Valve cusps are normally avascular and have a dense collagenous core (fibrosa) covered by loosely arranged connective tissue (spongiosa) which is rich in tissue mucins. The rate of connective tissue turnover is high. Mechanical trauma to cusps induces the production of excessive amounts of dense collagen by the spongiosa. Any derangement of cusp architecture therefore invokes non-specific secondary fibrosis, further affecting function.

The common cause of an acute valvulitis is rheumatic fever; at a simple level this is an example of immunological damage to valve tissues which share antigens with components of streptococcal cell walls. A streptococcal throat infection stimulates the production of antibodies which then cause valve damage: streptococcal antigens do not occur in the valve. Cellular immune responses are probably also involved. In the acute phase of rheumatic fever, the cusps are infiltrated with inflammatory cells (predominantly lymphocytes), and small platelet vegetations form along the apposition lines. Mortality in the acute stage relates to myocarditis rather than to valvulitis, although mitral regurgitation may develop following ventricular dilatation. Acute rheumatic fever may be followed by chronic valve disease after an interval of some years.

Chronic rheumatic valve disease is associated with vascularization and effacement of the cusp architecture by dense fibrosis. Fusion of the commissures, cusp fibrosis, and contraction are the three macroscopic hallmarks of chronic rheumatic disease. Depending on which predominates, all combinations of stenosis and regurgitation can result. The mechanism underlying continuing fibrosis may be recurrent subclinical attacks of acute rheumatic fever, or the end-stage of secondary fibrosis due to valve derangement.

Valves that are morphologically identical to those damaged by chronic rheumatic disease can be found in patients who have no history of acute rheumatic fever or chorea. It is uncertain whether these individuals have suffered subclinical streptococcal infections, or whether there are other causes of valvulitis. Both systemic lupus erythematosus and rheumatoid disease can be associated with a valvulitis which, in the end-stage, is very similar to chronic rheumatic disease.

Animal models of bacterial endocarditis show that platelet deposition and bacteraemia must coincide for infection to be established. Microscopic platelet deposition occurs on any valve in which there is regurgitant flow or abnormal cusp contact. Bicuspid aortic valves and prolapsed mitral valves are, therefore, at risk of bacterial endocarditis, as are valves affected by chronic rheumatic disease. Small platelet thrombi are also common on the apposition lines of the aortic cusps in subjects over the age of 70, even when the valve is tricuspid and functionally normal.

In established bacterial endocarditis the vegetations on valve cusps show a striking stratification. On the surface is thrombus rich in platelets with abundant dividing organisms; beneath this is a layer of thrombus rich in fibrin, containing dormant but viable organisms. The valve cusp underneath the thrombus shows an intense pleomorphic inflammatory infiltrate with numerous polymorphs. Fibroblastic proliferation and vascularization are also prominent. The inflammatory and repair processes are remote from the organisms ensconced in the thrombus. The tissue destruction in the underlying cusp is due to the activity of the polymorphs and monocytes in the inflammatory response.

Fragmentation of the collagen within the fibrosa produces open spaces filled with connective tissue mucins (myxoid change); this is a feature of the valve cusps in Marfan's disease. The cusps expand, rather than contract, but excess cusp movement and trauma leads to fibrosis in the spongiosa. Similar changes are also found as a cause of aortic or mitral disease in the absence of any systemic manifestation of connective tissues disease.

Age-related calcification develops within both the cusp fibrosa and the valve rings, and is related to mechanical stress on mature collagen. Each cusp in the aortic valve may develop a C-shaped mass of calcium at the point of maximal flexion; in the mitral valve calcium forms across the base of the posterior cusp. Calcification is enhanced by elevation of the calcium × phosphate product in blood.

The aortic valve cusps are contained within a short fibrous sleeve, the upper border of which is scalloped to form a three-pointed coronet; the commissures are attached to each point. There are two ‘rings’: the upper is the junction between the aortic sleeve and the base of the aorta; the lower is where the aortic sleeve joins both the anterior cusp of the mitral valve and the interventricular septum. The upper or supra-aortic ring (sinotubular junction) is marked by a distinct ridge at the level of the commissures, and has functional significance in that dilatation leads to regurgitation. The lower ‘ring,’ however, determines the size of prosthetic valve that can be inserted. In the closed position the aortic cusps abut and overlap on their ventricular faces. Fenestrations in the portion of the cusp above the closure line commonly develop with increasing age; they are of no functional significance.

Between 1 and 2 per cent of normal individuals have bicuspid aortic valves. These range between two anatomically normal cusps of equal size to unequal cusps, the larger of which may have a deep cleft in the free edge. A fibrous raphe may cross the aortic surface of a large bilobed cusp, or a fibrous chord may support the free edge. In bicuspid valves it is more common for there to be an anterior sinus from which both coronary arteries arise.

