EMPIRICAL RESEARCH
CARDIOVASCULAR REACTIVITY IN BLACK AND WHITE SIBLINGS VERSUS MATCHED CONTROLS 1,2
Dawn K. Wilson, Ph.D. Medical College of Virginia Samantha D. Holmes, B.S., Kristopher Arheart, Ed.D., and Bruce S. Alpert, M.D. University of Tennessee, Memphis
ABSTRACT
INTRODUCTION
Elevated cardiovcr~cular (CV) reactivity may be a marker or mechanism for the early development of essential hypertension (EH) and may contribute to the greater prevalence of E H observed in Black adults. Previous research has demonstrated that Black children show greater CV reactivity than White children to psychological stressors, however, the role of heritabifity in understanding these racial differences is still unknown. Evidence which supports a genetic influence on CV reactivity comes from animal studies, research on family history of EH, and from twin and sibfing studies. The present study expands on previous findings by examining racial differences in CV reactivity in 15 pairs of Black sibfings, 15 pairs" of age- and sex-matched unrelated Black control subjects, 17 pairs o f White sibfings, and 17 pairs of ageand sex-matched unrelated White control subjects. Systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR) measurements were obtained at rest and during a stress task (competitive video game). Black siblings demonstrated a significantly higher intraclass correlation for DBP reactivity than Black controls or White siblings (r = 0.73, versus 0.16, 0.14, respectively). Additionally, Black siblings demonstrated a steeper rise and then a plateau in DBP and HR reactivity to the video game task, while White siblings showed a more gradual increase in these measures over the course of playing three video games. The results for DBP and HR reactivity, however, were not consistent among either of the matched control groups. These results expand on previous research by suggesting a stronger genetic influence o f C V reactivity in Black than in White children.
Elevated cardiovascular (CV) reactivity [increases in blood pressure (BP) to stress] may be a marker or mechanism for the greater prevalence of essential hypertension (EH) observed in the Black community. The results of several longitudinal studies have shown that greater magnitudes of CV reactivity are characteristic of at-risk individuals who develop EH in later adulthood (1-5). Related studies have demonstrated that Black children show greater CV reactivity than White children to both psychological and physical stressors (6-9) which may increase their risk for developing EH. Little is known, however, about the heritability of CV reactivity patterns in Black versus White children. Familial aggregation of CV reactivity to stress may be influenced by both genetics and shared environmental factors. Three lines of evidence have provided genetic support for CV reactivity: (a) animal models of hypertension, (b) studies on family history of EH, and (c) studies examining twin and sibling patterns of CV reactivity (10). Environmental factors such as dietary factors, alcohol use, low physical activity, and psychological stress may be associated with increased risk for EH. Several of these factors have been shown to be significantly correlated between family members, suggesting that some of the familial aggregation of BP may be due to shared environmental factors (11). In a recent study by Williams et al. (12), however, only 7% of the total variance of diastolic blood pressure (DBP) among family members was attributable to shared environmental factors such as salt intake, alcohol use, and a m o u n t of exercise. The focus of the present study is on examining heritability of CV reactivity among family members such as nontwin sibling pairs. Evidence demonstrating genetic support for CV reactivity is strongest among the animal literature. Several studies conducted by Hallback and colleagues (13,14) examined the relationship between spontaneously hypertensive rats' (SHR) CV responses to stressful environmental stimuli and the subsequent development of hypertension. SHRs are bred such that they show substantial increases in blood pressure at about five to six weeks of age and develop established hypertension by six months of age (15). Hallback (13) has demonstrated that SHRs show greater CV responsiveness to acute noise and vibrating stressors in relation to normotensive control rats. Hallback and Folkow (14) have also shown that SHRs have greater sympathetic nervous system responses to aversive stimuli than control rats.
(Ann Behav Med 1995, 17(3):207-212)
Preparation of this manuscript was supported in part by the National Institutes of Health Grants HL-35788 and GCRC R07-32-359, as well as by a National Institutes of Health Underrepresented Undergraduate Minority Research Fellowship grant to the second author. 2 The authors would like to acknowledge Karen Williams, PhylLisRJchey, Nelda Smith, and Zondrah Williams for their assistance with data collection on this research project.
Reprint Address: D. K. Wilson, Ph.D., Medical College of Virginia, Department of Medicine, Division of Clinical Pharmacology and Hypertension, P.O. Box 980160, Richmond, VA 23298-0160. 9 1995 by The Society of Behavioral Medicine.
207
208
ANNALS
OF BEHAVIORAL
MEDICINE
Other investigators have also demonstrated similar results in S H R s which provide evidence o f a genetic influence on CV responsiveness ( 16,17). Previous research in h u m a n s has provided moderate evidence suggesting a genetic predisposition o f EH may be associated with elevated CV reactivity. For example, research has demonstrated that individuals with a positive family history o f EH show greater CV reactivity than individuals with no family history o f EH (1,18-21). Falkner et at. (18) demonstrated that adolescents who had a positive family history o f EH had greater systolic blood pressure (SBP), DBP, and heart rate (HR) responses to mental arithmetic than adolescents who had no family history o f EH. In another study, Falkner et al. (1) reported that adolescents with borderline hypertension and a genetic predisposition for EH showed more similar CV response patterns than individuals with labile hypertension. In a review by Alpert and Wilson (19), 73% (8/11) o f the studies on children demonstrated a positive relationship between family history and either SBP or DBP reactivity. Matthews and Rakaczky (20) have also provided a review o f the adult literature and reported that 70% o f 27 studies showed greater stress-induced BP responses in individuals with (versus without) a family history o f CV disease. Furthermore, a recent meta-analytic review by Fredrikson and Matthews (21) indicated that 36--43% o f the studies showed that individuals with a family history o f EH had greater BP and H R reactivity in response to active behavioral stressors than did individuals with no family history o f EH. Twin studies have also provided moderate support for a genetic influence on CV reactivity patterns. Carmelli et al. (22) studied 12 pairs o f monozygotic (MZ) and 21 pairs o f dizygotic (DZ) adult male twins (aged 54-64) and reported that M Z twins exhibited greater correlations for SBP, DBP, and H R reactivity to a mental arithmetic task than did DZ twins. A study by Smith, Turner, and Ford (23) o f adult male twins (88 pairs o f DZ, 82 pairs o f MZ; aged 21-61), one o f the largest twin studies conducted, compliments Carmelli et al.'s (22) findings. Their results showed greater correlations for both SBP and DBP reactivity in M Z than in DZ twins. In another large twin comparison study, MclIhany, Shaffer, and Hines (24) compared BP responses o f 200 pairs o f twins (87 MZ p a i r s - - 4 0 male, 47 female; 113 DZ p a i r s - - 3 2 male, 36 female, 45 mixed; mean age = 14 years) during a cold pressor task. Their findings showed that M Z twins shared greater intraclass correlations for SBP and DBP on the cold pressor task for like-sex twin pairs. Rose, Grim, and Miller (25) compared BP reactivity across several stress tasks (e.g. Stroop test, isometric handgrip) in an adolescent sample o f 111 pairs o f M Z twins, 66 pairs o f same-sex DZ twins, and 54 pairs o f non-related controls matched for gender. Their results showed significantly greater reactivity correlations for MZ twins than D Z twins, with no significant correlations for reactivity among the non-related controls. F r o m these studies, it is evident that there is a heritable quality to BP reactivity in twin populations. The degree to which familial aggregation is related to CV reactivity patterns remains an important question. Specifically, little is known about the generalizability o f the results from the twin studies to non-twin siblings. Only limited research has been conducted which compares correlations in CV reactivity among sibling pairs. In a study by Ditto (26), BP responses in 36 young adult non-twin sibling pairs (aged 18-20) demonstrated significant BP and HR correlations at rest and during a variety of tasks, such as the cold pressor task and a visual-verbal test. In another study by Matthews et al. (27), sibling resemblances in
W i l s o n et al. BP and H R reactivity were evaluated in 142 upper-middle-class White families. The results demonstrated sibling similarities in H R reactivity to mirror image tracing and in SBP reactivity to isometric handgrip exercise. Thus, these studies provide preliminary evidence that sibling comparisons may also suggest a genetic influence on CV reactivity. Racial differences in twin and sibling CV reactivity patterns have not previously been examined. The aim o f the present study was to examine racial differences in CV reactivity patterns in Black siblings and White siblings versus age, race, and sexmatched unrelated pairs o f control subjects. Substantial evidence indicates that there are differences in CV reactivity between Black and White children (6-9,19), however, little is known about racial differences in familial aggregation of CV reactivity. Additionally, there is a greater prevalence and severity o f EH in Black than in White adults (28,29), suggesting that greater aggregation o f CV reactivity m a y be more prevalent among Blacks (versus Whites). In light o f this evidence, we sought to investigate the familial aggregation o f CV reactivity in 15 pairs o f Black siblings, 15 pairs o f age- and sex-matched unrelated Black control subjects, i 7 pairs o f White siblings and in 17 pairs o f age- and sex-matched unrelated White control subjects.
