Effect of Treatment on Morbidity in Hypertension.
Veterans Administration Cooperative Study

on Antihypertensive Agents

Effect on the Electrocardiogram

By P. F. Poprere, M.D., M. C. Kyze, M.S., H. V. Presercer, M.D.,
AND E. D. Freis, M.D.

SUMMARY

Electrocardiographic (ECG) data.are presented from the Veterans Administration Cooperative
Study of 143 nontreated and 137 treated male patients with initial diastolic blood pressures
averaging 90-114 mm Hg. Average follow-up was 2.9 years and longest follow-up was 5 years.
Significant differences. were observed with respect to the ECG signs of left ventricular hyper-
trophy (LVH)..In the patients. not meeting criteria for LVH prior to randomization, the inci-
dence of abnormal QRS voltage, ST-segment depression or T wave flattening or inversion de-
veloping in the treated patients was approximately one-fourth that found in the control group.
In the-patients with ECG evidence of LVH prior to randomization, the reversion to normal of
QRS voltage or ST-segment depression was approximately 2.5 times greater in the treated than
in the control patients. Although no significant effect of treatment was observed on other ECG
changes such as Q and QS patterns, A-V or ventricular conduction defects or arrhythmias, the
incidence of such events was too low to make valid comparisons. However, the present results
indicate that antihypertensive drug treatment markedly improved the ECG changes specifically

related to hypertension.

Additional Indexing Words:
Systolic blood pressure
Electrocardiogram

REVIOUS REPORTS of serial electrocardio-
grams taken during the course of antihyperten-
sive drug treatment have indicated that certain
electrocardiographic abnormalities associated with
hypertension may revert toward normal following
treatment. All of these studies, however, were

Diastolic blood pressure
ST segment changes and hypertension

Antihypertensive treatment
T wave changes and hypertension

retrospective and did not include untreated con-
trols. Helmcke, Schneckloth and Corcoran, in their
study of 78 hypertensive patients, found that
although ECG improvement was more common in
those classified as blood pressure responders as
compared to the nonresponders the relationship was

 

From the Veterans Administration Hospital, Washington,
D.C.

Permanent members of the study group: Massimo
Calabresi, M.D., C. Hilmon Castle, M.D., Leo Elson, M.D.,
Edward D. Freis, M.D. (Chairman), Rudolph E. Fremont,
M.D., Michael A. Harris, M.D., David Littman, M.D., Eli A.
Ramirez, M.D., and J. R. Thomas, M.D., Other members:
Luis A. Arias, M.D., Mark L. Armstrong, M.D., Alston W.
Blount, M.D., Thomas Bruce, M.D., Ovid B. Bush, Jr., M.D.
(deceased), Eugene C. Clark, M.D., Annette Fitz, M.D.,
R.M. Freeman, M.D., Edward D. Frohlich, M.D., Arthur
Gear, M.D., John D. Kyriacopoulos, M.D., Alan R. Lyon,
M.D., Gloria D. Massaro, M.D., Donald McCaughan, M.D.,
Jean Morgan, M.D., Henry W. Overbeck, M.D., Elisea C.
Perez-Stable, M.D., Mitchell H. Perry, M.D, Roger Sutton,
M.D., and James Taguchi, M.D. Biostatisticians: Russel B.
Tewksbury, Sc.D. (deceased), Paul D. Williams, M.S., and

Circulation, Volume XLVIII, September 1973

481

Lawrence Shaw, A.M. Central Office Coordinator: Harold
W. Schnaper, M.D. Participating Veterans Administration
Hospitals: Allen Park, Michigan, Birmingham, Alabama,
Brooklyn, New York, Dayton, Ohio, Iowa City, Iowa,
Jackson, Mississippi, Memphis, Tennessee, Nashville, Ten-
nessee, Oklahoma City, Oklahoma, Pittsburgh, Pennsylvania,
Richmond, Virginia, Salt Lake City, Utah, St. Louis,
Missouri, San Juan, Puerto Rico, Washington, D.C., West
Haven, Connecticut and West Roxbury, Massachusetts.

Supported in part by Research Grant HL 15047 from the
National Heart and Lung Institute, National Institutes of
Health, U.S. Public Health Service.