The pathogenesis of aortic valve stenosis can best be ascertained by viewing the valve from above (Fig. 1) 1796. In adults over 40 the most common cause of isolated aortic stenosis is calcification in a bicuspid valve. The calcification is dystrophic in type and identical to that which occurs at a later stage in tricuspid aortic valves. It often begins in the raphe of the larger cusp, and extends throughout both cusps as bars and nodules of calcium. In the later stages all cusp architecture is obliterated, but it is usually possible to recognize the residual opening as a transverse slit extending across the whole width of the aortic root (Fig. 2) 1797. Commissural fusion is absent. The nodular masses of calcification in the cusps may erode on to the aortic surface, and become covered by thrombus; bacterial endocarditis is also a risk. The centres of large calcific nodules may contain soft gritty material, and spontaneous or iatrogenic embolism of calcific debris can occur.

Tricuspid calcific aortic valve stenosis is characterized by three distinct cusps without commissural fusion. Each cusp has a C-shaped nodule of calcium making the cusp rigid and leaving a triradiate lumen. Most cases occur in those over the age of 70.

True congenital aortic stenosis, where stenosis was present at or shortly after birth, ranges from a diaphragm with a central orifice and a hint of ridges where the three commissures should have formed, through what appear to be bicuspid valves with failure of separation of the commissures, to the uncommissural valve which has a dome-shaped diaphragm and an eccentric elliptical orifice. Aortic valve dysplasia in infants is characterized by replacement of the whole valve by nodular masses of disorganized mesenchymal tissue.

A small number of stenotic aortic valves cannot be fitted into these categories. Bicuspid valves with commissural fusion are assumed to derive from coexistent rheumatic disease (Fig. 4) 1799. Difficulty arises over valves which show dystrophic calcification in three cusps but fusion of one commissure: these raise the question of whether asymmetrical commissural fusion occurs in chronic rheumatic disease. If typical mitral rheumatic lesions are present the aortic lesion is usually considered to be rheumatic; if not, the question remains open. Acquired fusion of one commissure may also be confused with a bicuspid valve with a deep notch. In the former case both cusp edges can be traced up to the commissure; in the latter they actually reach the commissure.

In any surgical series the relative proportion of each form of adult aortic valve stenosis will depend on the selection of case material by geographic area and age.

Competence of the aortic valve depends on a critical relationship between total cusp area and aortic root area at the commissural level. The former normally exceeds the latter by up to 40 per cent, allowing the cusps to abut and support each other in the closed position. With age the aorta dilates, the diameter of the aortic root at the supra-aortic ridge increases, and some compensatory increase in cusp area occurs. Consideration of the pathogenesis of aortic regurgitation must always consider the relative proportions of root and cusp area.

In chronic rheumatic disease cusp fibrosis leads to a drastic reduction in overall cusp area. The presence or absence of concomitant commissural fusion determines whether coexistent stenosis is present. Perforations or tears in cusps, whether due to trauma or bacterial endocarditis, are also causes of regurgitation (Fig. 5) 1800. A small proportion of patients with bicuspid aortic valves develop regurgitation rather than stenosis. The risk is greatest where one cusp is significantly larger than the other, and has a deep cleft in the free edge. Rupture of a fibrous chord supporting a large cusp may precipitate regurgitation. Tricuspid aortic valves in which the cusps are soft, redundant, and prolapse, but in which the root is of normal size are often designated as ‘floppy’ aortic valves. The condition is regarded as analogous to the mitral floppy valve, but is very rare.

Any increase in cross-sectional area of the aortic root above that which can be compensated for by cusp stretching results in regurgitation. Viewed from above the cusps fail to meet, leaving a central gap in the closed position (Fig. 6) 1801. The reduction in mutual cusp support often allows one cusp to prolapse below its neighbour. This cusp will develop a localized fibrous, nodular, thickening of the free edge as a result of regurgitant flow (Fig. 7) 1802. Aortic root dilatation is a feature of Marfan's disease, but the majority of patients with isolated aortic root disease have no other evidence of connective tissue disorders. The condition is variously known as idiopathic non-inflammatory root dilatation or annulo-aortic ectasia, and may be familial.

In populations in which chronic rheumatic valve disease has vanished idiopathic root dilatation is becoming the single most common cause of aortic regurgitation. There is an undue association of aortic root dilatation with bicuspid aortic valves; root dilatation appears before and independently of regurgitation or stenosis and appears to be an associated anomaly rather than a secondary phenomenon.

Histological examination shows changes in the aortic media immediately above the supra-aortic ring. There is no inflammation. The elastic lamina appear straightened or fragmented, and smooth muscle is lost. In extreme cases, particularly in Marfan's disease, elastic laminae may fragment sufficiently to leave spaces filled with connective tissue mucins. The appearance is referred to as cystic medial necrosis, although necrosis is not present.

