[Skill Modules
>>
Heart Sounds & Murmurs
>>
Evidence Base
]
Evidence Base: Heart Sounds & Murmurs
Accuracy in Diagnosis of Diastolic Murmurs
The authors did extensive meta-analysis looking at the diagnosis
of aortic insufficiency. They found that cardiologists' precision (measuring
interobserver variability) for detecting diastolic murmurs was moderate when
they were using audiotapes (kappa=0.51) and was good in the clinical setting
(simple agreement, 94%). Not surprisingly, the precision of non-cardiologist
was less for all diastolic murmurs.
Most findings for aortic insufficiency were found to be neither sensitive nor
specific. The most useful finding for ruling in AI is the presence of
an early diastolic murmur and its absence for ruling out AI. Similarly,
the presence and absence of a mid-diastolic murmur significantly increases or
decreases the likelihood of mitral stenosis, respectively. The detection of
a typical PR murmur by a cardiologist significantly increased its probability
but its absence was not significant on the diagnosis.
| Clinical Findings for Detection of Heart Failure |
LR+ (95% CI) |
LR- |
| Presence of an early diastolic murmur to detect mild or greater AI |
8.8-32.0 (95% CI 2.8-32 to 16-630 |
|
| Presence of an early diastolic murmur to detect moderate or greater AI |
4.0-8.3 [95% CI, 2.5-6.9 to 6.2-11] |
|
| Absence of an early diastolic murmur to detect mild or greater AI |
|
0.1 [95% CI, 0.0-0.3] |
| Absence of an early diastolic murmur to detect moderate or greater AI |
|
0.1 [95% CI, 0.0-0.3] |
Source: data from two Grade A studies as summarized in Choudhry.
HCM = hypertrophic cardiomyopathy
VSD = ventricular septal defect
|
