report of committee on...

REPORT OF COMMITTEE ONELECTROCARDIOGRAPHY,

AMERICAN HEART ASSOCIATION

Recommendations for Standardization ofLeads and of Specifications for Instruments

in Electrocardiography andVectorcardiography

By COMMITTEE MEMBERS: CHARLES E. KOSSMANN, M.D., CHAIRMAN,DANIEL A. BRODY, M.D., GEORGE E. BURCH, M.D., HANs H. HECHT, M.D.,

FRANKLIN D. JOHNSTON, M.D., CALVIN KAY, M.D., EUGENE LEPESCHKIN, M.D.,HUBERT V. PIPBERGER, M.D., AND by MEMBERS OF THE SUBCOMMITTEE

ON INSTRUMENTATION: * HUBERT V. PIPBERGER, M.D., CHAIRMAN,GERHARD BAULE, PH.D., ALAN S. BERSON, M.S., STANLEY A. BRILLER, M.D.,

DAVID B. GESELOWITZ, Ph.D., LEO G. HORAN, M.D.,AND OTTO H. SCHMITT, Ph.D.

T HE LAST REPORT of the Committeeon Electrocardiography of the American

Heart Association appeared in 1954.1 It wasconcerned with instrumentation, techniques,and nomenclature of electrocardiographic andvectorcardiographic leads. Its objective wasto bring about, so far as possible, some stan-dardization which would ensure uniformityin clinical application without discouragingor restricting continued investigation whereindicated.

In the 12 years since its appearance, thereport seems to have accomplished its purposeso far as conventional scalar leads are con-cerned. Electrocardiographic practice has be-

*Richard McFee, Ph.D. served on the subcom-mittee during its earliest deliberations.The section of these "Recommendations" concerned

with Instruments (p. 585) and prepared by theSubcommittee is being simultaneously published inthe IEEE Transactions on Bio-Medical Engineeringto facilitate its access to the engineering profession.Circulation, Volume XXXV, March 1967

come quite uniform throughout the world.The comfortable feeling of this orderly and

standardized state goes back to 1938 whenthe first efforts to establish uniformity in elec-trocardiographic practice began.2' Rapidprogress in recent years has resulted in the notunexpected expression of dissatisfaction withthe recommendations as they stand now. TheCommittee, therefore, considered the time ap-propriate to review the previous report andto make new recommendations as indicated.

Several areas of endeavor appeared toneed particular attention. With the adventof magnetic tape, analog-to-digital converters,and a variety of programs for analysis ofelectrocardiographic data by computer, itwas clearly propitious to review the entire,intricate problem of instrumentation in pre-sent-day electrocardiography. The Committeefelt that, because of technical advances inrecording devices and methods, advice inthis area should be obtained from experts

583

by guest on May 14, 2018

http://circ.ahajournals.org/D

ownloaded from

http://circ.ahajournals.org/

COMMITTEE ON ELECTROCARDIOGRAPHY, AHA

who are, in addition, familiar with the prac-tical problems of electrocardiography. Accord-ingly a Subcommittee on Instrumentation wasappointed, composed of physicians and bio-medical engineers. In the present report, thegreatly expanded section on INSThUMENTS isalmost exclusively the work of this productiveSubcommittee.An effort has been made to define recom-

mended specifications of recorders as preciselyas possible and to indicate ranges of latitudewhen indicated. In order to ensure bettermaintenance of specified performance charac-teristics, a description of a relatively simpledevice for testing direct-writing electrocar-diographs has been included and is recom-mended for use at regular intervals. The testscan be performed by a trained electrocardio-graphic technician. In addition it is recom-mended that recorders be checked regularlyby engineering personnel for a more thoroughperformance evaluation.No attempt was made to give detailed jus-

tification for each recommendation regardinginstruments in this short report. It is plannedto include these in a more detailed reportto follow.The major portion of the recommendations

on instruments deals with direct-writing elec-trocardiographs which are the most convenientand most widely used at this time. However,such recorders possess significant limitationsin fidelity of data reproduction, particularlyin the frequency range. Recommended specifi-cations for direct-writing electrocardiographsrepresent minimum requirements. Wheneverpossible a broader frequency bandwidth thanthey provide is desirable, particularly in elec-trocardiographic investigations. Recommenda-tions for electrocardiographic preamplifiers tobe used for wide-band recording have, there-fore, been made also. Such preamplifiersshould be incorporated in vectorcardiographsand in any other electrocardiographic record-ing apparatus not limited in frequency re-sponse by a direct-writing mechanism.As a consequence of the increasing interest

in automatic processing of electrocardiograph-ic data, specifications for magnetic tape re-

corders and analog-to-digital data convertershave become necessary. Recommendations re-garding these have been included in thepresent report. Considerable progress may beanticipated in this rapidly changing field oftechnology. This portion of the "Recommenda-tions" should be considered as a guideline ofa general nature, and users are urged to obtainmost recent information concerning equip-ment capabilities.That there is redundant electrocardiograph-

ic information in the usually recorded 12leads4 is generally agreed. However, whenthe suggestion is made to the clinician thatthe number of leads he is using now is exces-sive, resistance to change is encounteredwhich results probably from the combinationof long-ingrained habit and the inadequacyof clinicopathological and other correlationswith orthogonal leads.The problem of reducing the number of

leads for clinical usage is inextricably involvedwith the problem of selecting a standard ref-erence frame for recording vectorcardiograms.In each instance the immediate objective ofleading from the surface of the body is toget a record of the true orthogonal compo-nents-x, y, and z-of the heart regarded asan electrical generator. Whether this recordshould be in a scalar, vector, or digital formwill probably be ultimately determined bypractical considerations such as convenience,speed, accuracy, and personal preference. Theacquisition of the data on magnetic tape makesreproduction in any form desired a possibilitylimited only by the amount and cost of thedata-handling equipment.The numerous systems for vectorcardio-

graphic recording can be divided into twoclasses: the uncorrected and the corrected.In the former, electrode placements for ac-quiring orthogonal components are based onanatomic considerations. In the latter theplacements are based on electrical considera-tions arising out of the concepts of the leadvector5 and of image space.6 In these systemsthe lead, regarded as a vector, is adjustedin length and direction for acquisition of the

Circulation. Volume XXXV, March 1967

584



ownloaded from



desired lead orthogonality; hence they aredesignated "corrected."Although the corrected leads are based on

the assumption that the cardiac generator be-haves as a dipolar source of electrical energy,and indeed are usually designed to ignoremultipolar components, nevertheless, it is theopinion of the Committee that they arefounded on firm physical principles. In con-trast, uncorrected leads, designed for the mostpart before the lead vector concept or theclosely related lead field concept7 was knownto electrocardiographers, are based on erro-neous principles relative to the distributionof electrical currents in volume conductorsas now understood. Although these uncorrect-ed systems were admirable beginnings in anew modality at the time of their origin,were and continue to be of considerable prac-tical value to their adherents, and were mostuseful in laying the groundwork for furthervectorcardiographic investigation, they are notrecommended for general use for the reasonsgiven.The Committee deliberated on which of the

several corrected systems thus far describedshould be recommended. In making such achoice, consideration must be given to suchmatters as the present general or restricteduse of the method; the practicality of it interms of the number of electrodes and correc-tive resistors, especially as related to the inputimpedance of recorders; reproducibility ofresults; and possibly the similarity of therecommended orthogonal leads to scalar leadsnow used in clinical medicine so that theinfinite amount of information collected withthe latter and clinically correlated over theyears will not be entirely wasted. The conclu-sion reached was that no single one of thecorrected systems could be designated at thistime, but that a more or less natural selectionbased on the above considerations wouldoccur in the future.

