548 Serum Cardiac Enzymes in Stroke

J. W. NORRIS, M.D., F.R.C.P.(C), V.C. HACHINSKI, M.D., F.R.C.P.(C),

M.G. MYERS, M.D., F.R.C.P.(C), J. CALLOW, M.D., T. WONG,

AND R.W. MOORE, PH.D.

SUMMARY Serum cardiac enzyme levels (CK, LDH, SCOT) were estimated and the ECG recorded for 4days following admission of 288 patients (Group I) to a stroke intensive care unit. Sixty-four of these patients,subsequently found not to have strokes, served as controls. Mean serum levels of all 3 cardiac enzymes wereelevated in 8% of the 224 patients with stroke. The mean serum enzyme levels in patients with transientischemic attacks (TIA) did not differ from controls.

In a second group of 230 patients with stroke (Group II) serum CK levels were measured and the iso-enzymes were fractionated to determine the tissue source of the enzymes. One hundred and one patients hadraised total CK values and 25 of these (11%) had raised CK-MB (heart) iso-enzyme, the remainder having CK-MM (skeletal muscle) fraction. No serum CK-BB (brain) iso-enzyme was detected in any patient. Patientswith positive serum levels of CK-MB had more evidence of acute myocardial ischemia on ECG (p < 0.05), andmore cardiac arrhythmias (p < 0.001) than those with normal CK levels. Scattered areas of myocytolysis werefound in the myocardium at autopsy in one patient.

The acute rise in serum cardiac enzymes which we have recorded in the initial stages of stroke suggest thatacute myocardial involvement is a commoner complication than is generally recognized. Also, since the CK-MB rises were modest and progressive, it is more likely that this acute myocardial dysfunction is a conse-quence, rather than a cause, of the acute cerebrovascular lesion.

Stroke Vol 10, No 5, 1979

SUBARACHNOID hemorrhage produces a varietyof cardiac disturbances in both animals and man.1"*These include electrocardiogram (ECG) changes,1 dis-turbances of cardiac function,4 focal myocardial ne-crosis,3 and an increase in circulating catechol-amines.2 Although similar cardiac changes occur inother forms of intracranial hemorrhage,4 the oc-currence of similar cardiac changes in non-hemor-rhagic forms of stroke is less certain.

Raised levels of serum cardiac enzymes — serumglutamic oxalo-acetic transaminase (SGOT), lactatedehydrogenase (LDH) and creatine phospho-kinase(CK) — have been reported in the initial phase ofstroke5-6 but without iso-enzyme fractionation, thetissue source of these enzymes has not been proven.Twenty-nine percent of patients with stroke in-vestigated by Dimant and Grob7 had raised serum en-zyme levels and those with elevated CK levels had ahigher incidence of ECG abnormalities, as well as ahigher mortality than those with normal CK levels.

The high incidence of cardiac arrhythmias inpatients with acute stroke810 may simply reflect un-derlying vascular disease common to both the heartand brain. However, the association of the cardiacarrhythmia with an acute rise in serum cardiac en-zyme levels at the onset of the stroke may indicate thepresence of an acute myocardial lesion.

In the present study, daily serum enzyme deter-minations, 12-lead ECG's and continuous cardiacmonitoring were performed to elucidate a possible

From the MacLachlan Stroke Unit, Department ofNeurosciences, Division of Cardiology (Dr. Myers) and Depart-ment of Biochemistry (Dr. Moore), Sunnybrook Medical Centerand University of Toronto, Ontario, Canada.

Mr. Wong is a medical student at the University of Toronto.Drs. Hachinski and Myers are Senior Research Fellows of the

Ontario Heart Foundation.

association between stroke and abnormalities of car-diac function.

Materials and Methods

All patients, initially admitted to the emergencyroom of a general hospital, were transferred to thestroke intensive care unit (Unit) when an acutecerebrovascular lesion was suspected. The Unit con-sists of 5 beds with facilities for continuous nursingand medical care, including cardiac monitoring.Details of this facility and classification of the types ofstrokes are described elsewhere.11 Diagnosis of thetype of stroke was based on clinical examination,cerebral spinal fluid analysis, isotope brain scan, com-puterized axial tomography of the brain and, where in-dicated, cerebral angiography.

