Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Heart Rate Variability as an Indicator of Chronic Stress Caused by Lameness in Dairy Cows

  • Levente Kovács ,

    Contributed equally to this work with: Levente Kovács, Fruzsina Luca Kézér, Viktor Jurkovich

    Affiliations MTA–SZIE Large Animal Clinical Research Group, Üllő-Dóra Major, Hungary, Institute of Animal Husbandry, Faculty of Agricultural and Environmental Science, Szent István University, Gödöllő, Hungary

  • Fruzsina Luca Kézér ,

    Contributed equally to this work with: Levente Kovács, Fruzsina Luca Kézér, Viktor Jurkovich

    Affiliations MTA–SZIE Large Animal Clinical Research Group, Üllő-Dóra Major, Hungary, Institute of Animal Husbandry, Faculty of Agricultural and Environmental Science, Szent István University, Gödöllő, Hungary

  • Viktor Jurkovich ,

    Contributed equally to this work with: Levente Kovács, Fruzsina Luca Kézér, Viktor Jurkovich

    Affiliation Department of Animal Hygiene, Herd Health and Veterinary Ethology, Faculty of Veterinary Science, Szent István University, Budapest, Hungary

  • Margit Kulcsár-Huszenicza,

    Affiliation Department and Clinics of Reproduction, Faculty of Veterinary Science, Szent István University, Budapest, Hungary

  • János Tőzsér

    Affiliation MTA–SZIE Large Animal Clinical Research Group, Üllő-Dóra Major, Hungary

Heart Rate Variability as an Indicator of Chronic Stress Caused by Lameness in Dairy Cows

  • Levente Kovács, 
  • Fruzsina Luca Kézér, 
  • Viktor Jurkovich, 
  • Margit Kulcsár-Huszenicza, 
  • János Tőzsér


4 Jan 2016: Kovács L, Kézér FL, Jurkovich V, Kulcsár-Huszenicza M, Tőzsér J (2016) Correction: Heart Rate Variability as an Indicator of Chronic Stress Caused by Lameness in Dairy Cows. PLOS ONE 11(1): e0146625. View correction


Most experimental studies on animal stress physiology have focused on acute stress, while chronic stress, which is also encountered in intensive dairy cattle farming–e.g. in case of lameness–, has received little attention. We investigated heart rate (HR) and heart rate variability (HRV) as indicators of the autonomic nervous system activity and fecal glucocorticoid concentrations as the indicator of the hypothalamic–pituitary–adrenal axis activity in lame (with locomotion scores 4 and 5; n = 51) and non-lame (with locomotion scores 1 and 2; n = 52) Holstein-Friesian cows. Data recorded during the periods of undisturbed lying–representing baseline cardiac activity–were involved in the analysis. Besides linear analysis methods of the cardiac inter-beat interval (time-domain geometric, frequency domain and Poincaré analyses) non-linear HRV parameters were also evaluated. With the exception of standard deviation 1 (SD1), all HRV indices were affected by lameness. Heart rate was lower in lame cows than in non-lame ones. Vagal tone parameters were higher in lame cows than in non-lame animals, while indices of the sympathovagal balance reflected on a decreased sympathetic activity in lame cows. All geometric and non-linear HRV measures were lower in lame cows compared to non-lame ones suggesting that chronic stress influenced linear and non-linear characteristics of cardiac function. Lameness had no effect on fecal glucocorticoid concentrations. Our results demonstrate that HRV analysis is a reliable method in the assessment of chronic stress, however, it requires further studies to fully understand the elevated parasympathetic and decreased sympathetic tone in lame animals.


Lameness is the number one welfare issue in the dairy industry due to pain, suffering and economic impact [1]. Recent studies found that in Hungarian dairies, on average 27–35% [2,3]—with extremes of 65% [4]—of lactating cows are clinically lame. Most of the physiological studies on dairy cattle welfare have focused on acute stress, while chronic stress, which has a more pronounced effect on welfare and production, has received little attention.

According to recent research on humans, geometric [5] and non-linear [6,7] measures of heart rate variability (HRV), i.e. the short-term fluctuations in successive cardiac interbeat intervals (IBI), are suitable for chronic stress assessment. Non-linear techniques are derived from chaos theory and non-linear system theory [8] and are effective to describe processes in biological systems [9]. The non-invasive measurement of heart rate and HRV has increasingly been used for the assessment of stress in dairy cattle [10].

Traditional methods for the evaluation of the cardiac IBI in time- and frequency-domains are extensively studied both in humans and animals. According to several papers on HRV analysis in farm animals certain cardiac parameters provide a powerful means for measuring the sympathetic and the parasympathetic activity of the autonomic nervous system (ANS) separately [1113]. The behavioral and physiological responsiveness of dairy cows and calves exposed to short-term stressors e.g. machine milking, veterinary procedures, disbudding or castration has been extensively studied using ANS-related linear HRV indices (see Ref. [10] for a review). The majority of these studies found traditional HRV measures useful for the detection and evaluation of acute stress. Although non-linear HRV and its relation with chronic effects of different milking systems have been studied in dairy cows [14], to our knowledge, no research has been published on the effects of chronic stress caused by multifactorial disorders such as lameness using geometric or non-linear HRV in adult cattle.

It is widely accepted that measuring cortisol concentrations is useful to detect the effects of stress in animal species [15]. The small number of works on cattle in response to chronic stress has mainly evaluated cortisol in blood. Studies considering serum cortisol concentrations has been published in lame cows, with contradictory findings [16,17]. A direct relationship between fecal glucocorticoid metabolites, blood cortisol, and adrenal activity has been demonstrated in dairy cattle [18]. Due to the intestinal passage time [19] the metabolites of cortisol appear in feces 12–24 hours after cortisol has been excreted into blood and thus has proven to be useful for the evaluation of chronic stress [15,20].

