Reduction of myocardial infarction after emergency coronary artery bypass grafting for failed coronary angioplasty with use of a normothermic reperfusion cardioplegia protocol

Reduction of myocardial infarction after emergency coronary artery bypass grafting for failed coronary angioplasty with use of a normothermic reperfusion cardioplegia protocol

J THORAC CARDIOVASC SURG 1991;101:1069-75 Reduction of myocardial infarction after emergency coronary artery bypass grafting for failed coronary angi...

810KB Sizes 0 Downloads 20 Views

J THORAC CARDIOVASC SURG 1991;101:1069-75

Reduction of myocardial infarction after emergency coronary artery bypass grafting for failed coronary angioplasty with use of a normothermic reperfusion cardioplegia protocol Emergency coronary artery bypass grafting is necessary in 2.7% to 13.S% of patients undergoing elective percutaneous transluminal coronary angioplasty. Myocardial infarction develops in 11 % to 49% of these patients, with 18% to 46% of infarcts resulting in new Q waves. Since February 1987 a revised protocol for myocardial preservation bas been used in 19 patients undergoing emergency bypass grafting for failed angioPlasty. Cardioplegia is induced with a normothermic blood cardioplegic solution. Multiple maintenance doses of cold (40 C) blood cardioplegic solution are then delivered through the aortic root and vein grafts. Before the aortic crossclamp is removed, normothermic reperfusion cardioplegic solution is delivered through the aortic root and vein grafts. This group was compared with all patients undergoing emergency bypass grafting for failed angioplasty before February 1987. These 4S patients received cold induction of cardioplegic solution, multiple maintenance doses of cold cardioplegic solution, and no reperfusion cardioplegic solution. The prevalence of myocardial infarction in the group receiving cold cardioplegic solution was 6S % versus 26 % in the group receiving normothermic cardioplegic solution (p < 0.007). Multivariate analysis identified the use of the normothermic cardioplegia protocol (p < 0.005), nontotaI occlusion of the angioplasty vessel (p < 0.03), and presence of collateral flow to the angioplasty vessel (p < 0.04) as being independently associated with absence of myocardial infarction.

Randy K. Bottner, MD, FACC, Robert B. Wallace, MD, FACS, Marc S. Visner, MD, FACC, FACS, Karen S. Stark, MD, FACC, Evelyn Recientes, BS, Nevin M. Katz, MD, FACC, FACS, Richard A. Hopkins, MD, GeoffreyA. Patrissi, MA, and Kenneth M. Kent, MD, PhD, FACC, Washington, D.C.

Acute complications resulting in compromised antegrade flow necessitate emergencycoronaryartery bypass grafting in 2.7% to 13.5% of patients undergoing percutaneous transluminal coronary angioplasty.!? Myocardial infarction develops in 11% to 49% of these patients, with 18% to 46% of infarcts resulting in new Q

From the Departments ofMedicine and Surgery, Georgetown University Hospital, Washington, D.C. Received for publication Jan. 31, 1990. Accepted for publication July 25, 1990. Address for reprints: Randy K. Bottner, MD, FACC, Department of Cardiology, USAF Medical Center Keesler, Keesler AFB, MS 39534.

12/1/24580

waves. I, 4-8 Mortality associated with emergency bypass grafting for failed angioplasty ranges from 0% to 8%.1-5,7-9 Experimental evidence suggests that, in the setting of acute myocardial ischemia, perfusion with a normothermic rather than a cold cardioplegic solution immediatelyon aortic crossclamping (induction)and just before removing the aortic crossclamp may limit the extent of myocardial injury and necrosis.P' II In this report we present our experienceusing these techniques for preservation and reperfusion of the jeopardized myocardium in 19consecutive patients whounderwent emergencybypassgrafting for failed angioplasty. This is compared with our previous experience with 45 consecutive patients who, after failed angioplasty, underwent emergency bypass grafting with multidose cold cardioplegia without reperfusion cardioplegia. 1069

The Journal.of Thoracic and Cardiovascular Surgery

10 7 0 Bottner et al.

