Published online before print March 9, 2009, doi: 10.1161/HYPERTENSIONAHA.108.126227
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 2009;53:819.)
© 2009 American Heart Association, Inc.
Original Articles |
Heightened Resting Neural Activity Predicts Exaggerated Stressor-Evoked Blood Pressure Reactivity
From the Department of Psychiatry (P.J.G., L.K.S., A.M.R., I.C.C.), University of Pittsburgh, Pa; Department of Psychiatry (H.D.C.), Brighton and Sussex Medical School, University of Sussex Campus, Falmer, Brighton, United Kingdom; and the Center for Functional Neuroimaging (J.W.), University of Pennsylvania, Philadelphia.
Correspondence to Peter J. Gianaros, Department of Psychiatry, University of Pittsburgh, 3811 O'Hara St, Pittsburgh, PA 15213. E-mail gianarospj{at}upmc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Individuals who express relatively large-magnitude or "exaggerated" blood pressure (BP) reactions to behavioral stressors are presumably at increased risk for cardiovascular disease. As shown by recent neuroimaging studies, individuals who express exaggerated stressor-evoked BP reactivity also express heightened neural activity in corticolimbic brain areas that centrally regulate the cardiovascular system. These studies, however, have exclusively examined BP reactivity and concomitant neural activity during stressor exposure. If exaggerated BP reactivity originates in part from a centrally regulated and dispositional cardiovascular response tendency, then heightened resting (prestressor) corticolimbic activity may predict the subsequent expression of exaggerated stressor-evoked BP reactivity. To test this hypothesis, perfusion MRI was used to quantify resting regional cerebral blood flow (an indirect metabolic measure of neural activity) in men (n=19) and women (n=20) aged 20 to 37 years who subsequently performed cognitive stressor tasks to evoke BP reactivity. Individuals who expressed larger task-induced rises in systolic and diastolic BP also expressed higher resting regional cerebral blood flow in 4 functionally related corticolimbic areas: the dorsal and perigenual anterior cingulate, medial prefrontal, and insular cortices. Specifically, resting regional cerebral blood flow in these areas accounted, respectively, for 40% and 31% of the variance in systolic (P=0.001) and diastolic (P=0.008) BP reactivity, after accounting for total resting cerebral blood flow, resting BP, task performance, and task-related ratings of unpleasantness, arousal, and perceived psychological control. Heightened resting corticolimbic activity may represent a neurobiological correlate of an individual’s predisposition for exaggerated stressor-evoked BP reactivity and possibly related cardiovascular risk.
Key Words: anterior cingulate cortex • blood pressure reactivity • individual differences • insula • medial prefrontal cortex • stress
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Individuals who express relatively large-magnitude or exaggerated blood pressure (BP) reactions to behavioral stressors are at moderately increased risk for hypertension,1,2 stroke,3 ventricular hypertrophy,4 preclinical atherosclerosis,5 and myocardial infarction.6,7 Although it is uncertain whether exaggerated stressor-evoked BP reactions are involved causally in the pathophysiology of these precursors and end points of cardiovascular disease,8 there is psychometric and genetic evidence that an individual’s tendency to express exaggerated stressor-evoked BP reactivity is a reliable9,10 and partly heritable11,12 dispositional cardiovascular stress response attribute. In parallel, there is human neuroimaging evidence that stressor-evoked BP reactions are associated with patterns of neural activity in corticolimbic brain areas that centrally regulate the cardiovascular system. More precisely, stressor-evoked BP reactions have been associated reliably with activity in the anterior cingulate cortex (ACC), medial prefrontal cortex (mPFC), and insula.13–18 Together, these corticolimbic areas can regulate stressor-evoked BP reactions by modulating autonomic and neurohormonal outflow to the myocardium and vasculature via functional neural connections with one another and with subcortical cell groups (eg, amygdala, pontine, and medullary nuclei) that are more proximally involved in cardiovascular control.19–21 In addition, individuals who express exaggerated stressor-evoked BP reactivity have been found to also express heightened stressor-evoked activity in the ACC, mPFC, and insula, broadly suggesting that these particular corticolimbic areas may represent components of a neural circuitry that is functionally involved in the dispositional expression of individual differences in cardiovascular stress reactivity.17,18
