Using infrared thermography to determine changes in teat skin surface temperature after machine milking in dairy cows

Mastitis is among the costliest diseases affecting dairy cows, partly due to the resulting permanent reduction in the quantity and quality of milk produced. Most mastitis cases involve pathogenic organisms entering the cow's mammary gland through the teat canal. The teat has natural defenses against these pathogens that can be disrupted during milk harvesting. These disruptions of the teat's circulatory system and tissue integrity can predispose them to mastitis. Traditionally, machine-milking induced changes in teat blood circulation and tissue integrity have been assessed by means of manual evaluation and ultrasonography. Infrared thermography (IRT) has previously been shown to produce precise and consistent measurements of skin surface temperatures (SSTs) on cows' hind teats. Our objective was to describe the variability in the teat SST following machine milking. Describing the variability in teat SST before and after milking could be useful to guide further studies to elucidate the physiology of the effects of milking on teat defense mechanisms. In this observational study, thermographic images of both hind teats from 140 cows immediately pre- and post-machine milking were analyzed. The average SSTs were subsequently determined at the proximal, middle, and distal aspects of each hind teat using image analysis software. The least squares means (95% CI) from general linear mixed models of the pre- and post-milking SSTs, respectively, were 33.6 (33.5–33.8) °C and 35.4 (35.3–35.5) °C at the proximal aspect, 33.2 (33.1–33.4) °C and 35.2 (35.1–35.3) °C at the middle aspect, and 32.3 (32.1–32.5) °C and 34.0 (33.9–34.1) °C at the distal aspect. The observed increase in SST from pre- to post-milking SSTs at all 3 aspects of the teat suggest that some of the variability in the SSTs can be attributed to the milking event. Future research is warranted to investigate the biological relevance of SST changes during machine milking and any potential change in teat defense mechanisms, risk of mastitis, or other pathologies. Mastitis is among the costliest diseases affecting dairy cows, partly due to the resulting permanent reduction in the quantity and quality of milk produced. Most mastitis cases involve pathogenic organisms entering the cow's mammary gland through the teat canal. The teat has natural defenses against these pathogens that can be disrupted during milk harvesting. These disruptions of the teat's circulatory system and tissue integrity can predispose them to mastitis. Traditionally, machine-milking induced changes in teat blood circulation and tissue integrity have been assessed by means of manual evaluation and ultrasonography. Infrared thermography (IRT) has previously been shown to produce precise and consistent measurements of skin surface temperatures (SSTs) on cows' hind teats. Our objective was to describe the variability in the teat SST following machine milking. Describing the variability in teat SST before and after milking could be useful to guide further studies to elucidate the physiology of the effects of milking on teat defense mechanisms. In this observational study, thermographic images of both hind teats from 140 cows immediately pre- and post-machine milking were analyzed. The average SSTs were subsequently determined at the proximal, middle, and distal aspects of each hind teat using image analysis software. The least squares means (95% CI) from general linear mixed models of the pre- and post-milking SSTs, respectively, were 33.6 (33.5–33.8) °C and 35.4 (35.3–35.5) °C at the proximal aspect, 33.2 (33.1–33.4) °C and 35.2 (35.1–35.3) °C at the middle aspect, and 32.3 (32.1–32.5) °C and 34.0 (33.9–34.1) °C at the distal aspect. The observed increase in SST from pre- to post-milking SSTs at all 3 aspects of the teat suggest that some of the variability in the SSTs can be attributed to the milking event. Future research is warranted to investigate the biological relevance of SST changes during machine milking and any potential change in teat defense mechanisms, risk of mastitis, or other pathologies. Mastitis, or the inflammation of the mammary gland, can result in a permanent decrease in milk quantity and quality, an increase in culling rates (animal loss costs), and an increase in treatment-related expenses. This, in addition to its prevalence, is why mastitis is among the most expensive diseases affecting the dairy industry. Most cases of mastitis are initiated by a pathogenic organism migrating into the mammary gland through the associated teat canal (Erskine, 2020Erskine, R. J. 2020. Mastitis in Cattle. Accessed 6/3/2023.Google Scholar). The cow's teat has natural defenses against pathogen migration, such as hydrophobic lipids, a continuously shedding, keratinized epithelial lining of the teat canal, healthy microbiota, and epithelial cell phagocytosis of inflammatory mediators (Katsafadou et al., 2019Katsafadou A.I. Politis A.P. Mavrogianni V.S. Barbagianni M.S. Vasileiou N.G.C. Fthenakis G.C. Fragkou I.A. Mammary Defences and Immunity against Mastitis in Sheep.Animals (Basel). 