Diagnosis and Management of Cough Executive Summary ACCP Evidence-Based Clinical Practice Guidelines

Richard S. Irwin, MD, FCCP, Chair; Michael H. Baumann, MD, FCCP (HSP Liaison); Donald C. Bolser, PhD; Louis-Philippe Boulet, MD, FCCP (CTS Representative); Sidney S. Braman, MD, FCCP; Christopher E. Brightling, MBBS, FCCP; Kevin K. Brown, MD, FCCP; Brendan J. Canning, PhD; Anne B. Chang, MBBS, PhD; Peter V. Dicpinigaitis, MD, FCCP; Ron Eccles, DSc; W. Brendle Glomb, MD, FCCP; Larry B. Goldstein, MD; LeRoy M. Graham, MD, FCCP; Frederick E. Hargreave, MD; Paul A. Kvale, MD, FCCP; Sandra Zelman Lewis, PhD; F. Dennis McCool, MD, FCCP; Douglas C. McCrory, MD, MHSc; Udaya B.S. Prakash, MD, FCCP; Melvin R. Pratter, MD, FCCP; Mark J. Rosen, MD, FCCP; Edward Schulman, MD, FCCP (ATS Representative); John Jay Shannon, MD, FCCP (ACP Representative); Carol Smith Hammond, PhD; and Susan M. Tarlo, MBBS, FCCP

(CHEST 2006; 129:1S–23S) Abbreviations: ACE ϭ angiotensin-converting enzyme; ACP ϭ American College of Physicians; A/D ϭ antihistamine/ decongestant; ATS ϭ American Thoracic Society; BPC ϭ bronchoprovocation challenge; CTS ϭ Canadian Thoracic Society; DPB ϭ diffuse panbronchiolitis; dTap ϭ acellular pertussis; FEES ϭ fiberoptic endoscopic evaluation of swallowing; GERD ϭ gastroesophageal reflux disease; HRCT ϭ high-resolution CT; HSP ϭ Health & Psychosocial Instruments; IBD ϭ inflammatory bowel disease; ICS ϭ inhaled corticosteroid; ILD ϭ interstitial lung disease; NAEB ϭ nonasthmatic eosinophilic bronchitis; NSCLC ϭ non-small cell lung cancer; SLP ϭ speech-language pathol- ogist; TB ϭ tuberculosis; UACS ϭ upper airway cough syndrome; URI ϭ upper respiratory infection; VC ϭ voluntary cough; VSE ϭ videofluoroscopic swallow evaluation

ecognition of the importance of cough in clinical (3) updates and expands, when appropriate, all pre- R medicine was the impetus for the original evi- vious sections; and (4) adds new sections with topics dence-based consensus panel report on “Managing that were not previously covered. These new sections Cough as a Defense Mechanism and as a Symptom,” include nonasthmatic eosinophilic bronchitis (NAEB); published in 1998,1 and this updated revision. Com- acute bronchitis; nonbronchiectatic suppurative air- pared to the original cough consensus statement, this way diseases; cough due to aspiration secondary to revision (1) more narrowly focuses the guidelines on oral/pharyngeal dysphagia; environmental/occupa- the diagnosis and treatment of cough, the symptom, tional causes of cough; tuberculosis (TB) and other in adult and pediatric populations, and minimizes the infections; cough in the dialysis patient; uncommon discussion of cough as a defense mechanism; (2) causes of cough; unexplained cough, previously re- improves on the rigor of the evidence-based review ferred to as idiopathic cough; an empiric integrative and describes the methodology in a separate section; approach to the management of cough; assessing Reproduction of this article is prohibited without written permission cough severity and efficacy of therapy in clinical from the American College of Chest Physicians (www.chestjournal. research; potential future therapies; and future di- org/misc/reprints.shtml). rections for research. Correspondence to: Richard S. Irwin, MD, FCCP, University of Massachusetts Medical School, Room S6-842, 55 Lake Ave North, To mitigate future diagnostic confusion, two new Worcester, MA 01655; e-mail address: [email protected] diagnostic terms have been introduced to replace two www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 1S older terms that may represent misnomers. The com- For an in-depth discussion or clarification of each mittee unanimously recommends that the term upper recommendation, readers are encouraged to read airway cough syndrome (UACS) be used in preference the specific section in question in its entirety. to postnasal drip syndrome (PNDS) when discussing cough that is associated with upper airway conditions Methodology and Grading of the Evidence for the because it is unclear whether the mechanism of cough Diagnosis and Management of Cough5 is postnasal drip, direct irritation, or inflammation of • The recommendations were graded, by consensus the cough receptors in the upper airway. The commit- by the panel, using the American College of Chest tee also recommends using the term unexplained cough Physicians Health and Science Policy Grading rather than idiopathic cough because it is likely that System, which is based on the following two compo- more than one unknown cause of chronic cough will be nents: quality of evidence; and the net benefit of the discovered. The term idiopathic implies that one is diagnostic and therapeutic procedure. dealing with only one disease. • The quality of evidence is rated according to the For managing adult patients with cough, the com- study design and strength of other methodologies mittee recommends an empiric, integrative diagnos- used in the included studies. tic approach, which is presented in the section • The net benefit of the recommendations is based entitled “An Empiric Integrative Approach to the on the estimated benefit to the specific patient Management of Cough”.3 Guidelines for managing population described in each recommendation acute, subacute, and chronic cough are presented in and not for an individual patient. Usually, the net algorithmic form (Fig 1–3). Guidelines with algo- benefit is a clinical benefit to the population of rithms for evaluating chronic cough in pediatric patients defined in the first phrase of the recom- patients Ͻ 15 years of age are presented in the mendation, but, in recommendations for future section entitled “Guidelines for Evaluating Chronic research or other nonclinical recommendations, it Cough in Pediatrics”2,4 [Fig 4, 5]. For a full discus- may be a societal benefit. sion on how to use the algorithms, please refer to • Both the quality of evidence and the net benefit these sections. components are listed after each recommenda- tion; their interaction defines the strength of the Summary and Recommendations recommendations. • The recommendations scale is as follows: A, Recommendations for each section of these guide- strong; B, moderate; C, weak; D, negative; I, lines are listed under their respective section titles. inconclusive (no recommendation possible); E/A,

Figure 1. Acute cough algorithm for the management of patients Ն 15 years of age with cough lasting Ͻ 3 weeks. For diagnosis and treatment recommendations refer to the section indicated in the algorithm. PE ϭ pulmonary embolism; Dx ϭ diagnosis; Rx ϭ treatment; URTI ϭ upper respiratory tract infection; LRTI ϭ lower respiratory tract infection. Section 7 ϭ Irwin8; Section 8 ϭ Pratter9; Section 9 ϭ Pratter10; Section 10 ϭ Pratter11; Section 11 ϭ Dicpinigaitis12; Section 12 ϭ Irwin13; Section 13 ϭ Braman14; Section 14 ϭ Braman15; Section 16 ϭ Rosen17; Section 22 ϭ Irwin et al.23

