Upper airway cough syndrome (UACS; post-nasal drip syndrome)
The American College of Chest Physicians 2006 guidelines adopted the term upper airway cough syndrome or UACS to replace the term post-nasal drip syndrome, as it was felt that the new terminology more effectively addressed the possibility that cough could be due, not only to the post-nasal drip itself, but to irritation or inflammation of upper airway structures that directly stimulate cough receptors independent of, or in addition to, post-nasal drip (Irwin et al., 2006). More recently, the concept of a ‘unified airway’ has been proposed based on epidemiological and physiological data suggesting that the upper and lower airways may be linked through local inflammatory processes. The model suggests that systemic propagation of inflammation occurs through trafficking of inflammatory mediators, thus promoting a system-wide response in the respiratory mucosa through which pathology in one portion of this system can stimulate and influence pathophysiological changes at a site distal to the initial site of inflammation (Krouse and Altman, 2010). Consistent with this hypothesis, recent studies incorporating induced cough in humans have demonstrated enhancement of cough reflex sensitivity in the presence of rhinosinusitis (Tatar et al., 2009).
UACS can result from a multiplicity of rhinosinus conditions (Table 2), and has been considered the most common cause of chronic cough in adults in the United States (Pratter, 2006a). Treatment for UACS-induced cough should be targeted at the underlying aetiology, and is often effective. However, when UACS is due to acute viral upper respiratory tract infection, or, is being treated empirically as the first step in a diagnostic-therapeutic algorithm for chronic cough of unclear aetiology, the combination of a first-generation antihistamine and decongestant is recommended (Irwin et al., 2006). A complicating factor in interpreting a response to this form of therapy, and thus assigning a diagnosis to the patient, is the possibility that symptomatic relief is achieved not through amelioration of post-nasal drip, but by a primary anti-tussive effect of the antihistamine. Indeed, diphenhydramine has been demonstrated to inhibit citric acid-induced cough in healthy volunteers (Packman et al., 1991), while more recently, dexbrompheniramine has been shown to inhibit activation of the transient receptor potential vanilloid-1 (TRPV1) ion channel in human TRPV1-expressing human embryonic kidney (HEK) cells and rat dorsal root ganglia neuron preparations (Sadofsky et al., 2008). TRPV1 has received significant attention recently as a particularly relevant receptor in human cough, and thus represents a compelling target for the development of potential new anti-tussive agents (Adcock, 2009).
Table 2. Potential causes of upper airway cough syndrome*
|Perennial non-allergic rhinitis|
| Vasomotor rhinitis|
| Nonallergic rhinitis with eosinophilia (NARES)|
| Following upper respiratory tract infection|
|Allergic fungal sinusitis|
|Rhinitis due to anatomic abnormalities|
|Rhinitis due to physical or chemical irritants|
|Rhinitis of pregnancy|
Eosinophilic airway inflammation (asthma and non-asthmatic eosinophilic bronchitis)
Multiple prospective studies have shown that asthma is among the most common aetiologies of chronic cough (24–29%) in adult non-smokers (Irwin et al., 1990; Pratter et al., 1993; McGarvey et al., 1998). Usually, cough is accompanied by the typical features of dyspnoea and wheezing, but may represent the sole or predominant symptom in the setting of an unremarkable physical examination and normal pulmonary function studies. Such a presentation describes a subgroup of asthmatics termed to have cough-variant asthma, or CVA (Dicpinigaitis, 2006a).
Asthmatic cough typically responds to standard asthma therapy of inhaled bronchodilators and inhaled corticosteroids. The leukotriene receptor antagonists (LTRA) zafirlukast (Dicpinigaitis et al., 2002) and montelukast (Spector and Tan, 2004; Kita et al., 2010) have been shown to be particularly effective in CVA, and should be added to a regimen of inhaled corticosteroids before escalation to therapy with systemic corticosteroids (Dicpinigaitis, 2006a). Given a paucity of data on the natural history of CVA, it remains unclear whether monotherapy with LTRAs is sufficient to prevent the sequelae of chronic airway inflammation as are seen in the typical form of asthma.
