ITALIAN JOURNAL OF ANIMAL SCIENCE 2021, VOL. 20, NO. 1, 1829–1836 https://doi.org/10.1080/1828051X.2021.1957724 PAPER Heart rate variability in newborn foals and its association with illness: a pilot study Valentina Vitalea,b, Giuseppe Contec , Paolo Baraglia, Eduard Jose-Cunillerasd and Micaela Sgorbinia,b a Dipartimento Scienze Veterinarie, University of Pisa, Pisa, Italy; bOspedale Didattico Veterinario “Mario Modenato”, University of Pisa, Pisa, Italy; cDipartimento di Scienze Agrarie, Alimentali e Agro-Alimentari, University of Pisa, Pisa, Italy; dDepartament de Medicina i Cirurgia Animals, Autonomous University of Barcelona, Barcelona, Spain ABSTRACT ARTICLE HISTORY This study aimed to investigate differences in heart rate variability (HRV) between healthy and Received 19 March 2021 sick neonatal foals with a variety of diagnosis and to verify whether some HRV parameters could Revised 4 June 2021 be associated with sepsis and/or survival. Twenty-one foals were included in the study: nine Accepted 9 July 2021 were healthy and 12 were sick. Retrospectively, sick foals were divided into sub-groups (non- KEYWORDS septic vs. septic and non-survivor vs. survivor) for statistical analysis. Heart rate was recorded Horse; heart rate variability; daily for 20 min and a 5-min period was used for heart rate variability analysis. Data on HRV var- neonatal disease; sepsis iables were analysed using a linear model. Least-square means with their standard errors were reported, and treatment effects were declared significant at p < .05. To isolate the group that differed from the others, Tukey’s test was used as a multiple comparison procedure. Pearson’s correlations between variables were calculated, to estimate their relationship. Standard Deviation of the RR intervals and Standard Deviation 2 by the Poincare plot was significantly lower in sick foals compared with healthy subjects, but there were no differences between survi- vors and non-survivors. Healthy foals presented a significantly lower Sample Entropy and heart rate compared with sick foals. HRV analysis is a non-invasive, rapid, and economical measure- ment tool that can be added to other parameters to improve the accuracy of predicting in-hos- pital mortality in foals. Further studies should be conducted with a larger sample of foals to confirm these findings and to evaluate the clinical usefulness of HRV analysis during neo- natal diseases. HIGHLIGHTS  This study investigated differences in heart rate variability (HRV) between sick and healthy neonatal foals  Standard Deviation of the RR intervals was significantly lower in sick compared with healthy foals, but there were no differences between survivors and non-survivors.  HRV analysis is a non-invasive, rapid, and economical tool that can be added to other param- eters to improve the accuracy of predicting mortality in foals. Introduction expensive, the early prognosis for survival on admis- Sepsis in neonatal foals remains a leading cause of sion or during the first few hours of hospitalisation morbidity and mortality despite the improvement in should be considered to identify neonates with a survival that has been reported over the last 25 years higher chance of survival (Castagnetti and Veronesi (Hytychova and Bezdekova 2015; Taylor 2015; Wong 2008; Wong and Wilkins 2015). et al. 2018). Identifying the right biomarkers is key to By assigning numerical values to several clinicopa- an earlier and accurate diagnosis of sepsis in the horse thological variables, Brewer and Koterba (1988) devel- (Pusterla et al. 2006; Bonelli et al. 2015a, 2015b; Taylor oped a sepsis scoring system that originally had an 2015; Bonelli et al. 2017). The survival of septicaemic 86% specificity and 93% sensitivity. More recently, the foals varies within a range of 45–81% (Corley and Furr accuracy of this system has been questioned (Corley 2003; Slack et al. 2005; Castagnetti and Veronesi and Furr 2003; Wong et al. 2018) and other prognostic 2008). Since intensive care for newborn foals is very indicators have been sought (Pusterla et al. 2006; CONTACT Valentina Vitale v.vitale_vet@yahoo.es Dipartimento Scienze Veterinarie, Via Livornese snc, 56122 San Piero a Grado, PI, Italy ß 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 1830 V. VITALE ET AL. Gold et al. 2007a, 2007b; Armengou et al. 2008; Hollis with sepsis or could represent early markers of surviv- et al. 2008; Hurcombe et al. 2008; Paltrinieri et al. als in neonatal foals referred to the hospital. 