The effect of flow-responsive pulsation on teat tissue condition and milking performance in Holstein dairy cows

The objectives of this study were to assess the effect of a reduced liner-open phase applied through flow-responsive pulsation (FRP), as a method to provide supplemental stimulation, on teat tissue conditions and milking characteristics in dairy cows. In 2 switch-back trials, 156 Holstein cows milked 3 times daily were assigned to the FRP or conventional (CON) group in alternating sequences. Trial I lasted for 35 d and was split into 5 alternating 1-week periods of FRP and CON. The duration of Trial II was 84 d, consisting of 4 alternating 3-week periods of FRP and CON. Pre-milking udder preparation for both groups consisted of predipping, forestripping and wiping the teats. Upon milking unit attachment, the FRP cows were milked at a pulsation rate of 50 cycles/min and a pulsation ratio of 30:70 until the preset milk flow threshold of 0.5 kg/min was reached. When the threshold value of 0.5 kg/min was reached, the pulsation was automatically switched to milking mode, which consisted of a pulsation rate of 60 cycles/min and a pulsation ratio of 70:30. Cows in the CON group were milked by milking mode (pulsation rate, 60 cycles/min; pulsation ratio, 70:30) immediately after attachment of the milking unit. We assessed machine milking-induced short-term changes to the teat tissue by palpation and visual inspection during Trial I, and we assessed teat-end hy-perkeratosis in Trial II. Electronic on-farm milk meters were used to assess milking characteristics [milk yield (kg/milking session), milking unit-on time (s), 2-min milk yield (kg), peak milk flow rate (kg/min), and duration of low milk flow rate (s)]. Generalized linear mixed models were used to analyze the effect of treat-ment on the outcome variables. The odds of machine milking-induced short-term changes to the teat tissue were lower for cows in the FRP group than for those in


INTRODUCTION
The advent of machine milking is one of the major technological breakthroughs that changed the scenario of dairy farming from small and medium household farming to large commercial dairy farms.Further refinement of the configuration and mode of operation during milking in recent decades has helped minimize the deleterious effects of its regular use on the cows' teats and udder health.Among those configurations, pulsation and vacuum settings offer a tradeoff between milking speed and teat tissue stress.These settings have been widely studied by several research groups (Smith and Petersen, 1946;Thomas et al., 1993;Gleeson et al., 2003) who have evaluated the different machine settings and various combination methods to increase milking speed, such as liner design (Spencer et al., 2007;Bade et al., 2009) or optimal premilking stimulation (Besier and Bruckmaier, 2016;Wieland et al., 2023;Singh et al., 2024).Premilking stimulation is recommended to induce oxytocin release and milk ejection before the start of milking, ensuring that the alveolar milk is available before the cisternal fraction is removed (Weiss and Bruckmaier, 2005;Watters et al., 2015).Results from recent studies showed that cows in late lactation and those in lower production stages require more time to achieve adequate milk ejection (Kaskous and Bruckmaier, 2011;Watters et al., 2012).Regardless of these findings, dairy operators usually apply a fixed premilking stimulation regimen because the application of different premilking stimulation regimens to groups of individuals is currently impractical.Consequently, the physiological requirements of late-lactation cows and low-producing animals are not met, resulting in delayed milk ejection.In a recent cross-sectional study of 5 New York State dairies, our group showed that 24% of cows were diagnosed with delayed milk ejection (Wieland et al., 2021).Delayed milk ejection has been associated with poor milking efficiency, impaired teat and udder health, and reduced milk yield (Bruckmaier, 2005;Erskine et al., 2019).
Premilking stimulation can be augmented with automated stimulation by applying different forms of pulsation to the teat via the teat cup's liner (Watters et al., 2015;Neuheuser et al., 2017;Neuheuser et al., 2018).Previous research has demonstrated the stimulatory effect of these systems with respect to the induction of oxytocin release and milk ejection when coupled with reduced vacuum (Neuheuser et al., 2018).Recently, a milk flow-controlled system for adjusting the pulsation rate and ratio (flow-responsive pulsation, FRP) during individual milking observations was introduced and evaluated by several researchers (Neuheuser et al., 2018;Tuor et al., 2022).The FRP system is linked with electronic on-farm milk meters.Through the application of this system, milking starts with a reduced lineropen phase (b-phase) immediately after milking unit attachment by means of a reduced pulsation rate (e.g., 50 cycles/min) and pulsation ratio (e.g., 30:70) until the milk flow reaches the assigned threshold.Then, the pulsation switches to the normal milking mode (e.g., pulsation rate 60 cycles/min; pulsation ratio 65:35).The results of a recent study (Tuor et al., 2022) showed that the FRP system is promising for optimizing the milk harvesting process in dairy operations.By means of the FRP system, cows whose physiological requirements for premilking stimulation are not met when the milking unit is attached (as indicated by a milk flow rate below the threshold value) could be subjected to supplemental stimulation to elicit their maximum milk ejection capacity.Tuor et al. (2022) reported that such systems may replace premilking stimulation, reduce the workload of milking technicians, and increase parlor efficiency.
During machine milking with a fixed vacuum setting, the teat tissues are likely to be exposed to higher vacuum levels during times of low milk flow (Rasmussen, 2004;Reinemann et al., 2013;Besier and Bruckmaier, 2016).Consequently, the risk of tissue alterations is increased under these milking conditions (Nyhan, 1968;Thiel et al., 1973).These alterations in the teat tissue can be visually observed as short-term changes (STC), such as swelling, discoloration, and firmness.Hamann and Mein (1996) showed that increasing the pulsation ratio from 50:50 to 80:20 caused increased teat tissue thickness.Conversely, a prolonged liner-closed phase (≥150 ms) has been shown to relieve teat congestion (Upton et al., 2016) and slow milk removal, therefore improving cisternal refill as long as milk ejection has not occurred (Neuheuser et al., 2018).The machine milking also imposes discomfort on the cows.Cows often express this discomfort in terms of increased hindleg activities such as stepping and kicking during milking (Rousing et al., 2004).Cerqueira et al. (2017) reported the association of these hindleg activities with overmilking.Supplemental stimulation provided through FRP may help to reduce cow discomfort during milking and decrease hindleg activity.
The effect of a dynamic control strategy to adjust pulsation during milking in an individual animal and its ability to provide supplemental stimulation have not been evaluated by rigorous methods.Additionally, there is little information regarding the efficiency of such milking systems and their effect on teat tissue conditions in Holstein cows milked thrice a day.Therefore, this study investigated the effect of FRP on the teat tissue conditions and milking performance of Holstein cows milked 3 times a day.The specific objectives of this study were to investigate the effects of FRP on postmilking teat tissue conditions and milking performance.We hypothesized that the application of FRP would augment meeting the individual cows' physiological requirements for teat stimulation and elicit their maximum milk ejection capacity, leading to a shorter milking duration, enhanced peak milk flow rate (PMF), reduced bimodality of milk flow, and improved postmilking teat tissue conditions.