Table I. Cold induction cardioplegic solution* Dextrose (5% in 0.2% NaCI, USP). Tromethamine (0.3 moljL) Citrate, phosphate, dextrose solution (USP) Citric acid (300 mg) Sodium citrate (2.63 gm) Sodium phosphate (222 mg) Dextrose (1.28 gm) Osmolality (438 mOsm/L) Potassium chloride (2 mEqjml) Temperature

550 ml 200ml 50ml

30 m! 4° C

'Cardioplegie solution is mixed with blood (I :4) to yield final cardioplegic infusionsolution.

Table II. Normothermic induction cardioplegic solution* Dextrose (5% in water) Tromethamine (OJ mol/L) Citrate, phosphate, dextrose solution (USP) Citric acid (300 mg) Sodium citrate (2.63 gm) Sodium phosphate (222 mg) Dextrose (1.28 gm) Osmolality (438 mOsmjL) Dextrose (50% in water) Potassium chloride (2 mEqjml) Temperature

470 m! 225 ml 225 ml

50 ml 40 ml 35°_37° C

'Cardioplegic solution is mixed with blood (1:4) to yield final cardioplcgic infu-

Methods Patient characteristics. The hospital records of all patients undergoing emergency bypass grafting for failed angioplasty at our institution from April 1981 through December 1989 were reviewed. Patients were included in this analysis if, in the course of an attempted elective angioplasty, emergency bypass grafting was necessary because of an acute complication. These complications included acute coronary occlusion (either during the angioplasty procedure or in the early postangioplasty period), coronary dissection with impaired antegrade flow,occlusion of a major branch vessel, or persistent chest pain with electrocardiographic evidence of ischemia. Patients were excluded if angioplasty was unsuccessful, and elective bypass grafting was performed at a later date without evidence of continuing myocardial ischemia. . On transfer of the patient to the operating room, cardiopulmonary bypass was initiated as quickly as possible. The aorta was crossclamped, the left ventricle vented, and the induction dose of cardioplegic solution administered through the aortic root. Cardioplegia protocols. Before February 1987 all patients (group I) undergoing emergency bypass grafting for failed angioplasty received cold (4 ° C) induction of cardioplegia after systemic cooling to 25 ° to 28 ° C. A 1:4 mixture of cold crystalloid cardioplegic solution (Table I) and blood (final hematocrit value 20% to 25%, temperature -4°C) was infused into the aortic root until cardiac arrest occurred. The infusion was then continued at 150 to 200 mil min for a total of 5 minutes. The total induction dose was usually 800 to 1500 ml. Maintenance doses of the same cardioplegic solution (200 to 500 ml) were administered through the aortic root and vein grafts every 20 minutes. On completion of the distal anastomoses, the aortic crossclamp was removed and core rewarming was begun. No additional doses of cardioplegic solution or perfusate were delivered through the vein grafts after removal of the crossclamp. The proximal anastomoses were performed with the aid of a partial occlusion clamp. Reperfusion through the grafts occurred on completion of the proximal anastomoses and release of the partial occlusion clamp. Since February 1987 all patients (group 2) having emergency bypass grafting for failed angioplasty have undergone normothermic induction cardioplegia followed by cold maintenance cardioplegia, followed by normothermic reperfusion cardioplegia as follows. On establishment of cardiopulmonary bypass and with the patient still warm (35° to 37° C), a 1:4 mixture of a crystalloid cardioplegic solution (Table II) and

sion solution.

