Evaluation of cerebral autoregulation performance in patients with arterial hypertension on drug treatment

Abstract

Cerebral autoregulation (AR) keeps cerebral blood flow constant despite fluctuations in systemic arterial pressure. The final common AR pathway is made up of vasomotor adjustments of cerebrovascular resistance mediated by arterioles. Structural and functional changes in the arteriolar wall arise with age and systemic arterial hypertension. This study evaluated whether AR is impaired in hypertensive patients and whether this impairment differs with disease control. Three groups of patients were prospectively compared: hypertensive patients under treatment with systolic blood pressure (SBP) <140 and diastolic blood pressure (DBP) <90 mm Hg (n = 54), hypertensive patients under treatment with SBP > 140 or DBP > 90 mm Hg (n = 31), and normotensive volunteers (n = 30). Simultaneous measurements of cerebral blood flow velocity (CBFV) and BP were obtained by digital plethysmography and transcranial Doppler, and the AR index (ARI) was defined according to the step response to spontaneous fluctuations in BP. Compared to the uncontrolled hypertension, the normotensive individuals were younger (age 43.42 ± 11.14, P < .05) and had a lower resistance‐area product (1.17 ± 0.24, P < .05), although age and greater arteriolar stiffness did not affect the CBFV mean of hypertensive patients, whether controlled or uncontrolled (62.85 × 58.49 × 58.30 cm/s, P = .29), most likely because their ARIs were not compromised (5.54 × 5.91 × 5.88, P = .6). Hypertensive patients under treatment, regardless of their BP control, have intact AR capacity.

1. INTRODUCTION

Despite representing only 2% of the total body weight, the human brain at rest uses approximately 15%‐20% of the cardiac output, that is, the same blood volume as the skeletal muscles, which weight 20 times more. ¹ Maintenance and restoration of ionic gradients, membrane potentials, and neurotransmission regulation are the main factors responsible for this high cerebral metabolic demand. ² As a result, several mechanisms act to maintain an adequate supply of oxygen and nutrients to satisfy the high energy consumption of brain tissue. ³ One of these mechanisms is autoregulation (AR), which refers to the ability of the brain to keep its blood flow constant despite oscillations in blood pressure (BP), and it acts through controlling cerebrovascular resistance. ⁴ , ⁵ AR can be evaluated by static or dynamic methods, which reflect, respectively, the overall efficiency and latency of the action of this vasoregulatory system. ⁶

Based on this relationship between BP and cerebral blood flow, several methods have been proposed over the last 30 years to evaluate AR. Conventionally, AR was analyzed in a static manner through changes in BP induced by pharmacological agents or carotid sinus compression. These interventions, however, could not be applied to all patients, especially severely ill or older patients, due to the risk of ischemia. Dynamic methods, including those that used spontaneous BP fluctuations, became the most widely adopted approach for this analysis. ⁷ Transcranial Doppler ultrasound noninvasively quantifies the flow velocity values, reflects the blood flow in the main intracranial arteries, and, when combined with other devices that allow concomitant and continuous BP measurement (eg, the Finapres^®), is one of the most important methods for the dynamic evaluation of AR. ⁸

Systemic arterial hypertension (SAH) is the main treatable risk factor for cerebrovascular disease and is related to structural and functional changes in cerebral hemodynamics. ⁹ Thus, despite the known beneficial effects of antihypertensives on most patients, some who have an impaired autoregulatory capacity ¹⁰ become potentially vulnerable to cerebral hypoperfusion and syncope under simple orthostatic stress. ¹¹

Thus, the present study evaluates whether AR is impaired in hypertensive patients compared to nonhypertensive patients and whether this impairment differs according to disease control.