In anticipation of the circumstance thateventually the same leads will be used forelectrocardiography and vectorcardiography,the Committee has changed the outline ofits recommendations compared to the reportCirculation, Volume XXXV, March 1967

of 1954. Rather than treat electrocardiographicand vectorcardiographic leads as entities inthemselves with a consideration of instru-ments, techniques, and nomenclature undereach, the report is simply divided intoINSTRUMENTs and LEADS with appropriatesubdivisions. Further, since most of the pre-viously reported recommendations on tech-niques and nomenclature are available at twopermanent sources," 8 only those aspectswhich have needed amplification or modifi-cation are considered in the present report.

InstrumentsA. Direct-Writing Electrocardiographs

1. System performance, linearity, and dis-tortionThe deviation of the recorded output

from an exact linear representation ofthe input signal shall not exceed 5% ofthe peak-to-peak output for amplitudesbetween 5 and 50 mm. For peak-to-peakamplitudes below 5 mm, this deviationshall not exceed 0.25 mm. The inputsignal may be comprised of frequencycomponents between 0.05 and 100 Hzin any combination. For signals contain-ing frequency components above 50 Hz,the peak-to-peak output amplitudes con-tributed by these components need beno greater than 5 mm to meet this per-formance requirement.

2. Input rangeThe instrument shall meet specifica-

tions with input amplitudes up to 10millivolts peak-to-peak and shall func-tion accurately for any input signal andelectrical or mechanical offset so longas the ideal response will require thewriting point to remain on the ruledportion of the recording chart.

3. Input impedance and currentIn each position of the lead switch

the magnitude of the input impedanceover the working frequency range shallbe no less than 500,000 ohms betweenany single patient electrode and ground.This measurement is to be made withall patient electrodes grounded exceptthe one under consideration.

585



ownloaded from



The instrument shall not cause cur-rents greater than 1.0 microampere toflow in the circuits to the patient.

Because of offset potentials which mayexist at the electrodes, the instrumentshall be capable of meeting all specifica-tions when differential offset voltages upto 100 millivolts and common mode off-set voltages up to 200 millivolts arepresent.

4. Central terminalThe magnitude of the deflections in

all leads, including the augmented leads,referred to the central terminal,4 mustnot deviate from their correct values bymore than an additional 2% from theallowable deviations specified in para-graph A1.*

5. GainThe gain is to be adjustable from the

panel of the instrument in three clearlylabeled fixed steps, having the followingvalues:

*The recommendations for input impedance (par.A3) and central terminal accuracy (par. A4) com-bined with the recommendation on common moderejection listed below (par. A9) impose certain strin-gent requirements on the input circuitry. If the latteris conventional in type, the values for the centralterminal resistors must be more than 300,000 ohmseach. If resistances less than this are used, voltagesof the leads are likely to be distorted. To meetthe specification for common mode rejection, a re-sistor which has one third the value of the central-terminal resistor may need to be inserted in serieswith the exploring electrode. Under such circum-stances the differential input impedance of the pre-amplifier would need to be greater than 20 megohms,with the input impedance of each input to signalreference balanced to a precision on the order of1.0%.The recommendations on input impedance and

common mode rejection may be met, using lowervalues for central terminal resistors and with a pre-amplifier of lower input impedance by using otherthan conventional circuitry, e.g., a buffer amplifierfor each active electrode.

These comments apply also to weighting networksother than the central terminal. For example, in acorrected orthogonal lead system, the value of theresistors chosen must be sufficiently high to meetthe input impedance specifications.

10 mm per millivolt5 mm per millivolt20 mm per millivoltAny continuous or vernier gain adjust-

ment should be available as a restrictedaccess control to be operated, for ex-ample, by screwdriver. Since the needto use this adjustment may often becaused by deterioration of the calibrating(standardizing) signal (par. A13), anote to this effect should be placednear this control.

6. Stability of gain and base lineTo verify stability of gain and base

line, the following test may be per-formed: Connect the right leg (RL)terminal, in each instance through a re-sistor of 20,000 ohms, to the terminalsof the right arm (RA), the left arm (LA),the left leg (LL), and the chest.a. With the lead selector switch on Lead

I and the sensitivity switch on 20 mmper millivolt, turn the machine onafter it has not been used for at least1 hour. After 3 minutes, center thetrace. During the next 12 minutes thebase line should not drift more than10 mm. During the following 45 min-utes, the base line should not driftmore than an additional 2 mm. After1 hour, turn the lead selector switchthrough all positions. In each casethe base-line level after the resetbutton has been pressed should notshift by more than 1 mm from itsvalue in Lead I.

b. With the lead selector switch on LeadI and the sensitivity switch on 20 mmper millivolt, turn the machine on afterit has not been used at least 1 hour.After 3 minutes, center the base line.Press the 1 millivolt calibration button.The deflection of approximately 20mm should differ by less than 1 mmfrom the deflection measured after 1hour.

c. Turn the instrument off. Turn it onagain 1 hour later. The deflectionmeasured in the Lead I position after

Circulation, Volume XXXV, March 1967

586



ownloaded from



OUTPUTI NPUT

UPPER LIMIT E

LOWER LIMIT

0.05 0.14 50 100FREQUENCY c.p.s.

Figure 1

Recommended minimum frequency response of direct-writing electrocardiographs. On theordinate the ratio between input and output is indicated in per cent. The response of the re-

corder shall be between the upper and lower limits as indicated by the solid lines in thefigure.

1 hour's warm-up should differ by lessthan 0.5 mm from that measured instep b.

7. OverloadThere shall be no damage to the in-

strument when it is subjected to 1.0 voltat any frequency from 47 to 63 Hzapplied for 2 seconds to the inputterminals with the controls set at anysensitivity or lead position.

8. Frequency response (fig. 1) *In addition to the linearity and dis-

*The Committee regards these as minimal frequen-cy characteristics applicable to direct-writing electro-cardiographs. In order to display high-frequency com-ponents of the electrocardiogram, instruments shouldhave a response extending to at least 500 Hz. Foranalysis of these components, paper or film speedsof 200 mm per second or more appear necessary.Presently available information suggests that therewill be no significant loss of electrocardiographic de-tail when a high-frequency cutoff of 2,500 Hz isemployed.Circulation, Volume XXXV, March 1967

tortion requirements specified in para-

graph Al, the following design charac-teristics are acceptable with constantamplitude sinusoidal input signals:a. From 0.14 to 50 Hz, the response shall

be flat to within +6% (+0.5 dB). Theresponse down to 0.05 Hz shall not bereduced by more than 30% (-3 dB)from the response at 0.14 Hz. Thisrequirement corresponds to a "timeconstant" of at least 3.2 seconds, where"time constant" refers to the time re-

quired for a direct current step input(such as the calibration voltage) todecay to 36.8% of its original magni-tude.

b. With an amplitude response of 5 mmpeak to peak at 50 Hz, the response toconstant amplitude sinusoidal inputsignals up to 100 Hz shall not be re-

duced by more than 30% (-3 dB),leaving an amplitude of at least 3.5mm at 100 Hz.

HO1009080706050403020100 I I

587

-1 I I I



ownloaded from



c. The response shall at no frequencyexceed the restraints specified for therange of 0.14 to 50 Hz.

9. Common mode rejectionFor each position of the lead switch

with the recorder gain set at 10 mmper millivolt, and with all active elec-trode leads connected together, a poten-tial difference of 100 millivolts peak-to-peak applied between them and theright leg lead shall cause no more than1.0 mm peak-to-peak deflection for fre-quencies from 45 to 65 Hz. In a likemanner the application of 10 millivoltsshall cause no more than 1.0 mm peak-to-peak deflection at any frequency. Thisspecification must be met when a re-sistance of 5,000 ohms is placed in serieswith any one electrode lead.These requirements correspond to a

common mode rejection of 1,000:1 at fre-quencies between 45 and 65 Hz, and100:1 at any other frequency.

10. Noise levelUpon simulating a subject by means

of completely shielded resistors of25,000 ohms placed between each patientlead and ground, the output noise, withthe recorder calibrated to 10 mm permillivolt, shall not exceed the equivalentof 0.1 mm root mean square in anyposition of the lead switch (10 micro-volts root mean square referred to theinput) .

