weakest in the control group and strongest in theisoflurane plus propofol group. (E) Content of myocardial eNOS protein expressionmeasured by Western blotting. Values are means+− S.D., n = 7–9 per group.*P < 0.05 and **P < 0.01 compared with group C. #P < 0.05 compared withgroup P.

DISCUSSION

The salient finding of the present clinical study is thatthe application of IsoPC in combination with propofoltreatment and postconditioning during ischaemia andearly reperfusion was superior to either an isofluraneor propofol anaesthesia regimen alone in reducingindices of oxidant stress, pro-inflammatory cytokines andmyocardial injury. Synergy of isoflurane and propofolalso reduced the need for inotrope support compared

Figure 4 Atrium tissue nitrotyrosine expression measuredby immunochemical staining (stained in brown as shown byarrow)Representative staining (n = 7–9 per group) from control (A), isoflurane (groupI) (B), propofol (group P) (C), and isoflurane plus propofol (group IP) (D) groups.(E) During reperfusion, protein expression of nitrotyrosine, an index of nitrativestress, was high in group C, and was relatively lower in groups P and I and wasfurther lower and hardly visible in group IP. Data are means+− S.D., n = 7–9per group. *P < 0.05 and **P < 0.01 compared with group C; $P < 0.05compared with group I.

with propofol or isoflurane alone in the present study.The isoflurane or propofol anaesthesia regimen alsoappeared to be moderately cardioprotective in that weobserved lower plasma levels of cTnI and/or CK-MBduring late reperfusion in these groups relative to thecontrol group. However, the marginal attenuation ofpostischaemic myocardial cellular injury conferred byisoflurane or propofol did not translate into immediateclinical and haemodynamic benefit. By contrast, IsoPCand propofol postconditioning significantly enhancedpostischaemic cardiac index and shortened ICU stay,which was coincident with profound reductions in plasmalevels of TNF-α and MDA content as well as reductionin myocardial levels of nitrotyrosine expression.

Our present finding that volatile anaesthetic pre-conditioning combined with intravenous anaesthetic



Table 4 Variations of perioperative plasma TNF-α, IL-6, SOD and MDAValues are means+− S.D., n = 29 in group C and n = 30 in the other groups. *P < 0.05 or P < 0.01 compared with baseline; �P < 0.05 compared with groupC; #P < 0.05 compared with group P; �P < 0.05 compared with group I.

Time post-ACC

Variable Group Baseline Pre-CPB 5 min 1 h 4 h 12 h 24 h

TNF-α C 2.62+− 0.49 2.74 +− 0.44 5.87+− 0.82* 9.38+− 1.12* 13.49+− 2.87* 10.33+− 1.76* 4.04+− 0.68*(pg/ml) P 2.47+− 0.59 2.52 +− 0.56 5.21+− 0.87* 8.27+− 1.06* 12.39+− 3.18* 10.06+− 2.49* 3.87+− 0.32*

I 2.28+− 0.65 2.44 +− 0.42 5.27+− 0.75* 8.47+− 0.91* 12.17+− 3.06* 11.07+− 2.88* 3.42+− 0.48*IP 2.40+− 0.40 2.42 +− 0.40 4.91+− 0.73* 5.82+− 0.80*�# 9.01+− 1.11*�#� 7.45+− 1.77*�#� 3.08+− 0.57�

IL-6 C 3.53+− 0.41 4.03 +− 0.52 15.12+− 3.19* 18.73+− 5.53*� 27.09+− 8.86* 15.55+− 5.43* 4.50+− 1.18(pg/ml) P 3.02+− 0.44 3.99 +− 0.45 14.55+− 4.02* 17.47+− 5.60* 24.84+− 6.55* 13.22+− 4.20* 4.25+− 0.65

I 3.16+− 0.45 3.64 +− 0.97 14.57+− 5.16* 17.61+− 4.33* 25.63+− 7.45* 15.25+− 4.69* 4.09+− 0.66IP 3.34+− 0.36 4.11 +− 0.38 14.20+− 6.04* 16.62+− 5.39* 17.67+− 4.04*�#� 9.22+− 2.41*�#� 4.11+− 0.60