Address for reprints: Dr. Edward D. Freis, Veterans
Administration Hospital, 50 Irving Street, N.W., Washing-
ton, D.C. 20422

Received February 13,
publication April 18, 1973.

1973; revision accepted for
482, Table 1

Classification of Patients According to Number of Post-
randomization Annual Electrocardiograms

 

 

Control group Treated group
No. No.

No. of annual exams

 

1 143 137
2 116 112
3 83 81
4 47 50
5 20 20

 

not consistent.? Dern et al. found improvement in
the electrocardiograms of 44 treated patients.‘
However, a valid estimation of benefit of antihyper-

tensive drug treatment based on such studies must .

be limited because of the absence of an untreated
control group with which to compare results.
Considerable electrocardiographic variability can
occur over time in both normal individuals® and in
patients with heart disease.? Without a control
group it is not possible to differentiate between
therapeutic effect and the effect of such sponta-
neous variability.

The present report describes the electrocardio-
graphic changes occurring during the course of the
Veterans Administration Cooperative Study on
antihypertensive agents. The patients were ran-
domly assigned, double blind to either active drugs
or placebos. Follow-up averaged 2.9 years and in
some cases was longer than 5 years. This study,
therefore, permitted a quantitative assessment of
the benefits of antihypertensive medications, if any,
on the electrocardiographic abnormalities associat-
ed with hypertension.

Methods and Materials

The design of the cooperative study and an analysis
of morbid events which occurred other than electrocar-
diographic changes have been described previously.®: 9
All patients were males who exhibited an average
diastolic blood pressure of 90 mm Hg or higher from
the fourth through the sixth day of hospitalization. Also,
the average diastolic blood pressure of the last two
posthospitalization clinic visits prior to randomization

POBLETE, KYLE, PIPBERGER, FREIS

was.in the range of 90-114 mm Hg. The patients were
then randomly assigned, double-blind to either active
drugs (hydrochlorothiazide 50 mg plus reserpine 0.1
mg combined in a single tablet given twice daily, and
hydralazine 25 or 50 mg three times daily) or to
placebos of these agents.

Copies of the prerandomization and one year
postrandomization 12 lead electrocardiograms were
obtained in 143 of 194 patients randomized into the
control or placebo treated group and in 137 of 186
treated patients. In addition, some of these patients’
electrocardiograms were obtained for 2, 3, 4 and 5
years postrandomization (table 1), the average being
2.9 years.

The background characteristics of the patients are
presented in tables 2 and 3. Approximately 42% of the
patients were black in both the control and treated
groups. Those showing cardiac enlargement by X-ray
were 22% in the control patients and 28% of the treated
patients (table 2). The average age at entrance in both
groups of patients was approximately 51 years (table
3). Average heights and weights were not significantly
different. The known duration of hypertension averaged
4.5 years in both groups of patients. The average
prerandomization blood pressure in the control group
was 164/104 and in the treated patients was 164/105
mm Hg. There were no significant differences in the
two groups of patients with respect to serum creatinine
or blood urea nitrogen levels, serum potassium
concentration, fasting blood sugar or serum cholesterol
values.

The changes in blood pressure following randomiza-
tion are shown in table 4. The average blood pressure of
the control patients changed little from the prerandom-
ization values except for a slight rise in systolic blood
pressure. In the treated patients the average blood
pressure fell from a prerandomization value of 164/105
mm Hg to a four month post-randomization average of
137/87 mm Hg and remained at essentially that level
or slightly lower for the duration of the study.

All of the collected electrocardiograms were read and
classified by one of the authors (P.P.) without his
knowledge of any other clinical features of the patients
including type of therapeutic regimen or changes in
blood pressure. The electrocardiographic characteristics
were classified according to the Minnesota code as
revised by Rose and Blackburn.'’ This classification
eliminates some of the ambiguities present in the
original code!1 and includes a more comprehensive set
of procedural rules.

