Title	:	Robotic Milking and its Effects of Fertility
Source	:	Reasearch for Animal Husb…
Author	:	Krulp, Morice , Robert
Date	:	2002
Content	:

Robotic Milking and Its Effect on Fertility and Cell Counts

T. A. M. Kruip,* H. Morice,* M. Robert,* and W. Ouweltjes†
*Institute of Animal Science and Health (ID-Lelystad),
PO Box 65, 8200 AB Lelystad, Netherlands
†Research Institute for Animal Husbandry,
PO Box 2176, 8203 AD Lelystad, Netherlands

ABSTRACT

The aim of this study was to analyze the effect of robotic milking (RM) on fertility and somatic cell counts (SCC) among dairy herds participating in the national Dutch milk recording system. It was hypothesized that RM, and a higher milking frequency in general, would have negative effects on fertility, due to expected and supposed deeper negative energy balance (NEB). Herds increasing milking frequency from two to three times daily consistently had increased production. Milk production during RM was intermediate between the amounts obtained by milking twice versus three times a day. Milking three times a day and the associated higher production had no significant effect on reproductive measures such as nonreturn rate at 56 d post insemination (NR56) or days to first service. Although RM did not affect NR56, use of the robot was associated with an increase in days to first service. An increase in milking frequency from two to three times daily did not affect SCC, but SCC were significantly increased after milking with the robot. Robotic milking has a signifi- cant positive effect on production and no negative effect on fertility as measured by NR56. The effect of RM in increasing days to first service appears due to reasons other than increased production and a more NEB. Increased SCC during RM is potentially of concern. From the data available, the relationship of RM to clinical mastitis could not be determined but this aspect needs further attention.

(Key words: robot, milking, fertility, cattle)

Abbreviation key:

RM= robotic milking, NEB= negative energy balance, NR-56 = nonreturn rate at 56 d after insemination, 2x, 3x = number of times cows were

INTRODUCTION

Because of the effect milking robots have on increasing milk production in dairy cattle, it was expected that robotic milking (RM) could have an effect on fertility. Experiments have been implemented on the problem of reduced fertility when milk production increases (Butler and Smith, 1989; Nebel and McGilliard, 1993). However, Stefanowska et al. (1995, 1996) reported that the increased milking frequency with a milking robot, leading to higher yield, does not delay the occurrence of estrus postpartum if a special herd management (Devir et al., 1993) is implemented. Other researchers (Barnes et al., 1990) found similar results: increasing milking frequency leads to higher milk yield and effi- ciency of production without loss of reproductive effi- ciency. However, all those studies on the effects of milking frequencies on production and fertility were performed in relatively small herds and under experimental circumstances.

Again under experimental circumstances, neither milking three times per day (3x; Waterman et al., 1983) nor automatic milking (Klungel et al., 2000) affected udder health. Yet, tendencies have been shown that when there is an increase in milking frequency, a transitory increase in milk SCC has been observed (Hillerton and Winter, 1992).

For these reasons, data on fertility and SCC of cows that have been milked by robot from parturition onwards under commercial farm conditions and over a long period of time were examined and analyzed.

MATERIALS AND METHODS

The data for this study came from the National Dutch data file (CR Delta, Arnhem, The Netherlands) for milk recordings collected from 1994 to June, 2000. Data from herds milking twice or three times per day (2x and 3x, respectively) were available over all those years. Data from herds using robotic milking were available from 1996 through June, 2000 (Table 1).

The number of farms with a milking frequency of 2 is just an arbitrary and a select number, meant as a reference. Table 1 illustrates that the number of farms with consistently 3? milking is relatively steady but low, whereas the number of farms with RM has increased, particularly since 1997. The dataset contains information about: number of lactations, milking

Milking Frequency	2x	3x	RM
1194	364	24	0
1995	358	29	0
1996	357	35	2
1997	352	43	4
1998	346	53	63
1999	278	47	87
2000	250	37	84

frequency for each test day, dates when milking frequency changed, number of samples taken (number of cows ? number of test days) per milking frequency, average test-day yields per herd, weighted average SCC (cells/ ml ? 1000) per herd along with calving and insemination dates. From those data, the following variables were calculated: average yield and cell counts for the three milking frequencies and days to first service and nonreturn rate at 56 d (NR56) after first insemination.

