Is There a Place for Cold Crystalloid Cardioplegiain the 1990s?Paul J. Hendry, MD, Roy G. Masters, MD, and Anne Hasped, RNDivision of Cardiac Surgery, University of Ottawa Heart Institute, Ottawa Civic Hospital, Ottawa, Ontario, Canada

New cardioplegic techniques have been advocated asadvantageous when compared with previously acceptedstandard methods. This study assessed results of a recentcohort of 388 consecutive patients undergoing coronaryartery bypass grafting between 1990 and 1992 using astandard cold crystalloid cardioplegia. The mean age ofpatients was 61.0 ± 0.5 years; there were 324 male and 64female patients. Risk factor distribution has not changedrecently. Preoperative left ventricular dysfunction (gradeIII/IV) was significant in 35.1% of patients, and 63.3%had previous myocardial infarctions. Intraoperative car­diac arrest was achieved using an average of 1,254 ±14 mL of S1.Thomas' solution in multiple doses. Patientshad a mean of 3.1 ± 0.04 grafts constructed, with 87.7%receiving at least one internal thoracic artery graft. An-

Since the advent of cardiac surgery in the 1950s,surgeons have desired a quiet and bloodless field for

their operations. In 1955, Melrose and colleagues [1]introduced a method to induce chemical arrest of theheart. Since that time, much attention has been paid todeveloping the best method for maintaining a goodoperative field without detriment to myocardial function.Many of these approaches involve arresting the heartusing solutions which may be classified as cold crystal­loid [2, 3], cold blood [4] and warm blood [5] cardiople­gias.

As techniques for cardiac operation and anesthesiawere refined, the variety and complexity of solutions,additives, and temperatures used for myocardial preser­vation increased. Comparisons have been made bothexperimentally and clinically to determine the best meth­od; however, many of these studies leave cliniciansconfused because comparisons are made between dis­similar techniques over different time periods.

In a recent editorial, Chiu [6] advised that surgeons useclinical endpoints including morbidity, mortality, andcost to determine the efficacy of myocardial preservation.At the University of Ottawa Heart Institute, cold crystal­loid cardioplegia has been used for many years. With theevolution of myocardial preservation, we wanted to de-

Accepted for publication June 1, 1994.

Presented at the Forty-Sixth Annual Meeting of the Canadian Cardiovas­cular Society in Vancouver, Canada, Oct 26-30, 1993.

Address reprint requests to Dr Hendry, University of Ottawa HeartInstitute, Ottawa Civic Hospital, Rm 207, 1053 Carling Ave, Ottawa, Ont,Canada K1Y 4E9.

© 1994 by The Society of Thoracic Surgeons

oxia and total cardiopulmonary bypass times were 40.6 ±0.6 and 90.3 ± 1.4 minutes, respectively. Perioperativemyocardial infarction was identified in 5.7% of patientswith elevated creatine kinase levels and electrocardio­graphic measurement changes, inotropic agents wererequired in 17.8%, and the intraaortic balloon was re­quired in 3.6% of patients. Hospital stays averaged 11.2 ±0.5 days. Overall 30-day mortality was 1.0%, whereasthose patients undergoing elective operation (n = 219)had a mortality of 0%. In conclusion, cold crystalloidcardioplegia has the advantages of being inexpensiveand simple to use. These results suggest that coronaryartery bypass grafting still may be performed very safelyusing this technique.

(Ann Thorae Surg 1994;58:1690-4)

termine whether a change was justified in our method ofcardioplegia. The purpose of this study was to review theresults of coronary artery bypass operation in terms ofclinically defined end points.

Material and Methods

Charts of patients undergoing first-time coronary arterybypass grafting (CABG) at the University of Ottawabetween July 1990 and June 1992 by two surgeons werereviewed retrospectively. Preoperative risk factors, pro­cedures, and outcomes were recorded.

CardioplegiaDuring the study period between July 1990 and Septem­ber 1991, St. Thomas' Hospital solution II (Plegisol; Ab­bott Laboratories, North Chicago, IL) was used as thecardioplegia (Table 1). After that date, the Heart Instituteconverted to a solution that was prepared by the hospitalpharmacy as follows: an electrolyte concentrate wasadded to 1 L of Ringer's lactate solution before infusion.The composition of the solution (HI solution) was similarto St. Thomas' Hospital solution I but has a lower sodiumand calcium content and has added HC03 - for buffering.The major reason for this change was a significant costsaving of the pharmacy-prepared solution (ie, $4.75 [Cdn]versus $25.54 for each liter of commercially preparedsolution).