2. METHOD

This was a prospective observational study conducted between November 2013 and December 2015, at public health service, the Heart Institute (Instituto do Coração – INCOR), São Paulo, Brazil. A total of 153 hypertensive patients followed up at the hypertension outpatient clinic of the institution were recruited. Seventy‐five volunteers with no known clinical diseases who were initially selected based on a manual BP measurement made up the control group. The BP determination was in agreement with the recommendations of the 7th Brazilian Guidelines for Hypertension (2016). ¹²

To assess the impact of hypertension on AR, hypertensive patients were divided into two groups according to previous studies ¹³ : (a) group 2 (controlled hypertension): systolic BP (SBP) <140 and systolic BP (DBP) <90 mm Hg and using one or more antihypertensives; (b) group 3 (uncontrolled hypertension): SBP > 140 or DBP > 90 mm Hg despite the use of antihypertensives. These groups were compared with group 1 (normotension), comprising individuals with SBP < 140 and DBP < 90 mm Hg and with no need for antihypertensives.

Information regarding the demographic and clinical aspects of each participant was obtained through directed anamnesis. The participants were of legal age, did not have diabetes or a history of other known cardiac or cerebrovascular diseases, and did not have carotid stenosis above 50%.

To obtain the cerebral hemodynamic parameters, the participants arrived at the service on a scheduled day and time, always in the afternoon, to undergo transcranial Doppler. The examination was performed in a room with controlled temperature (~25°C) and minimal external noise, and the patient was kept in the supine position with the head elevated to 30°. To measure the cerebral blood flow velocity (CBFV) of the middle cerebral arteries, the transtemporal windows were kept under manual insonation for 3 minutes, one side at a time, at a depth of 50 ± 5 mm, using a 2‐MHz transducer (DWL, Doppler‐BoX, Germany). The BP was continuously recorded noninvasively by digital plethysmography (Finometer^™, Finapres Medical Systems BV, Netherlands), with the arm kept at the same level in relation to the body for placement of the cuff on the middle finger or index finger of the left hand. Recording was only started after matching the plethysmography values with the manual BP measurement.

The simultaneous measurements of CBFV and BP, beat‐to‐beat, obtained during the continuous recording period, were transferred and stored in a computer for later analysis with specific software (Department of Cardiovascular Sciences, University of Leicester, UK). The mean SBP and DBP values obtained during this recording period were used to allocate the patients to the different groups.

All signals were visually inspected to identify artifacts or noise. Patients were excluded due to technical problems, such as the absence of an insonation window or Finometer malfunction. Patients whose records became unusable after the computerized analysis and those diagnosed with hypertension during the continuous BP recording period were also excluded.

Dynamic AR was evaluated using spontaneous fluctuations in mean BP as the input and changes in CBFV as the output. The AR index (ARI) was calculated through the appropriate step response profile, corresponding to one of the 10 existing curve models. Each curve corresponds to an ARI value, ranging from 0 (absence of AR) to 9 (best existing AR response). ⁸

The Shapiro‐Wilk test was used to see if data were normally distributed. Normally distributed data are expressed as the mean and standard deviation with the 95% confidence interval. Analysis of variance (ANOVA) was used with Bonferroni post hoc comparisons among the three study groups. A value of P < .05 was considered statistically significant. Statistical analysis was performed with Statistica Data Miner version 8.0 (Statsoft, Germany).

The study was approved by the local ethics committee, and all participants gave written informed consent.

3. RESULTS

Of the 151 hypertensive patients, 16 were excluded due to technical problems (15 because they did not have an insonation window and one due to Finometer malfunction) and 50 because they had unusable records after computerized analysis (22 controlled hypertensive patients and 28 uncontrolled hypertensive patients).

Among the 75 healthy volunteers, only 30 formed the normotensive group because 32 did not have usable records, five were excluded because they had been diagnosed with hypertension, and eight due to technical problems (Figure 1).

Figure 1.

Trial profile

The clinical and epidemiological characteristics of the participants are shown in Table 1.

Table 1.