2008; Hurcombe et al. 2009; Barsnick et al. 2011). Heart rate variability (HRV) is a non-invasive meas- Materials and methods urement that can be used to investigate the balance of sympathetic and vagal activity (Task Force of the A total of 21 foals were included in the study. Ethical European Society of Cardiology, North American approval (n 2825/2014) by the Ethics Committee on Society of Pacing and Electrophysiology 1996). HRV is Animal Experimentation of the University of Pisa and measured by determining the continuously changing an owner’s written consent were obtained. Nine of the time interval between successive heartbeats (R–R foals were healthy Standardbreds, that had been born intervals). It is based on the antagonistic oscillatory on the same stud farm, had undergone similar man- influences of the sympathetic and parasympathetic agement conditions, and were used as a control nervous systems on the nodus sinuatrialis (Stucke et al. group. The following inclusion criteria were set for the 2015). HRV is increasingly used in animal research to ‘control group’: (1) normal gestation length analyse changes in the sympathovagal balance related (>320 days) (Lester 2011); (2) unassisted delivery; (3) to disease, psychological and environmental stressors, mares treated for gastrointestinal parasites and vacci- or individual characteristics, such as temperament and nated for equine influenza, tetanus, and equine herpes coping strategies (Von Borell et al. 2007; Stucke virus-1; (4) Apgar Score 7 within 5 min of birth et al. 2015). (Stoneham 2006); (5) good passive transfer of immun- In human medicine, HRV has also been used to pre- ity at 24 h of age (Immunoglobulin G  800 mg/dL)1; dict prognosis in septic patients (Bonjorno Junior et al. (6) righting reflex immediately after birth and sucking 2019; Prabhakar et al. 2019), and several studies have reflex within 20 min, sternal recumbence within 2 min, reported that a decrease in HRV indices is associated standing position within 120 min, first suckling within with higher mortality in critically ill patients (de 180 min (Stoneham 2006); and (7) normal at physical Castilho et al. 2018). HRV analysis has been used espe- examination throughout the study period. cially in neonatal medicine to predict mortality in In addition, 12 sick client-owned foals were infants (Beuch ee et al. 2009; Lake et al. 2014). included. All the foals were referred with different Abnormal HRV measurements, such as decreased complaints to two different veterinary teaching hospi- approximate entropy, decreased sample entropy, tals providing secondary health care during one year. reduced variability, and transient decelerations have The foals were managed under similar circumstances been significantly associated with sepsis or sepsis-like in both the VTHs. Using data from clinical history, illness in premature neonates (Beuchee et al. 2009). physical examination, and clinicopathological analysis Early in the course of neonatal sepsis, there are collected on admission, a sepsis score was calculated reduced heart rate variability and transient decelera- for each foal (Wong et al. 2018). Based on the sepsis tions similar to the foetal distress that appears before score and results of blood culture, each foal was the clinical diagnosis of sepsis or systemic inflamma- assigned to 1 of 2 patient groups: a sick non-septic tory response syndrome (Griffin et al. 2005; Bohanon group (sepsis score <7 and negative results of bac- et al. 2015). In animal models, HRV assessment may teriological culture of blood; n ¼ 6 foals) or a septic provide insights into the acute effect of sympathetic group (sepsis score 7 with or without positive results and cholinergic anti-inflammatory pathways (Zila et al. of bacteriological culture of blood or sepsis score <7 2015). According to some clinical studies, HRV analysis with positive results of bacteriological culture of might play an important role in the continuous moni- blood; n ¼ 6 foals). Retrospectively, the foals enrolled toring of severe infection (Fairchild et al. 2009; in the study were grouped according to the outcome Jarkovska et al. 2015). HRV has been evaluated in new- into survivors (n ¼ 7) and non-survivors (n ¼ 5). The born foals (Nagel et al. 2012; Nagel et al. 2015) how- foals were considered survivors if they had been dis- ever, to the best of the authors’ knowledge, no charged from the hospital, while they were considered studies have been conducted to analyse HRV in sick or non-survivors if they died or had been humanely septic foals. euthanised due to their severe medical prognosis and The aim of this study was thus to investigate differ- not for economic reasons. For