Premature Ventricular Complexesin the Absence of Identifiable Heart Disease

JOHN B. KoSTIS, M.D., KATHRYN MCCRONE, R.N., A. E. MOREYRA, M.D.,

S. GOTZOYANNIS, M.D., NORA M. AGLITZ, R.N., N. NATARAJAN, M.D., AND P. T. Kuo, M.D.

SUMMARY To define the prevalence, frequency and characteristics of premature ventricular complexes(PVCs) in adults free of recognizable heart disease, we performed 24-hour ambulatory electrocardiography on

101 subjects (51 men and 50 women, mean age 48.8 years) in whom physical examination, chest x-ray, ECG,echocardiogram, maximal exercise stress test, right- and left-heart catheterization and coronary arteriog-raphy were normal. Thirty-nine subjects had at least 1 PVC/24 hours, but only four had more than 100PVCs/24 hours and fewer than five had more than five PVCs in any given hour. The probability of having atleast 1 PVC/24 hours increased with age (chi square = 11.789, p = 0.019). The number of PVCs/24 hourswas also positively associated with age (r - 0.33, p = 0.001). There was no consistent relationship between thepresence or number of PVCs/24 hours and sex, blood pressure, weight, height, body mass index, serum potas-sium or calcium, cholesterol and triglyceride, hemoglobin, the ingestion of coffee, tea or alcohol, and cigarettesmoking. Four subjects had multiform PVCs, two of whom had early PVCs.

AMBULATORY ELECTROCARDIOGRAPHY iscommonly performed to detect frequent or complexventricular ectopy, especially in patients with coro-nary artery disease.'`6 However, scant data are avail-able on the incidence, frequency and characteristics ofpremature ventricular complexes (PVCs) in normalsubjects. Kennedy and associates7 8 described a groupof apparently healthy subjects with frequent and com-plex ventricular irritability selected from a largereferral population, including subjects with coronaryartery disease and elevated left ventricular end-diastolic pressure. Hinkle and associates found PVCsin a high percentage of middle-aged men.9 10 In thesestudies, many subjects with frequent PVCs also hadclinical evidence of coronary heart disease, hyper-tension or chronic obstructive lung disease, and fre-quent PVCs were associated with a fatal event in thenext few years.

Raftery and Cashmanll found low frequency ofPVCs in 53 ambulatory, apparently normal subjects.Brodsky and co-workers'2 found PVCs in 25 of 50 ap-parently normal male medical students, ages 23-27years, in whom cardiac disease was excluded by non-invasive testing. Clarke and co-workers'3 found sig-nificant ventricular arrhythmias, including frequentventricular ectopic beats, R-on-T phenomenon, multi-focal ventricular ectopic beats, bigeminy and ventric-ular tachycardia in 10 of 86 subjects (12%) who wereapparently healthy by history, physical examination,electrocardiography, and biochemical and hema-tological screening. An exercise stress test or coro-nary arteriography was not performed to exclude

From the Department of Medicine, CMDNJ-Rutgers MedicalSchool, Piscataway, New Jersey.

Supported in part by a grant from the Somerset Country HeartAssociation.

Address for correspondence: John B. Kostis, M.D., CMDNJ-Rutgers Medical School, P.O. Box 101, Piscataway, New Jersey08854.

Received March 25, 1980; revision accepted October 6, 1980.Circulation 63, No. 6, 1981.

coronary artery disease. In a similar study, Glasserand associates" found PVCs in seven of 13 apparentlyhealthy subjects, ages 62-84 years. Gilsonl' found in-frequent ventricular ectopic beats in his early studiesof apparently normal males.

Because undetected coronary artery disease is prev-alent in middle age, and because coronary arteriog-raphy was not performed in these studies of ap-parently normal subjects, coronary atherosclerosiscannot be excluded as a cause of the observed arrhyth-mias, and these studies cannot be used to define thenormal. Using ambulatory monitoring for 10 wakinghours, DeMaria and associates'6 found PVCs in 10 of40 subjects with normal cardiac catheterization thatwere used as the control group in a study of mitralvalve prolapse. Details of this group were notreported. Therefore, to further define the prevalence,frequency and characteristics of PVCs in normal sub-jects we performed 24-hour ambulatory electrocar-diography on 101 subjects free of evident disease asverified by extensive noninvasive and invasive testing,including coronary arteriography.

