This work was performed at the Purdue University School of Veterinary Medicine, West Lafayette, IN. Previously presented in abstract form at the Veterinary Comparative Respiratory Society annual symposium, October 2009 by Kristina Goncarovs.
Corresponding author: Laurent Couetil, DVM, PhD, Purdue University School of Veterinary Medicine, 625 Harrison Street, West Lafayette, IN 47907; e-mail: email@example.com.
Background: One proposed nonmedical therapy for recurrent airway obstruction (RAO) in horses is a handheld acoustic device that propels sound waves from the nose down the tracheobronchial tree where it is intended to dislodge mucous and relax bronchospasm, permitting clearance of mucoid secretions.
Objective: To evaluate the effectiveness and safety of this device when used as per the manufacturer's recommendations as a treatment for RAO.
Animals: Nine adult horses previously diagnosed with RAO.
Methods: Prospective, cross-over clinical trial. Horses were exposed to a dusty environment until airway obstruction developed as defined by standard lung mechanics (SLM). Horses were randomly assigned to receive either acoustic therapy or a sham treatment for 4 weeks while being maintained in this environment. Horses were evaluated by clinical scores, SLM, and forced expiration regularly for 4 weeks. The opposite treatment was administered after a washout period.
Results: Seven horses received the treatment; 9 received the sham. There were no changes (P > .05) in clinical score, maximal change in transpulmonary pressure (ΔPLmax), lung resistance (RL), or the forced expiratory flow rate averaged over the last 75–95% of expiration (FEF75−95%) over the study period. The device was determined to be safe, although several minor adverse effects were noted, including head tossing, coughing, and chewing during treatment.
Conclusions and Clinical Importance: Treatment with this device did not improve clinical signs or lung function in horses with RAO kept in a dusty environment. Currently accepted treatments, including environmental management and medical therapy, should be recommended.
forced expiratory flow rate averaged over the last 75–95% of expiration
forced expiratory flow rate at 95% of expiration
forced expiratory volume at 1.5 seconds
forced expiratory volume at 1.5 seconds/forced vital capacity ratio
recurrent airway obstruction
total lung resistance
standard lung mechanics
Recurrent airway obstruction (RAO) is one of the most common chronic respiratory diseases of adult horses worldwide. Standard management practices, including stabling and feeding dusty hay, are often associated with exacerbation of disease.1 Clinical signs of RAO include exercise intolerance and an increased respiratory effort, with a prominent heave line in chronically affected horses. Many horses will also present with variable degrees of tracheal exudate, cough, or serous nasal discharge.
The etiology of RAO has been determined to be an allergy to inhaled dusts or molds with a synergistic effect of endotoxins.2,3 Pathophysiologic responses occurring within the lower respiratory tract include marked bronchospasm, severe airway inflammation, and excessive production of mucoid secretions that result in lower airway obstruction and hyperresponsiveness. Treatment of RAO is best achieved through the combination of environmental and medical management. Horses with RAO are ideally housed outside on pasture, away from dusts associated with hay and bedding, and fed pelleted feed or grass when available.4 This dramatic lifestyle change is not feasible for many owners, and medical management of RAO with inhaled or systemic bronchodilators and corticosteroids, which is efficacious, is used instead of management changes.4 However, these pharmaceuticals are not without adverse effects, and treatment can be expensive.
Several pulmonary diseases in humans are characterized by excessive and abnormal mucoid secretions including cystic fibrosis, primary ciliary dyskinesia, chronic bronchitis, and chronic obstructive pulmonary disease.5 Airway clearance techniques, formerly referred to as chest physiotherapy, are widely used as part of a multimodal treatment protocol for these patients.6 One such technique combines oscillating positive expiratory pressure with oscillations of airflow through a fluttering valve at the mouth.a Beneficial physiologic effects of this device include shearing of mucous from the airway wall, stabilization of airways preventing early closure, and changes in mucous quantity and consistency.7
Recently, a handheld acoustic deviceb was introduced to the veterinary market as a potential treatment for RAO specifically focusing on mucoid secretion clearance. The proposed mechanism of action is similar to the fluttering valve used in human medicine in that sound waves of gradually increasing frequencies are propelled from the nose via a handheld device down the tracheobronchial tree with the intent to vibrate the distal airways and facilitate mucous clearance. Sound waves are generated by a battery-powered speaker and electronic circuitry controls sound frequency, intensity, and modes of operation. The manufacturer also claims that vibrations relieve bronchospasm and airway inflammation.