details regarding all ences in HRV between healthy and sick neonatal foals foals included in this study, see Table 1. with a variety of diagnosis. We wanted to verify Heart rate was recorded in all foals using a heart whether some HRV parameters could be associated rate monitor2 (connected by a chest belt, as previously ITALIAN JOURNAL OF ANIMAL SCIENCE 1831 Table 1. Foals included in this study were divided into three groups (healthy, sick survivor, and sick non-survivor). Days of Age at 1st Survivor/ Foal Group Sex Breed recording Disease recording (days) not-survivor Hospital 1 Healthy M Standardbred 12 – 1 – 2 Healthy F Standardbred 12 – 1 – 3 Healthy M Thoroughbred 12 – 1 – 4 Healthy F Thoroughbred 12 – 1 – 5 Healthy M Thoroughbred 12 – 1 – 6 Healthy M Thoroughbred 12 – 1 – 7 Healthy M Standardbred 12 – 1 – 8 Healthy F Standardbred 12 – 1 – 9 Healthy M Standardbred 12 – 1 – 10 Non-septic F Standardbred 12 Enteritis 1 Survivor Pisa 11 Non-septic M Standardbred 12 Meconium impaction 1 Survivor Pisa 12 Septic F Thoroughbred 8 Failure of passive transfer immunity and septic arthritis 11 Survivor Pisa 13 Non-septic F Quarter horse 4 Hypoglycemia due to an aggressive mare 8 Survivor Pisa 14 Septic M Arabian 12 Septic arthritis and enteritis; Blood culture: positive 11 Survivor Barcelona 15 Non-septic F Thoroughbred 10 Failure of passive transfer immunity 1 Survivor Pisa 16 Non-septic M Standardbred 12 Omphalophlebitis, umbilical hernia and enteritis 3 Survivor Pisa 17 Septic M Thoroughbred 11 Septic arthritis and patent urachus 5 Non-survivor Pisa 18 Non-septic F Standardbred 4 Hypoxic ischaemic encephalopathy 4 Non-survivor Pisa 19 Septic F Standardbred 6 Omphaloflebitis and pneumonia 2 Non-survivor Pisa 20 Septic M Andalusian 12 Enteritis and omphaloflebitis; Blood culture positive 10 Non-survivor Barcelona 21 Septic F Arabian 1 Septicaemia and pneumonia 3 Non-survivor Barcelona M: male; F: female. described (Nagel et al. 2012; Nagel et al. 2015). In analysis4. The variables of HRV taken into account healthy foals, heart rate was recorded daily for 20 min were: Heart Rate (HR; bpm); Standard Deviation of the starting at 24 h of life for 12 consecutive days. The RR intervals (SDRR; ms); Square root of the mean of heart rate monitor belt was placed on the foals while the sum of the squares of differences between con- they were in their usual stall with their dam. A 15-min secutive RR intervals (RMSSD; ms); Geometric Standard period of familiarisation preceded the start of the Deviations by Poincare plot (SD1 and SD2; ms) and heart rate recording so that the foal could regain Sample Entropion (SampEn). calm. Using a closed-circuit television and a chronom- Data on HRV variables were assessed for distribu- eter, the same (VV) operator recorded from outside tion using the Shapiro-Wilk W test. Data relating to whether the foal was awake or asleep, standing or HRV variables were assessed by comparing first lying down, or if it was running around the stall. This healthy, surviving, and non-surviving foals. was used to select 5 min recordings for HRV analysis Subsequently, healthy, septic, and non-septic foals during which the foals were calm, but not sleeping. were compared. Data on HRV were analysed by an Data were always collected between 5:00 and 8:00 pm. ANOVA, to evaluate if there are significant differences In sick foals, the heart rate was recorded daily for between foals, classifying them as healthy, survivors, 20 min starting within 24 h from admission until dis- and non-survivors. Subsequently, the same statistical charge or death/euthanasia. As with the healthy sub- analysis was performed comparing healthy, septic, and jects, the Polar belt was strapped to the foals in the non-septic foals. hospital box in the presence of the dam. The record- In both cases, data on HRV were analysed using the ings were performed after a 15-min period of acclima- following factorial mixed model5: tion, between 5:00 and 8:00 pm, without interfering with clinical procedures and avoiding medication or yijz ¼ l þ Gi þ Tj þ Gi x Tj þ Hz ½Gi  þ eijz feeding times. Usually, the foals were lying down, however, if they were standing, an operator recorded where: its activity in the stall as with the healthy foals. The data were entered into a computer and under- yijz ¼ HRV variables: HR, SDRR, RMSSD, SD1, went automatic correction of artefacts using the avail- SD2, SampEn; able software3. A 5-min period from each 20 min Gi ¼ fixed effect of the ith group based on the follow- recording was selected, in which the foals were at ing classifications of horse