MATERIALS AND METHODS
The study consisted of 2 trials and was conducted at the Teaching Dairy Barn of the College of Veterinary Medicine, Cornell University (Ithaca, NY) from February to July 2023.The protocol was reviewed and approved by the Cornell University Institutional Animal Care and Use Committee (protocol no.2022-0028).

Animals and Housing
The lactating herd consisted of approximately 160 Holstein cows.Cows were housed year-round in 2 freestall pens with sand bedding and were fed a TMR formulated according to NRC requirements (2001).The herd key performance indicators before the start of Trial I were as follows: average milk production, 14,266 kg; mean test day SCC, 218,154 cells/mL; monthly clinical mastitis incidence, 1.4%; and 21-d pregnancy rate, 24.3%.

Milking System
Cows were milked 3 times per day at 0400, 1100, and 1900 h in a double 10 parallel milking parlor (P2100, DeLaval International AB, Tumba, Sweden).The vacuum pump (7.5 kW) was regulated by a variable frequency drive and set to supply a receiver operator vacuum of 45 kPa.The milking unit was composed of the cluster MC70 (DeLaval International AB) and a milking liner with a square barrel shape (ProSquare DPX2, IBA, Millbury, MA).Pulsators (EP100, DeLaval International AB) provided side-to-side alternating pulsation and added features to provide 2 pulsation control strategies through adjunct software (Delpro FarmManager 5.1, version 2021.08.004;DeLaval International AB).The farm had been using the pulsator in single pulsation mode with a pulsation rate of 60 cycles/min and a pulsation ratio of 70:30.The FRP was set based on the cluster milk flow rate.For a cluster milk flow rate below the threshold (<0.5 kg/min), the pulsation rate was set to 50 cycles/min at a ratio of 30:70.However, at a milk flow rate of and above the threshold (≥0.5 kg/min), the pulsation rate was set to 60 cycles/min at a ratio of 70:30.The automatic cluster removers were set to remove the units at a milk flow of 1.4 kg/min with a 0-s delay and a vacuum decay time of 2.3 s.The milk sweep was initiated 1.5 s after unit retraction and lasted for 4 s.The milk line was installed 75 cm below the cow standing level.All milking system settings were verified and assessed by the investigators according to the guidelines outlined by the National Mastitis Council (NMC, 2012) before the start of the study.
The milking routine was performed by 2 operators per milking session.Premilking udder preparation was performed with sets of 5 cows in 5 steps and performed by milking operators in 3 visits.The whole operation consisted of the following steps (Figure 1).Visit 1: step 1, wipe of teats with a clean dry towel (drywiping); step 2, predip with 1% iodine (Multi Dose MD; DeLaval International AB) using a teat dip applicator cup.Visit 2: step 3, sequentially forestrip 2 streams of milk per quarter; step 4, wipe (i.e., clean, and dry) of teats with a clean cloth towel.Visit 3: step 5, attach the milking unit.After milking-unit detachment, all the teats were dipped in 1% iodine (Multi Dose MD; DeLaval International AB) using the same method as for predipping.