2S Diagnosis and Management of Cough: ACCP Guidelines Figure 2. Subacute cough algorithm for the management of patients Ն 15 years of age with cough lasting 3 to 8 weeks. For diagnosis and treatment recommendations refer to the section indicated in the algorithm. AECB ϭ acute exacerbation of chronic bronchitis. See the legend of Figure 1 for abbreviations not used in the text. See Figure 1 for references to Sections.

strong recommendation based on expert opinion cough effectiveness. Level of evidence, expert only; E/B, moderate recommendation based on opinion; net benefit, substantial; grade of recom- expert opinion only; E/C, weak recommendation mendation, E/A based on expert opinion only; and E/D, negative 2. Individuals with neuromuscular weakness recommendation based on expert opinion only and no concomitant airway obstruction may benefit from mechanical aids to improve cough. Anatomy and Neurophysiology of the Cough Reflex6 Level of evidence, low; net benefit, intermediate; • There is clear evidence that vagal afferent nerves grade of recommendation, C regulate involuntary coughing. 3. In patients with ineffective cough, the cli- • Coughing, like swallowing, belching, urinating, nician should be aware of and monitor for and defecating, is unique because there is higher possible complications, such as pneumonia, at- cortical control of this visceral reflex. electasis, and/or respiratory failure. Level of • Cortical control can manifest as cough inhibition evidence, low; net benefit, substantial; grade of or voluntary cough. The implications of this are recommendation, B several-fold: because placebos can have a pro- found effect on coughing, treatment studies must Complications of Cough8 be placebo-controlled. Because cough can be an 1. In patients complaining of cough, evaluate affective behavior, psychological issues must be for a variety of complications associated with considered as a cause or effect of coughing. coughing (eg, cardiovascular, constitutional, GI, • There is a need to study the roles of consciousness genitourinary, musculoskeletal, neurologic, oph- and perception in coughing. thalmologic, psychosocial, and skin complica- tions), which can lead to a decrease in a patient’s Global Physiology and Pathophysiology of Cough7 health-related quality of life. Level of evidence, low; 1. In patients with endotracheal tubes, tra- benefit, substantial; grade of recommendation, B cheostomy need not be performed to improve 2. Patients with cough should have a thor- www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 3S Figure 3. Chronic cough algorithm for the management of patients Ն 15 years of age with cough lasting Ͼ 8 weeks. ACE-I ϭ ACE inhibitor; BD ϭ bronchodilator; LTRA ϭ leukotriene receptor antago- nist; PPI ϭ proton pump inhibitor. See the legend of Figure 1 for abbreviations not used in the text.

ough diagnostic evaluation, according to the Overview of Common Causes of Chronic Cough9 guidelines set forth in this document, to miti- gate or prevent these complications. Level of 1. In patients with chronic cough and a nor- evidence, low; net benefit, substantial; grade of mal chest roentgenogram finding who are non- recommendation, B smokers and are not receiving therapy with an

4S Diagnosis and Management of Cough: ACCP Guidelines Figure 4. Approach to a child Ͻ 15 years of age with chronic cough. There are limitations of the algorithm, which should be read with the accompanying text. Spirometry can usually be reliably performed in children Ͼ 6 years of age and in some children Ͼ 3 years of age if trained pediatric personnel are present. CXR ϭ chest radiograph.2

angiotensin-converting enzyme (ACE) inhibi- to result in a high rate of success in achieving tor, the diagnostic approach should focus on the cough resolution. Level of evidence, low; benefit, detection and treatment of UACS (formerly substantial; grade of recommendation, B called PNDS), asthma, NAEB, or GERD, alone 2. In all patients with chronic cough, regard- or in combination. This approach is most likely less of clinical signs or symptoms, because www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 5S Figure 5. Approach to a child Յ 14 years of age with chronic specific cough (ie, cough associated with other features suggestive of an underlying pulmonary and/or systemic abnormality). Children Ͼ 14 years of age should be managed as outlined in adult guidelines but there is no good evidence where the age cutoff should be. TEF ϭ tracheal esophageal fistula. See the legend of Figure 4 for abbreviation not used in the text.

UACS (formerly called PNDS), asthma, and accurate, and it should therefore be used in- GERD each may present only as cough with no stead of the term PNDS. Level of evidence, expert other associated clinical findings (ie, “silent opinion; benefit, substantial; grade of recommenda- PNDS,” “cough variant asthma,” and “silent tion, E/A GERD”), each of these diagnoses must be con- 2. In patients with chronic cough, the diag- sidered. Level of evidence, low; benefit, substantial; nosis of UACS-induced cough should be deter- grade of recommendation, B mined by considering a combination of criteria, 3. In patients with chronic cough, neither the including symptoms, physical examination find- patient’s description of his or her cough in ings, radiographic findings, and, ultimately, the terms of its character or timing, nor the pres- response to specific therapy. Because it is a ence or absence of sputum production, should syndrome, no pathognomonic findings exist. be used to rule in or rule out a diagnosis or to Level of evidence, low; benefit, substantial; grade of determine the clinical approach. Level of evi- recommendation, B dence, low; benefit, substantial; grade of recommen- 3. In patients in whom the cause of the dation, B UACS-induced cough is apparent, specific ther- apy directed at this condition should be insti- Chronic Upper Airway Cough Syndrome Secondary tuted. Level of evidence, low; benefit, substantial; to Rhinosinus Diseases (Previously Referred to as grade of recommendation, B Postnasal Drip Syndrome)10 4. For patients with chronic cough, an em- 1. In patients with chronic cough that is re- piric trial of therapy for UACS should be admin- lated to upper airway abnormalities, the com- istered because the improvement or resolution mittee considers the term UACS to be more of cough in response to specific treatment is the

