524
Cardiovascular Reactivity to the Cold Pressor
Test as a Predictor of Hypertension
Marilyn S. Menkesf, Karen A. Matthews, David S. Krantz, Ulf Lundberg,
Lucy A. Mead, Bahjat Qaqish, Kung-Yee Liang,
Caroline B. Thomas, and Thomas A. Pearson
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
Cardiovascular reactivity to stress is hypothesized to be a marker for subsequent neurogenic
cardiovascular disease, but few prospective studies of this hypothesis are available. We studied
910 white male medical students who had their blood pressure and pulse rate measured before
and during a cold pressor test in the years 1948-1964. Hypertensive status (requiring drug
treatment) was ascertained by annual questionnaires in the 20- to 36-year follow-up period. An
association was observed between maximum change in systolic blood pressure and later
hypertension, with a cumulative incidence of hypertension by age 44 of 6.7%, 3.0%, and 2.4%
for a change in systolic blood pressure in the upper, middle two, and lowest quartiles,
respectively (Kaplan-Meier, p<0.02). After adjustment for study entry age, Quetelet Index,
cigarette smoking, pretest systolic blood pressure, and paternal or maternal history of
hypertension in a Cox model, the association persisted. The excess risk associated with systolic
blood pressure reactivity was not apparent until the population aged some 20 years and was
most apparent among those in whom hypertension developed before age 45 (relative risk=2.5,
95% confidence intervals=1.47, 4.71 for a 20 mm Hg change). Diastolic blood pressure and
heart rate changes were not associated with later hypertension. These data suggest that persons
prone to later hypertension manifest an altered physiology at a young age. (Hypertension
1989;14:524-530)
C
ardiovascular reactivity to stress has been
hypothesized to be a marker for subsequent neurogenic hypertension.1 However,
few studies have tested the link between responses
to psychophysiological stimuli and later development of this disease in human populations. As one
approach to evaluation of this hypothesis, studies
have assessed the discriminative or prognostic value
of "excessive" reactivity to a standardized cold
stimulus, the cold pressor test. The empirical basis
for the cold pressor test rests on observations by its
From The Johns Hopkins Medical Institutions (M.S.M.,
L.A.M., B.Q., K.L., C.B.T.), Baltimore, Maryland, the University of Pittsburgh (K.A.M.), Pittsburgh, Pennsylvania, the Uniformed Services University of the Health Sciences (D.S.K.),
Bethesda, Maryland, the Karolinska Institutet (U.L.), Stockholm, Sweden, and The Mary Imogene Bassett Hospital (T.A.P.),
Cooperstown, New York.
tDeceased.
Supported in part by The John D. and Catherine T. MacArthur
Foundation and a Preventive Cardiology Academic Award (HL01113) from the National Heart, Lung, and Blood Institute.
Address for reprints: Karen A. Matthews, PhD, Department
of Psychiatry, University of Pittsburgh School of Medicine, 3811
O'Hara Street, Pittsburgh, PA 15213.
Received February 21, 1989; accepted in revised form July 3,
1989.
originators, Hines and Brown,2-4 and others5"12
who found: 1) that hypertensive persons show
greater lability of blood pressure under various
forms of stress than do normotensive persons,4'7 2)
that normotensive "hyper-reactors" to the cold
pressor test are more likely to have a positive family
history of hypertension than normotensive persons
who are less reactive,3-5-10 and 3) that hyperreactors to the cold pressor test may be predisposed
to the development of essential hypertension at a
later point in time.3-5-6
However, other studies have been unable to
confirm many of these observations,13-20 and few
long-term prospective studies of this hypothesis are
available. Previous prospective studies have been
limited by incomplete or short duration of follow-up
or by the small size of the cohorts initially tested for
reactivity to the cold pressor test.6-17-20 In all of
these studies, the analyses were based on arbitrary
definitions of reactivity introduced by Hines or
Brown. None took into consideration the impact of
time or age in the estimation of risk of development
of hypertension, and few controlled for confounding factors.
We report here on the results of a prospective
survival analysis of the relation of cold pressor
Menkes et al Cardiovascular Reactivity and Hypertension
reactivity to subsequent hypertension among a large
cohort of medical students who have been followed
since graduation for some 20 to 36 years; we took
into account the effects of age and other important
risk factors for hypertension.
