"Nonspecific" ST and T-Wave ChangesBy CHARLES K. FR1EDBERG, 1\.D., AND AiJnERT ZAGER, M.D.

ONE of the commonest problems in elec-trocardiography is the interpretation of

STf-seienmelnt depression and T-wvave inversionwithout other characteristic changes, particu-larly when there is clinical uncertainty re-garldiiig the presenmce and nature of an acutecoronary attack. A commoni solution, oftennnsatisfactory to the clinician receiving thereport, is the statement "'The ST-seginent andT-wave changes are nonspecific abnorinali-ties." A careful analysis of the degree of STdepression and character and depth of inver-sion of the T waves and a correlation of thesefindings with clinical conditions may enableus to arrive at a more definite electrocardio-g,raphic diagnosis. Since 110 such systematicanalysis has been reported, this study wasundertaken.

MethodThe electrocardiogramns of 1,000 consecutive

adult in-patients interpreted by the members ofthe cardiology staff of The M1ount Sinai Hospital,were reviewed and correlated with the clinicalstatus of the patients.The findings are representative of records of a

large, voluntary metropolitan hospital. The adultpatients were from both the ward and privateniedical and surgical services during a periodof approximately 4 months. Although usuallythe records were taken for specific clinical indica-tions, some represented routine preoperative elec-trocardiograins in the elderly age group. Recordsthat primarily involved abnormalities in rhythmwere excluded from the study.

The abnormal electrocardiograms were classifiedinto the following groups: acute and healed trans-mural myocardial infarction; bundle-branch block;ventricular hypertrophy; pericarditis. There re-mained a large abnormal group that did not satisfyrecognized criteria for any of the above groups,with changes that were designated nonspecificST-segment and T-wave abnormalities.The depth of ST-segmnenit depression and of

T-wave inversion was contrasted in those patients

From the Division of Cardiology, Department ofMledicinie, The Mount Sinai Hospital, New York,New York.

who had no recent cardiac palin, in those whohad pain of less than one-half hour duration andin those who had pain of longer duration. Thefrequency with which isolated T-wave abnormiali-ties, isolated ST-segmient abnormalities, or comin-bined abnormalities occurred in the various cate-gories was also analysed.A similar analysis w as imuide relating abnor-

mnalities in contour of the ST segment and Twaves in the same groups of patients. The recordeddepth of T-wave inversion or ST-segmnent depres-sion was the maximum change in precordial leadsV3 to V(. Tracings considered to have abnormalcontour of the ST segment exhibited an upwardconvexity of the ST segment or a long, straightenedor downward sloping ST segnent. The ternii"isehemic T wave" in this paper included boththe "coronary T wave" of Pardee,1 which is adownward sharply peaked T wave with syminnetri-c.11 limbs following an1cl upwardl convexity of tlimR-T interval and the "cove plane" T wave ofOppenheimer, Rothchild, and Mann2 with a curveddeseending limb, peaked alpex, and straight as-cending linib.

ResultsOf the study group of 1,000 electrocardio-

gramns there were normal finidiigs in 590 andabnormal in 410 cases. The abnormal ones

were subdivided into 93 cases of acute trans-

mural myocardial infaretion, 63 cases ofhealed mnyocardial infarction, 24 cases ofbundle-branch block, 18 cases of ventricularhypertrophy, three cases of pericarditis and209 cases in which the ST-segmnent and T-wavechanges were nonspecific abnormalities (ta-ble 1).

In the latter group of 209 patients withnonspecific ST-segmnienit and T-wave changes,46 were receiving digitalis, 22 had recentcardiac pain of less than 1/2 hour duration,35 patients had recent cardiac pain of 1/2hour or longer in duration (table 2). Twelltyadditional patients had had recent cardiacpain 1/ hour or longer but were already in-eluded in the previous categories becausethey had received digitalis or had other causes

for ST-segment and T-wave abnormalities

Circulation, Volume XXIII, May 1961 655

by guest on June 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

FRIEDBERG, ZAGER

Table 1Analysis of 1,000 Consecutive Adult Electrocardio-grams

Normal 590Abnormal 410Total 1000

Analysis of Electrocardiographic AbnormalitiesAcute infarct 93Old infarct 63Bilateral B.B.B. 1R.B.B.B. 17L.B.B.B. 6R.Y.H. 1L.V.H. 17Pericarditis 3Nonspecific ST and T changes 209Total 410

