Equine Veterinary Journal ISSN 0425-1644 DOI: 10.1111/evj.12325
Descriptive Clinical Reports
The prevalence of abnormal breathing patterns during exercise and associations with dynamic upper respiratory tract obstructions L. E. FITZHARRIS*, S. H. FRANKLIN† and K. J. ALLEN Langford House Equine Hospital, University of Bristol, UK. † School of Animal and Veterinary Sciences, University of Adelaide, South Australia, Australia. *Correspondence email: [email protected]; Received: 14.12.13; Accepted: 16.07.14
Summary Reasons for performing study: There is very limited published information on the prevalence and significance of abnormal breathing patterns adopted during canter/gallop. Objectives: The aim of this study was to report the prevalence of abnormal breathing patterns during canter/gallop and to investigate whether these may be associated with dynamic upper respiratory tract (URT) obstructions. Study design: Retrospective analysis of clinical records. Methods: Simultaneous audio and URT videoendoscopy recordings from 365 horses referred for treadmill evaluation were analysed. Results: Thirty percent of horses had an abnormal breathing pattern at canter or gallop, of which 23% had a 2:1 breathing pattern (i.e. one breath is taken over two strides). The prevalence of abnormal breathing patterns was higher at low canter speeds than during strenuous exercise. There was a significant association between breathing pattern during strenuous exercise and the presence of an URT obstruction. Conclusions: This study suggests that abnormal breathing patterns during canter and gallop may be associated with dynamic URT obstructions; therefore, horses identified by owners or veterinary surgeons to have an abnormal breathing pattern during exercise, in particular a 2:1 breathing pattern, may warrant investigations for diagnosis of respiratory tract disease. Keywords: horse; breathing pattern; exercise; upper airway
Introduction Temporal associations between respiration and locomotion have been reported in many mammals, including horses [1]. Even at racing speeds, trotters and pacers vary in the degree to which locomotion and respiration are synchronised [2,3]. In contrast, during canter and gallop the timing of the respiratory cycle is closely related to limb movements, with most horses adopting a breathing pattern of one breath every one stride [4]. This locomotor-respiratory coupling (LRC) is a breathing strategy that is thought to have both mechanical and energetic advantages [5]. It is suggested that flexion and extension of the back is the primary mechanical link between locomotion and respiration [6–9]. During canter inspiration occurs during the suspension phase of the gait, when the horse is entirely off the ground, or only supported by the hindlimbs, and expiration starts when the nonlead forelimb touches the ground [10]. There are occasions when either a transient or a sustained alteration in this 1:1 breathing pattern may occur. A transient locomotor-respiratory decoupling occurs during ‘big respiratory cycles’ (BRCs), when a single breath of a larger tidal volume is taken over 2 strides [11]. BRCs are thought to be a normal feature during exercise, and it is hypothesised that they result from a negative feedback mechanism providing a transient readjustment of the end expiratory lung volume or an improvement of pulmonary gas exchange [11,12]. It has been suggested that the frequency of BRCs during exercise is highly dependent on the individual horse but may also be inversely associated with fitness levels and respiratory frequency [11,12]. A more sustained alteration in breathing pattern has also been described whereby horses adopt a 2:1 LRC (i.e. each breath is taken over 2 strides). It has been suggested that disorders of the upper and lower respiratory tract may result in loss of the normal 1:1 breathing pattern leading to an abnormal 2:1 breathing pattern. One case report describes a 2:1 breathing pattern in a horse with a dynamic upper respiratory tract (URT) obstruction (intermittent epiglottic entrapment) [13]. This breathing pattern was also observed in research horses with experimentally induced dorsal displacement of the soft palate [14]. In addition, a 2:1 breathing pattern was observed in a research horse when inspired CO2 fraction was increased [15]. It has been reported that occasionally normal horses may not show 1:1 locomotor, respiratory coupling at slow canter speeds, but then adopt the coupling mechanism as speed increases [16]. However, Equine Veterinary Journal 47 (2015) 553–556 © 2014 EVJ Ltd
there is very limited published information on the prevalence of abnormal breathing patterns adopted during canter and gallop, particularly in clinical situations rather than under research situations. Therefore, the aim of this study was to report the prevalence of abnormal breathing patterns during exercise and to investigate whether these may be associated with URT obstructions.
