Blood flows down a pressure gradient

Blood, like any other fluid, will only flow from an area of relatively high pressure to somewhere where the pressure is lower. This is equivalent to pointing out that rivers only flow downhill and the water in a pond, with no outlet to a lower point, does not flow anywhere. Jamshed Patel (Case 8.1:1) has used this principle by putting his leg out of bed in order to try to increase blood flow.

Viscosity of blood

The viscosity of blood is mainly determined by the haematocrit, the percentage of blood volume which is occupied by the red blood cells. The viscosity of blood is frequently expressed as a relative viscosity, i.e. the viscosity compared to pure water taken as unity. On this basis, the relative viscosity of blood plasma, which contains protein but no red cells, is about 1.3. At a normal haematocrit, of the order of 45%, the relative viscosity of blood flowing through tissues is about 2.4 (Fig. 8.1). This assumes that the blood flow is fast enough to keep the red cells apart as, if blood flow is sluggish, the red cells tend to stick together. This is because the large plasma proteins (globulins and fibrinogen) form cross-bridges between slowly moving red cells. These bonds are disrupted in faster moving blood. Sometimes red cells pile up in a similar fashion to a pile of dinner plates and this is called rouleaux formation. Aggregation of red cells becomes an important factor increasing the resistance to blood flow in circulatory shock. It will tend to occur in the postcapillary blood vessels, the venules, when the velocity of flow is low. This may be part of the explanation for the swollen appearance (localized oedema) of Jamshed’s foot (see Case 8.1:1). Blood flow may cease completely if the resistance is too high and this contributes to the localized tissue oedema which sometimes develops during circulatory shock (see Chapter 14).

Laminar and turbulent flow

Blood flow may be either laminar or turbulent. ‘Laminar’, otherwise known as ‘streamline’, fluid flow means that all the particles in the fluid are flowing parallel to the wall of the tube (Fig. 8.2). However, they are not all moving at the same velocity. Those fluid particles in contact with the wall of the tube are theoretically stationary whilst those at the centre (axis) of the tube are flowing fastest (Fig. 8.2). The opposite of laminar flow is turbulent flow. In this case the fluid particles follow a much more irregular pathway and may develop vortices (whirlpools) in the blood vessel.

The conditions which result in the transition from laminar to turbulent flow were described mathematically by Osborne Reynolds in 1883. The essence of Reynolds’ law is that turbulence is more likely to occur in large tubes than in small tubes. Turbulence is more likely when the velocity of flow is high and when the viscosity of the fluid is low. The blood flow velocity at which there is a transition from laminar to turbulent flow is called the ‘critical velocity’ (Fig. 8.2). Laminar flow is essentially silent but turbulent flow sets up vibrations in the blood vessel wall which can be heard using a stethoscope. Turbulent flow in the circulation produces the noises which are called murmurs.

As the viscosity of blood depends primarily on the haematocrit and a low viscosity makes the development of turbulence more likely, anaemic patients are more likely to have murmurs in their circulation than those with a normal haematocrit. A murmur caused in this way would disappear once the anaemia was corrected. An example of this is during pregnancy. Maternal haematocrit decreases in pregnancy because the plasma volume expands by more than the red cell volume. Flow murmurs are therefore more likely to be heard in pregnant women and may cause temporary alarm, but they normally disappear when the baby is delivered and haematocrit returns to normal. We also use the development of artificially induced flow murmurs, produced by compressing an artery with a sphygmomanometer cuff so that flow velocity increases as the basis for non-invasive measurement of arterial blood pressure (see Chapter 10).

Within the normal circulation, blood flow can be thought of as approximating to a laminar flow pattern in most large vessels. The site where turbulence is most likely to occur in a normal person is in the first segment of the aorta because here the velocity of flow is high in a relatively large tube. During exercise, when the cardiac output, and therefore the velocity of flow, increases, turbulence will extend further down the aorta than at rest. Local changes in blood flow dynamics leading to turbulence are contributory factors in determining the location of endothelial cell damage which is a precursor to the development of atherosclerotic plaques in the circulation (see p. 91). Thrombus formation is more likely when blood flow is slow and there is no turbulence, i.e. in the veins.

In Jamshed’s case (see Box 8.1:1) a loud bruit (flow murmur produced by turbulent blood flow) was heard over the femoral artery. The probable explanation is that a narrowing of the vessel led to locally increased blood flow velocity through the constriction and hence the development of turbulence on the downstream side.