The composition of the herd groups included opportunities for two levels of comparison:

1) Comparison among the three milking systems over different farms: Farms with a milking frequency of two times per day (2?), farms with a milking frequency of three times per day (3?) and farms with RM. The three groups varied in terms of number of animals, duration of use, and management strategy. Differences caused by management strategy can be confounding. Therefore, within each farm where data were available, the differences between the results obtained with one way of milking, and the results obtained with another way of milking were calculated leading to the second approach:

2) Comparison of the farms that have changed their milking system in the period under investigation, i.e., a comparison between systems within farms. In this case, for each farm, the differences between the results after and before the change were calculated. The new groups for comparison were: from 2? to 3?; from 2? to RM; from 3? to RM and from 3? to 2?.

Table 2. The average yield expressed in kg/day in the three milking frequency groups: 2? and 3? per day and robotic milking (RM).

Milking Frequency	2x	3x	RM
Number of farms	376	64	87
Average number of registrations(tests days x number of cows)	2039	2091	1097
Average milk yield(in kg/d +- SD)	24.4+-1.76a	30.3+-1.84b	27.1+-1.73e
a,b,cMeans within rows with different superscripts are significantly different (P < 0.001).

Figure 1. Mean production increase on 30 farms at change from two to three times milking per day.

Statistics

The mean test-day yield, the mean SCC, the NR56, and the mean days to first service per herd and per milking system were analyzed by ANOVA. Differences between groups were tested for significance by pairwise comparison in the Student’s t-test (Genstat 5, fourth version for windows).

RESULTS

Milk Production

Across all farms in the dataset, the increase of milking frequency led to higher average test-day yield (P < 0.001). Farms with 3? milking had an average test-day yield higher than farms with RM and 2? (Table 2).

To approach the effects of the change in the milking frequency on a farm, i.e., within the same management, data before and after the change were registered and compared within each farm. Data from four types of milking frequency change are presented in Table 3

Significant increase in milk yield is observed when a change in milking frequency from 2? to 3? occurred (from 26.2 to 31.5 kg/d; P< 0.001) and when a change in milking frequency occurred from 2? to RM (from 25 to 27.2 kg/d; P < 0.001). Those increases per farm are

Groups	2x to 3x	2x to RM	3x to RM	3x to 2x
Milking frequency	2x 3x	2x RM	3x RM	3x 2x
Numberof farms	30 30	81 12	12 12	8 8
Average number of registrations	1062 1150	2043 995	275 1720	1444 1290
Average milk yield(in kg/d)+-SD	26.2+-1.5a31.5+-1.5b	25.0+-1.6a27.2+-1.7b	27.5+-1.7a26.7+-1.5a	30.1+-2.0a26.0+-2.2b

(a,bMeans within rows and pairs of columns with different superscripts are different (P < 0.001).)

illustrated in Figures 1 and 2, respectively. The observation that the change over from 2? to RM on some farms (see Figure 2) led to a lower milk yield had a strong effect on the overall average milk yield by the robot. There was no significant change in average milk yield after changing from 3? to RM. That small sample of farms (n = 12) had lower average 3? milk yields at the start than other herds milking 3?. After changing from 3? to 2?, the average yield decreased substantially (Table 3).

Fertility

The influence of milking frequency on fertility has been measured by monitoring the NR56 and number of days to first service (Table 4). If the four groups of changed milking frequency are considered, no signifi- cant differences in NR56 exist. The relation between NR56 and the average milk yield for the three milking frequencies is presented in Figure 3, illustrating that the NR56 rate decreased slightly as production increased, regardless of milking system. The number of days to first service increased significantly (P < 0.001) after the change from either 2? or 3? milking to RM (Table 4).

(Figure 2. Mean production on 81 farms at the change from 2? to robotic milking.)

SCC

The average SSC for the three different milking systems are presented in Table 5. The mean SCC was always higher (P < 0.05) in the RM group than when those same herds had been previously milked either at frequencies of 2? or 3?. Changing from 2? to 3? or vice versa did not significantly affect SCC.

