Triggers of breathlessness in inducible laryngeal obstruction and asthma

Inducible laryngeal obstruction (ILO) is often misdiagnosed as, or may coexist with, asthma. Identifying differences in triggering factors may assist clinicians to differentiate between the two conditions and could give mechanistic insights.


| INTRODUC TI ON
Inducible laryngeal obstruction (ILO) is an umbrella term for a group of conditions associated with acute breathlessness caused by episodic airflow obstructions arising in the larynx, most commonly manifesting as paradoxical adduction of glottic and/or supraglottic folds during inspiration, often referred to as vocal cord dysfunction. 1 This paradoxical adduction of glottic and/or supraglottic folds results in marked reduction in airflow through the larynx, producing distressing symptoms of dyspnoea, stridor, throat tightness and globus pharyngeus. [2][3][4] A recent systematic review highlighted ILO as a comorbidity in quarter of asthma cases where ILO was identified by visualization of laryngeal movement, with the value increasing to over one third when the ILO diagnosis included a provocation stimulus. 5 Further, in two prospective studies, of difficult-to-treat asthma and where ILO was suspected in asthmatic patients, up to 50% of patients were identified as suffering concomitant ILO and asthma. 6,7 Both asthma and ILO may present with dyspnoea and wheezing upon exposure to certain triggers such as physical exertion or inhaled irritants, making differentiation between the two conditions a clinical challenge. 7 Due to the similarities in presentation, many people with underlying ILO are misdiagnosed with refractory asthma. A retrospective study estimated that it took an average of 4.8 years for a patient with misdiagnosed asthma to be correctly diagnosed with ILO. 8 Treatment for asthma in misdiagnosed patients with underlying ILO is of little benefit, and patients are therefore unnecessarily suffering significant morbidity and are subject to high doses of potentially toxic treatments such as systemic steroids. 7,9 Aside from patient morbidity the economic impact of ILO misdiagnosis and untargeted treatment is likely high, 10,11 although robust cost-analysis data are lacking. It is therefore vital that clinicians are able to competently distinguish the two conditions. This study aimed to identify triggers for both ILO and asthma and compare the prevalence of each trigger between conditions, the outcome of which could assist clinicians in the diagnosis of ILO. In addition, the identification of disease-specific triggers may lead to novel hypotheses related to upper and lower airway hyperresponsiveness.

| ME THODS
Ethical approval was obtained from the Camden and Kings Cross Ethics Committee (16/LO/0911), and participants provided written informed consent.

| Recruiting site
Participants were recruited from a specialist airway clinic at the Royal Preston Hospital. The physicians in the service were respiratory specialists with a sub-specialty interest in asthma. This is a tertiary level service, where patients (over the age of 16 years old) with breathlessness are referred from GP surgeries or hospital consultants. This multidisciplinary service provides a variety of physiological tests and treatments, including laryngeal provocation, for the diagnosis of ILO, and bronchial provocation and bronchodilator reversibility tests, for the diagnosis of asthma. The standardized diagnostic workup for ILO is presented in Table 1.
The diagnosis of asthma in the service would typically be based on previous pulmonary function tests and expert clinical assessment, with further testing only occurring in case of diagnostic difficulty.
Comorbidities were self-reported and/or retrieved from patient's medical notes.

| Retrospective case note review: Phase one
Data from consecutive patients with laryngoscopy-confirmed ILO between January 2015 and May 2016 attending the specialist airway clinic were included in the study. A retrospective case note review of patient's referral letters and medical notes was performed.
Patient's demographics, comorbidities and triggers of breathlessness were identified and recorded. Triggers were then categorized under seven domains: scent, environment, mechanical, daily activities, emotions, temperature and others and used to inform the Breathlessness Trigger Survey.