Fecal cortisol metabolites have not been used in studies on stress caused by lameness in cattle so far. The aim of our study was to test whether concentrations of fecal glucocorticoid metabolites and/or HRV parameters are suitable to detect chronic stress conditions caused by lameness in dairy cows. We investigated non-linear characteristics of HRV as well as parameters from the time, frequency and geometric domain.

Materials and Methods

Animals and housing

This study was carried out on a commercial dairy farm in Hungary, with Holstein-Friesian cows. The herd had year-round calving patterns with annual rolling and an average herd size of 1200 cows and a 305-day milk yield of an average 8,840 kg/cow. This study was approved by the Department of Epidemiology and Animal Protection of the Directorate of Food Chain Safety and Animal Health at Central Agricultural Office (Permit Number: 22.1/1266/3/2010). All procedures involving animals were approved by the Ethics Committee of the Faculty of Veterinary Science, Szent István University.

The study was conducted during a 3-week period in November 2012. Hundred-and-forty-three animals were selected from the herd for this study. The farm was visited 3 to 4 times per week, and 8–10 animals were examined on each day. All selected cows were inspected physically before the HR recording. Animals with health problems or with a recent history (less than one month) of clinical mastitis or postpartum uterine diseases (n = 11) were excluded from the experiment to help exclude other possible causes of stress of pathological origin. Three of the cows were also excluded due to technical problems (dried out electrodes, being disturbed by group mates during data recording). Temperamental animals (n = 2) and cows in oestrus (n = 4) were not involved. Four cows were excluded due to equipment failure which prevented further analysis of HRV. Locomotion scorings were carried out twice. The first session was conducted 2 weeks before measurement and the second at the time of actual measurements on a 5-point scale according to Sprecher et al. [21] to ensure the scores had not drifted over time.

Locomotion scorings were carried out as cows exited the milking parlor and returned to the feeding bunk. Cows were observed walking on flat, nonslip concrete in a well-lit location [22]. The third author did the scoring, who was trained to use this locomotion scoring system with on-farm experience. Only those animals were involved in the lame group, which had a locomotion score 4 or 5 two weeks before HRV recordings. Animals with locomotion score 3 (n = 16), were not involved in the experiment. Finally, 51 lame (without any pharmacological treatment, with visible lesions on at least one hoof, with locomotion scores 4 and 5) and 52 non-lame (clinically healthy, without any pharmacological treatment, with locomotion scores 1 and 2) cows were included in the study (Table 1).

Table 1. Production data of experimental animals used for this study (Mean ± SD).

Cows were housed in modern free-stall barns bedded with straw. They were fed ad libitum with TMR consisting of 31.4% corn silage, 21.6% grass silage, 11.5% alfalfa hay, and 35.5% concentrate mash on a dry matter basis. Cows had free access to water. Animals were milked twice daily (between 5:00 am and 8:00 am and between 5:00 pm and 8:00 pm) in a 2 × 24 parallel milking parlor (BouMatic, Madison, WI, USA).

HR and HRV

Heart rate recordings were obtained using a mobile recording system which included a Polar Equine T56H transmitter with two integrated electrodes, a Polar H2 heart rate sensor and a Polar RS800 CX heart rate monitor (Polar Electro Oy, Kempele, Finland). Electrode sites were covered with ample ultrasound transmission gel (Aquaultra Blue, MedGel Medical, Barcelona, Spain). Transmitters and electrodes were positioned as advised by von Borell et al. [13] and fitted to cows with a leather girth after morning milking (between 6:00 am and 8:30 am). Data recording started 2 hours after preparation and lasted until the evening milking has finished (between 6:00 pm and 8:30 pm).

As advised in earlier reports [14,23,24] IBIs recorded during undisturbed lying bouts were involved in the analysis. Focal cows were not removed from their production groups and, consequently, from their familiar surroundings. Lying periods were recorded by visual observations, from a distance of at least 6 m from the resting area. Observers used watches synchronized with the heart rate receivers to later match behavior with the individual’s heart rate recordings.

For analysis, equal lengths of 5-min time windows were selected from the IBI signal stream [13,25]. A total of 1084 valid 5-min time windows were used for HRV analysis, 560 from lame cows [Means ± SD (range); 10.0 ± 1.3 (8–11) per cow] and 524 from sound ones [Means ± SD (range); 10.3 ± 1.5 (9–12) per cow]. The Kubios HRV software (version 2.1, Biomedical Signal Analysis Group, Department of Applied Physics, University of Kuopio, Finland) was used for HRV analysis [26]. For artefact correction, the custom filter was used and set at 0.3, identifying IBIs differing from the previous IBI by more than 30% as artefacts. After abnormal interval removal, the algorithm of the program substitutes detected errors with interpolated intervals calculated from the differences between previous and next accepted IBIs. In addition, a visual inspection of the corrected data was performed to edit out any artefacts still existing (generally Type 4 and Type 5 errors, for details see Ref. [12]). For removing slow nonstationary trend components, the ‘smoothness priors’ based detrending approach was chosen with λ = 1000 and fc = 0.029 Hz as used by a recent study on dairy cattle HRV [27].

