Relationship between Teat-End Shape, Teat-End Callosity and Somatic
Cell Count in Dutch Dairy Goats
E. Middeldorp , G. Koop
Abstract
A study in 5 herds, where 464 dairy goats were included, was set up to find out if there was a
relationship between teat-end shape, teat-end callosity and somatic cell count (SCC) in dairy
goats. During a visit all the teat-ends were scored on the rate of teat-end callosity thickness,
teat-end callosity roughness and teat-end shape. Also milk samples were taken from each goat
to determine the somatic cell count. Teat-end callosity thickness (TECT) was scored in a
range from 1 to 5 where 1=no callosity ring and 5=extreme callosity thickness and it was
found that TECT score 4 had a significant higher SCC than TECT score 1, TECT score 3 had
a significant higher SCC than TECT score 2, TECT score 4 had a significant higher SCC than
TECT score 2 and TECT score 4 had a significant higher SCC than TECT score 3.
Teat-end callosity roughness and teat-end shape were found to have no significant influence
on the somatic cell count.
Introduction
The dairy goat sector is upcoming in the Dutch agriculture business. The number of dairy
goats has increased from 98 077 in the year 2000 to 207882 in 2008 (CBS, 2009) while the
number of dairy goat farms has decreased from 838 in the 2000 to 655 in 2008 (CBS, 2009)
of which about 350 are professional dairy goat farms keeping about 170.000 dairy goats
(LTO, 2009; VKGN, 2009). These numbers show that the dairy goat sector has become more
and more intensified. Therefore it is necessary to understand more about udder conditions and
which aspects are responsible for changes in these conditions, to make sure that the quality of
the milk and the general health of the dairy goats stays at a high level or are brought up to a
higher level.
In dairy cows several studies were done about what factors can influence the somatic cell
count (SCC). Teat-end shape was found to have no significant effect on the SCC (Chrystal et
al., 2001; Chrystal et al., 1999). Teat-end lesions neither affected the SCC, however wider teat
diameters showed an increase in SCC (Chrystal et al.,1999).
Other studies in dairy cows have studied the effect of teat-end callosity (TEC) on occurrence
of clinical mastitis in dairy cows, and found that clinical mastitis cows had more TEC than
their healthy herd mates, particularly when clinical mastitis occurred between the second and
fifth months of lactation. (F. Neijenhuis et al, 2001).
Although the meaning of the SCC in dairy goat milk is not completely understood yet (E. A.
Droke et al, 1993) it is used in this study as a parameter for udder conditions. The SCC is
linked to TEC score and teat-end shape. The reason that also the teat-end shape has been
chosen while it had no significantly influence on the SCC is that this study was done in dairy
cows and not in dairy goats. The main goal of this study is to describe the relationship
between TEC and teat-end shape and the SCC.
Materials en methods
In this study 5 commercial Dutch dairy goat herds were included. A total of 464 dairy goats
was included in this study and they were equally divided over the herds. Each herd is visited
once for collecting data. This visit was used to score TEC and teat-end shape of both teats of
all goats involved in this study. The visual scoring took place immediately after cluster
removal and was done separately by two researchers. To score the TEC, a classification
system, described by F. Neijenhuis et al. (F. Neijenhuis et al, 2000; F. Neijenhuis et al, 2001),
was used. This classification system scores the TEC thickness (TECT) in a range from 1 to 5
(score 1 means no callosity ring and score 5 means extreme thick callosity ring), and also
scores the TEC roughness (TECR), scores are 1 or 2 (where 1 = rough callosity ring, and 2 =
smooth callosity ring). Teat-end shape is divided into four categories (1=round, 2=flat,
3=pointed, and 4=funnel shaped) as described by M.A. Chrystal et al (M.A. Chrystal et al,
1999; M. A. Chrystal et al, 2001).
Milk SCC determination
The SCC was determined by Fossomatic 5000 (FOSS), Hillerød, Denmark.
Analysis of data
For the analysis of the data only the goats were included where the data of both visits was
known. The analysis were carried out in several step, first 1) Relationship between TECT and
SCC, second 2) relationship between TECR and SCC and third 3) relationship between teatend shape and SCC. To test if there was a relationship an one-way anova was used and the
LSD (Least Significant Difference) test was used in the post hoc test. But before these tests
were carried out the data was checked for the assumptions of normality, with the
Kolmogorov-Smirnov test and homogeneity of variance was tested with the Levene’s test.
Therefore the SCC was converted into log(SCC).
Other analysis
Other factors such as days in milk (DIM) and parity were also examined for their relationship
with SCC and TEC. A linear regression model was used to test if there was a relationship
between DIM and the SCC. An one-way anova was used to test if there was a relationship
between parity and SCC and the LSD test was used in the post hoc test.