Relationship between blood vessel radius and blood flow

Where there is laminar flow, the flow rate is proportional to the fourth power of the radius of the blood vessel. Strictly, this relationship only applies to an ideal (Newtonian) fluid. Blood is anomalous, because it is not a homogeneous fluid, it has red cells suspended in it. Despite this, the fourth power relationship, as an approximation, provides a very useful concept for both physiological and pathological considerations. The relationship ‘flow is proportional to radius⁴’ was first derived by the French physician Poiseuille in 1846. To illustrate the effect of the relationship we can take a simple numerical example. If a blood vessel with a radius of 4 units is dilated to a radius of 5 units, this represents a 25% increase in the radius of the vessel. But, if we consider the effects on blood flow, if r = 4 then r⁴ = 256 and if r = 5 then r⁴ = 625. A 25% increase in size of the vessel would, therefore, lead to a 140% increase in blood flow.

The physiological consequence of the fourth power relationship is that small changes in the diameter of resistance blood vessels lead to relatively big changes in flow. Blood can, therefore, be diverted to match metabolic needs by constricting and dilating small blood vessels. These are known as the resistance vessels (see Chapter 9). In pathological terms, small changes in blood vessel diameter produced by, for example, atherosclerosis may result in large reductions in blood flow (see p. 90).

Narrowing of the femoral artery and hence reduction of blood flow has limited the delivery of blood to Jamshed’s calf muscle (see Box 8.1:1), hence causing hypoxic pain in the muscle. This became worse during exercise because the increased oxygen demand could not be satisfied by an adequate increase in blood flow.

Red cell distribution over the cross-section of a blood vessel

‘Axial accumulation’ of red cells is a consequence of laminar blood flow. Red cells are dragged into the part of the blood vessel which has the fastest flow, i.e. down the middle of the blood vessel. The blood flowing slowly near the wall of the blood vessel, therefore, has a lower haematocrit than the fast flowing blood at the centre of the vessel. Small branches from a large vessel hence receive blood which has a lower haematocrit than the average for the large vessel. Because the viscosity of blood is dependent on the haematocrit, the relative viscosity of blood in small vessels will be lower than that in large vessels. Experimentally, this was shown to become significant in blood vessels with a diameter of 300 μm or less and is known as the Fåhraeus–Lindqvist effect (Fig. 8.3). It provides an explanation for the fact that the relative viscosity of blood flowing through a vascular bed (which has lots of small-diameter vessels) is lower than the relative viscosity measured in a glass viscometer (which has a large tube diameter). A key point in understanding this complex phenomenon is to recognize that axial accumulation means that the red cells flow faster than the plasma in blood vessels.

Although blood does not behave as an ideal homogeneous fluid, the studies carried out by Poiseuille still provide us with a useful basis for understanding blood flow dynamics. However, Poiseuille’s law does assume that the tube is rigid.

Capillary blood flow

The smallest blood vessels which have the thinnest walls behave effectively as rigid tubes. This, at first, may seem surprising. In very small blood vessels concepts such as laminar flow have little meaning as capillaries have a diameter in the range of 3–8 μm. Red blood cells have a diameter of about 8 μm which means that they are a very tight fit passing through many capillaries. Clearly, either the capillary or the red cell has to deform and it is, in fact, the red cell which changes shape (see Chapter 11). As red cells are such a tight fit in capillaries, there is a region of trapped plasma in between successive cells. This is referred to as bolus flow.

Elasticity of blood vessel walls

Large blood vessels are distensible. For a rigid (e.g. glass) tube with a laminar flow pattern, there would be a linear relationship between pressure gradient across the ends of the tube and flow down the tube (Fig. 8.4A). In a large blood vessel, Figure 8.4B shows that the relationship between flow and pressure gradient is different in two ways:

The critical closing pressure means that there has to be a certain pressure inside the vessel in order to keep it inflated, i.e. if the pressure gradient across the wall of a blood vessel (transmural pressure) falls below a certain limiting value, then the vessel will collapse and flow will cease. Stimulation of sympathetic vasoconstrictor nerves to a blood vessel will increase the critical closing pressure and so a higher pressure will be needed inside the vessel in order to keep it open. This concept is important in understanding the shutdown of some blood vessels in circulatory shock. The fall in arterial blood pressure in shock leads, via the baroreceptor reflex, to sympathetically induced vasoconstriction in the arterioles (see Chapter 10). The pressure of blood in the vessels after this increased resistance will fall and, hence, there is a tendency for vessels to collapse (see Chapter 14).

