Acta Med Scand 210: 73-77, 1981

QT Interval in Relation to Ventricular Arrhythmias and Sudden Cardiac Death in Post myocardial Infarction Patients

Mogens Mvrller

From the Depariments ojCardiology and Clinical Physiology, Odense University Hospital, Odense, Denmurk

ABSTRACT. Ninety-one consecutive patients below the age of 70 years were subjected to a 60-second resting ECG and 24-hour ambulatory ECG monitor- ing two weeks and one, three and six months after an acute myocardial infarction. The corrected QT (QT,) interval decreased from the late hospital phase to the investigations three and six months after the infarc- tion (p<O.Ol, C0.05). ECG monitorings showing complicated ventricular ectopic beats (multiform, re- petitive, R-on-T) were associated with an insignifi- cantly longer QT,. than other recordings. Eleven pa- tients suffered a sudden cardiac death during a me- dian follow-up period of 24 months (range 22-27). The QT, intervals in patients who died suddenly were insignificantly longer than in the survivors. Only four patients, who all survived, had a constantly prolonged QT,. After exclusion of tracings during quinidine therapy, a QT, longer than 440 msec was found in 7 (23 %) of 31 recordings from patients who suffered a sudden cardiac death compared to 29 (10%) of 287 recordings from the survivors (pC0.05). A combina- tion of complicated ventricular ectopic beats and a QT, longer than 440 msec was demonstrated in 5 (16%) of 31 and 14 (5%) of 287 recordings from the two groups of patients (p<0.05). A trend towards longer QT, intervals was observed in patients with complicated ventricular arrhythmias and in those who died suddenly, but no well defined high-risk groups could be identified.

K e y words: QT, interval, ventricular arrhythmias, am- bulatory electrocardiography, acute myocardial infarc- tion, prognosis.

Acta Med Scand 210: 7 3 , 1981.

Congenital prolongation of the Q T interval is as- sociated with a high incidence of sudden cardiac death due to ventricular fibrillation (8, 10). Within recent years the QT interval has been shown to predict severe ventricular arrhythmias in the acute phase of a myocardial infarction (2), while reports dealing with the long-term prognostic implication of

the corrected QT (QT,) interval are more diverg- ing-one showing an association between a long QT, and sudden death (9) and another a shorter QT, in the deceased patients (1).

The object of the present study was to elucidate any relationship between the QT, interval and the occurrence of ventricular arrhythmias during re- peated 24-hour ambulatory electrocardiographic (ECG) monitoring in postinfarction patients and to relate this back to the occurrence of sudden cardiac death

PATIENTS AND METHODS One hundred consecutive patients (20 women, 80 men) below the age of 70 years (average 59, range 40-69) dis- charged alive from the Department of Cardiology, Odense University Hospital, following admission for definite acute myocardial infarction (AMI) were subjected to this study. Twenty-two patients had suffered a previous in- farction. The infarction was anterior in 44, inferior in 34, anterior + inferior in 15 and indefinite in 7 . Q waves developed in 68 patients.

On one of the last days prior to discharge as well as one, three and six months after the infarction, each patient was subjected to a 60-sec resting 12-lead ECG and to a 24-hour 2-lead ECG tape recording using a Medilog tape recorder (Oxford Instruments Ltd.). From the resting ECGs, re- corded at a paper speed of 50 mmlsec, the heart rate and QT interval were calculated from four and three consecu- tive beats, respectively. Using a Hewlett Packard 981OA Digitizer/98lOA Computer the R waves, the onset of the QRS complexes and the end ofthe T waves were marked,

T, was calculated by the formula of Bazett: QT

the T wave was defined a s the intersection between the hind limb of the T wave and the baseline. Duplicate mea- surements of QT intervals were made in 40 randomly cho- sen ECGs. The intrareader variability averaged 11 msec.

Correspondence to: M. Mdler, M.D., Department of Cardiology B, Odense University Hospital, DK-5000 Odense C, Denmark. Abbreviations: QT, = corrected QT interval, ECG = electrocardiography, electrocardiographic, AM1 = acute myocardial infarction.

?p- - R-R. Whenever possible, lead I1 was used. The end of

Acta Med Scand 210

74 M. Mollrr

Period following infarction

2 weeks 1 month 3 month5 6 months

No. of pats. Heart rate (min-') QT (msec) QT,. (msec)

91 86 87 84 7 5 t 1 3 7 5 t I 4 75+15 75 + I 4

381+37 376t36 366+36** 370?34* 421k31 417+30 405+34** 410t31'

* p<0.05, ** p<0.01 compared to measurement 2 weeks after infarction.