Serum enzyme levels were determined daily for thefirst 4 days following admission. CK was estimated bythe method of Rosalki,12 LDH was assayed by themethod of Wroblewski,13 and SGOT was estimated bythe Henry modification of the Karmen method.14

Serum cardiac enzyme levels were said to be raisedwhen at least 2 of these 3 enzyme levels were elevated.

When CK was found to be greater than 9 Okinakaunits, the CK iso-enzymes were fractionated by themethod of agar electrophoresis (Corning ACI). The 3bands which can be identified by this method are CK-MM (skeletal muscle), CK-MB (cardiac muscle), andCK-BB (brain). Significant MB iso-enzyme wasreported present if more than a trace was detected.

Both quantitative and qualitative values of CK canbe obtained by the Corning ACI method. Quantita-tion is achieved by photographing the fluorescencecaused by the iso-enzymes acting on a fluorescent sub-strate. The negative is scanned using a Beckman den-sitometer. The data have been verified in ourlaboratory using tissue extracts and commercial con-trol sera and found to be quantitative.

by guest on May 19, 2018

http://stroke.ahajournals.org/D

ownloaded from

SERUM CARDIAC ENZYMES IN STROKE/Mww et al. 549

We estimated the total CK units and proportion ofCK in each of the fractions on electrophoresis. Ournormal range for CK-MB is less than one (valuesusually undetectable). The specificity of this methodhas been tested against various tissue extracts in-cluding the heart, skeletal muscle, brain, liver andothers.

Results

Group I

In an initial series of 288 patients admitted to theUnit, serum CK (without iso-enzyme fractionation),SGOT and LDH were estimated daily for the initial 4days following admission. Sixty-four patients subse-quently found not to have stroke served as a controlgroup and they had a variety of illnesses includingMeniere's disease and cerebral tumors. The numbersof the" different types of strokes are shown in table 1.

When the total numbers of patients in the strokeand control groups were compared, there was nosignificant difference in the incidence of raised CK,SGOT and LDH levels (table 1). However, the meanvalues of all 3 enzymes were raised in the sub-groupwith cerebral hemorrhage (table 2). Significantlyelevated CK values were found in all stroke sub-groups except for TIA.

Control Group with Raised Enzyme Levels

Eight of the 64 control cases had a rise in 2 or moreserum cardiac enzyme levels (table 1). Two had acutemyocardial infarction and one had acute coronary in-sufficiency. These patients were admitted to the Unitbecause of syncope but no neurological lesion wasfound subsequently. The raised enzyme levels in theremaining 5 cases in this group could be accounted foras follows: 2 had post-seizure states, and there weresingle cases of acute pulmonary embolism, systemiclupus erythematosus and coma from drug overdosewith associated contusions. Cardiac monitoring in 3 ofthese patients showed cardiac arrhythmias.

TABLE 1 Comparison of Stroke and Control Groups ofPatients with Elevated Serum Levels of at Least 2 CardiacEnzymes (.Group 1) or with Elevated CK-MB iso-emymes(Group 2)

TABLE 2 Incidence of Mean Serum Cardiac Enzyme Levels(± SEM) in Each Type of Stroke (Group I)

Type of stroke

Hemispheric infarcts

Brain stem infarcts

Cerebral hemorrhage

Transient ischemiaattacks (TIAs)

Total Strokes

Control Group

Group I

No.

137

25

23

39

224

64

Raisedenzymes

11

2

4t

1

18

8

Group IINo. MB +

148

20

32

30

230

53

17

2

4t

2

25

3

Lesion

Cerebralinfarcts

Brain steminfarcts

Cerebralhemorrhage

TIAs

Controls

Difference from

7.2

7.4

11.0

4.9

4.7

CK

± 0.7

± 1.3

± 2.2

± 0.7

± 0.6

controls *p =*

206

176

• " 245

192

188

0.05, " p

LDH

± 5.1

±8.7

=•= 17.8

± 8.3

± 7.2

> 0.005,

SGOT

17.0 =

17.1 *

t 0.5

t 1.4

" • 20.7 ± 1.6"

18.2 * 1.2

15.7 =i

•••p > 0.00]

= 7.3

[.

tlncludes one patient with subarachnoid hemorrhage.