SOD C 110.1+− 13.4 97.1 +− 10.2 82.3+− 12.2 69.2+− 15.3* 61.5+− 10.2* 84.2+− 15.2*� 98.3+− 13.4(μm/l) P 103.3+− 11.41 91.4 +− 13.28 83.2+− 11.1 78.5+− 13.5* 72.2+− 15.3* 89.4+− 13.4 99.1+− 12.3

I 98.4+− 15.4 94.6 +− 9.83 88.5+− 13.2 80.2+− 11.5* 74.3+− 13.1* 87.3+− 11.2 95.6+− 14.3IP 113.6+− 13.5 96.2 +− 11.4 91.5+− 14.3 85.1+− 9.8* 82.8+− 14.3* 94.6+− 11.6 109.2+− 12.8

MDA C 5.36+− 0.83 5.83 +− 0.47 6.85+− 1.03* 9.88+− 0.84* 10.03+− 0.62* 8.54+− 0.81* 6.47+− 1.03(μm/l) P 5.76+− 0.34 6.04 +− 0.81 6.22+− 0.63 7.71+− 0.74*� 8.46+− 0.94*� 7.48+− 0.61* 6.26+− 0.48

I 5.86+− 0.52 6.24 +− 0.73 6.37+− 0.58 7.36+− 0.81*� 8.55+− 0.84*� 7.26+− 0.51* 6.35+− 0.57IP 5.56+− 0.47 6.17 +− 0.62 5.88+− 0.47� 6.25+− 0.41*�#� 7.12+− 0.46*�#� 6.13+− 0.81� 6.07+− 0.62

postconditioning offers synergistic or additional benefitin patients is novel and clinically important. AlthoughIsoPC can provide moderate [1] to minor [2] cardiacprotection during CABG, the effectiveness of volatileanaesthetic preconditioning is optimal when the durationof index ischaemia is between 30 and 35 min [3,35].In the current study, the average duration of ACCexceeded 60 min in all groups, which is well beyond therange during which volatile anaesthetic preconditioningalone would have been effective [3]. On the otherhand, a previous study shows that volatile anaestheticsevoflurane preconditioning combined with sevofluranepostconditioning offers no additional benefit overpreconditioning or postconditioning alone [36]. Similarly,ischaemic postconditioning has been shown not tooffer additional protection in ischaemic preconditionedhearts [37]. The lack of synergy between classic oranaesthetic pre- and post-conditioning is not surprising.Ischaemic as well as volatile anaesthetic post- and pre-conditioning share similar mechanisms [38] that rely onthe generation of small amount of ROS (in particular,superoxide anion) as an early trigger of protection [6,39].Therefore postconditioning may not confer additionalbenefit when the stimuli applied are the same andat their individual maximal strength (e.g. multipleischaemic preconditioning or high concentrations ofvolatile anaesthetic conditioning) [40]. However, thevolatile anaesthetic sevoflurane, when applied at relativelylower concentrations has been shown to decrease thetime threshold of classic ischaemic preconditioning [41].Furthermore, additive benefit by combining volatile

anaesthetic pre- and post-conditioning could be achievedin experimental settings when the anaesthetic was usedat relatively lower concentrations [42]. This suggeststhat synergy of anaesthetic pre- and post-conditioningis achievable depending on how it is modulated.

The intravenous anaesthetic propofol has antioxidantproperties [11], which may confer cardioprotectionvia mechanisms that differs from that of volatileanaesthetics. Propofol is a scavenger of the potent oxidantONOO− (peroxynitrite) (a product of superoxide anionand NO reaction) [43,44]. ONOO− at low levels iscardioprotective [45,46] and can serve as a trigger ofischaemic preconditioning [47]. However, ONOO−

is detrimental to the heart or cardiomyocytes whenits production is increased during reperfusion [48–50].Hypoxic postconditioning-mediated attenuation ofcardiomyocyte death following reoxygenation has beenshown to be attributable to reduction of ONOO−