2019; 9 (31561433): 726https://doi.org/10.3390/ani9100726Crossref PubMed Scopus (22) Google Scholar). However, an excessive vacuum, particularly during low milk flow, can result in excessive sloughing off of these natural defenses. Therefore, the milking procedure plays an active role in the prevention of, or predisposition of, cows to mastitis, and an effort should be made to protect the teat canal and associated tissues during all milking events. However, methods must exist on farm that distinguish healthy from compromised teat tissue associated with milking to ensure protection and proper milking parameters. One method that can be implemented is monitoring teats for short-term changes (STCs), which are teat tissue changes associated with the milking procedure, such as, change in morphology and color (Mir et al., 2015Mir A.Q. Bansal B.K. Gupta D.K. Short Term Changes in Teats Following Machine Milking with Respect to Quarter Health Status in Cows.Journal of Animal Research. 2015; 5: 467-471https://doi.org/10.5958/2277-940X.2015.00080.7Crossref Google Scholar). Changes in skin surface color could suggest poor perfusion of the tissue, which can result in hypoxia at the teat end (Mein, 2012Mein G.A. The role of the milking machine in mastitis control.Vet. Clin. North Am. Food Anim. Pract. 2012; 28 (22664210): 307-320https://doi.org/10.1016/j.cvfa.2012.03.004Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar). Hypoxia can lead to necrosis of the teat epithelial tissue, which weakens the first-line defenses of the mammary gland against mastitis. Other STCs include firmness, thickening or a ring at the teat base, and the degree of opening at the teat orifice, which are associated with harmful practices such as over milking, poor pulsation, or an excessive vacuum (Hillerton et al., 2000Hillerton J.E. Ohnstad I. Baines J.R. Leach K.A. Changes in cow teat tissue created by two types of milking cluster.J. Dairy Res. 2000; 67 (11037228): 309-317https://doi.org/10.1017/S0022029900004283Crossref PubMed Scopus (31) Google Scholar). In addition to its detrimental effect on udder health, these STCs have been reported to diminish animal well-being (Hillerton et al., 2002Hillerton J.E. Pankey J.W. Pankey P. Effect of over-milking on teat condition.J. Dairy Res. 2002; 69 (12047113): 81-84https://doi.org/10.1017/S0022029901005386Crossref PubMed Scopus (53) Google Scholar). Overmilking has been shown to lead to increased mastitis infections in unaffected quarters of infected cows during low milk flow (Natzke et al., 1982Natzke R.P. Everett R.W. Bray D.R. Effect of overmilking on udder health.J. Dairy Sci. 1982; 65 (7042783): 117-125https://doi.org/10.3168/jds.S0022-0302(82)82160-7Abstract Full Text PDF PubMed Google Scholar). Petersen, 1944Petersen W.E. The Action of the Mechanical Milker in Relation to Completeness of Milking and Udder Injury1.J. Dairy Sci. 1944; 27: 433-440https://doi.org/10.3168/jds.S0022-0302(44)92619-6https://doi.org/10.3168/jds.S0022-0302(44)92619-6Abstract Full Text PDF Google Scholar also showed that when the teat cup creeps upward on the teat, it can cut off milk flow before the teat sinus, which can damage the membranes of the teat, cause ischemic injury, and predispose the cow to an increased risk of mastitis. Monitoring for these STCs could allow for the detection of inappropriate machine milking protocols, provide an indirect measure of the teats' defense mechanisms, and offer opportunities for correction to decrease the risks associated with mastitis development. Ideally, every teat of every milked cow would receive individualized care and machine milking protocols to minimize the risk of mastitis. Each teat should be monitored at every milking for changes that could predispose the associated mammary gland to mastitis development. However, there are currently over 9.4 million dairy cows in the US and only approximately 160,000 individuals employed throughout all of dairy manufacturing (Pigott, 2023Pigott M. Dariy Farm in the US. IBISWorld.https://my.ibisworld.com/us/en/industry/11212/industry-at-a-glanceDate: 2023Date accessed: March 6, 2023Google Scholar). Shifting staff to monitor for STCs would be costly and time consuming and likely outweigh the benefits of individualized monitoring. However, an automated monitoring procedure could reap the benefits of individualized teat care without incurring the costs associated with increased technician time and salary. Digital imaging is capable