Materials and Methods
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
Study Population and Baseline Measurements
In 1946-1947, a longitudinal investigation on risk
factors in early adulthood for later cardiovascular
and other chronic diseases was initiated by one of
the authors (C.B.T.). The study population consisted of 1,130 white male medical students enrolled
in the 1948 to 1964 graduating classes of the Johns
Hopkins Medical School. More than 95% of eligible
participants were enrolled. The age range of the
total cohort at the initial examination was 19-49
years (mean, 23 years). Complete data for the cold
pressor test were obtained for 910 of the 1,130 white
male study subjects. Of the 220 nonparticipants, 50
had not participated in the overall study, and the
remainder failed to keep their appointment for the
cold pressor test.
Participants received a standardized medical examination that included weight and height measurements in the first medical school year, and before
graduation, they also completed questionnaires about
parental history of chronic disease and habits of
daily living, including number of cigarettes smoked
per day. No formal informed consent was obtained
at the time of baseline testing as this study was
initiated (1948-1964) before the development of
professional standards for obtainment of formal
informed consent. During their first year examination, the participants underwent the cold pressor
test. This test was performed with the subject
recumbent, with use of the technique of Hines and
Brown,2 and is described in more detail in a previous report on this study population.16 Briefly, blood
pressure and pulse rate were measured by a mercury sphygmomanometer by a single trained
observer. The blood pressure and pulse rate measurements were repeated until no further changes
were noted (+4 mm Hg and 4 beats/minute), a
common procedure in psychophysiological research
to ensure that a reliable baseline value has been
achieved. These control values were subsequently
used for the calculation of the cold pressor reactivity measures. The right hand was immersed above
the wrist for 1 minute in a bucket of ice water at
4° C, during which time the blood pressure and
pulse rate were measured in the opposite arm at
approximately 30 and 60 seconds and recorded.
Three measures of cold pressor reactivity were
calculated: the maximum changes in systolic blood
pressure, diastolic blood pressure, and heart rate
from the control values during the cold stimulus.
Although reliability data are not available from the
present study, three other investigations have reported
test-re test correlations for the cold pressor test. They
525
range from 0.39 to 0.83 for systolic blood pressure
changes, 0.23 to 0.75 for diastolic blood pressure
changes, and 0.23 to 0.50 for heart rate changes across
intervals of time from 2 weeks to 4 years.21-23 Among
the samples most comparable to that of first year
Johns Hopkins Medical School students (male and
female undergraduates), the test-retest reliability of
the cold pressor test was 0.83 for systolic blood
pressure changes, 0.23 for diastolic blood pressure
changes, and 0.50 for heart rate change.21
Follow-up Procedures
Since graduation, data on parental history were
updated by questionnaires that were mailed to study
participants each year. This follow-up information
was used in the analysis to define the number of
parents in whom the participants reported the development of hypertension before the age of 65, the
age of the youngest parents of participants, thus
allowing all parents to be at risk of becoming
hypertensive. Participation during the long followup period has remained reasonably high, with
response rates ranging from 68 to 78% for the return
of any one set of questionnaires.
The incidence of disease events and mortality in
this physician cohort were surveyed by questionnaire on an annual basis. For any 5-year period,
87-94% of any single class returned their morbidity
questionnaires. Subjects who reported receiving
drug therapy for hypertension were considered to
be hypertensive for the purposes of this study.
Other studies of the epidemiology of hypertension
have used self-reports of hypertension as the outcome variable and have found the hypertension
questions in the mail survey to have good sensitivity (82-98%) and specificity (98-99%).24>25 In addition, a small pilot study of 21 physicians enrolled in
the present study showed that self-reports of systolic blood pressure, diastolic blood pressure, pulse
rate, and weight were similar to (and not significantly different from) those measured by trained
observers. By 1984, after 20-36 years of follow-up,
113 study subjects reported receiving treatment for
hypertension; 105 of these had taken the cold pressor test. Thus, the statistical analysis was based on
data from 910 study subjects, 105 of whom reported
they were being treated for hypertension.