formation during reoxygenation [48]. Therefore theprotection against myocardial [19] as well as cerebral[20] IRI conferred by propofol postconditioning mightbe attributable to propofol’s ONOO− -scavengingproperty. In the present study, the preconditioningeffects of isoflurane (which generates small amount ofsuperoxide anion, NO and ONOO− before ischaemiaas early triggers that lead to reduced superoxideanion and ONOO− production during reperfusion)and the postconditioning effects of propofol (whichscavenges superoxide anion and ONOO− ) synergized toproduce enhanced cardioprotection. Indeed, during earlyreperfusion, atrial tissue protein levels of nitrotyrosine,



a footprint of ONOO− production, were much lowerin group IP relative to groups I or P. This wasaccompanied by a profound reduction in plasma levelsof MDA, an end product of an oxidant-mediatedincrease in lipid peroxidation. Synergy of IsoPC andpropofol postconditioning may have been achieved viathe activation of the PI3K/AKT pathway, indirectlyevidenced by the significantly elevated myocardial (atrial)levels of eNOS, a critical downstream mediator ofthe prosurvival PI3K/Akt pathway [25]. It shouldbe noted, we measured MDA using the conventionalspectrophotometric procedure based on absorbance ofthe thiobarbituric acid–MDA complex at 532 nm. Thismethodology is not highly specific to MDA and is proneto the interference of other non-lipid-related compounds.Therefore an HPLC-based assay of MDA [51] would besuperior and should be applied in future related studiesto confirm the finding the present study.

Myocardial necrosis can be recognized by theappearance in the blood of cardiac-specific proteinsreleased into circulation due to cardiomyocyte damage.The biochemical markers cTnI and CK-MB have high,to nearly absolute, myocardial tissue specificity. Thesignificantly lower levels of plasma troponin I and CK-MB seen in group IP are indicative of reduced myocardialcellular injury. In particular, the much significantly lowervalues of cTnI AUC value in group IP (∼40 % lowerthan that in group C) were in contrast with the marginaldifferences in group P and I (both were about 10 %lower than that in group C), which indicates synergisticeffects of the treatments in attenuating myocardial cellulardamage. The h-FABP is a rapid marker of perioperativemyocardial damage and peaks earlier than CK-MBor cTnI [31]. Furthermore, a previous study showsthat h-FABP predicts long-term mortality after acutecoronary syndrome and identifies high-risk patientsacross the range of troponin I values [52]. In our presentstudy, plasma h-FABP peaked at the first hour duringreperfusion and was significantly reduced by the IsoPCand propofol postconditioning regimen at this early stageof reperfusion. IsoPC and propofol postconditioningsignificantly reduced cTnI release at a relatively laterphase of reperfusion. Our results suggest that h-FABPcan serve as an early predictor of the anaestheticcardioprotective effects during cardiac surgery.

Although there were significant intergroup differencesin plasma levels of cTnI and h-FABP as well as indurations of ICU and hospital stay, the patients were notfollowed up after discharge. Further larger studies aremerited to determine the long-term clinical relevance ofthe changes in biomedical markers observed in the relatedstudy. The paucity of the available atrial tissue sample didnot allow us to pursue more detailed study regarding theactivation status of eNOS, nor the protein levels of PI3K,Akt and nitrotyrosine to identify/confirm the signallingpathway of the protection. Nevertheless, the results from

the present study are promising and provide clear cluesfor future in depth study.

In conclusion, we have shown that IsoPC, combinedwith propofol postconditioning, act synergistically inattenuating postischaemic myocardial reperfusion injuryas determined by the surrogate markers of myocardialinjury and function. Further large-scale and long-termstudies are required to confirm the clinical benefitof this novel anaesthesia regimen, which may offera promising therapeutic approach to cardioprotectionpatients undergoing cardiac surgery.