of detecting STCs such as color changes, the presence of a ring, and teat canal openness but is limited in its ability to determine teat end firmness due to congestion, which historically has been detected through manual palpation. Infrared thermography (IRT) could potentially provide more information about the teat by also measuring teat skin surface temperature (SST). Previous IRT studies have been able to differentiate between the presence and absence of mastitis when comparing the SST of an infected udder with that of other quarters or of the body SST (Sathiyabarathi et al., 2016Sathiyabarathi M. Jeyakumar S. Manimaran A. Jayaprakash G. Pushpadass H.A. Sivaram M. Ramesha K.P. Das D.N. Kataktalware M.A. Prakash M.A. Kumar R.D. Infrared thermography: A potential noninvasive tool to monitor udder health status in dairy cows.Vet. World. 2016; 9 (27847416): 1075-1081https://doi.org/10.14202/vetworld.2016.1075-1081Crossref PubMed Scopus (54) Google Scholar). IRT has also previously been shown to be useful in reflecting changes in the SST associated with different California Mastitis Test (CMT) scores (Colak et al., 2008Colak A. Polat B. Okumus Z. Kaya M. Yanmaz L.E. Hayirli A. Short communication: early detection of mastitis using infrared thermography in dairy cows.J. Dairy Sci. 2008; 91 (18946129): 4244-4248https://doi.org/10.3168/jds.2008-1258Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar). The goal of this project was to describe changes in teat SSTs that occur during machine milking to eventually evaluate its use as a monitoring tool for STCs and proper milking protocols. This observational study was conducted at the Teaching Dairy Barn at Cornell University, College of Veterinary Medicine (Ithaca, NY) in July 2021. The study protocol was reviewed and approved by the Cornell University Institutional Animal Care and Use Committee (protocol no. 2021–0005). During the study period, approximately 160 lactating Holstein cows were housed in 2 free-stall pens that were bedded with recycled sand. They were fed a total mixed ration formulated to meet or exceed the requirements outlined by the National Research Council (2001). Herd data were maintained in a dairy management software program (Dairy Comp 305, Valley Agricultural Software, Tulare, CA). The farm used National Dairy Herd Information Association services, including the individual-cow somatic cell count (SCC) option. The rolling herd key performance indicators were average milk production, 12,512 kg; bulk tank SCC, 225,000 cells/mL; monthly clinical mastitis incidence, 2.3%; 21-d pregnancy rate, 26%; and culling rate, 35.0%. 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 set to supply a receiver operator vacuum of 45 kPa regulated by a variable frequency drive. The milking unit was composed of a cluster MC70 (DeLaval International AB) and a milking liner with a square barrel shape (ProSquare DPX2, IBA, Millbury, MA). The pulsators (EP100, DeLaval International AB) were set to a pulsation rate of 60 cycles/min, a ratio of 70:30, and a side-to-side alternating pulsation. These settings resulted in an average claw vacuum during the peak milk flow period of 37 kPa. The automatic cluster removers were set to a cluster remover milk flow threshold of 1.4 kg/min, 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. The milking parlor was equipped with electronic on-farm flow-through milk flow meters using near-infrared technology (MM27BC, DeLaval International AB) for the assessment of milking characteristics. The milking system settings and milking characteristics were monitored with a dairy farm management software program (DelPro, DeLaval International AB). Before the start of the study, all system settings were verified and assessed by the investigators according to the guidelines outlined by the National Mastitis Council (2012). Lactating cows were eligible for enrollment if they were free of clinical mastitis for at least 2 wk, had no udder abnormalities such as nonlactating quarters or teat injuries, and had a record of normal ease of handling. We obtained cow characteristics such as lactation number, stage of lactation, and SCC on the last test day from the dairy management software program (DairyComp 305, Valley Agricultural Software, Tulare, CA). The milk flow characteristics were obtained with the electronic on-farm flow-through milk flow meters and recorded with the dairy farm management software (DelPro, DeLaval International AB): milk yield (i.e., yield of milk harvested from start of milking to detachment of the milking unit, kg), milking unit-on time (i.e., duration from start of milking to detachment of the milking unit, s), average milk flow rate (calculated as total/milking unit-on time, kg/min), and time spent in low milk flow rate (i.e., time spent below 1 kg/min milk flow