The root mean square of successive differences (RMSSD) between consecutive IBIs were evaluated in the time domain, which is useful for detecting tendencies related to vagal tone in dairy cattle [10] and was used in our previous studies as well [28,29]. For computing frequency-domain HRV, IBI data were subjected to Fast Fourier Transformation (FFT) for power spectrum analysis [30]. The HRV spectrum was calculated with the FFT based Welch’s periodogram method set to 256s overlapping segments with 50% window overlap. The interpolation rate of IBI series was 4 Hz. Spectral parameters included the normalized power of the high-frequency (HF) band and the relative power of the low frequency (LF) component and HF (LF/HF). The HF component is strongly associated with vagal tone [31] while LF/HF gives information about the sympathovagal balance of the ANS [25,32]. The limits of the spectral components were set as follows: LF: 0.05–0.20 Hz, and HF: 0.20–0.58 Hz as advised by von Borell et al. [13].

For graphical representation of the correlation between successive IBIs, where each interval in the time series (IBIi+1) is plotted against its successor (IBIi), standard deviation 1 (SD1) and the ratio between standard deviation 2 (SD2) and SD1 (SD2/SD1) were calculated by Poincaré plot analysis. SD1 represents vagal nerve activity [33], while SD2/SD1 reflects the sympathovagal balance. Both parameters have been used in dairy cattle [28,3436].

For geometric means of HRV the triangular interpolation of the IBIs (triangular interpolation of normal to normal, TINN) and the triangular index (RRtri index) were calculated. TINN approximates the IBI distribution by a linear function and the baseline width of this approximation triangle is used as a measure of the HRV [37,38]. The RRtri index is a measure, where the length of IBIs serves as the X-axis of the plot and the number of each IBI length serves as the Y-axis. The length of the base of the triangle is used and approximated by the main peak of the IBI frequency distribution diagram. This parameter is insensitive to artefacts and ectopic beats, because they are left outside the triangle [25].

For the calculation of non-linear parameters, the Recurrence Quantification Analysis (RQA) was used [39]. This procedure was already applied successfully in human medicine [40] and in animal models [41]. The LMAX and the Shannon Entropy indices were chosen from RQA measures previously used in dwarf goat [24] and in dairy cattle [42].

LMAX is the length of the longest line of recurrence points in a continuous row within the recurrence plot resulted from the RQA where a small LMAX means a large amount of ‘chaos’ [13]. Shannon Entropy means the Shannon information entropy of the line length distribution and measures the complexity of the signal.

In addition, we used the Detrended Fluctuation Analysis (DFA), which was introduced by Peng et al. [43]. DFA is useful for analyzing time series that appear to be long-memory processes for determining the statistical self-affinity of a signal by quantifying the fractal scaling properties of short-interval IBI data series. In this study we evaluated short-term (DFA1, computed for 3–11 IBIs) and long-term fluctuations (DFA2, computed for 12–20 IBIs). These parameters reflect the amount of randomness in the IBI time series [8]. Besides RQA and DFA measures the Correlation Dimension Analysis (CDA) parameter Correlation Dimension (CD) was calculated. The CD value will be high for the chaotic data and it decreases as the variation of the IBI signal becomes less or rhythmic [11] therefore CD was found to be useful in determining pathological signals [9]. For the more detailed description of the mathematical background of non-linear characteristics of HRV calculated with the RQA, DFA and CDA methods see the work of Acharya et al. [11].

Fecal cortisol metabolites

Since the concentration of cortisol metabolites in a fecal sample reflects the cortisol production after 12 hours in ruminants [15], feces samples for analysis of cortisol metabolites were collected two times daily for the evaluation of the activity of the hypothalamic-pituitary-adrenal (HPA) axis. About 50 to 100 g of feces was obtained manually from the rectum in the morning (immediately after fixing the heart rate monitors, between 6:00 am and 8:30 am) and in the evening (before removing the heart rate monitors, between 6:00 pm and 8:30 pm). Disposable gloves were changed after every cow. According to Palme [44], 10 to 15 g of feces from different locations on the glove was filled into sample tubes. Samples were stored on ice immediately and frozen at −20°C within 2 hours after collection until analysis as described by Möstl et al. [45]. For the extraction of the fecal glucocorticoid metabolites [46], samples were thawed at room temperature, stirred, and 0.5 g of feces was dispersed in 5 ml of 80% methanol [47]. The dispersion was vortexed for 30 min and centrifuged on +4°C at 3000 × g for 20 min [45]. The supernatant was transferred into Eppendorf tubes of 1.5 ml and stored at −20°C until further analysis. A H3 homemade radioimmunoassay was carried out to determine the concentration of fecal glucocorticoid metabolites following the description of Csernus [48]. All samples were analyzed in duplicate. Intraassay and interassay coefficients of variation were calculated.

Statistical analysis

Statistical evaluation was carried out in SPSS 18 (SPSS Inc., Chicago, IL). The GLM univariate procedure was used for modeling the linear relationship between dependent scale variables (HRV indices and fecal cortisol concentrations—one model for each), and one or more categorical and scale predictors. Averaged values of all HRV parameters calculated for 5-min IBI segments were used for each animal in the statistical evaluation to avoid pseudo-replication. Categorical predictors (main effects) were selected as fixed factors (parity: 1, 2, 3, 4 lactations, days in milk: between 140 and 160 days, body condition score: 2, 2.5, 3, 3.5 and lameness category: non-lame and lame) in the models. Age of cows and milk yield on experimental days were included in the models as covariates. Before each analysis, Levene’s test was used for testing equality of error variances. Tests of between-subjects effects were also determined for all sources of variances. The confidence intervals were adjusted with the post hoc test by Bonferroni. The level of significance was set at 0.05.


None of the main factors had a significant effect on mean fecal glucocorticoid concentrations of samples collected either in the morning or in the evening. Fecal cortisol concentrations of lame and non-lame cows are shown in Table 2.

Table 2. Fecal cortisol concentrations of lame (n = 51) and non-lame (n = 52) dairy cows.