Statistical Analysis
The frequency distributions of the three cold
pressor reactivity measures for the total population—
maximum change in systolic and diastolic blood
pressures and heart rate—were examined. The relation between reactivity to cold pressor and the
incidence of hypertension during 20-36 years of
follow-up was evaluated by an examination of
Kaplan-Meier survival curves,26 with age to hypertension as the outcome. For this analysis, the
reactivity variables were arbitrarily divided into
three categories: the lower 25%, the middle 50%,
and the upper 25% of the study population distribu-
526
Hypertension
Vol 14, No 5, November 1989
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
tions. The middle two quartiles did not differ in risk
of hypertension. The log rank test was used to
assess the statistical significance of the difference
between the three curves. 2 7
The three cold pressor reactivity measures were
also treated as continuous variables and were entered
into separate Cox proportional hazards models 28 to
obtain unadjusted estimates of association with the
development of hypertension. Estimates of relative
risk with corresponding 95% confidence limits were
obtained with the Cox model; these estimates were
adjusted for baseline values of age, Quetelet Index
({weight [kg]}/{height [m]}2), control systolic blood
pressure, cigarette smoking, and the updated histories of maternal and paternal history of hypertension. Evidence for interaction between the cold
pressor reactivity measures and family history of
hypertension was sought with this model because
family history of hypertension has been related to
cold pressor reactivity in some studies.
Finally, since there was some suggestion from
our data that cold pressor reactivity may be predictive of early rather than late onset of hypertension, a separate Cox model was developed to test
this hypothesis. The model incorporated a timedependent function, which had a value of zero or
one, respectively, if hypertension developed in a
subject before the age of 45 or at 45 years or later.
By using this model, separate estimates of association between cold pressor reactivity and subsequent " e a r l y " or " l a t e " hypertension w e r e
obtained. The model used in this analysis is
described by the following equation:
A[t; x,, x 2 , z(t)] = A0(t) • e " ^ 1 ^ * * '
where
A0(t)
is
the baseline hazard function;
Xj
is
cold pressor maximum systolic blood
pressure change;
x2
is
vector of the other covariates (baseline age, Quetelet Index, cigarette
smoking, control systolic blood pressure, maternal and paternal history
of hypertension);
z(t)
is
time-dependent function; value=0 or
1 when age is less than 45 or age is 45
or more years, respectively;
^
is
log relative risk of cold pressor reactivity for early hypertension;
/3 2
is
difference of t h e log relative risks of
§3
cold pressor reactivity for early and
late hypertension;
is vector of log relative risks for other
covariates.
Results
Baseline characteristics of the study population
undergoing the cold pressor test are presented in
TABLE 1. Baseline Characteristics of the Cardiovascular Reactivity Study Population
Subjects (N)
Mean age (±SD)
Mean quetelet (±SD)
Smoking history % (N)
Smokers
Ex-smokers
Parental history of hypertension % (N)*
Maternal+
Paternal +
Both+
Mean control blood pressure and pulse (±SD)
Systolic (mm Hg)
Diastolic (mm Hg)
Pulse (beats/min)
Mean cold pressor change in blood pressure and
pulse (±SD)
Systolic (mm Hg)
Diastolic (mm Hg)
Pulse (beats/min)
910
23.1 (2.5)
22.9 (2.6)
53.3 (486)
2.3 (21)
17.3 (158)
14.6 (133)
4.0 (36)
114.0 (10.4)
69.4 (8.0)
72.3 (9.9)
12.2 (8.7)
15.0 (9.1)
6.2 (8.7)
"Variable based on follow-up information to determine hypertension status of parents before age 65.
Table 1. The 220 subjects who did not complete the
test had characteristics similar to the participants
(p>0.10): mean age (23.4±2.7 years), mean Quetelet Index (23.0±2.5), smokers (59.8%), ex-smokers
(1.1%), positive maternal or paternal history of
hypertension (30.1%). About half of the study group
smoked cigarettes and over 30% had at least one
parent in whom hypertension developed before the
age of 65.
The distribution of the three cold pressor reactivity measures approximated a normal curve with
a mild skew toward higher values. The range of
values of the three measures of reactivity were:
- 8 and 76, - 6 and 50, -22 and 58 for the maximum
change in systolic pressure (mm Hg), diastolic
pressure (mm Hg), and heart rate (beats/min),
respectively.