Results
Study Population and Descriptive Results
During the study period, 464 different dairy goats were observed during the farm visits. Teatend callosity and teat-end shape of all animals was scored. Average TECT score was 2.26,
ranging per farm from 1.99 to 2.49. On average, 31.9% of the udder halves of the dairy goats
had rough callous rings, ranging from 2.2% to 55.4% among farms. The average partitioning
in teat-end shape among farms was round 87.8%, flat 1.9%, pointed 9.4% and funnel-shaped
0.9%.
Repeatability of Classification
TECT classifications by two observers were identical for 65% of the teats. For TECR, 82% of
teats were identically scored. For teat-end shapes, 77% teats were scored identically. The
Kappa coefficients for the two observers for the thickness of the callosity ring were 0.295 (SE
= 0.028), 0.576 for roughness (SE = 0.029) and 0.308 for teat-end shape (SE = 0.034).
Relationship between TECT and SCC
Arithmetic
Mean SCC
(x 1000)
1046
N
(Udderhalf)
49
SD
789
2
1170
598
1473
3
1514
266
1825
4
2683
14
1710
Total
1285
927
1575
TECT
1
Table 1. Mean SCC associated with the TECT score (score 1= no callosity ring, score 5= extreme thick callosity
ring). Score 5 was not given in this study.
Table 1 shows the arithmetic mean SCC for each TECT score. To show if there is a relation
between TECT and SCC an one-way anova was used.
TECT(I)
1
2
3
4
Mean
Difference (I-J)
0.084
-0.014
SE
0.079
0.083
P-value
0.288
0.870
4
-0.444(*)
0.162
0.006
1
-0.084
0.079
0.288
3
-0.098(*)
-0.528(*)
0.014
0.039
0.144
0.083
0.013
0.000
0.870
2
0.098(*)
0.039
0.013
4
-0.431(*)
0.146
0.003
1
0.444(*)
0.528(*)
0.431(*)
0.162
0.144
0.146
0.006
0.000
0.003
TECT(J)
2
3
4
1
2
3
* The mean difference is significant at the .05 level.
Table 2. Differences in TECT and log(SCC) calculated from the one-way anova model
Table 2 shows that the variance in SCC significantly differs between TECT scores. It was
found that TECT score 4 had a significant higher SCC than TECT score 1, TECT score 3 had
a significant higher SCC than TECT score 2, TECT score 4 had a significant higher SCC than
TECT score 2 and TECT score 4 had a significant higher SCC than TECT score 3.This
suggest that more TEC is related to a significantly higher mean SCC. There were no score 5
given in TECT.
Relationship between TECR and SCC
Arithmetic
Mean SCC
(x 1000)
N
(Udderhalf)
1545
296
1913
smooth
1163
631
1373
Total
1285
927
1575
TECR
rough
SD
Table 3. Mean SCC associated with TECR score
The values in table 3 did not pass the homogeneity of variance test. Therefore the MannWhitney test is used tot determine whether the differences in SCC can be associated with the
two TECR scorings. The result of this test had a P-value of 0.211. Therefore the conclusion is
that the difference in SCC between rough TEC and smooth TEC is not significant.
Relation between teat-end shape and SCC
Teat-end
shape
round
flat
pointed
funnel
shaped
Total
Arithmetic
Mean SCC
(x 1000)
1320
N
(Udderhalf)
814
SD
1612
996
18
1413
1053
87
1282
869
8
418
1285
927
1575
Table 4. Mean SCC associated with teat-end shape
Table 4 shows the arithmetic mean SCC for each teat-en shape. We found no significant
differences in SCC for the different teat-end shapes (round vs. flat P=0.113; round vs. pointed
P=0.113; round vs. funnel shaped P=0.839; flat vs. pointed P=0.443; flat vs. funnel shaped
P=0.290; pointed vs. funnel shaped P=0.497)
Relationship days in milk and SCC
To test if there was any relation between days in milk and SCC a linear regression model was
used. The R square value for all the data was R2=0.004. This has also been calculated within
each parity and the R2 for each parity were respectively for lactation 1=0.013, lactation
2=0.008, lactation 3=0.048 and lactation 4 and higher=0.114. Overall the R square is slightly
higher within each lactation but still relatively close to 0.
Relationship parity and SCC
Arithmetic
Mean SCC
(x 1000)
759
N
(Udderhalf)
286
SD
1229
2
1414
454
1614
3
1506
135
1392
4 and higher
2477
52
2255
Total
1285
927
1575
Parity
1
Table 5. Mean SCC associated with parity
Table 5 shows the arithmetic mean SCC for each parity. Parity 4 and higher are joined
together. To show if there is a relation between parity and SCC an one-way anova was used.