B.B.B. = Bundle branch block; R.V.H. = Rightventricular hypertrophy; L.V.H. = Left ventricularhypertrophy.

listed in table 1. One-half hour of cardiacpain was selected as the dividing point in aneffort to distinguish patients with anginapectoris from those in whom the cardiac painmight be associated with myocardial necrosis.The remaining 106 patients with elee-

trocardiographic ST-segment and T-wavechanges had had no digitalis therapy or recentcardiac pain, and the electrocardiographicabnormalities did not satisfy the criteria forany specific abnormality. By correlation withthe clinical histories a possible etiology forthe ST-segment and for T-wave changes wasuncovered in 57 of these 106 patients(table 3).Clinical Correlation with Depth of ST-Depressionand T-wave Inversion

In the group of patients with no apparentexplanation for the ST-segment and T-wavechanges, 84 per cent had less than 0.5 mm. ofST-segment depression and less than 1 mm.of T-wave inversion; only 32 per cent of thegroup for whom there was a postulated ex-planation had such minimal changes (table 4).Frequency of Combined and Isolated ST and T-waveChanges

In the patients with recent cardiac painof 1/2 hour or longer duration, 11.4 per centof the electrocardiograms were classified as

Table 2Clinica~l Analysis of 209 Cases with NonspecificST-segment or T-wave Changes

Digitalis therapyRecent cardiac painY2 hour or longerRecent cardiac painless than 1/2 hourNo recent cardiac pain

Total

46

35

20106209

abnormal solely on the basis of T-wave inver-sion (table 5); in the group with recent car-diac pain of less than one-half hour, 4.6 percent were classified as abnormal on this basis.By contrast, of the patients with no recentcardiac pain 42 per cent were classified asabnormal, solely because of T-wave inversion.Depth of T-wave Inversion and ST-Segment Depres-sion

The depth of T-wave inversion was 2 mm.or more in 10 (28.6 per cent) of the 35 caseswith recent cardiac pain lasting 1/2 hour orlonger, in 11 (50 per cent) of 22 cases withrecent cardiac pain for less than 1/2 hour,and in only 12 (11 per cent) of the 106cases without recent cardiac pain (table 6).A clinical review of the 12 patients withT-wave inversion of 2 mm. or greater, butno history of recent cardiac pain disclosedthat six patients had blood pressures of170/90 or above, two had abnormal serumelectrolytes, one had central nervous systemdisease, two had postoperative shock, and onearteriosclerotic heart disease.The depth of ST-segment depression was

also studied in relation to the same clinicalgroups (table 7). T-wave inversion between3/4 and 2 mm. in depth did not appear to besignificantly different in the groups with or

without recent cardiac pain. Recent cardiacpain had been present in all 10 patients withelectrocardiographic ST-segment depressionsof more than 2 mm. ST-segment depressionexceeding 0.5 mm., but less than 2 mm.occurred much more frequently in patientswith recent cardiac pain than in those withoutsuch pain. There were only 14 among the 106patients in the category with "no recent

Circulation, Volume XXIII, May 1961

656

by guest on June 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

"NONSPECIFIC' ST AND T-WAVE CHANGES

Table 3Possible Etiologies of One-Hundred and Six Cases with Nonspecific ST and T-WaveAbnormalities

ar.

i t0r.00 0

04 P.

U)~~~~~~~~~~~~~~~~.

CD

4)

Y.

0g

T 25 2 6 1 2 1 4 2 1 0 1 0 45ST and T 19 4 6 2 0 2 5 0 2 2 4 2 48ST 5 2 2 2 0 0 1 1 0 0 0 0 13Totals 49 8 14 5 2 3 10 3 3 2 5 2 106

ASHD, Arteriosclerotic (coronary) heart disease; HBP, hypertension; CNS, centralnervous system.

cardiac pain " with ST depression greaterthan 0.5 mm. Three of these patients had ablood pressure of 170/90 or more, four hadclinical arteriosclerotic (coronary) heartdisease, one had thyrotoxic heart disease, onehad abnormal serum electrolytes, two hadsevere cerebral disease, one had sprue withnormal serum electrolytes, and two had noclinical heart disease but were over 80 yearsof age.