METHOD
Subjects One hundred twenty-eight healthy Black and White siblings and control children (15 pairs o f Black siblings, 15 pairs o f unrelated Black controls, 17 pairs o f White siblings, and 17 pairs o f unrelated White controls) were recruited from schools, churches, social organizations, and through radio and newspaper advertisements. 3 The average age o f subjects was 13 _+ 2 years. Health status was evaluated based on each child's medical history elicited from his/her parent(s) to rule out the presence o f CV disease or other chronic illness. In addition, each child participated in a health screening which included a BP assessment and a urine specimen to rule out any abnormalities in red blood cells, white blood cells, urinary p l I , protein, or glucose. A sibling was defined as having the same biological mother and father. Each sibling pair consisted ofonly two siblings per family. In two cases where more than two siblings had participated, we randomly selected two siblings for inclusion in the data analyses. There was an equal breakdown ofgender across Black and White sibling pairs (White pairs = 2 female, 9 male, 6 mixed versus Black pairs = 3 female, 4 male, 8 mixed, X2 (df = 2) = 2.3, p = ns). All o f the sibling participants were living in the same household at the time o f the investigation. For each pair of siblings, a control pair o f children was created by identifying two unrelated children who were comparable to the siblings in age (within six months), sex, and r a c e : All participants were paid $30 for taking part in the study.
3 Data analyzed in this study are part of a larger data set from a study examining "'Familial Hypertension and Biracial Cardiovascular Reactivity in Youth": Bruce Alperl, M.D., Principal Investigator. Bccause the study testing protocol was quite time consuming, it was not possible for control subject pairs to be tested on the same day. 4 Because information was not obtained regarding family size and birth order, sibling and control pairs were nol matched on these variables.
CV Reactivity in Black and White Siblings Procedure Participants and their parents first provided us with demographic information which included obtaining measures o f annual family income, sex, age, height (cm), weight (kg), Quetelet index (kg/m2), and Tanner stages (30). Subjects had their BP measurements taken by a trained technician and were seated in a relaxed position with their legs uncrossed. Care was taken to select the proper size cuff for each subject based on the reco m m e n d a t i o n s outlined by the Second Task Force on BP control (31). F o r all participants, the BP cuff was placed on the nond o m i n a n t arm throughout the BP testing procedure. BP measurements were assessed using a D i n a m a p automated BP machine. Prior to baseline BP assessment, each subject's BP measurements were obtained from the automated machine and c o m p a r e d with a sphygrnomanometer to assure accuracy. Only subjects whose SBP measurements were within + 5 m m Hg were allowed to participate in the s t u d y : After a five-minute rest period, a total o f ten SBP, DBP, and H R measurements were taken with a 30-second interval between readings. The average o f these ten measurements were used as baseline values for the d a t a analyses. Baseline values o f BP were not used in calculating CV reactivity scores. Separate resting BP values were taken prior to beginning the CV reactivity test (described below) and were used to determine reactivity change scores.
Cardiovascular Reactivity Testing Competitive Video Game: The competitive video game pro-
VOLUME 17, N U M B E R 3, 1995
209
Study Design A total o f 32 pairs o f siblings (15 Black, 17 White) a n d 32 pairs o f matched unrelated control children were included in the final study sample. The independent variables were race (Black a n d White) and family type (sibling and control). F a m i l y type was treated as a repeated measure in the study design. The dependent variables were SBP, DBP, and H R reactivity. Children were nested within families, and both families and children were considered random. Separate analyses for each dependent variable were performed. SAS (SAS Institute Inc., Cary, NC) was used for all computations. RESULTS
Subject Characteristics Demographic information on the final sample is presented in Table l, sorted by race and family type. Income and sex were analyzed using the Cochran-Mantel-Haenszel test, comparing family type (sibling-control) while controlling for race and family affiliation (sibling families-control families). A repeated measures analysis o f variance with planned comparisons was used to test the remaining v a r i a b l e s : Planned comparisons tested for differences between control and sibling families within race and for racial differences within sibling families. Pearson correlations o f reactivity outcome variables with continuous d e m o graphic factors were computed to determine possible covariates for subsequent analyses. Tanner measures were significantly higher for Black siblings than White siblings (breast: p = .05; pubic: p < .04). The p values were non-significant for demographic, baseline, and performance measures o f annual family income, Quetelet index, baseline SBP, baseline DBP, baseline H R , average video points, and average video time. N o n e o f the Pearson correlations between reactivity and demographic and baseline variables were significant.
cedure has been previously described in detail and was adapted from Murphy and colleagues (6-8). Each participant was seated in front o f a 25-inch color television monitor. Next, four resting SBP, DBP, and H R readings were taken by a trained technician. The television m o n i t o r was turned on and the subject was given standard instructions for the Atari game "Breakout." Each participant was provided with a brief practice period to allow familiarization with the game. In G a m e 1 (personal challenge), participants were told to see how well they could do. In G a m e 2 (experimenter challenge), they were told they d i d pretty well in G a m e 1 but to try harder this time. In G a m e 3 (monetary incentive), they were told they could earn some m o n e y i f they beat their higher score obtained in Games 1 and 2. Participants were not told how much m o n e y they could earn, but were paid five cents to five dollars depending on their performance. During each game, one measurement o f SBP, DBP, and H R was taken 30 seconds after the participant began playing the game. A fiveminute recovery followed the video game task, however, no other rest period was provided for the participants during the entire reactivity task. Reactivity scores were calculated for each subject by subtracting the average o f the four measurements obtained during the resting phase from the average o f the three measurements obtained during the stress phase. Two measures o f video game performance were also assessed: (a) the average time in minutes that it took to complete all three games, and Co) the average number o f points obtained across all three games.
A repeated measures analysis o f variance with planned comparisons was used to analyze SBP, DBP, and H R reactivity
5 Because accurate automated BP readings were unobtainable on some children, we only included children in this study who demonstrated consistent readings across the automated and manual BP methods. Approximately 5% of the children gave inaccurate automated BP readings.
6 Planned comparisons were used to detect differences specifically between Black and White sibling groups and within race across sibling and control pairs because these comparisons were directly relevant to our research question. We expected that Black siblings would show greater aggregation of CV reactivity than Black controls and White siblings.
Intraclass Correlations The intraclass correlation coefficients and 95% confidence intervals (32) were computed for each race-family type c o m bination for SBP, DBP, and H R reactivity. The confidence intervals were inspected for coefficients significantly different from zero, and comparisons were m a d e to detect significant differences between control and sibling families within race and for racial differences within sibling families. Table 2 presents the intraclass correlations, means, and standard deviations for SBP, DBP, and H R reactivity. A significant intraclass correlation was found for DBP reactivity in Black siblings. Planned comparisons showed that Black siblings had a significantly higher DBP reactivity correlation than Black controls or White siblings. N o other correlations were significant.
Magnitude of BP Reactivity
210
ANNALS OF BEHAVIORAL
MEDICINE
Wilson et al.