Inflammatory destruction of the aortic media also leads to regurgitation. Syphilis causes both an increase in aortic root size and a separation of the valve cusps at their commissural attachments. All medial destruction, whether inflammatory or non-inflammatory, leads to a wrinkled scarred intima when viewed en face. In syphilis this is often confined very sharply to the first inch or two of the aorta, and the coronary ostia may be narrowed. The cusps are normal, apart from rolled free edges which occur as a secondary phenomenon. The histological changes of HLA-B27 related aortic disease in patients with ankylosing spondylitis are identical to those of syphilis, with intense perivascular infiltration of the media and adventitia by lymphocytes, plasma cells, and macrophages. In contrast to syphilis, fibrosis develops in the sinuses, and extends into the bases of the aortic cusps. Fibrosis also extends into the base of the anterior cusp of the mitral valve and into the atrioventricular nodal area. Most cases are associated with ankylosing spondylitis, but this valve lesion also occurs in individuals with trivial joint lesions. A pure aortic root and ascending aortic dilatation without cusp involvement occurs in the elderly, particularly in women. Large areas of medial necrosis appear, surrounded by giant cells which appear to be related to fragmented elastic. The condition is related to, and may be associated, with giant cell arteritis elsewhere.

For all practical purposes the sole cause is chronic rheumatic valve disease. In its simplest form, fusion of both commissures leads to a mobile diaphragm with a central oval aperture. Superimposed calcification may appear as nodules at the commissures (Fig. 8) 1803 or as more diffuse cusp deposits. Cusp rigidity thus contributes to stenosis. Cusp calcification frequently occurs in the absence of commissural fusion. Fusion of the chordae into solid pillars of collagen may superimpose an element of subvalve stenosis. Histological examination of the atrial appendage in mitral stenosis may show residual Aschoff bodies; these are long lasting granulomata which are an indication of previous rheumatic fever but do not indicate current activity. They may retain their characteristic histology for up to 20 years before becoming small fibrous scars.

Cusp fibrosis leading to retraction of the posterior cusp is typical of rheumatic disease and when associated with shortening of the chordae leads to regurgitation due to the virtual absence of a posterior cusp. The anterior cusp may be normal or even slightly expanded. Chordal fusion and shortening may also immobilize the anterior cusp (Fig. 9) 1804.

Perforation through the body of the cusp is typical of the healed stage of bacterial infection. The perforation may have a well-defined edge or an aneurysmal bulge which appears blown out like a wind sock. Healed bacterial infection leads to considerable fibrosis which may obscure any pre-existing valve lesion. Chordal rupture is also typical of bacterial endocarditis; in the healed stage residual vegetations are often present as calcified nodules on the ruptured ends.

Cusp expansion and chordal elongation are typical of the floppy mitral valve, the basis of which is myxomatous degeneration of the cusp fibrosa. Chordal elongation and cusp expansion allow cusp prolapse and lead to the characteristic ballooned cusps (Fig. 10) 1805. The cusps appear to be larger than normal, but are in fact thicker since considerable surface fibrosis occurs due to cusp impact. The cusps feel soft and hypermobile, in contrast to rheumatic disease in which the cusps are hard and rigid. The extent of cusp involvement varies. The most typical form involves one or more scallops of the posterior cusp, but both cusps or the anterior cusp alone may be affected. When disease is limited the remaining cusp tissue is often normal, and even with time it does not necessarily become involved. The central collagen of the chordae also undergoes cystic change, and loses its tensile strength. Chordal rupture is the most significant event precipitating clinically significant regurgitation. The ruptured chordae may fold back and become fused to the ventricular face of the valve cusp, or it may remain visible as a stump.

Microscopic deposition of platelet thrombi on the apposition lines of the expanded cusps is almost universal, but in the absence of bacterial endocarditis larger vegetations do not occur. The microthrombi may account for the minor transient ischaemic attacks that occur in patients with mitral valve prolapse.

The pathogenesis of the floppy mitral valve is uncertain. Diffuse involvement of both cusps is found in patients with genetic abnormalities of connective tissue synthesis, such as Marfan's disease and the joint hypermobility syndromes. Most subjects with isolated mitral prolapse, however, have no systemic disorder. Another view of the pathogenesis is that the myxoid change is a ‘wear and tear’ change in a valve which originally had minor anatomical abnormalities of the chordal support of the cusps.

There is a wide spectrum of severity of ischaemic mitral regurgitation. Papillary muscle necrosis occurs in up to 40 per cent of posterior–inferior acute infarcts, making ischaemic damage very common. At its most severe extreme there is rupture through either a whole papillary muscle at its base, or through one head or portion of a head of a papillary muscle closer to the chordal insertions. Rupture of a chord itself cannot be ischaemic since the chordae are avascular.