Stroke Group with Raised Enzyme Levels

Eighteen of 224 patients with stroke had raised en-zyme levels (table 1) including 11 in whom the ECGsnowed changes compatible with myocardial damage.Four of these 11 had evidence of acute transmuralmyocardial infarction, 3 had evolving ST-T wavechanges consistent with subendocardial ischemia orinfarction (fig. 1), while 4 showed left ventricularhypertrophy with marked ST depression and T-waveinversion.

Two of the 7 patients without ECG changes hadother reasons for enzyme elevation, one havingpolymyositis and the other fulminating septicemia.

All 5 of the remaining cases had cardiacarrhythmias and 4 of these died shortly after dischargefrom the Unit. Therefore, 16 of the 18 patients in thestroke group with raised enzyme levels had evidence ofpossible cardiac dysfunction.

Group II

After initiation of this study, it became apparentthat many of the raised enzyme levels observed weredue to non-cardiac effects such as trauma or in-tramuscular injections. A second group of patientswere, therefore, examined for the cardio-specific CK-MB iso-enzyme levels.

Two hundred and thirty patients with stroke and 53controls were admitted to the Unit. One hundred andone patients with stroke and 35 controls were found tohave raised total CK values (table 1). After iso-enzyme fractionation of the sera, 25 of the patientswith stroke and 3 of the controls had raised CK-MBlevels. CK-BB iso-enzyme was not detected in anypatient and the remainder of the raised CK levels weredue to elevated CK-MM.

Controls with Raised CK-MB Levels

All three patients without stroke were admitted in astuporous state to the Unit with an initial diagnosis ofacute cerebrovascular lesion. One patient had atransmural myocardial infarction, another hadmassive pulmonary embolism and the third had ter-minal hepatic coma. Cardiac arrhythmias were notdetected in any.

by guest on May 19, 2018

http://stroke.ahajournals.org/D

ownloaded from

550 STROKE VOL 10, No 5, SEPTEMBER-OCTOBER 1979

Mrs. L.C. ISU67S)

N-+4

5 .4 .77

I I

\J

^ ^ f i f e l VNW. L.C. fSU 678) 6.4.77

FIGURE 1. /I 64-year-old woman collapsed with a right hemiparesis and died 5 days after ad-mission. Lumbar puncture 6 hours after admission revealed an opening pressure of 140 mmwater (clear CSF with no cells). Day 1 CK = 25 (MBpresent). LDH = 137, SGOT = 23. Day 2CK = 26 (MB present), LDH = 174, SGOT = 30. ECG showed acute myocardial infarction.

by guest on May 19, 2018

http://stroke.ahajournals.org/D

ownloaded from

SERUM CARDIAC ENZYMES IN STROKE/Wo/rw el al. 551

Stroke Patients with Raised CK-MB Levels

Seventeen of the 25 patients with stroke withelevated cardiac iso-enzyme levels had ECG findingsconsistent with myocardial damage. Three showedtransmural myocardial infarction and 5 had evolvingrepolarization abnormalities compatible with suben-docardial infarction. Nine had ECG evidence of leftventricular hypertrophy and strain. During the periodof observation there were evolutionary changes show-ing progressively more ST depression and T-wave in-version.

In the remaining 8 patients in this group, no ECGabnormalities were seen, but 6 of them had cardiacarrhythmias and one had polymyositis.

The mean daily values of serum enzymes in these 25patients with elevated CK-MB levels showed agradual rise over the 4 days following admission (table3). This is quite unlike the acute rise (within 24-28 hs)seen in acute myocardial infarction and suggests aprocess of continuing low-grade myocardial necrosis.