Materials and methods Horses Audio and video recordings from 531 horses referred for treadmill evaluation for poor performance or abnormal respiratory noise to one centre during a 10 year period were reviewed. Only horses in which there was a complete simultaneous recording of both respiratory sounds and URT endoscopy were included in the study. In addition, only data obtained at the canter and gallop were included. Prior to treadmill endoscopy, horses were trained on the high-speed treadmill during 3 training sessions. Horses were habituated to wearing a mask during the third training session. Before testing, horses underwent an initial warm-up period, involving 20 min on a mechanical walker, followed by 2 min of walk at 1.8 m/s, 4 min of trot at 3.5 m/s and 1 min of canter at 6 or 7 m/s on the treadmill. After the warm-up, horses were removed from the treadmill and the endoscope and mask was fitted. Horses were returned to the treadmill and the exercise test and recording commenced. Racehorses performed a standardised incremental exercise test, which comprised of 1 min speed steps at each of 7, 8 and 10 m/s on a 10% incline, then an increase of 1 m/s every minute until fatigue. Nonracehorses performed variable exercise tests whereby the speed and incline were adjusted according to the discipline and level of fitness [17,18]. The step max was the highest speed step achieved.
Audio and video recordings The recording equipment comprised a close fitting facemask sealed with a neoprene shroud. The endoscope was passed into the right nostril to the level of the nasopharynx, positioned in line with the guttural pouch ostia and secured to the facemask. An omnidirectional microphone was placed within the facemask midway between the nares. The microphone was
553
L. E. Fitzharris et al.
Abnormal breathing patterns during exercise
Classification of breathing pattern All endoscopy video and audio recordings were reviewed. A breathing pattern grade was assigned to each horse for each one minute speed step. If more than one grade was observed within a speed step the highest grade was awarded. At the end of the exercise test when a speed step was only partially completed a grade was assigned for the part minute. The type of URT obstruction, or lower respiratory tract disease, was obtained from clinical records. Breathing pattern was graded according to the following definitions: Grade 1 – normal; 1:1 LRC with only 0 or 1 BRC. Grade 2 – normal; 1:1 LRC interrupted by occasional (2 or more) BRCs. Grade 3 – abnormal; baseline 1:1 LRC but rhythm disrupted by single, isolated BRCs, which were frequent and repetitive in nature. Consecutive BRCs did not occur. Grade 4 – abnormal; switching between periods of 2:1 and 1:1 breathing patterns. Grade 5 – abnormal; sustained 2:1 breathing pattern. The term ‘abnormal breathing pattern’ was used to refer to grades 3–5; the term ‘2:1 breathing pattern’ was used to refer to grades 4 and 5.