6S Diagnosis and Management of Cough: ACCP Guidelines pivotal factor in confirming the diagnosis of findings are nondiagnostic, MIC testing should UACS as a cause of cough. Level of evidence, low; be performed to confirm the presence of benefit, substantial; grade of recommendation, B asthma. However, a diagnosis of CVA as the 5. A patient suspected of having UACS-in- cause of cough is established only after the duced cough who does not respond to empiric resolution of cough with specific antiasthmatic antihistamine/decongestant (A/D) therapy with therapy. If MIC testing cannot be performed, a first-generation antihistamine should next un- empiric therapy should be administered; how- dergo sinus imaging. Although chronic sinusitis ever, a response to steroid therapy will not may cause a productive cough, it may also be exclude NAEB as an etiology of the patient’s clinically silent, in that the cough can be rela- cough. Quality of evidence, good; net benefit, sub- tively or even completely nonproductive and stantial; grade of recommendation, A none of the typical findings associated with 3. Patients with cough due to asthma should acute sinusitis may be present. Level of evidence, initially be treated with a standard antiasth- low; benefit, substantial; grade of recommendation, B matic regimen of inhaled bronchodilators and 6. In patients for whom a specific etiology of inhaled corticosteroids (ICSs). Quality of evi- chronic cough is not apparent, empiric therapy dence, fair; net benefit, substantial; grade of recom- for UACS in the form of a first-generation A/D mendation, A preparation should be prescribed before begin- 4. In patients whose cough is refractory to ning an extensive diagnostic workup. Level of treatment with ICSs, an assessment of airway evidence, low; benefit, intermediate; grade of rec- inflammation should be performed whenever ommendation, C available and feasible. The demonstration of persistent airway eosinophilia during such an Cough and the Common Cold11 assessment will identify those patients who may 1. Patients with acute cough (as well as PND benefit from more aggressive antiinflammatory and throat clearing) associated with the com- therapy. Quality of evidence, low; net benefit, mon cold can be treated with a first-genera- substantial; grade of recommendation, B tion A/D preparation (brompheniramine and 5a. For patients with asthmatic cough that is sustained-release pseudoephedrine). Naproxen refractory to treatment with ICSs and broncho- can also be administered to help decrease dilators, in whom poor compliance or another cough in this setting. Level of evidence, fair; contributing condition has been excluded, a benefit, substantial; grade of recommendation, A leukotriene receptor antagonist may be added 2. In patients with the common cold, newer to the therapeutic regimen before the escala- generation nonsedating antihistamines are inef- tion of therapy to systemic corticosteroids. Qual- fective for reducing cough and should not be ity of evidence, fair; net benefit, intermediate; grade used. Level of evidence, fair; benefit, none; grade of of recommendation, B recommendation, D 5b. Patients with severe and/or refractory 3. In patients with cough and acute URTI, cough due to asthma should receive a short because symptoms, signs, and even sinus-imag- course (1 to 2 weeks) of systemic (oral) cortico- ing abnormalities may be indistinguishable steroids followed by ICSs. Quality of evidence, from acute bacterial sinusitis, the diagnosis of low; net benefit,: substantial; grade of recommenda- bacterial sinusitis should not be made during tion, B the first week of symptoms. (Clinical judgment is required to decide whether to institute anti- Chronic Cough Due to Gastroesophageal Reflux biotic therapy.) Level of evidence, fair; benefit, Disease13 none; grade of recommendation, D 1. In patients with chronic cough due to gastroesophageal reflux disease (GERD), the Chronic Cough Due to Asthma12 term acid reflux disease, unless it can be defin- 1. In a patient with chronic cough, asthma itively shown to apply, should be replaced by should always be considered as a potential eti- the more general term reflux disease so as not to ology because asthma is a common condition mislead the clinicians into thinking that all with which cough is commonly associated. Qual- patients with cough due to GERD should im- ity of evidence, fair; net benefit, substantial; grade of prove with acid-suppression therapy. Level of recommendation, A evidence, expert opinion; benefit, substantial; grade 2. In a patient suspected of having CVA but of recommendation, E/A in whom physical examination and spirometry 2. In patients with chronic cough who also www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 7S complain of typical and frequent GI complaints “Discussion” section regarding esophageal im- such as daily heartburn and regurgitation, es- pedance monitoring). When this is the case, pecially when the findings of chest-imaging barium esophagography is the test of choice to studies and/or clinical syndrome are consistent reveal GER of potential pathologic significance. with an aspiration syndrome, the diagnostic Level of evidence, low; benefit, substantial; grade of evaluation should always include GERD as a recommendation, B possible cause. Level of evidence, low; benefit, 10. In patients with cough due to GERD, a substantial; grade of recommendation, B normal esophagoscopy finding does not rule out 3. Patients with chronic cough who have GI GERD as the cause of cough. Level of evidence, symptoms that are consistent with GERD or low; benefit, substantial; grade of recommendation, B who fit the clinical profile described in Table 1 11. For patients fitting the clinical profile for in Irwin13, should be considered to have a high cough due to GERD, it is recommended that likelihood of having GERD and should be pre- treatment be initially started in lieu of testing. scribed antireflux treatment even when they Level of evidence, low; benefit, substantial; grade of have no GI symptoms. Level of evidence, low; recommendation, B benefit, substantial; grade of recommendation, B 12. For patients fitting the clinical profile for 4. In patients with chronic cough, it should cough due to GERD, the performance of 24-h not be assumed that GERD has been defini- esophageal pH monitoring is recommended on tively ruled out as a cause of cough simply therapy when cough does not improve or re- because there is a history of antireflux surgery. solve to assist in determining whether the ther- Level of evidence, low; benefit, substantial; grade of apy needs to be intensified or if medical therapy recommendation, B has failed. Level of evidence, low; benefit, substan- 5. In patients with chronic cough, while tests tial; grade of recommendation, B that link GERD with cough suggest a potential 13. For patients with chronic cough, the fol- cause-effect relationship, a definitive diagnosis lowing tests are not routinely recommended to of cough due to GERD requires that cough link cough with GERD: (a) assessing for lipid- nearly or completely disappear with antireflux laden macrophages in BAL fluid and induced treatment. Level of evidence, low; benefit, substan- sputum, because this test has not been studied tial; grade of recommendation, B in patients with chronic cough and because a 6. In patients with chronic cough being eval- positive test result is not specific for aspiration; uated for GERD, the 24-h esophageal pH-mon- (b) exhaled nitric oxide measurements, because itoring test is the most sensitive and specific they do not appear to be helpful in diagnosing test; however, it is recommended that the test cough due to GERD; (c) a Bernstein test, be- results be interpreted as normal only when cause a negative Bernstein test result cannot be conventional indices for acid reflux are within used to exclude the diagnosis of cough due to the normal range and no reflux-induced coughs GERD; and (d) inhaled tussigenic challenges appear during the monitoring study. Level of with capsaicin, because they are not specific for evidence, low; benefit, substantial; grade of recom- coughs due to GERD and because the test mendation, B result can be positive in patients with GERD 7. In patients with cough who are undergoing without cough. Level of evidence, low; benefit, 24-h monitoring, a low percentage of coughs conflicting; grade of recommendation, I associated with (or induced by) reflux does not 14. In patients who meet the clinical profile exclude a diagnosis of cough due to GERD. Level predicting that silent GERD is the likely cause of evidence, low; benefit, substantial; grade of recom- of chronic cough or in patients with chronic mendation, B cough who also have prominent upper GI symp- 8. In patients with cough due to GERD, the toms consistent with GERD, an empiric trial of degree of abnormality noted in the esophageal medical antireflux therapy is recommended. pH-monitoring variables, such as the frequency Level of evidence, low; benefit, substantial; grade of and duration of reflux events, does not directly recommendation, B correlate with the severity of the patients’ 15. For treating the majority of patients with cough. Level of evidence, low; benefit, substantial; chronic cough due to GERD, the following grade of recommendation, B medical therapies are recommended: (a) di- 9. In diagnosing nonacid GERD as the cause etary and lifestyle modifications; (b) acid sup- of cough, barium esophagography may be the pression therapy; and (c) the addition of proki- only available test to reveal GER of potential netic therapy either initially or if there is no pathologic significance in this setting (see the response to the first two therapies. The re-