Days in milk and body condition had no effect on HRV parameters. Parity had a slight effect on RMSSD (univariate test: F3,87 = 3.18, P = 0.007) and on LMAX (univariate test: F3,87 = 9.60, P = 0.003). LMAX was lower in first lactation heifers than in fourth lactation cows (157 ± 20 vs. 265 ± 32 ms, P = 0.025). In second lactation cows lower RMSSD was found (15.2 ± 2.5 ms) than in third or fourth lactation cows (19.8 ± 3.0 ms and 35.5 ± 5.7 ms, P = 0.045 and P = 0.028, respectively).

Lameness had a pronounced effect on the baseline cardiac activity of cows. Heart rate was approximately 10 beats/min lower in lame cows than in non-lame animals. Parasympathetic measures both in the time- (RMSSD) and frequency-domain (HF) were higher in lame cows than in non-lame ones, while the indices of sympathovagal balance (LF/HF and SD2/SD1) were lower in lame cows than in sound cows. All geometric and non-linear HRV measures were lower in lame cows compared to non-lame ones. Lameness had no effect only on SD1 parameter. Values of HRV parameters in lame and non-lame cows are presented in Table 3.

Table 3. Heart rate and heart rate variability parameters of lame (n = 51) and non-lame (n = 52) dairy cows during lying posture.


We investigated time-domain, frequency-domain, geometric and non-linear HRV parameters as well as fecal glucocorticoid concentrations for the description of chronic changes in the cardiac IBI time series and HPA axis activity in dairy cattle. Although a variety of techniques based on non-linear theories has been proposed to quantify HRV, there are only a few number of studies on the evaluation of non-linear HRV in dairy cattle. According to recent research on HRV analysis methods, non-linear indices have many advantages compared to traditional linear parameters: 1) higher insensitivity to the presence of error IBIs [11]; 2) better reproducibility [49]; 3) higher sensitivity to smaller modulations in HR [50]. Since our paper is the first that describe the effects of chronic stress and HRV in adult cattle, it is difficult to compare our results to earlier findings in this field. Most of the studies on the effects of chronic physiological load on HRV has involved humans and lab animal species.

With the exception of SD1, all cardiac variables were affected by lameness. Pairwise comparisons of the values of linear indices highlighted marked differences in HR and ANS indices of HRV (HF and LF/HF) between lame and non-lame animals. Although heart rates of the lame cows were significantly lower than that of healthy cows, they were on average 18 beats/min higher than the threshold for pathological bradycardia (48 beats/min) [51]. Inconsistent with our results, earlier studies on conscious rats found that heart rate of animals exposed to prolonged intermittent stress [52] or chronic social stress [53] were similar to that of control animals. A recent human study on the cardiovascular consequences of chronic pain [7], found no differences in heart rate between affected and control patients. In line with our findings, Austin et al. [54] have reported on bradycardia in BSE positive cattle. The authors explain that though the disease affects the central nervous system primarily, including sites influencing heart rate, lower heart rates in infected animals is associated with low food intake. As lame cows tend to be spending more time lying and less time feeding [55], their lower heart rates may be associated with a low overall activity and food intake in our study as well. Since we found higher RMSSD and HF in lame animals compared to sound ones, it is possible that lameness-related bradycardia is mediated by increased vagal tone. Despite the conflicting reports on the functional roles of endogenous opioids in stress regulation [56], depressed heart rates of lame cows is possibly a result of the effect of endogenous opioids [5760]. Some studies have reported on opioid secretion in a prolonged pain situation [61,62] and others highlighted the adaptation of opioid systems to protect the animals from excessive physiological reactions to stress [63]. A limitation of our study, that opioid parameters were not measured.

We found higher parasympathetic and lower sympathetic activity in lame cows (higher RMSSD, HF and lower LF/HF and SD2/SD1 ratios), while earlier studies reported on the predominance of the vagal tone indicated by a significant increase of RMSSD [64] or HF [65] in cows positive for BSE. Both results reflect on elevated parasympathetic tone in case of chronic physiological loads, however, it is not clear whether this phenomenon and a lower HR is a direct consequence of chronic stress or associated with a lower locomotor activity of lame animals. According to Nanda et al. [66], the HPA axis appears to be under suppressive opioidergic control, however, cattle under long-term pain may react in an opposite manner. The limtation of our study is that we did not determine locomotor activity or lying time and frequency of focal animals. In a recent work we found that lying time and lying frequency did not influence ANS activity in healthy cows (unpublished results), however, in this paper it cannot serve as a basis of interpretation regarding HR and vagal tone. For exact conclusions on lameness-related ANS activity further investigations are required where the behavior of animals as well as their overall activity is controlled.

Geometric measures of HRV (RRtri index, TINN) were lower in lame cows than in sound animals. These indices describe the overall variability in IBIs [25] and a reduction in these parameters is indicative of stress [67]. Our observations in lame cows are consistent with similar data obtained in patients with chronic emotional distress [68,69], chronic obesity [70], chronic cardiovascular [71] and pulmonary diseases [5], observing lower values of RRtri index and TINN compared to control subjects.

It is well known from studies in human cardiology that the time course of HRV contains non-linear chaotic components [72,73] and it is generally recognized that non-linear techniques are effective when describing biological systems [9]. Non-linear parameters of HRV have been studied in relation with chronic alterations in cardiac activity examining periods of undisturbed lying and thus of minimal physical activity for the analysis of the cardiac IBI [23,24]. The indices of non-linear HRV in the present study referred to a loss of general freedom in cardiovascular dynamics. Both RQA and DFA methods showed a difference between groups. All non-linear parameters calculated with RQA (LMAX, Shannon entropy) or with DFA (DFA1, DFA2) as well as Correlation Dimension were statistically lower in lame cows than in non-lame cows. The decreased value of complexity measures (LMAX, Shannon entropy) in lame cows reflects a change towards a more periodic heart rate under stress, indicating stronger regularity and decoupling of multimodal integrated networks within the cardiovascular system. This reduction in heart rate complexity in affected animals may reflect a lower adaptability and fitness of the cardiac pacemaker [14].