normothermic blood (final hematocrit value 20% to 25%, temperature 35° to 37° C) is infused into the aortic root at a rate of250 to 300 ml/ min until cardiac arrest occurs. The rate is then decreased to 150 to 200 ml/rnin until a total of 5 minutes has elapsed. The patient is then cooled to 25° to 28° C, and 750 ml of a 1:4 mixture of cold (4 ° C) maintenance cardioplegic solution (Table III) and blood is given for 5 minutes, with additional doses of 100 to 200 ml given into the aortic root every 20 minutes thereafter. On completion of each distal anastomosis, 100 to 200 ml of cold maintenance blood cardioplegic solution is infused through each vein graft. During the completion of the final distal anastomosis, systemic rewarming is begun. A 1:4 mixture (450 ml) of cardioplegic solution (Table IV) and normothermic blood is then infused for 3 minutes into the aortic root and vein grafts, after which the aortic crossclamp is removed. Each vein graft is then perfused directly from the cardiopulmonary bypass circuit at a rate of 50 ml/rnin with the use of a multihead perfusion cannula until its proximal anastomosis is begun. The proximal anastomoses are performed with the aid of a partial occlusion clamp. On completion of the final proximal anastomosis, the partial occlusion clamp is removed and each vein graft is perfused with flow directly from the cardiopulmonary bypass circuit. Except for the use of the normothermic cardioplegia protocol, no changes in surgical technique and pharmacologic or mechanical support occurred throughout the study period. No patient received retrograde coronary sinus perfusion. Definitions. A significant coronary lesion was defined as >50% luminal diameter reduction. An estimate of the duration of myocardial ischemia was defined as the time (in minutes) from the last balloon inflation (or the onset of chest pain and electrocardiographic changes when a complication occurred after leaving the cardiac catheterization laboratory) to the initiation of cardiopulmonary bypass. A precise determination of this interval was available for only 40 patients. Therefore the data were analyzed for the group as a whole and for these 40 patients separately. Diagnosis of myocardial infarction. Standard twelve-lead electrocardiograms were obtained on admission to the hospital, immediately after bypass grafting, and, in most cases, daily until discharge. Q wave myocardial infarction was diagnosed when new Q waves of 40 msec duration and 25% the amplitude of the QRS complex developed in two or more contiguous leads, asso-

Volume 101

Reduction oj myocardialinfarction 1 0 7 1

Number 6 June 1991

Table III. Cold mainienance cardioplegic solution* Dextrose (5% in 0.2% NaCl, USP) Tromethamine (0.3 mol/L) Citrate, phosphate, dextrosesolution (USP) Citric acid (300 mg) Sodium citrate (2.63 gm) Sodium phosphate (222 mg) Dextrose (1.28 gm) Osmolality (438 mOsm/L) Potassium chloride (2 mliq/rnl) Temperature

500ml 200ml 50 ml

10ml 4° C

*Cardioplegie solution is mixed with blood (I :4) to yield final eardioplegie infusate.

Table IV. Normothermic reperjusion cardioplegic

solution*

Dextrose (5% in water) Tromethamine(0.3 mol/L) Citrate-phosphate-dextrose solution (USP) Citric acid (300 mg) Sodiumcitrate (2.63 gm) Sodium phosphate (222 mg) Dextrose (1.28 gm) Osmolality (438 mOsm/L) Dextrose 50% in water Potassium chloride(2 mfq/ml) Temperature

450ml 225 ml 225 ml

40ml 15 ml 35°-37" C

*Cardioplegic solution is mixed with blood (I :4) to yield final eardioplegie infusion solution.