rate between the start of milking and detachment of the milking unit, s). To minimize interference with the dairy's milking routine, we collected the data during 3 routine milking sessions (2 milking sessions at 1100 h and 1 milking session at 1900 h). To facilitate data collection, one trained investigator (MW) performed the routine milking procedures, including pre-milking udder preparation, milking unit attachment, and application of a post-milking teat disinfectant. The pre-milking teat sanitization and stimulation consisted of 3 steps. Step 1 was to wipe all 4 teats with a clean cloth towel and dip the teats with an iodine-based teat dip (Multi Dose MD; DeLaval International AB); step 2 was to manually fore-strip and dry all 4 teats with a clean cloth towel; and step 3 was to attach and align the milking unit. Thermographic images of both hind teats were obtained with a portable thermography camera (FLIR T530, Teledyne FLIR LLC, Wilsonville, OR) by 1 trained investigator (CD). Before the study, the camera was calibrated, and the imaging modes were identified and kept consistent throughout the trial. These included the autofocus function and the 'laser' option, where the focus is based on a laser distance. The laser distance meter was enabled to automatically determine the object distance. The reflective temperature was kept at 20°C and the emissivity at 0.95. The atmospheric temperature and relative humidity were retrieved from the local weather station: d 1, 23°C, 81%; d 2, 24°C, 79%; and d 3, 23°C, 79%. Images were taken from the caudal aspect of the udder in a caudo-to-cranial direction from an approximately 0.5 m distance. Teat scans were taken after completion of pre-milking udder preparation before milking unit attachment (T1) and directly after unit detachment (T2). All scans were labeled with the cow identification number and the sequence using the note function and subsequently stored on the integrated flash drive. The adjunct software program (FLIR Tools, Teledyne FLIR LLC) was used to conduct measurements of the teat SST of the left and right hind teats. The software program provides different measurement tools to calculate the temperature at a single location (i.e., spotmeter) or the average, minimum, and maximum temperatures of regions of interest outlined by different geometric structures (i.e., line, rectangle, ellipse). Using the rectangle tool, we determined the average teat SSTs in 3 regions of interest. Our reasoning for determining the average temperatures rather than a single temperature measurement was based on an attempt to decrease measurement error. Because we aimed to segment the teat into 3 different anatomical regions of interest as described previously (Paulrud et al., 2005Paulrud C.O. Clausen S. Andersen P.E. Rasmussen M.D. Infrared thermography and ultrasonography to indirectly monitor the influence of liner type and overmilking on teat tissue recovery.Acta Vet. Scand. 2005; 46 (16261926): 137-147https://doi.org/10.1186/1751-0147-46-137Crossref PubMed Google Scholar; Barkova et al., 2019Barkova A. Elesin A. Milshtein I. Barashkin M. Thermovision diagnostics of the milking equipment impact on the state of mammary glands of cattle.in: Proc. Advances in Intelligent Systems Research. 2019: 519-521Crossref Google Scholar; Tangorra et al., 2019Tangorra F.M. Redaelli V. Luzi F. Zaninelli M. The Use of Infrared Thermography for the Monitoring of Udder Teat Stress Caused by Milking Machines.Animals (Basel). 2019; 9 (31234510): 384https://doi.org/10.3390/ani9060384Crossref Scopus (10) Google Scholar), we obtained measurements from the proximal, middle, and distal aspects of the teat. One trained investigator (LH) consistently performed the following steps according to a standard operating procedure as previously described (DiLeo et al., 2022DiLeo C. Basran P.S. Porter I.R. Wieland M. Development and evaluation of a standardized technique to assess teat skin temperature of dairy cows using infrared thermography. JDS Communications.https://doi.org/10.3168/jdsc.2021-0181Date: 2022Google Scholar). First, the proximal and distal boundaries of the teat were defined. For this purpose, a rectangle was drawn from the teat base to the teat end. Subsequently, 3 separate rectangles of equal height were drawn at the proximal, middle, and distal thirds of the teat. The width of the teat at each aspect determined the width of the rectangle such that the corners did not exceed the boundary of the teat demarcation. Figure 1 depicts the regions of interest at the proximal, middle, and distal aspects of the teat used for the measurements of the average teat SST. Finally, the average temperatures of each region of interest outlined with the rectangle were obtained. We also assessed teat-end shape from the digital images, that were obtained automatically