Incidence of Cardiac Arrhythmias

Twenty-three (92%) of 25 patients in the CK-MB+group had cardiac arrhythmias compared to 102(50%) of 205 patients in the CK-MB- group (chronicarrhythmias excluded). This difference is significant at(p < 0.001 level, x2 test). There was also a significantdifference (p < 0.02) in the total number ofarrhythmias (table 4).

The 25 patients in the CK-MB + group were thenmatched for age and sex with 25 patients in the CK-MB— group. Seventeen patients had ECG changes ofevolving ST depression and T-wave inversion in theCK-MB+ group compared to 10 patients withoutdetectable cardiac iso-enzyme levels {p < 0.05, x2

test).

Pathological Correlation

A 70-year-old man with a fatal brain stem infarc-tion had raised CK-MB levels and evolving ST-T-wave changes on ECG. The patient had no previousheart disease and was admitted in a decerebrate andcomatose state, dying on the fourth day. On the day ofadmission, CK = 40.0 O.U. (MB present), LDH =222 I.U. and SGOT = 68 I.U. On the second day, CK= 16 O.U. (MB present), LDH = 263 I.U. and SGOT= 30 I.U. Serial daily ECG's showed widespread andprogressive depression of ST-T segments in the infero-lateral chest leads with prominent U waves.

TABLE 3 Mean Values (=*= Standard Error of the Mean) ofSerum Enzymes on Each Day Following the Stroke in the 25Patients with Elevated CK-MB

TABLE 4 Incidence and Type of Arrhythmic Episodes in 25CK-MB+ Patients Compared to 205 CK-MB- Patients{Group II)

Stroke dayEnzyme 3

CK-MB 16 ± 1.2 20 ±0 .4 24 ± 1.0 36 ± 4.7

LDH 228 ± 6.4 261 =•= 5.2 259 ± 3.9 304 ± 9.2

SGOT 24 ± 2.6 33 ± 4.9 32 ± 4.0 41 ± 12.6

Type of arrhythmiaCK-MB+

GroupCK-MB-

Group

Supraventricular ectopics 10 47

Ventricular ectopics 20 82

Paroxysmal atrial fibrillation 4 9

Ventricular tachycardia 1 0

Nodal rhythm 1 0

Total 36* 138

*V >0.02.

At autopsy there was extensive atherosclerosis ofthe coronary arteries and a recent subendocardial in-farction of the left ventricle. No mural thrombus waspresent but there were multiple scattered areas ofacute myocytolysis seen in the myocardium (fig. 2).Death was due to medullary compression from aswollen infarcted cerebellar hemisphere, associatedwith basilar artery thrombosis. There was no evidenceof embolism to either the brain or any other organ.

Discussion

Myocardial dysfunction with intracranial hem-orrhage is well documented, and includes evolving"ischemic" changes on ECG,16 a rise in serum cardiacenzymes16 and elevated serum catecholamine levels.17

Although autopsy examination of the heart in suchpatients may be suprisingly normal, subendocardiallesions are sometimes found.18 These myocardialchanges have been attributed to increased sympatheticactivity2- " and can be produced experimentally inanimals by infusion of catecholamines,19 and bystimulation of certain areas of the brain.4

Although cerebral infarction is a more commonform of stroke than cerebral hemorrhage, similaracute cardiac dysfunction has seldom been described.In 231 neurosurgical autopsies,3 18 had focalmyocytolysis but only one had cerebral infarction.However, raised serum CK levels with ischemicchanges on ECG were associated with significantly in-creased mortality in 78 patients with ischemicstrokes.7 Meyer et al.20 found an increase in serum andCSF catecholamines in a relatively small number ofpatients with acute cerebral infarction only whenhypertension was also present. They concluded thatthe catecholamine increase resulted from the non-specific effect of stress and hypertension rather thanthe cerebral lesions themselves. We also foundsignificantly higher levels of plasma catecholamines inpatients with acute cerebral infarction compared tocontrol (non-stroke) patients.21 Further, plasmacatecholamine levels were significantly elevated inboth normotensive and hypertensive stroke groups.22