The curvilinear relationship between blood pressure and flow (Figure 8.4B) is attributed to the fact that a blood vessel will be distended as the pressure inside it increases. Blood flow will also therefore increase. The extent to which a blood vessel can be distended by increasing the internal pressure inside it will depend on the blood vessel size and wall structure. Figure 8.5 shows a comparison between an ‘old’ and a ‘young’ aorta and a vein. Firstly, it can be seen that the vein is more distensible than the arteries. Second, Figure 8.5 shows that an aorta from a young person is easier to distend than an aorta from an older individual. Distensibility is the increase in volume of a blood vessel per unit increase of pressure inside it. Veins have relatively thin walls which contain only small amounts of collagen compared to arteries (see Chapter 1). Veins are easily distended and are, therefore, referred to as capacitance vessels. Normally, the veins contain approximately two thirds of the body’s total blood volume at any one time. Extra blood transfused into a person would primarily be accommodated in the veins and, correspondingly, blood loss would particularly result in a decrease in the volume held in the veins. Distribution of blood volume within the circulatory system is shown in Figure 1.7.

Congenital defects

Congenital abnormalities of arteries and veins are often related to an abnormal course, pattern of branching or anatomical relations of a vessel. This may be particularly important during surgery. An abnormally positioned coronary artery may predispose the patient to cardiac arrhythmias or even sudden death (see Chapter 3).

Berry aneuryms are a consequence of a congenital abnormality of the wall of one or more cerebral arteries, usually at a junction point between two vessels in the circle of Willis. The abnormality appears to be in the media layer of the blood vessel wall. It leads to saccular aneurysm formation which may rupture. This is often related to hypertension, which may only be short term as during physical exertion. Some berry aneurysms occur in the context of other pathologies such as polycystic kidney disease, Marfan’s syndrome or Ehlers–Danlos syndrome. Favoured sites for berry aneurysms include the junction of the internal carotid and posterior communicating artery and the junction of the anterior cerebral and anterior communicating arteries. Rupture of the aneurysm may lead to subarachnoid and/or intracerebral haemorrhage and possibly sudden death.

An arteriovenous fistula may also be a congenital vascular abnormality. In this case, there are well developed, abnormal communications between an artery and a vein which have the potential for thrombosis and rupture (e.g. cerebral arteriovenous fistulae).

Age-related changes in blood vessels

Many of the characteristic pathological changes seen in the vasculature of older individuals can be seen in younger populations with specific disease processes such as hypertension. The term ‘arteriosclerosis’ is often used to describe the ‘hardened’ or ‘thickened’ arteries of the elderly. This is due to the slow, progressive thickening of the intima, combined with medial fibrosis that occurs over many decades.

Included within the umbrella term arteriosclerosis are atherosclerosis (see below) and Monckeberg’s medial calcific sclerosis. This latter disease process is characterized by areas of medial calcification in small to medium-sized arteries, especially the arteries of the upper and lower limbs. No inflammation or necrosis is involved and it usually occurs in older patients over 50 years of age. Other than being a radiological oddity, the disease usually has no clinical effects.

Risk factors for atherosclerosis

A substantial number of risk factors have been identified which make the development of atherosclerotic lesions more likely. Constitutive (non-modifiable) risk factors include age, gender and some genetically determined factors such as familial hypercholesterolaemia. In the 35–55 age group fully formed atherosclerotic plaques are more common in men than in women. After the menopause there is an increased incidence and severity in women although men still remain marginally more affected than women. It is thought that oestrogens provide some protection in younger women.

Epidemiological research has identified some ‘major’ and some ‘minor’ (but still significant) modifiable risk factors. Major risk factors are diabetes, smoking, hyper-lipidaemia and hypertension, all of which are treatable or avoidable. Minor modifiable risk factors include lack of exercise, obesity, personality type and stress. A number of other potential risk factors have also been proposed.

Pathogenesis of atherosclerosis

Despite the investment of large amounts of money in research budgets, the precise definition of the triggering factors for atherosclerosis remains elusive. Part of the reason for this is that atherosclerosis develops over a time course of 40+ years. This means that it is impossible to find entirely satisfactory experimental animal models of atherosclerosis. Many theories have been proposed and currently the response to injury hypothesis is particularly favoured. This theory was proposed by Ross in 1973 although it does incorporate some of the older hypotheses.