with the thermography camera and categorized them into pointed, round, and flat as previously described (Wieland et al., 2017Wieland M. Nydam D.V. Virkler P.D. A longitudinal field study investigating the association between teat-end shape and two minute milk yield, milking unit-on time, and time in low flow rate.Livest. Sci. 2017; 205: 88-97https://doi.org/10.1016/j.livsci.2017.09.011https://doi.org/10.1016/j.livsci.2017.09.011Crossref Scopus (20) Google Scholar). During the measurements, any notes, including information on cow identification number and sequence remained obscured. Data were compiled in Microsoft Excel (2019, Microsoft Corporation, Redmond, WA). Before the statistical analyses, we screened the data for missing values and removed observations with a missing teat scan or missing milk flow data. To describe the SST at the proximal, middle, and distal aspects of the teat relative to the machine milking event (i.e., T1 and T2), we fitted 3 general linear mixed models with PROC MIXED in SAS (version 9.4, SAS Institute Inc. Cary, NC). To account for the dependence of repeated measurements over time, a REPEATED statement for quarter position nested within cow was included. Four covariance structures were tested (compound symmetry, variance components, autoregressive order 1, and unstructured) to model the covariance of repeated measurements, and the one with the smallest Akaike's information criterion was selected. Time was forced into the models as a fixed effect. We considered lactation number (1st, 2nd, or ≥ 3rd lactation), stage of lactation (≤100, 101–200, or > 200 d in milk (DIM), SCC (log10-transformed), milk yield (kg/milking session), milking unit-on time (s), and teat-end shape as independent variables and initially screened them for inclusion with univariable analysis. All variables with a P value |0.60| was considered indicative of collinearity. Manual backward elimination was used until each variable had a P value 0.5 was considered to indicate an influential value. For the final model, the assumptions of homoscedasticity and normality of residuals were assessed by inspecting residual plots versus corresponding predicted values and examining quantile-quantile residual plots. A total of 147 cows were enrolled in the study, but data from 7 cows were excluded due to missing data (missing teat scan, n = 2; missing milk flow data, n = 5). The average (mean ± SD) DIM of the remaining 140 cows was 166 ± 117 d, ranging from 2 to 494 d. The lactation number was distributed as follows: 53 (37.9%) animals were in the first lactation, 35 (25.0%) animals were in the second lactation, and 52 (37.1%) animals were in their third or greater lactation (24 in the third; 12 in the fourth; 12 in the fifth; 3 in the sixth; and 1 in the seventh lactation). The median SCC was 57,000 cells/mL and ranged from 8,000 to 5,021,000. The average mean ± SD (median; range) values of the various milking characteristics from the studied milking observations were as follows: milk yield, 12.7 ± 4.0 (12.3; 1.3–31.8) kg; average milk flow rate, 3.1 ± 0.8 (3.0; 0.4–5.1) kg/min; milking unit-on time, 247 ± 69 (240; 119–495) s; and time spent in low milk flow rate, 21.4 ± 27.5 (14.9; 0–290) s. Teat-end shape was distributed as follows: pointed, 38 (13.6%); round, 203 (72.5%); and flat, 39 (13.9%). A total of 294 hind teats were studied using thermography at 2 time points (T1 and T2). Fourteen measurements were excluded corresponding to the 7 excluded cows. Thus, 280 teats were analyzed before (T1; 140 images) and after (T2; 140 images) machine milking. The average (mean ± SD) teat SST values at the proximal, middle, and distal aspects of the left and right hind teats are shown in Table 1.Table 1Average (mean ± standard deviation) values of the teat skin surface temperature before (T1) and after (T2) machine milking at the proximal, middle, and distal aspects of the left and right hind teats of 140 Holstein dairy cows assessed with an infrared thermographic cameraLeft hind teatRight hind teatProximalMiddleDistalProximalMiddleDistalTime pointT1 (°C)33.6 ± 1.233.3 ± 1.332.4 ± 1.533.6 ± 1.233.2 ± 1.432.3 ± 1.6T2 (°C)35.4 ± 0.935.1 ± 0.934.0 ± 1.135.4 ± 0.935.2 ± 0.934.0 ± 1.1 Open table in a new tab For the proximal teat aspect, univariable analyses revealed the following results: lactation number P = 0.15; stage of lactation, P = 0.11; logSCC, P = 0.02; milk yield, P = 0.0003; milking unit-on time, P = 0.11, and teat-end shape, P = 0.65. Spearman correlation coefficients indicated no collinearity among the 5 variables (r ≤ |0.53|), and thus, all 5 variables were included in the initial model. The final model included lactation number (P = 0.04), milk yield (P < 0.0001) and time of measurement (P < 0.0001). The least squares means and 95% CIs were 34.6°C (34.5–34.8) for cows in lactation 