Table 2

Background of Randomized Patients: Numeration Data

 

 

Control group

Treated group

 

Characteristic No o No % Total
Number randomized 143 137 . 280
Black 59 41 57 42 116
Other 84 59 80 58. 164
Heart size by roentgenogram
Ungerleider enlarged 31 22 39 28 70

 

Circulation, Volume XLVIII, September 1973
EFFECT OF ANTIHYPERTENSIVE AGENTS ON ECG 483

Table 3

Measurement Data Prior to Randomization

 

 

Control group Treated group

 

Characteristic Mean Mean
Age (yrs) 51.6 50.3
Height (in) 69.1 68.4
Weight (ib) 181.8 176.6
Duration of known hypertension 4.4 4.5

(yrs)
Average clinic systolic BP 164.1 163.5
(mm Hg)
Average clinic diastolic BP 103.9 104.7
(mmHg)
Serum creatinine (mg/100 ml) 1.27 1.24
Blood urea nitrogen (mg/100 ml) 16.0 16.4
Serum potassium (mEq/liter) 4.5 4.4
Fasting blood sugar (mg/100 ml) 96.1 100.2
Cholesterol] (mg/100 ml) 245.5 247.6

 

In the analysis of Category 4 of the revised
Minnesota code (ST-junction and segment depression)
only items 4-1 to 4-3 were included. These items
require ST-segment depressions that are horizontal or
downward sloping which are considered to be more
representative of “ischemia” than is item 4-4 which
includes upward sloping or U-shaped ST-segments.12
Also excluded were optional code 5-4 of category 5 (T
wave items) and category 9 (miscellaneous items). In
addition to the above analyses the sum of the
magnitudes of the S wave in V, and the largest R wave
in either V; or Vg was calculated. A sum exceeding 35
mm was considered to be indicative of left ventricular
hypertrophy in accordance with Sokolow and Lyon’s
criteria.14

As shown in table 5 the prerandomization electrocar-
diograms were similar in the control and treated groups
of patients except that a higher percentage uf the
treated group exhibited high amplitude R waves in the
left side leads and deep S waves in V,. Evidence of Jeft
ventricular hypertrophy using Sokolow’s criteria was
present in 32.1% of the treated group as compared to

only 18.2% of the control patients prior to randomiza-
tion. The prevalence of T wave flattening or inversion,
however, was not significantly different in the two
groups of patients nor were there any significant
differences with respect to any of the other electrocar-
diographic characteristics that were analyzed.

Results
Incidence of Abnormalities

The incidence of electrocardiographic abnormal-
ities developing over an average follow-up of 2.9
years after randomization is shown in table 6 and
figure 1. Marked differences in voltage, T wave and
ST-segment changes were found in the control as
compared to the treated patients. Thus, of the
patients not meeting criteria for LVH prior to
randomization, 25% of the control and only 6.6% of
the treated group developed abnormal voltage. In
addition, 25.5% of the control patients as compared
to 7.5% of the treated group developed T wave
flattening or inversion. The incidence of’ ST-
segment depression was 19.3% in the placebo group
and 3.8% in the treated patients. All of these
differences were significant (P< .005). There were
no significant differences between the two groups of
patients in the incidence of Q and QS patterns, left
axis deviation, A-V or ventricular conduction
defects or arrhythmias.

Figure 2 shows the incidence of ST-T and voltage
changes by the life table method of analysis. This
method adjusts for the variable length of follow-up
of different patients. It also adjusts for any
differences in losses to observation between the
control and treated groups and it determines
whether the benefit of treatment occurs early or late

Table 4

Trends of Blood Pressure

 

 

Control group

Treated group

 

Average Average Average Average
systolic diastolic systolic diastolic
blood blood blood blood
Time of No. pressure pressure No. pressure pressure
observation observed (mm Hg) (mm Hg) observed (mm Hg) (mm Hg)
Prerandomization 143 164 104 137 164 105
Postrandomization '
4 months 142 168 105 137 137 87
8 months 141 170 107 137 137 88
12 months 143 168 105 137 136 86
16 months 116 171 106 112 135 86
20 months 116 173 106 112 133 86
24 months 113 173 107 109 134 86
28 months 83 175 107 81 134 86
32 months 83 172 104 80 135 85
36 months 82 170 103 79 134 86

 

Circulation, Volume XLVIII, September 1973
484

POBLETE, KYLE, PIPBERGER, FREIS

Table 5

Electrocardiographic Data Prior to Randomization

 

 

Control group

Treated group

 