Non-asthmatic eosinophilic bronchitis distinguishes itself from asthma by the absence of reversible airway obstruction and bronchial hyperresponsiveness, but the associated cough, as in asthma, is usually quite responsive to inhaled corticosteroid therapy (Brightling, 2010). Clinical response has been correlated with decreases in sputum eosinophil count and cough reflex sensitivity, thus supporting a causal relationship between cough and eosinophilic airway inflammation (Brightling, 2010). Only rarely is systemic treatment with oral corticosteroids required. The efficacy of LTRAs in chronic cough due to non-asthmatic eosinophilic bronchitis has not been investigated. Furthermore, the natural history of non-asthmatic eosinophilic bronchitis is yet to be elucidated fully. Preliminary long-term (>1 year) follow-up data from 32 patients suggest that although a minority (16%) of patients develop fixed airflow obstruction, the condition is rarely self-limiting; 66% of patients had persistent symptoms and/or ongoing airway inflammation (Berry et al., 2005).
Recently, exhaled nitric oxide has been proposed as a useful biomarker to confirm the presence of eosinophilic airway inflammation and thus, predict a favourable response to inhaled corticosteroid therapy in both asthmatic cough as well as cough due to non-asthmatic eosinophilic bronchitis (Lim, 2010). Absence of a response to systemic corticosteroids should prompt evaluation for other aetiologies of chronic cough.
Gastro-oesophageal reflux disease (GERD)
Perhaps the greatest challenges to the clinician evaluating a patient with chronic cough are the confirmation of the presence of gastro-oesophageal reflux and, more importantly, the determination that reflux, if present, is responsible for the cough (Table 3). Indeed, the diagnosis of GERD, including the use of empiric therapeutic trials with proton pump inhibitor medications (PPIs), remains problematic, even in the setting of typical reflux symptoms (Lacy et al., 2010). The response rate to PPI therapy in patients in whom cough is the sole or predominant symptom is even less (Chang et al., 2006; Pauwels et al., 2009).
Table 3. Challenges in the management of cough due to gastro-oesophageal refluxdisease (GERD)
| Poor sensitivity and specificity of empiric therapeutic trials with proton pump inhibitor medications (PPIs)|
| Standard diagnostic studies for acid reflux, such as 24-hour oesophageal pH measurement, may fall within normal range of number of reflux events, even when reflux is the cause of chronic cough.|
| Aggressive acid suppression therapy may be inadequate due to the presence of non-acid or weakly-acid refluxate as the cause of cough.|
| When aggressive acid suppression therapy fails, and reflux remains the suspected aetiology of chronic cough, addition of a prokinetic agent is appropriate. However, the available therapeutic options are severely limited.|
| In the setting of failure of maximal medical therapy, surgical intervention (laparoscopic Nissen fundoplication) may be considered. However, data on the outcome of surgical intervention for chronic cough due to GERD are limited, and do not demonstrate a uniformly positive outcome.|
Potential mechanisms proposed to explain GERD-induced cough include: (i) refluxate within the oesophagus triggering a distal oesophageal-tracheobronchial reflex; (ii) refluxate extending beyond the oesophagus into the larynx (laryngopharyngeal reflux; LPR) where sensory afferent cough receptors reside; and (iii) microaspiration (Figure 1). Furthermore, it has recently been appreciated that cough may be induced by non-acidic or weakly acidic refluxate, thus explaining why some patients whose cough was refractory to aggressive acid suppression experienced resolution of cough subsequent to anti-reflux surgery (laparoscopic Nissen fundoplication) (Irwin et al., 2002).
Recently, the diagnostic evaluation of the patient with unexplained cough has been improved by the increased availability of combined oesophageal pH-impedance monitoring that allows detection of both acid (pH > 4) and non-acid reflux events (Tokayer, 2008). Initial studies demonstrate that this modality significantly enhances the diagnostic yield for detecting gastro-oesophageal reflux in patients with atypical GERD symptoms, and has been proposed as the new diagnostic gold standard (Bajbouj et al., 2007; Kahrilas, 2010; Lee et al., 2010). Interestingly, multiple studies incorporating this technology to evaluate chronic cough sufferers have not demonstrated an increased number of reflux events compared with data from healthy volunteers, but proposed a causal relationship based on observed temporal associations between reflux events and cough (Smith et al., 2010a). Though most recent studies incorporating these newest diagnostic methods have demonstrated intra-oesophageal reflux as being most relevant (Smith et al., 2010a), other investigators have suggested a prominent role for laryngopharyngeal reflux in patients with chronic cough (Patterson et al., 2009).