1, 34.5°C (34.3–34.7) for cows in lactation 2, and 34.3°C (34.2–34.5) for animals in lactation 3 and greater. A 1-unit increase in milk yield was associated with an increase of 0.05°C (95% CI, 0.03–0.08). The least squares means and 95% CIs were 33.6°C (33.5–33.7) at T1 and 35.4°C (35.3–35.5) at T2. For the middle teat aspect univariable analyses revealed the following results: lactation number, P = 0.14; stage of lactation, P = 0.41, logSCC, P = 0.18; milk yield, P = 0.001; milking unit-on time, P = 0.08, and teat-end shape, P = 0.97. The final model included milk yield (P = 0.001) and time (P < 0.0001). A 1-unit increase in milk yield was associated with an increase of 0.04°C (95% CI, 0.02–0.07). The least squares means and 95% CIs were 33.2°C (33.1–33.4) at T1 and 35.2°C (35.1–35.3) at T2. For the distal teat aspect univariable analyses revealed the following results: lactation number, lactation number, P = 0.49; stage of lactation, P = 0.93 logSCC, P = 0.17; milk yield, P = 0.71; milking unit-on time, P = 0.89, and teat-end shape, P = 0.44. The final model included time (P < 0.0001). The least squares means and 95% CIs were 32.3°C (32.1–32.5) at T1 and 34.0°C (33.9–34.1) at T2. The inspection of Cook's distance revealed no influential outliers for any of the 3 final models. The assumptions of homoscedasticity and normality of residuals were met. Figure 2 shows the least squares means and 95% CIs for the SST at the proximal, middle, and distal aspects of the teat before (T1) and after (T2) machine milking. Our objective was to describe changes in the SST using IRT that occur during the machine milking procedure. At both time points, the SST (determined by IRT) was higher proximally and cooler distally. This is contrary to the findings of Tangorra et al., 2019Tangorra F.M. Redaelli V. Luzi F. Zaninelli M. The Use of Infrared Thermography for the Monitoring of Udder Teat Stress Caused by Milking Machines.Animals (Basel). 2019; 9 (31234510): 384https://doi.org/10.3390/ani9060384Crossref Scopus (10) Google Scholar but similar to those reported by Barth, 2000Barth K. Basic investigations to evaluate a highly sensitive infrared-thermograph-technique to detect udder inflammation in cows.Milchwissenschaft. 2000; 55: 607-609Google Scholar. Differences in the study population and the assessment of the SSTs, including image capturing and the subsequent technique for measuring the average SSTs at the different teat aspects, could help explain the discrepancies among the studies. We observed that the SSTs increased at all 3 locations post-milking relative to the pre-milking value. The increase in the SSTs of the teats and quarters post-milking has been described previously (Paulrud et al., 2005Paulrud C.O. Clausen S. Andersen P.E. Rasmussen M.D. Infrared thermography and ultrasonography to indirectly monitor the influence of liner type and overmilking on teat tissue recovery.Acta Vet. Scand. 2005; 46 (16261926): 137-147https://doi.org/10.1186/1751-0147-46-137Crossref PubMed Google Scholar; Vegricht et al., 2007Vegricht J. Machalek A. Ambroz P. Brehme U. Rose S. Milking-related changes of teat temperature caused by various milking machines.Res. Agric. Eng. 2007; 53: 121-125https://doi.org/10.17221/1954-RAECrossref Google Scholar; Yang et al., 2018Yang C. Li G. Zhang X. Gu X. Udder skin surface temperature variation pre- and post- milking in dairy cows as determined by infrared thermography.J. Dairy Res. 2018; 85 (29785909): 201-203https://doi.org/10.1017/S0022029918000213Crossref Scopus (8) Google Scholar; Barkova et al., 2019Barkova A. Elesin A. Milshtein I. Barashkin M. Thermovision diagnostics of the milking equipment impact on the state of mammary glands of cattle.in: Proc. Advances in Intelligent Systems Research. 2019: 519-521Crossref Google Scholar). Isaksson and Lind, 1994Isaksson A. Lind O. Milking-related changes in the surface temperature of the bovine teat skin.Acta Vet. Scand. 1994; 35 (7676928): 435-438https://doi.org/10.1186/BF03548319Crossref PubMed Scopus (6) Google Scholar listed 3 reasons that could lead to increased SST post-milking: (1) warm milk flowing through the lumen of the teat, (2) insulation provided by the teat cup of the milking unit, and (3) physiologic responses in vascular plexuses. To demonstrate conductive heat transfer, they flushed 38–39°C water through freshly slaughtered and excised teat cisterns and streak canals. They measured the temperature of the teat surface before and after flushing for less than 6 min with warm water and demonstrated an increase of 8°C (from 30 to 38°C) in the teat surface temperature (Isaksson and Lind, 1994Isaksson A. Lind O. Milking-related changes in the surface temperature of the bovine teat skin.Acta Vet. Scand. 