Materials and Methods

PopulationWe studied 101 subjects in whom physical ex-

amination of the cardiovascular system, chest x-ray,ECG, maximal exercise stress test, echocardiogram,right- and left-heart catheterization and coronaryarteriography were normal. Ninety-nine of these sub-jects were selected from 1500 patients referred for car-diac catheterization to the laboratories associatedwith CMDNJ-Rutgers Medical School for differen-tial diagnosis of chest pain. Two subjects were studiedbecause of hypercholesterolemia and strong familyhistory of cardiac disease. Cardiac catheterization wasperformed by the Sones technique.'7 Subjects withlesions of the coronary arteries compromising thelumen by only 5-10%, or with minor luminal irreg-ularities were excluded. Single-plane ventriculog-raphy was performed in the right anterior oblique

1351

by guest on June 20, 2018http://circ.ahajournals.org/

Dow

nloaded from

VOL 63, No 6, JUNE 1981

projection. Patients with abnormal left ventricularfunction as evidenced by an ejection fraction below50%, end-diastolic pressure above 12 mm Hg, or local-ized contraction abnormalities were excluded.Patients with mitral valve prolapse diagnosed by ven-triculography or echocardiography, valvular disease,abnormal right ventricular pressures, bundle branchblock, ventricular hypertrophy, abnormal ST or Twave, or other abnormalities of the resting ECG wereexcluded. M-mode echocardiography was performedin all subjects. Subjects with abnormal chamber size,increased septal thickness and mitral valve prolapsewere excluded. The subjects performed a maximaltreadmill stress test using a Balke protocol.'8 Fatigueor dyspnea was the end point in all subjects. Subjectswho developed horizontal or downsloping ST depres-sion more than 0.75 mm (0.075 mV) during or afterexercise were excluded. Tests of blood sugar, bloodurea nitrogen, serum creatinine, uric acid, total bili-rubin, calcium, alkaline phosphatase, SGOT, chol-esterol, triglycerides, total protein, albumin, LDH,sodium, potassium, chloride and CO2, as well as acomplete blood count and urinalysis, were per-formed. Significant extracardiac disease was ex-cluded by these tests, history and physical examina-tion. The height, weight and sex of the subjects and theuse of coffee, tea, alcohol and tobacco were noted. Thebody mass index was calculated by dividing the weightby the square of the height. The mean, range andstandard deviation of some of these variables areshown in table 1.There were 51 men and 50 women, mean age

48.8 * 10 years (SD) (range 16-68 years). The chestpain that prompted the coronary arteriography was in

TABLE 1. Range, Mean and Standard Deviation of SelectedVariables in Subjects with Normal Hearts

Variables Mean M1ax Min SD

Age (years) 48.8 68 16 10.0

Weight (lbs) 164.7 250 94 32.4

Height (inches) 66.9 74 59 3.9

Triglycerides (mg/dl) 148.9 720 32 108.9

Cholesterol (mg/dl) 218.1 480 120 55.8K (mEq/l) 4.3 5.8 3.0 0.5

Na+ (mEq/l) 141.2 1a8 130 5.9

Hemoglobin (g/dl) 14.2 16.9 11.2 1.4

Uric acid (mg/dl) 5.6 10.0 3.0 1.7

Calcium (mg/dl) 9.5 11.2 7.8 0.6

Maximal BP (mm Hg) 129.4 200 100 20.1

MIiniimal BP (mm Hg) 82.8 110 50 11.8

Cigarettes/day 8.5 80 0 15.5

Coffee (cups per day) 1.9 14 0 3.0

Body mass index(100 X lbs inl2) 0.4 5.3 3.4 0.5

Alcohol (M oz /week) 5.8 84 0 11.6

Glucose 100.2 173 7 5) 17.2

retrospect attributed to extracardiac causes in 51 sub-jects. Chest wall disorders were diagnosed in 13, es-ophageal disease in 29 and cervical osteoarthritis in19; psychological disturbances were thought to be thecause of pain in seven subjects.