TRIAL I: Treatment Allocation
The study used a switch-back design to evaluate short-term teat tissue changes and milking characteristics.A full crossover experimental design would have been ideal but considering the impracticability of splitting the herd into 2 milking groups for the simultaneous application of 2 treatments, a switch-back design was preferred.All the healthy lactating cows milked in the parlor were enrolled for the study.The cows were milked using conventional (CON) pulsation settings for a period of 7 d before the milking settings were changed to FRP mode for the next 7 d.This alternating pattern was followed for 5 wk leading to 3 CON periods and 2 FRP periods (Figure 1).The durations of various premilking udder preparation steps were recorded on d 1 of each period.
Milking Characteristics.The milking characteristics [the milk yield (kg; MY), first 15 s milk flow rate (kg/min), 15 to 30 s milk flow rate (kg/min), 30 to 60 s milk flow rate (kg/min), 60 to 120 s milk flow rate (kg/ min), average milk flow rate (kg/min; AMF), PMF (kg/min), 2-min milk yield (kg; 2MIN), duration of low milk flow rate (s; LMF), and milking unit-on time (s; MUOT)] were assessed at each milking with electronic on-farm milk meters using near-infrared technology (MM27BC, DeLaval International AB) and recorded using dairy farm management software (DelPro, DeLaval International AB).The detailed definitions of various milking characteristics and abbreviations used are listed in Table 1.For subsequent analysis, a new categorical variable (i.e., bimodal milk flow) was created and defined as previously described (Wieland and Sipka, 2023): bimodal milk flow was present if any of the incremental milk flow rates (flow rates during 15-30 s, 30-60 s or 60-120 s) were lower than any of the Milking Irregularities.The occurrences of a milking liner slip and a milking unit kick-off were monitored with electronic on-farm milk meters in conjunction with the software program.The threshold limit for milking liner slip was set to 175.The occurrence of milkingunit reattachment was registered via the milking point controller (MPC680, DeLaval International AB) and documented with the software program.
Teat Tissue Condition.The assessment of the postmilking teat tissue condition was performed during session 2 by 1 trained investigator (MS), who was not blinded to the treatment.The milking machine induced STC was recorded visually and by palpation on d 3 and 4 of each period.The teats were evaluated within 60 s postmilking for the presence of swelling at the base and consistency of the teat end.The scoring system was adopted from Hillerton et al. (2000) and scored as described by Wieland et al. (2020).Briefly, within 60 s after unit detachment, the condition of the teat base was observed and scored as follows: no visible mark present (score of 1), visible mark present (score of 2), or significant swelling (score of 3).The consistency of the teat end was evaluated and scored as soft (1), firm (2), or wedging present (3).An STC was considered present if the teat base score was 3 or if the consistency at the teat-end score was ≥2 for 1 or more teats; an STC was considered absent otherwise (Wieland et al., 2020).
Vacuum Measurements.We collected measurements of the vacuum dynamics of milking observations from a haphazard subset of cows on d 3 and 4 of each period.For this purpose, we used an electronic vacuum measurement device (VaDia, BioControl, Norway AS).Four VaDia devices were installed at the first 2 stalls on each parlor side (stalls 1, 2, 11 and 12).We analyzed the data using adjunct software (VaDia Suite, version 1.15.0.932;BioControl) and evaluated the overmilking period, average cyclical vacuum fluctuations (assessed for 10 pulsation cycles at 60 s after the start of the peak milk flow period), and average mouthpiece chamber vacuum during the peak milk flow period as previously described (Wieland et al., 2021).In addition, we manually evaluated the duration of pulsation stimulation [i.e., duration of reduced liner-open phase (b-phase)] in each milking observation during the FRP mode of milking.

TRIAL II: Treatment Allocation
The second trial was planned to evaluate the longterm effects of FRP on teat-end hyperkeratosis (HK) and milking performance.The design of the study was similar to that of Trial I.The duration of each period was 3 wk, and the study was conducted for 12 consecutive wk, resulting in 2 CON periods and 2 FRP periods (Figure 1).