11. Radio frequency interferenceManufacturers are urged to minimize

interference from radio and other highfrequencies through proper design ofcircuits, shielding, and filtering in thepower supply.

12. GroundingLine-operated electrocardiographs shall

be supplied with a three-terminal, power-line plug.The existence of multiple connections

between a subject and a line-operatedelectrocardiograph creates the potentialhazard of uncomfortable or even lethalelectric shock. Protection of the patient

and the operator from currents greaterthan 5 milliamperes must be providedby fuses or circuitry to furnish equal pro-tection under any of the following cir-cumstances present singly or in anycombination:a. When the subject is inadvertently or

purposely grounded by connection toa second device.

b. When defective or frayed insulationof the power cord of any transformer,motor, or other line-operated compo-nent may make the case containingthe electrocardiograph "hot."

c. When an incorrectly wired, three-wire receptacle is used to energizethe electrocardiograph.

d. When an internal short circuit orcomponent failure may occur in theamplifier connected to the patient.

e. When the subject may be wired di-rectly to the instrument case.Since satisfactory recording in electri-

cally noisy locations may be possible onlyby connecting the instrument directlyto a ground more suitable than can beobtained through the power-line plug,provision for this alternate mode of oper-ation must be made. With or withoutsuch a connection, leakage of currentback to the power-line ground or tothe auxiliary ground for either polarityof the power line, must be less than50 microamperes.

Currents considerably less than 5 mil-liamperes in magnitude may cause ven-tricular fibrillation in a patient whena saline-filled catheter or other exter-nally accessible, artificial conducting con-nection to the heart is present. The usermust be warned of the potential hazardof a line-operated or battery-operatedelectrocardiograph under these specialcircumstances since the protection againstelectrical shock ordinarily afforded bythe instrument may be inadequate.

13. Calibration ("standardization")The standardizing voltage shall be a

signal of 1.0 millivolt ±2% with a timeCirculation, Volume XXXV. March 1967

588



ownloaded from



constant no less than 100 seconds and an

output impedance of 1,000 ohms or less,which can be applied continuously to theinputs by a switch on the panel of theinstrument. When a multichannel re-

corder is used, means shall be providedfor the simultaneous application of thecalibration signal to all channels. Theinstrument shall be capable of recordingthis standardizing voltage as follows:a. Superimposed in series while re-

cording the electrocardiogram fromany lead position, and

b. WVith the patient disconnected fromthe instrument.In addition, the standardizing signal

shall be available through a suitable con-

nector to provide for checking of itsaccuracy.

If a battery is used for obtaining thestandardizing voltage, the instrumentshall be provided with an indicator (apush-to-test control is suggested) to alertthe user when the battery needs replace-ment; for example, battery is no longercapable of supplying the standardizingvoltage to the required accuracy.

14. Speed and speed accuracy

A minimum of 2 speeds, 25 mm per

second and 50 mm per second, shall beavailable. Accuracy of speed shall be+2% when operating from a 60-cyclesource.* Time markings at intervals of1 second +2% shall be recorded at allspeeds at an edge of the time rulingsof the recording paper by a device oper-

ating independently of the transportmechanism of the recorder.

15. Recording paper

The recording paper shall be ruledwith 1.0 mm divisions along both the timeand voltage axes. Every fifth division

*Accuracy requirements of ±2% have been specifiedto permit use of synchronous timing devices andsynchronously operated paper transports. If such de-vices are used, they will be unable to meet theaccuracy requirement when the line frequency de-viates appreciably from 60 cycles; therefore, a noteto this effect should be placed on the instrument.


shall be ruled darker than the others.The 1.0 mm divisions shall not deviateby more than +1% from 1.0 mm in eitheraxis. The ruled divisions shall cover atotal of 5 cm in width, and recordingsshall be rectilinear.These dimensional accuracies shall be

met throughout a temperature rangefrom 10 C to 50 C and for relativehumidities from 10 to 80%.

16. Skewa. Skew of recording, due to all causes,

shall not exceed 0.1 mm of horizontaldisplacement per 1.0 cm of verticaldeflection.

b. In multichannel recording, it is es-pecially important to ensure temporalalignment of traces. Therefore, withthe amplifiers for each channel set tothe same frequency response limits, alltraces shall fall dynamically withina 0.5 mm band of the ideal (zeroskew) response at all transport speedsfor the entire frequency range of theinstrument.

17. Recorded outputThe vertical width of the undeflected

trace shall not exceed 1.0 mm at anypaper transport speed. Both the upstrokeand downstroke of the writing deviceat rates of deflection less than 1,000 mmper second shall leave a visible, contin-uous trace with sharp edges.

18. Auxiliary outputA jack shall be available for gaining

access to the output. The signal at thisjack shall have a driving capability ofat least 1.0 milliampere for loads of1,000 ohms or more at ±1.0 volt, fullscale. The output impedance shall beless than 100 ohms. The output charac-teristics shall be the same as specifiedherein except that the frequency responseshall extend to at least 1,000 Hz (1 dBdown). This output shall have a dcoffset voltage no greater than +1 volt,which can be centered at zero volts.No damage.shall occur to the instru-

ment when the output is short circuited.

589



ownloaded from



Upon removal of the short circuit, theinstrument shall be capable of meetingspecifications.

19. Trace resetA capability for trace reset must be

provided by a switch or button on theinstrument. After a pulse of 10 millivoltshas been applied to the input for 10seconds at a gain of 20 mm per millivolt,depressing this button shall cause thetrace to return to within 1 mm of itsinitial position within 0.5 second. Re-leasing the button after it has beendepressed for not more than 0.5 secondshall result in no more than 1 mm ofadditional displacement.

20. Power-line variationsFor instruments operating at 60 Hz

and 120 volts, all specifications shall bemet when used in the range of power linevoltages from 95 to 135 volts and fre-quencies from 57 to 63 Hz. A disturbanceof +5 volts in the power line should notcause a deflection greater than that whichwould be obtained with an input signalof +50 microvolts.For battery-powered instruments, an

indicator shall be provided to alert theoperator when the batteries need replace-ment or recharging.

21. Temperature and altitudeThe instrument shall meet all specifi-

cations over the temperature range of10 C to 50 C, at altitudes from 0 to3,000 meters above sea level and at rela-tive humidities of 5% to 95%.

22. Electrode impedanceIt is desirable that any paste or jelly

and electrode combination which is usedresults in a skin-to-electrode impedanceof 5,000 ohms or less, measured at 60 Hzwith currents not exceeding 100 micro-amperes.*

*Electrodes of small size such as used in pediatricpractice, inadequate preparation of the skin, or poorquality electrode paste may lead to considerably high-er skin-to-electrode impedances, particularly at lowerfrequencies. Resulting signal distortions may be re-duced to acceptable levels by use of other than con-

23. Standardization of controls, cables, leg-ends, and recording format

Electrocardiographs meeting this spec-

ification shall be equipped with the fol-lowing controls labeled as shown inquotation marks.

"PaperSpeed""On-Off"

"1 mV""Lead"

"Center"

"Reset"

"Gainverier"

"Sensitivity"

The followingbe used:

"RA""LA""CL"ccRL"''IC"

Two-positionswitchTwo-positionswitch

Push buttonMultipositionswitch

Potentio-meter

Push buttonor switchThis poten-tiometer shallbe a recessedcontrol be-hind a hingedplate.Three-po-sition switch

"25 mm/s""50 mm/s"

(Light or

flag indicat-ing equip-ment is on.)

<I'l, sIII ,

"CaVR," "aVL,"aVF,"V

"5 mm/mV,"10 mm/mV.""20 mm/mV"

cable legends and colors shall

Right arm

Left arm

Left legRight legChest

whiteblackredgreenbrown

24. Testing device for direct-writing elec-cardiographsRandom tests have indicated that of

various types of direct-writing electrocar-diographs in common clinical use, a

ventional input circuitry, e.g. buffer amplifiers foreach active electrode.