No. No. with No. No. with
Llectrocardiographic tracings codable tracings codable
item reaa* items % read items %
Total randomized 143 137
Q & Q-5 patterns 143 12 8.4 136 10 7.4
Left axis deviation 137 10 7.3 133 8 6.0
High amplitude R waves (eft) 140 29 20.7 134 47 35.1
ST-segment depression 132 * 18 13.6 126 21 16.7
T wave flattening or 132 30 22.7 126 33 26.2
inversion
A-V conduction defect 143 4 2.8 137 2 1.5
Ventricular conduction 148 5 3.5 137 3 2.2
defect
Arrhythmias 143 10 7.0 137 6 4.4
Increased voltage 137 25 18.2 134 43 32.1
(S in Vi + R. in Vs or Ve
> 35 mm)

 

*Numbers vary because of exclusions for evaluating some of the codable items as described in Minnesota

code.®

or is continuous throughout the period of observa-
tion.

It is evident from figure 2 that a lower incidence
of ST-T and voltage changes was manifested early
in the treated group and that the difference in
incidence of such ECG changes between the control
and the treated group increased with the passage of
time. Thus, for ST-segment depression the 5 year
cumulative incidence rate for the control group was
29% as compared to 5% in the treated group. With
respect to T wave flattening or inversion at 5 years
the cumulative incidence rate was 38% for the
control group and 10% for the treated patients.

Finally, abnormal voltage change was 34% at 5 years
in the control group as compared to 9% in the
treated group.

Only a small proportion of patients developed
both voltage and ST-segment or T wave abnormal-
ities. Such a change occurred in 10 control patients
and in 2 treated patients. Four of these 10 control
patients exhibited neither voltage nor ST-T abnor-
malities prior to randomization while 6 showed one
or the other of these abnormalities in the preran-
domization period. The latter prerandomization
finding also was present in the two treated patients

Table 6

Patients with Normal Prerandomization Electrocardiograms Which Developed Abnormality (Codable Item)

Postrandomization

 

 

Contro! group

Treated group

 

randenized No. randossized No. Significance of
Electrocardiographie without developing without developing difference
item item abnormality % item abnormality % (P-value)*
Q & Q-S patterns 131 10 7.6 126 5 4.0 NS
Left axis deviation 127 3 2.4 125 1 0.8 NS
High amplitude R waves (left) 1i1 32 28.8 87 5 5.7 <0.005
ST-segment depression 114 22 19.3 105 4 3.8 <0.005
T wave flattening or inversion 102 26 25.5 93 7 7.5 <0.005
A-V conduction defect 139 9 6.5 135 2 1.5 NS
Ventricular conduction defect 138 4 2.9 134 3 2.2 NS
Arrhythmias 133 15 11.3 131 17 13.0 N85
Increased voltage 112 28 25.0 91 6 6.6 <0.005
(Sin V; + R in V; or V5
>35 mm)

 

*Hither the x? test or the Fisher exact probability test was used depending on sample size according to criteria given in refer-

ence 15.

Circulation, Volume XLVII, September 1973
EFFECT OF ANTIHYPERTENSIVE AGENTS ON ECG

PER CENT PATIENTS

100 I

1

30

 

 

MB CONTROL

P<0. 005

P<0. 005

(J TREATED

P<0, 005

 

° ABNORMAL QRS T-WAVE FLATTENING —S-T SEGMENT
VOLTAGE OR INVERSION DEPRESSION
Figure 1

Percent of patients developing either voltage or ST-T ab-
normalities following randomization who did not exhibit
such change prior to randomization.

who developed both voltage and ST-T changes
during the postrandomization period.

Reversions to Normal

Of the patients exhibiting voltage criteria for left
ventricular hypertrophy prior to randomization
74.4% of the treated patients as compared to 24.0%
of the control group reverted to normal voltage
(P <.005) (table 7, fig. 3). A favorable trend also
was evident with respect to flattened or inverted T
waves with 51.5% of the treated and 33.3% of the
control group reverting to normal, although this
difference was not statistically significant. A signifi-
cant difference was observed, however, with respect
to normalization of ST-segment depression. Such
reversion was seen in 57.1% of the treated patients
and 22.2% of the control group (P<.05). No

485

4
60

S-T SEGMENT DEPRESSION

40
20

 

 

T-WAVE FLATTENING OR INVERSION

 

VOLTAGE

40
20

| 2 3 4 5
years of observation

 