There is some anecdotal evidence supporting the efficacy of this product as a treatment for RAO. As it is user-friendly and nonpharmaceutical, it is an appealing alternative for horse owners. However, there are no scientific data available in regard to the efficacy or safety of this device. Therefore, the objectives of this study were (1) to evaluate the efficacy of this acoustic device as a treatment for RAO in horses and (2) to evaluate the safety of this device when used as per the manufacturer's recommended protocol.
Materials and Methods
This study was designed as a randomized, sham-controlled, cross-over clinical trial. Adult horses, part of a university research herd, that had been previously diagnosed with RAO were evaluated. The study was carried out in 2 phases as shown in Figure 1. Approval for this study was granted by the Purdue University Animal Care and Use Committee.
All horses had been previously diagnosed with RAO based on history, physical examination, standard lung mechanics (SLM), response to bronchodilators, and bronchoalveolar lavage cytology according to recognized criteria.8 Remission from airway obstruction was confirmed before commencement of either study phase. Disease status was evaluated by SLM and RAO remission was defined as a maximum change in transpulmonary pressure (ΔPLmax) <10 cmH2O. Airway obstruction was then induced by means of housing RAO-susceptible horses in individual stalls bedded with straw and feeding a combination of good quality and moldy hay. Horses were maintained in this environment for the duration of the treatment phase. Airway obstruction was confirmed with SLM, and horses were enrolled in the treatment trial when ΔPLmax > 15 cmH2O. The day that horses reached the critical degree of airway obstruction was considered day 0 of the phase.
In phase 1, horses were randomly assigned to receive either treatment with the acoustic device or the sham treatment. In phase 2, they received the opposite treatment. Treatment with the acoustic device was administered at frequencies recommended by the manufacturer as shown in Figure 2. The sham treatment was administered following the same protocol; however, the sound emitting device was hung on the stall door, while a silenced device was held to the horse's nose.
A washout period of at least 2 weeks was allotted between phases. Remission was again confirmed with SLM before commencement of phase 2. Horses were removed from the study if remission could not be confirmed after 4 weeks on pasture. They were subsequently treated with bronchodilators and corticosteroids. Additionally, horses that became anorectic during the study period for >24 hours were removed from the study and treated as above.
Acoustic Device Therapy
The acoustic device used during this study was a handheld, battery-powered device weighing less than 2 pounds (Fig 3).b At the outermost part of the device, a speaker emitted soundwaves of various frequencies. A plastic cone was separated from the speaker by a celephane diaphragm; this cone was applied to the nose of the horse during treatments as shown in Figure 4. Soundwaves were emitted at low, high, or massage frequencies as shown in Figure 2.
All horses were evaluated using 2 clinical scoring systems starting on day 0 and then 3 times per week by an investigator unaware of treatment allotment. Horses were first scored on a scale of 0–21 as described by Tesarowski et al (long score), evaluating respiratory rate (RR), nasal discharge, respiratory effort, tracheal sounds, lung sounds, and cough.9 They were subsequently scored on a scale of 2–8. (short score), evaluating nostril flaring and abdominal effort.10 In both scoring systems, higher scores indicate more severe clinical signs. Outcome variables analyzed included the long and short scores and the RR.
SLM, specifically evaluating ΔPLmax, total lung resistance (RL), dynamic compliance (Cdyn), and frequency of respiration (f), were performed on day 0 and then once per week.
Forced expiration maneuvers were performed on day 0 and then once every other week. Outcome variables analyzed included the forced expiratory volume at 1.5 seconds (FEV1.5), the FEV1.5/Forced Vital Capacity ratio (FEV1.5/FVC), the forced expiratory flow rate at 95% of expiration (FEF95%), and the forced expiratory flow rate averaged over the last 75–95% of expiration (FEF75−95%).
CBCs were performed on days 0 and 28 of the study period. Horses were examined daily, noting specifically appetite, urination, defecation, presence or absence of cough, and respiratory effort. Further evaluation of the respiratory tract was performed as described above on days in which horses were given clinical scores. Adverse events during administration of treatments were noted (eg, animal cooperation, head shaking, coughing, and nasal discharge).