Data Acquisition
The characteristics of the enrolled cows and their milking characteristics were collected as in Trial I.The vacuum measurements were performed on d 2, 9, and 16 of each period and processed using the same method from Trial I.The durations of various premilking udder preparation steps were recorded on d 1 of each period.

Teat-end Hyperkeratosis.
Effect of the FRP mode on long-term teat tissue changes was evaluated based on HK and assessed on d 18 of each period during session 2 by 1 trained investigator (MS) who was blinded to the treatment.The HK of the teat end was recorded on a 4-point scale as previously described by Mein et al. (2001).Briefly, teat end callosity was scored as no callosity ring present (score of 1); presence of callosity ring without roughness (score of 2); callosity ring and roughness present; keratin fronds extending 1 to 3 mm from the teat orifice (score of 3); and callosity ring present with excessive keratin fronds extending ≥4 mm from the teat orifice (score of 4).Teat-end hyperkeratosis was considered present if the cow had 1 or more teats with a score ≥3, whereas HK was considered absent otherwise.
Hindleg Activity Measurements.We also extended our research in Trial II to evaluate the effect of the FRP on cow discomfort.For this purpose, we recorded the hindleg activity of the cows during milking session 2 on d 2, 9 and 16 of each treatment period.Cows' hindleg activity was recorded manually as the number of steps and kicks expressed by each cow during a complete milking session.The number of steps and kicks were recorded visually by 4 observers standing 2 at each corner of the parlor and recording each a subset of 5 cows.The steps and kicks were differentiated based on measurements adopted from previous studies of hindleg activity (Meyer et al., 2021;Raoult et al., 2021).We considered a 'step' when any of the hind legs was lifted less than 15 cm off the standing level during milking.Lifting the leg to 15 cm or above during milking was considered a kick.The counting started immediately after the completion of milking unit attachment and continued until complete detachment of the milking unit.

Analytical Approach
The data were maintained in Excel spreadsheets (Microsoft 365 Excel; Microsoft Corp., Redmond, WA) and JMP Pro (ver.17.2.0;SAS Institute, Inc., Cary, NC).Before the statistical analyses, the data were screened for missing and erroneous values.Statistical analyses were performed with the software package SAS (version 9.4; SAS Institute, Inc.).Descriptive statistics were generated with PROC MEANS and PROC FREQ using SAS.Figures for graphical representations were created in GraphPad Prism (ver.10.1.0;GraphPad Software, San Diego, CA).To test the effect of FRP on the outcome variables of interest, we fitted generalized linear mixed models with an identical link for continuous variables and a logit link for binary variables.A detailed description of the model building is provided in the supplemental materials.Differences among groups in duration of overmilking, average cyclical vacuum fluctuations, and mouthpiece chamber vacuum values during the peak milk flow period were assessed with Student's t-tests using PROC TTEST.

TRIAL I: Description of the Study Population
The average (mean ± SD) DIM of the 156 study cows on the day of enrollment was 182 ± 106, ranging from 5 to 625 d.The lactation numbers were distributed as follows: 56 (35.9%) were in their 1st lactation, 40 (25.6%) were in their 2nd lactation, and 60 (38.5%) were in their 3rd or greater lactation.The average (mean ± SD) daily milk yield during the week before enrollment was 42.3 ± 10.0 kg and ranged from 4.5 to 66.7 kg.The median SCC before the study was 59,000 cells/mL, ranging from 8,000 to 7,372,000.Table 2 displays the baseline characteristics of the study cows.

Teat Tissue Condition
We evaluated the cows for STC and obtained data from 1,185 milking observations, that were used in the final analyses.The presence of STC was detected in 441/786 (56%) and 152/399 (38%) cow observations in the CON and FRP groups, respectively.The final model for the presence of STC is shown in Supplemental Table S1, and the adjusted probabilities are illustrated in Figure 2. The odds of STC were lower for cows in the FRP group than for cows in the CON group [odds ratio (95% CI) = 0.41 (0.31-0.55)].

Milking Characteristics
A total of 11,331 milking observations were obtained.Screening of the data resulted in the removal of 1,554 (13.7%) observations.Thus, a total of 9,777 (CON, 5,952; FRP, 3,825) milking observations were included in the final analyses.Manual milking was documented in 29 (0.3%) observations.
Bimodal milk flow was documented in 2,794/9,777 (28.6%) milking observations.The final model included treatment, period nested with treatment, lactation number, DIM, session, and log-transformed SCC (log-SCC; P < 0.0001; Supplemental Table S7).The FRP group had lower odds of bimodality than did the CON group.Figure 4 depicts the adjusted probabilities of bimodality in response to treatment.