Ambulatory Electrocardiography

Ambulatory electrocardiography was performedfor 24 hours using portable Avionics tape recorders(Model 445) to obtain two leads corresponding to amodified V1 and V,. Ambulatory electrocardiographyfor two additional 24-hour periods was performed ona random subset of 30 subjects. Satisfactory record-ings covering at least 18 of the 24 hours were avail-able on 99 subjects. The tapes were played back on a660A Avionics playback system by an experiencednurse. All disturbances of the rhythm were written outon electrocardiographic paper at 25 mm/sec andverified by a cardiologist. To ascertain that a signifi-cant arrhythmia was not missed during playback, 1hour of each recording and 10 24-hour completerecordings were played back at a paper speed of 5mm/sec. Five PVCs (2.2%) were missed by the nurse.The PVCs were classified as having right ventricularmorphology if the terminal QRS deflection was posi-tive in modified lead V, and negative in modified leadV1. PVCs with left ventricular morphology had a posi-tive terminal QRS deflection in modified lead V1 and anegative terminal QRS deflection in modified lead V,.PVCs not meeting either criterion were considered un-classifiable.1ll 20 The prematurity index was calcu-lated by dividing the coupling interval (RR') by theQT interval of the preceding normal beat. The aberra-tion index was derived by dividing the QRS durationof the PVC by the QRS duration of the normal beats.

For each hour, the maximal and minimal heartrates, total beats/hour and the number of PVCs ofeach morphology were tabulated. The average heartrate was obtained by dividing the number ofbeats/hour by 60. The data were entered into an IBM370/168 computer and analyzed statistically.Analyses included correlation coefficients using para-metric and nonparametric techniques, cross-tabula-tions, chi-square, independence tests, and t tests. Eachanalysis was limited to cases without missing valuesfor pertinent variables.

ResultsThirty-nine of the 101 subjects had at least one PVC

when monitored for 24 hours. However, the number ofPVCs/24 hours was rather small. Table 2 shows thepercentage of subjects that had a given number ofPVCs when monitored for different time intervals.Four subjects had more than 100 PVCs/24 hours andonly one had more than 500 in this time interval.Three subjects had more than five PVCs in any givenhour. The percentage of subjects who did not have anyPVCs during monitoring decreased when recordingslonger than 1 day were used. When 2 days of record-ing were analyzed, 10 of 30 subjects (33%) did not

CIRCULATION1352

by guest on June 20, 2018http://circ.ahajournals.org/

Dow

nloaded from

PVCs IN NORMAL SUBJECTS/Kostis et al.

TABLE 2. Probability (%c) of Observing a Given Number of Premature Ventricular Complexes During a GivenLength of Observation in Subjects with Normal Hearts

No. of PVCs0 >1 >5 >10 >50 >100 >500

1 hour (9-10 a.m.) 85 15 3 2 1 0 0

3 hours (9 a.m.-12 noon) 78 22 9 5 2 0 0

6 hours (9 a.m.-3 p.m.) 74 26 14 8 3 1 0

8 hours (9 a.m.-5 p.m.) 72 28 15 10 3 2 0

12 hours (9 a.m.-9 p.m.) 64 36 22 15 5 3 0

24 hours 61 39 25 20 9 4 1

Abbreviation: PVC = premature ventricular complex.

have any PVCs and only five of 30 (17%) did not haveany PVCs in 3 days. The percentage of subjects whohad a given number of PVCs/24 hours after 1, 2 and 3days of recording is shown in figure 1. The distribu-tion is unimodal and tends to assume a normal config-uration with increasing number of days of tape record-ing. The occurrence of PVCs varied considerably insuccessive 24-hour recordings. Of the 17 subjects whodid not have PVCs in the first 24-hour recording,10 did not have PVCs in two 24-hour recordings andfive did not have PVCs in any of the three 24-hour re-

cordings. Of the 13 subjects who had at least one PVCin the first recording, four did not have any PVCs in a

repeat recording.With increasing age, the number of PVCs recorded

in 24 hours increased significantly. The (nonpara-metric) Spearman's rank correlation coefficient was

0.33 (p = 0.001) and the Pearson product-momentcorrelation coefficient was 0.19 (p = 0.067). The prob-ability of having more than 1, more than 50, and more

61 Averoge Number of PVC per 24 hours

4740

o I 10 m0 Cw c, to ,0 Wo , 024 hours 48 hours 72 hours

Duration of Recording

FIGURE 1. Distribution of subjects according to theaverage number of premature ventricular complexes(PVCs)/24 hours after monitoring for 1 (99 subjects), 2 (30subjects), and 3 (30 subjects) days. With increasing durationof monitoring, the distribution approaches the normal andremains unimodal. The increase in the percentage of sub-jects with higher frequencies (> 100 PVCs/24 hours) withprolonged monitoring may be due to the disproportionatelyhigh number (three of 30, 10%) of subjects who had highPVCfrequency ofPVCs and were randomly allocated to thesubset of 30 subjects who had prolonged (48- or 72-hour)recordings.

