Objektivierung von Herzgeräuschen beim Rind mittels Phonokardiographie

The objective of the present study was to examine if heart murmurs
in adult cattle can be objectively diagnosed using
phonocardiography. The examinations were conducted using
phonocardiograms (PCG) and required a scheme for evaluation. Twenty
cattle older than two years of age that were hospitalised at the
Clinic for Ruminants, Oberschleissheim, were used for the study. In
all of them, heart murmurs were detected during auscultation. The
Meditron Analyzer 4.0 (Welch Allyn GmbH & Co KG, Jungingen)
allowed the simultaneous recording of electrocardiogram (ECG) and
PCG. The ECG using a thoracic lead served as a timeline for the
PCG. In a pre-test (n = 20) so called threshold values were
established which marked the division between the “background
noise” (amplitudes between the heart sounds) and murmurs. These
threshold values (for positive and negative range) were established
for four different locations per side of the body (in the third and
fourth intercostals spaces (ICS), dorsally and ventrally,
re-spectively), as well as for the presentation of the PCG with use
of a 50, 100 Hz filter, respectively. In the subsequent main study
twenty animals were examined clinically and according to the
results of auscultation assigned to the heart murmur grades (HM)
0-6 and categorised into groups: HM “faint” (HM grade 0-2), HM
“moderately loud” (HM grade 3-4) and HM “loud” (HM grade 5-6). PCG
were recorded on all eight locations mentioned above. The quality
and number of acceptable PCGs of the right body side was very low,
thus these recordings were not used in the analysis. For the PCG
evaluation certain selected heart actions were presented in a
larger scale and printed on graph paper. Systole and diastole were
marked using the ECG and the threshold values were drawn in
manually. In addition, systole and diastole were divided into 10
and 20 equal intervals for the 50 Hz and 100 Hz presentations,
respectively. At these determined points the amplitudes were
measured and recorded if they were above/below the threshold values
or if no values could be measured at these points. For amplitudes
that were outside the threshold values relative frequency at each
point of measurement for each location and for 50 and 100 Hz
presentations was calculated. Comparing the sum of the amplitudes
outside the threshold values between the groups “faint” and “loud”
significant differences were found for the location 1 (third ICS,
dorsally, left side) in the systole at 50 Hz and for the loca-tion
2 (third ICS, ventrally, left side) in the systole at 100 Hz (p =
0.041 and p = 0.034). A significant difference between the groups
“faint” and “loud” was found for the loca-tion 1 in the systole at
50 Hz (p = 0.042) concerning the area under the curve using the
amplitudes measured at these 10 and 20 points of measurement,
respectively. In addition, the mean values of the amplitudes that
were outside the threshold values were calculated. Comparing the
groups “faint” and “loud” a significant difference was also found
for the location 1 in the systole at 50 Hz in the positive range (p
= 0.034) as well as for location 2 in the systole at 50 Hz in the
negative range (p = 0.045). Further investigations could confirm
the tendencies found in the present study. Additional
investigations with a larger number of animals could help to
classify heart murmurs. Besides, they might help to increase the
chances of obtaining usable measurements from the right side. Also,
an algorithm for automatic evaluation of PCGs based on the data of
the present study would be very helpful.