When age at follow-up was considered in the
analysis, an association with incidence of hypertension was observed for one cold pressor reactivity
measure: the maximum change in systolic blood
pressure. Systolic blood pressure change was unrelated to control systolic blood pressure (r=-0.05,
p=0.09). The cumulative incidence of hypertension
by age, according to three levels of this reactivity
measure (upper, middle two, and lower quartiles)
are illustrated in Figure 1 and presented in more
detail in Table 2. No association of hypertension
with diastolic blood pressure or heart rate change
was seen. The curves demonstrated that the cold
pressor test does not distinguish between subjects
who are at increased risk of development of hypertension until the population is allowed to age some
20 years or more. Table 2 illustrates that by using
those with low cold pressor reactivity as a compar-
Menkes et al
Cardiovascular Reactivity and Hypertension
527
c
u
m
0.30
n
0.2S
c
d 0.20
H 0.15
y
J o.io J
t aos
e
a ao°
I
o
n
20
25
30
35
40
45
50
55
60
Age (years)
Low
Medium—
High
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
ison group, the risk of hypertension for subjects
who were age 44 and had high pressor reactivity to
cold (2.8) is greater than the risk for subjects who
were age 54 and also in the high reactivity category
(1.7).
Cold pressor systolic blood pressure change was
also observed to be significantly associated with
risk of hypertension when it was treated as a
continuous variable in a univariate Cox model
(B=0.027, SEM=0.011,/?=0.01), and in multivariate analysis after adjustment for other important
risk factors. Table 3 shows the regression coefficients and corresponding risk estimates for cold
pressor reactivity and other major risk factors that
were included in the multivariate model. The relative risk of hypertension for a cold pressor systolic
blood pressure change of 20 mm Hg is 1.93 after
adjustment for baseline age, Quetelet Index, pretest
control systolic blood pressure, cigarette smoking,
and maternal and paternal history of hypertension.
A significant interaction was observed between
maternal and paternal history in these data (relative
risk of 8.12 for hypertension if both parents had
hypertension). For this reason, a term was included
in the model to take it into account. Cold pressor
reactivity was not observed to be a stronger predictor in those with a positive maternal or paternal
FIGURE 1. Cumulative incidence of hypertension over 1530 years of follow-up according to level of systolic blood
pressure change during cold
pressor test at baseline. Three
groups are lowest quartile,
(dots), middle two quartiles
(solid line), and highest quartile (dashed line) according to
baseline systolic blood pressure change. Incidence of
hypertension in highest quartile is significantly different
from other two
groups
(p=0.021).
history of hypertension than in those without such
history. Note that having two parents with hypertension is a stronger predictor of subsequent hypertension than having either parent with hypertension
and that reactivity to the cold pressor test has a
relative risk similar to that of control systolic blood
pressure. Any comparison of relative risks between
continuous and categorical variables (e.g., family
history of hypertension vs. reactivity to the cold
pressor test) is somewhat flawed because it depends
on the arbitrary categorization of the continuous
variables.
In addition, we examined the relation of cold
pressor reactivity to incidence of hypertension,
using Hines' definition3 of a hyper-reactor to the
cold pressor test (a systolic blood pressure change
of 20 mm Hg or greater) in a multivariate model that
included the adjustment variables used previously.
The association remained (relative risk=1.76; 95%
confidence interval=1.12, 2.76), although it was
somewhat weaker than that observed when reactivity was treated as a continuous variable.
We examined the possibility that the observed
relation between cold pressor reactivity and incidence of hypertension differed for early (before 45
years) versus late (45 years or older) age of onset
with a time-dependent multivariate survival model,
TABLE 2. Cumulative Incidence of Hypertension at Age of Follow-up According to Level of Systolic Blood Pressure Reactivity to Cold
Pressor Test
Subjects at risk
Events of
hypertension (n)
257
424
229
910
28
42
35
105
Systolic blood pressure change to cold pressor test
Low (<6 mm Hg)
Moderate (7-17 mm Hg)
High(>18mmHg)
Total
Subjects (n) contributing person years at each age cutoff.