than 100 PVCs/24 hours in different age groups isshown in table 3. The probability of having at least 1PVC/24 hours also increased with age (chi-square = 11.789, p = 0.019).No significant association of the presence or ab-

sence of PVCs or the number of PVCs/24 hours andother variables (sex, systolic blood pressure, glucose,weight, height, body mass index, serum cholesterol,triglycerides, uric acid, potassium and calcium, bloodhemoglobin, the ingestion of coffee, tea or alcohol andcigarette smoking) was noted. No consistent relation-ship between the heart rate and the occurrence ofPVCs was found, although a positive correlation (r =

0.61) between the number of PVCs/hour and themaximal heart rate in the same hour was seen in onepatient. A diurnal pattern of the occurrence of PVCscould not be identified.Complex forms of ventricular ectopy were rare.

Multiform PVCs (bifocal) occurred in four subjects.Two of these subjects in addition had early PVCs (pre-maturity index less than 1.0). PVCs with prematurityindex less than 0.9 were not observed. Three subjectshad brief periods of bigeminy. No subject had two ormore PVCs in a row.The distribution of the prematurity index is shown

in figure 2. In this analysis all subjects who had PVCseither in the first 24-hour recording (39 subjects) or inrepeat recordings performed in a random subset of 30(12 additional subjects) are included. The mode is inthe interval between 1.21 and 1.40, the mean is 1.41and the standard deviation is 0.37. The range of pre-

TABLE 3. Effect of Age on the Probability (%) of HavingMIore Than a Given Number of Premature Ventricular Com-plexes per 24 Hours in Subjects with Normal Hearts

Age No. of PVCs(years) n >0 >50 > 100

10-29 6 16.7 0 0

30-39 11 18.2 0 0

40-49 29 27.6 3.3 0

50-59 39 51.3 12.8 5.1

60-69 12 58.3 25.0 16.7

Abbreviation: PVC = premature ventricular complex.

1 353

by guest on June 20, 2018http://circ.ahajournals.org/

Dow

nloaded from

VOL 63, No 6, JUNE 1981

maturity index is 0.9-2.63. The aberration index(QRS duration of PVC/QRS duration of normalbeat) had a mode between 1.81 and 2.10, a mean of1.84, a standard deviation of 0.37 and a range of1.2-3.0. The QRS duration of the PVC had a modebetween 0.1 1 and 0.13 second, a range of 0.09-0.2 sec-ond, a mean of 0.129 second and a standard deviationof 0.028 second (fig. 3). Four subjects had PVCs withduration longer than 0.16 second. Among the 51 sub-jects with PVCs, 25 had PVCs of right ventricularmorphology, 15 had PVCs of left ventricular mor-phology and four had PVCs both of right and left ven-tricular morphology. In seven subjects the origin couldnot be ascertained from the two monitored electro-cardiographic leads.

DiscussionThis study shows that the majority of subjects with

normal hearts have a small number of PVCs. Only 5%of the subjects had more than five PVCs in any givenhour and 4% had more than 100 PVCs/24 hours.These findings are similar to those of Brodsky andassociates'2 in a younger age group. The number ofPVCs in subjects with normal hearts in this study islower than that reported by Clarke et al.,13 who stud-ied apparently normal subjects, and by Hinkle et al.9Persons with undetected or known heart disease, es-pecially coronary artery disease, may have been in-cluded in these studies and may account for the higherincidence of PVCs.

Because the subjects in this study were catheterizedbefore they entered the study, the condition that led tocardiac catheterization might have caused some of thearrhythmias reported here. This is unlikely for sev-eral reasons. A meticulous search for cardiac disease

16

14

12

zLAJ

a

LA.

10'

8

6

4

UVER-RLIJ- I.I- 11 1.01- 1.1- Z. - OV1.2) 1.40 1.60 1.80 2.00 2.40 2.40

PREMATURITY INDEX

FIGURE 2. Frequency distribution ofprematurity index ofpremature ventricular complexes recorded in 51 subjectswith normal hearts.