Parity (I)
1
Parity (J)
2
3
4 and higher
2
3
SE
0.038
0.053
P-value
0.000
0.000
-0.63(*)
0.076
0.000
1
0.33(*)
0.038
0.000
3
-0.10
-0.31(*)
0.42(*)
0.050
0.074
0.053
0.050
0.000
0.000
4 and higher
1
2
4 and higher
4 and higher
Mean
Difference
(I-J)
-0.33(*)
-0.42(*)
1
2
3
0.10
0.050
0.050
-0.21(*)
0.082
0.010
0.63(*)
0.31(*)
0.21(*)
0.076
0.074
0.082
0.000
0.000
0.010
* The mean difference is significant at the .05 level.
Table 6. Differences in parity and log(SCC) calculated from the one-way anova model
Table 6 shows that there is a significant differences in SCC between each parity. It also shows
that an increasing parity number has a significant increase in SCC .
Discussion
First point of discussion is that this research is based on a moment prerecording. Other studies
have followed the proceedings in a period as long as 1.5 year (F. Neijenhuis et al, 2001) and
by doing this you get a better overview of what the different influences are. In this study no
significant influence was found between teat-end shape en SCC, this is in agreement with
other studies done with dairy cows (M. A. Chrystal et al, 1999; M. A. Chrystal et al, 2001).
TECR also does not seem to influence the SCC. Furthermore not all the data was collected at
the same time and there was sometimes a period of a month in between scoring the teats and
the date the milk sample was taken to determine the SCC. Little is known about how fast
these TEC rings develop and also it is not yet completely understood how fast the SCC
fluctuate in dairy goats and therefore it is possible that the relations found in this study are not
so clear as they seem.
Before the study started both researchers had little experience in scoring the teat-ends. By
assessing the scores it is notable that there is more variation in teat-ends in the last visits than
in the first visits. This rises the question whether there is really more variation in teat-ends in
the last visits or that the researches are just more refined in scoring the teat-ends during the
last visits. Overall the Kappa coefficients were quite low but this can also be the consequence
of the subjectivity of the test because every score was visually made and therefore the score of
a teat can be slightly different between the two researchers if they look at it from a slightly
different angle.
Also the influences of the SCC on production and health of the dairy goats is not entirely
known.
Conclusion
The most important conclusion of this study is that TECT is significant related to the SCC.
More study has to be done to completely understand the effect of the SCC on the health and
production of a dairy goat. Also there has to be more insight about what exactly causes an
increase in TEC. When this is done specific advise can be given back to the dairy goat owners
about how to handle these problems.
References
-
-
-
-
-
-
CBS (Centraal bureau voor Statistiek), published on 25-3-3009, <
http://statline.cbs.nl/StatWeb/publication/?VW=T&DM=SLNL&PA=71284ned&D1=
445-449,485-489&D2=a&HD=091005-1700&HDR=G1&STB=T >, visited on 10th
august 2009
Chrystal M. A., A. J. Seykora and L. B. Hansen, Heritabilities of teat end shape and
teat diameter and their relationships with somatic cell score, 1999, Journal of Dairy
Science 82, No. 9, pp. 2017–2022
Chrystal M. A., A. J. Seykora, L. B. Hansen, A. E. Freeman, D. H. Kelley and M. H.
Healey, Heritability of teat-end shape and the relationship of teat-end shape with
somatic cell score for an experimental herd of cows, 2001, Journal of Dairy Science
Vol. 84, No. 11, pp 2549–2554
Droke E. A., M. J. Paape and A. L. Di Carlo, Prevalence of High Somatic Cell Counts
in Bulk Tank Goat Milk, 1993, Journal of Dairy Science 76, No. 4, pp. 1035-1039
LTO (Land en tuinbouworganistatie Nederland), <http://www.lto.nl/nl/5141892Geitenhouderij.html>, visited on 10th august 2009
Neijenhuis F., H.W.Barkema, H. Hogeveen and J. P. T. M. Noordhuizen,
Classification and longitudinal examination of callused teat ends in dairy cows, 2000,
Journal of Dairy Science 83, No. 12, pp. 2795–2804
Neijenhuis F., H. W. Barkema, H. Hogeveen and J. P. T. M. Noordhuizen,
Relationship between teat-end callosity and occurrence of clinical mastitis, 2001,
Journal of Dairy Science 84, No.12, pp. 2664–2672
VKGN (Vereniging Kwaliteitszorg Geitenhouderij Nederland),
<http://www.kwaligeit.nl/melk.htm>, visited on 10th august 2009