Data analyses Data analysis was performed using PASW 18.0. Chi-squared tests were used to identify whether there was any significant difference in the prevalence of abnormal breathing patterns for racehorses vs. nonracehorses. Chi-squared testing was also used to investigate potential differences between abnormal breathing patterns and sex. An independent t test was used to investigate the differences in age between groups. Fisher’s exact tests were used to assess the prevalence of breathing pattern grades in horses with and without dynamic URT obstructions. A t test was also used to investigate the differences in step max between horses with and without an URT obstruction. Data from the 6 and 8 m/s speed step were combined to form a single grade for slow canter speeds and data from 10, 11 and 12 m/s were combined to form a single grade for gallop speeds. The speed steps were combined by using the breathing pattern of the highest grade. Subsequently a binomial analysis, using Chi-squared or Fisher’s exact tests as appropriate, was performed to investigate differences between abnormal breathing patterns (i.e. grades 3–5) or 2:1 breathing patterns (i.e. grade 4 or 5) and the presence of URT obstructions. Statistical significance was set at P
Descriptive Clinical Reports
The prevalence of abnormal breathing patterns during exercise and associations with dynamic upper respiratory tract obstructions L. E. FITZHARRIS*, S. H. FRANKLIN† and K. J. ALLEN Langford House Equine Hospital, University of Bristol, UK. † School of Animal and Veterinary Sciences, University of Adelaide, South Australia, Australia. *Correspondence email: [email protected]; Received: 14.12.13; Accepted: 16.07.14
Summary Reasons for performing study: There is very limited published information on the prevalence and significance of abnormal breathing patterns adopted during canter/gallop. Objectives: The aim of this study was to report the prevalence of abnormal breathing patterns during canter/gallop and to investigate whether these may be associated with dynamic upper respiratory tract (URT) obstructions. Study design: Retrospective analysis of clinical records. Methods: Simultaneous audio and URT videoendoscopy recordings from 365 horses referred for treadmill evaluation were analysed. Results: Thirty percent of horses had an abnormal breathing pattern at canter or gallop, of which 23% had a 2:1 breathing pattern (i.e. one breath is taken over two strides). The prevalence of abnormal breathing patterns was higher at low canter speeds than during strenuous exercise. There was a significant association between breathing pattern during strenuous exercise and the presence of an URT obstruction. Conclusions: This study suggests that abnormal breathing patterns during canter and gallop may be associated with dynamic URT obstructions; therefore, horses identified by owners or veterinary surgeons to have an abnormal breathing pattern during exercise, in particular a 2:1 breathing pattern, may warrant investigations for diagnosis of respiratory tract disease. Keywords: horse; breathing pattern; exercise; upper airway
Introduction Temporal associations between respiration and locomotion have been reported in many mammals, including horses [1]. Even at racing speeds, trotters and pacers vary in the degree to which locomotion and respiration are synchronised [2,3]. In contrast, during canter and gallop the timing of the respiratory cycle is closely related to limb movements, with most horses adopting a breathing pattern of one breath every one stride [4]. This locomotor-respiratory coupling (LRC) is a breathing strategy that is thought to have both mechanical and energetic advantages [5]. It is suggested that flexion and extension of the back is the primary mechanical link between locomotion and respiration [6–9]. During canter inspiration occurs during the suspension phase of the gait, when the horse is entirely off the ground, or only supported by the hindlimbs, and expiration starts when the nonlead forelimb touches the ground [10]. There are occasions when either a transient or a sustained alteration in this 1:1 breathing pattern may occur. A transient locomotor-respiratory decoupling occurs during ‘big respiratory cycles’ (BRCs), when a single breath of a larger tidal volume is taken over 2 strides [11]. BRCs are thought to be a normal feature during exercise, and it is hypothesised that they result from a negative feedback mechanism providing a transient readjustment of the end expiratory lung volume or an improvement of pulmonary gas exchange [11,12]. It has been suggested that the frequency of BRCs during exercise is highly dependent on the individual horse but may also be inversely associated with fitness levels and respiratory frequency [11,12]. A more sustained alteration in breathing pattern has also been described whereby horses adopt a 2:1 LRC (i.e. each breath is taken over 2 strides). It has been suggested that disorders of the upper and lower respiratory tract may result in loss of the normal 1:1 breathing pattern leading to an abnormal 2:1 breathing pattern. One case report describes a 2:1 breathing pattern in a horse with a dynamic upper respiratory tract (URT) obstruction (intermittent epiglottic entrapment) [13]. This breathing pattern was also observed in research horses with experimentally induced dorsal displacement of the soft palate [14]. In addition, a 2:1 breathing pattern was observed in a research horse when inspired CO2 fraction was increased [15]. It has been reported that occasionally normal horses may not show 1:1 locomotor, respiratory coupling at slow canter speeds, but then adopt the coupling mechanism as speed increases [16]. However, Equine Veterinary Journal 47 (2015) 553–556 © 2014 EVJ Ltd
there is very limited published information on the prevalence of abnormal breathing patterns adopted during canter and gallop, particularly in clinical situations rather than under research situations. Therefore, the aim of this study was to report the prevalence of abnormal breathing patterns during exercise and to investigate whether these may be associated with URT obstructions.