22 F

20t

18 -

16 F

14 -

z

WLAJ

12F

1o

2

PVC DURATION (seconds)FIGURE 3. Frequency distribution of the duration ofpre-mature ventricular complexes (PVCs) in 51 subjects withnormal hearts.

was performed by noninvasive and invasive methods.Coronary artery disease was excluded by coronaryarteriography, and patients with even minor abnor-malities of the coronary arteriogram were excludedfrom the study. Asymmetric septal hypertrophy andmitral valve prolapse were excluded by cardiaccatheterization, ventriculography and echocardiog-raphy. Valvular disease and congestive cardiomyop-athy were excluded by ventriculography and cardiaccatheterization, and Prinzmetal's angina with normalcoronary arteriography was excluded by ambulatoryelectrocardiography. Patients with the syndrome ofchest pain and an abnormal stress test were excludedby the normal findings on maximal exercise stress test.Therefore, it is unlikely that these subjects have anyknown cardiac disease.

Serious extracardiac disease was excluded as thecause of PVCs because the history, physical examina-tion, chest x-ray and biochemical screen failed to dis-close significant systemic disease.The cause of pain that led to cardiac catheteri-

zation was not related to the presence or number ofPVCs. For example, the subjects in whom the painwas attributed to esophageal disease did not have a

CIRCULATION1354

2

by guest on June 20, 2018http://circ.ahajournals.org/

Dow

nloaded from

PVCs IN NORMAL SUBJECTS/Kostis et al.

significantly different number of PVCs from subjectsin whom the pain was not attributed to esophageal dis-ease (t = 0.79, p = 0.43). Similar results were foundwhen other causes of chest pain, i.e., cervical osteo-arthritis (t = 0.52, p = 0.17), chest wall disorder (t =0.039, p = 0.97), and psychological abnormality (t =0.527, p = 0.61) were studied. These considerationsand the smaller incidence and frequency of PVCs thanreported in apparently normal subjects indicate thatthe data reported here pertain to normal subjects.Patients who presented with PVCs or palpitationswere excluded from the study. As some of these pa-tients have normal hearts, the incidence of PVCs innormal subjects may be higher than that reportedhere.

Several investigators have found that the incidenceand frequency of PVCs increases with age in ap-parently normal subjects.'0 11,12, 23 One explanation isthat undetected coronary artery disease becomes moreprevalent with increasing age and results in an in-crease of the incidence and frequency of PVCs. In thisstudy the incidence and frequency of PVCs were in-creased with increasing age, although coronary arterydisease was excluded by coronary arteriography.Therefore, a different explanation must be sought. Wespeculate that degenerative changes of the myo-cardium that increase with age, or small scars result-ing from overt or subclinical transient myocarditis,may cause PVCs.

Increased systolic blood pressure has been asso-ciated with PVCs in a large population.22 In this study,the PVCs in subjects with high systolic blood pressurecould have been caused by left ventricular hypertrophysecondary to the elevated blood pressure. Subjectswith left ventricular hypertrophy were excluded fromour study by echocardiography and electrocardiog-raphy, and PVCs were not associated with high bloodpressure. Using coronary arteriography, we also ex-cluded subjects with coronary atherosclerosis, a moreprevalent cause of PVCs in subjects with high systolicblood pressure.

Frequent and complex ventricular irritability hasbeen reported in subjects with normal hearts as diag-nosed by extensive noninvasive and invasive testing.This finding is not in direct contradiction with ourfindings because normal subjects with frequent PVCsmay represent the extreme of the distribution of nor-mal subjects (i.e., frequent PVCs may occur extremelyrarely in normal subjects). Alternately, these subjectsmay suffer from a form of cardiac disease that has notyet been identified. One can easily imagine that sub-jects with mitral valve prolapse and PVCs would havebeen classified as normal before the recognition of thissyndrome.

During prolonged ambulatory monitoring (3 days),the distribution of subjects according to the number ofPVCs is unimodal and approaches the normal curve,suggesting that PVC of low frequency are not nec-essarily a manifestation of disease, and that with pro-longed monitoring, PVCs can be identified in the ma-jority of normal subjects.