To date, however, neuroimaging studies of stressor-evoked BP reactivity have exclusively examined concomitant changes in BP and corticolimbic activity during stressor exposure. Critically, if individual differences in BP reactivity originate in part from a centrally regulated and dispositional cardiovascular stress response tendency, then relatively heightened resting (prestressor) corticolimbic activity may directly predict the subsequent expression of comparatively larger-magnitude (exaggerated) stressor-evoked BP reactions across individuals. To test this hypothesis, perfusion MRI22 was used to quantify resting regional cerebral blood flow (CBF [rCBF]; an indirect metabolic measure of neural activity) in young adults who subsequently performed cognitive stressor tasks designed to evoke BP reactivity. First tested was whether relatively heightened resting rCBF in corticolimbic areas would predict (prospectively) comparatively exaggerated BP reactions to the stressor tasks across individuals. Next tested was whether any prospective associations between resting rCBF and BP reactivity would persist after statistically accounting for plausible confounders, namely, total resting CBF, resting BP, task-related performance, and subjective ratings of task-related unpleasantness, arousal, and perceived psychological control. Last tested was whether the corticolimbic areas in which resting rCBF predicted BP reactivity across individuals would exhibit intercorrelations in their time-varying oscillations in the blood-oxygen level–dependent (BOLD) signal, putative indicators of coherent variation in metabolically linked neural activity.23 Such intercorrelations would provide evidence for so-called resting-state functional connectivity24 between specific corticolimbic areas that may partly constitute a neural circuitry for the dispositional expression of individual differences in stressor-evoked BP reactivity.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Participants
Participants were 20 men (mean age±SD: 24.8±5.1 years) and 20 women (mean age±SD: 23.8±3.8 years). All were right-handed, and none had any of the following: (1) a history of cardiovascular disease (including hypertension, stroke, myocardial infarction, congestive heart failure, and atrial and ventricular arrhythmias); (2) previous cardiovascular surgery (including coronary bypass, carotid artery, or peripheral vascular surgery); (3) a history of cancer, a chronic kidney or liver condition, type I or II diabetes mellitus, or any pulmonary or respiratory disease; (4) any current or previous self-reported psychiatric diagnosis of a substance abuse or mood disorder; (5) previous cerebrovascular trauma involving loss of consciousness; (6) previous neurosurgery or history of neurological conditions; (7) pregnancy (verified by urine test in females); (8) color blindness; (9) claustrophobia; (10) metallic implants; or (11) a self-reported history of using any psychotropic, lipid-lowering, or cardiovascular medications.
The participants’ average seated resting BP was 117/66 mm Hg (±10/9 SD), as determined by the mean of the last 2 of 3 BP readings obtained with an oscillometric device (Critikon Dinamap 8100, Johnson & Johnson) and taken 2 minutes apart after a 20-minute acclimation period before MRI testing. Data from 1 man were excluded because of excessive neuroimaging artifacts attributed to head movements. Thus, results herein are for the remaining 39 participants. Participants gave informed consent to study protocols, approved by the University of Pittsburgh Institutional Review Board. Supplemental information about participant characteristics and screening methods are provided online in Table S1 (please see http://hyper.ahajournals.org).
Study Protocols
Participants abstained from eating, exercising, and consuming caffeinated and tobacco products for 3 hours and drinking alcoholic beverages for 12 hours before testing. At testing, participants underwent a screening interview followed by protocols to assess anthropometric measures, demographic information, and seated BP. Participants then underwent an MRI protocol. For this protocol, participants were fitted with a BP cuff matched to arm size, inserted into the MRI scanner, and asked to rest for 
20 minutes; 12 to 15 minutes later, participants completed 2 stressor tasks while in the scanner.
Neuroimaging Data Acquisition
Neuroimaging data were acquired on a 3T Trio TIM whole-body scanner (Siemens), equipped with a 12-channel, phased-array head coil. Resting perfusion images were acquired with a pulsed arterial spin-labeling sequence. For this sequence, interleaved perfusion images with and without arterial spin labeling were obtained over a 5-minute, 28-second period using gradient-echo echo-planar imaging. The pulsed arterial spin-labeling sequence used a flow-sensitive alternating inversion recovery method,25 specifically applying a saturation pulse 700 ms after an inversion pulse. To reduce transit artifact, a 1000-ms delay separated the end of the labeling pulse and the time of image acquisition. Resting perfusion image acquisition parameters were: field of view: 240x240 mm; matrix size: 64x64 mm; repetition time: 4 seconds; echo time: 18 ms; and flip angle: 90°. Twenty-one sections (5 mm thick, 1 mm gap) were acquired sequentially in an inferior-to-superior direction, yielding 80 total perfusion images (40 labeled and 40 unlabeled; 2 initial discarded images allowing for magnetic equilibration). Resting BOLD images used to compute functional connectivity measures were acquired over a 5-minute, 6-second period with a gradient-echo echo-planar imaging sequence using the following parameters: field of view: 205x205 mm; matrix size: 64x64 mm; repetition time: 2 seconds; echo time: 28 ms; and flip angle: 90°. Thirty-nine sections (3-mm thick, no gap) were obtained sequentially in an inferior-to-superior direction, yielding 150 BOLD images (3 initial discarded images allowing for magnetic equilibration). For spatial coregistration of resting perfusion and BOLD images, T1-weighted 3D magnetization-prepared rapid gradient echo neuroanatomical images were acquired over 7 minutes, 17 seconds by these parameters: field of view: 256x208 mm; matrix size: 256x208 mm; repetition time: 2100 ms; inversion time: 1100 ms; echo time: 3.29 ms; and flip angle: 8° (192 sections, 1-mm thick, no gap).