AUTHOR CONTRIBUTION

Zhengyuan Xia designed the research. Zhiyong Huang,Xingwu Zhong, Shangyi Ji, Gordon Wong, Zhong-yuanXia and Zhengyuan Xia performed the research. ZhiyongHuang, Michael Irwin, Yanan Liu and Barry Finegananalysed the data, and Zhiyong Huang and ZhengyuanXia wrote the paper.

FUNDING

This study is supported, in part, by the National NaturalScience Foundation of China (NSFC) [grant number30872447, 30672033] and by a Society of CardiovascularAnesthesiologists (SCA) Foundation Starter grant (toZ.X.).

REFERENCES

1 Belhomme, D., Peynet, J., Louzy, M., Launay, J. M.,Kitakaze, M. and Menasche, P. (1999) Evidence forpreconditioning by isoflurane in coronary artery bypassgraft surgery. Circulation 100, II340–II344

2 Wang, X., Jarvinen, O., Kuukasjarvi, P., Laurikka, J.,Wei, M., Rinne, T., Honkonen, E. L. and Tarkka, M.(2004) Isoflurane produces only minor preconditioning incoronary artery bypass grafting. Scand. Cardiovasc. J. 38,287–292

3 Kevin, L. G., Katz, P., Camara, A. K., Novalija, E.,Riess, M. L. and Stowe, D. F. (2003) Anestheticpreconditioning: effects on latency to ischemic injury inisolated hearts. Anesthesiology 99, 385–391

4 Mio, Y., Bienengraeber, M. W., Marinovic, J., Gutterman,D. D., Rakic, M., Bosnjak, Z. J. and Stadnicka, A. (2008)Age-related attenuation of isoflurane preconditioning inhuman atrial cardiomyocytes: roles for mitochondrialrespiration and sarcolemmal adenosine triphosphate-sensitive potassium channel activity. Anesthesiology 108,612–620

5 Tanaka, K., Kehl, F., Gu, W., Krolikowski, J. G., Pagel, P.S., Warltier, D. C. and Kersten, J. R. (2002) Isoflurane-induced preconditioning is attenuated by diabetes. Am. J.Physiol. Heart Circ. Physiol. 282, H2018–H2023

6 Kevin, L. G., Novalija, E., Riess, M. L., Camara, A. K.,Rhodes, S. S. and Stowe, D. F. (2003) Sevofluraneexposure generates superoxide but leads to decreasedsuperoxide during ischemia and reperfusion in isolatedhearts. Anesth. Analg. 96, 949–955



7 Clermont, G., Vergely, C., Jazayeri, S., Lahet, J. J.,Goudeau, J. J., Lecour, S., David, M., Rochette, L. andGirard, C. (2002) Systemic free radical activation is amajor event involved in myocardial oxidative stress relatedto cardiopulmonary bypass. Anesthesiology 96, 80–87

8 Ansley, D. M., Xia, Z. and Dhaliwal, B. S. (2003) Therelationship between plasma free 15-F2t-isoprostaneconcentration and early postoperative cardiac depressionfollowing warm heart surgery. J. Thorac. Cardiovasc.Surg. 126, 1222–1223

9 Borgermann, J., Friedrich, I., Flohe, S., Spillner, J.,Majetschak, M., Kuss, O., Sablotzki, A., Feldt, T.,Reidemeister, J. C. and Schade, F. U. (2002) Tumornecrosis factor-α production in whole blood aftercardiopulmonary bypass: downregulation caused bycirculating cytokine-inhibitory activities. J. Thorac.Cardiovasc. Surg. 124, 608–617

10 Zhong, C., Zhou, Y. and Liu, H. (2004) Nuclear factor κBand anesthetic preconditioning during myocardialischemia-reperfusion. Anesthesiology 100, 540–546

11 Murphy, P. G., Myers, D. S., Davies, M. J., Webster, N. R.and Jones, J. G. (1992) The antioxidant potential ofpropofol (2,6-diisopropylphenol). Br. J. Anaesth. 68,613–618

12 Xia, Z., Godin, D. V. and Ansley, D. M. (2003) Propofolenhances ischemic tolerance of middle-aged rat hearts:effects on 15-F2t-isoprostane formation and tissueantioxidant capacity. Cardiovasc. Res. 59, 113–121