Clinical and epidemiological characteristics

	Group 1 (n = 30)	95% IC	Group 2 (n = 54)	95% IC	Group 3 (n = 31)	95% IC	P
Sex, male (no, %)	18 (60%)		32 (59%)		11 (35%)
Age, y (±SD)	43.46 (11.14)	39.3‐47.62	50.57 (11.7)	47.37‐53.77	50.87 (11.7)	46.56‐55.17	.01*
HTN time, y (±SD)			9.4 (6.78)	7.55‐11.25	8.8 (8.29)	5.76‐11.84	.71
Antihypertensive Class
ACEI (no, %)			25 (46%)		6 (19.3%)
ARB (no, %)			18 (33%)		20 (64%)
CCB (no, %)			20 (37%)		13 (41%)
BB (no, %)			18 (33%)		11 (35%)
Vasodilator (no, %)			2 (3.7%)		1 (3.2%)
Diuretics (no, %)			36 (66%)		18 (58%)
Number of drugs, mean (±SD)			2.3 (1.16)		2.5 (1.5)

P value refers of the means among the groups by the 1‐way ANOVA with Bonferroni post hoc comparisons. Group 1 (normotensive); Group 2 (controlled HTN); Group 3 (uncontrolled HTN).

Abbreviations: ACEI, angiotensin‐converting enzyme inhibitors; ARB, angiotensin receptor blockers; BB, beta blockers; CCB, calcium channel blockers; HTN, hypertension.

^*Significance difference (Group 1 vs Group 2 and Group 3).

The normotensive individuals were significantly younger than the two groups of hypertensive patients (P < .05).

In contrast, hypertensive patients with poor BP control were obese compared to normotensive individuals (BMI 30.45 and 27 respectively, P = .01, not shown in table 1), stressing obesity as one of the comorbidities that is usually associated and plays an important role in the pathophysiology of hypertension. ¹⁴

The drug treatment options most used by group 3 were angiotensin II receptor blockers (64.5%) and calcium channel blockers (41.9%). Although there was no difference between the mean number of antihypertensive drugs in use and the time of the hypertension diagnosis (P = .71), uncontrolled hypertensive patients had a mean SBP that was 35.38 mm Hg higher than controlled hypertensive patients had.

Nevertheless, neither hypertension, whether controlled or uncontrolled, nor advanced age had significant effects on AR (P = .6) within the BP limits studied (Table 2).

Table 2.

Hemodynamic parameters ^a

	Group 1 (n = 30)	95% IC	Group 2 (n = 54)	95% IC	Group 3 (n = 31)	95% IC	P
ARI mean (±SD)	5.54 (1.78)	4.87‐6.21	5.91 (1.66)	5.45‐6.36	5.88 (1.53)	5.55‐6.11	.6
CBFV mean (±SD)	62.85 (11.79)	54.98‐67.26	58.49 (12.87)	54.98‐62	58.30 (15.16)	52.73‐63.86	.29
RAP mean (±SD)	1.17 (0.24)	1‐1.27	1.33 (0.42)	1.21‐1.44	1.68 (0.65)	1.44‐1.92	0 *
PI mean (±SD)	1.59 (4.76)	−0.18 to 0.37	0.70 (0.13)	0.66‐0.73	0.78 (0.27)	0.68‐0.88	.25
CrCP mean (±SD)	9.49 (7.19)	6.8‐12.18	9.81 (12.09)	6.51‐13.11	11.38 (11.38)	7‐15.69	.75
SBP mean mm Hg (±SD)	122.56 (11.08)	118.42‐126.7	123.32 (2.12)	119‐127.59	158.70 (3.66)	151.21‐166.18	0 *
DBP mean mm Hg (±SD)	60.08 (10.51)	56.16‐64	61.06 (14.38)	58‐65.89	75.66 (13.26)	70.79‐80.52	0 *
ABP mean mm Hg (±SD)	80.91 (9.75)	77.27‐84.55	82.41 (13.71)	78.67‐86.16	103.34 (13.87)	98.25‐108.42	0 *

Abbreviations: ABP, average blood pressure; ARI, autoregulation index; CBFV, cerebral blood flow velocity; CrCP, critical closing pressure; DBP, diastolic blood pressure; PI, pulsatility index; RAP, resistance‐area product; SBP, systolic blood pressure.