DefinitionsElective operation was defined as that performed on astable patient more than 72 hours after cardiac catheter­ization. Urgent operation was that requiring operation

0003-4975/94/$7.00

Ann Thorac Surg1994;58:1690-4

Table 1. Composition of Cardioplegic Solutions UsedDuring Study

HENDRY ET AL 1691COLD CRYSTALLOID CARDIOPLEGIA

Table 2. Risk Factorsfor Coronary Artery Disease (n =388)

a Abbott Laboratories, North Chicago, IL.

HI = Heart Institute modification of St. Thomas' I solution.

within 24 to 72 hours of catheterization due to unstablesymptoms, severity of coronary anatomy, or intravenousdrug therapy requiring monitoring in the intensive careunit. Emergent operation was performed if it was re­quired within 24 hours after catheterization due to insta­bility of symptoms or hemodynamic status, or ongoingsigns of ischemia/acute myocardial infarction not re­sponding to medical therapy. Length of stay was calcu­lated from the time of operation to time of discharge.

Perioperative myocardial infarctions were diagnosed ifthe serum creatine kinase level was greater than 1,500units and the creatine kinase-MB level (if available) wasgreater than 5% of the total, as well as electrocardio­graphic changes including new Q waves or change inR-wave progression in the precordial leads. Conductionblocks were noted including right or left bundle-branchblocks and left anterior or posterior hemiblocks.

Plegisola HI Solution St. Thomas' IComponent

Na" (mmol)K+ (mmol)0- (mmol)Ca2 + (mmol)Mi+ (mmol)Procaine (mmol)HC03 - (mmol)Osmolality (mOsm/L)pH (at 20°C)Cost/Liter (Cdn)

120.016.0

160.4

1.2

16.0o

10.0324

7.8$25.54

127.419.6

183.0

1.515.7

0.985.0

3277.75

$4.75

144.020.0

200.42.4

16.01.0

o300-3205.5-7.0

?

Factor Yes (%) No (%) Unknown (%)

Smoking 77.3 21.9 0.8

Diabetes 19.6 80.4 0

Hypertension 44.1 55.9 0

Hypercholesterol 36.1 61.1 2.8

Family history 61.3 27.8 10.8

bladder temperature was greater than 36°C, the patientwas weaned from cardiopulmonary bypass. Generally,inotropic agent support was used if the cardiac outputwas less than 2.0 Llmin, although the use of these agentswas often at the discretion of the anesthetist and surgeonand depended on a variety of relatively subjective crite­ria. Intraaortic balloon counterpulsation was used preop­eratively in patients who were hemodynamically unsta­ble and intraoperatively if increasing doses or more thantwo inotropic agents were required in a patient withunstable hemodynamic parameters.

Patients initially were monitored in the intensive careunit and subsequently followed a prescribed course ofprogressive ambulation on the surgical ward. They weredischarged from the hospital when fully ambulatory.

Data Collection and Statistical AnalysisCharts were reviewed by a research nurse, and preoper­ative, intraoperative, and postoperative data were col­lected using a standardized form. Descriptive statisticswere reviewed and comparisons were made betweendiscrete variable using>? analysis. Comparisons betweensolutions were made using both>? and Student's t testanalysis, with a p value less than 0.05 considered asindicating statistical significance.

Results

A total of 388 patients underwent first-time coronarybypass operation over the 2-year period. The meanpatient age was 61 ± 0.5 years with 324 male (83.5%) and64 female patients (16.5%). The distribution of risk factorsis shown in Table 2.