590



ownloaded from



Al tests to be performed with lead switch In "V" position

To Electrocardiograph Patient Cable

R.L. L.A. R.A. L.L. CHEST

p..~~~~7r /

rC / F-Ov /

All resistors are ± 1% toleranceCapacitor is t2%tolerance

relatively large number failed to meetthe American Heart Association's perfor-mance recommendations of 1954.1 Sincedeterioration of the electrical character-istics of an instrument is not readilyapparent from examination of a stan-dardization pulse or a clinical record,a simple device for testing electrocardio-graphs is advocated. The testing system,which may be an accessory, or be self-contained within the electrocardiograph,employs a switch, seven resistors, and acapacitor (fig. 2A and B). By seriallyadvancing the switch and depressing thecalibration button, the low-frequency re-sponse, linearity, high-frequency re-sponse,* input impedance, and common

*A modification of the procedure proposed byDower and co-workers.9

Figure 2Suggested device for checking periodically the lowand high-frequency response, linearity, and input im-pedance and common mode rejection of direct-writingelectrocardiographs. The circuit of the testing device(A) is shown on the left above with the test signals,tests, and required readings in the respective columnsbelow (B).

SWITCHPOSITION

0.

TESTSIGNAL

None

1. 1-mv step tochest lead

2. 2-mv step tochest lead

TESTS

Low frequency response(time constant)

Linearity

READING

When calibrate button is helddown, the time for a 10mm/mvsignal to decay to 3mm shouldbe equal or longer than 3.2 sec.

Upward deflection of20mm ± 0.5 mm

3. 2-mv step tocentral ter-minal

Linearity Downward deflection of20 mm ± 0.5 mm

4. 1f -1112 myvI---I i-l.0 mSe-c

5. 10 mv step to chestlead through 10 kQresistor and to centraltermino! through 5 kQ


High frequency response

Input impedance andcommon mode rejectionratio

Deflection shou!d be equalto or larger than 5rrmm

Def'ection should beless than Imm

591



ownloaded from



mode rejection can be evaluated. Theresult of each test can be read as anamplitude of the deflection of the stylus.The one exception is the test of low-frequency response which is evaluated interms of the time required for a 10 mmper millivolt signal to decay to 3 mm.Although the foregoing system will per-

mit regular testing to be done easily byan electrocardiographic technician, it isnot intended that such tests be regardedas equivalent to a complete performanceevaluation necessary to satisfy all rec-ommendations in this report.

B. Recommendations for ElectrocardiographicPreamplifiers for Use with Recording DevicesOther than Direct-Writers

1. System performance, linearity, and dis-tortionThe deviation of the output signal

from an exact linear representation of theinput signal shall not exceed 1% of thepeak-to-peak output amplitude for am-plitudes between 2 and 20 volts. Forpeak-to-peak output amplitudes below2 volts, this deviation shall not exceed0.02 volt. The input signal may becomprised of frequency components be-tween 0.05 and 2,500 Hz in any com-bination.

2. GainThe gain is to be adjustable in fixed

steps between 200 and 2,000. Any con-tinuous or vernier gain adjustment shallbe available as a recessed control foradjustment by screwdriver or similardevice. Gain is defined from the single-ended output (or one of the two sidesof a double-ended output) to the dif-ferential input.

3. Frequency responseIn addition to the linearity and dis-

tortion requirements specified in para-graph B1, the following design charac-teristics are acceptable with constantamplitude sinusoidal input signals:From 0.14 to 950 Hz, the response shall

be flat to within ±6% (+0.5 dB). Theresponse from 2,500 Hz down to 0.05 Hz

shall not be reduced by more than 30%(-3 dB). At no frequency shall theresponse exceed the restraints specifiedfor the 0.14 to 950 Hz range.

Provisions shall be made on the con-trol panel for lowering the high-frequen-cy cutoff in several steps down to 100Hz.

4. Common mode rejectionWith the amplifier set at a gain of

1,000 and the input leads connectedtogether, a potential difference of 100millivolts peak-to-peak applied betweenthem and the input signal reference shallcause no more than 100 millivolts ofpeak-to-peak potential difference at theamplifier output at any frequency. Thisspecification must be met when a resis-tance of 5,000 ohms is inserted in serieswith either of the input leads. Thisrequirement corresponds to a commonmode rejection ratio of 1,000:1.

5. Noise levelUpon simulating the subject with com-

pletely shielded resistors of 25,000 ohmsbetween either or both input terminalsand ground, the equivalent input noiseat any gain shall be less than 10 micro-volts root mean square. Performancecharacteristics shall be maintained withbalanced source impedances up to100,000 ohms.

6. Calibration ("standardization")A calibration signal shall be provided

for insertion into the input as describedfor direct-writers (par. A13). When sev-eral preamplifiers are being used simul-taneously for multiple leads, means shallbe provided for simultaneous applica-tion of the calibration signal to all pre-amplifiers. A separate input connector forreceiving this signal shall be provided onthe instrument.

7. OutputThe output shall have a driving cap-

ability of 10 milliamperes into loads of1,000 ohms or greater and shall havean internal impedance of less than 100ohms. Means should be provided to

Circulation, Volauze XXXV, March 1967

592



ownloaded from



center the output voltage at approxi-mately zero volts.

8. Reset of dc output levelA capability for reset of the dc output

level must be provided by a switch orbutton on the instrument. Pressing thisbutton after a signal of 10 millivolts hasbeen applied to the input for 10 secondsshall return the output to within 0.1volt of its initial position within 0.5second or less, and releasing the buttonshall result in no more than 0.1 voltadditional displacement.

9. Differential input impedance and currentThe differential input impedance shall

be greater than 20 megohms, and thecommon mode input impedance shall begreater than 200 megohms. The impe-dance of each input to signal referencemust be balanced to within 1%.When the instrument is properly con-

nected to the patient, it shall not causecurrents greater than 1.0 microampere toflow in the circuits to the patient.

10. Weighting networksWhen a central terminal,3 corrective

networks for orthogonal leads, or otherlead systems requiring weighting resis-tors are used, consideration must begiven to the choice of resistor value asnoted in the section pertaining todirect-writing instruments (par. A4 andfootnote).

11. Recommendations for input range, gainand base-line stability, overload, ground-ing, short circuit protection, radio-fre-quency interference, and allowable varia-tions of power, temperature, and altitudeshall be followed as described above fordirect-writers. Any reference in theseparagraphs to output deflection of thewriting device should be interpreted asvoltage output with an amplifier gainof 2,000 corresponding to the direct-writer gain of 20 mm per millivolt.

C. Vectorcardiographs1. Frequency response

Recommendations for electrocardio-

graphic preamplifiers for use with record-ing devices other than direct-writers(pars. Bi to 11) shall be followed. Inorder to facilitate display of P and Tloops at high gain, provisions for re-cording with low-pass filters down to200 Hz shall be added. These shouldbe available through a switching devicewith clearly labeled steps.

2. Calibration and phase coherenceFor calibration, plane projections and

component leads of each plane shouldbe clearly identified. Since this can beaccomplished in various ways, only oneof many possible procedures will be de-scribed here to illustrate the principlesinvolved.The calibration signals may be lin-

early rising ramp signals of 1 millivoltamplitude, successively applied to x, y,and z amplifiers, each with appropriatepositive sense. Application of brightnessmodulation of the cathode ray oscillo-scope beam will serve to identify in-dividual channels by characteristic pat-terns as shown below:

x

y

Increased brightness area half way upramp for Lead Y

z

Increased brightness 1/3 and 2/3 way upramp for Lead Z

By simultaneous application of oneramp signal to all inputs, phase coher-ence between channels, a pair at a timefor each projection plane (xy, xz, zy),may be checked. Perfect phase coher-ence will result in a straight-line displayat an angle of 450 to the horizontal.The deviation from 450 is an indicationof the lack of phase coherence betweenthe channels. Using a-ramp rate of 0.05millivolt per millisecond, the straight-linedisplay should be 450 ± 3°.