status, which were consid- rest, also taking into account signal quality. Heart rate ered independently: (1) healthy vs. survivors vs. no variability during this 5 min was subsequently analysed survivors; (2) healthy vs. septic vs. no septic. with a free software program designed for HRV Tj ¼ fixed effect of the jth day of the survey (12 days) 1832 V. VITALE ET AL. Hz[Gi] ¼ random effect of the zth horse nested in When comparing the average values of the three the group. groups, the healthy foals showed a significantly lower eijz ¼ random residual HR (96.23 vs. 114.79 bpm) and SampEn (0.82 vs. 1.22 ms), and higher SDRR (47.27 vs. 20.93 ms) and SD2 Least-square means with their standard errors were (65.72 vs. 27.89 ms) compared with non-survivor foals. reported, and treatment effects were declared signifi- Survivor foals showed intermediate values for all the cant at p < .05. To isolate the group that differed from variables considered. No significant differences were the others, HSD Tukey’s test was used as a multiple observed for RMSSD and SD1 in any of the groups. comparison procedure. Finally, Pearson’s correlations An analysis of the differences over time revealed between variables were calculated, to estimate their that non-survivor foals had a higher HR on all the relationship. days except on admission, while the survivor subjects had similar values to healthy foals except on days 7 and 8 when their HR was higher. SDRR and SD2 were Results significantly lower in sick foals compared with healthy subjects, but there were no differences between survi- Foals included in the study are described in Table 1. In vors and non-survivors. However, after 7 days, in sur- the control group, HR was recorded for 11 consecutive vivor foals, these parameters were similar to those of days starting from 24 h of life. In the study group, 4 out the healthy groups. of 5 non-survivor foals were euthanised, while one died Healthy foals presented a significantly lower naturally. In the survivors, the HR was recorded until SampEn (0.82 vs. 1.19 ms) and higher SDRR (47.27 vs. discharge for 11 days in three foals, for 8 days in two 25.51 ms) and SD2 (65.72 vs. 34.24 ms) compared with foals, and 4 days in the other two subjects. In non-survi- septic foals. Non-septic subjects showed intermediate vors, the heart rate was recorded until death in each values for all the variables considered. No significant foal for 11, 6, 4, and 1 day, respectively. differences were observed for HR, RMSSD, and SD1 in HRV parameters for healthy and sick foals, divided any of the groups. into survivor/non-survivors and septic/non-septic are An analysis of the differences over the time period reported in Tables 2 and 3. in healthy foals revealed that the HR was lower from Table 2. HRV parameters presented as mean ± SEM for healthy, survivors and non-survivors and results of the statis- tical comparison. Days p-Value HRV parameter 1 2 3 4 5 6 7 8 9 10 11 SEM G T GxT HR Healthy 101.84 cd 97.60 cd 95.18 cd 101.84 cd 97.60 cd 100.30 cd 96.86 cd 93.72 ad 91.36 ad 92.35 ad 89.85 ad 5.66  ns  Survivors 101.34 cd 101.02cd 93.49 ad 95.57 cd 102.91 cd 96.86 cd 108.05 bc 110.11 bc 97.11 cd 108.75 bc 98.96 cd Not survivors 111.39 ab 116.53 ab 114.30 ab 104.89 ac 102.56 cd 116.58 ab 123.99 ab 121.54 ab 108.69 bc 126.54 aa 116.07 ab SDRR Healthy 44.08 ab 52.80 aa 51.37 aa 49.77 ab 49.12 ab 46.60 ab 47.15 ab 46.37 ab 46.75 ab 41.16 ab 44.82 ab 6.52  ns  Survivors 21.15 ac 27.28 bc 44.42 ab 41.83 ab 35.34 ab 29.64 bc 41.58 ab 41.46 ab 45.89 ab 41.41 ab 37.91 ab Not survivors 24.14 ac 25.77 bc 19.22 ac 20.17 ac 22.25 ac 28.72 bc 24.30 ac 28.15 bc 8.65 ac 16.95 ac 15.25 ac RMSSD Healthy 16.49aa 17.69aa 18.61aa 13.68aa 17.16aa 13.56aa 16.51aa 16.31aa 19.04aa 15.04aa 15.40aa 3.57 ns ns ns Survivors 10.11aa 17.64aa 21.60aa 15.29aa 12.54aa 16.74aa 14.60aa 12.75aa 17.15aa 12.87aa 17.49aa Not survivors 12.68aa 10.99aa 11.01aa 10.91aa 13.66aa 9.43aa 14.54aa 15.84aa 6.19aa 11.04aa 14.04aa SD1 Healthy 11.68aa 12.50aa 13.18aa 9.69aa 12.12aa 10.36aa 11.67aa 11.50aa 13.46aa 10.66aa 11.89aa 2.23 ns ns ns Survivors 7.16aa 12.49aa 15.30aa 10.82aa 8.87aa 11.87aa 10.34aa 9.01aa 11.12aa 9.15aa 12.39aa Not survivors 8.98aa 7.77av 7.77aa 7.75aa 9.69aa 6.69aa 10.25aa 8.30aa 9.40aa 7.85aa 9.90aa SD2 Healthy 61.08 ab 73.57 aa 71.30 aa 69.66 aa 68.24 ab 65.04 ab 65.53 ab 64.50 ab 64.61 ab 57.15 ab 62.20 ab 8.53  ns  Survivors 28.85 ac 36.31 ac 60.61 ab 58.03 ab 48.96 ab 39.86 ab 57.55 ab 57.81 ab 64.96 ab 57.36 ab 51.84 ab Not survivors 32.74 ac 35.42 ac 25.12 ac 27.40 ac 29.59 ac 40.09 ab 32.56 ac 35.31 ac 11.31 ad 22.66 