Hypertension Risk and Caffeine's Effect on Cardiovascular Activity During Mental Stress in Young Men William R. Lovallo and Gwendolyn A. Pincomb Veterans Affairs Medical Center and Department of Psychiatry and Behavioral Sciences University of Oklahoma Health Sciences Center
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Bong Hee Sung Department of Medicine and College of Pharmacy University of Oklahoma Health Sciences Center
Susan A. Everson Veterans Affairs Medical Center and Department of Psychiatry and Behavioral Sciences University of Oklahoma Health Sciences Center
Richard B. Passey Department of Pathology University of Oklahoma Health Sciences Center
Michael F. Wilson Veterans Affairs Medical Center and Department of Medicine University of Oklahoma Health Sciences Center Examined the cardiovascular effects of caffeine plus behavioral stress in men low versus high in risk of essential hypertension. Caffeine (3.3 mg/kg, equivalent to 2 to 3 cups of coffee) or placebo was given on alternate days to 19 low-risk men (negative for parental hypertension and low-normal resting blood pressure, BP) and 20 high-risk men (positive history, high-normal BP). Forty minutes later, each worked for 15 min on a demanding psychomotor task during which BP, cardiac output, and vascular resistance were determined. During rest, caffeine raised vascular resistance in both groups. During the task, it supra-additively increased the systolic BP response by enhancing the rise in cardiac output, producing equivalent BP rises in both groups. Due to the higher resting pressures of the high-risk men, caffeine plus the task resulted in 50% of these having transient BP of 140/90 mg Hg or greater. Caffeine in combination with mental stress may produce undesirable BP in those at risk for hypertension. Key words: caffeine, hypertension, stress, cardiovascular, diet, impedance cardiography
The present study examined the cardiovascular effects of caffeine, during rest and work on a demanding psychomotor task, in men designated either low or high in risk for essential hypertension. Caffeine is a substance widely consumed in coffee, tea, soft drinks, and medications and has effects on the central nervous system and on cardiovascular and neuroendocrine functions. Therefore, caffeine may interact with processes occurring during states of stress and may do so to different degrees in different persons. Caffeine's best documented cardiovascular effect is its ability to increase blood pressure (BP) at rest (Greenberg & Requests for reprints should be sent to William R. Lovallo, Veterans Affairs Medical Center (151A), 921 Northeast 13th Street, Oklahoma City, OK 73104.
Shapiro, 1987; Whitsett, Christensen, & Hirsh, 1980), which occurs as a result of an increased systemic vascular resistance (Pincomb et al., 1985). Caffeine also elevates BP during conditions of behavioral stress (Greenberg & Shapiro, 1987; Lane & Williams, 1985). During a reaction-time task, caffeine elevated cardiac output and contractility, but vascular resistance was equivalent to that seen under placebo (Pincomb, Lovallo, Passey, & Wilson, 1988). Caffeine may thus influence BP by means of vascular or cardiac effects depending on the activational state of the individual. In a study of young men with a parental history of hypertension, caffeine has been shown to increase BP in an additive fashion with the application of mental stress (Greenberg & Shapiro, 1987). However, the hemodynamic factors underlying the BP response to caffeine plus stress have not been assessed in men at risk for hypertension. In the
CAFFEINE AND MENTAL STRESS
present study, we examined caffeine's effects on cardiovascular dynamics of high-risk and low-risk men at rest and during work on a challenging psychomotor task. We predicted that high-risk men would respond more to caffeine's vasoconstrictive influence both at rest and during the task.