| Participants
The prospective study was conducted over a short window (May to June 2016) in order to minimize any seasonal variability. Participants with endoscopically proven ILO and/or asthma were recruited from the specialist airways clinic. A participant information leaflet regarding the study was given to each eligible patient when they arrived at the clinic for their outpatient appointment. Study exclusion included (a) no formal diagnosis of ILO or asthma and (b) other conditions that may cause breathlessness. Participants with both ILO and asthma were eligible to take part in the study provided that they were able to differentiate between triggers of breathlessness for each condition.

| Questionnaires
The Breathlessness Triggers Survey is a 23-item survey made specifically for this project and based on the findings from the first part of the study. It evaluates the likelihood of each item being a trigger for breathlessness in either ILO or asthma. Patients respond to each item using a five-point Likert scale (1 = strongly agree, 2 = agree, 3 = neutral, 4 = disagree, 5 = strongly disagree). If a single participant had both asthma and ILO, the participant was asked to complete this survey twice (one for each condition). The design of the Breathlessness Triggers Survey is described in the phase one results section below.
To characterize the study population, participants were asked to complete the St. George's Respiratory Questionnaire (SRGQ) 12

| Statistical analysis
The responses for each question in the Breathlessness Triggers Survey were grouped into strongly agreed/agreed, neutral and disagreed/strongly disagreed. The item was considered a trigger if the participant indicated either strongly agreed or agreed.
A chi-square test was performed for each trigger to compare the percentages of those who agreed/strongly agreed between the two conditions. If the chi-square test indicates more than 20% have an expected count of less than five, the Fisher's exact was used to compare the difference between the two conditions. All statistical analyses were performed using SPSS v26.0 (IBM Corp, Chicago, Ill).

| Participants
Data from 202 patients with ILO (73% female, mean [SD] age 53.1 [15.7] years) were included in the retrospective study ( Table 2). The most common comorbidities were gastro-oesophageal reflux disease (GORD), asthma and mental health conditions such as depression and general anxiety disorder.

| Number of triggers for breathlessness
The number of triggers indicated in patients' letters was recorded for each patient; the median (interquartile range) number of triggers per patient was 3 (3), with a maximum of 8. There was no mention of any triggers in the patients' notes for 18 patients.

| Types of triggers
Reported triggers for ILO are presented in Table 3. Of note, the most

| Survey design
The data collated from the retrospective study were used to design a survey to evaluate the potential triggers of breathlessness for phase two of the study. All the triggers listed in Table 3

| Participants
A total of 89 participants were recruited for this study: 38 participants with ILO only, 39 participants with asthma only and 12 participants with both ILO and asthma. Table 4 shows the demographics and relevant comorbidities of the participants for the prospective study.

| Questionnaires
A worse quality of life total score was reported by participants with both asthma and ILO compared to participants who had a diagnosis of only one of the conditions (Table 4, P < .05). However, no such between-group differences were noted on the MRC dyspnoea scale (Table 4). Likewise, no difference was noted between participants with both ILO and asthma and individuals with only one condition using the VCDQ and the ACQ (Table 4).

Number of triggers
The number of triggers indicated as agree/strongly agree in the Breathlessness Triggers Survey was summed for each participant.

Others 17
Not indicated in letters 9 Patient unable to identify any triggers 8

Types of triggers
The proportions of participants in each cohort reporting each trigger of breathlessness are presented in Figure 1A-F and summarized below.

Smells
The majority of participants with ILO or asthma agreed that perfumes/ deodorants (76% and 61% respectively) and cleaning sprays (74% and 63%) triggered their breathlessness with no difference in frequency between conditions (P > .05). Although the majority of participants in both cohorts disagreed that vinegar would trigger their breathlessness, it was more likely to trigger breathlessness in patients with ILO compared to asthma (22% versus 6%, P = .019) ( Figure 1A).

Environmental factors
Smoke/fumes affected more than 80% of both cohorts without a between-group difference in frequency (P > .05), whereas damp and pollen/flowers were more problematic for the asthma group compared to those with ILO (71% versus 46%, P = .012 and 78% versus 52%, P = .005, respectively) ( Figure 1B).