ciated with enzymatic evidence of myocardial infarction. Non-

Q wave myocardial infarction was diagnosed when enzymatic

evidence of myocardial infarction developed in the absence of new diagnostic Q waves. . Blood sampling for total creatine kinase (CK) and CK myocardial band (MB) isoenzyme activity was obtained immediately on return to the intensive care unit and each morning thereafter for 2 days. In the absence of enzymatic evidence of myocardial infarction, no further samples were obtained. When enzymatic evidence of myocardial infarction was present, blood sampling continued each morning until serum enzyme levels normalized r, During the study period, total CK activity was determined with the Astra System (Beckman Instruments, Inc., Brea, Calif.). Normal range activity with this system is 22 to 180 IU IL.lsoenzyme analyses were performed with the use of three different methods. Before April 1985, isoenzyme electrophoresis (Beckman) was used (normal ::s4% of total activity). From April 1985 to March 1987, Isomune-CK (Roche Diagnostic Systems, Nutley, N.J.) was used (normal ::sIS lUlL). Since March 1987 the CK-MB system (DuPont Laboratories, Wilmington, Del.) has been used (normal ::s15 Il.l/L). The reproducibility and comparability of these three systems have been demonstrated in our laboratory with control solutions and patient samples. Enzymatic evidence of myocardial infarction was defined as peak total CK and MB isoenzyme activity >1.5 X upper limit of normal for the method employed. A crude assessment of myocardial infarct size was attempted by comparing peak MB isoenzyme levels (in the case of the isoenzyme electrophoresis method, the product of the MB% X total CK activity was used to determine peak MB). For patients who died during bypass grafting, neither enzymatic nor electrocardiographic data could be obtained. These patients were considered to have had a Q wave myocardial infarction. Statistical methods. Data were analyzed by means of standard statistical techniques. A two-tailed Fisher's exact test was used to compare frequencies of categoric variables. Continuous variables were analyzed with the use of the Mann-Whitney U test. Stepwise multiple logistic regression analysis was performed to assess the independent contributions of multiple variables to the development of myocardial infarction. A significance level of 0.05 was used for all testing.

Results

A total of 64 patients wereincluded in the study. In 45 patients (group I) emergency bypass grafting was per-

Table V. Clinical characteristics

Age (mean ± SD) Gender (% male) CAD (mean/patient)

Group I (n = 45)

Group 2

58.2 ± 7.5 69% 1.6 ± 0.6

58.8 ± 10.0 74% 1.6 ± 0.6

(n

= 19)

p

Value

NS NS NS

SO. Standard deviation; NS, not significant; CAD. number of coronary arteries with> 50% stenosis.

formed with the cold cardioplegia protocol, and in 19 patients (group 2) the normothermiccardioplegia protocolwasused.The clinicalcharacteristicsofthe twogroups are presentedin Table V. There were no significant differences in the sex distribution, average age, or average numberofsignificantly diseased coronaryvessels between the two groups. Table VI demonstrates the anatomic and procedural characteristicsof the two groups. Except for the higher incidence ofangiographically determinedseverecoronary dissection (p < 0.003) in group 2, there were no significant differences. No patient received an internal mammary artery graft. Outcome data are presentedin Table VII. In group 1 the prevalence of myocardialinfarctionas determined by enzymeanalysis was65%comparedwith 26%in group 2 (p < 0.007).There werereductions in both Q wave(38% in group 1versus16%in group 2) and non-Q wave(27% in group 1versus11% in group 2) myocardialinfarctions that trended toward, but did not attain, statistical significance. Mean total CK levels were not different between the two groups. In contrast, the difference betweenmean peak CK-MB levels was highlysignificant (p < 0.0004). The median peak.CK-MBlevel was 57 lUlL (range 1 to 550) in group I versus 10 lUlL (range 0 to 82) in group 2, suggesting not only fewer but also smaller infarcts. In this small sample,neither total death, cardiac death, nor

I072

The Journal of Thoracic and Cardiovascular Surgery

Bonner et al.

Table VI. Procedure characteristics PTCA vessel Right LAD Circumflex Collaterals Dissection Occlusion IABP Perfusion catheter No. of grafts (mean ± SD) Bypass time (min) Aortic crossclamp time (min) Ischemic time (min)

Group I (n = 45)

Group 2 (n = 19)

p Value

7 (16%) 26 (58%) 12 (26%) 5 (11%) 18 (40%) 33 (73%) 35 (78%) 1(2%) 1.8 ± 0.7 77 ± 53 32 ± 20 140 ± 36

6 (32%) 7 (58%) 2 (10%) I (5%) 16 (84%) 14 (74%) 18 (95%) 3 (16%) 1.7 ± 0.6 84 ± 36 40 ± 13 135 ± 45

NS NS NS NS <.0003 NS NS NS NS NS NS NS

PTCA. Percutaneous transluminal coronary angioplasty; NS. not significant; LAD. left anterior descending; IABP. intraaortic balloon pump; SO, standard deviation.