Stressor Tasks
After the resting period, participants completed 2 counterbalanced stressor tasks designed to evoke BP reactivity: a modified Stroop color-word interference task18 and a modified multisource interference task.26 The tasks were separated by a 10- to 12-minute recovery period, during which subjective ratings of the first task were obtained (see below). Each task lasted 9 minutes, 20 seconds and was composed of trials defining 2 alternating conditions, a less demanding congruent condition and a more demanding incongruent condition. The congruent and incongruent conditions lasted 52 to 60 seconds and were preceded by a 10- to 17-second period during which participants fixated on a cross-hair. Briefly, the congruent and incongruent conditions of both tasks were matched on motor response requirements and visual stimulus characteristics. In addition, the incongruent condition of each task was performance titrated, such that task accuracy was adaptively maintained at 50% within and between individuals. In this way, task engagement and performance were experimentally approximated across participants. Supplemental task details and trial illustrations and provided online in Figure S1.
Task Accuracy and Subjective Ratings
Task performance (accuracy) was computed as the percentage of trials correctly completed. Posthoc, we verified that mean accuracy during the incongruent condition of each task was titrated across participants to 55.4% (±6.8% SD), as compared with 89.8% (±3.9% SD) during the congruent condition, t38=34.8 (P<0.001). In conjunction with this performance titration, mean response times to trials delivered during the incongruent condition compared with the congruent condition of each task were slowed by 465.8 ms (±152.7 ms SD; t38=19.1; P<0.001).
To assess ratings of valence (1: very unhappy; 9: very happy), arousal (1: very calm; 9: very aroused), and perceived control (1: very little control; 9: very much control), participants completed a modified self-assessment manikin scale27 after the resting (prestressor) period and each task. Supplemental summaries of task accuracy and ratings are provided online in Figure S2. For the purpose of ancillary analyses, participants also completed inventories before testing to assess dispositional attributes of negative emotionality (trait anxiety and hostility) and recent levels of perceived life stress, which could plausibly covary with individual differences in stressor-evoked BP reactivity; however, scores on these inventories did not correlate significantly with resting BP or with stressor-evoked BP reactivity in this sample (Table S1). Hence, scores on these inventories were omitted from further analyses.
BP Measurement
In the MRI scanner, participant BP was measured during the resting (prestressor) and stressor task periods from the brachial artery of the nondominant (left) arm, which was not used for task responding. BP measurements were taken with an oscillometric device (Multigas 9500, MedRad, Inc), set to inflate every 2.5 minutes during the resting period and once during each condition of the Stroop task and multisource interference task. To compute resting BP, the final 3 measurements were averaged. To compute task-related BP, measurements from the demanding incongruent condition of the Stroop task and multisource interference task were averaged. The incongruent condition (minus) resting BP difference score was used to compute BP reactivity following previous work.17,18 In this sample, men and women did not differ in resting or reactivity measures of systolic BP (SBP) or diastolic BP (DBP; t values: 
1.6; P
0.11). In addition, across individuals, SBP and DBP changes from the resting period to the incongruent conditions of the Stroop task and multisource interference task were correlated (
SBP r=0.70, P<0.001; DBP r=0.73, P<0.001), indicating that the tasks evoked reliable individual differences in BP reactivity. Following previous guidelines,10 task-averaged SBP and DBP reactivity scores were used for subsequent analyses.
Preprocessing of Neuroimaging Data
Resting perfusion images were preprocessed with computational routines implemented in Statistical Parametric Mapping software (SPM2, Wellcome Trust Centre for Neuroimaging). For preprocessing, perfusion images were realigned to the first image of the series, coregistered to each participant’s magnetization-prepared rapid gradient echo image, spatially normalized to the International Consortium for Brain Mapping 152 template (Montreal Neurological Institute), and resliced to an isotropic voxel size of 3 mm3. Images were then smoothed with a 12-mm full-width at half-maximum isotropic Gaussian kernel, after which the 40 labeled and 40 unlabeled perfusion images were submitted to pairwise subtraction. Subtraction images were converted to an absolute CBF image series using a validated algorithm.28 This perfusion series was then averaged, generating for each individual a single resting voxel-wise rCBF image and a total CBF value, both in units of milliliters per 100 grams per minute.