8S Diagnosis and Management of Cough: ACCP Guidelines sponse to these therapies should be assessed months of intensive therapy (Table 3 in Irwin13), within 1 to 3 months. Level of evidence, expert and serial esophageal pH-monitoring studies or opinion; benefit, substantial; grade of recommenda- other objective studies (eg, barium esophagog- tion, E/A raphy, esophagoscopy, and gastric-emptying 16. In patients in which this empiric treat- study with solids) performed while the patient ment fails, it cannot be assumed that GERD has receives therapy show that intensive medical been ruled out as a cause of chronic cough; therapy has failed to control the reflux disease rather, the objective investigation for GERD is and that GERD is still the likely cause of cough; then recommended because the empiric ther- and (d) patients express the opinion that their apy may not have been intensive enough or persisting cough does not allow them a satisfac- medical therapy may have failed. Level of evi- tory quality of life. Level of evidence, expert dence, expert opinion; benefit, substantial; grade of opinion; benefit, substantial; grade of recommenda- recommendation, E/A tion, E/A 17. In some patients, cough due to GERD will favorably respond to acid suppression ther- Chronic Cough Due to Acute Bronchitis14 apy alone; proton pump inhibition may be ef- fective when H2-antagonism has been ineffec- 1. In a patient with an acute respiratory infec- tive; prokinetic therapy and diet, when added to tion manifested predominantly by cough, with or proton pump inhibition, may be effective when without sputum production, lasting no more than proton pump inhibition alone has been ineffec- 3 weeks, a diagnosis of acute bronchitis should not tive. Level of evidence, low; benefit, substantial; be made unless there is no clinical or radio- grade of recommendation, B graphic evidence of pneumonia, and the common 18. Patients requiring an intensive medical cold, acute asthma, or an exacerbation of COPD treatment regimen should be treated with the have been ruled out as the cause of cough. Quality following: (a) antireflux diet that includes no of evidence, expert opinion; benefit, substantial; grade > 45 g of fat in 24 h and no coffee, tea, soda, of recommendation, E/A chocolate, mints, citrus products, including to- 2. In patients with the presumed diagnosis of matoes, or alcohol, no smoking, and limiting acute bronchitis, viral cultures, serologic assays, vigorous exercise that will increase intraab- and sputum analyses should not be routinely dominal pressure; (b) acid suppression with a performed because the responsible organism is proton pump inhibitor; (c) prokinetic therapy; rarely identified in clinical practice. Quality of and (d) efforts to mitigate the influences of evidence, low; benefit, intermediate; grade of rec- comorbid diseases such as obstructive sleep ommendation, C apnea or therapy for comorbid conditions (eg, 3. In patients with acute cough and sputum nitrates, progesterone, and calcium channel production suggestive of acute bronchitis, the blockers) whenever possible. Level of evidence, absence of the following findings reduces the expert opinion; benefit, substantial; grade of recom- likelihood of pneumonia sufficiently to elimi- mendation, E/A nate the need for a chest radiograph: (1) heart 19. In patients with chronic cough due to rate > 100 beats/min; (2) respiratory rate > 24 GERD that has failed to improve with the most breaths/min; (3) oral body temperature of maximal medical therapy, which includes an > 38°C; and (4) chest examination findings of intensive antireflux diet and lifestyle modifica- focal consolidation, egophony, or fremitus. tion, maximum acid suppression, and prokinetic Quality of evidence, low; benefit, substantial; grade therapy, and the rest of the spectrum of treat- of recommendation, B ment options in Table 3 in Irwin,13 cough may 4a. For patients with the putative diagnosis only improve or be eliminated with antireflux of acute bronchitis, routine treatment with an- surgery. Level of evidence, low; benefit, substantial; tibiotics is not justified and should not be of- grade of recommendation, B fered. Quality of evidence, good; benefit, none; 20. In patients who meet the following crite- grade of recommendation, D ria, antireflux surgery is the recommended 4b. For these patients, the decision not to use treatment: (a) findings of a 24-h esophageal an antibiotic should be addressed individually pH-monitoring study before treatment is posi- and explanations should be offered because tive, as defined above; (b) patients fit the clini- many patients expect to receive an antibiotic cal profile suggesting that GERD is the likely based on previous experiences and public ex- cause of their cough (Table 1 in Irwin13); (c) pectation. Quality of evidence, expert opinion; ben- cough has not improved after a minimum of 3 efit, intermediate; grade of recommendation, E/B www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 9S 5. Children and adult patients with con- with increased cough, sputum production, spu- firmed and probable whooping cough should tum purulence, and/or shortness of breath, receive a macrolide antibiotic and should be which are often preceded by symptoms of an isolated for 5 days from the start of treatment; upper respiratory tract infection, should be early treatment within the first few weeks will considered to have an acute exacerbation of diminish the coughing paroxysms and prevent chronic bronchitis, as long as conditions other spread of the disease; the patient is unlikely to than acute tracheobronchitis are ruled out or respond to treatment beyond this period. Level are considered unlikely. Level of evidence, expert of evidence, good; net benefit, substantial; grade of opinion; net benefit, substantial; grade of recom- evidence, A mendation, E/A 6a. In most patients with a diagnosis of acute 5. In patients with chronic cough who have ␤ bronchitis, 2-agonist bronchodilators should chronic exposure to respiratory irritants, such not be routinely used to alleviate cough. Quality as personal tobacco use, passive smoke expo- of evidence, fair; benefit, none; grade of recommen- sure, and workplace hazards, avoidance dation, D should always be recommended. It is the most 6b. In select adult patients with a diagnosis of effective means to improve or eliminate the acute bronchitis and wheezing accompanying cough of chronic bronchitis. Ninety percent of ␤ the cough, treatment with 2-agonist broncho- patients will have resolution of their cough dilators may be useful. Quality of evidence, fair; after smoking cessation. Level of evidence, benefit, small/weak; grade of recommendation, C good; net benefit, substantial; grade of recommen- 7. In patients with a diagnosis of acute bron- dation, A chitis, antitussive agents are occasionally useful 6. In stable patients with chronic bronchi- and can be offered for short-term symptomatic tis, there is no role for long-term prophylactic relief of coughing. Quality of evidence, fair; bene- therapy with antibiotics. Level of evidence, low; fit, small/weak; grade of recommendation, C benefit, none; grade of recommendation, I 8. In patients with a diagnosis of acute bron- 7. In patients with acute exacerbations of chitis, because there is no consistent favorable chronic bronchitis, the use of antibiotics is rec- effect of mucokinetic agents on cough, they are ommended; patients with severe exacerbations not recommended. Quality of evidence, fair; ben- and those with more severe airflow obstruction efit, conflicting; grade of recommendation, I at baseline are the most likely to benefit. Level of evidence, fair; net benefit, substantial; grade of recommendation, A Chronic Cough Due to Chronic Bronchitis15 8. In stable patients with chronic bronchitis, 1. Adults who have a history of chronic the clinical benefits of postural drainage and cough and sputum expectoration occurring on chest percussion have not been proven, and most days for at least 3 months and for at least they are not recommended. Level of evidence, 2 consecutive years should be given a diagnosis fair; net benefit, conflicting; grade of recommenda- of chronic bronchitis when other respiratory or tion, I cardiac causes of chronic productive cough are 9. In patients with an acute exacerbation of ruled out. Level of evidence, low; net benefit, chronic bronchitis, the clinical benefits of pos- substantial; grade of recommendation, B tural drainage and chest percussion have not 2. The evaluation of patients with chronic been proven, and they are not recommended. cough should include a complete history re- Level of evidence, fair; net benefit, conflicting; grade of garding exposures to respiratory irritants in- recommendation, I cluding cigarette, cigar, and pipe smoke; passive 10a. In stable patients with chronic bronchi- smoke exposures; and hazardous environments tis, therapy with short-acting ␤-agonists should in the home and workplace. All are predispos- be used to control bronchospasm and relieve ing factors of chronic bronchitis. Level of evi- dyspnea; in some patients, it may also reduce dence, low; net benefit, substantial; grade of recom- chronic cough. Level of evidence, good; net bene- mendation, B fit, substantial; grade of recommendation, A 3. Smoke-free workplace and public place 10b. In stable patients with chronic bronchi- laws should be enacted in all communities. tis, therapy with ipratropium bromide should Level of evidence, expert opinion; net benefit, sub- be offered to improve cough. Level of evidence, stantial; grade of recommendation, E/A fair; net benefit, substantial; grade of recommenda- 4. Stable patients with chronic bronchitis tion, A who have a sudden deterioration of symptoms 10c. In stable patients with chronic bronchi-