A recent study combining simultaneous acoustic cough recording with pH-impedance monitoring, in addition to cough reflex sensitivity measurement, in a group of unselected patients with chronic cough has shed further light on the association of reflux and cough (Smith et al., 2010b). Subjects in this study demonstrated a symptom association probability (SAP) of cough preceded by reflux, as well as reflux preceded by cough. This observation supports the concept that not only may reflux induce cough, but cough may cause or exacerbate reflux, through mechanisms including increased intra-thoracic pressure, thus generating a self-perpetuating process of cough stimulation. Notably, SAP-positive subjects did not have a greater degree of oesophageal acid exposure or degree of erosive disease compared with SAP-negative subjects, but they did have enhanced cough reflex sensitivity, thus suggesting a mechanism of sensitization at the level of sensory afferents in the oesophagus and/or central sensitization at the level of the brainstem.
For patients with chronic cough in whom GERD is confirmed or suspected, acid suppressive therapy is indicated, and can be escalated to an aggressive regimen of a twice-daily PPI, taken one-half hour before breakfast and dinner, with or without a histamine-2 antagonist (i.e. ranitidine 150 mg) at bedtime. Recently, liquid alginate suspensions have been evaluated as alternate or complementary therapy for symptoms of LPR. The alginate is believed to form a physical barrier floating above the gastric contents, thus inhibiting backflow of refluxate into the oesophagus (McGlashan et al., 2009). In addition to pharmacological therapy, anti-reflux lifestyle measures should be stressed: sleep with head elevated, do not eat within 2 h of bedtime, and avoid reflux-promoting foods and beverages including alcohol, caffeine, chocolate, peppermint, and spicy or greasy foods (Irwin, 2006).
If cough persists despite strict adherence to the above-described regimen for 2–3 months, and reflux remains the presumed aetiology, then empiric addition of a prokinetic agent is appropriate. Unfortunately, the clinician's options in this regard are extremely limited (Merati, 2010). One prospective study demonstrated the ability of a prokinetic agent (cisapride or metoclopramide) to reduce or eliminate cough in subjects whose cough did not respond to PPI therapy alone (Poe and Kallay, 2003). Cisapride was removed from the United States market in 2000 due to its association with QT interval prolongation and ventricular arrhythmias (Layton et al., 2003). Other available prokinetic agents include bethanechol, a muscarinic agonist that must be avoided in asthmatics, and erythromycin, the use of which raises concern of numerous drug interactions, cardiac arrhythmias and emergence of resistant bacteria (Berthet et al., 2010). Furthermore, neither bethanechol nor erythromycin has ever been specifically evaluated in cough associated with GERD.
Hence, physicians in the United States and elsewhere are essentially left with metoclopramide to treat reflux-induced cough refractory to PPI therapy, but this agent is often associated with sedation and other undesirable effects and has been implicated as a cause of tardive dyskinesia after prolonged use (Rao and Camilleri, 2010). Domperidone, a dopaminergic antagonist like metoclopramide, has been used as a prokinetic outside the United States. Published data for the efficacy of this drug in reflux-induced cough are also lacking. Furthermore, domperidone has been associated with serious ventricular arrhythmias and sudden cardiac death (Johannes et al., 2010). Thus, the lack of safe and effective prokinetic agents for physicians to use in the setting of presumed reflux-induced cough refractory to acid suppression therapy represents a significant unmet clinical need.
An area of current investigation that may yield therapeutic options beyond that of acid suppression involves the γ-aminobutyric acid type B (GABA-B) receptor agonists. Transient lower oesophageal sphincter relaxation (TLESR) promotes the occurrence of gastro-oesophageal reflux. The GABA-B agonist baclofen has been shown to inhibit TLESR, and reduces the number of reflux episodes in subjects with GERD, reduces GERD symptoms in one study of chronic administration and increases basal lower oesophageal sphincter pressure in healthy volunteers and subjects with GERD (Kuo and Holloway, 2010). In addition, baclofen has been shown to be an anti-tussive agent independent of its effect on reflux by inhibiting capsaicin-induced cough in healthy volunteers (Dicpinigaitis and Dobkin, 1997). Despite such compelling evidence, baclofen has not gained widespread use because of undesirable side effects including sedation, dizziness, headache and confusion. Hence, newer GABA-B agonists with more specific peripheral activity and less central effects are currently the subject of active research (Kuo and Holloway, 2010).