Milking Irregularities
The frequency distributions of milking irregularities were 772/9,777 (7.9%) for liner slip, 207/9,777 (2.1%) for milking unit kick-off, and 510/9,777 (5.2%) for milking unit reattachment.Table 3 shows the frequency distributions for milking irregularities stratified by period.The results of the final models for milking liner slip, milking unit kick-off, and milking unit reattachment are shown in Supplemental Tables S8-S10.The adjusted probabilities of milking irregularities in response to treatment are depicted in Figure 4.There was no effect of treatment on milking liner slip (P = 0.40), milking unit kick-off (P = 0.75), or milking unit reattachment (P = 0.84).The odds ratios (95% CI) for cows that received FRP compared with those that did not (CON cows) were as follows: milking liner slip, 1.07 (0.91-1.26); milking unit kick-off, 1.05 (0.76-1.46); and milking unit reattachment, 1.02 (0.85-1.23).

TRIAL II: Description of the Study Population
We enrolled 156 lactating cows with an average (mean ± SD) DIM of 179 ± 113, ranging from 1 to 617 d.The lactation distributions were 61 (39.1%), 36 (23.1%), and 59 (37.8%) cows in the 1st, 2nd, and ≥ 3rd lactation, respectively.The average (mean ± SD; range) daily milk yield during the week before enrollment was 42.0 ± 10.7; 13.6−69.9kg, and the median SCC before the study was 48,000 cells/mL, with a range of 9,000 to 7,140,000.Table 2 displays the baseline characteristics of the study cows.

Teat-end Hyperkeratosis
We evaluated the cows for HK and obtained data from 506 milking observations, that were used in the final analyses.Teat-end hyperkeratosis was detected in 93/263 (35.4%) and 87/243 (35.8%) cow observations in the CON and FRP groups, respectively.The model results for the presence or absence of HK are shown in Supplemental Table S11 and illustrated in Figure 2.There was no effect of treatment on the odds of HK (P = 0.87).

Teat Tissue Conditions
In this study, we investigated the effect of FRP on teat tissue conditions in high-producing Holstein dairy cows milked 3 times per day.Our data showed that cows that received supplemental stimulation through FRP had lower odds of STC than cows that were milked using conventional mode of milking with a single pulsation (at a rate of 60 cycles/min and a pulsation ratio of 70:30) during the entire process of milking.Previous researchers reported that STC can increase the risk of new intramammary infections (Zecconi et al., 1996) and impact the well-being of dairy cows (Hillerton et al., 2002).Our results are thus of particular importance, as they suggest an efficient method to improve the udder health and well-being of dairy cows.
We speculate that the decreased odds of STC in cows receiving FRP could be attributed to the following factors.First, FRP cows had lower odds of bimodal milk flow than CON cows.This decreased the duration that the teats were subjected to vacuum-induced strain, which can lead to STC in the teat tissue (Williams and Mein, 1982).Second, cows in the FRP group had higher peak milk flow rates.Due to the inverse relationship between the milk flow rate and vacuum-induced strain on the teat tissue (Bade et al., 2009;Ambord and Bruckmaier, 2010), FRP cows likely experienced a lower magnitude of force that creates STC.Third, cows in the FRP group were given additional stimulation to elicit the milk-ejection reflex.This may have resulted in the ejection of milk into the gland cisterns and teat cisterns, thereby increasing the intracisternal pressure by the time of the start of actual milking.Increased cisternal pressure could also increase teat diameter and result in a better seal between the teat and the milking liner barrel.This can result in a lower mouthpiece chamber vacuum, decreasing the risk of STC, as suggested by Penry et al. (2017).Fourth, the extended d-phase during the stimulation phase may have counteracted the mechanical forces exerted by the vacuum at the Our results showed no effect of FRP on HK.Teatend hyperkeratosis affects teat canal closure, enhances lodging of pathogenic bacteria, and consequently is associated with an increased risk of new IMI (Neijenhuis et al., 2001).A previous study by our group (Wieland et al., 2023) reported a greater risk of HK in cows receiving automated stimulation, with a pulsation rate of 300 cycles/min and a ratio of 25:75 for a maximum duration of 80 and 99 s.Based on these results, we would have expected higher odds of HK in the FRP group.Our results do not support this theory.We suggest that the combination of FRP with a premilking stimulation regimen, as implied herein, has no negative effect on the teat-end condition.