593



ownloaded from



3. Time markingA sawtooth pattern of brightness should

be used to indicate the direction of ro-tation of vector loops by teardrop-shapeddots with the blunt edge leading. Inter-vals should be available in multiples of200 per second (for example, 200, 400,600, 800, 1,000 per second).

Provisions shall be made to inactivatethe timing device temporarily shouldthis be necessary in order to recordhigh-frequency components of vectorloops.

4. Instrumentation relative to the leads usedVectorcardiographs should be pro-

vided with either a simple switch orplug-in device to facilitate transfer fromone lead system to another. Easily ex-changeable modules will allow use notonly of currently available referenceframes but also of their successors with-out significant modification of the equip-ment. The component weighting resis-tors of corrected lead systems shouldbe readily accessible and clearly labeledto facilitate periodic recheck of theirvalues. This feature is also desirablebecause the values of the resistors mayhave to be increased from presently ac-

cepted ones in order to meet the inputimpedance recommendations describedabove (pars. A3 and A4).

5. Certain of the recommendations madefor direct-writers and preamplifiers are

also recommended for vectorcardio-graphs. These are covered in paragraphsB1 to 5 and 7 to 11.

D. Magnetic Tape Recording ofthe Electrocardiogram

In order to assure comparability oftape-recorded data, the instrumentationmust meet certain minimum require-ments. Furthermore, for faithful repro-duction of these data the input specifi-cations of data processing equipment, forexample, analog-to-digital data convert-ers, must be met.

1. Tape transport electronicsThe recording and reproducing elec-

tronics must be such that frequenciesfrom zero to over 100 Hz will be re-

produced. This requires that systems such

as frequency-modulation (FM), pulse-duration modulation or pulse-positionmodulation, digital recorder, or similarmodulation system be used. The FM sys-

tem is most commonly used in electro-cardiography and the recommendationsbelow apply mainly to this type of sys-

tem.2. Tape transports

a. In FM systems the family formatsof the Inter-Range InstrumentationGroup ("IRIG') are desirable.'0 TheIRIG standards apply to the followingitems: tape widths (3% or 1 inch),*tape thickness, track geometry, headand head stack configuration (includ-ing head and stack numbering, gap

scatter, and individual azimuth align-ment), head polarity, tape guidance,tape speeds, and tape speed toler-ances, center carrier frequencies anddeviations, deviation direction andfrequency drift, and band widths.

b. The low or intermediate-band versionof the IRIG standards shall be accept-able. These systems shall be capableof reproducing tapes recorded on low-band systems without major equip-ment modification.

c. It is desirable that reels and hubsbe in accordance with the latest issuein effect of Federal Specifications."The capability of handling reels of102 inches in diameter is desirable.

3. Voltage range of inputThe full scale range of input to the

tape transport electronics should be stan-

dardized to a span of 2 volts and shouldbe capable of extension to a span of10 volts through the use of an externaladjustable control.

*Tape width of ii inch is not specified in the IRIGstandards. However, if used, the electrical specifica-tions should conform to these standards.

Circwlation, Volume XXXV, March 1967

594



ownloaded from



4. OverloadingIn order to avoid error due to over-

loading at the input, the instrument mustbe provided with an indicator to dis-close when input overvoltage does occur.

5. Output voltage rangeThe full scale output voltage of the

"reproduce" amplifiers should extendover 10 volts which can be centeredat zero volts. They should also be ad-justable from the central panel down-ward to a span of 2 volts in severalsteps. The instrument must be capableof providing a ratio of output-to-inputvoltage of one to one.

6. Linearity and distortionZero-based linearity from "record" in-

put to "reproduce" output shall be within+1% of the full scale output voltage.Harmonic distortion shall meet the IRIGrequirements for single-carrier FMsystems.

7. NoiseThe signal-to-noise ratio shall meet the

IRIG requirement for Single Carrier FMSystems, e.g., the signal-to-noise ratioof a signal of 0.1 of the maximum modu-lation frequency recorded at full devia-tion shall be 40 dB minimum.

If flutter compensation is needed toaccomplish this, a statement to this effectshall be conspicuously placed on theinstrument.

8. Time displacement errorThe time displacement error between

any two channels due to gap scatter,tape skew, and all other causes shallbe kept below 100 microseconds at 732inches per second.

9. Start-up timeThe time required for the tape trans-

port to reach a stable speed shall beless than 1.0 second. If greater than this,provision shall be made for disabling therecording amplifiers until stable speed isreached.

10. Monitoring while recordingThe capability of monitoring during,

as well as after, recording shall be pro-Circulation, Volume XXXV, March 1967

vided. Signal-to-noise ratio, as specifiedabove, shall not be reduced when repro-ducing either during or after recording.

11. Input impedanceThe recording electronics shall have a

minimum input impedance of 40,000ohms.

12. Output impedanceThe output impedance of the repro-

ducing electronics shall be less than 100ohms and the output driving capabilityshall be at least 10 milliamperes whenconnected into a load of 1,000 ohms.

13. AlignmentThe instrument shall be provided with

a built-in capability for checking the fre-quency alignment of the recording andreproducing electronics with provisionsfor checking actual frequency deviations.

14. Recording "disable switch"The instrument shall be provided with

the capability of disabling individual re-cording channels through switches onthe control panel. This will permiterasing or recording of data on selectedchannels.

15. Magnetic tapeIt is desirable that high quality in-

strumentation tape with polyester baseand with "A" oxide characteristic beused. Tapes conforming to the latest issuein effect of Federal Specifications"1 arerecommended.

E. Analog-to-Digital Conversion of theElectrocardiogram

For automatic processing of electro-cardiographic data, it is necessary toconvert the deflections and intervals todigital form for presentation to the com-puter. Two general types of analog-to-digital conversion systems will accom-plish this. First, where there is a needto process and analyze the data in "realtime," an on-line system is required. Thistakes the form of an analog-to-digitalconverter which receives the electrocar-diographic data directly from the subjectand passes it on to the digital computer,operating in this regard as a unit of an

595



ownloaded from



integrated system. Second, when delaysof minutes, hours, or days are of noconsequence, the analog-to-digital con-verter may be used separately from thedigital computer. In this system the con-verter receives data either directly fromthe subject or from previously recordedanalog tape. Its output is a digital tapewith a format compatible with the digitalcomputer which is to analyze the data.The requirements for the analog-to-

digital conversion equipment are some-what different in each system and willbe outlined in separate sections below:

1. On-line converters

a. Input(1) The instrument must be capable

of accepting electrocardiographicdata either directly from a subjector from analog tape.

(2) The input impedance for eachinput line must be greater than10,000 ohms.

(3) The input must be capable of ac-cepting a full-scale input voltagespan of 2 volts which can be cen-tered about zero volts and ad-justable in steps to a span of 10volts.

(4) When spatial characteristics ofthe electrocardiogram are to beanalyzed, the instrument has tobe capable of accepting morethan one of the simultaneously re-corded electrocardiographic leads(most commonly three). Theserecords when converted shallmaintain temporal relationships.

b. OutputSince the instrument is usually in-

tegrated with a digital computer, theoutput will, by design, be of theproper form on tape or in anotherform of memory in the computer.However, digital-to-analog conversionshould be provided both for visualmonitoring and analog recording.

c. AccuracyAccuracy shall be no less than

+1.0% (7bits*).d. Precision

Precision, that is, resolution, shallbe no less than ±0.2% (9 bits).

e. Sampling rateA minimum sampling rate of 500

samples per second shall be providedfor each electrocardiographic lead forproper measurement of conventionalparameters.Computer analysis of high fre-

quency components of the electrocar-diogram requires considerably highersampling rates.

f. TestingThe instrument shall be provided

with a built-in facility for testing ac-curacy and precision.