The significant elevations of serum SGOT, LDHand CK found in 8% of our patients with stroke,suggests a much higher myocardial involvement in

by guest on May 19, 2018

http://stroke.ahajournals.org/D

ownloaded from

552 STROKE VOL 10, No 5, SEPTEMBER-OCTOBER 1979

FIGURE 2. Section of myocardium of left ventricle showing focal area of myocytolysis (seecase report) HE stain X 400 (Courtesy of Dr. A. Kolin, Department of Anatomical Pathology,University of Toronto, Sunnybrook Medical Centre A: 140-77).

ischemic or hemorrhagic stroke, than hithertobelieved. The progressive ischemic changes on ECGand the timing of elevated cardio-specific CK iso-enzyme are further evidence that these cardiacchanges follow closely upon the heels of the acutecerebrovascular lesion. Also, the increased incidenceof cardiac arrhythmias found in Group II is consistentwith our previous finding10 of increased arrhythmias ina group of patients with acute stroke matched for age,sex and duration of stay in the Unit against a controlgroup. The cerebral lesions themselves, however, donot appear to elevate serum cardiac enzyme levels inthe absence of cardiac lesions or other causes knownto affect these enzymes.

The incidence of raised cardiac enzyme levels wasthe same whether the lesion was ischemic orhemorrhagic so that embolism of a mural thrombus tothe brain is unlikely to be the explanation. In any case,

the massive cerebral ischemic lesions encountered canscarcely be attributed to the small subendocardiallesions as indicated by the minor serum enzymeelevations, since the degree of serum CK-MB rise isproportional to the volume of infarcted cardiacmuscle.23 Although the specificity of the MB iso-enzyme for cardiac muscle is controversial,"1 " it stillremains the most sensitive indicator of acute myocar-dial necrosis.26

Circulating levels of serum CK-BB could not bedetected by earlier workers5 but increasingly sensitivefiuorometric and radioimmunoassay techniques haveshown that trace amounts do exist and may risefollowing a variety of cerebral lesions."1 *" The earlypeak (within a few hours) and transient nature of thebrain iso-enzyme in the blood may explain our in-ability to detect it in our patients.

The present findings indicate that acute ischemic

by guest on May 19, 2018

http://stroke.ahajournals.org/D

ownloaded from

SERUM CARDIAC ENZYMES IN STROKE/'Norris et al. 553

and hemorrhagic strokes may be accompanied byacute myocardial ischemia or infarction with raisedserum cardiac enzyme levels and associated with car-diac arrhythmias. These coincidental findings raise thepossibility that the acute cardiac abnormalitiessometimes seen in stroke patients are a direct conse-quence of the neurological lesion.

Acknowledgment

We thank Professor Adele Csima for statistical advice.

References1. Byer E, Ashman R, Toth LA: Electrocardiograms with large

upright T waves and long Q-T intervals. Am Heart J 33:796-806, 1947

2. Tomomatsu T, Ueba Y, Matsumoto T et al: ECG observationsand urinary excretion of catecholamines in cerebrovascular ac-cident. Jpn Circ J 28: 905-912, 1964

3. Connor RCR: Focal myocytolysis and fuchsinophilic degenera-tion of the myocardium of patients dying with various brainlesions. Ann NY Acad Sci 156: 261-270, 1969

4. Weidler DJ: Myocardial damage and cardiac arrhythmias afterintracranial hemorrhage. A critical review. Stroke 5: 759-764,1974

5. Dubo H, Park DC, Pennington RJT et al: Serum creatinekinase in cases of stroke, head injury and meningitis. Lancet 2:743-748, 1967

6. Eisen AA, Sherwin AL: Serum creatine phosphokinase activityin cerebral infarction. Neurology (Minneap) 18: 263-268, 1968

7. Dimant J, Grob D: Electrocardiographic changes and myocar-dial damage in patients with acute cerebrovascular accidents.Stroke 8: 448-455, 1977

8. Reinstein L, Gracey JG, Kline JA et al: Cardiac monitoring inthe acute stroke patient. Arch Phys Med S3: 311-314, 1972

9. Lavy S, Yaar 1, Melamed E et al: The effect of acute stroke oncardiac functions as observed in an intensive care stroke unit.Stroke 5: 775-780, 1974