The 24-hour ECG monitorings were analyzed at 60 times real time by means of a Pathfinder High Speed ECG Analyser (Reynolds Medical Electronics Ltd.), and the ventricular arrhythmias were described both qualitatively and quantitatively. Uniform ventricular ectopic beats were, irrespectively of their number, termed uncompli- cated ventricular arrhythmias, while multiform, repetitive and R-on-T ventricular ectopic beats were termed compli- cated ventricular arrhythmias. Sudden cardiac death was defined a s sudden cardiovascular collapse causing death or requiring cardiopulmonary resuscitation with docu- mented ventricular fibrillation within 24 hours of the onset of symptoms. A QT,. exceeding 440 msec was defined a s prolonged.

Serum electrolytes (sodium, potassium, calcium) were determined at the investigations two weeks and six months after the infarction and, further, in relation to the other examinations if the patients were on treatment with digoxin and/or diuretics.

The unpaired 1-test (two-tailed) and the X'-test were employed for statistical analyses. Significant differences were defined a s p<0.05.

RESULTS

Due to bundle branch block or flat and biphasic T waves, data were available from 91 patients only. Four of them died suddenly within one month after

the infarction, seven after the final ECG examina- tions. Two patients died from other causes. A total of 348 resting ECGs and corresponding 24-hour ECG monitorings were subjected to analysis.

The electrolytes did not show hypokalemia, while serum potassium was slightly increased in five examinations (5.1-5.5 mmol/l). No significant hypo- or hypercalcemia was observed. The heart rate, measured on the resting ECGs, was unchanged throughout the period of investigation (Table I ) . The QT,. interval was significantly shorter three and six months after the infarction than in the late hospi- tal phase (p<O.OI, <0.05).

Table I 1 shows the QT, intervals i n relation to drug therapy at the four occasions. The use of dig- oxin, P-blocking agents and quinidine did not dif- fer significantly from time to time (~"4.17, df=9, p=0.90). One patient was on procainamide during the first three registrations. From Table I I it can be seen that patients on P-blocking agents and patients on quinidine had a shorter respectively longer QT,. interval than patients off these drugs. Digoxin alone did not seem to affect the QT interval. The patient on procainamide had a QT,. varying from 440 to 460

Period folio wi ng infarction

2 weeks 1 month 3 months 6 months

N o . of ECGs Digoxin

P-Blocker ( + digoxin)

Quinidine (i digoxin)

None of these

YO" 420+33 ( n = 14) 411+12 (n=14) 443 524% ( n = 8 ) 419+?7 (ti =54)

85 " 425 +30 ( n = l l ) 398+31* (n=16) 437 236 (n=7) 417+26 ( n = 5 1 )

86" 410+36 (n=13) 385+33* (n=16) 436 +42** ( n = 8 ) 404+28 (n=49)

84 413531 (n=19) 399k33 ( n = 1 6 ) 436+26* (n=7) 40Y+2Y (n=42)

' I One patient on procainamide excluded. * p<O.OS, ** p<0.01 compared to patients not treated with digoxin. P-blocker or quinidine.

1, r l i I / S l (id 210

QT and sudden cardiac drath 75

Table 111. QT,. (metin k S . D . ) in rrlution to ventriculur ectopic' h r ~ t s (VEB) during 24-hour ECG tape recording

Period following infarction

2 weeks 1 month 3 months 6 months

No. of pats. 91 N o or uncomplicated VEB 419+33

(n=59) Complicated VEB 424+28

( n = 3 2 ) Significance NS Mean VEB/h < 10 417+31

(n=64) Mean VEB/h >I0 429+31

(n=27) Significance NS

86 41 I k37 (n=45) 419L26 (n=41) NS 410k32 (n=46) 425k27 (n =40) p<0.05

87 403k35 (n =54) 410k37 (n=33) NS 399+33 (n=48) 413235 (n=39) NS

84 404i33 (n=51) 416k30 (n=33) NS 407232 (n=49) 414229 (n=35) NS

NS = not significant @>0.05)

msec. The 62 ECGs recorded during /?-blocking treatment showed a mean QT, of 398k35 (S.D.) msec compared to 438f31 msec for the 30 record- ings obtained from patients receiving quinidine

N o significant association between QT, and the Occurrence of complicated ventricular arrhythmias was found (Table I I I ) , but a tendency is clear. One month after the infarction, patients with frequent ventricular ectopic beats had a significantly longer QT,. than the others (p<0.05) and a similar trend was demonstrated in the other examination. The prescribed therapy with /?-blocking agents and quinidine did not differ significantly between the two groups.