Materials and methods Horses Audio and video recordings from 531 horses referred for treadmill evaluation for poor performance or abnormal respiratory noise to one centre during a 10 year period were reviewed. Only horses in which there was a complete simultaneous recording of both respiratory sounds and URT endoscopy were included in the study. In addition, only data obtained at the canter and gallop were included. Prior to treadmill endoscopy, horses were trained on the high-speed treadmill during 3 training sessions. Horses were habituated to wearing a mask during the third training session. Before testing, horses underwent an initial warm-up period, involving 20 min on a mechanical walker, followed by 2 min of walk at 1.8 m/s, 4 min of trot at 3.5 m/s and 1 min of canter at 6 or 7 m/s on the treadmill. After the warm-up, horses were removed from the treadmill and the endoscope and mask was fitted. Horses were returned to the treadmill and the exercise test and recording commenced. Racehorses performed a standardised incremental exercise test, which comprised of 1 min speed steps at each of 7, 8 and 10 m/s on a 10% incline, then an increase of 1 m/s every minute until fatigue. Nonracehorses performed variable exercise tests whereby the speed and incline were adjusted according to the discipline and level of fitness [17,18]. The step max was the highest speed step achieved.
Audio and video recordings The recording equipment comprised a close fitting facemask sealed with a neoprene shroud. The endoscope was passed into the right nostril to the level of the nasopharynx, positioned in line with the guttural pouch ostia and secured to the facemask. An omnidirectional microphone was placed within the facemask midway between the nares. The microphone was
553
L. E. Fitzharris et al.
Abnormal breathing patterns during exercise
Classification of breathing pattern All endoscopy video and audio recordings were reviewed. A breathing pattern grade was assigned to each horse for each one minute speed step. If more than one grade was observed within a speed step the highest grade was awarded. At the end of the exercise test when a speed step was only partially completed a grade was assigned for the part minute. The type of URT obstruction, or lower respiratory tract disease, was obtained from clinical records. Breathing pattern was graded according to the following definitions: Grade 1 – normal; 1:1 LRC with only 0 or 1 BRC. Grade 2 – normal; 1:1 LRC interrupted by occasional (2 or more) BRCs. Grade 3 – abnormal; baseline 1:1 LRC but rhythm disrupted by single, isolated BRCs, which were frequent and repetitive in nature. Consecutive BRCs did not occur. Grade 4 – abnormal; switching between periods of 2:1 and 1:1 breathing patterns. Grade 5 – abnormal; sustained 2:1 breathing pattern. The term ‘abnormal breathing pattern’ was used to refer to grades 3–5; the term ‘2:1 breathing pattern’ was used to refer to grades 4 and 5.
Data analyses Data analysis was performed using PASW 18.0. Chi-squared tests were used to identify whether there was any significant difference in the prevalence of abnormal breathing patterns for racehorses vs. nonracehorses. Chi-squared testing was also used to investigate potential differences between abnormal breathing patterns and sex. An independent t test was used to investigate the differences in age between groups. Fisher’s exact tests were used to assess the prevalence of breathing pattern grades in horses with and without dynamic URT obstructions. A t test was also used to investigate the differences in step max between horses with and without an URT obstruction. Data from the 6 and 8 m/s speed step were combined to form a single grade for slow canter speeds and data from 10, 11 and 12 m/s were combined to form a single grade for gallop speeds. The speed steps were combined by using the breathing pattern of the highest grade. Subsequently a binomial analysis, using Chi-squared or Fisher’s exact tests as appropriate, was performed to investigate differences between abnormal breathing patterns (i.e. grades 3–5) or 2:1 breathing patterns (i.e. grade 4 or 5) and the presence of URT obstructions. Statistical significance was set at P