13 Xia, Z., Godin, D. V., Chang, T. K. and Ansley, D. M.(2003) Dose-dependent protection of cardiac function bypropofol during ischemia and early reperfusion in rats:effects on 15-F2t-isoprostane formation. Can. J. Physiol.Pharmacol. 81, 14–21

14 Ko, S. H., Yu, C. W., Lee, S. K., Choe, H., Chung, M. J.,Kwak, Y. G., Chae, S. W. and Song, H. S. (1997) Propofolattenuates ischemia–reperfusion injury in the isolated ratheart. Anesth. Analg. 85, 719–724

15 Kokita, N., Hara, A., Abiko, Y., Arakawa, J., Hashizume,H. and Namiki, A. (1998) Propofol improves functionaland metabolic recovery in ischemic reperfused isolated rathearts. Anesth. Analg. 86, 252–258

16 Xia, Z., Godin, D. V. and Ansley, D. M. (2004)Application of high-dose propofol during ischemiaimproves postischemic function of rat hearts: effects ontissue antioxidant capacity. Can. J. Physiol. Pharmacol. 82,919–926

17 Luo, T. and Xia, Z. (2006) A small dose of hydrogenperoxide enhances tumor necrosis factor-α toxicity ininducing human vascular endothelial cell apoptosis:reversal with propofol. Anesth. Analg. 103, 110–116

18 Kokita, N. and Hara, A. (1996) Propofol attenuateshydrogen peroxide-induced mechanical and metabolicderangements in the isolated rat heart. Anesthesiology 84,117–127

19 He, W., Zhang, F. J., Wang, S. P., Chen, G., Chen, C. C.and Yan, M. (2008) Postconditioning of sevoflurane andpropofol is associated with mitochondrial permeabilitytransition pore. J. Zhejiang Univ. Sci. B 9, 100–108

20 Wang, H. Y., Wang, G. L., Yu, Y. H. and Wang, Y. (2009)The role of phosphoinositide-3-kinase/Akt pathway inpropofol-induced postconditioning against focal cerebralischemia–reperfusion injury in rats. Brain Res. 1297,177–184

21 Wang, B., Shravah, J., Luo, H., Raedschelders, K., Chen,D. D. and Ansley, D. M. (2009) Propofol protects againsthydrogen peroxide-induced injury in cardiac H9c2 cellsvia Akt activation and Bcl-2 up-regulation. Biochem.Biophys. Res. Commun. 389, 105–111

22 Lim, K. H., Halestrap, A. P., Angelini, G. D. andSuleiman, M. S. (2005) Propofol is cardioprotective in aclinically relevant model of normothermic bloodcardioplegic arrest and cardiopulmonary bypass. Exp.Biol. Med. 230, 413–420

23 Xia, Z., Huang, Z. and Ansley, D. M. (2006) Large-dosepropofol during cardiopulmonary bypass decreasesbiochemical markers of myocardial injury in coronarysurgery patients: a comparison with isoflurane. Anesth.Analg. 103, 527–532

24 Flier, S., Post, J., Concepcion, A. N., Kappen, T. H.,Kalkman, C. J. and Buhre, W. F. (2010) Influence ofpropofol-opioid vs isoflurane-opioid anaesthesia onpostoperative troponin release in patients undergoingcoronary artery bypass grafting. Br. J. Anaesth. 105,122–130

25 Tsang, A., Hausenloy, D. J. and Yellon, D. M. (2005)Myocardial postconditioning: reperfusion injury revisited.Am. J. Physiol. Heart Circ. Physiol. 289, H2–H7

26 Amour, J., Brzezinska, A. K., Weihrauch, D., Billstrom,A. R., Zielonka, J., Krolikowski, J. G., Bienengraeber,M.W., Warltier, D. C., Pratt, Jr, P. F. and Kersten, J. R.(2009) Role of heat shock protein 90 and endothelial nitricoxide synthase during early anesthetic and ischemicpreconditioning. Anesthesiology 110, 317–325