ad 19.31 ad SampEn Healthy 0.81 ac 0.72 ac 0.79 ac 0.70 ac 0.77 ac 0.81 ac 0.81 ac 0.90 ac 0.91 ac 0.93 ac 0.85 ac 0.13  ns  Survivors 1.21 ab 1.35 ab 1.06 bc 0.93 ac 0.98 ac 1.20 ab 0.90 ac 0.76 ac 0.96 ac 0.97 ac 1.09 bc Not survivors 1.05 bc 1.12 bc 1.16 bc 1.36 ab 1.17 bc 1.18 ab 1.22 ab 1.66 aa 1.11 bc 1.15 bc 1.29 ab HR: heart rate (beats/min); SDRR: standard deviation of RR (ms); RMSSD: square root of the mean of the sum of the squares of differences between con- secutive RR (ms); SD1 and SD2: geometric standard deviation by Poincare plot; SampEn: sample entropion; SEM: standard error mean; G: effect of group; T: effect of time; ns: not significant. p < .05, p < .01. a, b, c means within an HRV parameter with different letters differ. ITALIAN JOURNAL OF ANIMAL SCIENCE 1833 Table 3. HRV parameters presented as mean ± SEM for healthy, non-septic, and septic foals and results of the statis- tical comparison. Days p-Value HRV parameter 1 2 3 4 5 6 7 8 9 10 11 SEM G T GxT HR Healthy 101.84 bc 97.60 ac 95.17 cd 101.84 bc 97.60 ac 100.30 bc 96.86 ac 93.72 cd 91.36 ad 92.35 cd 89.85 ad 5.94 ns   Non-septic 107.26 bc 106.61 bc 99.61 bc 99.66 bc 106.95 bc 99.37 bc 108.82 ab 112.10 ab 105.80 bc 112.93 ab 103.14 bc Septic 103.10 bc 108.78 b 105.64 bc 99.34 bc 100.25 bc 111.43 ab 122.95 aa 119.98 ab 95.91 cd 121.16 aa 110.69 ab SDRR Healthy 44.09 aa 52.80 aa 51.38 aa 49.78 aa 49.12 aa 46.60 aa 47.16 aa 46.38 aa 46.76 aa 41.17 aa 44.82 aa 6.52  ns ns Non-septic 19.53 aa 25.46 aa 41.51 aa 38.41 aa 31.97 aa 30.69 aa 39.89 aa 41.15 aa 42.53 aa 39.73 aa 36.23 aa Septic 26.42 aa 28.78 aa 24.13 aa 25.41 aa 28.26 aa 26.16 aa 30.04 aa 30.28 aa 23.84 aa 21.00 aa 19.30 aa RMSSD Healthy 16.49 aa 17.69 aa 18.61 aa 13.68 aa 17.16 aa 13.56 aa 16.51 aa 16.31 aa 19.04 aa 15.04 aa 15.40 aa 3.57 ns ns ns Non-septic 8.33 aa 14.79 aa 20.19 aa 14.28 aa 11.47 aa 17.51 aa 11.05 aa 12.15 aa 15.40 aa 11.57 aa 16.19 aa Septic 15.18 aa 16.17 aa 13.67 aa 12.39 aa 14.54 aa 10.12 aa 19.39 aa 21.39 aa 11.66 aa 12.85 aa 15.85 aa SD1 Healthy 11.68 aa 12.50 aa 13.18 aa 9.69 aa 12.12 aa 10.36 aa 11.67 aa 11.50 aa 13.46 aa 10.66 aa 11.89 aa 2.23 ns ns ns Non-septic 5.90 aa 10.46 aa 14.30 aa 10.09 aa 8.10 aa 12.42 aa 7.82 aa 8.58 aa 9.68 aa 8.20 aa 11.44 aa Septic 10.74 aa 11.46 aa 9.66 aa 8.81 aa 10.33 aa 7.18 aa 13.70 aa 15.13 aa 8.27 aa 9.14 aa 11.19 aa SD2 Healthy 61.09 aa 73.58 aa 71.30 aa 69.67 aa 68.24 aa 65.04 aa 65.53 aa 64.50 aa 64.61 aa 57.16 aa 62.20 aa 8.53  ns ns Non-septic 26.84 aa 34.21 aa 56.59 aa 53.29 aa 44.34 aa 41.20 aa 55.70 aa 57.46 aa 60.62 aa 55.18 aa 49.66 aa Septic 35.56 aa 38.82 aa 31.89 aa 34.69 aa 38.07 aa 36.22 aa 39.99 aa 39.42 aa 32.46 aa 28.19 aa 24.84 aa SampEn Healthy 0.81 aa 0.72 aa 0.80 aa 0.70 aa 0.77 aa 0.82 aa 0.81 aa 0.90 aa 0.92 aa 0.94 aa 0.86 aa 0.13  ns ns Non-septic 1.13 aa 1.30 aa 1.05 aa 1.02 aa 0.98 aa 1.19 aa 0.85 aa 0.77 aa 0.95 aa 0.96 aa 1.08 aa Septic 1.17 aa 1.21 aa 1.19 aa 1.23 aa 1.12 aa 1.21 aa 1.21 aa 1.36 aa 1.09 aa 1.13 aa 1.27 aa HR: heart rate (beats/min); SDRR: standard deviation of RR (ms); RMSSD: square root of the mean of the sum of the squares of differences between con- secutive RR (m); SD1 and SD2: geometric standard deviation by Poincare plot; SampEn: sample entropion; SEM: standard error mean; G: effect of group; T: effect of time; ns: not significant. p < .05, p < 0.01. a, b, c means within an HRV parameter with different letters differ. the 7th day onwards compared with sick foals, but no other differences were observed. Furthermore, no dif- ferences were detected between septic and non-sep- tic patients. Discussion Our study describes an HRV analysis on neonatal Figure 1. healthy and sick foals with a variety of clinical diagno- ses. We found a statistically significant effect of illness on SDRR, SD2, and SampEn. However, no differences (Griffin et al. 2005; Prabhakar et al. 2019), experimental were found between survivors and non-survivors, or animals (Jarkovska et al. 2015; Zila et al. 2015), and between septic and non-septic groups on admission. adult horses (McConachie et al. 2016a, 2016b; Vitale Our results on HRV parameters in healthy foals are et al. 2020). As no changes were observed in RMSSD similar to those already described by Nagel and col- and SD1, we cannot associate the decrease in HRV in leagues (Nagel et al. 2015). Data were collected at the sick animals with a decreased parasympathetic tone same time of the day to prevent circadian but rather with a possible autonomic dysfunction interference. (Vitale et al., 2020). Bonjorno Junior et al. (2019) The decrease in SDRR and SD2 in sick foals