10. Norris JW, Froggatt GM, Hachinski VC: Cardiac arrhythmiasin acute stroke. Stroke 9: 392-396, 1978

11. Norris JW, Hachinski VC: Intensive care management ofstroke patients. Stroke 7: 573-577, 1976

12. Rosalki SB: An improved procedure for serum creatine

phosphokinase determination. J Lab Clin Med 4: 696-705,1967

13. Wroblewski F, LaDue JS: Lactic dehydrogenase activity inblood. Proc Soc Exp Biol Med 90: 210-213, 1955

14. Henry RJ, Chiamori N, Golub OJ et al: Revised spec-trophotometric methods for the determination of glutamicoxalo-acetic transaminase, glutamic pyruvic transaminase andlactic acid dehydrogenase. Am J Clin Path 34: 381-398, 1960

15. Koskelo P, Punsar S, Sipila W: Subendocardial haemorrhageand ECG changes in intracranial bleeding. Br Med J 1:1479-1480, 1964

16. Fabinyi G, Hunt D, McKinley L: Myocardial creatine kinaseiso-enzyme in serum after subarachnoid haemorrhage. J NeurolNeurosurg Psychatry 40: 818-820, 1977

17. Cruickshank J, Neil-Dwyer G, Stott A: Possible role ofcatecholamines, corticosteroids, and potassium in production ofelectrocardiographic abnormalities associated with sub-arachnoid haemorrhage. Br Heart J 36: 699-706, 1974

18. Doshi R, Neil-Dwyer G: Hypothalamic and myocardial lesionsafter subarachnoid haemorrhage. J Neurol NeurosurgPsychiatry 40: 821-826, 1977

19. Reichenbach DD, Benditt EP: Catecholamines and car-diomyopathy. Hum Pathol 1: 125-150, 1970

20. Meyer JS, Stoica E, Pascu I: Catecholamine concentration inCSF and plasma of patients with cerebral infarction andhaemorrhage. Brain 96: 277-288, 1973

21. Myers MG, Norris JW, Sole MJ et al: Cardiac abnormalitiesfollowing acute stroke. Circulation 58: 11-92, 1978

22. Norris JW, Hachinski VC, Myers MG.: Cardiac lesions follow-ing acute stroke. 9th International Salzburg Conference onCerebral Vascular Disease, Sept 1978 (in press)

23. Cairns JA: Myocardial infarction size: measurement andmodification. Can Med Assoc J 6: 255-262, 1977

24. Adornato BT, Engel WK: MB-creatine phosphokinase iso-enzyme elevation not diagnostic of myocardial infarction. ArchInt Med 137: 1089-1090, 1977

25. Schmidt WA: Myocardial Ischemia (letter) N Engl J Med 297:448, 1977

26. Rapaport E: Serum enzymes and iso-enzymes in the diagnosisof acute myocardial infarction. Modern Concepts of Car-diovascular Disease, Dallas, American Heart Assoc, No 10, p47-50, 1977

27. Kaste M, Somer H, Kontinnen A: Brain-type creatine kinaseiso-enzyme. Arch Neurol 34: 142-144, 1977

28. Bell RD, Rosenberg RN, Ting R et al: Creatine kinase BB iso-enzyme levels by radio-immuno-assay in patients withneurological disease. Ann Neurol 3: 52-59, 1978

by guest on May 19, 2018

http://stroke.ahajournals.org/D

ownloaded from

J W Norris, V C Hachinski, M G Myers, J Callow, T Wong and R W MooreSerum cardiac enzymes in stroke.

Print ISSN: 0039-2499. Online ISSN: 1524-4628 Copyright © 1979 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Stroke doi: 10.1161/01.STR.10.5.548

1979;10:548-553Stroke. 

http://stroke.ahajournals.org/content/10/5/548World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://stroke.ahajournals.org//subscriptions/

is online at: Stroke Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer available in the

Permissions in the middle column of the Web page under Services. Further information about this process isOnce the online version of the published article for which permission is being requested is located, click Request

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Stroke Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on May 19, 2018

http://stroke.ahajournals.org/D

ownloaded from