Control group
-----. Treated group
Figure 2

Estimated cumulative incidence of either ST-T or voltage
abnormalities over a fwe-year period as calculated by the
life-table method.

significant differences were found with respect to
reversion to normal of other electrocardiographic

Table 7

Patients with Prerandomization Electrocardiograms Showing Abnormality (Codable Item) which Reverted to

Normal Postrandomization

 

Control group

Treated group

 

 

randosnized No. randsonized No. Significance of

Electrocardiographic with reverting to with reverting to difference

item item normal % item norinal % (P-value}*
Q & Q-5 patterns 12 3 25.0 10 2 20.0 NS
Left axis deviaticn 10 2 20.0 8 4 50.0 NS
High amplitude R waves (left) 29 8 27.6 47 35 74.5 <0.005
ST-segment depression 18 4 22.2 21 12 57.1 0.029
T wave flattening or inversion 30 10 33.3 33 17 51.5 NS
A-V conduction defect 4 3 75.0 2 0 0.0 NS
Ventricular conduction defect 5 1 20.0 3 0 0.0 NS
Arrhythmias 10 8 80.0 6 5 83.3 NS
Increased voltage 25 6 24.0 43 32 74.4 <0.005

(Sin Vi + R in V; or V5
> 35 mm)

 

*Same notation as in table 6.

Circulation, Volume XLVIIL, September 1973
486

HM CONTROL [__] TREATED

2
0
4

®
oO
T

P<0. 005

~
°
lj

P<0. 05

PER CENT PATIENTS
rn WwW BB Q Q
oO O O09 68 6S
T q I ' I

5
T

 

 

 

 

 

ABNORMAL QRS
VOLTAGE

S-T SEGMENT
DEPRESSION

Figure 3

Percent of patients reverting from abnormal to normal ST-T
or voltage change during the postrandomization period.

characteristics. The prevalence of some of these
characteristics was too low, however, to permit a
reliable estimate (table 7).

Figure 4 shows the incidence of reversions to
normal of ST-T and voltage changes during the
postrandomization period as indicated by the life-
table method of analysis. The figure indicates a
relatively high incidence of reversion to normal. in
the first year of follow-up as compared to
subsequent years. Few reversions occurred after the
third year of follow-up. With respect to ST-segment
depression the 5 year cumulative reversion rate to
normal was 63% in the treated group as compared to
22% in the control group (fig. 4). The reversion rate
of T wave changes for the cumulative 5 year period
was 56% in the treated patients and 36% in the
control patients. With respect to voltage the 5 year
cumulative reversion rate was 87% of the treated
group and 26% of the control group.

Nine patients in the control and 15 in the treated
group exhibited abnormal voltage plus T wave or
ST-segment criteria for left ventricular hypertrophy
prior to randomization. Disappearance of all of
these evidences of left ventricular hypertrophy was
noted following randomization in 9 or 60% of the
treated group as compared to none of the control
patients (P< 0.005).

 

 

POBLETE, KYLE, PIPBERGER, FREIS
S-T SEGMENT DEPRESSION

a

 

 

 

years of observation

 

Control group

eon Treated group

Figure 4

Estimated cumulative incidence of either ST-T or voltage
reversion from abnormal to normal over a five-year period
as calculated by the life-table method.

Relationship of Effectiveness of Treatment
to Level of Prerandomization Blood Pressure
and to Age

Approximately one-fourth of the control group
developed voltage criteria for left ventricular
hypertrophy during the postrandomization period.
However, in the treated group a smaller percentage
of patients with blood pressures of 165 systolic and
above or diastolic pressures of 105-114 mm Hg
developed increased voltage than did those with
lower blood pressures (table 8). For this reason the
percent “effectiveness of treatment” (difference
between percent incidences in control and treated
groups divided by percent incidence in control

Circulation, Volume XLVIII, September 1973
EFFECT OF ANTIHYPERTENSIVE AGENTS ON ECG 487

Table 8
Development of Major QRS Voltage Change with Respect to Prerandomization Blood Pressure

 

 

 