The majority of patients (63.3%) had a myocardialinfarction remotely, and 14 (3.6%) patients presentedwith an acute myocardial infarction as the indication forbypass operation. Six patients (1.5%) presented in cardio­genic shock. Otherwise, the major presenting symptomwas angina in 95.9% of patients with 71.6% havingCanadian Cardiovascular Society (CCS) class III or IVsymptoms. The distribution of preoperative symptomclass is as follows:

Operative ProcedureAfter anesthetic induction, patients were intubated andhemodynamic monitoring lines were established. Afteradministration of intravenous prophylactic antibioticsand standard chest/leg preparation and draping, mediansternotomy was performed. Internal thoracic arteries andlong saphenous veins were harvested. After intravenousheparin administration, patients were placed on cardio­pulmonary bypass using aortic and two-stage venouscannulation and a membrane oxygenator. The aorta wascross-clamped and 1 L of cold (4°C) crystalloid cardiople­gia was administered to attain cardiac arrest. Body tem­perature was maintained at 28° to 32°C depending on thepreference of the surgeon. Topical cold saline solution orslush (or both) was used concomitantly. Distal anastomo­ses between conduit and coronary artery were made firstusing 6-0 or 7-0 Prolene (Ethicon, Somerville, NJ) contin­uous sutures. When all of the bypass grafts were con­structed, the cross-clamp was removed and the proximalanastomoses were made. For those requiring defibrilla­tion, direct-current cardioversion was initiated at 10 J.The patient was rewarmed during this time, and when

CCS angina classIIIIIIIV

6.2%22.2%38.1%33.5%

1692 HENDRY ET ALCOLD CRYSTALLOID CARDIOPLEGIA

New York Heart Association (NYHA)congestive heart failure classIIIIIIIV

Left ventricular wall motion scoreIIIIIIIV

PriorityElectiveUrgentEmergency

26.2%50%

11.9%11.9%

33.3%31.6%23.4%11.7%

56.4%32.7%10.8%

Ann Thorac Surg1994;58:1690-4

Table 3. Comparison of Patient Variables by CardioplegicSolution Used

HIPlegisol Solution p

Factor (n = 221) (n = 167) Value

Age (y) 62.0::':: 0.6 59.8::':: 0.8 0.04Hospital stay (days) 12.4 ::':: 0.7 9.9::':: 0.5 0.01Incidence of heart 19.5 13.9 NS

block po stop (%)

% requiring defib 87.3 25.8 0.0001Periop MI (%) 5.9 5.4 NSMortality (%) 0.9 1.2 NS

defib = defibrillation; HI = Heart Institute modified St. Thomas' Icardioplegic solution; MI = myocardial infarction; NS = not sig­nificant; postop = postoperatively.

Intraoperative data are as follows: The average numberof grafts/patient was 3.1 ± 0.04. Left internal thoracicartery was used in 87.7% of patients, and left and rightinternal thoracic arteries were used in 1.3% of patients.Endarterectomy and grafts were used in 2.6% of patients.Average anoxia time was 40.6 ± 0.6 minutes, and averagetotal pump time was 90.3 ± 1.4 minutes. The amount ofcardioplegia used per case was 1,254 ± 14 mL. Postoper­ative complications and mortality are as follows: periop­erative myocardial infarctions, 5.7%; reopening for bleed­ing/tamponade, 4.6%; incidence of conduction blocks,17.0%; usage of one or more inotropic agent, 17.8%; use ofintraaortic balloon pump, 3.6%; and cerebrovascular ac­cidents, 2.3%. The length of intensive care unit stay was3.3 ± 0.3 days, and the length of hospital stay was 11.2 ±0.5 days. Overall in-hospital mortality was 4 (1.0%).Mortality by status was as follows: elective cases, 0 (0%);urgent, 1 (0.8%); and emergency, 3 (7.1%; P = 0.006).Mortality in patients with acute myocardial infarctionswas 2/14 (14.3%; P = 0.02). Four patients died within30 days after operation of the following causes: 2 patientswith left main disease had development of preoperativeshock, underwent operation, and died of heart failureand arrhythmias; 1 patient died suddenly due to arrhyth­mia secondary to a small perioperative myocardialinfarction; and 1 patient with severe left ventriculardysfunction preoperatively died postoperatively of respi­ratory failure and sepsis.

Patients also were compared with respect to solutionused for cardioplegia during their operations. Differencesin age, hospital stay, the incidence of conduction blocksafter operation, the need for defibrillation after cross­clamp removal, perioperative myocardial infarctionrates, and overall 30-day mortality are shown in Table 3.