Resting BOLD images were preprocessed using Statistical Parametric Mapping software (SPM5; Wellcome Trust Centre for Neuroimaging). As for perfusion images, BOLD images were realigned to the first image of the series, coregistered to each participant’s magnetization-prepared rapid gradient echo image, normalized to the Montreal Neurological Institute 152 template, and smoothed with a 6-mm full-width at half-maximum isotropic Gaussian kernel. After preprocessing, a BOLD signal time series was extracted from 4 empirically determined regions of interest (ROIs) for each individual, according to methods detailed previously.18 Specifically, a time series was extracted from the mean BOLD signal of all of the voxels in a 6-mm sphere surrounding the Montreal Neurological Institute coordinates identified in the voxel-wise regression analyses of resting perfusion (as quantified by rCBF) and stressor-evoked BP reactivity reported below. These regions (shown in Figure 1) included the dorsal ACC (dACC; x, y, z Montreal Neurological Institute coordinates in millimeters: –6, 27, 33), perigenual ACC (pACC; –9, 54, 12), mPFC (mPFC; 24, 51, 18), and insula (42, –12, 12). Each time series extracted from each region for each participant was mean centered, drift corrected, and inspected for outliers. Any values >3 SD of the series mean were replaced by averaging 2 surrounding values. Each outlier-corrected time series was then band-pass filtered from 0.01 to 0.10 Hz to remove nonneural sources of noise using a linear-phase finite impulse-response Hamming filter of length 51 (102 seconds, based on the 2-second BOLD signal acquisition repetition time). Each filtered time series was then submitted to a cross-correlation routine described below. A supplemental description of this routine is provided online in Figure S3.
|
Data Analysis
To test whether resting (prestressor) levels of perfusion predicted subsequent stressor-evoked BP reactions across individuals, individual rCBF images were first submitted to a voxel-wise multiple regression model in SPM5. In the model, stressor-evoked SBP reactivity (computed by the average SBP change from the prestressor period to the incongruent Stroop task and multisource interference task conditions) was entered as a regressor of interest, with resting SBP entered as a covariate to account for initial BP. In the model, voxel-wise rCBF images were scaled to each individual’s total resting CBF value. To correct for multiple voxel-wise statistical testing, we maintained a whole-brain significance threshold of P0.005 with a cluster (k) extent of 20 contiguous voxels (3x3x3 mm3).29
After testing whether resting perfusion predicted stressor-evoked SBP reactivity, 2 hierarchical regression models were executed outside of SPM5 using SPSS software (version 16, SPSS Inc). These hierarchical models tested specifically whether the prospective associations between resting perfusion and stressor-evoked SBP (model 1) or DBP (model 2) reactivity would persist after accounting for the potential confounding influence of individual differences in total resting CBF, resting BP, task accuracy, and ratings of task-related valence, arousal, and control. For the models, we extracted the mean rCBF values from the 4 ROIs in which resting perfusion predicted SBP reactivity in the voxel-wise SPM5 regression analysis described above. We then entered these extracted rCBF values as a set of predictors in the second step of the 2 models. In step 1 of both models, we entered total CBF, resting SBP (model 1) or resting DBP (model 2), task-averaged accuracy, and task-averaged ratings of valence, arousal, and control. The dependent variable for the models was the task-averaged SBP (model 1) or DBP (model 2) reactivity value for each individual. The unique percentage of variance in BP reactivity explained by the set of extracted rCBF values was evaluated by the R2 from step 1 to step 2.
To determine the intercorrelations in the resting BOLD signal time series across the 4 ROIs, we executed a cross-correlation routine detailed online in Figure S3. Briefly, pairwise cross-correlation coefficients (r values) were computed across the 4 ROI BOLD signal time series for each individual. The cross-correlation coefficients were then transformed to Fisher’s z values, averaged across individuals, and used to compute 99% CIs using a bootstrapping method. This method permitted a test of whether the averaged z values differed from 0 across individuals, indicating so-called functional connectivity between the ROIs. To aid interpretability, the averaged z values and 99% CIs were transformed back to r values for illustration in Figure 2.
|
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Task Ratings and BP Reactivity
The stressor tasks elicited moderate levels of subjective distress sufficient to evoke individual differences in BP reactivity. Specifically, using 9-point rating scales, participants reported that they felt less happy, more aroused, and less in control, while performing the tasks, as compared with the resting period (t values for all task versus resting comparisons:
3.7; P0.001; Figure S2). In addition, both SBP and DBP increased on average while participants performed the tasks, as compared with the resting period (t values for all of the task versus resting comparisons: 3.0; P0.006; Figure S4).