indi- already speculated that the reduced HRV in sepsis cates the decreased long-term variability of cardiac may indicate early autonomic dysfunction due to activity (Von Borell et al. 2007). Similarly, in human attenuation of the adrenergic response at the cardio- infants, short-term variability has not been found to myocyte level. change in the presence of sepsis, while long-term vari- When considering the HR in foals we must take ability is significantly lower (Bohanon et al. 2015). into account that the physiology of the cardiovascular The depression of HRV in conditions of systemic system is slightly different from those of adults. Foals inflammation has been repeatedly shown in humans cannot readily increase their stroke volume, thus, to 1834 V. VITALE ET AL. maintain an adequate cardiac output they have to When comparing foals according to their septic sta- increase their HR (Carr, 2014). This might influence the tus, the changes observed were similar. This is prob- HRV which is difficult to quantify at this point due to ably because the majority of septic foals died (4/6) the low number of cases included. Nevertheless, HR and the majority of non-survivors were septic (4/5). was not found different between septic, non-septic Several limitations need to be considered in the and healthy foals while other HRV parameters were. interpretation of the current results. Firstly, the recruit- Our study demonstrates that also in foals, HRV is ment of patients was limited as it was related to the different according to the state of health. The lack of equipment and manpower available. Secondly, the significant differences between the different groups of ages of sick foals are quite different, and this could sick foals during the initial 24 h suggests that HRV have influenced the results. Thirdly, although we analysis performed on admission may not be useful selected our HRV parameters based on a literature when predicting survival or septicaemia. It is also true review, there may have other suitable HRV variables that the number of sick foals enrolled in this study for a comparison between groups. was low which might have been a bias for the statis- tical analysis. Conclusions A recent study reported that when the HRV responses to porcine moderate sepsis without organ In conclusion, HRV analysis is a non-invasive, rapid, failure and to a porcine progressive lethal septic and economical measurement tool, which requires a shock were compared, no differences were found common heart rate monitor which is nearly always (Jarkovska et al. 2015). This probably indicates that freely available in a referral equine hospital, as the the sensitivity of HRV to systemic inflammation is software for HRV analysis. Adding HRV to other com- high but, on the other hand, HRV does not allow the monly used parameters may improve the accuracy of illness to be scaled. A reliable staging of inflamma- predicting in-hospital mortality in foals. Further studies should be conducted with a larger sample of foals to tory processes likely requires an integrative approach confirm these findings and to evaluate the clinical use- with additional parameters, such as clinical, hemato- fulness of HRV analysis during neonatal diseases. logical, and biochemical variables (Peek et al. 2004; Bohanon et al. 2015; Jarkovska et al. 2015; Wong et al. 2018). Acknowledgements Although the accuracy of the sepsis score devel- We would like to thank Dr. Paola Marmorini and the staff of oped by Brewer and Koterba (1988), has been ques- the breeding farm La Piaggia for allowing the healthy foals tioned (Corley and Furr 2003; Slack et al. 2005; to be recorded. Castagnetti and Veronesi 2008; Wong and Wilkins 2015; Wong et al. 2018), it remains the best predictor of survival, with the survival rate decreasing as the Disclosure statement sepsis score increases (Peek et al. 2004; Wilkins 2018; Wong et al. 2018). No potential conflict of interest was reported by the authors. Nevertheless, a variety of ‘new vital signs’, such as tissue oxygenation and HRV, have been emerging in ORCID medicine. These can be associated with blood analysis Giuseppe Conte http://orcid.org/0000-0002-7257-4762 and other vital parameters to obtain more detailed Micaela Sgorbini http://orcid.org/0000-0002-0294-1803 information on the health status of the patient, and thus produce a more reliable prognosis (Bohanon et al. 2015). In addition, in studies on humans, several Data availability statement data points are usually obtained to predict the risk of The authors confirm that the data supporting the findings of death in