27 Ge, Z. D., Pravdic, D., Bienengraeber, M., Pratt, Jr, P. F.,Auchampach, J. A., Gross, G. J., Kersten, J. R. andWarltier, D. C. (2010) Isoflurane postconditioningprotects against reperfusion injury by preventingmitochondrial permeability transition by an endothelialnitric oxide synthase-dependent mechanism.Anesthesiology 112, 73–85

27a Xia, Z., Huang, Z., Ji, S. Y., Xia, Z. Y. and Irwin, M. G.(2009) Isoflurane and propofol synergy in reducingmyocardial ischemia-reperfusion injury in patients.Anesthesiology 111, A285

28 Thygesen, K., Alpert, J. S. and White, H. D. (2007)Universal definition of myocardial infarction. J. Am. Coll.Cardiol. 50, 2173–2195

29 Xia, Z., Kuo, K. H., Nagareddy, P. R., Wang, F., Guo, Z.,Guo, T., Jiang, J. and McNeill, J. H. (2007)N-acetylcysteine attenuates PKCβ2 overexpression andmyocardial hypertrophy in streptozotocin-induceddiabetic rats. Cardiovasc. Res. 73, 770–782

30 Xia, Z., Luo, T., Liu, H. M., Wang, F., Xia, Z. Y., Irwin,M. G. and Vanhoutte, P. M. (2010) l-Arginine enhancesnitrative stress and exacerbates tumor necrosis factor-αtoxicity to human endothelial cells in culture: preventionby propofol. J. Cardiovasc. Pharmacol. 55, 358–367

31 Petzold, T., Feindt, P., Sunderdiek, U., Boeken, U.,Fischer, Y. and Gams, E. (2001) Heart-type fatty acidbinding protein (hFABP) in the diagnosis of myocardialdamage in coronary artery bypass grafting. Eur. J.Cardiothorac. Surg. 19, 859–864

32 Sun, Y., Oberley, L. W. and Li, Y. (1988) A simple methodfor clinical assay of superoxide dismutase. Clin. Chem. 34,497–500

33 Wasowicz, W., Neve, J. and Peretz, A. (1993) Optimizedsteps in fluorometric determination of thiobarbituricacid-reactive substances in serum: importance ofextraction pH and influence of sample preservation andstorage. Clin. Chem. 39, 2522–2526

34 Lerman, J. (1996) Study design in clinical research: samplesize estimation and power analysis. Can. J. Anaesth. 43,184–191

35 De Hert, S. G., Van Der Linden, P. J., Cromheecke, S.,Meeus, R., ten Broecke, P. W., De Blier, I. G., Stockman,B. A. and Rodrigus, I. E. (2004) Choice of primaryanesthetic regimen can influence intensive care unit lengthof stay after coronary surgery with cardiopulmonarybypass. Anesthesiology 101, 9–20

36 Deyhimy, D. I., Fleming, N. W., Brodkin, I. G. and Liu,H. (2007) Anesthetic preconditioning combined withpostconditioning offers no additional benefit overpreconditioning or postconditioning alone. Anesth.Analg. 105, 316–324

37 Manintveld, O. C., Hekkert, M. L., Van Der Ploeg, N. T.,Verdouw, P. D. and Duncker, D. J. (2009) Interactionbetween pre- and postconditioning in the in vivo ratheart. Exp. Biol. Med. 234, 1345–1354



38 Crisostomo, P. R., Wairiuko, G. M., Wang, M., Tsai, B.M., Morrell, E. D. and Meldrum, D. R. (2006)Preconditioning versus postconditioning: mechanismsand therapeutic potentials. J. Am. Coll. Surg. 202, 797–812

39 Lebuffe, G., Schumacker, P. T., Shao, Z. H., Anderson, T.,Iwase, H. and Vanden Hoek, T. L. (2003) ROS and NOtrigger early preconditioning: relationship tomitochondrial KATP channel. Am. J. Physiol. Heart Circ.Physiol. 284, H299–H308