Bimodal distributions suggesting that there weretwo types of subjects, one with low and one with highfrequency of PVCs (probably one with healthy andone with diseased hearts), were observed in previousstudies of subjects in whom cardiac disease was not ex-cluded. This conforms to the common observationthat practically all persons of middle or advanced agehave experienced occasional PVCs.PVCs detected in this population were more often

of right than left ventricular morphology. Twenty-fiveof 51 subjects with PVCs (39 with PVCs in the first 24hour recording and 12 with PVCs only in a repeatrecording) had PVCs of right ventricular morphologyand only 15 had PVCs of left ventricular morphology.This tendency of normal subjects to have right ventric-ular PVCs has been reported,'9' 23, 24 and an adequateexplanation for it is not known. It is reasonable to ex-pect that patients with disease of the left ventriclewould have PVCs originating in that ventricle.However, the reason for the increased frequency ofright ventricular PVCs in normal subjects is not ap-parent. This study indicates that the predilection ofnormal subjects to have PVCs of right ventricularmorphology is not absolute and that many normalsubjects have PVCs of left ventricular morphology. Inaddition, the electrocardiographic configuration ofPVCs depends not only on the focus of origin of thearrhythmia, but also on the site of exit, which may bein either ventricle.26 Therefore, the electrocardio-graphic morphology of the PVC is not a reliable indi-cator of the presence or absence of cardiac disease.26Complex forms of ventricular ectopy that carryserious prognostic implications in patients withischemic heart disease (e.g. PVCs interrupting the Twave, multiform PVCs, and salvos) were rare.27 Onlyfour subjects had multiform PVCs, two of whom hadearly PVCs. Increased QRS duration of the PVC (>0.16 second) has also been associated with heart dis-ease.28 In our study, only four of 101 subjects hadPVCs longer than 0.16 second. Considerable day-to-day variation in the occurrence of PVCs was ob-served. Similar variations of the frequency of PVCs inpatients with heart disease have been reported.29We did not find a significant correlation between the

presence and frequency of PVCs and variables thathave been associated with PVCs in previous reports(e.g., hypokalemia and use of coffee or alcohol). PVCshave been reported in subjects with markedhypokalemia and are evident in subjects with alco-holic cardiomyopathy, and occasionally in subjectswithout alcoholic cardiomyopathy who consumealcohol. The lack of correlation between PVCs andthese variables in our study is probably because thepotassium, sodium, hemoglobin, uric acid, and othervariables we studied did not show extreme deviationsfrom the normal, and only a few subjects consumedlarge amounts of alcohol and coffee (table 1).

AcknowledgmentThe authors thank Joanne DiPietro and Julianne Smith-Farese

for their statistical and editorial assistance.

1 355

by guest on June 20, 2018http://circ.ahajournals.org/

Dow

nloaded from

VOL 63, No 6, JUNE 1981

References1. Harrison DC, Fitzgerald JW, Winkle RA: Ambulatory elec-

trocardiography for diagnosis and treatment of cardiac arrhyth-mias. N Engl J Med 294: 373, 1976

2. Kennedy HL, Caralis DG: Ambulatory electrocardiography: aclinical perspective. Ann Intern Med 87: 729, 1977

3. Kotler MN, Tabatznik B, Mower MM, Tominaga S: Prog-nostic significance of ventricular ectopic beats with respect tosudden death in the late postinfarction period. Circulation 47:959, 1973

4. The Coronary Drug Project Research Group: Prognostic im-portance of premature beats following myocardial infarctions:experience of the Coronary Drug Project. JAMA 223: 1116,1973

5. Ruberman W, Weinblatt E, Goldberg JD, Frank CW, ShapiroS: Ventricular premature beats and mortality after myocardialinfarction. N Engl J Med 297: 750, 1977

6. Bigger JT Jr, Dresdal RJ, Heissenbuttel RH, Weld RM, WitAL: Arrhythmias in ischemic heart disease: mechanism, prev-alence, significance, and management. Prog Cardiovasc Dis 19:255, 1977

7. Kennedy HL, Underhill SJ: Frequent or complex ventricularectopy in apparently healthy subjects. Am J Cardiol 38: 141,1976

8. Kennedy HL, Pescarmona JE, Bouchard RJ, Goldberg RJ:Coronary artery status of apparently healthy subjects with fre-quent and complex ventricular ectopy. Ann Intern Med 92: 179,1980

9. Hinkle LE Jr, Carver ST, Stevens M: The frequency of asymp-tomatic disturbances of cardiac rhythm and conduction in mid-dle-aged men. Am J Cardiol 24: 629, 1969