Table VII. Outcome data Cardioplegia protocol Myocardial infarction Q wave Non-Q wave Peak CK (lUlL) (mean ± SD) Peak MB (lUlL) (mean ± SD) Death Cardiac death Hospital stay (days)

Group I (n = 45)

Group 2 (n = 19)

p Value

29 (65%) 17 (38%) 12 (27%) 1412 ± 2530 111 ± 139 7 (15%) 7 (15%) 8.3 ± 3.3

5 (26%) 3 (16%) 2 (11%) 1292 ± 1967 18 ± 22 2 (11 %) 1 (5%) 9.8 ± 5.2

<0.007 =0.07 =0.13 NS <0.0004 NS NS NS

CK. Creatine kinase; SO, Standard deviation; NS, not significant; MB. creatine kinase myocardial band.

duration of hospital stay after operation was significantly different between the groups. The results of stepwise logistic regression analysis to identify factors associated with absence of myocardial infarction are shown in Tables VIII and IX. The use of the normothermic cardioplegia protocol, nontotal occlusion of the angioplasty vessel, and presence of collateral flow to the angioplasty vessel on the preangioplasty angiogram were all independently associated with the absence of myocardial infarction in the group as a whole (column I) and in the subgroup of 40 patients for whom an accurate assessment of the duration of ischemic time was known (column 2). The p value for the goodness of fit ratio was <0.05, indicating that these three factors accounted for the majority of variability between the two groups. The log likelihood ratios, improvement in x2, and goodness of fit p values strongly support these three factors as the most important contributors to absence of myocardial infarction in the model.

Discussion Our results with normothermic induction of blood cardioplegia, cold maintenance blood cardioplegia, and normothermic reperfusion of blood cardioplegic solution during emergency bypass grafting in the setting of failed angioplasty suggest that this method may resultin significant reduction of myocardial infarction when compared with our previous method of myocardial preservation with cold cardioplegia. Previous reports of emergency bypass grafting for failed angioplasty with cold-induction cardioplegia and blood reperfusion have demonstrated myocardial infarction rates similar to those seen in our cold cardioplegia group.s3, 4, 7, 8 Caution must be used in interpreting our results. The nonrandomized and nonconcurrent nature of this study introduces the possibility that baseline differences in the two study groups or changes over time may be responsible for the observed differences in prevalence of myocardial infarction. Analysis of baseline variables (Tables V

Volume 101

Number 6 June 1991

and VI), however, demonstrates no significant differences between the groups with the exception that angiographically determined severe coronary dissection was more prevalent in the group receiving normothermic cardioplegia, a difference that would be expected to impact negatively on the results in this group. Although the study period spans 8V2 years, no significant changes in preoperative support or surgical technique were noted. Furthermore, only two cardiac surgeons were responsible for the vast majority of surgical procedures (60/64), eliminating the confounding effect of multiple surgical approaches to a similar clinical situation. To further ensure that changes in surgical technique or postoperative support did not influence the results, we analyzed the rates of mortality and myocardial infarction in patients undergoing elective, isolated coronary artery bypass grafting during the study period. The mortality rate in this group of patients was 1.6% in 1981 and remained essentially unchanged throughout the study period. Furthermore, the rate of total myocardial infarction (Q wave and non-Q wave) remained stable between 2% and 4% throughout the study period. The results in patients undergoing elective bypass suggest that the reduction in myocardial infarction rates seenin group 2 are not attributable to changes in routine intraoperative or postoperative care, but reflect a change in the approach to these patients. The only significant documentable change between the two study groups was the use of the normothermic cardioplegia protocol. Imponderables, such as changes in clinical judgment by the cardiologist performing percutaneous transluminal coronary angioplasty as to when to operate or improvements in the handling of the patient by the anesthesiologist during the operation, cannot be excluded as having significantly contributed to the results. Another limitation of the interpretation of the data is the lack of information regarding the precise duration of ischemia in all patients. In the 40 patients (24 in the cold group and 16 in the normothermic group) for whom this interval was known, however, no significant differences in ischemic time or other variables were noted by either univariate or multivariate analysis from the group as a whole. The mechanism by which the normothermic induction of cardioplegia and the use of normothermic reperfusion of cardioplegic solution limits or prevents the development of myocardial necrosis is uncertain. Prolonged ischemic injury results in depletion of myocardial energy stores and conversion to anaerobic metabolism, with the subendocardial region being most severely affected.'? Rosenkranz and colleagues!' demonstrated in dogs that recovery of high-energy phosphate stores and reversal of

