Abdominal Bruits: Introduction
Toward the end of an unusually busy clinic, a clinical clerk greets the final patient of the day, a man with a recently documented increase of blood pressure. With all the enthusiasm that remains after 4 years of medical training, she compulsively listens for abdominal bruits. Almost surprised, she hears a soft systolic-diastolic epigastric bruit and is faced with the inevitable question: so what?
Why Is This an Important Question to Answer With a Clinical Examination?
As we have gained insight into the origin and meaning of vascular bruits, detailed auscultation of the abdomen has become more common. Once detected, an abdominal bruit often is characterized according to pitch, timing, amplitude, and location in an effort to detect and document pathologic states, such as renovascular disease, splenic enlargement, hepatic cirrhosis, carcinoma of the pancreas and liver, splenic and hepatic vascular abnormalities, intestinal vascular insufficiency, and aortic disease. More recently, abdominal bruits have been documented in a substantial percentage of healthy individuals.
Although the search for an abdominal bruit has become part of the general physical examination, it also has been recommended as a key element of the examination of the hypertensive patient, in whom the presence of an abdominal bruit is considered to be an important feature of renovascular hypertension.1-3
To be of value, a diagnostic investigation (such as eliciting an abdominal bruit in the setting of hypertension) must reliably predict the presence or absence of disease (in this case, renovascular hypertension). This process should influence the course of management or prognosis. With this in mind, the reliability and accuracy of auscultating for an abdominal bruit in a patient with hypertension will be examined.
The Anatomic and Physiologic Origin of the Abdominal Bruit
Whereas turbulent flow within a vessel is the physiologic basis for a bruit, the pitch and radiation are a function of the flow and direction of the turbulent stream. Intrinsic or extrinsic abnormalities can produce turbulence, and although these abnormalities usually arise from within the abdomen, they can also arise from the inguinal area, retroperitoneum, or thorax.
Prevalence of Abdominal Bruits
The prevalence of bruits in different groups is summarized in Table 3-1. In “normal” populations (individuals without hypertension), the presence of any abdominal bruit has been detected in 6.5% to 31% of patients, with a predilection for the younger age groups (Figure 3-1). Among normal individuals older than 55 years, the prevalence was 4.9%. It is generally believed that the short, faint, and midsystolic bruit heard in these asymptomatic patients is “innocent.”7
Table 3-1The Prevalence of Abdominal Bruits |Favorite Table|Download (.pdf) Table 3-1 The Prevalence of Abdominal Bruits
|Reference, y ||Age, y ||No. and Study Group ||Prevalence, % |
|General Population |
|Edwards et al,4 1970 ||17-30 ||200 healthy volunteers ||6.5 |
|Julius and Stewart,5 1967 ||Unknown ||170 volunteers ||16 |
|Rivin,6 1972 ||16-85 ||426 patients without cardiovascular or intra-abdominal disease ||18 |
|Watson and Williams,7 1973 ||13-71 ||161 psychiatric patients ||31 |
|13-78 ||200 patients referred with gastrointestinal complaints ||27 |
|Patients With Hypertension |
|Julius and Stewart,5 1967 || ||155 patients referred with hypertension ||28 |
|Patients With Angiographically Proven Renal Stenosis |
|Hunt at al,8 1974 ||6-63 ||100 patients referred for investigation of hypertension ||87 |
|Perloff et al,9 1961 ||17-72 ||54 patients referred with sustained hypertension ||78 |
The Prevalence of Bruits Varies With Age in Normal Populations
In patients with angiographically proven renal artery stenosis, bruits have been documented in 77.7% to 86.9% of cases, with higher prevalence than the 28% observed among unselected patients referred for hypertension.5, 8, 9 In a study by Grim et al,10 the systolic-diastolic bruit was never detected in 379 normal subjects and was found in 1 of 199 patients with essential hypertension.
Eppier et al11 distinguished the presence of abdominal bruits in fibromuscular hyperplasia of the renal artery from that in atherosclerotic lesions. Their retrospective medical record review of 87 patients with surgically treated renal artery stenosis revealed a bruit in 77% of patients with fibromuscular disease and in 35% of patients with atherosclerotic disease.