10S Diagnosis and Management of Cough: ACCP Guidelines tis, treatment with theophylline should be con- 19. In patients with chronic bronchitis, cen- sidered to control chronic cough; careful mon- tral cough suppressants such as codeine and itoring for complications is necessary. Level of dextromethorphan are recommended for short- evidence, fair; net benefit, substantial; grade of term symptomatic relief of coughing. Level of recommendation, A evidence, fair; benefit, intermediate; grade of evi- 11. For patients with an acute exacerbation dence, B of chronic bronchitis, therapy with short-acting ␤-agonists or anticholinergic bronchodilators Chronic Cough Due to Nonasthmatic Eosinophilic should be administered during the acute exac- Bronchitis16 erbation. If the patient does not show a prompt 1. In patients with chronic cough who have response, the other agent should be added after normal chest radiograph findings, normal spi- the first is administered at the maximal dose. rometry findings, and no evidence of variable Level of evidence, good; net benefit, substantial; airflow obstruction or airway hyperresponsive- grade of recommendation, A ness, the diagnosis of NAEB should be consid- 12. For patients with an acute exacerbation ered. Level of evidence, expert opinion; benefit, of chronic bronchitis, theophylline should not substantial; grade of recommendation, E/A be used for treatment. Level of evidence, good; 2. In patients with chronic cough with nor- net benefit, none; grade of recommendation, D mal chest radiograph findings, normal spirom- 13. For stable patients with chronic bronchi- etry findings, and no evidence of variable tis, there is no evidence that the currently airflow obstruction or airway hyperresponsive- available expectorants are effective and there- ness, the diagnosis of NAEB as the cause of the fore they should not be used. Level of evidence, chronic cough is confirmed by the presence of low; net benefit, none; grade of recommendation, I airway eosinophilia, either by sputum induction 14. In stable patients with chronic bronchitis, or bronchial wash fluid obtained by bronchos- treatment with a long-acting ␤-agonist when copy, and an improvement in the cough follow- coupled with an ICS should be offered to con- ing corticosteroid therapy. Level of evidence, trol chronic cough. Level of evidence, good; net expert opinion; benefit, substantial; grade of recom- benefit, substantial; grade of recommendation, A mendation, E/A 15. For stable patients with chronic bronchi- 3. In patients with chronic cough due to tis and an FEV of < 50% predicted or for those 1 NAEB, the possibility of an occupation-related patients with frequent exacerbations of chronic cause needs to be considered. Level of evidence, bronchitis, ICS therapy should be offered. Level expert opinion; benefit, substantial; grade of recom- of evidence, good; net benefit, substantial; grade of mendation, E/A recommendation, A 4. For patients with chronic cough due to 16. For stable patients with chronic bronchi- NAEB, the first-line treatment is ICSs (except tis, long-term maintenance therapy with oral when a causal allergen or sensitizer is identified corticosteroids such as prednisone should not [see recommendation 5]). Level of evidence, low; be used; there is no evidence that it improves benefit, substantial; grade of recommendation, B cough and sputum production, and the risks of 5. For patients with chronic cough due to serious side effects are high. Level of evidence, NAEB, when a causal allergen or occupational expert opinion; net benefit, negative; grade of rec- sensitizer is identified, avoidance is the best ommendation, E/D treatment. Level of evidence, expert opinion; ben- 17. For patients with an acute exacerbation of efit, substantial; grade of recommendation, E/A chronic bronchitis, there is no evidence that the 6. For patients with chronic cough due to currently available expectorants are effective, NAEB, if symptoms are persistently trouble- and therefore they should not be used. Level of some and/or the natural history of eosinophilic evidence, low; net benefit, none; grade of recommen- airway inflammation progresses despite treat- dation, I ment with high-dose ICSs, oral corticosteroids 18. For patients with an acute exacerbation should be given. Level of evidence, expert opinion; of chronic bronchitis, a short course (10 to 15 benefit, substantial; grade of recommendation, E/A days) of systemic corticosteroid therapy should be given; IV therapy in hospitalized patients Chronic Cough Due to Bronchiectasis17 and oral therapy for ambulatory patients have both proven to be effective. Level of evidence, 1. In patients with suspected bronchiectasis good; net benefit, substantial; grade of recommen- without a characteristic chest radiograph find- dation, A ing, an high-resolution CT (HRCT) scan of the www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 11S chest should be ordered because it is the diag- selection of agents depending on the likely nostic procedure of choice to confirm the diag- pathogens. Level of evidence, low; benefit, substan- nosis. Level of evidence, low; benefit, substantial; tial; grade of recommendation, B grade of recommendation, B 2. In patients for whom there is no obvious Chronic Cough Due to Nonbronchiectatic cause, a diagnostic evaluation for an underlying Suppurative Airway Disease (Bronchiolitis)18 disorder causing bronchiectasis should be per- formed, because the results may lead to treat- 1. In patients with cough and incomplete or ment that may slow or halt the progression of irreversible airflow limitation, direct or indirect disease. Level of evidence, low; benefit, substantial; signs of small airways disease seen on HRCT grade of recommendation, B scan, or purulent secretions seen on bronchos- 3. In patients with bronchiectasis with air- copy, nonbronchiectatic suppurative airways flow obstruction and/or bronchial hyperreactiv- disease (bronchiolitis) should be suspected as ity, therapy with bronchodilators may be of the primary cause. Level of evidence, expert opin- benefit. Level of evidence, expert opinion; benefit, ion; benefit, substantial; grade of recommendation, small; grade of recommendation, E/C E/A 4. In patients with bronchiectasis caused by 2. In patients with cough in whom more cystic fibrosis (CF), rhDNase should be used to common causes have been excluded, because improve spirometry. Level of evidence, low; ben- bacterial suppurative airways disease may be efit, small; grade of recommendation, C present and clinically unsuspected, bronchos- 5. In patients with CF, prolonged treatment copy is required before excluding it as a cause. with systemic corticosteroids should not be of- Level of evidence, low; benefit, substantial; grade of fered to most patients because of significant recommendation, B side effects. Level of evidence, low; benefit, con- 3. In patients in whom bronchiolitis is sus- flicting; grade of recommendation, I pected, a surgical lung biopsy should be per- 6. In patients with CF, prolonged courses of formed when the combination of the clinical ibuprofen should not be used. Level of evidence, syndrome, physiology, and HRCT findings do low; benefit, conflicting; grade of recommendation, I not provide a confident diagnosis. Level of evi- 7. In patients with idiopathic bronchiectasis, dence, expert opinion; benefit, substantial; grade of the prolonged systemic administration of anti- recommendation, E/A biotics may produce small benefits in reducing 4. In patients with bacterial bronchiolitis, sputum volume and purulence, but may also be prolonged antibiotic therapy improves cough associated with intolerable side effects. Level of and is recommended. Level of evidence, low; evidence, low; benefit, conflicting, grade of recom- benefit, substantial; grade of recommendation, B mendation, I 5. In patients with toxic/antigenic exposure 8. In patients with CF, therapy with aerosol- or drug-related bronchiolitis, cessation of the ized antipseudomonal antibiotics are recom- exposure or medication plus corticosteroid mended. Level of evidence, low; benefit, interme- therapy for those with physiologic impairment diate; grade of recommendation, C is appropriate. Level of evidence, expert opinion; 9. In patients with idiopathic bronchiectasis, benefit, substantial; grade of recommendation, E/A aerosolized antibiotics should not be used. Level 6. In the inflammatory bowel disease (IBD) of evidence, low; benefit, negative; recommendation, D patient with cough, bronchiolitis should be sus- 10. In patients with conditions associated pected as a potential cause. Level of evidence, with the hypersecretion of mucus and the in- low; benefit, substantial; grade of recommendation, B ability to expectorate effectively, chest physio- 7. In patients in whom IBD-related bronchi- therapy should be used and patients should be olitis is suspected, both adverse drug reaction monitored for symptom improvement. Level of and infection should be specifically considered. evidence, expert opinion; benefit, small/weak; grade Level of evidence, expert opinion; benefit, substan- of recommendation, E/C tial; grade of recommendation, E/A 11. In selected patients with localized bron- 8. In patients with IBD, therapy with both chiectasis that causes intolerable symptoms de- oral corticosteroids and ICSs may improve spite maximal medical therapy, surgery should cough, and a trial of therapy is suggested. Level be offered. Level of evidence, low; benefit, substan- of evidence, low; benefit, substantial; grade of rec- tial; grade of recommendation, B ommendation, B 12. In patients with exacerbations of bronchi- 9. In patients with chronic cough who have ectasis, antibiotics should be used, with the recently lived in Japan, Korea, or China, diffuse