Provisions for periodic verificationof other dynamic characteristics aredesirable.

g. Visual monitoringProvision shall be made for visual

monitoring of the input signal.h. Alarm and checking features

(1) Provision shall be made for asuitable indication when the inputhas caused the analog-to-digitalconverter to reach full scale.

(2) An indication shall be providedfor alerting the operator when arecord has been digitized with aparity bit error.

(3) If a record is improperly digi-tized, an indication shall be pro-vided from the analog-to-digitalconversion equipment or the com-puter.

2. Off-line converters

a. OutputThe instrument must be capable of

*"Bit" is a contraction for "binary digit." It is aunit of information equal to the result of a choicebetween two alternatives (see GLOSSARY).

Circulation, Volume XXXV. March 1967

596



ownloaded from



providing the digital output in a for-mat compatible with the digital com-puter to be used for data processing.

b. Alarm and checking featuresSame as for on-line converters (par.

Elh).c. Other inputs

Provision shall be made for inser-tion of identification information, suchas patient number, date, and so forth,into each record in such a manner thatboth this information and the digi-tized electrocardiographic data can becarried on the same record.

d. The following features shall be thesame as for on-line converters: input,accuracy, precision, sampling rate,testing, and visual monitoring.

Leads

A. Techniques, Scalar

1. The techniques described in the pre-vious report1 8 for bipolar extremityleads, unipolar extremity leads, unipolarprecordial leads, and unipolar esophagealleads are recommended with the follow-ing additions and modifications:a. Unipolar endocardial leads

The exploring or endocardial elec-trode may be a small cylinder, ap-proximately 2 mm by 3 mm, of non-corrosive metal placed at the end ofa standard solid or hollow cardiaccatheter. It is connected by insulatedwire within the catheter to a clip orjack at its proximal end to which aninput terminal of the recording de-vice may be attached. As a substitute,a polyethylene catheter containing aconducting solution may be used. Alead-wire from the input may beclipped to an occluded hypodermicneedle in contact with the solutionat the proximal end of the catheter(see INSTRUMENTS, par. A12 on haz-ards of grounding).Placement of the endocardial elec-

trode is done most accurately andsafely with the aid of the image-in-


tensified fluoroscope. However, in pa-tients too ill to be moved, the poly-ethylene catheter may be insertedpreferably via a saphenous vein, andentry of the tip into the right atriumor right ventricle may be determinedby the voltage and configuration ofthe continuously monitored endocar-dial or intravascular electrocardio-gram.The central terminal4 is to be used

as the indifferent electrode as forother unipolar leads. Response of theelectrocardiograph to the introductionof 1.0 millivolt must usually be re-duced from the customary 1.0 cm be-cause of the increased magnitude ofintracardiac deflections. The calibra-tion should be recorded on each leadmade.The leads obtained are designated

by the symbol V followed by theupper case subscript IVC, SVC, RA,RV, or PA, depending upon whetherthe recording is from the inferior orsuperior vena cava, right atrium, rightventricle, or pulmonary artery, e.g.,VIVCi, VSVC, VRA7 VRVr, VPA.

b. With high impedance amplifiers inelectrocardiographs the requirementof a resistance of 5,000 ohms in eachof the arms of the central terminal4is inadequate and should be increasedby an approximate factor of 60 to100 (see INSTRUMENTS, pars. A3, A4,and footnote).

B. Techniques, Vector

1. Polarity of leadsThe previous principles laid down rela-

tive to the use of unipolar leads forvectorcardiographic recording still ap-ply.1 With regard to bipolar leading, no

problems have been apparent in makingthe proper connection of the input termi-nals to the electrodes involved either in

the recording of the frontal or transverse

vectorcardiogram or of the transverse

(x) or longitudinal (y) axial leads.

597



ownloaded from



Custom and increasing knowledge of theexcitatory process have long dictatedthat, when the right arm is electronega-tive with respect to the left, or when thecephalic electrode is electronegative withrespect to the caudal, the recorded de-flection shall be upright in the correspon-ding bipolar scalar record. Observingthese same polarity conventions and mak-ing connections to the oscilloscopicdeflection plates by assuming that thescreen is the plane of the body beingstudied, with the observer facing it, therecorded frontal planar loop will havean orientation as it is known to havein the body.

In the case of the sagittal lead2' 3and of the planar vectorcardiograms ofwhich it is a component (transverse orxz, and sagittal or zy), conventions arenot so firmly established. The problemis complicated by the fact that the meandirection of excitation of the ventriclesin normal subjects is usually backwardin these planes. To incorporate the sagit-tal component with the proper polarityinto a sagittal or transverse vectorcardio-gram, therefore, the input terminals ofthe scalar recorder must be connectedin such a way that the lead reflects thetrue sagittal direction of the electro-motive force in the body. Such connec-tions-negative terminal anterior, posi-tive terminal posterior-will result intransverse and sagittal vectorcardiogramswith the same relative positions to ortho-gonal axes on the oscilloscopic screenas they have to such axes in the body.But if the scalar sagittal lead is con-

nected in the manner indicated to anordinary electrocardiograph, the scalarrecord will have the approximate ap-pearance, depending on the system used,of an average unipolar precordial leadinscribed upside down. The reason is thatunipolar thoracic leads have customarilybeen obtained only from the front of thethorax; the polarity in the bipolar sagittallead necessary for recording a properly

oriented vectorcardiogram results in ascalar lead simulating what would beobtained with a unipolar electrode placedon the back of the thorax.Although simulation of previously re-

corded experience would facilitate clini-cal usage and interpretation, switchingthe polarity of a lead depending on itsscalar or vector use would be undesirable.In the interest of a consistent view ofthe principal spatial force which nor-mally is downward, leftward, and back-wvard, it is recommended that the sagittalcomponent of the vectorcardiogram, whenrecorded as a scalar lead, be inscribedwith the input terminals of the recorderattached to the body in such a waythat electropositivity of the posteriorelectrode results in an upward deflectionin the finshed record. The bipolar con-nections to the sagittal lead for scalarrecording are then the same as those usedfor oscilloscopic recording of the sagittalplane viewed from the left.

2. Selection of viewsThe frontal view and the transverse

view from above seem to have been gen-erally accepted as standard althoughthere would be one advantage to viewingthe latter from below (see par. D3 be-low). The sagittal display has been pub-lished as viewed from both the rightside (right sagittal) and left side (leftsagittal) with almost equal frequency,and with the inevitable difficulty of com-paring one study with another. Cogentarguments12 have been put forth for theuse of either view. However, in keep-ing with the recommended polarity ofthe scalar sagittal lead, the sagittal planarvectorcardiogram viewed from the leftis preferred.

C. Nomenclature, Scalar

The nomenclature previously outlined" 8

is again recommended. It has been notedthat the suggestion for labeling the atrialT wave as T,I has not been generally accep-ted over the older nomenclature, Ta. The

Circulation, Volume XXXV, March 196 7

598



ownloaded from



remarks on the measurement of the mostnearly correct P-R interval will undoubtedlybecome superfluous as simultaneous record-ing of scalar leads becomes routine. Undersuch circumstances the earliest occurringP wave and the earliest occurring QRSdeflection in a variety of leads can bedetermined, and the difference in time be-tween the two will be the most nearlycorrect P-R interval. The nomenclature asotherwise outlined will be valid for ortho-gonal leads (x, y, z) if these or other leadseventually supplant the present 12 standardscalar leads.

D. Nomenclature, Vector

1. It is recommended that the componentsof the vectorcardiogram be designated as"loops" or "forms" and that the lettersP, Tp, QRS, T, and U be used to de-scribe the five forms ascribable to thesame electrical processes which accountfor the similarly designated deflectionsin the scalar electrocardiogram.The symbol or abbreviation for the

word, vectorcardiogram, is VCG.2. Recommendations previously made with

respect to designation of axes (x, y,z-transverse, longitudinal, sagittal) andplanes (xy, zy,* xz-frontal, sagittal,transverse) are continued. Neither themore commonly used term, "horizontal,"or the rarely used term, "superior," inreferring to the transverse axis or plane,though acceptable, is preferred becausethe axis or plane thus designated is onlyhorizontal or superior when the patientis standing or sitting. Similar reasoninglies behind the preference for longitu-dinal to vertical when referring to thelong axis of the body.