Resting rCBF and BP Reactivity
In a voxel-wise, multiple regression analysis, larger-magnitude, stressor-evoked SBP reactions were predicted across individuals by relatively higher levels of resting rCBF in the dorsal and perigenual areas of the left ACC (dACC and pACC), the dorsal area of the right mPFC (mPFC), and the posterior area of the right insula (Figure 1). Subsequent hierarchical multiple regression analyses further demonstrated that resting rCBF values extracted from the dACC, pACC, mPFC, and insula continued to predict both stressor-evoked SBP and DBP reactivity after accounting for total resting CBF, resting BP, task accuracy, and task ratings of valence, arousal, and control. Specifically, in step 1 of a 2-step hierarchical regression analysis, total CBF, resting SBP, task accuracy, and task ratings accounted for 14% of the variance in SBP reactivity (F6,32=0.88; R2 adjusted=–0.02; P=0.52). In another 2-step regression analysis, total CBF, resting DBP, task accuracy, and task ratings accounted for 18% of the variance in DBP reactivity (F6,32=1.2; R2 adjusted=0.03; P=0.33). In step 2 of these regression analyses, resting rCBF values extracted from the dACC, pACC, mPFC, and insula accounted collectively for remaining variance in both SBP (F4,28=6.1; R2=0.40; P=0.001) and DBP (F4,28=4.2; R2=0.31; P=0.008) reactivity.
In addition, exploratory regression analyses including the same step 1 variables as above demonstrated that resting rCBF values extracted from the dACC, pACC, mPFC, and insula accounted individually for unique variance in both SBP and DBP reactivity (Table S2). Across these exploratory analyses, however, only resting rCBF extracted from the insula did not account for unique variance in DBP reactivity after family wise error rate correction for multiple (posthoc) statistical testing (P family wise error=0.08). By contrast, resting rCBF from the insula did account for unique variance in SBP reactivity; and, resting rCBF in all of the remaining areas accounted for unique variance in both SBP and DBP reactivity after family wise error correction (Table S2).
Here, we recognize that, during the resting period, participants may have been apprehensive about the forthcoming stressor tasks, leading to anticipatory unpleasantness or arousal and possibly increased BP reactivity or resting rCBF. This is unlikely, however, because resting ratings of valence and arousal did not correlate significantly with SBP or DBP reactivity or with resting rCBF in the dACC, pACC, mPFC, or insula (r values ranged from –0.14 to 0.25; P0.12). In addition, resting SBP and DBP did not correlate significantly with resting rCBF in the dACC, pACC, mPFC, or insula (r values: –0.21 to 0.18; P0.20).
Functional Connectivity Between Areas Where Resting rCBF Predicted BP Reactivity
Across individuals, the dACC, pACC, mPFC, and insula exhibited moderate and directionally positive cross-correlations in their time-varying BOLD signal fluctuations, indicating resting "functional connectivity" between these areas. Figure 2 shows the aggregate cross-correlation coefficients across these areas, along with their 99% bootstrapped CIs, generated by 5000 Monte Carlo simulations.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The central finding of this study was that relatively heightened resting (prestressor) rCBF in the dACC, pACC, mPFC, and insula uniquely predicted comparatively larger-magnitude (exaggerated) BP reactions to subsequently completed cognitive stressor tasks across individuals (Figure 1). In addition, these corticolimbic areas exhibited resting intercorrelations in their time-varying BOLD signal oscillations, indicating that they exhibited so-called functional connectivity24 with one another (Figure 2). As such, the present study provides further evidence that the dACC, pACC, mPFC, and insula could represent components of a neural circuitry that is functionally involved in the dispositional expression of individual differences in stressor-evoked BP reactivity.
An individual’s tendency to express exaggerated stressor-evoked BP reactivity has long been viewed as a dispositional attribute that may be linked to the pathophysiology of cardiovascular disease.30 Heritability studies11,12 further support the converging view that individual differences in stressor-evoked BP reactivity arise in part from genetic factors, which may involve the familial transmission of allelic polymorphisms that modify myocardial and vascular sensitivity to centrally regulated patterns of peripheral efferent autonomic and neurohormonal outflow.31–33 A largely unsupported view, however, is that individual differences in stressor-evoked BP reactivity are solely attributable to interindividual variation in self-reported levels of distress or negative emotionality, because subjective ratings of stress, emotionality, and arousal are weakly (and rarely significantly) associated with BP reactivity in laboratory studies.10,34 In agreement, individual differences in stressor-evoked BP reactivity were not significantly correlated with task-related unpleasantness, arousal, or perceived control ratings in this study, nor were the prospective associations between resting corticolimbic activity and stressor-evoked BP reactivity accounted for by such ratings. Finally, self-reported ratings of dispositional negative emotionality (trait anxiety and hostility) and recent levels of perceived life stress were also not significantly correlated with stressor-evoked BP reactivity in this sample (Table S1). Hence, one interpretation of our findings in synthesis with previous work is that a tendency to express exaggerated stressor-evoked BP reactivity could arise in part from interacting central and peripheral neurobiological factors that are subject to genetic modification and relatively independent of subjective (consciously reportable) states of stress, arousal, or emotionality. In this regard, the present findings and those from previous neuroimaging studies may specifically suggest that neurobiological measures, including rCBF and functional connectivity measures, may have advantages over subjective rating measures in understanding the origins of individual differences in stressor-evoked BP reactivity.