How To Examine for Abdominal Bruits
The patient should be relaxed in a supine position, with the room quiet and with the examiner initially auscultating in the epigastrium, with moderate pressure applied to the diaphragm of the stethoscope. All 4 quadrants should be auscultated anteriorly. The auscultation should continue over the spine and flanks in the areas between T12 and L2 to rule out bruits that may be heard best posteriorly. However, no data exist that would support the routine auscultation of the back for abdominal or retroperitoneal bruits. Once detected, bruits can be correlated to the cardiac cycle by palpation of the carotid upstroke, with the systolic-diastolic bruit being more prolonged and extending into diastole.
Because the kidneys lie retroperitoneally and the renal arteries leave the aorta in the area cephalad to the umbilicus, attention should be given to auscultation in the epigastric area for the bruit of renovascular disease, a pancreatic neoplasm, or an innocent bruit (Figure 3-2). The bruit of a hepatic carcinoma has been heard in the right upper quadrant, whereas that of a splenic arteriovenous fistula has been described in the left upper quadrant. Periumbilical bruits are at times heard in the setting of mesenteric ischemia, and venous hums are from portosystemic hypertension. Finally, in the older population, an abdominal bruit may be associated with an abdominal aortic aneurysm. Estes,12 in a study of 102 patients with abdominal aortic aneurysms, demonstrated the presence of an associated bruit in 28% of cases.
Appropriate Areas of Auscultation
The Precision of Abdominal Auscultation for Bruits
Neither intraobserver nor interobserver variations in the way we elicit this sign have been evaluated in detail. However, Watson and Williams7 reported 92% (149/161) agreement when patients with celiac artery compression were prospectively examined by 2 examiners for the presence of an abdominal bruit. With standardization, auscultation of the abdomen can be performed with the appropriate degree of precision.
The Accuracy of Abdominal Auscultation in Renovascular Hypertension
This discussion will concentrate on abdominal bruits in fibromuscular and atherosclerotic renovascular disease. Because abdominal bruits occur in healthy individuals and in those with the nonrenovascular conditions listed in Table 3-2, they may occasionally yield false-positive findings in hypertensive patients.
Table 3-2Reported Nonrenovascular Causes of an Abdominal Bruita |Favorite Table|Download (.pdf) Table 3-2 Reported Nonrenovascular Causes of an Abdominal Bruita
|Reference, y ||Condition |
|Arida,13 1977 ||Splenic arteriovenous fistula |
|Bloom,14 1950 ||Hepatic cirrhosis |
|Clain et al,15 1966 ||Alcoholic hepatitis, hepatoma |
|Estes,12 1950 ||Abdominal aortic aneurysm |
|Goldstein,16 1968 ||Celiac artery compression syndrome |
|Lee,17 1967 ||Bacterial gastroenteritis |
|Matz and Spear,18 1969 ||Unilateral renal hypertrophy |
|McLoughlin et al,19 1975 ||Celiac artery stenosis |
|Sarr et al,20 1980 ||Chronic intestinal ischemic |
|Serebro and W'srand,21 1965 ||Pancreatic neoplasia |
|Shumaker and Waldhausen,22 1961 ||Hepatic arteriovenous fistula |
|Smythe and Gibson,23 1963 ||Tortuous splenic arteries |
Many studies describe the accuracy of the abdominal bruit in detecting renovascular disease in patients referred for hypertension, but only 3 demonstrate sufficient methodologic rigor (Table 3-3). These reports were of sufficient size and uniform clinical assessment, and the angiogram was the criterion standard. A further study by Julius and Stewart5 reported a sensitivity of 20%; however, specificity could not be estimated.