Reduction of myocardial infarction

10 7 3

Table VIII. Stepwise multiple logistic regression analysis to identify factors independently associated with absence of myocardial infarction Entire group In = 60) Variable Reperfusate N ontotal occlusion Collateral flow Dissection CAD Age Bypass time Perfusion catheter No. of grafts IABP Sex Aortic crossclamp time Ischemic time

P 0.0045 0.0257 0.0309 0.2171 0.2231 0.2556 0.2818 0.4805 0.5212 0.6630 0.7313 0.8952

Patients in whom ischemic time is known In = 40)

P 0.0005 0.0030 0.0647 0.3492 0.7024 0.8562 0.9252 0.8946 0.3854 0.9631 0.6441 0.7041 0.8502

CAD. Numberofcoronary arteries with >50';;. stenosis; IASP. intraaorticballoon pump.

anaerobic metabolism, as well as recovery of global left ventricular function, were improved when a 5-minute period of normothermic induction of cardioplegia was employed, suggesting that normothermic induction may result in greater metabolic reserve in the jeopardized myocardial bed. In contrast, Grondin and coworkers l 3 demonstrated significantly less regional myocardial recovery (as assessed by percentage and maximal velocity of fiber shortening) in dogs receiving cold potassium cardioplegic solution with critical coronary stenoses versus control animals without stenoses. Lazar and colleagues'? have shown that, in dogs undergoing reversible ischemic damage, a reactive hyperemia ensues when the myocardium is still viable, with consumption of oxygen beyond that necessary to meet the demands of mechanical function. It has been postulated that oxygen may be used in the repletion of high-energy phosphate stores and in cellular repair processes. These authors demonstrated that myocardial oxygen uptake markedly exceeded basal requirements during normothermic reperfusion of blood cardioplegic solution. Inhibiting electromechanical work with cardioplegic solution infusion while maintaining normothermia may allow maximum diversion of oxygen to oxygen-requiring reparative processes. Findings consistent with these hypotheses have been reported by Teoh and associates 14 in 20 patients undergoing elective bypass grafting. The infusion of normothermic cardioplegic solution before crossclamp release resulted in improved metabolic recovery, with

The Journal of Thoracic and Cardiovascular

I 0 7 4 Bottner et al.

Surgery

Table IX. Summary of stepwise multiple logistic regression results Improvement Step No.

Variable

Log likelihood

Reperfusate Nonocclusion Collaterals

-24.9 -18.7 -15.2 -13.5

0 I

2 3

higher levels of high-energyphosphates and reduced left atrial pressures during recovery. The characteristics of the cardioplegicreperfusate also favor maximal myocardial recovery. Tromethamine is a physiologic buffer that restores acid-base balance and promotes an optimal milieu for enzymatic processes. Sodium citrate serves as an ionic calcium chelator and may prevent or minimizethe cellular damage induced by ionic calcium. Finally, the high osmolality of the cardioplegicsolution reperfusate has been shown to prevent myocardial edema, 15 an effectthat can exacerbate reperfusioninjury by increasingcoronary resistance,impairing mitochondrial function, and reducing left ventricular compliance.l'<'? Whether substrate enhancement with glutamate and aspartate.P or the addition of the calcium channel blocking agent diltiazem-! to the cardioplegic solution would have improved the results cannot be assessed by the present study. In our protocol only a 3-minute period of normothermic reperfusionwas used as opposedto a 20-minute period used by Allen and coworkers.P It is possible that a longer duration of normothermic reperfusion may have resulted in greater myocardial salvage in the present study. Conversely, it may be that prolongingthe duration of normothermic reperfusionbeyond 3 minutes results in no further myocardial salvage. The answer to this question cannot be determined by the present study. A possible explanation for the differencein infarction rates between the two groups that is unrelated to the use of the normothermic cardioplegia protocol is the earlier time of reperfusion in group 2. In this group, after the administration of the 3-minute warm reperfusiondoseof cardioplegicsolution,the grafts and distal coronary beds were immediately perfusedwith flow from the cardiopulmonary bypass circuit via the multihead perfusion cannula, whereas in group I reperfusionfrom the cardiopulmonary bypass circuit was delayed until the proximal anastomoses were completed and the side-biting aortic clamp was removed. Although this differencecannot be ignored, the approximate time to construct all proximal anastomoses rarely exceeds 15 minutes, a small percentage of the total ischemictime. It seems unlikelythat this