TABLE 1 Demographic, Basefine, and Video Performance Measures (Mean ___SD) by Family Type and Race
Variables Annual Income Under $10,000 $10,000-20,000 $20,001)-30,000 over $30,000 Quetelet Index (kg/m2) Tanner Stage Breast* Tanner Stage Pubic* Baseline Systolic BP (mm Hg) Baseline Diastolic BP (ram Hg) Baseline Heart Rate (bpm) Average Video Points Average Video Time (minutes)
Black Sibs
Black Cs
White Sibs
White Cs
13% 20% 17% 50% 20 + 4 4+ 1 4 __+1 110 + 11 67 + 8 77 + 10 39 + 54 6+ 3
17% 33% 27% 23% 22 + 6 4+ 1
0% 6% 12% 82% 20 + 4 3+ 1
3% 6% 24% 67% 20 + 5 3+ 1
4 + 1
3 + 1
110 ___12 68 + 8 77 + 9 22 + 18 7 _+ 3
107 + 8 66 + 8 83 + 13 41 + 35 5 -+ 2
3 + 1
108 66 81 44 5
+ 9 + 9 _ 11 +__36 + 2
*p < .05 (Black siblings > White siblings)
across G a m e s 1 to 3. Planned comparisons tested differences for each game between control and sibling families within race and for racial differences within sibling families. Planned comparisons were also made to detect significant differences between Games 1 and 2 and between Games 2 and 3 within each racefamily type. Previous research by Murphy et al. (8) has demonstrated that Black children show a greater change in BP reactivity as compared to White children during Games 1 and 2. Thus, we expected that Black siblings, in particular, would show this greater increase in reactivity early on in the task than either the Black controls or White siblings. Table 3 presents the mean change in SBP, DBP, and H R reactivity during Games 1, 2, and 3 for Black and White siblings and for their matched controls. SBP reactivity at Game 3 was significantly higher than at Game 2 for all groups (Black siblings: p < .03; Black controls: p < .02; White siblings: p < .01; and White controls: p < .01). In Black siblings, DBP reactivity was significantly higher at Game 2 than at Game 1 (p < .03), and in White siblings, DBP reactivity was significantly higher in Game 3 than in Game 2 (p < .03). H R reactivity at Game 2 was significantly higher for Black siblings than White siblings (p < .01). H R reactivity in Black siblings was significantlyhigher in G a m e 2 than in Game 1 09 < .03). H R reactivity in Black controls and White siblings was significantly higher in Game 3 than in G a m e 2 (p < .04 and p < .01, respectively). H R reactivity in White controls was significantly higher in Game 2 than in G a m e 1 a n d in Game 3 than in G a m e 2 (p < .03 and p < .01, respectively).
DISCUSSION The results of the present study demonstrated that Black siblings showed a greater correlation for DBP reactivity than Black control subjects and White siblings. Similar, though nonsignificant, correlational patterns were also found for SBP and H R reactivity. These results provide preliminary evidence for a stronger genetic influence on CV reactivity patterns in Black than in White siblings. Previous twin studies have demonstrated that BP and H R reactivity correlations for MZ twins range from .42 to .77 and for DZ twins f r o m . 14 to .46 (22-25). Ditto's (26) research on non-twin siblings demonstrated significant reactivity correlations ranging from .27 to .68 and Matthews et al. (27) showed significant reactivity correlations among siblings ranging from .24 to .34. The magnitude of these correlations across both the twin and sibling studies have appeared to be somewhat dependent on the sample size employed in each study. For example, McIlhany et al. (24) studied 200 pairs of MZ and D Z twins and their correlations ranged from .73 to .77 for MZ twins and from .34 to .46 for DZ twins. However, in a study of only 12 pairs of MZ and 21 pairs of DZ twins, Carmelli et al. (22) found that their correlations ranged from .51 to .80 for MZ twins and from 914 t o . 19 for DZ twins. Furthermore, in a study of 36 non-twin siblings, Ditto (26) reported correlations which ranged from .27 to .62. Although the magnitude of the correlations have tended to be weaker for studies with smaller sample sizes, this has not reduced their power to detect statistically significant relationships across twin and sibling pairs. The magnitude of the cor-
TABLE 2 Intraclass Correlations (Lower 95%-Upper 95% Confidence Interval), Means, and Standard Deviations of BP Reactivity Measures by Family Type and Race
Blacks
Whites
Intraclass Correlations
Siblings
Controls
Siblings
Controls
SBP Reactivity (mm Hg) DBP Reactivity (ram Hg) HR Reactivity (bpm) Means + SD SBP Reactivity (ram Hg) DBP Reactivity (mm Hg) HR Reactivity (bpm)
.27 (.00-.68) .73* (.37-.90) .39 (.00-.74)
.01 (.00-.38) .16 (.00-.61) .19 (.00-.63)
.01 (.00-.41) .14 (.00-.57) .19 (.00-.60)
.10 (.00-.54) .11 (.00-.55) .01 (.00-.20)
10.3 + 8.7 10.2 _+ 8.2 7.2 _+ 9.6
8.7 + 9.2 8.7 + 7.1 5.6 -+ 9.0
9.4 + 8.3 10.5 ___8.0 2.1 + 7.2
9.9 _+ 7.8 8.5 _+ 7.3 3.6 + 7.4
* p < .05 (Black siblings > Black controls and White siblings)
C V Reactivity in B l a c k and W h i t e Siblings
V O L U M E 17, N U M B E R 3, 1995
211
TABLE 3 Mean and Standard Error of the Mean (+ SEM) Increase in SBP, DBP, and HR Reactivity by Family Type and Race Blacks Variable Mean SBP (mm Hg) Game 1 Game 2 Game 3 Mean DBP (mm Hg) Game 1 Game 2 Game 3 Mean HR (bpm) Game 1 Game2 Game 3
Whites
Siblings
Controls
Siblings
Controls
6.8 + 1.6 9.4 + 2.1 14.7 _+ 2.2*
7.8 ___2.4 6.9 + 1.7 13.1 + 1.5"
6.4 + 1.2 7.6 + 1.6 14.8 _+ 2.2*
6.4 + 2.5 6.7 + 1.5 17.0 + 2.2*
5.8 + 1.9 11.4 _ 2.4* 13.3 --+2.1
7.5 + 1.8 8.1 + 1.6 10.6 ___1.4
7.5 + 1.3 9.4 ___2.4 14.7 + 2.8*
8.5 + 1.9 6.9 --_ 1.3 10.5 + 2.2
-.8 + 1.6 - 1 . 0 + 1.8 7.6 + 2.6*
-2.6 + 1.5 2.8 + 1.6" 10.0 + 2.1"
1.7 + 1.5 7.7 + 2.1" 12.2 + 2.9
2.3 + 2.2 4.5 + 1.9 9.8 + 2.1"
*p < .05 Note: * Indicates a significant change from the preceding game to the * game.