To elaborate, it is noteworthy that acute stressors have been demonstrated in previous neuroimaging studies to evoke patterns of neural activity in the dACC, pACC, mPFC, and insula that covary reliably with concomitant changes in BP.14,16–18 There is cumulative human and nonhuman animal evidence that these corticolimbic areas are anatomically networked and that they are instrumental for regulating autonomic, neurohormonal, and cardiovascular reactions to stressors, presumably in support of adaptive behavioral action (eg, the canonical "fight-or-flight" response).13,19,20 In previous studies, however, measures of corticolimbic activity and BP reactivity were examined concomitantly and exclusively during stressor exposure. As a result, it was unknown whether resting (prestressor) patterns of corticolimbic activity would prospectively predict individual differences in subsequently evoked BP reactivity. Thus, extending previous work, the present findings indicate that corticolimbic activity relates to individual differences in stressor-evoked BP reactivity, even when this activity is measured before stressor exposure. In extension, the cross-correlational (connectivity) findings illustrated in Figure 2 suggest that activity in the dACC, pACC, mPFC, and insula is likely to be functionally integrated within a broader neural circuitry. This notion is consistent with invasive human and nonhuman animal studies that have demonstrated functional and anatomic connections between these corticolimbic areas, which are important for autonomic, neurohormonal, and cardiovascular regulation.19,21,35–40 In addition, it is notable that 2 recent neuroimaging studies18,41 have specifically demonstrated that exaggerated stressor-evoked BP reactivity and increased carotid intima-media thickness, an indicator of preclinical atherosclerosis linked to stressor-evoked BP reactivity,8 are both associated with a dysregulated pattern of functional connectivity between the pACC and one of its subcortical projection sites involved in cardiovascular regulation: the amygdala. In view of these recent studies and the present study, we are currently testing whether particular patterns of resting state connectivity (eg, time-lagged, directional, or multivariate) exhibited among the pACC, the other corticolimbic areas identified here, and their subcortical targets specifically predict individual differences in stressor-evoked BP reactivity or covary with associated indicators of cardiovascular risk.
Although novel, we appreciate both inferential limitations and unresolved questions raised by the present study. First, by testing younger men and women in good cardiovascular health, extrapolations to older and more heterogeneous samples are precluded. Second, by measuring resting perfusion on just one occasion that involved later stress testing, conjectures that relatively heightened resting corticolimbic activity may predispose toward exaggerated stressor-evoked BP reactivity remain tenuous, particularly until individual differences in resting corticolimbic activity (eg, perfusion and functional connectivity) are shown to be reproducible over multiple occasions that do and do not involve later stress testing. Third, by using a study design that did not include assessments of family history of hypertension or cardiovascular disease or other familial assessments permitting inferences regarding genetic influences on our findings, predictions about whether resting corticolimbic activity predicts BP reactivity in part via heritable and risk-related factors are largely premature. Finally, by not measuring preclinical cardiovascular disease indicators or cardiovascular risk factors that have been associated with stressor-evoked BP reactivity, assumptions that our findings are relevant to clinical risk remain untested.
Perspectives
To end, individuals expressing relatively exaggerated stressor-evoked BP reactivity also expressed heightened resting rCBF in the cingulate, medial prefrontal, and insular cortices, corticolimbic brain areas that also exhibited functional connectivity with each other. Collectively, these findings could be interpreted from a neurobiological perspective, which views resting measures of rCBF42 and functional connectivity23 as corresponding in part with the level of metabolic "preparedness" of functionally related brain areas to respond to future cognitive, emotional, or otherwise behaviorally salient stimuli. In particular, resting rCBF in the cingulate, medial prefrontal, and insular cortices (along with the degree of their functional connectivity) may partially correspond with their "preparedness" to respond to behaviorally salient stressor stimuli and communicate with subcortical cell groups that are involved in regulating BP. These notions could be tested empirically by determining whether and how measures of resting corticolimbic activity and connectivity relate to patterns of neural activity elicited directly by behaviorally salient stressors that evoke BP reactivity. Arguably, testing such notions may aid in further explicating the neurobiological factors and neural circuitries that predispose some individuals toward exaggerated stressor-evoked BP reactivity and perhaps related cardiovascular risk.
|
Acknowledgments |
|---|
We thank Qi Shi for his technical assistance and Dr Jeanne McCaffery for her constructive comments on a draft of this article.
Sources of Funding
This work was supported by National Institutes of Health grants K01-MH070616 and R01-HL089850, the Pittsburgh Mind-Body Center (HL076852/076858), and a Commonwealth Universal Research Enhancement Grant from the Pennsylvania Department of Health.
Disclosures
None.
Received November 17, 2008; first decision December 9, 2008; accepted February 13, 2009.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Knox SS, Hausdorff J, Markovitz JH. Reactivity as a predictor of subsequent blood pressure: racial differences in the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Hypertension. 2002; 40: 914–919.