these patients (Bonjorno Junior et al. 2019). this study are available within the article. The values of SDRR and SD2 in the sick survivors became similar to those of the healthy foals starting References from the seventh day of observation, while in the Armengou L, Monreal L, Taranco n I, Navarro M, Rıos J, non-survivor group the values remained higher. These Segura D. 2008. Plasma D-dimer concentration in sick results support the hypothesis that HRV parameters newborn foals. J Vet Intern Med. 22(2):411–417. return to normal values in foals that recover Barsnick RJIM, Hurcombe SDA, Smith PA, Slovis NM, from illness. Sprayberry KA, Saville WJA, Toribio RE. 2011. Insulin, ITALIAN JOURNAL OF ANIMAL SCIENCE 1835 glucagon and leptin in critically ill foals. J Vet Int Med. in critically ill neonatal foals. J Vet Int Med. 22(5): 25(1):123–131. 1223–1227. Beuchee A, Carrault G, Bansard JY, Boutaric E, Betremieux P, Hurcombe SDA, Toribio RE, Slovis N, Kohn CW, Refsal K, Pladys P. 2009. Uncorrelated randomness of the heart rate Saville W, Mudge MC. 2008. Blood arginine is associated with sepsis in sick premature infants. vasopressin, adrenocorticotropin hormone and cortisol Neonatology. 96(2):109–114. concentrations at admission in septic and critically ill foals Bohanon FJ, Mrazek AA, Shabana MT, Mims S, and their association with survival. J Vet Int Med. 22(3): Radhakrishnan GL, Kramer GC, Radhakrishnan RS. 2015. 639–647. Heart rate variability analysis is more sensitive at identify- Hurcombe SDA, Toribio RE, Slovis NM, Saville WJ, Mudge ing neonatal sepsis than conventional vital signs. Am J MC, MacGillivray K, Frazer ML. 2009. Calcium regulating Surg. 210(4):661–667. hormones and serum calcium and magnesium concentra- Bonelli F, Meucci V, Divers T, Radcliffe R, Jose-Cunilleras E, tions in septic and critically ill foals and their association Corazza M, Guidi G, Tognetti R, Castagnetti C, Intorre L, with survival. J Vet Int Med. 23(2):335–343. et al. 2015a. Evaluation of plasma procalcitonin concentra- Hytychova T, Bezdekova B. 2015. Retrospective evaluation of tions in healthy foals and foals affected by septic systemic blood culture isolates and sepsis survival rate in foals in inflammatory response syndrome. J Equine Vet Sci. 35(8): the Czech Republic: 50 cases (2011–2013). Vet Emerg Crit 645–649. Care. 25(5):660–666. Bonelli F, Meucci V, Divers TJ, Jose-Cunilleras E, Corazza M, Jarkovska D, Valesova L, Chvojka J, Benes J, Sviglerova J, Tognetti R, Guidi G, Intorre L, Sgorbini M. 2015b. Plasma procalcitonin concentration in healthy horses and horses Florova B, Nalos L, Matejovic M, Stengl M. 2015. Heart affected by systemic inflammatory response syndrome. J rate variability in porcine progressive peritonitis-induced Vet Intern Med. 29(6):1689–1691. sepsis. Front Physiol. 6:412. Bonelli F, Meucci V, Divers TJ, Wagner B, Intorre L, Sgorbini Lake DE, Fairchild KD, Moorman JR. 2014. Complex signals M. 2017. Kinetics of plasma procalcitonin, soluble CD14, bioinformatics: evaluation of heart rate characteristics CCL2 and IL-10 after a sublethal infusion of lipopolysac- monitoring as a novel risk marker for neonatal sepsis. J charide in horses. Vet Immunol Immunopathol. 184:29–35. Clin Monit Comput. 28(4):329–339. Bonjorno Junior JC, Caruso FR, Mendes RG, da Silva TR, Lester GD. 2011. Prematurity, dysmaturity and assessment of Biazon T. M P d C, Rangel F, Phillips SA, Arena R, Borghi- maturity. In: McKinnon AO, Squires EL, Vaala WE, Varner Silva A. 2019. Non invasive measurements of hemo- DD, editors. Equine reproduction. 2nd ed. Wiley-Blackwell, dynamic, autonomic and endothelial function as predic- Philadelphia. p. 121–127. tors of mortality in sepsis: a prospective cohort study. McConachie E, Giguere S, Barton MH. 2016a. Scoring system PLOS One. 14(3):e0213239. for multiple organ dysfunction in adult horses with acute Brewer BD, Koterba AM. 1988. Development of a scoring sys- surgical gastrointestinal disease. J Vet Intern Med. 30(4): tem for the early diagnosis of equine neonatal sepsis. 1276–1283. Equine Vet J. 20(1):18–22. McConachie E, Giguere S, Rapoport G, Barton MH. 2016b. Carr EA. 2014. Field triage of the neonatal foal. Vet Clin Heart rate variability in horses with acute gastrointestinal North Am Equine Pract. 30(2):283–300. disease requiring exploratory laparotomy. J Vet Emerg Crit Castagnetti C, Veronesi MC. 2008. Prognostic factors in the Care. 26(2):269–280. sick neonatal foal. Vet Res Commun. 