Ischemic ST Segment and Ischemic T Waves

The clinical groups with and without car-diac pain were correlated with the presence ofthe ischemic type of ST segment and ischemicT wave (table 8). There was no significantdifference in frequency of the ischemic con-tour of the ST segment among the threegroups. On the other hand, there was adefinitely higher incidence of ischemic typeof T wave in the groups with recent cardiacpain than in those without pain, and the dif-ference was even greater in terms of thecombination of ischemic ST segment andischemic T wave.

DiscussionOf the 410 abnormal electrocardiograms

more than 50 per cent were reported as show-ing nonspecific ST-segment or T-wave abnor-mnalities. The 209 cases in the latter category,when correlated with the clinical history fellinto four groups:

1. Forty-six (22 per cent) had receiveddigitalis, which could account for the ST-Tchanges;Circulation, Volume XXIII, May 1961

2. Forty-nine (24 per cent) had no ap-parent cause for the ST-T abnormalities.

3. Fifty-seven (27 per cent) had an acuteepisode of cardiac pain within 5 days priorto the electrocardiogram; the pain lasted ahalf-hour or longer in 35 and less than 1/2hour in 22;

4. Fifty-seven (27 per cent) had no recentcardiac pain but suffered from some diseasethat could be regarded as a possible cause forthe electrocardiographic changes.A consideration of these groups suggests

an approach to a more definite diagnosis than"nonspecific ST and T-wave abnormalities."It seems desirable first to determine whetherthe patient has been receiving digitalis. Al-though characteristic ST-T-wave changeshave been described following digitalis thechanges are not always characteristic.

If digitalis is excluded as the cause ofelectrocardiographic changes, the problem isto determine whether the ST-T changes de-veloped in connection with recent cardiac painand may therefore indicate myocardial is-chemia or necrosis. If there is no history of arecent acute attack of cardiac pain, it isnecessary to distinguish between ST and T-wave changes due to some competent clinicalcause (other than acute myocardial ischemiaor necrosis) and ST and T-wave changes thatare of no clinical significance. This study hasindicated that such differentiation can almostalways be made on the basis of the clinicalhistory and the observation that the combina-tion of ST-segment depression of less than

657

p

by guest on June 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

FRIEDBERG, ZAGER

Table 4Contrast of ST-Segment Depression or T'-Wave Inversion in Group with No ApparentExplanation and Group with Possible Explanation for Abnormalities

Insignificant ST depression Isolated T Isolated STST and T and inversion depressionchanges* T inversiont more than 1 mm. more than % mm.No. % No. % No. % No. % Total

Apparentexplanation 18 32 20 35 16 28 3 5 57No apparentexplanation 41 83.7 0 0 7 14.3 1 2 49

*ST depression less than 0.5 mm.; T flat or inverted less than 1 mm.tST depressed more than 0.5 mm.; T inverted more than 1 mm.

Table 5Classification Based on Isolated T-Wave Inversion, ST-Segment Depression, or Com-bined T-Wave Inversion and ST-Segment Depression

Both ST and T Only T-wave Only ST segmentabnormalities inversion depressionNo. % No. % No. % Total

Recent cardiac pain1/2 hr. or longer induration 25 71.4 4 11.4 6 17 3.5Recent cardiac painless than 1/2 hr. induration 19 86.4 1 4.6 2 9 22No recent cardiac pain 57 53.7 45 42.4 4 3.6 106

0.5 m111n1. and T-wave inversion of less than 1mm. is unaccompanied by a reasonable causefor electrocardiographic changes and thatsuch minor ST-T changes are probably withinthe range of normal. The hyperventilationsyndrome or unrecognized physiologic changesmust be excluded.