The QT, intervals were insignificantly shorter in the 78 patients surviving the follow-up period than in those I I who died suddenly (Table IV). The difference was most pronounced within the first month after the infarction. The use of /?-blocking agents was quite similar during ECG recording in

( p <0.00 I ).

Table IV. Relutionship betw>een QT,. (msec) and outcome (mean k S . D . )

Period following Survivors Non-survivors infarction (n=78) ( n = l l )

2 weeks 419k30 435k25 ( n =78) ( n = l l )

I month 417k31 426+28 (n=77) (n=7)

3 months 405k33 416k43 (n=77) (n=7)

6 months 410229 416248 (n=75) (n=7)

the two groups of patients. On the other hand, only one (3 %) of 32 recordings from patients dying sud- denly was made during quinidine therapy compared to 29 (9%) of the remaining 316 (NS). If all ECGs recorded during quinidine therapy were excluded, the difference of mean QT, between the survivors and non-survivors did increase, but a significant level was not reached.

Fig. 1 shows the distribution of QT, intervals in relation to outcome. In a total of 49 ECGs the QT, interval exceeded 440 msec. Eight (25%) of these ECGs were among the 32 ECGs from patients who subsequently suffered a sudden death, which means that 41 (13%) of the 316 ECGs from the survivors showed an abnormally long QT, interval (NS). Af- ter excluding quinidine-affected registrations, the corresponding figures were 7/3 1 (23 %) and 29/287 ( lo%) , respectively (p<0 .05) . When all recordings made during treatment with digoxin, p-blocker, quinidine and procainarnide were excluded, the proportions were 3/12 (25%)and 15/184 (8%) (NS).

0 Survivors (n-3071

Oereased (n- 321

r

300 320 340 360 380 LOO 4 2 0 4 4 0 460 480 500 OT, irnsecl

Fig. 1 . Distribution of QTc values in 78 patients surviving the follow-up period and in I I patients who died suddenly.

M . M d l e r 76

20

Ln + c 01

+ -

B 10

L (Y n

9 z

0

Survivors (n=78I

Deceased ln.71

I I

I I I I I I

L 320 340 360 380 400 420 440 L ~ O 480

Mean QT, lrnsecl

Fig. 2 . Mean QT, values in 78 patients surviving the follow-up period and in 7 patients dying suddenly after the final ECG examination.

Nine (27%) of the 33 ECGs showing left anterior hemiblock had a QT, longer than 440 msec com- pared to 40/315 (13%) of the remaining ECGs (pC0.05). A combination of complicated ventricu- lar arrhythmias and a QT, interval longer than 440 msec was observed during 5 (16%) of 32 examina- tions in patients who suffered a sudden cardiac death compared to 22/316 (7%) of the remainder (NS). Apart from the quinidine-treated patients, the proportions were 5/31 and 14/287 (p<O.OS).

Fig. 2 shows the mean QT, of the 78 patients surviving the follow-up period and the seven pa- tients who died suddenly more than six months after the infarction. It is obvious that a considerable overlap exists between the two groups. However, it may be mentioned that four of the 10 patients with a mean QT, longer than 440 msec were receiving quinidine during at least three of the four regis- trations and one patient was treated with pro- cainamide. The QT, was constantly prolonged in only four patients. Digoxin was prescribed to one of these patients at all four monitorings, digoxin to- gether with quinidine to the three others. None of these patients died.

DISCUSSION

Ventricular arrhythmias are common in patients with a recent myocardial infarction and have in some investigations been associated with an in- creased risk of sudden cardiac death ( 5 , 7 , 11) . As a prolonged QT, may be an indicator of asynchro- nous repolarization (3), the combination of long QT, and certain types of ventricular ectopic beats may possibly facilitate the development of fatal ar- rhythmias due to re-entry mechanisms. In the pres-

Acta Med Scand 210

ent study, most QT, intervals did not exceed an upper limit of 440 msec. This is in accordance with the findings reported by others (1, 4, 6), but in contrast to those in a study by Schwartz and Wolf (9). The latter authors found a mean QT, interval longer than 440 msec during repeated registrations over several years in 38% of 55 postinfarction pa- tients, while Haynes et al. (4) found a prolonged QT, in 18% of their 98 patients and Ahnve et al. ( I ) found a QT, exceeding 440 msec in 3 1 7% of their 463 patients examined once. This difference possibly depends on medication, which is not discussed in detail by Schwartz and Wolf, whereas 6 % of the patients of Haynes et al. and 11 7% of the patients of Ahnve et al. received quinidine.