METHOD
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Subjects The volunteers were 40 White men—20 designated high risk for essential hypertension and 20 designated low risk. Highrisk subjects had one or both parents with essential hypertension, as verified by physician report, and had two or more recordings of resting BP of 135 mm Hg systolic (SBP) and/or 85 mm Hg diastolic (DBP) to 154 mm Hg SBP and/or 94 mm Hg DBP, inclusive (greater than 90th/85th percentiles); lowrisk subjects had a negative history of hypertension in both parents, as verified by physician report and no screening BP above 132 mm Hg SBP or 84 mm Hg DBP (less than 86th/81st percentiles based on published norms; National Institutes of Health, 1980). Inclusion criteria were age of 21 to 35 years; weight within 20% of normal, according to Metropolitan Life Insurance Company norms; absence of self-reported major illnesses; no prior treatment for essential hypertension or current use of any prescription medication; usual caffeine consumption equivalent to 1 to 5 cups of coffee per day with no reported intolerance or negative side effects; smoking of fewer than 10 cigarettes per day; and alcohol consumption of fewer than 15 drinks per week. All volunteers signed a consent form approved by the Institutional Review Board of the University of Oklahoma Health Sciences Center and Veterans Affairs Medical Center and were paid for their participation. Consent was obtained from parents to request medical data from their physicians. Protocol Screening occurred in two phases. Phase 1 consisted of three automated BP measurements in 5 min after the subject had been seated upright at rest for 5 min. A second set of three readings was taken over 5 min after 15 min at rest. The subject provided his personal medical history and his parents' hypertension history. Habitual caffeine consumption was estimated by structured recall using a comprehensive questionnaire of daily, weekly, or monthly intake of dietary and medicinal sources. Phase 2 consisted of a physical exam, anthropometric measurements, and a symptom-limited exercise tolerance test using the Bruce protocol (Committee on Exercise, 1972). Test sessions occurred 2 or more days apart at the same time of day. Subjects abstained for 24 hr from alcohol and for 12 hr from caffeine and maintained a fasting state for 12 hr prior to testing. Compliance was checked by having subjects complete a form so stating. Adherence to caffeine restriction was further verified by caffeine assays on plasma samples. Each
237
subject was then assigned at random to one of two caffeine-placebo test orders. The test procedure was as follows. The subjects sat semirecumbent in a recliner chair after having been fitted with a BP cuff and instrumented for recording of heart rate (HR) and cardiac output. A heparinized intravenous catheter was placed in a forearm vein. Testing included quiet rest and baseline (20 min), caffeine or placebo administration, drug absorption (40 min), task instructions, psychomotor task (15 min), and recovery (20 min). The subject was allowed to read general-interest reading materials during nontask periods. Caffeine (USP, anhydrous; Amend Drug and Chemical Co., Irvington, NJ) was administered orally in a 3.3 mg/kg dose mixed with 6 oz of unsweetened grapefruit juice (Texsun, Weslaco, TX). Placebo consisted of the grapefruit juice alone. The psychomotor test consisted of an unsignaled, simple visual reaction-time task having 60 trials with intertrial intervals ranging randomly from 4 to 30 sec, with a mean interval of 15 sec. Programming used digital logic apparatus (Med Associates, East Fair field, VT). The subject was provided a response key near his right hand and was told that each "very rapid" response (less than 270 msec) would result in his earning a 50
.18, or between risk groups (Ms = 4.75 jitg/ml and 5.39 jig/ml, respectively), F(l, 32) = 1.73,/? > .19. These concentrations agree with those reported by others using comparable doses (Smits, Boekema, DeAbreu, Thien, & van't Laar, 1987). One low-risk person's data were eliminated from the cardiovascular analyses due to elevated plasma caffeine concentrations (1.7 to 2.2 /xg/ml) at all samples on his placebo day—these were at the threshold for producing cardiovascular effects (Whitsett et al.,
TABLE 1 Subject Characteristics Subjects b
All Subjects* Variable Age (years) Height (cm) Weight (kg) Body fat (percentage)6 Quetelet index (gm/cm 2 ) f Treadmill time (min) Caffeine intake (mg/day) 8
M
High Risk0
Low Risk
SEM
Low Versus High Risk
M
SEM
M
SEM
t
Pd
26.7 180 79 22.6
0.71 1.1 1.7 0.7
26.2 180 76 21.5
1.04 1.5 1.9 0.8
27.3 180 82 23.6
0.98 1.6 2.6 1.1
0.72 0.00 1.69 1.55
ns ns ns ns
2.43 14.4 344
0.04 0.43 49
2.35 14.7 390
0.04 0.59 82
2.51 14.1 307
0.06 0.63 60
2.36 0.71 0.85
.03 ns ns
1.9
6.04
.0001
1.3
5.14
.0001
1.3
7.49
.0001
1.7
3.43
.002
2.2
6.02
.0001
1.4
4.89
.0001
BP at Phase 1 Screening After 5 Minh SBP (mm Hg) Range DBP (mm Hg) Range
126
1.7
119
1.6
134
101 to 130 72
1.4
67
112 to 150 1.8
78
48 to 77
65 to 88
BP at Phase 1 Screening After 15 Minh SBP (mm Hg) Range DBP (mm Hg) Range
123
1.4
116
1.2
130
102 to 123 71
1.3
67
121 to 144 1.6
75
49 to 76
61 to 85
BP at Phase 2 Screening' SBP (mm Hg) Range DBP (mm Hg) Range
120
2.1
110
2.1
129
92 to 126 78
1.2
73
110 to 140 1.3
60 to 80
82 70 to 90
Notes. Student's t tests compared low- versus high-risk groups. BP data show means of three readings over 5 min; ranges show extremes of single readings. a N = 39. bn = 19. cn = 20. d ns = nonsignificant. eBased on skinfold thickness; data available for only 16 low-risk and 15 high-risk men. f Weight/height 2 . g Data available for only 16 low-risk and 19 high-risk men. h BP recorded by Critikon Dinamap (Tampa, FL) automated monitor. 'BP recorded by auscultation using disappearance of Korotkoff sounds as DBP.