Comment

It often has been difficult to reconcile the differencesbetween numerous studies attempting to compare myo­cardial preservation techniques. For those centers usingsome form of cardioplegia during heart operation, thechoice is between cold crystalloid, cold blood, and, mostrecently, warm blood. There are abundant animal studiesassessing different solutions, temperatures, and addi-

tives. However, these studies often fail to replicate thesurgical conditions for which myocardial preservation isrequired. The presence of coronary stenoses, the re­quired manipulation of the heart which tends to kinkvessels, and the intermittent cessation of cardioplegicinfusions are infrequently duplicated in the laboratory tomake these studies clinically relevant. It is for this reasonthat clinical data must be reviewed and preservationtechniques assessed by any given center.

In this study, we reviewed our recent experience withcold crystalloid cardioplegia. Hearse and associates [3]developed St. Thomas' solution through a large series ofrat experiments. It has proved to be a very useful solu­tion; however, numerous other crystalloid solutions cur­rently are being used [7-9].

The overall mortality seen in this group of patients was1.0%, with the lowest mortality in the elective group of0%. This compares favorably with other published seriesusing different cardioplegic techniques [10-14). Mortalityof patients with acute myocardial infarction was high at14.3%, but these patients present at various stages afterthe event. Intervention in terms of "resuscitation" usingwarm blood techniques certainly seems potentially use­ful in this group of patients. Lichtenstein and co-workers[15] reported a 0% mortality rate in a group of patientswho had a myocardial infarction from 6 hours to 7 dayspreoperatively. Because it often is difficult to get a patientundergoing an acute myocardial infarction into the op­erating room within 4 to 6 hours after onset (ie, latency ofpatient coming to hospital with pain, time to diagnose,and time to mobilize the team), there is a decreasedlikelihood of reversing the process by emergency revas­cularization operation. Further refinement of techniquesfor this difficult group of patients is required.

In terms of postoperative complications, the periopera­tive myocardial infarction rate was 5.7%. Again, this iscomparable with other published series [13, 14, 16], butwe continue to try to reduce this number. The absolutecreatinine kinase concentration after operation often iscompared between patients in whom either cold crystal­loid or cold blood cardioplegias are used as an indicatorfor myocardial necrosis. There is often less creatine

Ann Thorac Surg1994;58:1690-4

kinase released during blood cardioplegia but myocar­dial infarction rates are usually the same in both groups[4]. Perioperative infarction rates have been reported tobe as high as 14% to 17% with crystalloid cardioplegia [8,17], but the composition of such solutions (ie, majordifferences in electrolyte composition) used by thesegroups may be questioned as not being ideal for preser­vation. In addition, the degree of hypothermia may varybetween surgeons/institutions.

The incidence of low cardiac output may be judged byrates of use of inotropic agents and intraaortic balloonpumps. In this study, inotropic agents were used in 17.8%and intraaortic balloon pumps were used in 3.6% ofpatients. Although the rate of intraaortic balloon pumpusage is quite acceptable, inotropic agents were usedfairly liberally, thus making them a less accurate indica­tor of myocardial dysfunction. The threshold for inotro­pic agent use often varies from center to center, thusmaking comparisons difficult.

Of interest, there were differences seen between thetwo cardioplegia groups. Although age and length ofhospital stay were different, it is unlikely that this wasdue to the cardioplegia alone. The recent decrease inhospital stay likely represents a trend to discharge pa­tients earlier. We have been experimenting with an earlydischarge program over the last year of the study, whichshould reduce the hospital stay even further. This alsoemphasizes the problem of comparing cohorts of patientstaken from different time periods as other factors such asanesthetic techniques, nursing care, and perfusion prac­tices may change over time.

Spontaneous resumption of sinus rhythm duringreperfusion has been used as an indicator for adequacyof preservation by advocates of warm cardioplegia [18].The results presented here suggest that the addition of amembrane stabilizer such as procaine to the cardioplegicsolution may result in an increased likelihood of thespontaneous return of sinus rhythm after release of thecross-clamp.