Milking Characteristics
Our second objective was to study the effect of FRP on milking characteristics.We found that cows subjected to FRP had lower 2MIN, longer LMF, slightly longer MUOT, higher PMF, and lower odds of bimodality.In contrast, no meaningful differences in the MY were detected between the 2 groups.The observed differences are most likely due to the supplemental stimulation provided during the early stage of milking through the pulsatory action of the milking liner during milk flow below the threshold in FRP cows.This additional stimulation likely provided additional time for oxytocin to be released, thus also allowing more time for transport and binding to the receptors of the myoepithelial cells and for the ejection of milk from the alveoli to the duct system and teat cisterns.This resulted in better milk ejection before the actual milking started.The stimulatory phase in the FRP treatment (milk flow rate <0.5 kg/min at the beginning of milking) restricted the harvest of milk at the usual pace, leading to a lower 2MIN and longer duration of LMF.This can also be established from the data of milk flow rates in Table 3, which showed lower milk flow rates of 15S and 30S during the FRP periods (periods 2 and 4) than during the CON periods (periods 1, 3, and 5).However, this approach could have helped to meet the cows' physiological need for the establishment of milk ejection.As a result, a slightly higher PMF was observed in the FRP cows.Interestingly, the difference in MUOT between the FRP and CON groups was subtle (only 2 s) in Trial I and absent in Trial II.The cows were able to catch up in the later phase of miking, compensating for the lower milk harvest during the stimulation phase.We believe that adequate milk ejection increased the amount of cisternal milk available for harvesting and thus accelerated the milk harvesting process during the milking phase in FRP cows.Our results also support the current recommendations for adequate milk ejection that teats should be stimulated for at least 15 s (Kaskous and Bruckmaier, 2011;Vetter et al., 2014;Wieland et al., 2020) combined with a preparation lag time of 60-120 s (NMC, 2013).
The absence of an effect of different supplemental stimulation regimens on the MY has been previously reported in several studies in which mechanical stimulation was applied (Neuheuser et al., 2018;Tuor et al., 2022;Wieland et al., 2023).For example, Tuor et al. (2022) conducted a study to evaluate the effect of FRP and FRP coupled with reduced vacuum on the milking performance of Holstein cows.The cows were milked at a lower pulsation rate and ratio (50 cycles/min, 30:70) or a combination of both lower pulsation and lower vacuum (34 kPa) during the low milk flow rate (<0.4 kg/min) than at single pulsation and vacuum throughout milking (60 cycles/min, 65:35; 44 kPa).They also reported no differences in MY or MUOT.In contrast to our study, Tuor et al. (2022) reported no differences in PMF in cows milked using single pulsation and vacuum throughout milking, from the cows milked either using FRP or combined flow-responsive pulsation and vacuum.Differences in the derivation of PMF among studies could have led to this discrepancy.For example, Tuor et al. (2022) defined PMF as the maximum milk flow maintained for at least 22.4 s, and they used a portable milk flow meter system (i.e., LactoCorder).However, in this study, we considered the average of the maximum 60 s milk flow rate as the PMF.Additionally, the short preparation lag time (5 s) and lower threshold value for switching to the milking mode might have rendered supplemental stimulation through pulsation very short in the study of Tuor et al. (2022), which could have failed to elicit milk ejection to a level that produced a difference from the single pulsation mode.No differences in MY or PMF were reported by Neuheuser et al. (2018) when reduced pulsation and vacuum settings (50 cycles/min, 30:70; 33 kPa) were applied during a milk flow rate of <0.3 kg/min during milking.Ipema et al. (2007) also reported no differences in the MY when cows were milked at a lower b-phase (486 ms) compared with 536 and 623 ms.Ambord and Bruckmaier (2009) studied 29 Holstein cows and applied different pulsation rates and ratios based on instantaneous milk flow rates for a 1-wk period.The authors reported no differences in the MY or main milking time (duration of milking at a milk flow rate >0.5 kg/min in the beginning until <0.22 kg/min at the end) among groups with different peak flow rates and different treatments.They reported higher AMF and PMF with the application of FRP in cows with a PMF >3.2 kg/min.
The absence of a meaningful difference in the MY with different premilking stimulation methods in recent studies could be due to the continuous evolution of the udder and duct systems in cows.That is, we speculate that breeding cows for higher production has modified the udder anatomy of the milk duct system over time.This altered udder and teat anatomy may facilitate complete drainage of milk irrespective of the premilking stimulation regimen.This speculation is also supported by recent studies in which no differences in strip yield (residual milk) were reported at the end of milking when different premilking stimulations were applied (Merrill et al., 1987;Rasmussen et al., 1992;Edwards et al., 2013).
The slightly higher MUOT and PMF in the FRP may not have a profound impact on the parlor throughput and efficiency.However, decreased bimodality and STC in FRP are desirable from the perspectives of udder health and animal well-being.The variation among lactation, DIM, milking session and logSCC for the MUOT followed the corresponding MY of the respective categories in both trials.The beneficial effects of FRP on MUOT, 2MIN, and LMF could not be observed in the FRP group, as these parameters were measured from the time of milking unit attachment, which also included the duration of supplemental stimulation through the pulsation system, leading to deflation in the values of these parameters.
Differences in the odds of bimodality were observed between groups, as expected.The lower odds of bimodality in the FRP were inherent to the pulsation system, which kept the milk flow lower during the stimulation phase (which was also evident from the LMF and 2MIN data).This maintained a greater milk reserve in the teat and gland cistern than in the CON group.This larger repository of milk took longer to drain during the actual milking phase, bridging the gap between exhaustion of the cisternal reserve and availability of alveolar milk, which reduced the incidence of bimodality.Ambord and Bruckmaier (2009) also reported a lower percentage of bimodality when FRP was applied to cows with lower PMF (<3.2 kg/min).It is important to indicate that during the measurement of the incremental milk flow rates used for the detection of bimodality we were unable to decipher between the stimulation phase and the milking phase.Due to the reduced liner-open phase, the milk flow was lower during the stimulation phase.Thus, it is likely that the observed reduction in bimodality was due to the inherent system, as well as the detection method used in this study rather than to a stimulatory effect only.
However, Tuor et al. (2022) and Neuheuser et al. (2018) did not find a similar effect of FRP on bimodality.The differences in the methodology to assess the bimodality and the preparation lag times among their study and the one described herein could also explain this discrepancy.