Control! group Treated group
No. showin No now
Subgroup randomized change % randomized ‘change % Effectiveness*
I. Normal initial voltage
becoming abnormal.
Prerandomization BP
Systolic < 165 mm Hg 63 17 27 57 5 9 67
Systolic 165+ mm Hg 49 11 22 34 1 3 86
Total 112 28 25 91 6 7 72
Diastolic 90-104 mm Hg 51 13 25 42 4 10 60
Diastolic 105-114 mm Hg 61 15 25 49 2 4 84
Total 112 28 25 91 6 7 72
II. Abnormal initial voltage
reverting to normal.
Prerandomization BP
Systolic <165 mm Hg 9 3 33 21 12 57 42
Systolic 165+ mm Hg 16 3 19 22 2 91 79
Total 25 6 24 43 32 74 68
Diastolic 90-104 mm Hg 10 1 10 16 14 88 89
Diastolic 105-114 mm Hg 15 5 33 27 18 67 51
Total 25 6 24 43 32 74 68

 

*% Effectiveness
Incidence % control — % treated

% control
Reversion % treated — % control

% treated

 

for subgroup I

 

for subgroup II

group) was greater in the patients with the higher
levels of blood pressure.

The situation was reversed with respect to ST-
segment depression. In the prevention of this
abnormality treatment was 100% effective in the
patients with systolic levels below 165 mm Hg and
69% effective in those with systolic blood pressures
above this level (table 9); and was 91% effective in
the patients with diastolic levels below 105 mm Hg
as compared to 69% in the 105-114 diastolic
subgroup. Similar differences were observed with
regard to T wave flattening or inversion. However,
these differences in the effectiveness of treatment at
different levels of blood pressure between voltage
changes on the one hand and ST-T changes on the
other may have been due to chance. Nonrepresenta-
tive trends can occur as a result of subdividing the
patients into such small sample sizes.

As would be expected, a greater percentage of
patients with average blood pressures above 164
systolic or 104 diastolic exhibited abnormal voltage
prior to randomization than did those with initial
blood pressures below these levels (table 8).
However, no consistent pattern of change with
relation to blood pressure was evident in the
postrandomization period. Treatment was effective

Circulation, Volume XLVI, September 1973

in both groups. The same was true with respect to
reversion of ST-segment depression to normal
(table 9). Treatment had approximately the same
effectiveness irrespective of the level of initial blood
pressure.

Twenty-five patients in the contro! group exhibit-
ed abnormal voltage prior to randomization. Six of
these patients reverted to normal during the
postrandomization period (table 8). This im-
provement was unrelated to changes in blood
pressure. For example, at one year following
randomization the average blood pressure of the six
patients who reverted was 7/0 mm Hg higher than
their average blood pressure prior to randomization.
This was not significantly different from the blood
pressure change observed in the control patients
who did not revert to normal. The latter exhibited
an increase of 4/3 mm Hg at one year after
randomization. The reversion to normal seen in six
of the control patients probably reflects sponta-
neous variation that may occur over time with
respect to electrocardiographic changes.** How-
ever, its occurrence in the control group was
considerably lower (P< .005) than the 74% rever-
sion to normal observed in the treated patients
(table 8).
 

 

 

488 POBLETE, KYLE, PIPBERGER, FREIS
Table 9
Development of ST-Segment Change with Respect to Prerandomization Blood Pressure
Control group Treated group
No. No.
No. showing No. showing %
Subgroup randomized change % randomized change % Effectiveness*
I. Normal base line ST-segment
becoming abnormal.
Prerandomization BP
Systolic < 165 mm Hg 70 8 11 64 0 0 100
Systolic 165+ mm Hg 44 14 32 41 4 10 69
Total 114 22 19 105 4 4 79
Diastolic 90-104 mm Hg 53 12 23 48 1 2 91
Diastolic 105-114 mm Hg 61 10 16 57 3 5 69
Total 114 22 19 105 4 4 79
II. Abnormal base line ST-
segment reverting to normal.
Prerandomization BP
Systolic < 165 mm Hg 5 1 20 12 5 42 52
Systolic 165+ mm Hg 13 3 23 9 7 78 71
Total 18 4 22 21 12 57 61
Diastolic 90-104 mm Hg 8 2 25 7 4 57 56
Diastolic 105-114 mm Hg 10 2 20 14 8 57 65
Total 18 4 22 21 12 57 61

 

*Same notations as in table 8.