Table 3-3Accuracy of the Abdominal Bruit in Renovascular Hypertension |Favorite Table|Download (.pdf) Table 3-3 Accuracy of the Abdominal Bruit in Renovascular Hypertension
|Reference, y ||Type of Bruit ||Sensitivity, % (95% CIa) ||Specificity, % ||LR |
| || || || ||If Present ||If Absent |
|Grim et al,10 1979 ||Systolic and diastolic abdominal bruit ||25/64 = 39 (27-51) ||197/199 = 99 (98-100) ||39 ||0.6 |
|Fenton et al,24 1966 ||Any epigastric or flank bruit, including isolated systolic bruit ||17/27 = 63 (45-81) ||82/91 = 90 (84-96) ||6.4 ||0.4 |
|Perloff et al,9 1961 ||Systolic bruit ||78 ||64 ||2.1 ||0.35 |
Presence of Abdominal Bruits
The most useful study10 of the accuracy of abdominal auscultation assembled a consecutive series of patients referred to a university medical center for hypertension. All patients healthy enough for surgery underwent careful abdominal auscultation, with positive findings confirmed by a second examiner, plus other tests for renovascular hypertension, including arteriography. Of 64 patients with renovascular hypertension (an abnormal angiogram result and a renal vein renin ratio >1.5), 25 had combined systolic-diastolic abdominal bruits, for a sensitivity of 39% (95% confidence interval [CI], 27%-51%). Of 199 hypertensive patients with normal arteriogram results, 2 had systolic-diastolic bruits, for a specificity of 99% (95% CI, 98%-100%). Thus, although the absence of a systolic-diastolic bruit did not rule out renovascular hypertension, the presence of a systolic-diastolic bruit helped to rule it in, with a likelihood ratio (LR) of 39 (95% CI, 9.4-160).
A second study recorded any epigastric or flank bruits in a series of hypertensive patients undergoing arteriography.24 Not surprising, the sensitivity of 63% (95% CI, 45%-81%) for any bruit was higher than in the previous study, whereas the specificity for any bruit was somewhat lower, at 90% (95% CI, 84%-96%). Consequently, the presence of any systolic bruit confers a lower LR for renovascular hypertension (LR = 6.4; 95% CI, 3.2-13). Thus, the systolic-diastolic abdominal bruit is less sensitive (P = .04; χ21 = 4.36) and more specific (P< .01; χ21 = 13.5) than the combination of both isolated systolic and combined systolic-diastolic bruits.
Other than these studies and that by Perloff et al,9 additional studies of the accuracy of abdominal bruits in patients with hypertension are less rigorous and are not reported.
In summary, there is a substantial prevalence of systolic bruits in young, healthy patients, which increases in hypertensive patients, especially those with documented renovascular disease. In instances when the accuracy of the abdominal bruit has been rigorously assessed in evaluating patients with renovascular disease, the sensitivity has been reported to be between 20% and 78%, whereas the specificity has been between 64% and 90%. Systolic-diastolic bruits are seldom heard in healthy people or in patients with essential hypertension, but they are more common in individuals with renovascular disease. In patients with fibromuscular disease, there is an increased prevalence for all types of bruits.
Auscultatory Characteristics of Bruits
Although many bruits have been characteristically described as having a certain pitch, intensity, and location, the data to support this have been questioned.11, 19 Moser and Caldwell25 demonstrated a slightly increased prevalence of high-pitched bruits in association with renal artery disease (87%) when compared with the prevalence of medium-pitched or low-pitched bruits (57%). This finding supports the results of Julius and Stewart,5 who reported an increased prevalence (64%) of high-pitched bruits in these patients.
In the study by Moser and Caldwell,25 the intensity of the bruit described in patients with renovascular disease was less discriminatory, with 80% (17/21) of cases having loud bruits and 55% (16/29) having quiet bruits. These same authors described their results in predicting the localization of the stenosis. In their study, of the 13 patients in whom renovascular disease was isolated to 1 vessel, stenosis was correctly localized beforehand in 6 (46%). Eppier et al11 reported slightly better results because the site of the renovascular lesion was correctly localized in 70% of patients with fibromuscular disease and 43% of patients with atherosclerotic renovascular disease. Julius and Stewart5 directly auscultated the renal artery by using a sterile stethoscope at the time of renovascular surgery, demonstrating that, of 18 patients with bruits, in 9 the bruits were confined to the correct renal artery and in 7 the renal artery bruits were combined with additional vascular bruits. In 2 patients (11%), the bruits heard before surgery were secondary to other vascular abnormalities, and there were no bruits associated with the renal artery.