Daily activities
The majority of participants with ILO (72%) and asthma (78%) agreed that exercise was a trigger for their breathlessness; however, there were no between-cohort differences in frequency (P = .454, Figure 1D).

Emotions
A large proportion of participants with ILO (86.0%) and asthma (73%) agreed that stress was a trigger of their breathlessness (P = .096) ( Figure 1E).

Temperature
The majority of participants in both cohorts agreed that the extremes of temperatures could trigger their symptoms, especially cold air, with no difference between the two groups ( Figure 1F).

| D ISCUSS I ON
To our knowledge, this is the first study to investigate self-reported triggers of breathlessness in both ILO and asthma. were more prevalent in asthma and one "scent" (vinegar) was more prevalent in ILO. Exercise was a trigger for the majority of patients  The most evident finding of this study was the significantly higher prevalence of mechanical factors such as talking, shouting and swallowing in triggering symptoms of breathlessness in ILO participants compared to patients with asthma ( Figure 1C). The high occurrence of mechanical triggers in ILO is in keeping with our understanding of the pathophysiology of the condition. 16 Indeed, neural hypersensitivity and/or mechanical insufficiency are proposed as key underlying mechanisms in ILO. 4 Laryngeal hypersensitivity is thought to be a consequence of underlying inflammation which could be caused by associated comorbidities such as GORD or post-nasal drip, or extrinsic irritants such as dust particles. 17,18 Chronic inflammation may modify neural pathways, altering the perceptions and response of higher centres to afferent signals, leaving the neural pathways in a perpetually hyper-excitable state. In the context of hyper-excitability, the mechanical movements in the laryngeal region during speech and/or swallowing may be sufficient to trigger local reflexes which produce paradoxical adduction of vocal folds during inspiration. This may be particularly apparent in patients with mechanical insufficiencies, for example malfunctioning laryngeal abductor muscles or laxity of ligaments or laryngeal cartilage. 4 As asthma involves bronchial hyper-reactivity and not extra-thoracic/laryngeal hyper-reactivity, mechanical factors would be less likely to trigger symptoms of breathlessness in asthmatic patients.
In contrast, environmental triggers of breathlessness such as pollen/flowers and damp weather were more prevalent in patients with asthma compared with ILO ( Figure 1B). The role of pollen/ flowers in triggering atopic asthma is well-established in literature.
The role of damp weather in triggering asthma symptoms is less certain, but could be mediated through increased ozone, airborne and (d) unlike EIB, inhaled beta-2 agonists are usually ineffective in for the treatment of EILO. 21 In addition, our patients were older than those typically found to have EILO 22 and a study targeting this younger age group may reach different conclusions. Further, exercise-induced breathlessness can be a symptom in asthma and ILO, but causes by factors not directly linked to laryngeal or bronchial airflow obstruction, such as obesity and/or deconditioning. We would therefore propose that primary care physicians explore exercise as a trigger for breathlessness in more detail to assist in the differentiation ILO and asthma. Where inhaled therapy is largely ineffective, patients should be referred for assessment of ILO as a possible cause of exertional dyspnoea.
The ILO cohort in this study showed a greater percentage of participants with comorbidities such as GORD and mental health conditions (depression, anxiety). It is proposed that the acidic contents from GORD damage the laryngeal mucosa. The prolonged inflammatory insults may increase laryngeal sensitivity resulting in hypersensitivity of the laryngeal region. Thus, GORD may be a contributing factor in the development of ILO as well as a trigger for breathlessness. The association of ILO with mental health conditions such as depression, anxiety and stress has been extensively discussed in literature. [23][24][25][26] A strength of this study lies in the robust diagnostic workup for patients with ILO (Table 1) Triggers Survey that was used in this study is not a validated questionnaire, and, due to the nature of the study design, the list of potential triggers should not be considered exhaustive. To rectify this short-coming, a section at the end of the survey was present to allow participants to fill in other triggers which were not listed.
In conclusion, this study has taken the first step to identify triggers of ILO that could help differentiate the condition from asthma. Mechanical factors (ie talking, shouting and swallowing)