12S Diagnosis and Management of Cough: ACCP Guidelines panbronchiolitis (DPB) should be considered in 4c. In patients with postinfectious cough, the evaluations of the cause. Level of evidence, when the cough adversely affects the patient’s low; benefit, substantial; grade of recommendation, B quality of life and when cough persists despite 10. In patients with suspected DPB, an ap- use of inhaled ipratropinin, consider the use of propriate clinical setting and characteristic inhaled corticosteroids. Level of evidence, expert HRCT scan findings may obviate the need for opinion; net benefit, intermediate; grade of evi- invasive testing and a trial of macrolide therapy dence, E/B (erythromycin or other 14-member ring macro- 4d. For severe paroxysms of postinfectious lides such as clarithromycin and roxithromycin) cough, consider prescribing 30 to 40 mg of is appropriate. Level of evidence, expert opinion; prednisone per day for a short, finite period of benefit, substantial; grade of recommendation, E/A time when other common causes of cough (eg, 11. In patients with DPB, prolonged treat- > UACS due to rhinosinus diseases, asthma, or ment ( 2 to 6 months) with erythromycin (or gastroesophageal reflux disease) have been other 14-member ring macrolides such as clar- ruled out. Level of evidence, low; net benefit, ithromycin and roxithromycin) is recom- intermediate; grade of evidence, C mended. Level of evidence, low; benefit, substan- 4e. Central acting antitussive agents such as tial; grade of recommendation, B codeine and dextromethorphan should be con- sidered when other measures fail. Level of evi- 19 Postinfectious Cough dence, expert opinion; net benefit, intermediate; 1. When a patient complains of cough that grade of evidence, E/B has been present for at least 3 weeks, but not 5. When a patient has a cough lasting for > 2 more than 8 weeks, consider a diagnosis of weeks without another apparent cause and it is postinfectious cough. Quality of evidence, expert accompanied by paroxysms of coughing, post- opinion; net benefit, intermediate; strength of rec- tussive vomiting, and/or an inspiratory whoop- ommendation, E/B ing sound, the diagnosis of a B pertussis infec- 2. In patients with subacute postinfectious tion should be made unless another diagnosis is cough, because there are multiple pathogenetic proven. Level of evidence, low; net benefit, sub- factors that may contribute to the cause of stantial; grade of evidence, B cough (including postviral airway inflammation 6a. For all patients who are suspected of with its attendant complications such as bron- having whooping cough, to make a definitive chial hyperresponsiveness, mucus hypersecre- diagnosis order a nasopharyngeal aspirate or tion and impaired mucociliary clearance, upper polymer (Dacron; INVISTA; Wichita, KS) swab airway cough syndrome [UACS], asthma, and of the nasopharynx for culture to confirm the gastroesophageal reflux disease), judge which presence of B pertussis. Isolation of the bacteria factors are most likely provoking cough before is the only certain way to make the diagnosis. considering therapy. Quality of evidence, expert Level of evidence, low; net benefit, substantial; opinion; net benefit, intermediate; strength of rec- grade of evidence, B ommendation, E/B 6b. PCR confirmation is available but is not 3. In children and adult patients with cough following an acute respiratory tract infection, if recommended as there is no universally ac- cough has persisted for > 8 weeks, consider cepted, validated technique for routine clinical diagnoses other than postinfectious cough. testing. Level of evidence, low; net benefit, conflict- Quality of evidence, low; net benefit, intermediate; ing; grade of evidence, I strength of recommendation, C 7. In patients with suspected pertussis infec- 4. For adult patients with postinfectious tion, to make a presumptive diagnosis of this cough, not due to bacterial sinusitis or early on infection, order paired acute and convalescent in a Bordetella pertussis infection, while the sera in a reference laboratory. A fourfold in- optimal treatment is not known: crease in IgG or IgA antibodies to PT or FHA is 4a. Therapy with antibiotics has no role, as consistent with the presence of a recent B the cause is not bacterial infection. Level of pertussis infection. Level of evidence, low; net evidence, expert opinion; net benefit, none; grade of benefit, intermediate; grade of evidence, C evidence, I 8. A confirmed diagnosis of pertussis infec- 4b. Consider a trial of inhaled ipratropium as tion should be made when a patient with cough it may attenuate the cough. Level of evidence, has B pertussis isolated from a nasopharyngeal fair; net benefit, intermediate; grade of evidence, B culture or has a compatible clinical picture with www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 13S an epidemiologic linkage to a confirmed case. and/or chemotherapy should usually be offered. Level of evidence, low; net benefit, substantial; Level of evidence, good; benefit, intermediate; grade grade of evidence, B of recommendation, A 9. Children and adult patients with con- 5. For patients with dyspnea or hemoptysis firmed or probable whooping cough should due to endobronchial tumors, cough may also receive a macrolide antibiotic and should be be present. Endobronchial methods should be isolated for 5 days from the start of treatment considered for the palliation of these symptoms, because early treatment within the first few but cough alone is seldom a reason to offer such weeks will diminish the coughing paroxysms treatment. Level of evidence, fair; benefit, small; and prevent spread of the disease; treatment grade of recommendation, C beyond this period may be offered but it is 6. For patients with cough and lung cancer, unlikely the patient will respond. Level of evi- the use of centrally acting cough suppressants dence, good; net benefit, substantial; grade of evi- such as dihydrocodeine and hydrocodone is dence, A recommended. Level of evidence, low; benefit, 10. Long-acting ␤-agonists, antihistamines, intermediate; grade of recommendation, C corticosteroids, and pertussis Ig should not be offered to patients with whooping cough be- cause there is no evidence that they benefit Cough and Aspiration of Food and Liquids Due to these patients. Level of evidence, good; net benefit, Oral-Pharyngeal Dysphagia21 none; grade of evidence, D 1. In patients with cough, a medical history 11. All children should receive prevention against pertussis infection as part of a complete particularly directed at identifying conditions diphtheria, tetanus, acellular pertussis (dTap) increasing the likelihood of oral-pharyngeal primary vaccination series. This should be fol- dysphagia and aspiration (as indicated in Table 21 lowed by a single dose dTap booster vaccination 1 in Smith Hammond and Goldstein ) should early in adolescence. Level of evidence, good; net be obtained. Patients with high-risk conditions benefit, substantial; grade of evidence, A should be referred for an oral-pharyngeal swal- 12. For all adults up to the age of 65, regular lowing evaluation. Level of evidence, low; benefit, vaccination with dTap should be administered substantial; grade of recommendation, B routinely at intervals of 10 years. Level of evi- 2a. Patients with cough and their caregivers dence, expert opinion; net benefit, substantial; grade should be questioned regarding perceived swal- of evidence, E/A lowing problems, including an association of cough while eating or drinking and a fear of Chronic Cough Due to Lung Tumors20 choking while eating and drinking. If a patient with cough reports swallowing problems, further 1. In a patient with cough who has risk fac- evaluation for oral-pharyngeal dysphagia is indi- tors for lung cancer or a known or suspected cated. Level of evidence, low; benefit, substantial; cancer in another site that may metastasize to grade of recommendation, B the lungs, a chest radiograph should be ob- 2b. Further evaluation, including a chest ra- tained. Level of evidence, expert opinion; benefit, diograph and a nutritional assessment, should substantial; grade of recommendation, E/A be considered in patients with cough or condi- 2. In patients with a suspicion of airway in- tions associated with aspiration. Level of evi- volvement by a malignancy (eg, smokers with dence, low; benefit, substantial; grade of recommen- hemoptysis), even when the chest radiograph findings are normal, bronchoscopy is indicated. dation, B Level of evidence, low; benefit, substantial; grade of 3. Patients with oral-pharyngeal dysphagia recommendation, B and cough should be referred, ideally to a 3. For patients with stage I and II non-small speech-language pathologist (SLP), for an oral- cell lung cancer (NSCLC), surgery to remove the pharyngeal swallow evaluation. Level of evi- NSCLC is the treatment of choice. If cough was dence, low; benefit, substantial; grade of recommen- caused by a NSCLC that can be surgically re- dation, B moved, the cough will typically cease. Level of 4. Patients with cough related to pneumonia evidence, low; benefit, substantial; grade of recommen- and bronchitis who have received medical diag- dation, B noses and conditions associated with aspiration 4. For patients with more advanced NSCLC (Table 1 in Smith Hammond and Goldstein21) (stages III and IV), external beam radiation should be referred, ideally to an SLP, for an