3. A simple standard method for indicatingthe direction of the vector in three planesis desirable. In the frontal plane, custom,since Einthoven, is quite fixed. Regarding

*The order of the letters has been reversed com-

pared to the earlier report in order to indicate theproper reference axis as z in the sagittal plane.12Circulation, Volume XXXV, March 1967

the frontal plane as a circle, it maybe divided into upper and lower, rightand left halves. Beginning at the leftextremity of the ground line (equator,x-axis), angles are indicated as positivefrom 00 to +180° clockwise on the perim-eter of the lower half, and negativefrom 00 to -180° proceeding counter-clockwise on the perimeter of the upperhalf. It is recommended that the anglebetween a cardiac vector and the x-axisin the frontal plane be designated F.0It is identical with Einthoven's anglealpha.In considering a method of indicating

the direction of a sagittal or transverseplanar vector, it is well to recognizethat the electrocardiographic orthogonalreference frame recommended above isone which is rotated, as contrasted tomathematical custom, through 180°around the x-axis. The positive end ofthe latter remains on the left side of thesubject, but the positive ends of they-axis and of the z-axis are inferior andposterior, respectively.

If the left sagittal view is displayedon an oscilloscope, the polarities (sign)of the axes will be as they are in thefrontal view, and the same conventionscan be applied. The plane may bedivided into positive lower and negativeupper halves with the posterior extremityof the z-axis regarded as being at 0.'Beginning at this extremity (observer'sright), angles inscribed clockwise arepositive from 0° to +180° and thoseinscribed counterclockwise are negativefrom 00 to -180.° It is recommendedthat the angle between the vector andthe z-axis be designated S.'With regard to the transverse view of

a vector or vectorcardiogram, perfectconsistency with the principles of mea-suring F' anad S' just elaborated wouldbe obtained if the oscilloscopic displaywere made as though the plane were

viewed from below. The habit of view-ing this plane from above is so in-grained

599



ownloaded from



that a change is not likely to be accepted.Therefore, in the display of the trans-verse plane viewed from above, theupper and lower halves are reversed insign as contrasted to the frontal andleft sagittal views. Angles written in acounterclockwise direction beginning atthe left end of the x-axis have a positivevalue from 00 to +180,0 and those writtenin a clockwise manner have a value from0° to -180.° It is recommended thatthis angle, as in a spherical coordinatesystem (see below), be designated "Ho ."*

If the spherical coordinate system isused for designating the location andmagnitude of spatial vectors, it is recom-

mended that the symbols for elevation(latitude), azimuth (longitude), andmagnitude be V0, Ho, and Mt, respec-tively. Azimuth is the angle between thehorizontal (transverse) projection of a

spatial vector and the left half of thex-axis and it is identical with H° as

defined for the transverse plane above.Anterior angles (00 to -1800) are re-

garded as negative, and posterior angles(00 to +1800) as positive. Elevation isthe angle between the spatial vector andthe transverse plane. Inferior angles are

positive (0° to +900).4. Communications would be facilitated if

some common symbols were used in re-

ferring to specific portions of the vector-cardiogram. Since Einthoven's time, E hasbeen used to indicate an instantaneousforce. The letter A has been used to indi-cate an integrated value (area) of theforce to include its duration as well as

magnitude. These letters have been em-

bellished in a variety of ways by differentauthors but always to indicate, in reality,a value of all or a part of the electromo-

*H1, though not altogether accurate,1 is usedinstead of TO because of the possible confusionwith T as a designation of the final ventriculardeflection.tM may be regarded as a general symbol which

applies to both instantaneous and integrated vectors(see par. B4 below).

tive force of the heart as reflected in apresumed axis (x, y, z) or plane (xy, xz,zy), or in three-dimensional space (xyz).A uniform, easily typed symbolic

nomenclature with as close a relation topast custom as possible is visualized asinvolving the use, without spacing, of adecimal and digits or the capital letterA when indicated, but not E (which isrendered redundant by the digits); ofthe usual capital letters for specific ex-citatory or recovery events (P, Q, R, S,T, U, C for QRST, J for S-T junction,S-T for S-T segment); and of lover caseletters as indicated for the axes andplanes as defined above. For example,.03Px means the instantaneous voltage0.03 sec after the beginning of atrialexcitation manifest in the transverse axiallead; AQRSz signifies the time-voltagearea of QRS in the sagittal axial lead.With oscilloscopic visualization or re-

cording of the vectorcardiogram inplanes or in space, only the P, QRS, andT loops and the point J are easily identi-fied. The components of the QRS loopcorresponding to the scalar deflectionsQ, R, and S are so variable in thedifferent axes and planes that use of theindividual letters for referring to por-tions of the vector configuration is notsufficiently specific. Greater accuracy isachieved by designating vectors tempo-rally, zero time being at the beginningof the P, QRS, or T loop. In the pres-ent nomenclature this may be indicat-ed by a decimal followed by digits placedbefore the designated instantaneous pla-nar or spatial P, QRS, or T vector. Forexample, .05QRSxy is the vector occurring0.05 sec after the beginning of the QRSloop in the frontal plane. In the systemvector symbols are omitted.

GlossaryIn order to present meaningful specifications in

brief form, the language of the enginieer has beenemployed in much of this report. There arecharacteristics of engineering language besides a


600



ownloaded from



distinctive vocabulary. The uninitiated physicianor life scientist will notice a tendency to employnouns singly or in groups as adjectives; verbssuch as "read" or "record" are converted intomodifiers for designating apparatus which doesthose jobs. The meaning of such wording, thoughsometimes ambiguous, is usually clear; communi-cative difficulty between the various groups whoseinterests converge in electrocardiographic instru-mentation lies largely in new words or old wordsused in new ways. Because the subject matteris of interest to physicians and other nonengineers,a short glossary of terms is herewith attached tofacilitate their study of these recommendations.

Accuracy: A degree of conformity to some rec-ognized standard value, or more specifically,the deviation of a result obtained by a par-ticular method from the true value. It isusually expressed as a percentage of full scale.Thus, if the input signal varies between zeroand a maximum of 10 millivolts, then thevalue assigned to the output should be notfurther from the actual by more than 0.1millivolt for the accuracy to be within ± 1%.

Bit: Contraction for "binary digit"; a unit ofinformation equal to the result of a choicebetween two alternatives. In terms of thebinary system an eight bit word is capableof holding 28 possibilities or values varyingfrom zero to 255 in whole numbers.

Carrier wave: A continuous electromagnetic orradio wave, typically sinusoidal in form andof relatively high frequency, which is alteredeither in amplitude (AM) or frequency (FM)by a modulating signal of much lower fre-quency. Thus, it "carries" the informationcontained in the signals superimposed upon it.When transmitted or recorded, the modulatingsignal is added to the carrier, and when re-ceived or reproduced the carrier is removedleaving the desired signal.

Common mode rejection: The ability of a differ-ential amplifier to reject common signals. Thecommon mode rejection ratio is defined as:

CM rej gain of amplifier to difference signalgain of amplifier to common signal

Differential amplifier: A type of amplifier ordin-arily employed in electrocardiography designedto amplify the difference between two voltagesmeasured with respect to a common point,usually the right leg. Ideally, no signal whichis common to the two active input terminalsshould appear in the output. In a practicalamplifier, however, such a common signal willproduce an output (see "Common mode re-jection" above).