Cumulative incidence
(% of subjects) of
hypertension
Age at follow-up
44 years
54 years
2.4
3.0
6.7
9.2
9.5
15.4
815
346
528
Hypertension
Vol 14, No 5, November 1989
TABLE 3. Regression Coefficients and Relative Risks Relating Cold Pressor Systolic Blood Pressure and Other Risk
Factors to Incidence of Hypertension Over 20-36 Years of Follow-up
Variable
Cigarette smoking
Maternal
hypertension
Paternal hypertension
Both parents with
hypertension
Control systolic blood
pressure (mm Hg)t
Cold pressor systolic
blood pressure
change (mm Hg)t
95% confidence
interval
(1.15, 2.60)
0.547
0.209
Relative
risk*
1.73
0.097
0.758
0.280
0.256
1.10
2.13
(0.64, 1.91)
(1.29, 3.52)
1.239
0.438
8.12
(4.47, 14.10)
0.036
0.009
2.04
(1.43, 2.90)
0.033
0.011
1.93
(1.26, 2.98)
SE
'Regression coefficients and relative risks estimated from a Cox Proportional Hazards Model that also includes
baseline age (p>0.05) and Quetelet Index (p>0.05).
tRelative risks estimated for an increase in systolic blood pressure of 20 mm Hg.
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
the results of which are shown in Table 4. For the 33
subjects who reported that hypertension had developed before age 45 years, the relative risk for a 20
mm Hg change in systolic pressure during the cold
pressor test was 2.63, a strong and statistically
significant association. In contrast, for the 72 subjects who reported the development of hypertension at age 45 years or later, the relative risk
associated with a similar systolic change was 1.53,
with a 95% confidence interval that included one.
Discussion
A significant independent association between
blood pressure reactivity to the cold pressor test
and incidence of subsequent hypertension was
observed in this study. This is in agreement with the
results of two previous prospective studies. Wood
et al5 followed 47% of a study cohort of 300 school
children for 45 years and found a 3.7-fold risk of
hypertension for those classified as hyper-reactors
to the cold pressor test at either of two time periods.
Barnett and associates6 followed a group of Hines'
original cohort for 27 years and found a 10% (4/40)
incidence of hypertension among patients classified
as hyper-reactors at the time of the initial cold
pressor test, but no hypertension among the patients
originally classified as normoreactors. However,
TABLE 4. Relative Risks Relating Cold Pressor Systolic Blood
Pressure Reactivity to Incidence of Subsequent Hypertension by
Age at Follow-up
Age at
follow-up
Number of events
of hypertension
Relative risk of subsequent
hypertension (95%
confidence interval)'
33
2.63 (1.47, 4.71)
<45 years
72
1.53 (0.86, 2.71)
>45 years
Relative risks estimated from a Cox Proportional Hazards
Model with adjustments for baseline age, Quetelet Index, cigarette smoking, maternal and paternal history of hypertension,
and control systolic blood pressure.
•Relative risks estimated for an increase in systolic blood
pressure of 20 mm Hg.
the low rate of follow-up and the lack of a uniform
definition of test positives in the former study have
been sharply criticized,29 and the low incidence of
hypertension among high reactors provides rather
weak support for the clinical usefulness of the cold
pressor test.
Four other prospective studies have not confirmed a positive association between cold pressor
reactivity and later hypertension.17-20 One study
was an 18-year follow-up study of a cohort of
aviators,17 a group that may have been selected to
be healthy and physically fit, including low, nonlabile blood pressures and heart rates. By the end of
the 18-year follow-up period, the mean age of the
cohort was only about 41 years, an age that may
have been too young to uncover many hypertensive
cases. A second study of air force officers18 may
have had similar selection biases and reported no
correlation between cold pressor response and hypertension status after a follow-up period of only 7
years. Similarly, the remaining two studies, of
university19 and medical student20 populations, determined the hypertensive status of their study subjects after only 5-10 years of follow-up, before most
of them had reached their fourth decade of life.
Since the prevalence of hypertension increases dramatically with age, our data suggest that assessment
of this end point should not be made until the
majority of the population reaches the age of 45 or
50 years if a fair test of the predictive value of the
cold pressor test is to be made.
In addition to having followed a large proportion
of the study cohort for a time span of up to three
and a half decades, the present investigation differs
from those previous to it with respect to the statistical methods of analyses that were used. Past
investigations have examined cold pressor reactivity in relation to incidence of hypertension without
taking the dimension of time or age into account in
their analyses.5'6-16-20 When this analytic approach
is applied to our own data, our study also shows no
Menkes et al Cardiovascular Reactivity and Hypertension
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
independent association between cold pressor blood
pressure reactivity and hypertension (/?>0.10).16
There are two points to be made in favor of the
survival analyses we chose to use in our study.