DISCUSSION

Milk Production

The presented data confirm the results obtained by others, demonstrating that more frequent milking leads to more milk yield (Amos et al., 1985; Barnes et al., 1990; Hillerton and Winter, 1992; Erdman and Varner, 1995; Kruip et al, 2000). The average milk yield in kilograms per day for a milking frequency of 3? was higher compared to the group with the milking frequency of 2?. This finding is in agreement with the conclusion of Erdman and Varner (1995), which was based on 19 reports in the literature. The mean production obtained with the RM lies in between the average productions of groups milked at 2? and 3?. Although the actual milking frequency of RM was not recorded in the current study, the average yield fits into the estimation of the average milking frequency that we made in an earlier study of RM: in which the frequency was 2.7 times per day (Kruip et al., 2000). The changes in average milk yields in kilogram per day were most obvious when the milking frequency was changed from 2? to 3? and from 2? to RM. The increase of average yield in kilograms per day after changing over from 2? to 3? is in theory about 15%, varying from about 10 (Klei et al., 1997) to about 25% (Amos et al., 1985; Campos et al., 1994) and in agreement with our finding of 20%. The increase in average milk yield in kilograms per day after change from 2? to RM was unexpectedly low (Table 2). This was probably due to the fact that the change to RM did not always result in an increase in milk yield in all herds. In this respect, it is relevant to mention that the intervals between milkings for 2? and 3? were very regular, whereas milking intervals

Table 4. Changes in nonreturn rates by 56 d after insemination (NR56) and the number of days to first service after the change to another milking frequency, 2? or 3? per

Groups	2x to 3x	2x to RM	3x to RM	3x to 2x
Milking frequency	2x 3x	2x RM	3x RM	3x 2x
Numberof first insemination	178 228	309 82	67 141	134 154
NR56 after first insemination+-SD	62.2+10.3a61.2+-8.6a	62.1+10.1a63.0+-8.7a	64.9+10.8a62.3+-8.5a	67.6+12.8a71.1+-9.2a
Days to first services+-SD	83+-13.7a88+-15.1a	85 +-14.2a 94+-20.1b	84+-10.8a98+-22.3b	90+-18.5a95+-23.0a

(Figure 3. Nonreturn rates for the three milking groups as a function of production.)

in the RM system might be much more irregular (Winter and Hillerton, 1995; Ketelaar-de Lauwere et al., 1999). We would like to speculate that the regularity of the intervals between milkings has some benefits and leads to more production (Figure 1). On the contrary, irregular intervals between milkings might lead to milk accumulation in the alveoli and less production by the secretory cells due to the autocrine down regulation of production (Knight and Wilde, 1993; Knight et al., 1994; Knight and Dewhurst, 1994). Fewer and irregular visits to the robot might count for the lower production in the change from normal milking to RM for some farms.

Fertility

From the literature, interactions between milk yield and reproduction have been observed (Butler and Smith, 1989; DeVries and Veerkamp, 2000). Shortly, yield increase leads to a deeper negative energy balance (NEB) postpartum and by that to lower levels of both glucose and insulin. In some respect, insulin is the driving force of reproduction (Kruip et al., 1999; 2001). Because a higher frequency of milkings leads to a higher yield but not to a proportional increase in feed intake, it consequently leads to a deeper NEB. Indeed, Butler and Smith (1989) explained the fact that the depth of the NEB was directly related to a longer postpartum interval to first ovulation and lower conception rates. Based on that observation we expected a reduction in fertility as milking frequency increased. However, under experimental conditions, most of the studies showed that more frequent milking does not delay the occurrence of estrus (Poole, 1982; Stefanowska et al., 1996;). Amos et al. (1985) also indicated that reproductive performance was not influenced by the milking frequency.

Groups	2x to 3x	2x to RM	3x to RM	3x to 2x
Milking frequency	2x 3x	2x RM	3x RM	3x 2x
Numberof farms	30 30	81 81	12 12	8 8
Number of registration	1062 1150	2043 995	275 1720	1445 1290
Average SCC+-SD	152*10 3+-7.5a157*103+-6.2a	149*103+-6.9a193*103+-7.8b	153*103 +-7.4a228*103+-8.3b	151*103+-6.1a95+-23.0a

Source: Research for Animal Husbandry
Authors: Krulp, Morice, Robert