Q J Med 2000; 93: 685-688
© 2000 Association of Physicians
Commentary papers |
The patient with a systolic murmur: severe aortic stenosis may be missed during cardiovascular examination
From the Department of Cardiology, Guy's and St. Thomas' Hospitals, London, UK
Summary
Significant aortic stenosis is prevalent amongst elderly people. It may be subclinical, manifesting only as a murmur, but can still cause unexpected death with little warning after symptoms develop. Recent studies have highlighted the unreliability of the classical clinical signs of severe aortic stenosis, leading to concern that some patients may not be referred appropriately for echocardiography. Here, we review the evidence for the accuracy of each sign. We suggest that the assessment of the patient with a systolic murmur should be reappraised, and offer guidelines toward improving the recognition of aortic stenosis in the community.
Introduction
Aortic stenosis is common. As many as 5% of all people aged over 75 in the Helsinki Ageing Study had moderate or severe stenosis.1 The natural history of aortic stenosis consists of a long latent period during which sudden death is uncommon. Mortality rises sharply soon after the onset of symptoms to 3% in the first few months2 and around 50% at 3 years.3 In addition, mortality on the surgical waiting list for aortic valve replacement has been reported to be as high as 7%.4 The detection of severe aortic stenosis is therefore important, but the diagnosis may be missed until the post-mortem examination in up to one half of cases.5 Two important reasons for missing the diagnosis are variations in clinical skill among physicians and, more importantly, limitations in the clinical assessment of severe aortic stenosis.
The clinical assessment of aortic stenosis
We are taught by clinical textbooks and in preparation for college examinations that severe aortic stenosis causes a loud ejection systolic murmur, a soft or absent second heart sound, a slow-rising carotid pulse and systemic hypotension with a reduced pulse pressure. All of these signs, however, are unreliable.
Systolic murmur
Almost all patients with moderate or severe aortic stenosis have an audible systolic murmur.6 However, the site of maximum intensity does not aid differentiation from mitral regurgitation, and the murmur of aortic stenosis may be most easily audible in the ‘mitral area’. This may cause a mistaken diagnosis of ischaemic mitral regurgitation in a patient with severe aortic stenosis and angina. The absence of a murmur over the right clavicle can help to exclude aortic stenosis.7 It has been suggested that the grade of murmur and timing of peak intensity may correlate to the severity of stenosis. However, in 123 asymptomatic patients with aortic stenosis examined by a single consultant cardiologist, a grade 3/6 murmur or above predicted a peak gradient of more than 64 mmHg with 90% specificity but only 29% sensitivity.8 In this series, mild stenosis was usually associated with a short soft murmur, but murmur intensity can be a poor predictor of the severity of aortic stenosis in an unselected population if patients with left ventricular failure are included.9
Carotid upstroke and amplitude
Amongst 781 unselected elderly patients in a long-term health care facility examined by a consultant cardiologist, a prolonged carotid upstroke time was reported in 53% with severe but also in 33% with moderate aortic stenosis.9 Similarly, 68% of the patients with severe aortic stenosis in Munt's group had normal or near-normal carotid amplitude.8 However a prolonged carotid upstroke occurred in only 3% with mild aortic stenosis.9
Soft or absent second heart sound
In a selected group of 231 patients referred for catheterization, 73% of all patients with aortic stenosis and 24% of those without were judged to have a soft or single second heart sound.10 In a further series of 397 patients with isolated aortic stenosis, the second heart sound was absent in only 9%.11 In Aronow's group of unselected cases, the absence of the second sound did not distinguish between moderate and severe aortic stenosis.9
Hypotension and reduced pulse pressure
Between 22% and 40% of patients requiring valve replacement for isolated aortic stenosis have a systolic blood pressure of 130 mmHg or more.11,12 In post-mortem studies, 
50% of patients with aortic stenosis have systemic hypertension.13 Similarly, only 6.6% of patients referred for aortic valve replacement had an aortic pulse pressure, measured invasively, of <35 mmHg.11
Combinations of signs
Etchells et al. recently applied a clinical prediction rule to a selected group of 124 patients referred for echocardiography.7 Whilst the combination of a right clavicular murmur and at least three of reduced carotid upstroke, reduced carotid volume, quiet second heart sound and maximal murmur intensity at the right sternal edge strongly predicted moderate or severe aortic stenosis, the presence of only two of these four signs was inconclusive.