Xl 12.4 7.1 3.4

Goodness offit p

Xl

P

<0.0001 0.008 0.065

49.8 37.4 30.3 26.9

0.05 0.31 0.60 0.76

delay can account for the greater prevalence of myocardial infarction observed. Multiple logisticregressionanalysisdemonstrated that the use of the normothermic bloodcardioplegiaprotocol, nontotal occlusion of the angioplastyvessel, and the presence of collateral flow to the angioplasty vessel (in decreasing order of significance) were all independently associated with a lower incidence of myocardial infarction, in the group as a whole and in the subgroup of 40 patients for whom an accurate assessmentof the ischemic time could be made. Thus a logical approach to the patient with failed angioplasty and active myocardial ischemiacan be advocated.All such patients undergoing emergencybypassgrafting may benefitby use of the normothermic cardioplegia protocol. If the angioplasty vesselisoccludedand nocollateral flow is present,placement of a perfusioncatheter in the catheterization laboratory, if possible, may be of benefit,22,23 If collateral flow is present,the placementof a perfusioncatheter may notbe necessary. An intraaortic balloonpump should beplaced to reduce myocardial oxygen demand.i" The marked reduction in perioperative myocardial infarction as assessed by enzyme and electrocardiographic evaluationsuggestsa "salvage" effectfor the normothermiccardioplegiastrategy for ischemicmyocardiumin the setting of failed angioplasty. Patients undergoing emergencybypassgrafting for acute ischemiain other settings might also benefitwith reduction in infarct size and cardiac death by applicationof this technique. Basedon the results of this trial, we believe that a prospective, randomized trial comparing the normothermicand coldcardioplegia strategies is warranted. REFERENCES 1. Bredlau CE, Roubin GS, Leimgruber PP, Douglas JS, King SB, Gruentzig AR. In-hospital morbidity and mortality in patients undergoing elective coronary angioplasty. Circulation 1985;72:1044-52. 2. Reul GJ, Cooley DA, Hallman GL, et al. Coronary artery bypass for unsuccessful percutaneous transluminal coronary angioplasty. J THORAe CARDIOVASC SURG 1984; 88:685-94.

Volume 101 Number 6 June 1991

3. Pelletier LC, Pardini A, Renkin J, David PR, Hebert Y, Bourassa MG. Myocardial revascularization after failure of percutaneous transluminal coronary angioplasty. J THORAC CARDIOVASC SURG 1985;90:265-71. 4. Golding LAR, Loop FD, Hollman JL, et al. Early results of emergency surgery after coronary angioplasty. Circulation 1986;74(Pt 2):III26-9. 5. Cowley MJ, Dorros G, Kelsey SF, Van Raden M, Detre KM. Emergency coronary bypass surgery after coronary angioplasty: the National Heart, Lung, and Blood Institute's Percutaneous Transluminal Coronary Angioplasty Registry experience. Am J Cardiol I984;53:22c-6c. 6. Jones ELK, Craver JM, Gruentzig AR, et al. Percutaneous transluminal coronary angioplasty: role of the surgeon. Ann Thorac Surg 1982;34:493-503. 7. Parsonnet V, Fisch D, Gielchinsky I, et al. Emergency operation after failed angioplasty. J THORAC CARDIOVASC SURG 1988;96:198-203. 8. Talley JD, Jones EL, Weintraub WS, King SB. Coronary artery bypass surgery after failed elective percutaneous transluminal coronary angioplasty: a status report. Circulation I989;79(Pt 2):1126-31. 9. Connor AR, Vlietstra RE, Schaff HV, Ilstrup DM, Orszulak T A. Early and late results of coronary artery bypass after failed angioplasty. J THORAC CARDIOVASC SURG

Reduction of myocardial infarction

15.