Milking Irregularities
We also studied the effect of treatment on milking irregularities, in accordance with our previous study (Singh et al., 2024).Our results did not reveal an effect of FRP on milking liner slip, milking unit kick-off, or milking unit reattachment.

Hindleg Activity
We extended our objectives to analyze cow discomfort through the evaluation of hindleg activity during milking, as described by previous research groups (Kutzer et al., 2015;Cerqueira et al., 2017).The percentages of cows that showed at least one step or kick event were 90.6% and 53%, respectively, which were greater than the reported proportions in previous studies (Cerqueira et al., 2017;Meyer et al., 2021).Meyer et al. (2021) reported one or more steps in 65% to 82% of cows and one or more kicks in 10% to 65% of cows in 10 dairy herds in Northern Germany during the months from November to February.In the study reported by Cerqueira et al. (2017), the percentage of cows with at least one step or kick ranged from 10% to 91% for step and from 0% to 38% for kick.The observed discrepancy in the proportion of cows that showed steps or kicks in this study could be due to the individual variation in their expression of behavior during milking (Van Reenen et al., 2002), herd size (Rousing et al., 2004) or prevailing environmental conditions of the milking parlor (Cerqueira et al., 2017).We found a greater number of steps and kicks in the FRP group than in the CON group during milking.The observed differences between the groups are less likely to be due to the milking machine settings, as we also observed differences among the 4 periods.The number of steps and kicks increased during the study period.We also noticed a greater number of flies and insects in the parlor as the summer advanced during the study period.This could have caused disturbance of the animals, in response to which the cows would have also expressed steps and kicks.This may have added to the milking machine-induced discomfort, leading to increased steps and kicks from periods 1 to 4. Since the FRP period always succeeded the CON period, the overall effect in the FRP was more pronounced.Vitela-Mendoza et al. ( 2016) also reported an increase in tail movements, leg stamps, and kicks in beef cattle with an increased population of stable flies.Cerqueira et al. (2017) also reported an increased frequency of kicks in dairy cows with increased milking room temperature above 27°C.

Study Limitations and Future Directions
Although the study was designed to simulate the current milking practices in modern dairy operations, some limitations must be considered.First, the study used a switch-back design for the sake of practicality and feasibility.A contemporary controlled study could have provided a better perspective of the treatment effect by controlling for the random effects of season and management.Second, although most of the outcome variables were measured using mechanical methods, the observer was not blinded to the treatment during the assessment of STC.This could have led to information bias.Third, our results showed a subtle difference in the parameters related to milking efficiency (i.e., higher PMF in the FRP group by 0.1 kg/min with 2 s longer MUOT).With these differences, the usefulness of the FRP for milking efficiency may not be noticeable in practical dairy operations.Fourth, milking characteristics such as the MUOT, 2MIN, PMF and LMF of the FRP group were assessed in combination with the pulsation and milking phases.A study involving separate analyses of these milking characteristics during the actual milking phase of FRP with CON could provide better insight into the use of the FRP system in milking.Furthermore, a study involving the application of FRP with reduced vacuum during the low-milk-flow phase could also help explore its effect as a supplemental stimulation.Fifth, as this study did not involve any difference in the premilking udder preparation between the treatments, it is apparent that no differences in parlor efficiency could be achieved.Though our results indicated the role of FRP in supplementing the premilking stimulation, the efficacy of FRP as a replacement for manual stimulation needs to be evaluated.Such studies could advance parlor efficiency and lower the dependence on manual work.Finally, we conducted our study on 1 New York dairy farm with Holstein cows that were milked 3 times per day.Therefore, the external validity may be limited to similar operations in this region.