The relation between blood pressure and voltage
changes was also examined in the treated group of
patients. In the treated patients who exhibited
reversion to normal voltage the average blood
pressure at one year postrandomization was 31/19
mm Hg lower than during the prerandomization
period. In the patients with abnormal voltage which
did not revert, the one year average blood pressure
was 21/16 mm Hg lower than the prerandomization
blood pressure. This difference in the extent of
average blood pressure reduction between the two
groups was not significant.

Patients aged 50 and above responded to
treatment as well as or better than the patients

below 50 years of age. The percent effectiveness of
treatment in the prevention of abnormal QRS
voltage was 68 for the patients below age 50 and 77
for the older patients (table 10). Reversion of
abnormal voltage to normal also was greater in the
older subjects, the percent effectiveness of
treatment being 79 in the patients age 50 and above
as compared to 56 for those below 50 years of
age.

Discussion
The present data indicate that antihypertensive
therapy not only protects against the development
of increased QRS voltage, ST-segment depression

 

 

 

Table 10
Development of QRS Voltage Change with Respect to Age
Control group Treated group
No. No.
No. showing No. showing %
randomized change % randomized change % Effectiveness™
I. Normal initial QRS
voltage becoming abnormal
Age < 50 yrs 54 15 28 47 4 9 68
Age 50 + yrs 58 13 22 44 2 5 77
Total 112 28 25 91 6 7 72
II. Abnormal initial QRS
voltage reverting to normal
Age < 50 yrs 15 4 27 26 16 62 56
Age 50 + yrs 10 2 20 17 16 94 79
Total 25 6 24 43 32 74 68

 

*Same notations as in table 8.

Circulation, Volume XLVIN, September 1973
EFFECT OF ANTIHYPERTENSIVE AGENTS ON ECG 489

and T wave flattening or inversion in hypertensive
patients but also causes a significant reversal of
these abnormalities when they are present prior to
treatment. These manifestations of left ventricular
hypertrophy are specifically although not exclusive-
ly related to hypertension and are probably the
direct consequence of a chronically increased
afterload.

In contrast to the marked changes associated
with treatment, particularly on QRS voltage and
ST-segment depression, there were no statistically
significant changes in Q or Q-S patterns or in
conduction defects between the placebo and
treatment groups. However, the low incidence of
these latter events in both the control and treated
groups did not permit an accurate assessment of the
effectiveness of treatment. Although as shown in
table 6 treatment appeared to lower the incidence
of Q or Q-S patterns and of A-V conduction defects,
the total incidence of these events was too low to
achieve statistical significance.

The various electrocardiographic changes ob-
served during the study are consistent with other
data from the trial relating to morbid events.* Thus,
treatment greatly reduced the incidence of “hyper-
tensive” complications such as hemorrhagic stroke,
congestive heart failure, progressive renal damage
and accelerated hypertension. However, treatment
did not appear to favorably influence the athero-
sclerotic complication of myocardial infarction. This
negative result with respect to atherosclerosis may
only mean that treatment must be initiated at an
earlier stage in the hypertensive process.

The mechanism of the ST-T alterations in left
ventricular hypertrophy is uncertain. Traditionally,
these changes are considered to reflect a severe load
on the left ventricle or “left ventricular strain”
causing increased myocardial oxygen demand. The
ST-T changes could, therefore, reflect chronic
ischemia which may be prevented or reversed by
reducing the blood pressure. It has also been
suggested that the ST-T changes are secondary to
higher voltages in these patients. However, some
studies have shown that the ST-T changes are
usually greater than would be expected as a
consequence of the magnitude of the QRS voltages,
suggesting that ST-T changes in patients with left
ventricular hypertrophy may represent a primary
disturbance in the recovery process of the left
ventricular myocardium.

In a recent report Hamer, Shinebourne and
Fleming" correlated hemodynamic findings with
electrocardiographic abnormalities in 17 hyperten-

Circulation, Volume XLVUI, September 1973

sive patients. On the basis of the electrocardiogram
they were divided into two groups. Nine (Group 1)
had either a normal tracing or only QRS voltage
changes of LVH with normal T waves and ST-
segments. Group 2 (8 patients) had, in addition, ST
depression and/or T wave inversion in the left chest
leads. The results indicated no significant difference
in the two groups with respect to cardiac output,
peripheral resistance, and diastolic blood pressure.
However, the systolic blood pressure was signifi-
cantly higher in Group 2 as compared to Group 1.
The present data (table 9) also indicate that the
control patients with systolic blood pressure levels
of 165 mm Hg and above developed ST-segment
depression three times more often than the control
patients with lower systolic levels. This difference
was significant at the P < .025 level.