2. Matthews KA, Woodall KL, Allen MT. Cardiovascular reactivity to stress predicts future blood pressure status. Hypertension. 1993; 22: 479–485.
3. Everson SA, Lynch JW, Kaplan GA, Lakka TA, Sivenius J, Salonen JT. Stress-induced blood pressure reactivity and incident stroke in middle-aged men. Stroke. 2001; 32: 1263–1270.
4. Kapuku G, Treiber FA, Davis HC, Harshfield GA, Cook BB, Mensah GA. Hemodynamic function at rest, during acute stress, and in the field: predictors of cardiac structure and function 2 years later in youth. Hypertension. 1999; 34: 1026–1031.
5. Jennings JR, Kamarck TW, Everson-Rose SA, Kaplan GA, Manuck SB, Salonen JT. Exaggerated blood pressure responses during mental stress are prospectively related to enhanced carotid atherosclerosis in middle-aged Finnish men. Circulation. 2004; 110: 2198–2203.
6. Krantz DS, Santiago HT, Kop WJ, Bairey-Merz CN, Rozanski A, Gottdiener JS. Prognostic value of mental stress testing in coronary artery disease. Am J Cardiol. 1999; 84: 1292–1297.[CrossRef][Medline] [Order article via Infotrieve]
7. Manuck SB, Olsson G, Hjemdahl P, Rehnqvist N. Does cardiovascular reactivity to mental stress have prognostic value in postinfarction patients? A pilot study. Psychosom Med. 1992; 54: 102–108.
8. Treiber FA, Kamarck T, Schneiderman N, Sheffield D, Kapuku G, Taylor T. Cardiovascular reactivity and development of preclinical and clinical disease states. Psychosom Med. 2003; 65: 46–62.
9. Swain A, Suls J. Reproducibility of blood pressure and heart rate reactivity: a meta-analysis. Psychophysiology. 1996; 33: 162–174.[Medline] [Order article via Infotrieve]
10. Kamarck TW, Lovallo WR. Cardiovascular reactivity to psychological challenge: conceptual and measurement considerations. Psychosom Med. 2003; 65: 9–21.
11. De Geus EJ, Kupper N, Boomsma DI, Snieder H. Bivariate genetic modeling of cardiovascular stress reactivity: does stress uncover genetic variance? Psychosom Med. 2007; 69: 356–364.
12. Turner JR, Hewitt JK. Twin studies of cardiovascular response to psychological challenge: a review and suggested future directions. Ann Beh Med. 1992; 14: 12–20.
13. Critchley HD. Neural mechanisms of autonomic, affective, and cognitive integration. J Comp Neurol. 2005; 493: 154–166.[CrossRef][Medline] [Order article via Infotrieve]
14. Critchley HD, Corfield DR, Chandler MP, Mathias CJ, Dolan RJ. Cerebral correlates of autonomic cardiovascular arousal: a functional neuroimaging investigation in humans. J Physiol. 2000; 523: 259–270.
15. Critchley HD, Mathias CJ, Josephs O, O'Doherty J, Zanini S, Dewar BK, Cipolotti L, Shallice T, Dolan RJ. Human cingulate cortex and autonomic control: converging neuroimaging and clinical evidence. Brain. 2003; 126: 2139–2152.
16. Gianaros PJ, Derbyshire SW, May JC, Siegle GJ, Gamalo MA, Jennings JR. Anterior cingulate activity correlates with blood pressure during stress. Psychophysiology. 2005; 42: 627–635.[CrossRef][Medline] [Order article via Infotrieve]
17. Gianaros PJ, Jennings JR, Sheu LK, Derbyshire SW, Matthews KA. Heightened functional neural activation to psychological stress covaries with exaggerated blood pressure reactivity. Hypertension. 2007; 49: 134–140.
18. Gianaros PJ, Sheu LK, Matthews KA, Jennings JR, Manuck SB, Hariri AR. Individual differences in stressor-evoked blood pressure reactivity vary with activation, volume, and functional connectivity of the amygdala. J Neurosci. 2008; 28: 990–999.
19. Dampney RA. Functional organization of central pathways regulating the cardiovascular system. Physiol Rev. 1994; 74: 323–364.