Essentially, the response to injury theory suggests that atheroma occurs because of long-term ‘grumbling’ injury to the endothelium. Causative factors for this endothelial injury leading to endothelial dysfunction include hyperlipidaemia (particularly low-density lipoprotein—LDL cholesterol), diabetes (damage as a result of hyper-glycaemia causing glycosylation of proteins and dyslipidaemia), hypertension, toxins acquired as a result of cigarette smoking, increased plasma homocysteine and infective agents. In the latter group cytomegalovirus (CMV), Chlamydia pneumoniae and Helicobacter pylori have been particularly prominent as proposed causative organisms for atheroma formation. Many of these insults to the endothelial layer may be linked to the increased production of reactive oxygen species (ROS). This means primarily the formation of the O^–₂ anion but the ROS group also includes hydrogen peroxide (H₂O₂), the hydroxyl anion (OH^–), and a range of lipid radicals formed from interaction with the peroxynitrite anion (ONOO^–), itself a product of the superoxide anion and nitric oxide (NO). Reduction in the availability of NO as a result of oxidative stress may be a key event in the development of atheroma at several points. Therapeutic strategies targeting a reduction in oxidative stress and an increase in NO availability are under investigation.

Altered function of the endothelial cells leads to increased permeability to LDL cholesterol and increased white blood cell adherence. Monocytes enter the intima layer and transform into macrophages. The macrophages accumulate oxidized LDL cholesterol and are then known as foam cells. Together with infiltration of T lymphocytes these events lead to the initial atherosclerotic lesion, the fatty streak. These often develop very early in life and have been found at autopsy in the arteries of stillborn babies. It is widely presumed that the fatty streak is the precursor of the later plaques but this is difficult to prove conclusively. Although high levels of LDL cholesterol present an important risk factor for atherogenesis, high levels of HDL (high-density lipoprotein) are considered to be protective.

The smooth muscle cells in the blood vessel wall are normally located in the media layer. In a normal arterial blood vessel there is a sparse population of cells with a dual contractile and fibroblast phenotype in the intimal layer. These myointimal cells start to proliferate under the influence of cytokines released from foam cells and from platelets which adhere to the damaged endothelial surface. The myointimal cells start to secrete collagen and the now stiffened atheromatous structure is referred to as a lipid plaque at this stage.

Final development of the plaque into a hard, white fibrolipid plaque (Fig. 8.6) follows further production of collagen and sometimes calcification of accumulated extracellular lipid. The damaged endothelial layer may ulcerate providing a site for thrombosis to form.

The process of atherosclerosis described above has many of the characteristics of an inflammatory response. These include invasion of monocytes which become macrophages, the involvement of T lymphocytes, the production of cytokines and growth factors and, in the later stages, focal necrosis in the blood vessel wall. Answering the question: ‘An inflammatory response to what?’ may prove productive in the future.

Pathological consequences of atheroma

Major blood vessels may become narrowed by atheroma hence reducing blood flow. This may range from a fairly modest change which is insufficient to cause symptoms to almost complete interruption of flow (Fig. 8.7). This frequently results in ischaemic heart disease, or inadequacy of cerebral blood flow or peripheral vascular disease as described in the history of Jamshed in Case 8.1:1.

Exposure of collagen in the vessel wall to blood constituents when the endothelial layer is damaged initiates thrombus formation. This may suddenly totally occlude an already narrowed artery or the thrombus may embolize and block a vessel further downstream.

Weakening of the blood vessel wall as a consequence of atheroma may lead to the formation of an aneurysm, a region where the weakened wall balloons out (see p. 94). The most common site for this to occur is in the abdominal aorta (Fig. 8.8).

• Large vessel vasculitis:

• Medium sized vessel vasculitis:

• Small vessel vasculitis:

• macrovasculopathy

• microvasculopathy.

Macrovasculopathy

Diabetic patients, particularly those that are poorly controlled, tend to have more severe and widespread atheroma than non-diabetics. Thus complications such as ischaemic heart disease (coronary artery atheroma), stroke (carotid and cerebral arteries), lower limb ischaemia leading to gangrene in the toes and feet and renal ischaemia tend to be more common in diabetic patients.

Microvasculopathy

Hyaline arteriolosclerosis is a frequent finding in the arteriolar (and capillary) circulation of diabetic patients. This thickening of the vessel wall, which looks very pink in a section stained with haematoxylin and eosin, may be due to increased flow of plasma proteins across the vessel wall leading to deposition of high relative molecular mass proteins such as fibrinogen and LDL cholesterol. It is important to realize that hyaline arteriolosclerosis is also seen in amyloidosis, benign hypertension and in the arterioles of the elderly. This microvasculopathic process causes ischaemic lesions in the retina associated with haemorrhage and vessel proliferation and in peripheral nerves with the resulting neuropathy leading to skin insensitivity to pain and ulceration.

Arteriolosclerosis is also responsible for renal and cerebral changes. The basement membrane of the glomerulus thickens and this is associated, paradoxically, with increased permeability of the glomerular filter leading to proteinuria. Renal failure can occur as nephrons are lost due to glomerular sclerosis. The arterioles of the kidney often show hyaline mural change.