Hamer postulated that because of the higher
systolic blood pressure seen in Group 2 patients the
myocardium must generate a greater force. This
leads to a primary change in the ST-segments and T
waves by producing localized subendocardial
changes which may alter the direction of repolariza-
tion or interfere with the spread of the activation
process. However, this author also pointed out that
the ST-T changes must represent more than a
mechanical effect, as they do not revert to normal
immediately when the left ventricular pressure is
reduced. However, as indicated in the present
study, many do revert with treatment after a period
of time. Thus, the pathogenesis of the ST-T changes
is still not completely clarified.

The present results, with respect to the effect of
antihypertensive drugs on the electrocardiogram,
provide further support for the value of treating
patients with essential hypertension. It should be
emphasized, however, that the blood pressure was
reduced to or near normotensive levels in the great
majority of the treated group. Less effective
reduction of blood pressure may not have resulted
in such a salutary effect.

Acknowledgment

The clerical assistance of Bert Silverman is gratefully
acknowledged.

References

1. Bripces WC, Jounson AJ, SmiTHwick RH, WuiITE
PD: Electrocardiography in hypertension; Study of
patients subjected to lumbodorsal splanchnicectomy.
JAMA 131: 1476, 1946

2, Leisoman AWD: The electrocardiogram in hyperten-
sion. Q J Med 20: 1, 1951
490

3.

Hetmcxe JG, ScunecxiorH R, Corcoran AC:
Electrocardiographic changes of left ventricular
hypertrophy: Effects of antihypertensive treatment.
Amer Heart J 53: 549, 1957

Dern PL, Pryor R, WALKER SH, SEaRLs DT: Serial
electrocardiographic changes in treated hypertensive
patients with reference to voltage criteria, mean QRS
vectors, and the QRS-T angle. Circulation 36: 823,
1967

Grorce FC, BreckennipceE AM, Dotiery CT: Value
of routine electrocardiography in hypertensive pa-
tients. Brit Heart J 34: 618, 1972

Wi.LeMs JL, PopLere PF, Presercer HV: Day-to-day
variation of the normal orthogonal electrocardiogram
and vectorcardiogram. Circulation 45: 1057, 1972

Srmonson E: Electrocardiogram in coronary heart
disease. Minn Med 38: 871, 1955

VETERANS ADMINISTRATION COOPERATIVE STUDY
Group ON ANTIHYPERTENSIVE AGENTS: Effects of
treatment on morbidity in hypertension: II. Results in
patients with diastolic blood pressure averaging 90
through 114 mm Hg. JAMA 213: 1143, 1970

9.

10.

11.

12.

13.

14.

15.

POBLETE, KYLE, PIPBERGER, FREIS

VETERANS ADMINISTRATION COOPERATIVE STUDY
Grovr ON ANTIHYPERTENSIVE AGENTS: Effects of
treatment on morbidity in hypertension: Results in
patients with diastolic blood pressures averaging 115
through 129 mm Hg. JAMA 202: 1028, 1967

Rose GA, Biacxsurn H: Cardiovascular Survey
Methods. W.H.O. Monograph 56, Geneva, 1968

Biacxspurn H, Keys A, Srmonson E, RAvuTAHAR]JU P,
Punsar S: The electrocardiogram in population
studies. A classification system. Circulation 21: 1160,
1960

Master AM, RosENFIELD I: Two-step exercise test:
Current status after twenty-five years. Mod Conc
Cardiov Dis 36: 19, 1967

SoxoLow M, Lyon TP: The ventricular complex in left
ventricular hypertrophy as obtained by unipolar
precordial and limb leads. Amer Heart J 37: 161,
1949

Hamer J, SHINEBOURNE E, FLEMinG J: Significance of
electrocardiographic changes in hypertension. Brit
Med J 1: 79, 1969

S1EcEL S: Nonparametric Statistics for the Behavioral
Sciences. New York, McGraw-Hill Book Co., 1956

Circulation, Volume XLVI, September 1973