In this series of 106 cases of nonspecific STand T-wave changes in patients without recentacute cardiac pain there appeared to be a

reasonable explanation in 57 patients withsuch clinical abnormalities as pericarditis,postoperative shock, clinical chronic cor pul-monale, pulmonary embolism, electrolyte dis-turbances, and chronic angina etc. (table 3).Intraeranial disease, which included 10 casesin this series, has been reported as a cause ofabnormalities of the ST segment or T wave,or both, when as a direct result of cerebrallesions or mediated through an alteration inserum electrolytes.4 In almost all of the 49eases in which there was no apparent clinicalVanise, the ST-segment and T-wave changes

were of minimal degree and may have beeninsignificant. In the cases of ST-segment andT-wave inversion, which were found in pa-tients with recent acute cardiac pain, theproblem was to determine whether the electro-cardiographic changes actually developed inconnection with the acute pain or were pre-viously present. This could be resolved by astudy of previous and subsequent serialelectrocardiograms.

Cases with recent cardiac pain could gener-ally be differentiated from those without suchpain by an ST-segment depression of morethan 1/2 mm. in the former. In no instancewas the ST segment depressed more than 2mm. in cases without recent cardiac pain,whereas this degree of depression was notuncommon in cases with recent cardiac pain.The difference between cases with and withoutcardiac pain was more striking if 5 mm. ofinversion of the T wave was used as the cri-terion. In the cases with recent cardiac painST-segment depression alone or in combina-

Circulation, Volume XXIII, May 1961

658

by guest on June 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

"NONSPECIFIC" ST AND T-WAVE CHANGES

Table 6Depth of T-Wave Iniersion

More than5 mm. or 0.5 mm. up 0.5 mm. Normal Totalgreater 2-5 mm. to 2 mm. or less T wave cases in

No. % No. % No. % No. % No. % group

Recent cardiac pain1/2 hr. or longer induration 6 17.1 4 11.4 12 34.3 1" 34.3 1 2.9 335Recent cardiac painless than 1/2 hr. induration 6 27.4 5 22.7 7 31.8 3 13.6 1 4.5 22No recent cardiac pain 4 3.8 8 7.8 43 42.1 47 46 4 3.8 106

tion with T-wave inversion occurred moreregularly than in those without recent cardiacpain; in the latter group T-wave inversionwas more likely to appear as an isolatedfinding.

Attention was directed to deep T-wave in-versions in the midprecordial leads as de-scribed by Pruitt et al.4 They reported astudy of 62 cases with at least 5 mm. of T-wave inversion which was maximal in lead V3.In 38 of these there was clinical evidencesuggesting myocardial infarction or severeischemia, but in 24 no such clinical evidencewas present. Four of the patients in the groupwith cardiac pain who died and were studiedat autopsy showed evidence of subendocardialinfarcts. In our series there were 12 casesthat showed electrocardiographic changessimilar to those in Pruitt's cases. In seven ofthese there had been recent cardiac pain of1/2 hour or longer duration; in two, recentcardiac pain of less than 1/2 hour duration;and in three, no recent cardiac pain.An analysis of the ST and T contours dis-

closed no significant difference between thefrequency of an "ischemic" type of ST seg-ment in patients with and without recentcardiac pain. But if the criterion of 3/4 mm.or greater depression of the ST segment isadded to that of ischemic contour, the combi-nation was found in 78 per cent of the casesof recent cardiac pain and in only 21 per centof those without such pain. The presence ofan "ischemic" contour of T wave appearedto show a good correlation with recent cardiacpain, since such a T wave was observed in 49per cent of the cases with recent cardiac painCirculation. Volume XXIII, May 1961

and in only 10 per centpain.