1994; 35 (7676928): 435-438https://doi.org/10.1186/BF03548319Crossref PubMed Scopus (6) Google Scholar). Of course, this demonstration was on excised tissue and does not account for heat dissipation through the vascular system of the teat tissue back to the body. This dissipation of heat through the vascular system and through conduction to the teat cup, milking unit and environment post-milking unit detachment may account for the more minor rise in SST in our study than that of Isaksson and Lind, 1994Isaksson A. Lind O. Milking-related changes in the surface temperature of the bovine teat skin.Acta Vet. Scand. 1994; 35 (7676928): 435-438https://doi.org/10.1186/BF03548319Crossref PubMed Scopus (6) Google Scholar. In addition, post-milking, there is often a thin film of milk left on the teat that could dissipate heat through evaporation. Isaksson and Lind, 1994Isaksson A. Lind O. Milking-related changes in the surface temperature of the bovine teat skin.Acta Vet. Scand. 1994; 35 (7676928): 435-438https://doi.org/10.1186/BF03548319Crossref PubMed Scopus (6) Google Scholar proposed that milking induces relaxation, vasodilation, and an enhanced blood flow rate, which could increase the SST during and after milking. We previously demonstrated increased teat tissue perfusion post-milking unit detachment compared with pre-stimulation through power Doppler sonography (Wieland et al., 2020Wieland M. Shirky S. Gioia G. Sipka A. Virkler P.D. Nydam D.V. Älveby N. Porter I.R. Blood perfusion of teat tissue in dairy cows: Changes associated with pre-milking stimulation and machine milking.J. Dairy Sci. 2020; 103 (32389482): 6588-6599https://doi.org/10.3168/jds.2020-18219Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar). We also reported an increase in tissue perfusion post-teat preparation before milking; however, the study involved placing the teat into a warmed lube to scan with ultrasound, which could have relaxed the musculature in the tissue and simulated milking unit attachment, resulting in increased perfusion (Wieland et al., 2020Wieland M. Shirky S. Gioia G. Sipka A. Virkler P.D. Nydam D.V. Älveby N. Porter I.R. Blood perfusion of teat tissue in dairy cows: Changes associated with pre-milking stimulation and machine milking.J. Dairy Sci. 2020; 103 (32389482): 6588-6599https://doi.org/10.3168/jds.2020-18219Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar). Increased teat tissue perfusion may be encouraged by the vacuum and a physiologic response to body-temperature milk flowing through the teat canal to prevent localized hyperthermia. Persson, 1991Persson K. Microcirculation in the bovine teat skin, measured by laser Doppler flowmetry.Acta Vet. Scand. 1991; 32 (1950846): 131-133https://doi.org/10.1186/BF03547005Crossref PubMed Scopus (9) Google Scholar also found an increase in blood flow following machine milking at the teat end compared with 30 min before milking using laser Doppler flowmetry. Kunc et al., 2007Kunc P. Knizkova I. Prikryl M. Maloun J. Infrared thermography as a tool to study the milking process: a review.Agricultura Tropica et Subtropica (Czech Republic). 2007; 40: 29-32Google Scholar demonstrated that the teat SST increases post-calf suckling, as well as after machine milking. Suckling is often considered a normal physiologic process, which may suggest that a rise in teat SST post-milking is not pathological. However, Neijenhuis et al., 2001Neijenhuis F. Klungel G.H. Hogeveen H. Recovery of cow teats after milking as determined by ultrasonographic scanning.J. Dairy Sci. 2001; 84 (11814016): 2599-2606https://doi.org/10.3168/jds.S0022-0302(01)74714-5Abstract Full Text PDF PubMed Google Scholar, NMC, 2012NMC Procedures for Evaluating Vacuum Levels and Air Flow in Milking Systems. National Mastitis Council, Verona, WI2012Google Scholar, NRC, 2001NRC Nutrient Requirements of Dairy Cattle.Seventh Revised Edition, 2001. The National Academies Press, Washington, DC2001: 408Google Scholar demonstrated that post-milking teat recovery could last longer than 8 h. Notably, teat-wall thickness took 6 h post-milking to return to baseline, which could be a proxy for vascular hyperemia to congestion or edema (Neijenhuis et al., 2001Neijenhuis F. Klungel G.H. Hogeveen H. Recovery of cow teats after milking as determined by ultrasonographic scanning.J. Dairy Sci. 2001; 84 (11814016): 2599-2606https://doi.org/10.3168/jds.S0022-0302(01)74714-5Abstract Full Text PDF PubMed Google Scholar). This could suggest that the increased SST we observed post-milking may persist and merits further study in addition to determining the biological significance of prolonged teat tissue hyperemia. Another factor that may have influenced the relative change in teat SST pre- and post-milking seen in our study is a relative decrease in SST following pre-milking teat preparation and stimulation. Isaksson and Lind, 1994Isaksson A. Lind O. Milking-related changes in the surface temperature of the bovine teat skin.Acta Vet. Scand. 