P value refers of the means among the groups by the 1‐way ANOVA with Bonferroni post hoc comparisons. Group 1 (normotensive); Group 2 (controlled HTN); Group 3 (uncontrolled HTN).

^aThe hemodynamic parameters presented here are to the means of the values obtained after the processing of continuous and simultaneous BP and CBFV records.

^*significance difference (Group 3 vs Group 1 and Group 2).

Uncontrolled hypertensive patients had a higher have greater arterial stiffness, as demonstrated by their greater resistance‐area product (RAP), but similar critical closing pressure (CrCP), which assesses the vasomotor reactivity of small cerebral arteries, one of the functional parameters of cerebral hemodynamics, compared to normotensive or controlled hypertensive (P = .75), suggesting that chronic adaptations to elevated BP involve changes in vascular distensibility, rather than increased vascular smooth muscle tone, without causing any significant repercussions on CBFV mean (P = .29).

4. DISCUSSION

AR is important to protect the brain against ischemia or excessive capillary pressure and hyperperfusion. ¹⁵ This regulation is mediated by the action of metabolic vasodilators from hypoxic brain tissue when cerebral perfusion pressure (CPP) is reduced, acting in combination with the myogenic response of vascular smooth muscle cells. For example, the major cerebral arteries and the pial and penetrating arterioles dilate in response to moderate reductions in systemic pressure. In response to the increased cerebral transmural pressure, over half of the reduction in this pressure occurs in the major arteries, which results in approximately 75% attenuation of the CPP before the blood reaches the arterioles, which are responsible for the remaining regulation of CBF and capillary pressure. ¹⁶

Hypertension and age induce structural and functional changes in the cerebral arteriolar wall, such as stiffness, arteriolar narrowing, and reduced endothelial nitric oxide synthesis. This pathophysiological process leads to increased cerebrovascular resistance and higher RAP and is one of the mechanisms by which hypertension can affect cerebral AR, resulting in reduced blood flow and increased cerebral oxygen extraction fractions. ¹⁵ , ¹⁷ Nevertheless, the present study showed that patients with chronic SAH, mainly uncontrolled, despite presenting higher RAP compared to normotensive (P = 0) and older age (P < .05), had no impairment of their CBF, as evidenced by the CBFV mean that was practically the same among the three groups (Table 2), regardless of the patient's gender.

Although CBFV in normotensive individuals to be higher in women than men, ¹⁸ we did not observe a significant increase in CBFV in uncontrolled hypertensive women (65%) compared to men in group 2 (59%), probably because women were using more antihypertensives with potential arterial systemic vasodilation (64.5% and 41.9%, respectively ARB and CCB). ¹⁹ , ²⁰

Other studies have shown that even with increased arterial stiffness and reduced elasticity, hypertensive adults showed a constant or even increased ability to attenuate changes in the mean CBFV despite variations in SBP, ²¹ , ²² , ²³ probably due to some compensatory mechanisms. According to Lipsitz et al, ¹¹ one of these mechanisms seems to be the observed dissociation between the reactivity of small arteries to CO₂ and the cerebral hemodynamic response. Postural hypocapnia is known to be associated with reductions in CBF, whereas hypertensive patients, despite having greater hypocapnia in orthostasis, have a small reduction in CBF. ¹³ Therefore, it is possible that this mechanism is already underway in the hypertensive patients of our study, as they presented CrCP without significant changes in relation to that found in normotensive individuals (P = .75).

Unlike Kastrup et al, ²⁴ who found that cerebrovascular reactivity (CVR) decreased with age, especially in women, our study showed no difference in autoregulatory integrity with age. This suggests that there may be an AR capacity reserve as a function of individual degeneration of the multiple physiological mechanisms that compose it over time, and that the impairment of AR capacity only emerges when this reserve reaches its limit. ⁶

The lower limits of AR are increased in hypertensive patients, and there is evidence to suggest that this can be prevented or even modified by good pressure control. ²⁵ The effect of antihypertensive treatment on AR capacity is not well established, nor is whether any potential effect can be reversed by the suspension and/or duration of treatment. Lipsitz et al ¹⁰ demonstrated that 6 months of antihypertensive treatment with ACE inhibitors increased carotid elasticity and blood flow through the cerebral circulation. According to Périard et al, ²⁶ the protective effect of the antihypertensive agent, regardless of class, is probably not related to the restoration of AR but is due to the prevention of structural changes in the arterial wall through reductions in atheromatosis and shear stress.