Why do these observations differ from the clinical findings traditionally expected in aortic stenosis? This has chiefly been explained by the changing aetiology and demographic profile of the condition, now largely degenerative and prevalent in the elderly. In 1947, all cases were judged to be rheumatic on pathological examination, and the average age at death was 55 to 65 years.14 The elderly are more likely to be hypertensive, and atherosclerosis of the large arteries can result in a relatively sharp carotid upstroke. Signs may vary according to flow rate across the aortic valve with changing heart rate and cardiac output and concomitant or secondary left ventricular dysfunction will result in a shorter and quieter systolic murmur.15
Variation in clinical skills
Are studies that employ consultant cardiologists applicable to general and primary care physicians? Although systemic blood pressure can be reliably measured, cardiac auscultation and carotid palpation are highly subjective and poorly reproducible. No studies have compared the auscultatory skills of consultant cardiologists against those of junior doctors, and a retrospective analysis of 169 patients referred for echocardiography by general physicians showed that the auscultatory findings were not predictive of echocardiographic abnormalities.16 Similarly, a group of 453 trainees in internal medicine and family practice were poor at identifying digitized recordings of murmurs and additional heart sounds.17
Clinical implications
If the clinical signs of aortic stenosis are unreliable, should every patient with a systolic murmur be selected for echocardiography?
Severe aortic stenosis is common, but may be asymptomatic. A small number of patients with aortic stenosis present late with overt symptoms and signs of cardiac failure,18 which may be precipitated by another clinical event such as a chest infection. Early features may be relatively non-specific, such as palpitations, dizziness and fatigue, and some people may be unaware of a gradual decline in their exercise capacity. It is therefore important to be aware of the possibility of aortic stenosis, and we suggest that auscultation at the apex and right parasternal position should be performed in every patient aged over 70 years who presents with such symptoms. We would further advocate screening all patients above 70 if they visit a general practitioner or clinic even in the absence of cardiac symptoms, because the sharp mortality rise after symptom onset and the length of surgical waiting lists makes advance awareness of severe aortic stenosis important. Echocardiography should be requested in patients with a loud murmur or any suggestion of exertional symptoms. The current American College of Cardiology/American Heart Association guidelines for the management of valvular heart disease recommend echocardiography for ‘holosystolic or late systolic’ and ‘grade 3 or greater midsystolic’ murmurs.19
It is vital to be aware that supervening heart failure causes a fall in the grade of murmur. All patients with clinical signs of heart failure and a murmur require echocardiography. This must be performed before starting treatment with an ACE inhibitor. On the other hand, systolic murmurs are common, occurring in up to half of elderly patients, and mild aortic stenosis may take 15 or more years to progress to severe. Patients with soft, short ejection systolic murmurs and well-heard second sounds do not need echocardiography.20 Auscultation, although poor at differentiating moderate from severe aortic stenosis, is more reliable at confirming mild stenosis in those without symptoms.
Conclusion: which patients with a systolic murmur require an echo?
Aortic stenosis is common in the elderly and potentially fatal soon after or even before the onset of noticeable symptoms. The classical signs of severe aortic stenosis are often absent, and, in particular, systemic hypertension is common.
Every patient aged over 70 years should be auscultated routinely if they visit their general practitioner, and, if a systolic murmur is detected, questioned carefully for exertional symptoms. The genuinely asymptomatic patient with a combination of a quiet grade 1 or 2/6 murmur, normal carotid amplitude and a normal second heart sound is very unlikely to have severe aortic stenosis, and does not require an echocardiogram. For all other patients, a routine echocardiogram should be requested to exclude significant aortic stenosis. All patients with suspected heart failure and a murmur should have an echo before an angiotensin-converting-enzyme inhibitor is commenced.
