Does blood viscosity affect cerebral blood flow? A study in a population living in a high-altitude city

Introduction

Blood viscosity (BV) indicates the resistance of blood when it is moving; thus, BV has been studied as a factor that could modify cerebral blood flow (CBF), although results have not been consistent in healthy people ¹. This could be due to the methodologies used to measure CBF or BV, the populations selected, or the presence of autoregulatory mechanisms that allow CBF to remain stable despite variations in BV. By contrast, clinical studies indicated that cerebral small vessel infarction is correlated with elevated BV level, suggesting that it may have an influence on the BV ^2,3. This finding suggests that it could have a negative influence on CBF, probably when autoregulatory mechanisms are decompensated.

These inconsistent results could also be due to the physics and dynamics of blood as a fluid; the blood is composed of several heterogeneous components, such as red blood cells, platelets, proteins, etc., which means that its behaviour as a fluid does not obey Newtonian laws. ^4,5. At different tensions, the BV will change; therefore, during systole or diastole or when passing through a large vessel or capillary, the BV level varies, which makes it different from other haematological parameters that are usually stable. This aspect, which has been widely studied from the rheology and physics point of view, has not been completely ^6-8 and fully integrated into the field of clinical study.

In order to provide a middle ground between a healthy and a diseased settler, our study aimed to assess settlers whose physiological self-regulatory mechanisms have already been exhausted. Hence, young and clinically healthy people were included in this study, who were expected to show preservation in their self-regulatory mechanisms, but who live in a city at high altitude and are exposed to physiological challenges, acclimatisation, and chronic hypoxia ⁹. The main compensatory mechanism consists of increasing the haemoglobin levels, which increases the BV values above those observed in people living at sea level ¹⁰. This study aimed to evaluate the variations in BV levels and to determine their correlation with some parameters of CBF.

Methods

Results

A total of 131 volunteers were included, 78 (59.5%) were women, with a median age of 24 [IQR: 22–29] years. The median haemoglobin level was 16.1 mg/dl [IQR: 14.8–17.9], with variations between men (18.1 mg/dl, IQR: 17.5–18.7) and women (15.3 mg/dl, IQR: 14.1–16.0). The other haematological parameters measured are presented in Table 1.

The blood viscosity levels ranged from 5.01 cP [IQR 4.45–5.73cP] at 300 sec⁻¹ to 6.16 cP [5.58-7.20 cP] at 75 sec⁻¹ (Table 2). Meanwhile, the mean CBF velocity was 57 cm/s [IQR: 50–65], and the PI was 0.9 [0.86–1.02].

The BV levels had a weak and negative correlation with mean CBF velocity, but not with the other parameters evaluated, with small variations according to the viscosity measurement, ranging from a value of r = −0.132 at 75 sec⁻¹ to a value of r = −0.170 at 300 sec⁻¹ (Table 3).

Discussion

Our study has some strengths. It included an understudied population, expanded the sample size of previous studies, and used several levels of BV following the current recommendations for its measurement ¹⁵. Initially, our results may not seem to provide a contribution, due to the low correlation identified in only one of the parameters evaluated. However, the information must be evaluated in the context in which it is presented. First, the study was conducted in a young population and in those with no symptoms or without described diseases; hence, their autoregulatory mechanisms should be unscathed and should correctly handle hemodynamic alterations to which they have been exposed to. Second, they have been chronically exposed to hypoxia, as evidenced by their elevated haemoglobin levels, while they must have undergone other physiological changes to acclimatise. Third, the sample size was extremely large. According to the estimates, the statistical power was superior to that reported in previous studies. Fourth, despite the previous points, a small impact on CBF was identified.

The association found, although small and only for mean FV, suggests that BV has an impact on CBF, but it is compensated by the self-regulatory mechanisms present as this is a young population ²⁰. In this regard, we have to analyse that the systolic and diastolic FVs are pulsatile components of CBF, while the mean flow velocity is a continuous component and calculated based on the systolic and diastolic flow velocity values ²¹. These aspects should be interpreted with caution because the absence of association between BV and systolic and diastolic values, but the presence of association between BV and mean flow values, suggests a regulation for continuous values, but not for pulsatile values; this may be because the variations of BV at this level are relatively small and because the mechanisms of autoreactivity or vasoreactivity also remain constant when cerebral perfusion is adequate ^22,23.