Another marker for adequacy of myocardial preserva­tion is the presence of conduction disturbances duringthe recovery period after cardioplegia administration.The overall incidence of heart block in this series was17% and was not different between cardioplegic solu­tions. This rate compares favorably with those of bloodcardioplegia. In fact, some centers using blood cardiople­gia report higher incidences of heart block comparedwith crystalloid cardioplegia [16, 19]. The use of cardio­plegia itself can cause conduction problems when com­pared with non-cardioplegia techniques such as inter­mittent cross-clamping [2]. The conduction problemsmay be related to the use of potassium in these solutions[20].

In a review of cold blood and crystalloid cardioplegias,it is difficult to objectively identify major clinical differ­ences between these techniques [4]. Whereas blood car­dioplegia composition is generally uniform between cen­ters (ie, 4:1 blood/crystalloid, similar concentrations ofelectrolytes), this is not true of crystalloid cardioplegias.Therefore, comparisons are difficult when one solution

HENDRY ET AL 1693COLD CRYSTALLOID CARDIOPLEGIA

may be based on dextrose in water whereas the other isbased on a Ringer's lactate solution, with or withoutcalcium, magnesium, and varying buffers. Further, car­dioplegic solutions are administered at different temper­atures, which may affect outcome [4]. However, compar­ing clinical series assessing warm blood cardioplegiasand our results shows little difference [21, 22]. It is ofinterest that Yau and colleagues [23] have shown that allof the aforementioned methods of myocardial preserva­tion using cardioplegias cause some myocardial isch­emia. This suggests that there is still no ideal method formyocardial protection during cardiac operation. How­ever, in comparison with other techniques, cold cardio­plegias have the advantage of providing a better opera­tive field with low risk to the patient, which was theoriginal intent of a cardioplegia at its inception [24].

Of great importance to our institution is the reductionin cost of preparing and delivery of cardioplegic solu­tions. The cost has been reduced from $25.54 to $4.75(Cdn) for each liter of cardioplegia. The delivery systemfor this method of cardioplegia is simple and inexpen­sive, whereas tubing for cold and warm blood increasesthe cost of each case by $100 to $200 (Cdn), depending oninstitution. A considerable expense is incurred with theuse of these delivery systems (ie, $100,000 to 200,000(Cdn) per 1,000 cases per year). With comparable mor­tality, morbidity, and intensive care unit and hospital staybetween techniques, this cost may be difficult to justify inthe current climate of medical economic constraints.

In conclusion, cold crystalloid cardioplegia provides aquiet, bloodless operative field. It may be administered ata very low cost and enables cardiac operation to beperformed with very acceptable results in the majority ofpatients. Improvements in cold crystalloid cardioplegictechniques are required in a subset of patients who havesevere myocardial injury preoperatively.

Doctor Hendry is a Career Scientist sponsored by the Ministry ofHealth of Ontario.

We thank Sharon Finlay, RN, for her assistance in data retrieval.

References

1. Melrose DG, Dreyer B, Bentall HH, Baker JBE. Electivecardiac arrest. Preliminary communication. Lancet 1955;2:21.

2. Flameng W, Van der Vusse GJ, De Meyere R, et al. Intermit­tent aortic cross-clamping versus St. Thomas' Hospital car­dioplegia in extensive aorta-coronary bypass grafting. Arandomized clinical study. J Thorac Cardiovasc Surg 1984;88:164-73.

3. Ledingham SJM, Braimbridge MY, Hearse DJ. The St.Thomas' Hospital cardioplegic solution. A comparison of theefficacy of two formulations. J Thorac Cardiovasc Surg 1987;93:240-46.

4. Barner HB. Blood cardioplegia: a review and comparisonwith crystalloid cardioplegia. Ann Thorac Surg 1991;52:1354-67.

5. Lichtenstein SV. Warm heart surgery: concept, concerns,and future course. J Cardiac Surg 1993;8:161-6.

6. Chiu RC-J. Cardioplegia: from the bedside to the laboratoryand back again. Ann Thorac Surg 1991;52:1209-10.

7. Fremes SE, Christakis GT, Weisel RD, et al. A clinical trial of

1694 HENDRY ET ALCOLD CRYSTALLOID CARDIOPLEGIA

blood and crystalloid cardioplegia. J Thorac Cardiovasc Surg1984;88:726- 41.