Notes
Address correspondence to Dr P. Das, Department of Cardiology, 6th Floor, East Wing, St. Thomas’ Hospital, Lambeth Palace Road, London SE1 7EH. e-mail: pauldas{at}compuserve.com 
References
1. Lindroos M, Kupari M, Heikkila J, Tilvi R. Prevalence of aortic valve abnormalities in the elderly: an echocardiographic sample of a random population. J Am Coll Cardiol1993; 21:1220–5.[Abstract]
2. Pellika PA, Nishimura RA, Bailey KR, Tajik AJ. The natural history of adults with asymptomatic, haemodynamically significant aortic stenosis. J Am Coll Cardiol1990; 15:1012–17.[Abstract]
3. Horstkotte D, Loogen F. The natural history of aortic valve stenosis. Eur Heart J1988; 9(Suppl E):57–64.
4. Lund O, Nielsen TT, Emmertsen K, Flo C, Rasmussen B, Jensen FT, Pilegaard HK, Kristensen LH, Hansen OK. Mortality and worsening prognostic profile during waiting time for valve replacement in aortic stenosis. Thoracic Cardiovasc Surg1996; 44:289–95.[ISI][Medline]
5. Andersen JA, Hansen BF, Lyngborg K. Isolated valvular aortic stenosis. Clinico-pathological findings in an autopsy material of elderly patients. Acta Med Scand1975; 197:61–4.[ISI][Medline]
6. Roldan CA, Shively BK, Crawford MH. Value of cardiovascular physical examination for detecting valvular heart disease in asymptomatic subjects. Am J Cardiol1996; 77:1327–31.[ISI][Medline]
7. Etchells E, Glenns V, Shadowitz S, Bell C, Siu S. A bedside clinical prediction rule for detecting moderate or severe aortic stenosis. J Gen Int Med1998; 13:699–704.[ISI][Medline]
8. Munt BI, Legget ME, Kraft CD, Miyake-Hull CY, Fujioka M, Otto CM. Physical examination in valvular aortic stenosis: correlation with stenosis severity and prediction of clinical outcome. Am Heart J1999; 137:298–306.[ISI][Medline]
9. Aronow WS, Kronzon I. Prevalence and severity of valvular aortic stenosis determined by Doppler echocardiography and its association with echocardiographic and electrocardiographic left ventricular hypertrophy and physical signs of aortic stenosis in elderly patients. Am J Cardiol1991; 67:776–7.[ISI][Medline]
10. Hoagland PM, Cook EF, Wynne J, Goldman L. Value of noninvasive testing in adults with suspected aortic stenosis. Am J Med1986; 80:1041–9.[ISI][Medline]
11. Lombard JT, Seltzer A. Valvular aortic stenosis. A clinical and haemodynamic profile of patients. Ann Int Med1987; 106:292–8.
12. Mautner GC, Mautner SL, Cannon RO, Hunsberger SA, Roberts WC. Clinical factors useful in predicting aortic valve structure in patients >40 years of age with isolated valvular aortic stenosis. Am J Cardiol1993; 72:194–8.[ISI][Medline]
13. Wren C, Petch MC. Calcific aortic stenosis. J R Coll Physicians Lond1983; 17:192–5.[ISI][Medline]
14. Wood P. Diseases of the Heart and Circulation. Eyre and Spottiswoode, 1950.
15. Come PC, Riley MF, Ferguson JF, Morgan JP, McKay RG. Prediction of severity of aortic stenosis: accuracy of multiple noninvasive parameters. Am J Med1988; 85:29–37.[ISI][Medline]
16. Fink JC, Schmid CH, Selker HP. A decision aid for referring patients with systolic murmurs for echocardiography. J Gen Int Med1994; 9:479–84.[ISI][Medline]
17. Mangione S, Nieman LZ. Cardiac auscultatory skills of internal medicine and family practice trainees. A comparison of diagnostic proficiency. JAMA1997; 278:717–22.[Abstract]
18. Morgan DJR, Hall RJC. Occult aortic stenosis as a cause of intractable heart failure. BMJ1979; 1:784–7.
19.
Bonow R, Carabello B, de Leon A. ACC/AHA guidelines for the management of patients with valvular disease. Executive summary. Circulation1998; 98:1949–84.
20. Prendergast BD, Banning AP, Hall RJC. Valvular heart disease. Investigation and management. Recommendations of a working group of the British Cardiac Society and The Research Unit of the Royal College of Physicians. London, RCP, 1996.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. H Little, K.-L. Chan, and I. G Burwash Impact of blood pressure on the Doppler echocardiographic assessment of severity of aortic stenosis Heart, July 1, 2007; 93(7): 848 - 855. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J Bermejo The effects of hypertension on aortic valve stenosis Heart, March 1, 2005; 91(3): 280 - 282. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O.M.P. Jolobe Systolic murmur and aortic stenosis QJM, January 1, 2001; 94(1): 49 - 49. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||