Prognosis of Patients With Hypertension and Bruits
Finally, the importance of identifying the location, pitch, and intensity of a bruit is questionable, and this issue awaits further clarification with larger prospective studies. Two reports have linked the presence of bruits to the outcome of renovascular surgery but with conflicting results. Eppier et al11 found that 84% of patients with systolic-diastolic bruits had favorable surgical results, compared with 55% of patients with only systolic bruits or no bruits. This result was replicated in patients whose renal artery stenoses were due to atherosclerosis, but the presence of diastolic bruits and the recent onset of hypertension correlated with favorable surgical outcomes in patients with both fibromuscular and atherosclerotic vascular disease. In contrast, Simon et al26 were unable to attach prognostic importance to abdominal bruits in patients with fibromuscular or atherosclerotic renovascular disease.
In view of the high prevalence (7%-31%) of innocent abdominal bruits in the younger age groups, if a systolic abdominal bruit is detected in a young, normotensive, asymptomatic individual, no further investigations are warranted. In view of the low sensitivity, the absence of a systolic bruit is not sufficient to rule out the diagnosis of renovascular hypertension. In view of the high specificity, the presence of a systolic bruit (in particular a systolic-diastolic bruit) in a hypertensive patient is suggestive of renovascular hypertension. Subsequent investigation should take into consideration the pretest likelihood of renovascular disease and full cost and potential benefits of any management decision. In view of the lack of evidence to support characterizing bruits as to pitch, intensity, and location, bruits should be reported only as systolic or systolic/diastolic. Existing information does not permit a definitive statement pertaining to the prognostic implication of a renal bruit.
In summary, the critical review of the literature pertaining to the abdominal bruit would suggest that the routine auscultation of the abdomen for the presence or absence of an abdominal bruit in the healthy asymptomatic population is of little value in view of the high prevalence of benign bruits. However, for our troubled clinical clerk, the presence of a systolic-diastolic bruit would provide supportive evidence of an underlying diagnosis of renovascular disease and should lead her to more aggressive investigation for this disorder.
Author Affiliation at the Time of the Original Publication
From the Department of Medicine, Ottawa (Ontario) Civic Hospital, Ottawa, Ontario, Canada.
The author and editors thank E. K. M. Smith, MD, for his helpful review of the manuscript.
EF. Renal artery disease: diagnosis and therapy. Geriatrics. 1963;8:761.
et al.. Experience coeliac axis compression syndrome
BH. Diagnostic significance of and murmurs. N Engl J Med.
PB. Epigastric bruits in patients without celiac axis compression. Ann Intern Med.
et al.. Hypertension secondary to renal artery occlusive disease. Circulation. 1961;24:1286–1304.
et al.. Abdominal bruits in renovascular hypertension. Am J Cardiol. 1976;37(1):48–52.
E. Abdominal aortic aneurysm. Circulation.
EJ. Splenic arteriovenous fistula with hypertension, varices
, and ascites. N Y State J Med.
S. Abdominal arterial murmurs in liver disease. Lancet.
R. Provision of health services: past, present, future. N Engl J Med.
P. Abdominal bruit due to unilateral renal hypertrophy. Lancet.
. Diastolic bruit in chronic intestinal ischemia. Dig Dis Sci.
MB. Diagnostic sign
of carcinoma of the body of the pancreas. Lancet.
JA. Intrahepatic arteriovenous fistula of hepatic artery and portal vein. Surg Gynecol Obstet. 1961;112:497–501.
DB. Upper-quadrant bruit due to tortuous artery. N Engl J Med. 1963;2:1308–1309.
JF. Diagnosis and results of surgery in renovascular hypertension. Lancet.
JR. Abdominal murmurs. Ann Intern Med. 1962;56:471–483.