A relation between the occurrence of ventricular tachycardia or ventricular fibrillation and a long QT, before the event has been found in the CCU ( 2 ) . Studies in the posthospital phase of the infarction have not been conducted before on the relationship between these two parametres, but it has been clearly demonstrated that the mean QT, of survivors of out-of-hospital ventricular fibrillation is significantly longer than that of postinfarction patients without this arrhythmia (4). In the present study, no significant association could be demon- strated between the length of the QT, interval and the occurrence of complicated ventricular ar- rhythmias during long-term ECG registration in the late hospital or the early posthospital period of a myocardial infarction, but a persistent trend to- wards an interrelation was noticed.

The long-term prognostic implication of the QT, interval measured after the acute phase of a myocardial infarction is controversial. Schwartz and Wolf (9) performed repeated QT, measure- ments in 55 postinfarction patients during a seven- year follow-up period. In their study, the average mean QT, interval of the 28 patients who died sud- denly was significantly longer than that of the sur- vivors, and 77% of their 21 patients with a mean QT, exceeding 440 msec died suddenly. In contrast to this the very recent report by Ahnve et al. (1 ) dealing with QT, measurement in the late hospital phase of 463 survivors of an AMI demonstrated a significantly shorter QT, in the deceased patients than in patients alive after 3-6 years. After exclu- sion of all patients with bundle branch block and those on quinidine or digitalis, mean QT, was al- most identical in survivors and non-survivors. Fi- nally, in a subgroup of patients below 66 years of

QT and sudden cardiac death 77

age fulfilling the above criteria, QT, tended to be longest in those who died suddenly-<specially in those who died within six months of the infarction. This subgroup is in many ways comparable with the patients in the present study.

In the present study, few patients died suddenly. The association between a long QT, and cardiac death might have reached a significant level if the number of patients had been larger. On the other hand, the study, in agreement with Ahnve et al. ( I ) , demonstrates that only a minority of QT, measure- ments in patients who subsequently suffered a sud- den death exceeded 440 msec. This means that QT, could not be expected to be a sensitive and specific predictor of the outcome.

The combination of prolonged QT, and the oc- currence of complicated ventricular arrhythmias during long-term ambulatory ECG monitoring was of some help in predicting the outcome, but a well defined high-risk group could not be identified.

ACKNOWLEDGEMENT

Supported by a grant from the Danish Heart Foundation (Hjerteforeningen).

REFERENCES

I . Ahnve, S., Helmers, C. & Lundman, T.: QT, inter- vals at discharge after acute myocardial infarction and long-term prognosis. Acta Med Scand 208: 5 5 , 1980.

2. Ahnve, S. , Lundman, T. & Shoaleh-var, M.: The relationship between QT interval and ventricular ar- rhythmias in acute myocardial infarction. Acta Med Scand 204: 17, 1978.

3. Han, J. , Millet, D., Chizzonitti, B. & Moe, G. K.: Temporal dispersion of recovery of excitability in atrium and ventricle as a function of heart rate. Am Heart J 71: 481, 1966.

4. Haynes, R. E . , Hallstrom, A. P. & Cobb, L. A.: Repolarization abnormalities in survivors of out-of- hospital ventricular fibrillation. Circulation 57: 654, 1978.

5 . Moss, A. J., Davis, H. T., DeCamilla, J. & Bayer, L. W.: Ventricular ectopic beats and their relation to sudden and non-sudden cardiac death after myocar- dial infarction. Circulation 6 0 988, 1979.

6. Nyberg, G., Vedin, A. & Wilhelmsson, C.: QT time in patients treated wit: alprenolol or placebo after myocardialinfarction. Br Heart J 41: 452, 1979.

7. Rehnqvist, N., Lundman, T. & Sjogren, A.: Prognos- tic implications of ventricular arrhythmias registered before discharge and one year after acute myocardial infarction. Acta Med Scand 204: 203, 1978.

8. Schwartz, P. J., Periti, M. & Malliani, A.: The long Q-T syndrome. Am Heart J 89: 378, 1975.

9. Schwartz, P. J. & Wolf, S.: QT interval prolongation as predictor of sudden death in patients with myocar- dial infarction. Circulation 57: 1074, 1978.

10. Vincent, G. M., Abildskov, J. A. & Burgess, M. J.: The Q-T interval syndromes. Prog Cardiovasc Dis 16: 523, 1974.

1 I . Vismara, L. A., Vera, Z., Foerster, J. A., Amster- dam, E. A. & Mason, D. T.: Identification of sudden death risk factors in acute and chronic artery disease. Am J Cardiol39: 821, 1977.

Acra Med Scand 210