relations in the present study were comparable with correlations shown in previous sibling studies but only for DBP reactivity a m o n g Black sibling pairs (range .27 to .73). G i v e n a larger sample size, we may have had more power to detect reliable differences in Black sibling pairs across SBP (r = .27) and H R (r = .39) reactivity measures. Another interesting finding in the present study was that the Black sibling pairs demonstrated a steeper rise and then a plateau in both DBP and H R reactivity to the video game, while White sibling pairs showed a more gradual increase in reactivity with the increasing levels o f challenge to the mental stress task. This pattern of results was not consistent, however, among either of the control groups. These findings suggest that the magnitude of BP response to stress may be heritable and more acute in Blacks. These results are consistent with the findings of Murphy et at. (8) who demonstrated that Black children show a greater change in BP reactivity as compared to White children during Games 1 and 2. Light et al. (5) have also demonstrated that Black subjects exhibit greater vasoconstriction than White subjects across various psychological stressors (i.e. competitive reaction time task), as evidenced by their total peripheral resistance responses. Treiber and colleagues (33) found that Black male children exhibited significantly greater increases in total peripheral resistance and DBP and significantly longer plateaus in these responses during recovery to a cold face stimulus task than did White male children. Results of another study by Treiber and colleagues (34) demonstrated that Blacks had greater systemic vascular resistance than Whites to the cold face task. In view of these data, Black children who show greater levels of CV reactivity to psychological stress may be at the highest risk for developing EH, however, little is known about the role of genetics as a potential causal influence underlying these physiological response patterns. Further research is warranted to determine whether the acute increases in BP and H R responses to stress among Black siblings in the present study can be replicated in a larger population a n d associated with underlying changes in peripheral resistance. There are several limitations to the present study. First, the sample size was relatively small, which resulted in lowering our power to detect statistically significant relationships across SBP and H R reactivity measures. Perhaps more robust findings could have been apparent, especially for SBP reactivity, with a larger
n u m b e r of subjects. Future research should employ larger sample sizes to confirm our findings. A n additional limitation of the present study was that we did not control for shared environmental factors, such as diet and exercise. We also did not control for variables such as family history of EH, birth order, and family size. Accounting for these influences may have yielded more convincing support for the genetic influence of CV reactivity in Black siblings. In conclusion, these results provide preliminary evidence which suggests that the influence o f genetics on CV reactivity may be stronger in Black than in White children. Moreover, they expand on previous findings regarding racial differences in reactivity. Further research is needed to better understand the underlying mechanisms that may account for the greater correlations in CV reactivity in Black versus White siblings and to see if these findings replicate in larger sample populations. Because Black children are at particularly high risk for developing EH in early adulthood, further studies should investigate the link between heredity and CV reactivity specifically in these children. REFERENCES (1) Falkner B, Kushner H, Onesti G, Angelakos ET: Cardiovascular characteristics in adolescents who develop essential hypertension. Hypertension. 1981, 3(5):521-527. (2) Wood DL, Sheps SG, Elveback LR, Schringer A: Cold pressor test as a predictor of hypertension. Hypertension. 1984, 6:301-306. (3) Menkes MS, Matthews KA, Krantz DS, et al: Cardiovascular reactivity to the cold pressor test as a predictor of hypertension. Hypertension. 1989, 14:524-530. (4) Rose RJ, Chesney MA: Cardiovascular stress reactivity: A behavior-genetic perspective. Behavior Therapy. 1986, 17:31 4-323. (5) Light KC, Sherwood A, Turner JR: High cardiovascular reactivity to stress: A predictor of later hypertension development. In Turner JR, Sherwood A, Light KC (eds), Individual Differences in Cardiovascular Response to Stress: Applications to Models of Cardiovascular Disease. New York: Plenum, 1992, 281-293. (6) Murphy JK, Alpert BS, Moes DM, Somes GW: Race and cardiovascular reactivity: A neglected relationship. Hypertension. 1986, 8:1075-1083. (7) Murphy JK, Alpert BS, Walker SS, Willey ES: Race'and cardiovascular reactivity: A replication. Hypertenszon. 1988, 11:308311.
212
A N N A L S OF B E H A V I O R A L M E D I C I N E
(8) Murphy JK, AIpert BS, Willey ES, Somes GW: Cardiovascular reactivity to psychological stress in healthy children. Psychophysiology. 1988, 25:144-152. (9) Treiber FA, Musante L, Braden D, et al: Racial differences in hemodynamic responses to the cold face stimulus in children and adults. Psychosomatic Medicine. 1990, 52:286-296. (10) Saab PG, Schneiderman N: Biobehavioral stressors, laboratory investigation, and the risk of hypertension. In Blascovich J, Katkin ES (eds), Cardiovascular Reactivity to PsychologicalStress and Disease. Washington, DC: American Psychological Association, 1993, 49-82. (11) Hunt SC, Hasstedt SJ, Kuida H, ct al: Genetic heritability and common environmental components of resting and stressed blood pressures, lipids, and body mass index in Utah pedigrees and twins. American Journal of Epidemiology. 1989, 129:625-638. (12) Williams RR, Hunt SC, Hasstedt SJ, et al: Are there interactions and relations between genetic and environmental factors predisposing to high blood pressure? Hypertension. 1991, 18:I-29-I-37. (13) Hallback M: Consequence of social isolation on blood pressure, cardiovascular reactivity, and design in spontaneously hypertensive rats. Acta Physiologica Scandinavica. 1975, 93:455-465. (14) Hallhack M, Folkow B: Cardiovascular responses to acute mental "stress" in spontaneously hypertensive rats. Acta Physiologica Scandinavica. 1974, 90:684-698. (15) Schneiderman N, McCabe PM: Psychophysiologic strategies in laboratory research. In Schneiderman N, Weiss SM, Kaufman PG (eds), Handbook of Research Methods in CardiovascularBehavioral Medicine. New York: Plenum, 1989, 349-364. (l 6) Hallback-Norlander M, Lundin S: Background of hyperkinetic circulatory state in young spontaneously hypertensive rats. In Meyer P, Schmitt H (eds), Nervous System and Hypertension. New York: Wiley, 1979. (17) McCarty R, Kopin IJ: Alterations in plasma catecholamines and behavior during acute stress in spontaneously hypertensive and Wistar-Kyoto normotensive rats. Life Science. 1978, 22:997. (18) Falkner B, Onesti G, Angelakos ET, Fernandes M, Langman C: Cardiovascular response to mental stress in normal adolescents with hypertensive parents: Hemodynamics and mental stress in adolescents. Hypertension. 1979, 1:23--30. (19) Alpert BS, Wilson DIC Stress reactivity in childhood and adolescence. In Turner JR, Sherwood A, Light K (eds), Individual Differences in CardiovascularResponse to Stress: Applications to Models of Cardiovascular Disease. New York: Plenum, 1992, 187-201. (20) Matthews KA, Rakaczky CJ: Familial aspects of the Type A behavior pattern and physiologic reactivity to stress. In Schmidt TH,
Wilson et al. Dembroski TM, Blumchen G (eds), Biological and Psychological Factors in CardiovascularDisease. Berlin, Germany: Springer-Verlag, 1986, 228-245. (21) Fredrikson M, Matthews KA: Cardiovascular responses to behavioral stress and hypertension: A meta-analytic review. Annals of Behavioral Medicine. 1990, 12:30-39. (22) Carmelli D, Chesney MA, Ward MM, Rosenman RH: Twin similarity in cardiovascular stress response. Health Psychology. 1985, 4:413-423. (23) Smith TW, Turner CW, Ford MH: Blood pressure reactivity in adult male twins. Health Psychology. 1987, 6(3):209-220. (24) McIlhany ML, Shaffer JW, Hines EH: Heritability of blood pressure: An investigation of 200 pairs of twins using the cold pressor test. Johns Hopkins Medical Journal. 1975, 136:57-64. (25) Rose RJ, Grim CE, Miller JZ: Familial influences on cardiovascular stress reactivity: Studies of normotensive twins. Behavioral Medicine Update. 1984, 6:21-24. (26) Ditto B: Sibling similarities in cardiovascular reactivity to stress. Psychophysiology. 1987, 24(3):353-360. (27) Matthews KA, Manuck SB, Stoney CM, et al: Familial aggregation of blood pressure and heart rate response during behavioral stress. Psychosomatic Medicine. 1988, 50:341-352. (28) U.S. Department of Health and Human Services: Cardiovascular and Cerebrovascular Disease: Report to the Secretary's Task Force on Black and Minority Health. Washington, DC: U.S. Department of Health and Human Services, 1986, 267-274. (29) Anderson NB, McNeilly M, Myers H: Toward understanding race differences in autonomic reactivity: A proposed contextual model. In Turner JR, Sherwood A, Light KC (eds), Individual Differences in Cardiovascular Response to Stress. New York: Plenum, 1992, 125-145. (30) Tanner JM: Normal growth and techniques of growth assessment. Clinics in Endocrinology and Metabolism. 1986, 15:411-451. (31) U.S. Department of Health and Human Services: Report of the Second Task Force on Blood Pressure Control in Children. Washington, DC: U.S. Government Printing Office, 1987. (32) Koch GG: Intraclass correlation. In Kotz S, Johnson NL (eds), Encyclopedia of Statistical Sciences (Vol. 4). New York: John Wiley & Sons, 1983, 212-217. (33) Treiber FA, Musante L, Braden D, et al: Racial differences in hemodynamic responses to the cold face stimulus in children and adults. Psychosomatic Medicine. 1990, 52:286-296. (34) Treiber FA, Davis H, Musante L, et al: Ethnicity, gender, family history of myocardial infarction, and hemodynamic responses to laboratory stressors in children. Health Psychology. 1993, 1:6-15.