CONCLUSIONS
The use of flow-responsive pulsation in milking cows was associated with lower odds of machine milkinginduced short-term teat tissue changes and lower odds of bimodality, whereas no differences in the odds of milking kick-off or reattachment were observed.Thus, using flow-responsive pulsation coupled with manual premilking udder preparation provided additional stimulation to accommodate the cows' physiological requirements for the elicitation of the milk-ejection reflex.We conclude that FRP may foster adequate teat stimulation in cows before the initiation of milk harvest and has the potential to improve teat tissue conditions.

Figure 1 .
Figure 1.Experimental design, premilking udder preparation and milking regimens (CON, conventional pulsation setting; FRP, flowresponsive pulsation; STC, short-term teat tissue changes; HK, teat-end hyperkeratosis; HLA, hindleg activity) Singh et al.: AUTOMATED SUPPLEMENTAL STIMULATION Table 1.Definitions of milking characteristics recorded with the electronic on-farm milk meter (MM27BC, DeLaval International AB, Tumba, Sweden) Milking characteristic Abbreviation Definition Milk yield (kg) MY Yield of milk recorded from the start of milking 1 to detachment of milking unit First 15 s milk flow rate (kg/min) 15S Average milk flow rate recorded in the first 15 s after the start of milking 1 15 to 30 s milk flow rate (kg/min) 30S Average milk flow rate between 15 to 30 s after the start of milking 1 30 to 60 s milk flow rate (kg/min) 60S Average milk flow rate between 30 to 60 s after the start of milking 1 60 to 120 s milk flow rate (kg/min) 120S Average milk flow rate between 60 to 120 s after the start of milking 1 Average milk flow rate (kg/min) AMF Total milk yield/milking unit-on time Peak milk flow rate (kg/min) PMF Maximum 60-s average milk flow rate 2-min milk yield (kg) 2MIN Milk yield harvested within the first 2 min after the start of milking 1 Duration of low milk flow rate (s) LMF Duration (s) with a milk flow rate <1 kg/min Milking unit-on time (s) MUOT Time recorded from the start of milking 1 to unit detachment 1 Start of milking: Start of the milking procedure as recorded by the milking point controller induced by pushing the start button.
Singh et al.: AUTOMATED SUPPLEMENTAL STIMULATION

Figure 2 .
Figure 2. Adjusted probabilities from generalized linear mixed models showing the effect of flow-responsive pulsation (FRP) and conventional (CON) milking on (A) short-term teat tissue changes (STC) and (B) teat-end hyperkeratosis (HK).Error bars represent 95% CIs.For other main effects, see Supplemental TableS1 and S11.

Figure 3 .
Figure 3. Least squares means from general linear mixed models showing the effect of 2 different milking pulsation settings, flow-responsive pulsation (FRP) and conventional (CON), on the milk yield (A), milking unit-on time (B), 2-min milk yield (C), peak milk flow rate (D), and duration of low milk flow rate (E).E: P-values for the effect of treatment between cows with different lactation numbers are derived from Tukey-Kramer post hoc tests.Error bars represent 95% CIs.For other main effects, see Supplemental Tables S2-S6.
Singh et al.: AUTOMATED SUPPLEMENTAL STIMULATIONbeginning of milking during the period of milk flow below the threshold in FRP cows.This likely reduced the risk of STC in FRP cows.
Tuor et al. (2022) used a transient decrease of milk flow rate at least by 100 g/min for a minimum of 2.8 s during the initial phase of milking to identify bimodality.Neuheuser et al. (2018) considered a transient decrease of milk flow rate by > 200 g/min occurred in 95 s after the milk flow increased above 500 g/min.
Singh et al.: AUTOMATED SUPPLEMENTAL STIMULATION

Table 2 .
Baseline characteristics of Holstein cows subjected to 2 different milking machine pulsation mode in two different trials.Cows in Trial I alternated between 1-week periods in the conventional (CON) and flowresponsive pulsation (FRP) modes for a total of 5 weeks.Cows in Trial II alternated between CON and FRP modes of milking for 12 weeks with each period of 3 weeks duration.Days 1 and 2 of each treatment period were classified as an adjustment period and excluded from the analyses 1 logSCC: Somatic cell count before the study, log 10 transformed.2Dailyaverage milk yield for the last 7 d before enrollment.3Definition of a bimodal milk flow curve: If any of the incremental milk flow rates (30S, 60S, or 120S) were lower than any of the previous ones (15S, 30S, or 60S).
for abbreviations and definitions