14S Diagnosis and Management of Cough: ACCP Guidelines oral-pharyngeal swallow evaluation. Level of ev- ommendations should be prescribed when indi- idence, low; benefit, substantial; grade of recommen- cated, and can be refined by testing with foods dation, B and liquids simulating those in a normal diet 5. Patients with a reduced level of conscious- during the VSE or FEES. Level of evidence, low; ness are at high risk for aspiration and should benefit, substantial; grade of recommendation, B not be fed orally until the level of consciousness 14. For patients with muscular weakness dur- has improved. Level of evidence, low; benefit, ing swallowing, muscle strength training, with substantial; grade of recommendation, B or without electromyographic biofeedback, and 6. Alert patients with cough who are in high- electrical stimulation treatment of the swallow- risk groups for aspiration (Table 1 in Smith ing musculature are promising techniques but Hammond and Goldstein21) should be observed cannot be recommended at this time until fur- drinking small amounts of water (3 oz). If the ther work in larger populations is performed. patient coughs or shows clinical signs that are Level of evidence, low; benefit, conflicting; grade of associated with aspiration (Tables 2, 3 in Smith evidence, I Hammond and Goldstein21), the patient should 15. Patients with intractable aspiration may be referred for a detailed swallowing evalua- be considered for surgical intervention. Level of tion, preferably to an SLP. Level of evidence, low; evidence, low; benefit, substantial; grade of recom- benefit, substantial; grade of recommendation, B mendation, B 7. In patients with cough, the value of the subjective assessment of voluntary cough (VC) Angiotensin-Converting Enzyme Inhibitor-Induced as the sole predictor of aspiration is uncertain Cough22 because of poor reliability and an unclear asso- ciation with evaluation. Level of evidence, low; 1. In patients presenting with chronic cough, benefit, conflicting; grade of evidence, I in order to determine that the angiotensin- 8. The assessment of the reflexive cough re- converting enzyme (ACE) inhibitor is the cause sponse to inhaled irritants as a predictor of of the cough, therapy with ACE inhibitors aspiration risk and subsequent pneumonia is should be discontinued regardless of the tem- not recommended due to a lack of adequate poral relation between the onset of cough and supportive studies. Level of evidence, low; benefit, the initiation of ACE inhibitor therapy. The conflicting; grade of evidence, I diagnosis is confirmed by the resolution of 9. In acute stroke patients, the expulsive cough, usually within 1 to 4 weeks of the cessa- phase rise time of VC may predict aspiration. tion of the offending agent; however, the reso- The use of this test has not been validated in lution of cough may be delayed in a subgroup of other patient groups, and further studies com- patients for up to 3 months. Quality of evidence, paring the accuracy of objective measures of low; net benefit, substantial; grade of recommenda- VC to the clinical swallow evaluation to identify tion, B aspiration risk are needed. Level of evidence, low; 2. In patients presenting with chronic ACE benefit, small; grade of recommendation, C inhibitor-induced cough, discontinue therapy 10. Patients with dysphagia should undergo with the drug because it is the only uniformly videofluoroscopic swallow evaluation (VSE) or effective treatment. Quality of evidence, low; net fiberoptic endoscopic evaluation of swallowing benefit, substantial; grade of recommendation, B (FEES) to identify appropriate treatment. Level 3. In patients whose cough resolves after the of evidence, low; benefit, substantial; grade of rec- cessation of therapy with ACE inhibitors, and ommendation, B for whom there is a compelling reason to treat 11. Patients with dysphagia should be man- with these agents, a repeat trial of ACE inhibi- aged by organized multidisciplinary teams that tor therapy may be attempted. Quality of evi- may include a physician, a nurse, an SLP, a dence, fair; net benefit, substantial; grade of recom- dietitian, and physical and occupational thera- mendation, A pists. Level of evidence, low; benefit, substantial; 4. In patients for whom the cessation of ACE grade of recommendation, B inhibitor therapy is not an option, pharmacologic 12. In patients with dysphagia, VSE or FEES therapy, including that with sodium cromogly- can be useful for determining compensatory cate, theophylline, sulindac, indomethacin, amlo- strategies enabling patients with dysphagia to dipine, nifedipine, ferrous sulfate, and picota- safely swallow. Level of evidence, low; benefit, mide that is aimed at suppressing cough should substantial; grade of recommendation, B be attempted. Quality of evidence, fair; net benefit, 13. In patients with dysphagia, dietary rec- intermediate; grade of recommendation, B www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 15S 5. In patients in whom persistent or intoler- should not be used to diagnose or exclude able ACE inhibitor-induced cough occurs, ther- psychogenic cough. Level of evidence, expert apy should be switched, when indicated, to an opinion; benefit, substantial; grade of recommenda- angiotensin receptor blocker, with which the tion, E/A incidence of associated cough appears to be 5. In adult and pediatric patients with chronic similar to that for control drug, or to an appro- unexplained cough, common psychosocial prob- priate agent of another drug class. Quality of lems such as anxiety, depression, domestic vio- evidence, good; net benefit, substantial; grade of lence, and child abuse/neglect that are often as- recommendation, A sociated with somatization disorders should be evaluated. Level of evidence, expert opinion; benefit, Habit Cough, Tic Cough, and Psychogenic Cough substantial; grade of recommendation, E/A in Adult and Pediatric Populations23 6. In adult and pediatric patients with chronic cough associated with troublesome psy- 1a. In adult patients with chronic cough, the chological manifestations, psychological coun- diagnoses of habit cough or psychogenic cough seling or psychiatric intervention should be can only be made after an extensive evaluation encouraged, after other causes have been ruled has been performed that includes ruling out tic out. Level of evidence, expert opinion; benefit, disorders and uncommon causes (as described small/weak; grade of recommendation, E/C in the section “Uncommon Causes of Cough”), and cough improves with specific therapy such Chronic Cough Due to Chronic Interstitial as behavior modification or psychiatric therapy. Pulmonary Diseases24 Level of evidence, expert opinion; benefit, substan- tial; grade of recommendation, E/A 1. In patients with chronic cough, before 1b. In adult patients with chronic cough that diagnosing interstitial lung disease (ILD) as the remains persistently troublesome despite an sole cause, common etiologies such as UACS, extensive and thorough evidence-based evalua- which was previously referred to as PNDS, tion, and after behavior modification and/or asthma, and GERD should be considered. As psychiatric therapy have failed, unexplained these common causes may also share clinical cough should be diagnosed rather than a habit features with specific ILDs, a diagnosis of ILD cough or psychogenic cough. Level of evidence, as the cause of cough should be considered a expert opinion; benefit, substantial; grade of recom- diagnosis of exclusion. Level of evidence, expert mendation, E/A opinion; benefit, substantial; grade of recommenda- 1c. In children with chronic cough, the diag- tion, E/A noses of habit cough or psychogenic cough can 2. In patients with cough secondary to an only be made after tic disorders and Tourette ILD, because of the prognostic implications, syndrome have been evaluated and cough im- primary treatment should be dictated by the proves with specific therapy such as behavior specific disorder. Level of evidence, low; benefit, modification or psychiatric therapy. Level of substantial; grade of recommendation, B evidence, expert opinion; benefit, substantial; grade 3. In patients with cough secondary to idio- of recommendation, E/A pathic pulmonary fibrosis, corticosteroids may 2. In adult patients with cough, the diagnosis lead to symptomatic improvement; however, as of habit cough should not be made unless bio- they have been shown to neither prolong survival logical and genetic tic disorders associated with nor improve quality of life and may be associated coughing such as Tourette syndrome have been with significant side effects, their use requires an ruled out. Level of evidence, expert opinion; bene- individualized analysis of the overall benefits and fit, substantial; grade of recommendation, E/A risks. Level of evidence, expert opinion; benefit, inter- 3. In adults with chronic cough, the presence mediate; grade of recommendation, E/B or absence of nighttime cough or cough with a 4. In patients with cough and characteristic barking or honking character should not be clinical and radiographic features, sarcoidosis used to diagnose or exclude a diagnosis of should be considered as a cause. Level of evi- psychogenic cough. Level of evidence, low; bene- dence, expert opinion; benefit, substantial; grade of fit, substantial; grade of recommendation, B recommendation, E/A 4. In children with chronic cough, the char- 5. In patients with cough secondary to sarcoid- acteristics of the cough may be suggestive of, osis, although therapy with oral corticosteroids but are not diagnostic of, psychogenic cough. may lead to symptomatic improvement, as they The presence or absence of nighttime cough have not been proven to have a durable benefit