According to human studies, health state and well-being [6,74], or the function of the ANS [75,76] can be described by means of DFA indices. Krstacic et al. [77] observed a decrease in DFA parameters parallel with a decreased SNS activity in men with cardiomyopathy. This is in accordance with our findings on lower LF/HF and SD2/SD1 ratios in lame cows than in non-lame ones and a reduction in DFA1 and DFA2. In line with our results, Correlation Dimension was lower in patients with cardiac disorders [78] or with chronic mental stress disorder [79].

In this work, besides lameness, we involved production variables in the analysis to evaluate their possible effects on HRV. Days in milk and body condition score did not affect HRV parameters. Parity was the only factor that influenced cardiac activity but only in case of two parameters. RMSSD and LMAX were higher in cows with higher parity. These results seem to be indicative for a lower vagal tone and a lower degree of complexity in cardiac dynamics in younger animals, however, the differences are so minor that no serious effect of parity on cardiac function can be testified.

Measuring fecal cortisol metabolites as a feedback-free method is referred to as a valuable tool in several research fields, e.g. animal welfare [15]. In case of monitoring prolonged periods of elevated glucocorticoid production, the precise sampling protocol is not that critical [80] and the concentrations of cortisol metabolites in a fecal samples reflect the cortisol production after 10–12 hours in cattle [15], we decided to collect feces samples two times daily. This regime enabled the undisturbed IBI recording for the entire experimental days. None of the production variables affected mean fecal glucocorticoid concentrations in the present study. In opposite to our expectations, we did not find a significant effect of lameness on fecal glucocorticoid concentrations either in samples collected in the morning, or in the evening. Walker et al. [17] also found similar cortisol concentrations in lame and sound cows, albeit they measured milk for cortisol. In earlier studies cortisol determination for feces has been judged a crude and highly variable parameter for stress analysis [81] due to the high inter-animal variations [80]. We tried to overcome this problem by involving a large number of animals in this study, however, our results might mirror the combined effects of several housing-associated stressors which could have masked the effects of lameness on the HPA activity. Another explanation of our results might be that cortisol responses gradually diminish with repeated stressor exposure [82]. Supposedly, a longer-term monitoring of fecal cortisol metabolites or cortisol concentrations in hair [83] would be demonstrative of HPA axis activation due to long-term stress caused by lameness.

Based on our findings, linear and non-linear analysis of HRV are promising approaches for chronic stress detection. Further research might be beneficial in the evaluation of treatments or newly developed technologies targeted on the reduction of the effects of chronic stress in dairy housing systems.


Chronic intermittent stress—e.g. lameness—induces distinct changes in linear and non-linear components of HRV. Heart rate was lower, while vagal tone was higher in lame animals. The decline of non-linear heartbeat dynamics indicate a loss of complexity in heart rate and a more deterministic control of HRV in reaction to chronic stress load. It requires further research to understand whether elevated parasympathetic tone and depressed HR are associated with lower locomotor activity or the expression of long-term stress caused by lameness. It appears that chronic pain does not necessarily lead to a change in fecal cortisol concentrations.


The study was supported by the Research Centre of Excellence—9878/2015/FEKUT project.

Author Contributions

Conceived and designed the experiments: LK. Performed the experiments: LK FLK VJ MK-H. Analyzed the data: LK JT. Contributed reagents/materials/analysis tools: MK-H. Wrote the paper: LK.