• Rupture: this may lead to torrential and often fatal haemorrhage into the peritoneal cavity (haemoperitoneum).

• Embolus formation: variable sized fragments may split off from the intimal thrombus and cause occlusion of the small arteries in the legs and feet leading to gangrenous necrosis.

• Infective aneurysm: also known as a mycotic aneurysm, this rare condition occurs when an infected embolus lodges in an artery and allows seeding of the vessel wall by the microbes. This classically occurs as a consequence of an embolus from an infected vegetation on a heart valve as in infective endocarditis (see Chapter 3). The vessel wall can become inflamed, soft and rupture. It is also possible for infective aneurysms to occur in septicaemia.

• Syphilitic aneurysms occur in the tertiary stage of syphilis, typically in the ascending thoracic (rather than abdominal) aorta. The disease process, which starts as an inflammatory process around the vasa vasorum of the adventitia, may lead to a thick-walled, dilated aorta which may extend back as far as the aortic root at the heart causing valvular incompetence.

• Vasculitic aneurysms. Macroscopic polyarteritis nodosa may lead to aneurysm formation by causing inflammation, necrosis, thrombosis, fibrosis and weakening of the vessel wall. With fragmentation of the elastic lamina, aneurysm formation can occur.

Dissecting aneurysm

This should be considered as a special category of aneurysm. In fact, there is not usually any significant dilatation of the vessel lumen. The pathology is due to a tear in the vessel wall. Blood enters the arterial wall through the tear in the intima and tracks into the media, shearing off the inner third from the outer two thirds. The blood may then re-enter the vessel through a second tear in the wall (‘double-barrelled aorta’) or may rupture to the outside of the vessel potentially causing sudden death by massive haemorrhage. Dissecting aneurysms most often occur in the thoracic aorta. The patient may present with severe chest pain (felt between the shoulder blades). Examination may reveal absent pulses in the area of dissection. The patient is often a middle-aged male with a known history of hypertension. Less commonly, dissection may occur in the context of a young patient with Marfan’s syndrome. In Marfan’s syndrome, there is an inherited reduction of a protein in elastic tissue. This leads to ‘cystic medial degeneration’ of the vessel wall, which becomes weak and can tear.

Rarely, dissection may occur in smaller calibre vessels because of invasive procedures, such as during or after coronary artery angioplasty.

Assmann, G., Nofer, J-R. Atheroprotective effects of high density lipoproteins. Annu. Rev. Med.. 2003; 54:321–341.

Bass, P., Burroughs, S., Carr, N., Way, C. Master Medicine: General and Systematic Pathology, third ed. Churchill Livingstone: Edinburgh; 2009.

Becker, A. E., de Boer, O. J., van der Wal, A. C. The role of inflammation and infection in coronary artery disease. Annu. Rev. Med.. 2001; 52:289–297.

Donnelly, R., London, N. J. M. ABC of Arterial and Venous Disease. London: BMJ Books; 2000.

Forbes, C. D., Jackson, W. F. Colour atlas and text of clinical medicine, third ed. Edinburgh: Mosby; 2002.

Gallagher, P. J. Cardiovascular system. In Underwood J. C. E., ed. : General and Systematic Pathology, fourth ed., Edinburgh: Churchill Livingstone, 2004.

Hamilton, C. A., Miller, W. H., Al-Benna, S., et al. Strategies to reduce oxidative stress in cardiovascular disease. Clin. Sci.. 2004; 106:219–234.

Hansson, G. K., Robertson, A-K. L., Soderberg-Naucler, C. Inflammation and atherosclerosis. Annu. Rev. Pathol. Mech. Dis.. 2006; 1:297–329.

Hunt, B. J., Poston, L., Schachter, M., Halliday, A. W. Introduction to Vascular Biology: From Basic Science to Clinical Practice, second ed. Cambridge: Cambridge University Press; 2002.

Jennette, J. C., Falk, R. J. Medical progress: small vessel vasculitis. N. Engl. J. Med.. 1997; 337:1512–1523.

Levick, J. R. An introduction to cardiovascular physiology, fifth ed. London: Arnold; 2009.

Smith, J. J., Kampine, J. P. Circulatory Physiology—The Essentials, third ed. Baltimore: Williams and Wilkins; 1990.

Stevens, A., Lowe, J. Pathology, second ed. Edinburgh: Mosby; 2000.

Underwood, J. C. E. General and Systematic Pathology, fourth ed. Edinburgh: Churchill Livingstone; 2004.