First, they are a preferred method of analysis for
determination of the relation of prognostic factors
to outcome over time in longitudinal studies.30
Secondly, since our data suggest that the importance of the cold pressor test as a predictor of
hypertension varies with age, a consideration of
time or age is crucial in the analysis.
Our treatment of the indexes of blood pressure
reactivity to the cold pressor test as continuous
rather than categorical variables, unlike other investigators, increased the statistical power of our data
to detect an association with incidence of hypertension. However, this difference in treatment of the
exposure variable does not, by itself, explain the
difference in results obtained between studies, since
we also observed an independent association
between the maximum cold pressor blood pressure
change (or increment) when this variable was categorized according to similar criteria used by other
investigators and analyzed with the Cox survival
models. It is also important to note that the association with blood pressure change was independent
of the basal level of blood pressure, since these data
were adjusted for basal levels as well as other
potential confounders in the models.
Although we did find a relation between systolic
blood pressure response to cold stress and later
hypertension, we did not observe a similar independent association for diastolic blood pressure or
heart rate. These results appear to be in agreement
with several studies,18-20 although they disagree
with some of Hines' work, who in later years based
his definition of hyper-reactivity solely on diastolic
pressor responses.31 Perhaps the nonsignificant
results for diastolic blood pressure in the present
study during the cold pressor result from the low
reliability relative to that of systolic blood pressure
in this age group.21-23
It is also important to note here that the cold
pressor test elicits different hemodynamic mechanisms from stressors with a more substantial mental
component. Results from two studies that measured
blood catecholamine levels in response to two stress
tests, the cold pressor test, and mental arithmetic
test32-33 showed that the mental arithmetic test
elicited higher heart rate responses and levels of
plasma epinephrine than did the cold pressor test,
whereas the cold pressor test was associated with
higher levels of plasma norepinephrine and elevations in blood pressure comparable with those
obtained during the mental arithmetic test. Norepinephrine levels were found to be more strongly
correlated with blood pressure levels than with
heart rates. Similarly, several investigators34-37
suggest that persons with a family history of
hypertension in fact possess a heightened cardiovascular response potential expressed during the
529
performance of tests that require "active coping"
in the mental or psychosocial sense. Such tasks
include the mental arithmetic test, shock avoidance, or other situational demands that trigger
affective behaviors.34-38 Although some affective
response is elicited by the cold pressor test, it is
primarily a physical stressor that passively involves
the participant.
The present study does not and cannot elucidate
why episodic hyperresponsiveness is related to risk
for hypertension. Hyperresponsiveness may represent one pathogenic mechanism in the development
of essential hypertension, be a marker for a central
defect in the autonomic control of the cardiovascular system, or reflect early changes in arterial compliance of future hypertensive individuals.38-39 One
intriguing mechanism that has been proposed40-41 is
that pressor hyper-reactivity is a manifestation of a
widespread basic membrane transport disorder that
disrupts cellular cation homeostasis. Sodiumlithium transport abnormalities have been observed
in blood cell membranes of patients with essential
hypertension,42-43 and a recent study44 has found
that the cold pressor test elicited a greater sodium
efflux in leukocytes of subjects with a positive
family history of hypertension compared with those
with a negative history.
Although the present study suggests that a state
of hyperresponsiveness may precede essential hypertension and that the cold pressor test could be
useful as a predictor of future hypertension in a
young study population, the findings must be considered to be preliminary until they are replicated
by other large, long-term prospective studies that
are able to achieve complete follow-up. Even then,
several factors would need to be considered, particularly if the test is ever to be used for screening
purposes. First, the cold pressor test may not be a
valid and reliable test in all populations. Second,
the cold pressor test is reasonably acceptable to
most study subjects and simple to administer. A
final consideration is the relative strength of the
association between cold pressor reactivity and
incidence of hypertension. Although having two
parents with hypertension is a stronger prognostic
factor in this cohort and other cohort studies,6 the
risk of hypertension associated with a 20 mm Hg
increase in cold pressor systolic response is
increased by over 90% and closely approximates
the increased risk observed for a similar increase in
basal systolic blood pressure.
In sum, the present study suggests that systolic
blood pressure reactivity to the cold pressor test may
prove to be an important independent predictor of
early hypertension. The cold pressor tests may thus
identify a subgroup of individuals with an occult
physiological abnormality that predisposes them to
hypertension decades later. Preventive interventions
may be particularly warranted in these individuals.