Elongation (expansion) of infarcted papillary muscles may occur in the acute phase of infarction and, although they remain intact, fibrosis perpetuates this elongation leading to regurgitation. Fibrous scarring at the bases of the papillary muscle also alters the regulation of cusp apposition in systole and leads to mild regurgitation.

Mitral ring calcification is common in those over the age of 70 but is a misnomer because the calcification is just below the insertion of the cusp into the atrioventricular junction. Calcium splints the valve orifice leading to a combination of mild stenosis and regurgitation. The age of the patients precludes consideration of the condition as a surgical problem.

Ring dilatation is a cause of regurgitation. The mitral valve ring to which the cusps are attached in the atrioventricular junction is not a solid fibrous structure with sufficient innate tensile strength to maintain the normal ring size. Ring size is dependent more on the structural and functional state of the left ventricle.

Tricuspid rheumatic disease is analogous to that in the mitral valve and due to commissural fusion and cusp contraction. Rheumatic tricuspid disease is usually concomitant with both aortic and mitral disease and virtually never occurs in isolation. Right ventricular and ring dilation rather than cusp fibrosis is a major factor in regurgitation. Tricuspid stenosis and regurgitation may also develop in subjects with carcinoid tumours. The pathological processes are striking in that fibrosis is superimposed on the cusp leaving the fibrosa intact. Identical lesions develop in the pulmonary valve.

Allen W, Matloff J, Fishbein M. Myxoid degeneration of the aortic valve and isolated aortic regurgitation. Am J Cardiol 1985; 55: 439–44.

Bulkley B, Roberts W. Ankylosing spondylitis and aortic regurgitation—description of the characteristic cardiovascular lesion from study of eight necropsy patients. Circulation 1973; 68: 1914–27.

Davies M, Moore B, Braimbridge M. The floppy mitral valve: study of incidence, pathology and complications in surgical, necropsy and forensic material. Br Heart J 1978; 40: 468–81.

Edwards J. The congenital bicuspid aortic valve. Circulation 1961; 23: 485–8.

Sadee AS, Becker AE, Verheul HA, Bouma B, Hoedemaker G. Aortic valve regurgitation and the congenitally bicuspid aortic valve: a clinico-pathological correlation. Br Heart J 1992; 67: 439–42.

Guiney T, Davies M, Parker D, Leech G, Leatham A. The aetiology and course of isolated severe aortic regurgitation: a clinical, pathological and echocardiographic study. Br Heart J 1987; 58: 358–68.

Hanson T, Edwards B, Edwards J. Pathology of surgically excised mitral valves. One hundred consecutive cases. Arch Pathol Lab Med 1985; 109: 823–8.

Lachman A, Roberts W. Calcific deposits in stenotic mitral valves—extent and relation to age, sex, degree of stenosis in a study of 164 operatively excised valves. Circulation 1978; 57: 808–15.

Nishimura R, Schaff H, Shub C, Gersh B, Edwards W, Takik A. Papillary muscle rupture complicating acute myocardial infarction—analysis of 17 patients. Am J Cardiol 1983; 51: 373–8.

Olsen L, Subramanian R, Edwards W. Surgical pathology of pure aortic insufficiency—a study of 225 cases. Mayo Clin Proc 1984; 59: 835–41.

Passik C, Ackermann D, Pluth J, Edwards W. Temporal changes in the causes of aortic stenosis: a surgical pathological study of 646 cases. Mayo Clin Proc 1987; 62: 119–23.

Perloff J, Roberts W. The mitral apparatus. Functional anatomy of mitral regurgitation. Circulation 1972; 46: 227–39.

Peterson M, Roach R, Edwards J. Types of aortic stenosis in surgically removed valves. Arch Pathol Lab Med 1985; 109: 829–32.

Roberts W, Morrow A, McIntosh C. Congenitally bicuspid aortic valve causing severe, pure aortic regurgitation without superimposed infective endocarditis: analysis of 13 patients requiring aortic valve replacement. Am J Cardiol 1981; 47: 206–9.

Sadee AS, Becker AE, Verheul HA, Bouma B, Hoedemaker G. Aortic valve regurgitation and the congenitally bicuspid aortic valve: a clinico-pathological correlation. Br Heart J 1992; 67: 439–42.

Waller B, Morrow A, Maron B. Etiology of clinically isolated severe, chronic, pure mitral regurgitation: analysis of 97 patients over 30 years of age having mitral valve replacement. Am Heart J 1982; 104: 276–88.

William RC. Molecular mimicry and rheumatic fever. Clin Rheum Dis 1985; 11: 573–91.