Data sets from phases 1 and 2 were first considered as unpaired measurements, thus permitting all data collected to be included for each horse, even if they did not trigger in 1 arm of the study. Data from each phase were additionally analyzed as paired measurements with data excluded for the 2 horses that did not complete both phases of the study. Comparisons of outcome variables between treatment groups were done on day 0 with the use of a t-test.c A repeated measures analysis of variance was performed to evaluate the effect of each treatment on outcome variables over time. Posthoc analysis with a Tukey's test was carried out when appropriate. Data was expressed as mean ± SD. Significance was set at P < .05.
Nine horses were included in this study, with a mean age of 24 (range 17–29) years. Six horses were mares and 3 were geldings. Horses were Quarter Horses or Quarter Horse-type breeds (7), 1 Icelandic pony, and 1 Arabian cross. From these horses, 16 data sets were collected; 2 horses failed to develop sufficient airway obstruction in the 2nd phase, both after having received the sham treatment in phase 1. In total, 7 horses received the acoustic device treatment, while 9 received the sham treatment. Two horses were removed from the study because of exacerbation of airway obstruction, demonstrated by anorexia and severe clinical signs; both horses were in the acoustic device treatment arm of the study (days 14 and 27).
There was no statistically significant difference between treatment groups in the total score for either clinical scoring system or the RR on day 0 (Table 1).
Table 1. Baseline data.
Treatment (mean ± SD)
RR, respiratory rate; ΔPLmax, maximal change in transpulmonary pressure; RL, total lung resistance; Cdyn, dynamic compliance; F, frequency of respiration; FEV1.5, forced expiratory volume at 1.5 seconds; FEV1.5/FVC, ratio of FEV1.5 to forced vital capacity; FEF95%, forced expiratory flow at 95% of expiration; FEF75−95%, forced expiratory flow averaged over the last 75–95% of expiration.
Clinical Score (Rush)
6 ± 1
7 ± 1
Clinical Score (Tesarowski)
11 ± 3
10 ± 3
24 ± 45
26 ± 10
30.1 ± 8.4
31.9 ± 5.9
1.95 ± 0.49
1.96 ± 0.47
0.63 ± 0.23
0.67 ± 0.28
18.27 ± 6.28
19.50 ± 4.45
32.4 ± 9.6
30.9 ± 11.6
0.86 ± 0.06
0.83 ± 0.06
9.74 ± 6.03
7.37 ± 4.39
3.29 ± 1.20
3.32 ± 1.40
There was no statistically significant effect of treatment over time in the long score (P= .59), the short score (P= .97), or the RR (P= .62). The acoustically treated group had a significantly higher (P= .02) RR on day 25 as compared with day 2 (Fig 5). Statistical analysis was repeated with data excluded for the 2 horses that did not trigger for their treatment arm (thereby evaluating the sham group with 7 horses) and no statistical differences were noted between groups in any outcome variable measured.
There was no statistically significant difference between treatment groups in ΔPLmax, RL, Cdyn, or f on day 0 (Table 1). There was no statistically significant effect of treatment on ΔPLmax (P= .49), RL (P= .99), Cdyn (P= .78), or f (P= .69) at any of the time points. Although it was not statistically significant, the acoustically treated group had higher mean ΔPLmax values as compared with the sham group throughout the duration of the study by an average of approximately 35% (Fig 6). Statistical analysis was repeated with sham data excluded for the 2 horses that did not trigger for their treatment arm and no significant differences were noted between groups in any of the outcome variables measured.
There was no statistically significant difference between treatment groups in FEV1.5, FEV1.5/FVC, FEF95%, and FEF75−95% on day 0 (Table 1). There was no statistically significant effect of treatment on FEV1.5 (P= .81), FEV1.5/FVC (P= .20), FEF95% (P= .68), or FEF75−95% (P= .61; Fig 7) at any of the time points. Statistical analysis was repeated with sham data excluded for the 2 horses that did not trigger for their treatment arm and no significant differences were noted between groups in any of the outcome variables measured.