20. Loewy AD. Forebrain nuclei involved in autonomic control. Prog Brain Res. 1991; 87: 253–268.[Medline] [Order article via Infotrieve]
21. Neafsey EJ. Prefrontal cortical control of the autonomic nervous system: anatomical and physiological observations. Prog Brain Res. 1990; 85: 147–165.[Medline] [Order article via Infotrieve]
22. Detre JA, Wang J. Technical aspects and utility of fMRI using BOLD and ASL. Clin Neurophysiol. 2002; 113: 621–634.[CrossRef][Medline] [Order article via Infotrieve]
23. Fox MD, Raichle ME. Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging. Nat Rev Neurosci. 2007; 8: 700–711.[CrossRef][Medline] [Order article via Infotrieve]
24. Rogers BP, Morgan VL, Newton AT, Gore JC. Assessing functional connectivity in the human brain by fMRI. Magn Reson Imaging. 2007; 25: 1347–1357.[CrossRef][Medline] [Order article via Infotrieve]
25. Kim SG. Quantification of relative cerebral blood flow change by flow-sensitive alternating inversion recovery (FAIR) technique: application to functional mapping. Magn Reson Med. 1995; 34: 293–301.[Medline] [Order article via Infotrieve]
26. Bush G, Shin LM. The Multi-Source Interference Task: an fMRI task that reliably activates the cingulo-frontal-parietal cognitive/attention network. Nat Protoc. 2006; 1: 308–313.[Medline] [Order article via Infotrieve]
27. Bradley MM, Lang PJ. Measuring emotion: the self-assessment manikin and the semantic differential. J Behav Ther Exp Psychiatry. 1994; 25: 49–59.[CrossRef][Medline] [Order article via Infotrieve]
28. Wang J, Aguirre GK, Kimberg DY, Roc AC, Li L, Detre JA. Arterial spin labeling perfusion fMRI with very low task frequency. Magn Reson Med. 2003; 49: 796–802.[CrossRef][Medline] [Order article via Infotrieve]
29. Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC. Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): use of a cluster-size threshold. Magn Reson Med. 1995; 33: 636–647.[Medline] [Order article via Infotrieve]
30. Krantz DS, Manuck SB. Acute psychophysiologic reactivity and risk of cardiovascular disease: a review and methodologic critique. Psychol Bull. 1984; 96: 435–464.[CrossRef][Medline] [Order article via Infotrieve]
31. McCaffery JM, Bleil M, Pogue-Geile MF, Ferrell RE, Manuck SB. Allelic variation in the serotonin transporter gene-linked polymorphic region (5-HTTLPR) and cardiovascular reactivity in young adult male and female twins of European-Am descent. Psychosom Med. 2003; 65: 721–728.
32. McCaffery JM, Pogue-Geile MF, Ferrell RE, Petro N, Manuck SB. Variability within alpha- and beta-adrenoreceptor genes as a predictor of cardiovascular function at rest and in response to mental challenge. J Hypertens. 2002; 20: 1105–1114.[CrossRef][Medline] [Order article via Infotrieve]
33. Li GH, Faulhaber HD, Rosenthal M, Schuster H, Jordan J, Timmermann B, Hoehe MR, Luft FC, Busjahn A. Beta-2 adrenergic receptor gene variations and blood pressure under stress in normal twins. Psychophysiology. 2001; 38: 485–489.[CrossRef][Medline] [Order article via Infotrieve]
34. Feldman PJ, Cohen S, Lepore SJ, Matthews KA, Kamarck TW, Marsland AL. Negative emotions and acute physiological responses to stress. Ann Behav Med. 1999; 21: 216–222, discussion 223–216.[Medline] [Order article via Infotrieve]
35. Buchanan SL, Powell DA. Cingulothalamic and prefrontal control of autonomic function. In: Vogt BA, Gabriel M, eds. Neurobiology of the Cingulate Cortex and Limbic Thalamus: A Comprehensive Handbook. Boston, MA: Birkhauser; 1993: 381–414.
36. Verbene AJM, Owens NC. Cortical modulation of the cardiovascular system. Prog Neurobiol. 1998; 54: 149–168.[CrossRef][Medline] [Order article via Infotrieve]
37. Chiba T, Kayahara T, Nakano K. Efferent projections of infralimbic and prelimbic areas of the medial prefrontal cortex in the Japanese monkey, Macaca fuscata. Brain Res. 2001; 888: 83–101.[CrossRef][Medline] [Order article via Infotrieve]
38. Öngür D, Price J. The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys, and humans. Cereb Cortex. 2000; 10: 206–219.
39. Vogt BA, Pandya DN. Cingulate cortex of the rhesus monkey II: cortical afferents. J Comp Neurol. 1987; 262: 271–289.[CrossRef][Medline] [Order article via Infotrieve]
40. Resstel LB, Correa FM. Involvement of the medial prefrontal cortex in central cardiovascular modulation in the rat. Auton Neurosci. 2006; 126–127: 130–138.
41. Gianaros PJ, Hariri AR, Sheu LK, Muldoon MF, Sutton-Tyrrell K, Manuck SB. Preclinical atherosclerosis covaries with individual differences in reactivity and functional connectivity of the amygdala. Biol Psychiatry. In press.
42. Gur RC, Ragland JD, Reivich M, Greenberg JH, Alavi A, Gur RE. Regional differences in the coupling between resting cerebral blood flow and metabolism may indicate action preparedness as a default state. Cereb Cortex. 2009; 19: 375–382.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||