Distortion: A change in shape of an electrocardio-Circulation, Voluume XXXV, March 1967

graphic waveform introduced by the recordinginstrument. It is useful to distinguish lineardistortion from nonlinear distortion. Linear dis-tortion results when the gain of the instrumentfor a sinusoidal input is not constant, butdepends on the frequency of the sinusoidalsignal. The frequency response of an instru-ment is a record of its gain versus frequencyof sinusoidal input. Nonlinear distortion is char-acterized by an output which departs from asinusoidal wave shape when the input issinusoidal.

Frequency modulation: The modification of thefrequency of a carrier wave of relatively highfrequency in accordance with a signal. Forexample, if the carrier wave is a sine waveof 4,000 Hz and the signal is an electrocardio-graphic waveform, the carrier frequency willbe progressively raised above 4,000 Hz pro-portionate to the height of the R wave anddropped proportionally below 4,000 Hz duringthe S wave.

Gain: The ratio of the amplitude of the outputto the input, usually expressed in milli-meter/millivolt for a direct-writer, and volts/volt for an amplifier or preamplifier.

Impedance: The total opposition to flow of analternating current in a circuit. Analogous tothe actual electrical resistance to a direct cur-rent, the impedance is the ratio of the effectivevoltage to the effective current. In additionto actual resistance, impedance includes react-ance which results from capacitative or induc-tive properties of the circuit.

Input: (1) The energy (signal) put into anelectronic device. (2) The terminal or ter-minals for the input signal of such a device.(3) The part of the electronic circuit im-mediately adjacent to the input terminals.

Linearity: Linearity provides a measure of thenonlinear distortion in an instrument (see"Distortion"). Since nonlinear distortions mayenter in a number of ways and result inpeculiar effects, the linearity specification maybe written in several ways. Commonly it dealswith the departure from a sinusoidal waveshape for a sinusoidal input. The characteristicsof the writing device pose a particular problemin establishing specifications for direct-writers.Linearity as used in this report refers to zero-based linearity. That is, a graphic plot ofoutput voltage (or deflection in the case ofdirect-writers) versus input voltage will yielda series of points. If a straight line is drawnwhich passes through the origin, the deviationof any of the actual points from this straightline is a measure of the nonlinearity of thesystem.

601



ownloaded from



Modulation: The modification of a basic, con-tinuous cairier signal by another information-containing signal. Most commonly the termrefers to alterations in amplitude (AM) orfrequency (FM) of a sinusoidal carrier waveby the modulating signal. However, the carriermay be a train of pulses which may be modifiedin width or in frequency by the added in-formational signal.

Noise: Noise refers to unwanted "signal" intro-troduced from other sources, including theelectrocardiograph itself. An example is 60cycle alternating current "interference."

Output: (1) The power, energy, signal, or in-formation delivered from an electronic deviceafter conversion of the input into another formor modification (such as amplification). (2)The terminals or terminal of an amplifier fordelivery of the output. (3) That part of theelectronic circuit immediately adjacent to theoutput terminals.

Parity bit channel: In digital recording systems(paper or magnetic tape), an added bit whichacts as an internal checking mechanism fordetecting errors in recording.

Precision: The degree of agreement betweenrepeated measurements of a quantity and thusthe repeatabilitv of the performance of aninstrument. It is desirable for an instrument tobe more precise than it is accurate so thatthe deviation between successive measurementswill never exceed the deviation from the valuebeing measured.

Record electronics: In a magnetic tape system,the amplifiers and magnetic heads which con-vert the input signal into magnetic patternson the tape.

Reproduce electronics: In a magnetic tape system,a pick-up head and attached amplifiers whichconvert the varying magnetic pattern frommoving tape back into an electrical waveform.

Signal: Signal in this discussion refers to theelectrocardiographic waveform attributable tothe electrical activity of myocardium.

Signal-to-noise ratio: The ratio of an appropriateparameter characterizing the amplitude of thewanted signal to a similar parameter of theunwanted noise.

Skew: (1) For direct-writers, the departure ofthe writing stylus from an absolutely verticalswing up or down, that is, from a path perpen-dicular to the direction of movement of thepaper. (2) For magnetic tape, the departureof the moving tape from an absolutely straightcourse past the magnetic heads. The term iscomparable in some degree to "parallax" as

used in photographic recording of the elec-trocardiogram.

References1. WILSON, F. N., ET AL.: Recommendations for

standardization of electrocardiographic andvectorcardiographic leads. Circulation 10: 564,1954.

2. JoINr RECOMMENDATIONS OF THE AMERICANHEART ASSOCIATION AND THE CARDIAC SOCIETYOF GREAT BRITAIN AND IRELAND: Standardiza-tion of precordial leads. Amer Heart J 15:107, 1938.

3. COMMITTEE OF THE AMERICAN HEART ASSOCIA-TION FOR THE STANDARDIZATION OF PRECORDIALLEADS: Standardization of precordial leads(a): Supplementary report. Amer Heart J 15:235, 1938; (b) Second supplementary report.JAMA 121: 1349, 1943.

4. WILSON, F. N., JOHNSTON, F. D., MACLEOD,A. G., AND BARKER, P. S.: Electrocardiogramsthat represent the potential variations of asingle electrode. Amer Heart J 9: 1, 1934.

5. BURGER, H. C., AND VAN MILAAN, J. B.: Heartvector and leads: I, II, and III. Brit Heart J8: 157, 1946; 9: 154, 1947; 10: 229, 1948.

6. FRANK, E.: The image surface of a homogeneoustorso. Amer Heart J 47: 757, 1954.

7. MCFEE, R., AND JOHNSTON, F. D.: Electrocardio-graphic leads: I. Introduction. II. Analysis.III. Synthesis. Circulation 8: 554, 1953; 9:255, 1954; 9: 868, 1954.

8. KOSSMANN, C. E., ABILDSKOV, J. A., BERGER,A. R., BRUMLEK, J., DOLGIN, M., AND MILNOR,W. R.: Part V. The electrocardiographic diag-nosis. In Diseases of the Heart and BloodVessels; Nomenclature and Criteria for Diag-nosis, ed 6, by the Criteria Committee, NewYork Heart Association. Boston, Little, Brown,& Co., 1965.

9. DOWER, G. E., ZIEGLER, W. G., BERRY, F. G.,ANTD MooRE, A. D.: A simple test of speedof response of electrocardiographs. Amer HeartJ 67: 524, 1964.

10. INTER-RANGE INSTRUMENTATION GROUP (TELEM-ETRY WORKING GROUP): Telemetry standards(revised 1966). IRIG Document No. 106-66,Defense Documentation Center, Cameron Sta-tion, Alexandria, Virginia.

11. IN"TERIM FEDERAL SPECIFICATIONS (NAVY-SHIPS):W-R-00175/4: Reel, Aluminum and Magnesi-um; WV-T-0070/4: Tape, InstrumentationTelemetering, Polyester Base. Superintendentof Documents, U. S. Government PrintingOffice, Washington, D. C.

12. HELM, R. A.: Vectorcardiographic notation. Cir-culation 13: 581, 1956.

Cifcu4ation, Volume XXXV, March 1967

602



ownloaded from


and OTTO H. SCHMITTS. BERSON, STANLEY A. BRILLER, DAVID B. GESELOWITZ, LEO G. HORAN HUBERT V. PIPBERGER, HUBERT V. PIPBERGER, GERHARD BAULE, ALAN

HECHT, FRANKLIN D. JOHNSTON, CALVIN KAY, EUGENE LEPESCHKIN, CHARLES E. KOSSMANN, DANIEL A. BRODY, GEORGE E. BURCH, HANS H.

Instruments in Electrocardiography and VectorcardiographyRecommendations for Standardization of Leads and of Specifications for

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1967 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.35.3.583

1967;35:583-602Circulation.

http://circ.ahajournals.org/content/35/3/583.citationlocated 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. and Rights Question and Answer

Permissionsthe 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 column ofClearance 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:



ownloaded from

http://circ.ahajournals.org/content/35/3/583.citation

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

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