CARDIOVASCULAR REACTIVITY IN BLACK AND WHITE SIBLINGS VERSUS MATCHED CONTROLS 1,2
Dawn K. Wilson, Ph.D. Medical College of Virginia Samantha D. Holmes, B.S., Kristopher Arheart, Ed.D., and Bruce S. Alpert, M.D. University of Tennessee, Memphis
ABSTRACT
INTRODUCTION
Elevated cardiovcr~cular (CV) reactivity may be a marker or mechanism for the early development of essential hypertension (EH) and may contribute to the greater prevalence of E H observed in Black adults. Previous research has demonstrated that Black children show greater CV reactivity than White children to psychological stressors, however, the role of heritabifity in understanding these racial differences is still unknown. Evidence which supports a genetic influence on CV reactivity comes from animal studies, research on family history of EH, and from twin and sibfing studies. The present study expands on previous findings by examining racial differences in CV reactivity in 15 pairs of Black sibfings, 15 pairs" of age- and sex-matched unrelated Black control subjects, 17 pairs o f White sibfings, and 17 pairs of ageand sex-matched unrelated White control subjects. Systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR) measurements were obtained at rest and during a stress task (competitive video game). Black siblings demonstrated a significantly higher intraclass correlation for DBP reactivity than Black controls or White siblings (r = 0.73, versus 0.16, 0.14, respectively). Additionally, Black siblings demonstrated a steeper rise and then a plateau in DBP and HR reactivity to the video game task, while White siblings showed a more gradual increase in these measures over the course of playing three video games. The results for DBP and HR reactivity, however, were not consistent among either of the matched control groups. These results expand on previous research by suggesting a stronger genetic influence o f C V reactivity in Black than in White children.
Elevated cardiovascular (CV) reactivity [increases in blood pressure (BP) to stress] may be a marker or mechanism for the greater prevalence of essential hypertension (EH) observed in the Black community. The results of several longitudinal studies have shown that greater magnitudes of CV reactivity are characteristic of at-risk individuals who develop EH in later adulthood (1-5). Related studies have demonstrated that Black children show greater CV reactivity than White children to both psychological and physical stressors (6-9) which may increase their risk for developing EH. Little is known, however, about the heritability of CV reactivity patterns in Black versus White children. Familial aggregation of CV reactivity to stress may be influenced by both genetics and shared environmental factors. Three lines of evidence have provided genetic support for CV reactivity: (a) animal models of hypertension, (b) studies on family history of EH, and (c) studies examining twin and sibling patterns of CV reactivity (10). Environmental factors such as dietary factors, alcohol use, low physical activity, and psychological stress may be associated with increased risk for EH. Several of these factors have been shown to be significantly correlated between family members, suggesting that some of the familial aggregation of BP may be due to shared environmental factors (11). In a recent study by Williams et al. (12), however, only 7% of the total variance of diastolic blood pressure (DBP) among family members was attributable to shared environmental factors such as salt intake, alcohol use, and a m o u n t of exercise. The focus of the present study is on examining heritability of CV reactivity among family members such as nontwin sibling pairs. Evidence demonstrating genetic support for CV reactivity is strongest among the animal literature. Several studies conducted by Hallback and colleagues (13,14) examined the relationship between spontaneously hypertensive rats' (SHR) CV responses to stressful environmental stimuli and the subsequent development of hypertension. SHRs are bred such that they show substantial increases in blood pressure at about five to six weeks of age and develop established hypertension by six months of age (15). Hallback (13) has demonstrated that SHRs show greater CV responsiveness to acute noise and vibrating stressors in relation to normotensive control rats. Hallback and Folkow (14) have also shown that SHRs have greater sympathetic nervous system responses to aversive stimuli than control rats.
(Ann Behav Med 1995, 17(3):207-212)
Preparation of this manuscript was supported in part by the National Institutes of Health Grants HL-35788 and GCRC R07-32-359, as well as by a National Institutes of Health Underrepresented Undergraduate Minority Research Fellowship grant to the second author. 2 The authors would like to acknowledge Karen Williams, PhylLisRJchey, Nelda Smith, and Zondrah Williams for their assistance with data collection on this research project.
Reprint Address: D. K. Wilson, Ph.D., Medical College of Virginia, Department of Medicine, Division of Clinical Pharmacology and Hypertension, P.O. Box 980160, Richmond, VA 23298-0160. 9 1995 by The Society of Behavioral Medicine.
207
208
ANNALS
OF BEHAVIORAL
MEDICINE
Other investigators have also demonstrated similar results in S H R s which provide evidence o f a genetic influence on CV responsiveness ( 16,17). Previous research in h u m a n s has provided moderate evidence suggesting a genetic predisposition o f EH may be associated with elevated CV reactivity. For example, research has demonstrated that individuals with a positive family history o f EH show greater CV reactivity than individuals with no family history o f EH (1,18-21). Falkner et at. (18) demonstrated that adolescents who had a positive family history o f EH had greater systolic blood pressure (SBP), DBP, and heart rate (HR) responses to mental arithmetic than adolescents who had no family history o f EH. In another study, Falkner et al. (1) reported that adolescents with borderline hypertension and a genetic predisposition for EH showed more similar CV response patterns than individuals with labile hypertension. In a review by Alpert and Wilson (19), 73% (8/11) o f the studies on children demonstrated a positive relationship between family history and either SBP or DBP reactivity. Matthews and Rakaczky (20) have also provided a review o f the adult literature and reported that 70% o f 27 studies showed greater stress-induced BP responses in individuals with (versus without) a family history o f CV disease. Furthermore, a recent meta-analytic review by Fredrikson and Matthews (21) indicated that 36--43% o f the studies showed that individuals with a family history o f EH had greater BP and H R reactivity in response to active behavioral stressors than did individuals with no family history o f EH. Twin studies have also provided moderate support for a genetic influence on CV reactivity patterns. Carmelli et al. (22) studied 12 pairs o f monozygotic (MZ) and 21 pairs o f dizygotic (DZ) adult male twins (aged 54-64) and reported that M Z twins exhibited greater correlations for SBP, DBP, and H R reactivity to a mental arithmetic task than did DZ twins. A study by Smith, Turner, and Ford (23) o f adult male twins (88 pairs o f DZ, 82 pairs o f MZ; aged 21-61), one o f the largest twin studies conducted, compliments Carmelli et al.'s (22) findings. Their results showed greater correlations for both SBP and DBP reactivity in M Z than in DZ twins. In another large twin comparison study, MclIhany, Shaffer, and Hines (24) compared BP responses o f 200 pairs o f twins (87 MZ p a i r s - - 4 0 male, 47 female; 113 DZ p a i r s - - 3 2 male, 36 female, 45 mixed; mean age = 14 years) during a cold pressor task. Their findings showed that M Z twins shared greater intraclass correlations for SBP and DBP on the cold pressor task for like-sex twin pairs. Rose, Grim, and Miller (25) compared BP reactivity across several stress tasks (e.g. Stroop test, isometric handgrip) in an adolescent sample o f 111 pairs o f M Z twins, 66 pairs o f same-sex DZ twins, and 54 pairs o f non-related controls matched for gender. Their results showed significantly greater reactivity correlations for MZ twins than D Z twins, with no significant correlations for reactivity among the non-related controls. F r o m these studies, it is evident that there is a heritable quality to BP reactivity in twin populations. The degree to which familial aggregation is related to CV reactivity patterns remains an important question. Specifically, little is known about the generalizability o f the results from the twin studies to non-twin siblings. Only limited research has been conducted which compares correlations in CV reactivity among sibling pairs. In a study by Ditto (26), BP responses in 36 young adult non-twin sibling pairs (aged 18-20) demonstrated significant BP and HR correlations at rest and during a variety of tasks, such as the cold pressor task and a visual-verbal test. In another study by Matthews et al. (27), sibling resemblances in
W i l s o n et al. BP and H R reactivity were evaluated in 142 upper-middle-class White families. The results demonstrated sibling similarities in H R reactivity to mirror image tracing and in SBP reactivity to isometric handgrip exercise. Thus, these studies provide preliminary evidence that sibling comparisons may also suggest a genetic influence on CV reactivity. Racial differences in twin and sibling CV reactivity patterns have not previously been examined. The aim o f the present study was to examine racial differences in CV reactivity patterns in Black siblings and White siblings versus age, race, and sexmatched unrelated pairs o f control subjects. Substantial evidence indicates that there are differences in CV reactivity between Black and White children (6-9,19), however, little is known about racial differences in familial aggregation of CV reactivity. Additionally, there is a greater prevalence and severity o f EH in Black than in White adults (28,29), suggesting that greater aggregation o f CV reactivity m a y be more prevalent among Blacks (versus Whites). In light o f this evidence, we sought to investigate the familial aggregation o f CV reactivity in 15 pairs o f Black siblings, 15 pairs o f age- and sex-matched unrelated Black control subjects, i 7 pairs o f White siblings and in 17 pairs o f age- and sex-matched unrelated White control subjects.