10. Hinkle LE, Carver ST, Argyros DC: The prognostic sig-nificance of ventricular premature contractions in healthy peo-ple and in people with coronary artery disease. Acta Cardiol(suppl 18): 5, 1974

11. Raftery EB, Cashman PMM: Long-term recording of the elec-trocardiogram in normal population. Postgrad Med J 52 (suppl7): 32, 1976

12. Brodsky M, Wu D, Denes P, Kanakis C, Rosen KM: Arrhyth-mias documented by 24-hour continuous electrocardiographicmonitoring in 50 male medical students without apparent heartdisease. Am J Cardiol 39: 390, 1977

13. Clarke JM, Shelton JR, Hamer J, Taylor S: The rhythm of thenormal human heart. Lancet 2: 508, 1976

14. Glasser SP, Clark PT, Applebaum HJ: Occurrence of frequent

complex arrhythmias detected by ambulatory monitoring.Chest 75: 565, 1979

15. Gilson J: Electrocardiocorder-AVSEP patterns in 37 normaladult men: a four-year experience. Am J Cardiol 16: 789, 1965

16. DeMaria AN, Amsterdam EA, Vismara LA, Neumann A,Mason DT: Arrhythmias in the mitral valve prolapse syn-drome: prevalance, nature, and frequency. Ann Intern Med 84:656, 1976

17. Sones FM, Shirey EK: Cine coronary arteriography. ModConcepts Cardiovasc Dis 31: 735, 1962

18. Balke B, Ware RW: An experimental study of physical fitnessof air force personnel. US Armed Forces Med J 10: 675, 1959

19. Rosenbaum MB: Classification of ventricular extrasystoles ac-cording to form. J Electrocardiol 2: 289, 1969

20. Talbot ST, Dreifus LS: Characteristics of ventricular extra-systoles and their prognostic importance: a reappraisal of theirmethod of classification. Chest 67: 665, 1975

21. Barr AJ, Goodnight JH, Sall JP, Blair WH, Chilko DM: Sta-tistical Analysis System. Raleigh, North Carolina, SASInstitute, 1979

22. Blackburn H, Debacker G, Crow R, Prineas R, Jacobs D:Epidemiology and prevention of ventricular ectopic rhythms.Adv Cardiol 18: 208, 1976

23. Hiss RG, Averill KH, Lamb LE: Electrocardiographic find-ings in 67,375 asymptomatic subjects. Am J Cardiol 96: 1960

24. Manning GW, Ahuja SP, Gutierrez MR: Electrocardiographicdifferentiation between ventricular ectopic beats from subjectswith normal and diseased hearts. Acta Cardiol 23: 462, 1968

25. Kaplinsky E, Ogawa S, Kmetzo J, Dreifus LS: Origin of so-called right and left ventricular arrhythmias in acute myo-cardial ischemia. Am J Cardiol 42: 774, 1978

26. Bodenheimer MM, Banka VS, Helfant RH: Relation betweenthe site of origin of ventricular premature complexes and thepresence and severity of coronary artery disease. Am J Cardiol40: 865, 1977

27. Lown B, Calvert AF, Armington R, Ryan M: Monitoring forserious arrhythmias and high risk of sudden death. Circulation52 (suppl III): Ill-189, 1975

28. Huppert VF, Berliner K: The intraventricular conduction time(QRS duration) of ventricular premature systoles. Cardiology27: 87, 1955

29. Morganroth J, Michelson EL, Horowitz LN, Josephson ME,Pearlman AS, Dunkman WB: Limitations of routine long-termelectrocardiographic monitoring to assess ventricular ectopicfrequency. Circulation 58: 408, 1978

1356 CIRCULATION

by guest on June 20, 2018http://circ.ahajournals.org/

Dow

nloaded from

J B Kostis, K McCrone, A E Moreyra, S Gotzoyannis, N M Aglitz, N Natarajan and P T KuoPremature ventricular complexes in the absence of identifiable heart disease.

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1981 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.63.6.1351

1981;63:1351-1356Circulation. 

http://circ.ahajournals.org/content/63/6/1351the World Wide Web at:

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

  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 information about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. FurtherEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:

by guest on June 20, 2018http://circ.ahajournals.org/

Dow

nloaded from