239
I
,401
• I
Ba««lin«
I"'"'[]
Recovery
™RTT"k
I SYSTOLIC BLOOD PRESSURE
A
( m m Hg)
7%
' o to ' o eg ' o »-
100
' o o
Li ~ m
85
'
" DIASTOLIC DlASTOLIC BLOOD BLOOD PRESSURE PRESSURE (mm HgJ
B 1%
I
m
I ' 10
55
' m c o O 10
1980). Thus, the final sample size was 39 subjects —19 low risk and 20 high risk. The risk groups were compared for resting cardiovascular activity by averaging the predrug baseline activity across caffeine and placebo days. As shown in Table 2, the high-risk group had significantly higher BP at rest, in line with their screening days. This higher BP was due to an elevated systemic vascular resistance with modestly lower HR, cardiac output, and index of contractility. Rate-pressure product was marginally greater among the high-risk men. The reaction-time task was seen to produce highly significant changes from rest on the placebo day as follows: SBP and DBP ( + 6%/+ 3%), systemic vascular resistance (-9%), HR ( + 12%), cardiac output (+19%), contractility index (+14%), minute work ( + 23%), and rate-pressure product (+18%), all Fs(l, 37) > 10.00, ps < .004. The task thus produced a mild defense reaction characterized by enhanced cardiac activity, lowered vascular resistance, and a net increase in BP, as seen in earlier work (Lovallo, Pincomb, & Wilson, 1986). The caffeine effects at rest, in contrast to placebo, are shown in Table 3. Caffeine elevated systemic vascular resistance, leading to significantly increased BP and minute work of the heart, all Fs(l, 37) > 7.40, ps < .009. This is in agreement with earlier studies showing that caffeine's primary effect at rest is on vascular resistance and not on cardiac function (Pincomb et al., 1985; Pincomb et al., 1988). Caffeine affected the risk groups similarly, although the high-risk men showed numerically larger rises in vascular resistance and DBP. The effect of caffeine on responses to the psychomotor task may be seen in Figures 1 and 2. SBP and DBP were both elevated above placebo levels, Fs(l, 37) > 56.18, ps < .0001
o t
SYSTEMIC VASCULAR RESISTANCE INDEX (Dyne-sec. cm" 5 /m 2 )
C -11%
O in t
TABLE 2 Comparison of Risk Groups at Rest Risk Groupb Low Variable*
M
O 10 o
High SEM
M
SEM
t
1.4 5.68 111 121 SBP (mm Hg) 1.0 71 1.4 6.48 0.8 61 DBP (mm Hg) 59 1.8 0.14 60 HR (beat/min) 1.6 3.16 0.13 CI (1/min/m 2 ) 2..94 0.12 1.30 0.77 12.9 HI (ohms/sec2) 11..4 0.65 1.46 SVRI 2.70 81 2,,498 (dyne-sec • cm"Vm 2 ) 2,078 131 MWI 34 34 1.4 (joules/min/m2) 1.3 0.46 RPP (mm Hg/min) 1.83 227 199 7,,167 6,613 a
Pc .0001 .0001 ns ns ns .01 ns .08
CI = cardiac index, HI = Heather's index of cardiac contractility, SVRI = systemic vascular resistance index, MWI = minute work index, RPP = rate-pressure product. b Entries show predrug resting baseline values averaged over placebo and caffeine days; these values did not differ significantly across days for any variable. Student's t tests compared low- versus high-risk groups. High risk = essential hypertension in one or both parents and resting BP of 135/85 mm Hg or greater; low risk = normotensive parents and resting BP of 132/84 mm Hg or less. c ns = nonsignificant.
O 10