8. Beyersdorf F, Krause E, Sarai E, et al. Clinical evaluation ofhypothermic ventricular fibrillation, multi-dose blood car­dioplegia, and single-dose Bretschneider cardioplegia incoronary surgery. Thorac Cardiovasc Surgeon 1990;38:20-9.

9. Schaper J, Scheid HH, Schmidt U, Hehrlein F. Ultrastruc­tural study comparing the efficacy of five different methodsof intraoperative myocardial protection in the human heart.J Thorac Cardiovasc Surg 1986;92:47-55.

10. Buckberg GD. Antegrade/retrograde blood cardioplegia toensure cardioplegic distribution: Operative techniques andobjectives. J Cardiac Surg 1989;4:216-38.

11. Salerno TA, Houck JP, Barrozo CAM, et al. Retrogradecontinuous warm blood cardioplegia: a new concept inmyocardial protection. Ann Thorac Surg 1991;51:245-47

12. Vaughn CC, Opie JC, Florendo FT, Lowell PA, Austin J.Warm blood cardioplegia. Ann Thorac Surg 1993;55:1227-32.

13. Iverson LIG, Young IN, Ennix CL Jr, et al. Myocardialprotection: a comparison of cold blood and cold crystalloidcardioplegia. J Thorac Cardiovasc Surg 1984;87:509-16.

14. Fremes SE, Weisel RD, Mickle DAG, et al. Myocardialmetabolism and ventricular function following cold potas­sium cardioplegia. J Thorac Cardiovasc Surg 1985;89:531-46.

15. Lichtenstein SV, Abel JG, Salerno DM. Warm heart surgeryand results of operation for recent myocardial infarction.Ann Thorac Surg 1991;52:455-60.

16. Louagie YAG, Collard E, Gonzalez M, et al. Initial experi­ence with low-potassium cold blood cardioplegia: a clinicalcomparative study. Ann Thorac Surg 1992;53:628-34.

Bound volumes available to subscribers

Ann Thorac Surg1994;58:1690-4

17. Codd JE, Barner HB, Pennington DG, et al. Intraoperativemyocardial protection: a comparison of blood and asanguin­eous cardioplegia. Ann Thorac Surg 1985;39:125-33.

18. Christakis GT, Koch JP, Deemar KA, et al. A randomizedstudy of the systemic effects of warm heart sUJrgery. AnnThorac Surg 1992;54:449-59.

19. Gundry SR, Sequeira A, Coughlin TR, McLaughlin JS. Post­operative conduction disturbances: A comparison of bloodand crystalloid cardioplegia. Ann Thorac Surg 1989;47:384-90.

20. Cohen NM, Allen CA, Belz MK, Nixon TE, Wise RM,Damiano RJ Jr. Electrophysiological consequences of hypo­thermic hyperkalemic elective cardiac arrest. J Cardiac Surg1993;8:156-60.

21. The Warm Heart Investigators. Prospective randomized trialcomparing normothermic to hypothermic coronary bypasssurgery [Abstract]. Circulation 1993;88(Suppl 1):1-288.

22. Mellitt RJ, Weintraub WS, Craver JM, et al. The interrela­tionship of age and normothermic blood vs cold crystalloidcardiplegia on the incidence of stroke in elective coronaryartery bypass grafting: results of the Emory RandomizedTrial [Abstract]. Circulation 1993;88(Suppl 1):1-288.

23. Yau TM, Ikonomidis JS, Weisel RD, et al. Which techniquesof cardioplegia prevent ischemia? Ann Thorac Surg 1993;56:1020-8.

24. Lajos TZ, Espersent CC, Lajos PS, Fiedler RC, Bergsland J,Joyce LT. Comparison of cold versus warm cardioplegia.Crystalloid antegrade or retrograde blood? Circulation 1993;88(Suppl 2):344-9.

Bound volumes of the 1994 issues of The Annals of Thoracic Surgery are available only to subscribers from the Publisher. The costis $96.00 (outside US add $25.00 for postage) for volumes 57 and 58. Each bound volume contains a subject and author index, andall advertising is removed. The binding is durable buckram with the name of the journal, volume number, and year stamped onthe spine. Payment must accompany all orders. Contact Elsevier Science Inc, 655 Avenue of the Americas, New York, NY 10010; ortelephone (212) 633-3950 (facsimile: (212) 633-3990).