METHOD
Subjects One hundred twenty-eight healthy Black and White siblings and control children (15 pairs o f Black siblings, 15 pairs o f unrelated Black controls, 17 pairs o f White siblings, and 17 pairs o f unrelated White controls) were recruited from schools, churches, social organizations, and through radio and newspaper advertisements. 3 The average age o f subjects was 13 _+ 2 years. Health status was evaluated based on each child's medical history elicited from his/her parent(s) to rule out the presence o f CV disease or other chronic illness. In addition, each child participated in a health screening which included a BP assessment and a urine specimen to rule out any abnormalities in red blood cells, white blood cells, urinary p l I , protein, or glucose. A sibling was defined as having the same biological mother and father. Each sibling pair consisted ofonly two siblings per family. In two cases where more than two siblings had participated, we randomly selected two siblings for inclusion in the data analyses. There was an equal breakdown ofgender across Black and White sibling pairs (White pairs = 2 female, 9 male, 6 mixed versus Black pairs = 3 female, 4 male, 8 mixed, X2 (df = 2) = 2.3, p = ns). All o f the sibling participants were living in the same household at the time o f the investigation. For each pair of siblings, a control pair o f children was created by identifying two unrelated children who were comparable to the siblings in age (within six months), sex, and r a c e : All participants were paid $30 for taking part in the study.
3 Data analyzed in this study are part of a larger data set from a study examining "'Familial Hypertension and Biracial Cardiovascular Reactivity in Youth": Bruce Alperl, M.D., Principal Investigator. Bccause the study testing protocol was quite time consuming, it was not possible for control subject pairs to be tested on the same day. 4 Because information was not obtained regarding family size and birth order, sibling and control pairs were nol matched on these variables.
CV Reactivity in Black and White Siblings Procedure Participants and their parents first provided us with demographic information which included obtaining measures o f annual family income, sex, age, height (cm), weight (kg), Quetelet index (kg/m2), and Tanner stages (30). Subjects had their BP measurements taken by a trained technician and were seated in a relaxed position with their legs uncrossed. Care was taken to select the proper size cuff for each subject based on the reco m m e n d a t i o n s outlined by the Second Task Force on BP control (31). F o r all participants, the BP cuff was placed on the nond o m i n a n t arm throughout the BP testing procedure. BP measurements were assessed using a D i n a m a p automated BP machine. Prior to baseline BP assessment, each subject's BP measurements were obtained from the automated machine and c o m p a r e d with a sphygrnomanometer to assure accuracy. Only subjects whose SBP measurements were within + 5 m m Hg were allowed to participate in the s t u d y : After a five-minute rest period, a total o f ten SBP, DBP, and H R measurements were taken with a 30-second interval between readings. The average o f these ten measurements were used as baseline values for the d a t a analyses. Baseline values o f BP were not used in calculating CV reactivity scores. Separate resting BP values were taken prior to beginning the CV reactivity test (described below) and were used to determine reactivity change scores.
Cardiovascular Reactivity Testing Competitive Video Game: The competitive video game pro-
VOLUME 17, N U M B E R 3, 1995
209
Study Design A total o f 32 pairs o f siblings (15 Black, 17 White) a n d 32 pairs o f matched unrelated control children were included in the final study sample. The independent variables were race (Black a n d White) and family type (sibling and control). F a m i l y type was treated as a repeated measure in the study design. The dependent variables were SBP, DBP, and H R reactivity. Children were nested within families, and both families and children were considered random. Separate analyses for each dependent variable were performed. SAS (SAS Institute Inc., Cary, NC) was used for all computations. RESULTS
Subject Characteristics Demographic information on the final sample is presented in Table l, sorted by race and family type. Income and sex were analyzed using the Cochran-Mantel-Haenszel test, comparing family type (sibling-control) while controlling for race and family affiliation (sibling families-control families). A repeated measures analysis o f variance with planned comparisons was used to test the remaining v a r i a b l e s : Planned comparisons tested for differences between control and sibling families within race and for racial differences within sibling families. Pearson correlations o f reactivity outcome variables with continuous d e m o graphic factors were computed to determine possible covariates for subsequent analyses. Tanner measures were significantly higher for Black siblings than White siblings (breast: p = .05; pubic: p < .04). The p values were non-significant for demographic, baseline, and performance measures o f annual family income, Quetelet index, baseline SBP, baseline DBP, baseline H R , average video points, and average video time. N o n e o f the Pearson correlations between reactivity and demographic and baseline variables were significant.
cedure has been previously described in detail and was adapted from Murphy and colleagues (6-8). Each participant was seated in front o f a 25-inch color television monitor. Next, four resting SBP, DBP, and H R readings were taken by a trained technician. The television m o n i t o r was turned on and the subject was given standard instructions for the Atari game "Breakout." Each participant was provided with a brief practice period to allow familiarization with the game. In G a m e 1 (personal challenge), participants were told to see how well they could do. In G a m e 2 (experimenter challenge), they were told they d i d pretty well in G a m e 1 but to try harder this time. In G a m e 3 (monetary incentive), they were told they could earn some m o n e y i f they beat their higher score obtained in Games 1 and 2. Participants were not told how much m o n e y they could earn, but were paid five cents to five dollars depending on their performance. During each game, one measurement o f SBP, DBP, and H R was taken 30 seconds after the participant began playing the game. A fiveminute recovery followed the video game task, however, no other rest period was provided for the participants during the entire reactivity task. Reactivity scores were calculated for each subject by subtracting the average o f the four measurements obtained during the resting phase from the average o f the three measurements obtained during the stress phase. Two measures o f video game performance were also assessed: (a) the average time in minutes that it took to complete all three games, and Co) the average number o f points obtained across all three games.
A repeated measures analysis o f variance with planned comparisons was used to analyze SBP, DBP, and H R reactivity
5 Because accurate automated BP readings were unobtainable on some children, we only included children in this study who demonstrated consistent readings across the automated and manual BP methods. Approximately 5% of the children gave inaccurate automated BP readings.
6 Planned comparisons were used to detect differences specifically between Black and White sibling groups and within race across sibling and control pairs because these comparisons were directly relevant to our research question. We expected that Black siblings would show greater aggregation of CV reactivity than Black controls and White siblings.
Intraclass Correlations The intraclass correlation coefficients and 95% confidence intervals (32) were computed for each race-family type c o m bination for SBP, DBP, and H R reactivity. The confidence intervals were inspected for coefficients significantly different from zero, and comparisons were m a d e to detect significant differences between control and sibling families within race and for racial differences within sibling families. Table 2 presents the intraclass correlations, means, and standard deviations for SBP, DBP, and H R reactivity. A significant intraclass correlation was found for DBP reactivity in Black siblings. Planned comparisons showed that Black siblings had a significantly higher DBP reactivity correlation than Black controls or White siblings. N o other correlations were significant.
Magnitude of BP Reactivity
210
ANNALS OF BEHAVIORAL
MEDICINE
Wilson et al.