  1. 1. Ventura BA, von Keyserlingk MAG, Weary DM. Animal welfare concerns and values of stakeholders within the dairy industry. J Agric Environ Ethics. 2014.
  2. 2. Jurkovich V, Fóris B, Végh Á, Kovács P, Könyves L, Brydl E. Assessing welfare quality in Hungarian dairy cattle herds. Magy Állatorv L. 2012;134: 605–613. (In Hungarian).
  3. 3. Gudaj R, Brydl E, Lehoczky J, Komlósi I. Different management methods on prevalence of lameness in 25 Holstein-Friesian herds in Hungary. Acta Vet Beogr. 2013;63: 405–420.
  4. 4. Jurkovich V, Olaszy K, Lehoczky J, Könyves L, Tirián A, Brydl E. Prevalence of claw disorders in dairy herds. Magy Állatorv L. 2007;129: 468–473. (In Hungarian).
  5. 5. Carvalho de TD, Pastre CT, Rossi RC, Abreu de LC, Valenti VE, Vanderlei LCM. Geometric index of heart rate variability in chronic obstructive pulmonary disease. Rev Portug Pneumol. 2011;17: 260–265.
  6. 6. Stadnitsuki T. Some critical aspects of fractality research. Non-linear Dynam Psychol. 2012;16: 137–158.
  7. 7. Meeus M, Goubert D, De Backer F, Struyf F, Hermans J, Coppieters I, et al. Heart rate variability in patients with fibromyalgia and patients with chronic fatigue syndrome: a systematic review. Semin Arthritis Rheum. 2013;43: 279–287. pmid:23838093
  8. 8. Voss A, Schulz S, Schroeder R, Baumert M, Caminal P. Methods derived from non-linear dynamics for analysing heart rate variability. Philos Trans A Math Phys Eng Sci. 2009;367: 277–296. pmid:18977726
  9. 9. Acharya UR, Joseph KP, Kannathal N, Lim ChM, Suri JS. Heart rate variability: a review. Med Bio Eng Comput. 2006;44: 1031–1051.
  10. 10. Kovács L, Jurkovich V, Bakony M, Póti P, Szenci O, Tőzsér J. Welfare assessment in dairy cattle by heart rate and heart rate variability—Literature review and implications for future research. Animal. 2014;8: 316–330.
  11. 11. Acharya ÚR, Joseph KP, Kannathal N, Lim CM, Suri JS. Heart rate variability: a review. Med Bio Eng Comput. 2006;44: 1031–1051.
  12. 12. Marchant-Forde RM, Marlin DJ, Marchant-Forde JN. Validation of a cardiac monitor for measuring heart rate variability in adult female pigs: accuracy, artefacts and editing. Physiol Behav. 2004;80: 449–458. pmid:14741229
  13. 13. von Borell E, Langbein J, Després G, Hansen S, Leterrier C, Marchant-Forde J, et al. Heart rate variability as a measure of autonomic regulation of cardiac activity for assessing stress and welfare in farm animals: a review. Physiol Behav. 2007;92: 293–316. pmid:17320122
  14. 14. Hagen K, Langbein J, Schmied C, Lexer D, Waiblinger S. Heart rate variability in dairy cows—influences of breed and milking system. Physiol Behav. 2005;85: 195–204. pmid:15894344
  15. 15. Möstl E, Palme R. Hormones as indicators of stress. Dom Anim Endocrinol. 2002;23: 67–74.
  16. 16. Almeida PE, Weber PSD, Burton JL, Zanella AJ. Depressed DHEA and increased sickness response behaviours in lame cows with inflammatory foot lesions. Domest Anim Endocrinol. 2008;34: 89–99. pmid:17229542
  17. 17. Walker SL, Smith RF, Jones DN, Routly JE, Dobson H. Chronic stress, hormone profiles and estrus intensity in dairy cattle. Horm Behav. 2008;53: 493–501. pmid:18206887
  18. 18. Morrow CJ, Klover ES, Verkerk GA, Mathews LR. Fecal glucocorticoid metabolities as a measure of adrenal activity in dairy cattle. Gen Comp Endocrinol. 2002;126: 229–241. pmid:12030779
  19. 19. Palme R, Fischer P, Schildorfer H, Ismail MN. Excretion of infused 14C-steroid hormones via faeces and urine in domestic livestock. Anim Repr Sci. 1996;43: 43–63.
  20. 20. Merl S, Scherzer S, Palme R, Möstl E. Pain causes increased concentrations of glucocorticoid metabolites in horse faeces. J Equine Vet Sci. 2000;20: 586–590.
  21. 21. Sprecher DJ, Hostetler DE, Kanneene JB. A lameness scoring system that uses posture and gait to predict dairy cattle reproductive performance. Theriogenology. 1997;47: 1179–1187. pmid:16728067
  22. 22. Whay H. Locomotion scoring and lameness detection in dairy cattle. In Pract. 2002;24: 444–449.
  23. 23. Langewitz W, Rüddel H, Schächinger H, Lepper W, Mulder LJM, Veldman JHP, et al. Changes in sympathetic and parasympathetic cardiac activation during mental load: an assessment by spectral analysis of heart rate variability. Homeostasis. 1991;33: 1–2.
  24. 24. Langbein J, Nürnberg G, Manteuffel G. Visual discrimination learning in dwarf goats and associated changes in heart rate and heart rate variability. Physiol Behav. 2004;82: 601–609. pmid:15327907
  25. 25. ESC-NASPE Task Force (Task Force of the European Society of Cardiology, North American Society of Pacing and Electrophysiology). Heart rate variability: standards of measurement, physiological interpretation, and clinical use. Circulation. 1996;93: 1043–1065. pmid:8598068
  26. 26. Tarvainen MP, Niskanen J-P, Lipponen JA, Ranta-aho PO, Karjalainen PA. Kubios HRV—Heart rate variability analysis software. Comp Meth Prog Biomed. 2014;113: 210–220.
  27. 27. Frondelius L, Järvenranta K, Koponen T, Mononen J. The effects of body posture and temperament on heart rate variability in dairy cows. Physiol. Behav. 2015;139: 437–441. pmid:25481355
  28. 28. Kovács L, Tőzsér J, Szenci O, Póti P, Kézér FL, Ruff F, et al. Cardiac responses to palpation per rectum in lactating and nonlactating dairy cows. J Dairy Sci. 2014;97: 6955–6963. pmid:25200771