of those without such

To summarize, the probability that the ST-segment depressions and T-wave inversionswere associated with and due to the episodeof recent acute cardiac pain was indicated bythe combination of ST depression ad T-waveinversion with a depth of more than 1/2 mm.of the ST depression and more than 1 mm. in-version of the T wave, by isolated T-waveinversion of more than 2 mm. and especiallyby more than 5 mm. in the midprecordialleads, and by an ischemic contour of the STsegment that is depressed more than 1/2 mm.The combination of such electrocardio-

graphic findings and a history of recent acutecardiac pain usually indicates that the elec-trocardiographic changes were caused by re-cent myocardial ischemia or necrosis. How-ever, it is desirable to confirm such aninterpretation by a comparison with previouselectrocardiograms if available. Furthermore,serial electrocardiograms showing progressivechange or disappearance of the abnormalitiesusually attest to a relationship to the recentcardiac pain but care should be taken to besure that there is identical electrode place-ment and minimal changes should not be em-phasized. The effort to differentiate ST-Tchanges denoting acute myocardial ischemiafrom acute myocardial necrosis was made bycorrelating these changes with recent cardiacpain of less than or more than 1/2 hour dura-tion. It was thought that the pain of brieferduration might represent myocardial ischemiaand that of longer duration might indicatemyocardial necrosis. But no significant dif-

659

by guest on June 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

FRIEDBERG, ZAGER

Table 7Depth of ST-Segment Depression

More than Isoelec-Greater 1/2 mm. up to %2 mm. or tric ST Total

than 2 mm. 2 mm. less segment cases inNo. % No. % No. % No. % group

Recent cardiac pain1/2 hr. or longer induration 6 17.1 16 45.7 11 31.4 2 5.7 35Recent cardiac painless than Y2 hr. iniduration 4 18.2 10 45.4 8 36.4 0 0 22No recent cardiac pain 0 0 14 13.2 47 44.3 45 42.5 106

Table 8Contrast of Contour Abnormalities in Cases with and without Recent Cardiae Pain

NeitherIschemic ischemicST seg- Ischemic ST seg-ment and Ischemic ST ment norischemic T wave segment ischemic TotalT wave only only T wave cases In

No. % No. % No. % No. % group

Recent cardiac pain1/2 hr. or longer induration 14 40 6 17.1 10 28.6 3) 14.3 33Recent cardiac painless than 1/2 hr. induration 6 27.3 2 9.1 7 31.8 7 31.8 22No recent cardiac pain 8 7.5 3 2.8 29 97.3 66 62.2 106

ferences could be found in the degree ofcharacter of ST and T-wave changes in thesetwo groups.

Several considerations may explain the lackof electrocardiographic difference in these twogroups of cases classified according to theduration of acute cardiac pain. Many of thepatients with recent cardiac pain for morethan 1/2 hour unquestionably experienced ex-tensive myocardial necrosis. But when this wasindicated by Q waves as well as ST-T changesthey were separated from the group with non-specific ST-T abnormalities. Evidence hasbeen presented by Cook and others5 that acutemyocardial necrosis is associated with ab-normal Q waves as well as ST and T changeswhen more than 75 per cent of the thicknessof the myocardial wall underwent necrosis.When subendocardial necrosis did not exceed50 to 75 per cent of the thickness of theventricular wall there was only ST-segmentdepression and T-wave inversion. Suchchanges are indistinguishable from subendo-

cardial ischemia in the same area. In fact it isprobable that the ST-segment depression andT-wave inversion observed in patients withsubendocardial necrosis are due to the associ-ated ischemia rather than to the necrosis itself.Thus, if subendocardial necrosis occurred inthe patients with pain lasting more than 1/2hour, the electrocardiographic changes wouldbe undistinguishable from those in patientsexperiencing recent cardiac pain for lessthan 1/2 hour even if ischemia and no necrosiswas present in the latter group. A differentia-tion between subendocardial ischemia andsubendocardial necrosis would be indicated bydifferences in the duration and severity of thepain and by various observations and lab-oratory findings. The electrocardiographicchanges themselves could only be reported asindicating subendocardial ischemia or ne-crosis.

SummaryA study was made of 1,000 consecutive

adult in-patient electrocardiograms to deter-Circuleation, Volume XXIII, May 1961

660

by guest on June 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

"NONSPECIFIC" ST AND T-WAVE CHANGES

mine the possibility of making a more precisediagnosis than "nonspecific ST and T-wavechanges. " More than 50 per cent (209) ofthe 410 abnormal electrocardiograms (exclu-sive of arrhythmias) were characterized bynonspecific depression of ST segment or Twave inversion, or both.