40 Pagel, P. S. (2007) When lack of addition really does addup. Anesth. Analg. 105, 1863

41 Toller, W. G., Kersten, J. R., Pagel, P. S., Hettrick, D. A.and Warltier, D. C. (1999) Sevoflurane reducesmyocardial infarct size and decreases the time thresholdfor ischemic preconditioning in dogs. Anesthesiology 91,1437–1446

42 Obal, D., Dettwiler, S., Favoccia, C., Scharbatke, H.,Preckel, B. and Schlack, W. (2005) The influence ofmitochondrial KATP-channels in the cardioprotectionof preconditioning and postconditioning bysevoflurane in the rat in vivo. Anesth. Analg. 101,1252–1260

43 Mouithys-Mickalad, A., Hans, P., Deby-Dupont, G.,Hoebeke, M., Deby, C. and Lamy, M. (1998)Propofol reacts with peroxynitrite to form a phenoxylradical: demonstration by electron spin resonance.Biochem. Biophys. Res. Commun. 249, 833–837

44 Kahraman, S. and Demiryurek, A. T. (1997)Propofol is a peroxynitrite scavenger. Anesth. Analg. 84,1127–1129

45 Novalija, E., Varadarajan, S. G., Camara, A. K., An, J.,Chen, Q., Riess, M. L., Hogg, N. and Stowe, D. F. (2002)Anesthetic preconditioning: triggering role ofreactive oxygen and nitrogen species in isolated hearts.Am. J. Physiol. Heart Circ. Physiol. 283, H44–H52

46 Lefer, D. J., Scalia, R., Campbell, B., Nossuli, T.,Hayward, R., Salamon, M., Grayson, J. and Lefer, A. M.(1997) Peroxynitrite inhibits leukocyte–endothelial cellinteractions and protects against ischemia–reperfusioninjury in rats. J. Clin. Invest. 99, 684–691

47 Altug, S., Demiryurek, A. T., Kane, K. A. and Kanzik, I.(2000) Evidence for the involvement of peroxynitrite inischaemic preconditioning in rat isolated hearts.Br. J. Pharmacol. 130, 125–131

48 Wang, H. C., Zhang, H. F., Guo, W. Y., Su, H., Zhang,K. R., Li, Q. X., Yan, W., Ma, X. L., Lopez, B. L.,Christopher, T. A. and Gao, F. (2006) Hypoxicpostconditioning enhances the survival and inhibitsapoptosis of cardiomyocytes following reoxygenation:role of peroxynitrite formation. Apoptosis 11, 1453–1460

49 Wang, X. L., Liu, H. R., Tao, L., Liang, F., Yan, L., Zhao,R. R., Lopez, B. L., Christopher, T. A. and Ma, X.L.(2007) Role of iNOS-derived reactive nitrogen species andresultant nitrative stress in leukocytes-inducedcardiomyocyte apoptosis after myocardialischemia/reperfusion. Apoptosis 12, 1209–1217

50 Lopez, B. L., Liu, G. L., Christopher, T. A. and Ma, X. L.(1997) Peroxynitrite, the product of nitric oxide andsuperoxide, causes myocardial injury in the isolatedperfused rat heart. Coron. Artery Dis. 8, 149–153

51 Bird, R. P., Hung, S. S., Hadley, M. and Draper, H. H.(1983) Determination of malonaldehyde in biologicalmaterials by high-pressure liquid chromatography. Anal.Biochem. 128, 240–244

52 Kilcullen, N., Viswanathan, K., Das, R., Morrell, C.,Farrin, A., Barth, J. H. and Hall, A. S. (2007) Heart-typefatty acid-binding protein predicts long-term mortalityafter acute coronary syndrome and identifies high-riskpatients across the range of troponin values. J. Am. Coll.Cardiol. 50, 2061–2067

Received 25 August 2010/16 December 2010; accepted 3 February 2011Published as Immediate Publication 3 February 2011, doi:10.1042/CS20100435


synergy of isoﬂurane preconditioning and propofol ... · china, and department of anesthesiology...

Documents