TABLE 1 Demographic, Basefine, and Video Performance Measures (Mean ___SD) by Family Type and Race
Variables Annual Income Under $10,000 $10,000-20,000 $20,001)-30,000 over $30,000 Quetelet Index (kg/m2) Tanner Stage Breast* Tanner Stage Pubic* Baseline Systolic BP (mm Hg) Baseline Diastolic BP (ram Hg) Baseline Heart Rate (bpm) Average Video Points Average Video Time (minutes)
Black Sibs
Black Cs
White Sibs
White Cs
13% 20% 17% 50% 20 + 4 4+ 1 4 __+1 110 + 11 67 + 8 77 + 10 39 + 54 6+ 3
17% 33% 27% 23% 22 + 6 4+ 1
0% 6% 12% 82% 20 + 4 3+ 1
3% 6% 24% 67% 20 + 5 3+ 1
4 + 1
3 + 1
110 ___12 68 + 8 77 + 9 22 + 18 7 _+ 3
107 + 8 66 + 8 83 + 13 41 + 35 5 -+ 2
3 + 1
108 66 81 44 5
+ 9 + 9 _ 11 +__36 + 2
*p < .05 (Black siblings > White siblings)
across G a m e s 1 to 3. Planned comparisons tested differences for each game between control and sibling families within race and for racial differences within sibling families. Planned comparisons were also made to detect significant differences between Games 1 and 2 and between Games 2 and 3 within each racefamily type. Previous research by Murphy et al. (8) has demonstrated that Black children show a greater change in BP reactivity as compared to White children during Games 1 and 2. Thus, we expected that Black siblings, in particular, would show this greater increase in reactivity early on in the task than either the Black controls or White siblings. Table 3 presents the mean change in SBP, DBP, and H R reactivity during Games 1, 2, and 3 for Black and White siblings and for their matched controls. SBP reactivity at Game 3 was significantly higher than at Game 2 for all groups (Black siblings: p < .03; Black controls: p < .02; White siblings: p < .01; and White controls: p < .01). In Black siblings, DBP reactivity was significantly higher at Game 2 than at Game 1 (p < .03), and in White siblings, DBP reactivity was significantly higher in Game 3 than in Game 2 (p < .03). H R reactivity at Game 2 was significantly higher for Black siblings than White siblings (p < .01). H R reactivity in Black siblings was significantlyhigher in G a m e 2 than in Game 1 09 < .03). H R reactivity in Black controls and White siblings was significantly higher in Game 3 than in G a m e 2 (p < .04 and p < .01, respectively). H R reactivity in White controls was significantly higher in Game 2 than in G a m e 1 a n d in Game 3 than in G a m e 2 (p < .03 and p < .01, respectively).
DISCUSSION The results of the present study demonstrated that Black siblings showed a greater correlation for DBP reactivity than Black control subjects and White siblings. Similar, though nonsignificant, correlational patterns were also found for SBP and H R reactivity. These results provide preliminary evidence for a stronger genetic influence on CV reactivity patterns in Black than in White siblings. Previous twin studies have demonstrated that BP and H R reactivity correlations for MZ twins range from .42 to .77 and for DZ twins f r o m . 14 to .46 (22-25). Ditto's (26) research on non-twin siblings demonstrated significant reactivity correlations ranging from .27 to .68 and Matthews et al. (27) showed significant reactivity correlations among siblings ranging from .24 to .34. The magnitude of these correlations across both the twin and sibling studies have appeared to be somewhat dependent on the sample size employed in each study. For example, McIlhany et al. (24) studied 200 pairs of MZ and D Z twins and their correlations ranged from .73 to .77 for MZ twins and from .34 to .46 for DZ twins. However, in a study of only 12 pairs of MZ and 21 pairs of DZ twins, Carmelli et al. (22) found that their correlations ranged from .51 to .80 for MZ twins and from 914 t o . 19 for DZ twins. Furthermore, in a study of 36 non-twin siblings, Ditto (26) reported correlations which ranged from .27 to .62. Although the magnitude of the correlations have tended to be weaker for studies with smaller sample sizes, this has not reduced their power to detect statistically significant relationships across twin and sibling pairs. The magnitude of the cor-
TABLE 2 Intraclass Correlations (Lower 95%-Upper 95% Confidence Interval), Means, and Standard Deviations of BP Reactivity Measures by Family Type and Race
Blacks
Whites
Intraclass Correlations
Siblings
Controls
Siblings
Controls
SBP Reactivity (mm Hg) DBP Reactivity (ram Hg) HR Reactivity (bpm) Means + SD SBP Reactivity (ram Hg) DBP Reactivity (mm Hg) HR Reactivity (bpm)
.27 (.00-.68) .73* (.37-.90) .39 (.00-.74)
.01 (.00-.38) .16 (.00-.61) .19 (.00-.63)
.01 (.00-.41) .14 (.00-.57) .19 (.00-.60)
.10 (.00-.54) .11 (.00-.55) .01 (.00-.20)
10.3 + 8.7 10.2 _+ 8.2 7.2 _+ 9.6
8.7 + 9.2 8.7 + 7.1 5.6 -+ 9.0
9.4 + 8.3 10.5 ___8.0 2.1 + 7.2
9.9 _+ 7.8 8.5 _+ 7.3 3.6 + 7.4
* p < .05 (Black siblings > Black controls and White siblings)
C V Reactivity in B l a c k and W h i t e Siblings
V O L U M E 17, N U M B E R 3, 1995
211
TABLE 3 Mean and Standard Error of the Mean (+ SEM) Increase in SBP, DBP, and HR Reactivity by Family Type and Race Blacks Variable Mean SBP (mm Hg) Game 1 Game 2 Game 3 Mean DBP (mm Hg) Game 1 Game 2 Game 3 Mean HR (bpm) Game 1 Game2 Game 3
Whites
Siblings
Controls
Siblings
Controls
6.8 + 1.6 9.4 + 2.1 14.7 _+ 2.2*
7.8 ___2.4 6.9 + 1.7 13.1 + 1.5"
6.4 + 1.2 7.6 + 1.6 14.8 _+ 2.2*
6.4 + 2.5 6.7 + 1.5 17.0 + 2.2*
5.8 + 1.9 11.4 _ 2.4* 13.3 --+2.1
7.5 + 1.8 8.1 + 1.6 10.6 ___1.4
7.5 + 1.3 9.4 ___2.4 14.7 + 2.8*
8.5 + 1.9 6.9 --_ 1.3 10.5 + 2.2
-.8 + 1.6 - 1 . 0 + 1.8 7.6 + 2.6*
-2.6 + 1.5 2.8 + 1.6" 10.0 + 2.1"
1.7 + 1.5 7.7 + 2.1" 12.2 + 2.9
2.3 + 2.2 4.5 + 1.9 9.8 + 2.1"
*p < .05 Note: * Indicates a significant change from the preceding game to the * game.