  29. 29. Kovács L, Tőzsér J, Kézér FL, Ruff F, Aubin-Wodala M, Albert E, et al. Heart rate and heart rate variability in multiparous dairy cows with unassisted calvings in the periparturient period. Physiol Behav. 2015;139: 281–289. pmid:25449409
  30. 30. Akselrod S, Gordon D, Madwed JB, Snidman NC, Shannon DC, Cohen RJ. Hemodynamic regulation: investigation by spectral analysis. Am J Physiol. 1985;249: 867–875.
  31. 31. Porges SW. The polyvagal theory: phylogenetic contributions to social behavior. Physiol Behav. 2003;79: 503–513. pmid:12954445
  32. 32. van Ravenswaaij-Arts CMA, Kollée LAA, Hopman JCV, Stoelinga GBA, Vangeijn HP. Heart-rate-variability. Ann Int Med. 1993;118: 436–447. pmid:8439119
  33. 33. Tarvainen MP, Niskanen JP. Kubios HRV Version 2.2 User’s Guide. 2014. Available:
  34. 34. Minero M, Canali E, Ferrante V, Carenzi C. Measurement and time domain analysis of heart rate variability in dairy cattle. Vet Rec. 2001;149: 772–774. pmid:11808666
  35. 35. Kovács L, Bakony M, Tőzsér J, Jurkovich V. Short communication: The effect of milking in a parallel milking parlor with non-voluntary exit on the HRV of dairy cows. J Dairy Sci. 2013;96: 7743–7747.
  36. 36. Kovács L, Kézér FL, Tőzsér J. Measuring stress level of dairy cows during milking using by geometric indices of heart rate variability. SPASB. 2013;46: 213–217.
  37. 37. Malik M, Farrell T, Cripps T, Camm AJ. Heart rate variability in relation to prognosis after myocardial infarction: selection of optimal processing techniques. Eur Heart J. 1989;10: 1060–1074. pmid:2606116
  38. 38. Farrell TG, Bashir Y, Cripps T, Malik M, Poloniecki J, Bennett ED, et al. Risk stratification for arrhythmic events in post infarction patients based on heart rate variability, ambulatory electrocardiographic variables and signal averaged ECG. J Am Coll Cardiol. 1991;18: 687–697. pmid:1822090
  39. 39. Zbilut JP, Thomasson N, Webber CL. Recurrence quantification analysis as a tool for non-linear exploration of nonstationary cardiac signals. Med Eng Phys. 2002;24: 53–60. pmid:11891140
  40. 40. Giuliani A, Piccirillo G, Marigliano V, Colosimo A. A non-linear explanation of ageing-induced changes in heartbeat dynamics. Am J Physiol Heart Circ Physiol. 1998;275: 1455–1461.
  41. 41. González JJ, Cordero JJ, Feria M, Pereda E. Detection and sources of non-linearity in the variability of cardiac R–R intervals and blood pressure in rats. Am J Physiol Heart Circ Physiol. 2000;279: 3040–3046.
  42. 42. Mohr E, Langbein J, Nürnberg G. Heart rate variability: A noninvasive approach to measure stress in calves and cows. Physiol Behav. 2002;75: 251–259. pmid:11890975
  43. 43. Peng CK, Buldyrev SV, Havlin S, Simons M, Stanley HE, Goldberger AL. On the mosaic organization of DNA sequences. Phys Rev. 1994;49: E1685–E1689.
  44. 44. Palme R. Measuring fecal steroids: Guidelines for practical application. Ann N Y Acad Sci. 2005;1046: 75–80. pmid:16055844
  45. 45. Möstl E, Maggs JL, Schrötter G, Besenfelder U, Palme R. Measurement of cortisol metabolites in faeces of ruminants. Vet Res Comm. 2002;26: 127–139.
  46. 46. Rumball CW, Oliver MH, Thorstensen EB, Jaquiery AL, Husted SM, Harding JE, et al. Effects of twinning and periconceptional undernutrition on late-gestation hypothalamic-pituitary-adrenal axis function in ovine pregnancy. Endocrinology. 2008;149: 1163–1172. pmid:18048493
  47. 47. Palme R, Möstl E. Measurement of cortisol metabolites in faeces of sheep as a parameter of cortisol concentration in blood. Int J Mammal Biol. 1997;62: 192–197.
  48. 48. Csernus V. Antibodies of high affinity and specificity for RIA determination of progesterone, testosterone, estradiol-17b and cortisol. In: Görög S. editor. Budapest: I. Academy Press: Advances in steroid analysis; 1982. pp. 171–177.
  49. 49. Mourot L. Could Non-Linear Heart Rate Variability Analysis of Short RR Intervals Series Give Clinically Valuable Information in Heart Disease? J Clin Exp Res Cardiol. 2014;104.
  50. 50. Makikallio TH, Tapanainen JM, Tulppo MP, Huikuri HV. Clinical applicability of heart rate variability analysis by methods based on nonlinear dynamics. Card Electro Physiol Rev. 2002;6: 250–255.
  51. 51. Radostits OM, Gay CC, Hinchcliff KW, Constable PD. Veterinary Medicine: A textbook of the diseases of cattle, horses, sheep, pigs and goats. 10th ed. St Louis: Saunders Ltd.; 2007.
  52. 52. Flak JN, Jankord R, Solomon MB, Krause EG, Herman JP. Opposing effects of chronic stress and weight restriction on cardiovascular, neuroendocrine and metabolic function. Physiol Behav. 2011;104: 228–234. pmid:21396386
  53. 53. Lemaire V, Mormède P. Telemetered recording of blood pressure and heart rate in diffrenet strains of rats during chronic social stress. Physiol Behav. 1995;58: 1181–1188. pmid:8623018
  54. 54. Austin AR, Pawson L, Meek S, Webster S. Abnormalities of heart rate and rhythm in bovine spongiform encephalopathy. Vet Rec. 1997;141: 352–357. pmid:9351181
  55. 55. Galindo F, Broom DM. The effects of lameness on social and individual behaviour of dairy cows. J Appl Anim Welf Sci. 2002;5: 193–201.
  56. 56. Steckler T, Kalin NH, Reul JMHM. Handbook of Stress and the Brain Part 1: The Neurobiology of Stress, 1st ed. Amsterdam: Elsevier Science Publication; 2005.