530
Hypertension
Vol 14, No 5, November 1989
Acknowledgments
We gratefully acknowledge the technical assistance of Sally Weida, RN, Susan Hardesty, Evelyn
Severn, and Betty Ann Remley and the helpful
comments on an earlier version of this manuscript
of Dr. Michael Klag, Director of the Precursors
Study. This article is dedicated to Marilyn Menkes,
who is deeply missed by her coworkers.
References
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
1. Krantz DS, Manuck SB: Acute psychophysiologic reactivity
and risk of cardiovascular disease: A review and methodologic critique. Psychol Bull 1984;96:435-464
2. Hines EA Jr, Brown GE: The cold pressor test for measuring
the reactibility of the blood pressure: Data concerning 571
normal and hypertensive subjects. Am Heart J 1936; 11:1-9
3. Hines EA Jr: The significance of vascular hyperreaction as
measured by the cold-pressor test. Am Heart J 1940;
19:408-416
4. Hines EA Jr: Range of normal blood pressure and subsequent development of hypertension. JAMA 1940;115:271-274
5. Wood DL, Sheps SG, Eleback LR, Schirger A: Cold pressor
test as a predictor of hypertension. Hypertension 1984;
6:301-306
6. Barnett PH, Hines EA, Schirger A, Gage RP: Blood pressure and vascular reactivity to the cold pressor test. JAMA
1963;183:845-848
7. Levy RL, Hillman CC, Stroud WD, White PD: Transient
hypertension: Its significance in terms of later development
of sustained hypertension and cardiovascular-renal disease.
JAMA 1944;126:829-833
8. Steptoe A, Melville D, Ross A: Behavioral response demands,
cardiovascular reactivity, and essential hypertension. Psychosom Med 1984;46:33-48
9. Falkner B, Kushner H, Onesti G, Angelakos ET: Cardiovascular characteristics in adolescents who develop essential
hypertension. Hypertension 1981;3:521-527
10. Briggs JF, Oerting H: Vasomotor response of normal and
hypertensive individuals to thermal stimulus (cold). Minn
Med 1981;16:481-486
11. Cuddy RP, Smulyan H, Keighley JF, Markason CR, Eich
RH: Hemodynamic and catecholamine changes during a
standard cold pressor test. Am Heart J 1966;71:446-454
12. Ayman D, Goldshine D: Cold as a standard stimulus of blood
pressure: A study of normal and hypertensive subjects. N
EnglJ Med 1938;219:650-655
13. Russek HI: The significance of vascular hyperreaction as
measured by the cold pressor test. Am Heart J 1943;26:398-404
14. Drummond PD: Cardiovascular reactivity in mild hypertension. Ann Psychosom Res 1983;27:291-297
15. Eliasson K, Hjemdahl P, Kahan T: Circulatory and sympathoadrenal responses to stress in borderline and established
hypertension. J Hypertens 1983;1:131-139
16. Thomas CB, Duszynski KR: Blood pressure levels in young
adulthood as predictors of hypertension and the fate of the
cold pressor test. The Johns Hopkins Med /1982;151:93-100
17. Harlan WR Jr, Osborne RK, Graybiel A: Prognostic value of
the cold pressor test and the basal blood pressure based on an
eighteen year follow-up study. Am J Cardiol 1964;13:832-837
18. Armstrong HR, Rafferty JA: Cold pressor test follow-up
study for seven years on 166 officers. Am Heart J 1950;
39:484-490
19. Eich RH, Jacobsen EC: Vascular reactivity in medical
students followed for 10 years. J Chronic Dis 1967;20:583-592
20. Thacker EA: A comparative study of normal and abnormal
blood pressures among university students, including the
cold pressor test. Am Heart J 1940;20:89-97
21. Faulstich ME, Williamson DA, McKenzie SJ, Duchmann
EG, Hutchinson KM, Blovin DC: Temporal stability of
responding: A comparative analysis of mental and physical
stressors. Int J Neurosci 1986;30:65-72
22. McKinney ME, Miner MH, Ruddel H, Mcllvain HE, Witte
H, Buell JC, Eliot RS, Grant LB: The standardized mental
stress protocol: Test-retest reliability and comparison with
ambulatory blood pressure monitoring. Psychophysiology
1985;22:453-463
23. Lacey JI, Lacey BC: The law of initial values in the
longitudinal study autonomic constitution: Reproducibility
of autonomic responses and response patterns over a fouryear interval. Ann New YorkAcad Sci 1962;98:1257-1290
24. Blair SM, Goodyear NN, Gibbons LW, Cooper KH: Physical fitness and incidence of hypertension in healthy normotensive men and women. JAMA 1984;252:487-490
25. Paffenbarger RS, Wing AL, Hyde RT, Jung DL: Physical
activity and incidence of hypertension in college alumni. Am
J Epidemiol 1983;117:245-257
26. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-481
27. Peto R, Peto J: Asymptomatically efficient rank invariant
test procedures. J R Stat Soc Ser (A) 1972;135:185-198
28. Cox DR: Regression models and life-tables. / R Stat Soc Ser
(B) 1972;34:187-202
29. Horowitz RI: Methodologic standards and the clinical usefulness of the cold pressor test. Hypertension 1984;6:295-296
30. Brookmeyer R: Time and latency considerations in the
quantitative assessment of risk, in Gordis L (ed): Epidemiology and Health Risk Assessment. New York, Oxford
University Press, 1987
31. Windesheim JH, Roth GM, Hines EA Jr: Direct arterial
study of the blood pressure responses to cold of normotensive subjects and patients with essential hypertension before
and during treatment with various antihypertensive drugs.