Adverse effects noted during the study in the acoustically treated animals included immature neutrophilia (n = 1, on day 28), mature neutrophilia (n = 1, on day 28), head tossing during treatment (n = 5), coughing during treatment (n = 5), chewing and snorting during treatment (n = 1). Head tossing (n = 5), coughing (n = 5), chewing (n = 1), and snorting (n = 1) were also noted in the same horses during the sham phase. All of these behaviors varied in frequency of occurrence, with head tossing and coughing the most consistently demonstrated. One horse developed a severe pharyngitis and laryngitis shortly after being removed from the treatment arm of the study because of an exacerbation of RAO. During treatment for this exacerbation and a suspected large colon impaction, repeated nasogastric intubations were performed on this horse to administer oral fluids. The horse was reluctant to swallow the tube during each treatment. Subsequently, anorexia and respiratory difficulty worsened and a stridor became evident. Severe pharyngitis and laryngitis were noted upon endoscopy that required tracheostomy as well as systemic antimicrobial and anti-inflammatory therapy. We considered it likely that upper airway inflammation developed secondary to trauma because of difficult and repeated intubation.
The results of this study show that based on clinical evaluation, SLM, and forced expiration, the acoustic device tested is not an effective treatment for RAO in horses maintained in a dusty environment. Notably, 2 horses exhibited worsening of disease during their treatment phase after having completed the sham phases of the study without complication. The device appears to be relatively safe, with adverse effects being behavioral and generally harmless (head shaking, coughing, snorting, chewing during treatment) and commonly encountered in the sham and treatment groups. These manifestations were attributed to restraint and application of the nose piece to the horses. However, these behaviors could well interfere with appropriate treatment, as holding of the cone onto the nose could become difficult in these situations. One horse developed severe pharyngitis and laryngitis; however, it is believed that this complication was secondary to repeated nasogastric intubation shortly after removal from the study, and not directly related to use of the device. Systematic endoscopy of the upper respiratory tract of horses before and after completion of each treatment phase would have been helpful to definitively address this issue.
Based on the use of manual chest percussion and oscillating pressure-based techniques in human medicine, specifically in the management of cystic fibrosis, the mechanism of action of this device proposed by the manufacturer appeared promising. It is recognized that excessive and thickened mucoid secretions block lower airways and impair airflow and ventilation. Bronchospasm, another important clinical manifestation of RAO, further interferes with removal of these mucoid plugs. In human medicine, a handheld fluttering valve device is available that, when held to the mouth during respiration, uses a patient's normal inspiratory and expiratory patterns to transmit oscillating movements to the bronchial tree.b This device was initially shown to be as effective at airway clearance as compared with the gold standard of manual chest percussion,7 and can be easily performed by patients independently in their own homes. In chronic bronchitis, this fluttering valve device was shown to be more effective than manual chest percussion and postural drainage at removing airway secretions as determined by total sputum wet weight.11 It is interesting to note, however, that the only study evaluating the long-term use (1 year) of human flutter valve actually indicated that patients had greater deterioration in pulmonary function and a higher number of hospital visits when compared with other devices.12 In the present study, mucous clearance was not measured by sputum production as in these human studies; however, because neither clinical signs nor lung function testing improved with treatment, it is presumed that mucous clearance was also not affected positively.
The total length of airways in a horse as compared with length of airways of a human is dramatically different. It is possible that the primary reason the acoustic device failed to show efficacy was because of dampening of sound waves through the tracheobronchial tree, insufficient sound energy delivered to mobilize airway secretions, or that sound frequencies or amplitude used were not appropriate for equine airways.
The conflicting results obtained with acoustic devices in human medicine in combination with this study in horses lends credence to continued use of proven medical therapies for RAO in horses with bronchodilators and corticosteroids. Medical therapy in conjunction with environmental management is the most effective treatment of this chronic disease.4 The environment in this study was challenging for RAO horses because they were housed indoors in a dusty barn and fed moldy hay. Therefore, it would be interesting to evaluate if the use of this device in horses with exacerbated disease after the environment is improved (outside on pasture, fed pelleted feeds) is beneficial, compared with environmental changes alone. If this device were to show clinical efficacy in this context, it would be prudent to repeat a safety analysis with a larger study population.
a FLUTTER, Axcan Scandipharm Inc, Birmingham, AL
b VibraVM, Vibra Lung Inc, Denver, CO
c Statistica, Statsoft, Tulsa, OK
The study was funded by the state of Indiana and a Purdue University School of Veterinary Medicine research account, funded by the total wagers tax and VibraLung Inc.
We thank Donna Griffey, RVT, for her technical assistance and George Moore, DVM, MS, PhD, for his statistical consultation.