1994; 35 (7676928): 435-438https://doi.org/10.1186/BF03548319Crossref PubMed Scopus (6) Google Scholar review of SST of teats associated with milking showed a repeatable decrease in SST following pre-milking teat stimulation. Our T1 followed pre-milking teat stimulation, cleaning with iodine, and fore-stripping. The manipulation of the teat end and dipping the teat in iodine likely resulted in a temporary physiological vasoconstriction that reduced heat dissipation through the vasculature and contraction of the teat, reducing skin surface area and further reducing heat loss. Similarly, because the iodine solution was not warmed, its impact on the SST was likely similar to the isopropyl alcohol challenge performed by Paulrud et al., 2005Paulrud C.O. Clausen S. Andersen P.E. Rasmussen M.D. Infrared thermography and ultrasonography to indirectly monitor the influence of liner type and overmilking on teat tissue recovery.Acta Vet. Scand. 2005; 46 (16261926): 137-147https://doi.org/10.1186/1751-0147-46-137Crossref PubMed Google Scholar, in which teat skin surface temperatures dropped for up to 10 min after the exposure. Paulrud et al., 2005Paulrud C.O. Clausen S. Andersen P.E. Rasmussen M.D. Infrared thermography and ultrasonography to indirectly monitor the influence of liner type and overmilking on teat tissue recovery.Acta Vet. Scand. 2005; 46 (16261926): 137-147https://doi.org/10.1186/1751-0147-46-137Crossref PubMed Google Scholar and Hamann and Dück, 1984Hamann J. Dück M. Preliminary report on measurement of teat skin temperature using infrared thermography.Milchpraxis. 1984; 22: 148-152Google Scholar reported a drop in temperature post-milking and dry or wet cleaning of the teat. This suggests that our T1 measurement may have been reduced following teat preparation, enhancing the change in temperature between T1 and T2. Interestingly, teat-end shape was not associated with teat SST, which contrasts with previous findings from our group using B-mode ultrasonography (Wieland et al., 2019Wieland M. Virkler P.D. Borkowski A.H. Älveby N. Wood P. Nydam D.V. An observational study investigating the association of ultrasonographically assessed machine milking-induced changes in teat condition and teat-end shape in dairy cows.animal. 2019; 13: 341-348https://doi.org/10.1017/S1751731118001246Crossref PubMed Scopus (11) Google Scholar) and manual evaluation (Wieland et al., 2018Wieland M. Nydam D.V. Älveby N. Wood P. Virkler P.D. Short communication: Teat-end shape and udder-level milking characteristics and their associations with machine milking-induced changes in teat tissue condition.J. Dairy Sci. 2018; 101 (30316606): 11447-11454https://doi.org/10.3168/jds.2018-15057Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar) to assess changes in teat traits that occur relative to machine milking. We believe that the observed discrepancies can be attributed to differences in the study population and diagnostic techniques. Our study had some limitations. First, the study was conducted at a single farm with Holstein dairy cows that are milked 3 times per day. Thus, the external validity of the results is limited to similar operations in the region with the same milking schedule and similar milking machine settings (e.g., vacuum, pulsation, and take-off settings). Second, although we used a standard protocol for the assessment of the teat SST, the boundaries of the proximal, middle, and distal aspects may have changed slightly (between the pre- and post-milking images) due to, for example, changes in teat dimensions that occur during milking. Throughout this work, we demonstrated repeatable increases in the teat SST at all teat aspects post-milking relative to pre-milking unit attachment and conclude that IRT can be used to reliably measure changes in teat SST that occur during machine milking. Future work should evaluate the biological significance of this change in temperature, determine how long it persists, and how it relates to machine milking. Such work could include more sophisticated methods such as thermal radiomics (Basran et al., 2022Basran P.S. DiLeo C. Zhang Y. Porter I.R. Wieland M. Delta thermal radiomics: An application in dairy cow teats.JDS Commun. 2022; 3 (36339742): 132-137https://doi.org/10.3168/jdsc.2021-0179https://doi.org/10.3168/jdsc.2021-0179Abstract Full Text Full Text PDF Google Scholar) and facilitate the evaluation of whether increased perfusion and temperature post-milking are pathological or protective against pathogens and intramammary infections. We thank the staff at the Cornell Teaching Dairy Barn (Ithaca, NY). This work benefitted from the support of the Cornell Institute for Digital Agriculture (CIDA; Ithaca, NY). The authors have not stated any conflicts of interest.