relations in the present study were comparable with correlations shown in previous sibling studies but only for DBP reactivity a m o n g Black sibling pairs (range .27 to .73). G i v e n a larger sample size, we may have had more power to detect reliable differences in Black sibling pairs across SBP (r = .27) and H R (r = .39) reactivity measures. Another interesting finding in the present study was that the Black sibling pairs demonstrated a steeper rise and then a plateau in both DBP and H R reactivity to the video game, while White sibling pairs showed a more gradual increase in reactivity with the increasing levels o f challenge to the mental stress task. This pattern of results was not consistent, however, among either of the control groups. These findings suggest that the magnitude of BP response to stress may be heritable and more acute in Blacks. These results are consistent with the findings of Murphy et at. (8) who demonstrated that Black children show a greater change in BP reactivity as compared to White children during Games 1 and 2. Light et al. (5) have also demonstrated that Black subjects exhibit greater vasoconstriction than White subjects across various psychological stressors (i.e. competitive reaction time task), as evidenced by their total peripheral resistance responses. Treiber and colleagues (33) found that Black male children exhibited significantly greater increases in total peripheral resistance and DBP and significantly longer plateaus in these responses during recovery to a cold face stimulus task than did White male children. Results of another study by Treiber and colleagues (34) demonstrated that Blacks had greater systemic vascular resistance than Whites to the cold face task. In view of these data, Black children who show greater levels of CV reactivity to psychological stress may be at the highest risk for developing EH, however, little is known about the role of genetics as a potential causal influence underlying these physiological response patterns. Further research is warranted to determine whether the acute increases in BP and H R responses to stress among Black siblings in the present study can be replicated in a larger population a n d associated with underlying changes in peripheral resistance. There are several limitations to the present study. First, the sample size was relatively small, which resulted in lowering our power to detect statistically significant relationships across SBP and H R reactivity measures. Perhaps more robust findings could have been apparent, especially for SBP reactivity, with a larger
n u m b e r of subjects. Future research should employ larger sample sizes to confirm our findings. A n additional limitation of the present study was that we did not control for shared environmental factors, such as diet and exercise. We also did not control for variables such as family history of EH, birth order, and family size. Accounting for these influences may have yielded more convincing support for the genetic influence of CV reactivity in Black siblings. In conclusion, these results provide preliminary evidence which suggests that the influence o f genetics on CV reactivity may be stronger in Black than in White children. Moreover, they expand on previous findings regarding racial differences in reactivity. Further research is needed to better understand the underlying mechanisms that may account for the greater correlations in CV reactivity in Black versus White siblings and to see if these findings replicate in larger sample populations. Because Black children are at particularly high risk for developing EH in early adulthood, further studies should investigate the link between heredity and CV reactivity specifically in these children. REFERENCES (1) Falkner B, Kushner H, Onesti G, Angelakos ET: Cardiovascular characteristics in adolescents who develop essential hypertension. Hypertension. 1981, 3(5):521-527. (2) Wood DL, Sheps SG, Elveback LR, Schringer A: Cold pressor test as a predictor of hypertension. Hypertension. 1984, 6:301-306. (3) Menkes MS, Matthews KA, Krantz DS, et al: Cardiovascular reactivity to the cold pressor test as a predictor of hypertension. Hypertension. 1989, 14:524-530. (4) Rose RJ, Chesney MA: Cardiovascular stress reactivity: A behavior-genetic perspective. Behavior Therapy. 1986, 17:31 4-323. (5) Light KC, Sherwood A, Turner JR: High cardiovascular reactivity to stress: A predictor of later hypertension development. In Turner JR, Sherwood A, Light KC (eds), Individual Differences in Cardiovascular Response to Stress: Applications to Models of Cardiovascular Disease. New York: Plenum, 1992, 281-293. (6) Murphy JK, Alpert BS, Moes DM, Somes GW: Race and cardiovascular reactivity: A neglected relationship. Hypertension. 1986, 8:1075-1083. (7) Murphy JK, Alpert BS, Walker SS, Willey ES: Race'and cardiovascular reactivity: A replication. Hypertenszon. 1988, 11:308311.
212
A N N A L S OF B E H A V I O R A L M E D I C I N E
(8) Murphy JK, AIpert BS, Willey ES, Somes GW: Cardiovascular reactivity to psychological stress in healthy children. Psychophysiology. 1988, 25:144-152. (9) Treiber FA, Musante L, Braden D, et al: Racial differences in hemodynamic responses to the cold face stimulus in children and adults. Psychosomatic Medicine. 1990, 52:286-296. (10) Saab PG, Schneiderman N: Biobehavioral stressors, laboratory investigation, and the risk of hypertension. In Blascovich J, Katkin ES (eds), Cardiovascular Reactivity to PsychologicalStress and Disease. Washington, DC: American Psychological Association, 1993, 49-82. (11) Hunt SC, Hasstedt SJ, Kuida H, ct al: Genetic heritability and common environmental components of resting and stressed blood pressures, lipids, and body mass index in Utah pedigrees and twins. American Journal of Epidemiology. 1989, 129:625-638. (12) Williams RR, Hunt SC, Hasstedt SJ, et al: Are there interactions and relations between genetic and environmental factors predisposing to high blood pressure? Hypertension. 1991, 18:I-29-I-37. (13) Hallback M: Consequence of social isolation on blood pressure, cardiovascular reactivity, and design in spontaneously hypertensive rats. Acta Physiologica Scandinavica. 1975, 93:455-465. (14) Hallhack M, Folkow B: Cardiovascular responses to acute mental "stress" in spontaneously hypertensive rats. Acta Physiologica Scandinavica. 1974, 90:684-698. (15) Schneiderman N, McCabe PM: Psychophysiologic strategies in laboratory research. In Schneiderman N, Weiss SM, Kaufman PG (eds), Handbook of Research Methods in CardiovascularBehavioral Medicine. New York: Plenum, 1989, 349-364. (l 6) Hallback-Norlander M, Lundin S: Background of hyperkinetic circulatory state in young spontaneously hypertensive rats. In Meyer P, Schmitt H (eds), Nervous System and Hypertension. New York: Wiley, 1979. (17) McCarty R, Kopin IJ: Alterations in plasma catecholamines and behavior during acute stress in spontaneously hypertensive and Wistar-Kyoto normotensive rats. Life Science. 1978, 22:997. (18) Falkner B, Onesti G, Angelakos ET, Fernandes M, Langman C: Cardiovascular response to mental stress in normal adolescents with hypertensive parents: Hemodynamics and mental stress in adolescents. Hypertension. 1979, 1:23--30. (19) Alpert BS, Wilson DIC Stress reactivity in childhood and adolescence. In Turner JR, Sherwood A, Light K (eds), Individual Differences in CardiovascularResponse to Stress: Applications to Models of Cardiovascular Disease. New York: Plenum, 1992, 187-201. (20) Matthews KA, Rakaczky CJ: Familial aspects of the Type A behavior pattern and physiologic reactivity to stress. In Schmidt TH,
Wilson et al. Dembroski TM, Blumchen G (eds), Biological and Psychological Factors in CardiovascularDisease. Berlin, Germany: Springer-Verlag, 1986, 228-245. (21) Fredrikson M, Matthews KA: Cardiovascular responses to behavioral stress and hypertension: A meta-analytic review. Annals of Behavioral Medicine. 1990, 12:30-39. (22) Carmelli D, Chesney MA, Ward MM, Rosenman RH: Twin similarity in cardiovascular stress response. Health Psychology. 1985, 4:413-423. (23) Smith TW, Turner CW, Ford MH: Blood pressure reactivity in adult male twins. Health Psychology. 1987, 6(3):209-220. (24) McIlhany ML, Shaffer JW, Hines EH: Heritability of blood pressure: An investigation of 200 pairs of twins using the cold pressor test. Johns Hopkins Medical Journal. 1975, 136:57-64. (25) Rose RJ, Grim CE, Miller JZ: Familial influences on cardiovascular stress reactivity: Studies of normotensive twins. Behavioral Medicine Update. 1984, 6:21-24. (26) Ditto B: Sibling similarities in cardiovascular reactivity to stress. Psychophysiology. 1987, 24(3):353-360. (27) Matthews KA, Manuck SB, Stoney CM, et al: Familial aggregation of blood pressure and heart rate response during behavioral stress. Psychosomatic Medicine. 1988, 50:341-352. (28) U.S. Department of Health and Human Services: Cardiovascular and Cerebrovascular Disease: Report to the Secretary's Task Force on Black and Minority Health. Washington, DC: U.S. Department of Health and Human Services, 1986, 267-274. (29) Anderson NB, McNeilly M, Myers H: Toward understanding race differences in autonomic reactivity: A proposed contextual model. In Turner JR, Sherwood A, Light KC (eds), Individual Differences in Cardiovascular Response to Stress. New York: Plenum, 1992, 125-145. (30) Tanner JM: Normal growth and techniques of growth assessment. Clinics in Endocrinology and Metabolism. 1986, 15:411-451. (31) U.S. Department of Health and Human Services: Report of the Second Task Force on Blood Pressure Control in Children. Washington, DC: U.S. Government Printing Office, 1987. (32) Koch GG: Intraclass correlation. In Kotz S, Johnson NL (eds), Encyclopedia of Statistical Sciences (Vol. 4). New York: John Wiley & Sons, 1983, 212-217. (33) Treiber FA, Musante L, Braden D, et al: Racial differences in hemodynamic responses to the cold face stimulus in children and adults. Psychosomatic Medicine. 1990, 52:286-296. (34) Treiber FA, Davis H, Musante L, et al: Ethnicity, gender, family history of myocardial infarction, and hemodynamic responses to laboratory stressors in children. Health Psychology. 1993, 1:6-15.