  57. 57. Feuerstein G, Siren AL. The opioid system in cardiac and vascular regulation of normal and hypertensive states. Circulation. 1987;75: 1–125.
  58. 58. Schmid M, Kuessel L, Klein K, Metz V, Fischer G, Krampl-Bettelheim E. First-trimester fetal heart rate in mothers with opioid addiction. Addiction. 2010;105: 1265–1268. pmid:20642509
  59. 59. Akil H, Watson SJ, Young E, Lewis ME, Khachaturian H, Walker JM. Endogenous opioids: biology and function. Annu Rev Neurosci. 1984;7: 223–255. pmid:6324644
  60. 60. Faden AI, Feuerstein G. Hypothalamic regulation of the cardiovascular and respiratory systems: role of specific opiate receptors. Br J Pharmacol. 1983;79: 997. pmid:6317125
  61. 61. Parikh D, Hamid A, Friedman TC, Nguyen K, Tseng A, Marquez P, et al. Stress-induced analgesia and endogenous opioid peptides: The importance of stress duration. Eur J Pharmacol. 2011;650: 563–567. pmid:21044625
  62. 62. Fredricksen RCA, Geary LE. Endogenous opioid peptides: Review of physiological, pharmacological and clinical aspects. Progr Neurobiol. 1982;19: 19–69.
  63. 63. Loijens CJ, Janssensb JG, Schoutena WGP, Wiegant VM. Opioid activity in behavioral and heart rate responses of tethered pigs to acute stress. Physiol. Behav. 2002;75: 621–626. pmid:12020727
  64. 64. Pomfrett CJD, Glover DG, Bollen BG, Pollard BJ. Perturbation of heart rate variability in cattle fed BSE-infected material. Vet Rec. 2004;154: 687–691. pmid:15200074
  65. 65. Little CJL, Julu POO, Hansen S, Mellor DJ, Milne MH, Barrett DC. Measurement of cardiac vagal tone in cattle: a possible aid to the diagnosis of BSE. Vet Rec. 1996;139: 527–528. pmid:8953696
  66. 66. Nanda AS, Dobson H, Ward WR. Opioid modulation of the hypothalamo-pituitary-adrenal axis in dairy cows. Dom Anim Endocrinol. 1992;9: 181–186.
  67. 67. Hejjel L, Gál I. Heart rate variability analysis. Acta Physiol Hung. 2001;88: 219–230. pmid:12162580
  68. 68. Ramaekers D, Ector H, Demyttenaere K, Rubens A, Van de Wefr F. Association between cardiac autonomic function and coping style in healthy subjects. Pacing Clin Electrophysiol. 1998;21: 1546–1552. pmid:9725152
  69. 69. Orsilia R, Virtanen M, Luukkaala T, Tarvainen M, Karjalainen P, Viik J, et al. Perceived mental stress and reactions in heart rate variability—a pilot study among employees of an electronics company. Int J Occup Saf Ergo. 2008;14: 275–283
  70. 70. Vanderlei LC, Pastre CM, Frietas IF Jr, Godoy MF. Geometric indexes of heart rate variability in obese and eutrophic children. Arq Bras Cardiol. 2010;95: 35–40. pmid:20585738
  71. 71. Li L, Li K, Cheng S, Liu C, Liu C. Five-class density histogram and its application in short-term heart rate variability. J Med Biol Eng. 2012;32: 287–291.
  72. 72. Signorini MG, Cerutti S, Guzzetti S, Parola R. Non-linear dynamics of cardiovascular variability signals. Meth Inf Med. 1994;33: 81–84.
  73. 73. Kanters JL, Hojgaard MV, Agner E, Holsteinrathlou NH. Short- and long-term variations in non-linear dynamics of heart rate variability. Cardiovasc Res. 1996;31: 400–409. pmid:8681327
  74. 74. Goldberger AL, Amaral LAN, Hausdorff JM, Ivanov PC, Peng CK, Stanley HE. Fractal dynamics in physiology: Alterations with disease and aging. Proc Nat Acad Sci. 2002;99: 2466–2472. pmid:11875196
  75. 75. Perkiomaki JS, Zareba W, Badilini F, Moss AJ. Influence of atropine on fractal and complexity measures of heart rate variability. Ann Noninv Electrocardiol. 2002;7: 326–331.
  76. 76. Castiglioni P, Parati G, Di Rienzo M, Carabalona R, Cividjian A, Quintin L. Scale exponents of blood pressure and heart rate during autonomic blockade as assessed by detrended fluctuation analysis. J Physiol. 2011;589: 355–369. pmid:21115648
  77. 77. Krstacic G, Parati G, Gamberger D, Castiglioni P, Krstacic A, Steiner R. Heart rate variability and non-linear dynamic analysis in patients with stress-induced cardiomyopathy. Med Biol Eng Comp. 2012;50: 1037–1046.
  78. 78. Acharya UR, Kannathal N, Krishnan SM. Comprehensive analysis of cardiac health using heart rate signals. Physiol Meas. 2004;25: 1130–1151.
  79. 79. Melillo P, Bracale M, Pecchia L. Non-linear Heart Rate Variability features for real-life stress detection. Case study: students under stress due to university examination. BioMed Eng OnLine. 2011;10: 96. pmid:22059697
  80. 80. Palme R, Robia C, Meßmann S, Hofer J, Möstl E. Measurement of faecal cortisol metabolites in ruminants: a non-invasive parameter of adrenocortical function. Wien Tierärztl Mschr. 1999;86: 237–241.
  81. 81. Ohmura H, Hiraga A, Aida H. Changes in heart rate and heart rate variability in Thoroughbreds during prolonged road transport. Am J Vet Res. 2006;67: 455–462. pmid:16506907
  82. 82. Schrader L, Ladewig J. Temporal differences in the responses of the pituitary adrenocortical axis, the sympathoadrenomedullar axis, heart rate, and behavior to a chronic intermittent stressor in domestic pigs. Physiol Behav. 1999;66: 775–783.
  83. 83. Comin A, Peric T, Corazzin M, Veronesi MC, Meloni T, Zufferli V, et al. Hair cortisol as a marker of hypothalamic-pituitary-adrenal axis activation in Friesian dairy cows clinically or physiologically compromised. Livest Sci. 2013;152: 36–41.