Studies in individuals who ascend rapidly to high altitudes show that CBF is altered by acute exposure to high altitude; in individuals who waited until acclimatisation to high-altitude areas resulted in hypoxia, a dynamic modification process of CBF was observed ^24,25. In individuals who waited until acclimatisation to high-altitude areas resulted in hypoxia, dynamic modification of CBF due to autoregulation was observed; however, this mechanism is not completely understood, and it may not have an impact on the CBF ²⁶ and may not have a significant impact on brain physiology ²⁷. However, in all studies analysed, the effect of BV, which also varies according to acclimatisation status, was not evaluated. Our study, however, suggests that BV has a small impact on CBF dynamics and it is related to the BV reported in chronic high-altitude dwellers. Although our results are limited, if we incorporate them to those obtained in clinical settings, we could suggest that BV is related to BV in chronic high-altitude residents ^2,3. Although our results are limited, if we incorporate them to those obtained in clinical settings, we could suggest that BV does play a role in the alterations of CBF levels and could be manifested when cerebral autoregulation mechanisms are overcome, although these have a lower expression in young and healthy people.

Although the effect of BV on CBF has not been studied further, other investigations have evaluated the impact of haematocrit on cerebral autoregulation, finding an association between higher haematocrit levels and alterations in CBF levels ^28,29 and have even modelled these phenomena ^30,31. Our population has higher BV and haematocrit levels than those living at sea level, but without demonstrating excessive erythrocytosis (men: Hb ≥20 mg/dl, women: Hb ≥18 mg/dl) ³². Moreover, our results are consistent with those of other evaluations conducted in populations living in high-altitude areas ¹⁰. The drawback of these indirect approaches is that BV is a dynamic phenomenon; therefore, even if we maintain a constant haematocrit, the BV level will change; hence, the impact of BV is probably not significant.

The low impact identified in our study could be attributed to the method we used, as previous studies developed several methods for measuring the CBF and some processes for measuring BV were not standardised ²³. Transcranial Doppler ultrasound was performed, which is a cerebral perfusion study and measures some components of blood flow ^1,33. It measures some components of CBF and has a good correlation with direct or invasive studies ¹⁷. The association between BV and CBF has been investigated since the studies of Grotta, Ackerman, Correia, Fallick and Chang ¹² and Brown and Marshall ³⁴. Although they reported that these two factors had a positive correlation, they presented technical limitations for both BV and CBF measurements due to the technological development that occurred during that time. Grotta, Ackerman, Correia, Fallick and Chang ¹² did not directly measure the BV levels, but estimated its value using a formula based on the haematocrit and fibrinogen levels; meanwhile, Brown and Marshall ³⁴ measured CBF using xenon-133. These variations, in addition to the use of small sample size, may explain why other studies have shown inconsistent data ¹.

In addition, BV levels are variable according to the share rate to which they are subjected, but this is not the only factor that could alter the BV levels. A component that is usually mentioned is the calibre of the vessel through which the blood transits, although BV can vary significantly according to this calibre; this is limited to vessels with less than 0.3 mm in diameter (arterioles and capillaries), which is known as the Fahræus-Lindqvist effect ³⁵. In our case, we evaluated the middle cerebral artery, whose average diameter is theoretically 3.2 mm, although the diameter can vary in patients with hypocapnia or hypercapnia ³⁶. These changes would have no impact on changes in BV levels. Although our results are only limited only to the large arteries and do not provide details on the events that can occur at the arteriole or capillary level, the physiological impact of BV on small vessels remain controversial ³⁵. Clinically, this impact is associated with small vessel infarctions ².

Conclusion

Only a slight variation was observed in BV levels. However, this variation had a significant impact on mean flows, but not on the other parameters. This indicates that, in young and clinically healthy individuals, there are autoregulatory mechanisms that compensate for BV variations, although it is not completely understood. This could play a role in the pathogenesis of cerebrovascular diseases that requires further clinical studies in older population.