16.

17.

18.

19.

20.

1988;9~:191-7.

10. Lazar HL, Buckberg GD, Manganaro AJ, et al. Reversal of ischemic damage with secondary blood cardioplegia. J THORAC CARDIOVASC SURG 1979;78:688-97. II. Rosenkranz ER, Vinten-Johansen J, Buckberg GD, Okamoto F, Edwards H, Bugyi H. Benefits of normothermic induction of blood cardioplegia in energy-depleted hearts, with maintenance of arrest by multidose cold blood cardioplegic infusions. J THORAC CARDIOVASC SURG 1982;84:667-77. 12. Jones RN, Peyton RB, Sabina RC, et al. Transmural gradient in high-energy phosphate content in patients with coronary artery disease. Ann Thorac Surg 1981;32:546-53. 13. Grondin CM, Helias J, Vouhe PR, Robert P. Influence of a critical coronary artery stenosis on myocardial protection through cold potassium cardioplegia. J THORAC CARDIOVASC SURG 1981;82:608-15. 14. Teoh KH, Christakis GT, Weisel RD, et al. Accelerated myocardial metabolic recovery with terminal warm blood

21.

22.

23.

24.

10 7 5

cardioplegia. J THORAC CARDIOVASC SURG 1986;91:88895. Follette DM, Fey K, Buckberg GD, et al. Reducing postischemic damage by temporary modification of reperfusate calcium, potassium, pH, and osmolarity. J THORAC CARDIOVASC SURG 1981;82:221-38. Jennings RB, Ganote CEo Mitochondrial structure and function in acute myocardial ischemic injury. Circ Res 1976;38(suppl 1):180-91. Frame LH, Powell WJ. Progressive resistance to coronary blood flow in the low flow ischemic state. Circulation 1975;52(Pt 2):II183. Willerson JT, Watson JT, Hutton I, Templeton GH,'Fixler DE. Reduced myocardial reflow and increased coronary vascular resistance following prolonged myocardial ischemia in the dog. Circ Res 1975;36:771-81. Nelson RL, Goldstein M, McConnell DH, Maloney JV Jr, Buckberg GD. Studies of the effects of hypothermia on regional myocardial blood flow and metabolism during cardiopulmonary bypass. J THORAC CARDIOVASC SURG 1977;73:201-7. Allen BS, Buckberg GD, Schwaiger M, et al. Studies of controlled reperfusion after ischemia. XVI. Early recovery of regional wall motion in patients following surgical revascularization after eight hours of acute coronary occlusion. J THORAC CARDIOVASC SURG 1986;92:636-48. Melendez FJ, Gharagozloo F, Sun SC, Benfell K, et al. Effects of diltiazem cardioplegia on global function, segmental contractility, and the area of necrosis after acute coronary artery occlusion and surgical reperfusion. J THORAC CARDIOVASC SURG 1988;95:613-7. Hinohara T, Simpson JB, Phillips HR, Stack RS. Transluminal intracoronary reperfusion catheter: a device to maintain coronary perfusion between failed coronary angioplasty and emergency coronary bypass surgery. J Am Coli Cardiol 1988;11:977-82. Werter C, EI Gamal M, Bonnier H, Michels R, Van Gelder L, Krieken AYD. Coronary reperfusion with a new catheter in six patients with acute occlusion after angioplasty. Cathet Cardiovasc Diagn 1988;14:238-42. Murphy DA, Craver JM, Jones EL, et al. Surgical revascularization following unsuccessful percutaneous transluminal coronary angioplasty. J THORAC CARDIOVASC SURG 1982;84:342-8.