These 209 cases comprised four groups: (1)46 patients (22 per cent) had received digi-talis, which could account for the ST-Tchanges; (2) 57 patients (27 per cent) hadhad an acute episode of cardiac pain within5 days prior to the electrocardiogram; (3) 57patients (27 per cent) had had no recent car-diac pain but suffered from some disease thatcould be regarded as a possible cause for theST-T changes; and (4) 49 patients (24 percent) had no apparent cause for the electro-cardiographic changes. In the last group, incontrast with the others, the ST-segment de-pression was less than 0.5 mm. and the T-waveinversion less than 1 mm.

In patients with recent cardiac pain, ascontrasted with those without such pain, theelectrocardiograms were characterized by acombination of ST depression and T-waveinversion, with a depth of more than 1/2 mm.of the ST depression and more than 1 mm.inversion of the T wave, by isolated T-waveinversion of more than 2 mm. and especiallyby more than 5 mm. in the midprecordialleads, and by an ischemic contour of the STsegment that is depressed more than 2 mm.

There was no definite difference in theelectrocardiographic findings in the patientswith recent cardiac pain of less than ½2 hourand those with recent cardiac pain of morethan ½/2 hour when the cases with transmuralinfarction (Q-wave changes) were excluded.This is compatible with the concept that theelectrocardiogram usually does not distin-guish between subendoeardial ischemia andsubendocardial necrosis. The latter differenti-ation would depend on multiple clinieal andlaboratory findings.

References1. PARDEE, H. E. B.: Heart disease and abnormal

electrocardiograms with special reference tocoronary "T" wave. Am. J. M. Sc. 169: 270,1925.

2. ROTHSCHILD, M. A., MANN, H., AND OPPEN-HEIMER, B. S.: Successive changes in theelectrocardiogram following acute coronaryartery occlusion. Proc. Soc. Exper. Biol. &Med. 23: 253, 1926.

3. BuRcH, G. E., MYERs, R., AND ABILDSKOV, J. A.:A new electrocardiographic pattern observedin cerebrovascular accidents. Circulation 9:719, 1954.

4. PRUITT, R. D., KLAKEG, C. H., AND CHIAPIN,L. E.: Certain clinical states and pathologicchanges associated with deeply inverted Twaves in the precordial electrocardiogram.Circulation 11: 517, 1955.

5. COOK. R. W., EDWARDS, J. E., AND PRUITT, R. D.:Electrocardiographic changes in acute sub-endocardial infarction. Parts I and II. Circu-lation 18: 603, 1958.

The fantastic economies of successful research swiftly go beyond any method ofaccurate accounting. As an example, I would estimate that the work on the sulfa drugsup to and including the proof of their efficiency in the treatment of lobar pneumonia didnot cost more than $150,000. But in as short a time as three years after their value inpneumonia was shown, the saving in life insurance policies that would have been paidby one life insurance company for deaths from lobar pneumonia alone only, on the WestCoast of the United States, over one year only, was calculated at $3,000,000. In such waysresearch gives a continuing and a permanent lift to Great Medicine.-ALAN GREGG, M.D.,Challenges to Contemporary Medicine. New York, Columbia University Press, 1956, p. 71.

Circulation, Volume XXIII, Meay 1961

661

by guest on June 13, 2018http://circ.ahajournals.org/

Dow

nloaded from

CHARLES K. FRIEDBERG and ALBERT ZAGER"Nonspecific" ST and T-Wave Changes

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1961 American Heart Association, Inc. All rights reserved.

75231is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TXCirculation

doi: 10.1161/01.CIR.23.5.6551961;23:655-661Circulation. 

http://circ.ahajournals.org/content/23/5/655located on the World Wide Web at:

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

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation 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

of the Web page under Services. Further information about this process is available in thewhich permission is being requested is located, click Request Permissions in the middle columnClearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of the CopyrightCirculationoriginally published in Requests for permissions to reproduce figures, tables, or portions of articlesPermissions:

by guest on June 13, 2018http://circ.ahajournals.org/

Dow

nloaded from