When maximal medical therapy has failed to alleviate reflux-induced cough, surgical intervention may be considered. Data on outcomes of surgical fundoplication are still limited, and results to date confirm that a satisfactory outcome is not guaranteed. In one retrospective study of long-term (median 53 months) outcomes of laparoscopic anti-reflux surgery, atypical GERD symptoms including cough were reported to have improved in 65–75% of 128 patients. Of note, 33% of these post-operative patients were receiving daily antacid therapy at the time of interview (Kaufman et al., 2006). Another retrospective study evaluated 51 patients, out of a total of 240 who underwent surgical fundoplication, with predominantly extra-oesophageal symptoms of GERD. The 40 patients available for analysis (at a mean of 53.3 months post-surgery) were asked to grade their overall quality of life (QOL); 25% reported QOL as excellent, 32.% as good, 32.5% as satisfactory and 10% reported overall QOL as bad (Iqbal et al., 2009). One prospective study describing a community hospital experience with laparoscopic Nissen fundoplication found, as in other studies, greater improvement in typical rather than atypical GERD symptoms after long-term follow up (mean 24.6 months) of 84 patients. Comparison of pre- and post-operative symptom scores demonstrated a 58% improvement in cough. In the study group as a whole, 74% reported significant atypical GERD symptoms preoperatively; only 7% reported exclusively atypical symptoms (Ranson et al., 2007). Thus, further guidance is awaited from prospective trials of anti-reflux surgery in patients with refractory, GERD-induced chronic cough to assist the physician and patient in the difficult decision of whether to pursue a surgical option. Of particular significance from future studies would be the identification of specific characteristics predictive of a successful outcome after surgery.
In the evaluation of a patient with chronic cough, most physicians will appropriately seek to eliminate potential exogenous causative agents such as primary inhaled cigarette smoke and ACE inhibitor medications. In addition, a thorough history must be performed to identify other potentially relevant factors in the environment or workplace that could be causing or enhancing chronic cough. Epidemiological studies have demonstrated that environmental exposure to particulate matter, irritant gases, second-hand tobacco smoke, mixed pollutants and moulds is associated with increased cough (Joad et al., 2007). Furthermore, chronic cough is among the most prevalent work-related airway disorders, as documented in, for example, coal miners, hard-rock miners, tunnel workers and concrete manufacturing workers (Groneberg et al., 2006). In addition, some individuals appear to experience irritation and cough after exposure to non-toxic amounts of common substances such as perfumes, pesticides, paint and automobile exhaust, among many others (Brooks, 2010). This group presents a particular challenge in terms of discerning the underlying cause(s) of chronic cough.
In their 2006 cough management guidelines, the American College of Chest Physicians adopted the term unexplained cough to replace the previously used term idiopathic cough, as it was felt that the term unexplained cough better describes a condition that is likely due to multiple underlying aetiologies rather than a single entity (Pratter, 2006b). The challenge facing the clinician is to ascertain that a complete and thorough evaluation has been performed, including diagnostic-therapeutic trials with appropriate medications at adequate doses and for sufficient duration, before a diagnosis of unexplained cough is conferred. Nevertheless, there appears to exist a group of patients with genuine unexplained cough, and recent investigations have allowed the description of a particular phenotype of such individuals. Patients with apparent true unexplained chronic cough are often middle-aged (perimenopausal) women with prolonged dry cough and demonstrable cough reflex hypersensitivity (McGarvey, 2008). Additional studies have shown that such patients tend to have airway inflammation characterized by increased numbers of mast cells (McGarvey et al., 1999) and lymphocytes (Birring et al., 2003), and an increased incidence of organ-specific autoimmune disorders (Birring et al., 2004). Interestingly, the recently proposed entity of post-viral vagal neuropathy, in which chronic cough associated with laryngeal symptoms such as throat clearing follow an URI, appears to affect predominantly women in their fifth decade of life (Rees et al., 2009; Greene and Simpson, 2010).
Recently, the Cough Hypersensitivity Syndrome has been proposed to explain why a subgroup of individuals with common conditions such as rhinitis, asthma and GERD develop chronic, refractory cough, whereas the majority of the population with the same underlying conditions does not (Morice, 2010). At present, satisfactory therapeutic agents to manage chronic, unexplained cough are lacking. New safe and effective anti-tussive medications, as well as pharmacological agents aimed at the hypersensitised cough reflex likely underlying this condition, are eagerly awaited.