Our study did not aim to evaluate the mechanisms of antihypertensive therapy on cerebral AR capacity, but we observed that hypertensive patients under treatment and at approximately the same time since diagnosis (P = .71), regardless of adequate BP control, did not exhibit AR impairment (P = .6). This result is in agreement with previous studies ¹¹ , ²¹ showing that AR is not impaired in hypertensive patients.

The vertebrobasilar territory was not evaluated. There are regional differences in brain AR capacity. Resistance in the major brain arteries increases and decreases in the brainstem during a moderate increase in BP. Thus, AR depends primarily on the brainstem arterioles, while in the brain, the large vessels play the most important role. Therefore, a pathophysiological alteration that affects hypertensive patients would occur more intensely in the brainstem, suggesting that this region, by being less dependent on the large vessels, may have more impaired AR or that its compensatory mechanisms may not depend on the histological structure. ²⁷

The present study has other limitations: (a) Among the hypertensive patients, there may have been patients with occult cerebrovascular disease, as the patients were not subjected to neuroradiological evaluation. CVR can be modified by the presence of changes in the cerebral white matter; (b) RCV reactivity to CO2 has not been evaluated, which could provide additional information on a possible decoupling between the metabolic response and CBF in hypertensive patients; (c) it is possible that the result obtained on the performance of RA, linked to CVR, is not so precise, because to estimate CBF through CBFV, it is necessary to consider that there are no variations in the caliber of the great cerebral arteries, justly those responsible for significant portion of this cerebrovascular resistance; (d) during the recording period, the transducer was manually held, increasing the susceptibility to variations in the CBFV resulting from possible variations in the insonation angle. As no difference was found between the AR of the normotensive and hypertensive groups, it is unlikely that the results were significantly altered by this technical limitation.

5. CONCLUSION

Hypertensive patients undergoing treatment, regardless of their blood pressure control, have intact cerebral AR capacity.

Research should be intensified in order to try know better the effects that each antihypertensive has on the AR cerebral capacity helping the choice of one that, in addition to properly controlling BP, will also contribute to maintaining the integrity AR cerebral.

CONFLICT OF INTEREST

On behalf of all authors, the corresponding author states that there is no conflict of interest.

AUTHORS’ CONTRIBUTIONS

Prof^a. Monica Sanches Yassuda, BSc ǀ Edson Bor‐Seng‐Shu, MD, PhD ǀ Ayrton Roberto Massaro MD, PhD ǀ Ricardo Nitrini MD, PhD ǀ Luiz Aparecido Bortolotto MD, PhD involved in study concept and design. Michel Ferreira Machado, MD ǀ Henrique Cotchi Simbo Muela MD, PhD ǀ Valeria Aparecida Costa‐Hong, BSc, PhD ǀ Natalia Cristina Moraes BSc, MSc ǀ Claudia Maia Memória, BSc involved in acquisition of data for the work. Michel Ferreira Machado, MD ǀ Henrique Cotchi Simbo Muela MD, PhD ǀ Ricardo de Carvalho Nogueira, MD, PhD involved in drafting of the manuscript and interpretation of data for the work. Ricardo de Carvalho Nogueira, MD, PhD involved in study supervision and statistical analysis. All authors have read, revised, and approved the manuscript.

ETHICAL APPROVAL

The study was approved by the institutional board review.

CONSENT FOR PUBLICATION

The subjects gave informed consent.

PATIENT AND PUBLIC INVOLVEMENT

Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Machado MF, Muela HCS, Costa‐Hong VA, et al. Evaluation of cerebral autoregulation performance in patients with arterial hypertension on drug treatment. J. Clin. Hypertens. 2020;22:2114–2120. 10.1111/jch.14052