Circulation 1955;ll:878-888
32. Andren L, Hannson L: Circulatory effects of stress in essential
hypertension. Ada Med Scand (suppl) 1981;646:69-72
33. LeBlanc J, Cote J, Jobin M, Labrie A: Plasma catecholamines and cardiovascular responses to cold and mental
activity. JAppl Physiol 1979;47:1207-1211
34. Hastrup JL, Light KC, Obrist PA: Parental hypertension
and cardiovascular response to stress in healthy young
adults. Psychophysiology 1982;19:615-622
35. Manuck SB, Prioietti JM, Rader SJ, Polefrone JM: Parental
hypertension, affect, and cardiovascular response to cognitive challenge. Psychosom Med 1985;47:189-200
36. Ditto B: Parental history of essential hypertension, active
coping, and cardiovascular reactivity. Psychophysiology 1986;
23:62-70
37. Drummond PD: Cardiovascular reactivity in borderline hypertensives during behavioral and orthostatic stress. Psychophysiology 1985;22:621-628
38. Fredrikson M, Matthews KA: Cardiovascular responses to
behavioral stress and hypertension: A meta-analytic review.
Ann Behav Med (in press)
39. Matthews KA, Weiss TM, Detre T, Dembroski TM, Falkner
B, Manuck SB, Williams RB (eds): Handbook of Stress,
Reactivity, and Cardiovascular Disease. New York, John
Wiley & Sons, Inc, 1986, pp 19-29, 137-138
40. Folkow BS: Physiological aspects of primary hypertension.
Physiological Reviews 1982;62:347-504
41. Weder AB, Julius S: Behavior, blood pressure variability,
and hypertension. Psychosom Med 1985;47:406-414
42. Canessa M, Adranga N, Solomon H, Connolly T, Tosteson
D: Increased sodium-lithium countertransport in red cells of
patients with essential hypertension. N Engl J Med 1980;
302:772-776
43. Edmondson R, Thomas R, Hilton R, Patrick J, Jones N:
Abnormal leukocyte composition and sodium transport in
essential hypertension. Lancet 1975;l:1003-1005
44. Riozzi A, Heagerty AM, Bing RF, Thurston H, Swales JD:
Effect of cold pressor test-induced stress on leukocyte
sodium transport and norepinephrine. Hypertension 1987;
9:13-17
KEY WORDS • neurogenic hypertension
cold pressor test • physical stress
• epidemiology •
Cardiovascular reactivity to the cold pressor test as a predictor of hypertension.
M S Menkes, K A Matthews, D S Krantz, U Lundberg, L A Mead, B Qaqish, K Y Liang, C B
Thomas and T A Pearson
Hypertension. 1989;14:524-530
doi: 10.1161/01.HYP.14.5.524
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
Hypertension is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1989 American Heart Association, Inc. All rights reserved.
Print ISSN: 0194-911X. Online ISSN: 1524-4563
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://hyper.ahajournals.org/content/14/5/524
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Hypertension can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Hypertension is online at:
http://hyper.ahajournals.org//subscriptions/