32(S1):87–91. Nagel C, Erber R, Bergmaier C, Wulf M, Aurich J, Mo €stl E, Corley KTT, Furr MO. 2003. Evaluation of a score designed to Aurich C. 2012. Cortisol and progestin release, heart rate predict sepsis in foals. J Vet Emerg Crit Care. 13(3): and heart rate variability in the pregnant and postpartum 149–155. mare, fetus and newborn foal. Theriogenology. 78(4): de Castilho FM, Ribeiro ALP, Nobre V, Barros G, de Sousa 759–767. MR. 2018. Heart rate variability as predictor of mortality in Nagel C, Erber R, Ille N, Wulf M, Aurich J, Mo €stl E, Aurich C. sepsis: a systematic review. PLoS One. 13(9):e0203487. 2015. Heart rate and salivary cortisol concentrations in Fairchild KD, Saucerman JJ, Raynor LL, Sivak JA, Xiao Y, Lake foals at birth. Vet J. 203(2):250–252. DE, Moorman JR. 2009. Endotoxin depresses heart rate Paltrinieri S, Giordano A, Villani M, Manfrin M, Panzani S, variability in mice: cytokine and steroid effects. Am J Veronesi MC. 2008. Influence of age and foaling on Physiol Regul Integr Comp Physiol. 297(4):R1019–R1027. plasma protein electrophoresis and serum amyloid A and Gold JR, Divers TJ, Barton MH. 2007a. Plasma adrenocortico- tropin, cortisol and adenocorticotropin/cortisol ratios in their possible role as markers of equine neonatal septicae- septic and normal-term foals. J Vet Int Med. 21:791–796. mia. Vet J. 176(3):393–396. Gold JR, Perkins GA, Erb HN, Ainsworth DM. 2007b. Cytokine Peek SF, Darien BJ, Semrad SD, McGuirk S, Lien L, Riseberg profiles of peripheral blood mononuclear cells isolated A, Marques F, Slack JA, Coombs D. 2004. A prospective from septic and healthy neonatal foals. J Vet Int Med. 21: study of neonatal septicemia and factors influencing sur- 481–488. vival. AAEP Proc. 50:60–62. Griffin MP, Lake DE, Bissonette EA, Harrell FE, O’Shea TM, Prabhakar SM, Tagami T, Liu N, Samsudin MI, Ng JCJ, Koh Moorman JR. 2005. Heart rate characteristics: novel physi- ZX, Ong MEH. 2019. Combining quick sequential omarkers to predict neonatal infection and death. organ failure assessment score with heart rate variability Pediatrics. 116(5):1070–1074. may improve predictive ability for mortality in septic Hollis AR, Furr MO, Magdesian KG, Axon JE, Ludlow V, patients at the emergency department. PLOS One. 14(3): Boston RC, Corley KTT. 2008. Blood glucose concentration e0213445. 1836 V. VITALE ET AL. Pusterla N, Magdesian KG, Mapes S, Leutenegger CM. 2006. Prunier A, et al. 2007. Heart rate variability as a measure Expression of molecular markers in blood of neonatal of autonomic regulation of cardiac activity for assessing foals with sepsis. Am J Vet Res. 67(6):1045–1049. stress and welfare in farm animals – a review. Physiol Slack JA, McGuirk SM, Erb HN, Lien L, Coombs D, Semrad Behav. 92(3):293–316. SD, Riseberg A, Marques F, Darien B, Fallon L, et al. 2005. Vitale V, Viu J, Armengou L, Rıos J, Jose-Cunilleras E. 2020. Biochemical markers of cardiac injury in normal, surviving Prognostic value of measuring heart rate variability at the septic, or nonsurviving septic neonatal foals. J Vet Intern time of hospital admission in horses with colic. Am J Vet Med. 19(4):577–580. Res. 81(2):147–152. Stoneham SJ. 2006. Assessing the newborn foal. In: Paradis Wilkins PA. 2018. What’s in a word? The need for SIRS and MR, editor. Equine neonatal medicine: a case-based sepsis definitions in equine medicine and surgery. Equine approach. 1st ed. Philadelphia (PA): Elsevier Inc. p. 1–11. Vet J. 50(1):7–9. Stucke D, Große Ruse M, Lebelt D. 2015. Measuring heart Wong DM, Wilkins PA. 2015. Defining the systemic inflamma- rate variability in horses to investigate the autonomic ner- tory response syndrome in equine neonates. Vet Clin vous system activity – pros and cons of different methods. North Am Equine Pract. 31(3):463–481. Appl Anim Behav Sci. 166:1–10. Wong DM, Ruby RE, Dembek KA, Barr BS, Reuss SM, Task Force of the European Society of Cardiology, North Magdesian KG, Olsen E, Burns T, Slovis NM, Wilkins PA. American Society of Pacing and Electrophysiology. 1996. 2018. Evaluation of updated sepsis scoring systems and Heart rate variability: standards of measurement, physio- systemic inflammatory response syndrome criteria and logical interpretation, and clinical use. Circulation. 93: their association with sepsis in equine neonates. J Vet 1043–1065. Intern Med. 32(3):1185–1193. Taylor S. 2015. A review of equine sepsis. Equine Vet Educ. Zila I, Mokra D, Kopincova J, Kolomaznik M, Javorka M, 27(2):99–109. Calkovska A. 2015. Heart rate variability and inflammatory von Borell E, Langbein J, Despres G, Hansen S, Leterrier C, response in rats with lipopolysaccharide-induced endotox- Marchant-Forde J, Marchant-Forde R, Minero M, Mohr E, emia. Physiol Res. 64(5):S669–S676.