16S Diagnosis and Management of Cough: ACCP Guidelines and are associated with significant side effects, an Chronic Cough Due to Tuberculosis and Other individualized analysis of the overall benefit and Infections26 risk is necessary. Level of evidence, fair; benefit, 1. In areas where there is a high prevalence intermediate; grade of recommendation, B of TB, chronic cough should be defined as it is 6. In patients with cough secondary to sar- in the World Health Organization Practical Ap- coidosis, therapy with oral corticosteroids fol- proach to Lung Health program as being 2 to 3 lowed by ICSs may improve symptoms. Level of weeks in duration. Level of evidence, low; benefit, evidence, fair; benefit, conflicting; grade of recom- substantial; grade of recommendation, B mendation, I 2. In patients with chronic cough who live in 7. In patients with cough, ILD, and a con- areas with a high prevalence of TB, this diagnosis cerning environmental, occupational, or avoca- should be considered, but not to the exclusion of tional exposure, hypersensitivity pneumonitis the more common etiologies. Sputum smears and should be considered as a potential cause. Level cultures for acid-fast bacilli and a chest radio- of evidence, expert opinion; benefit, substantial; graph should be obtained whenever possible. grade of recommendation, E/A Level of evidence, low; benefit, substantial; grade of 8. In patients with cough due to hypersensi- recommendation, B tivity pneumonitis, treatment should include 3. In patients with suspected TB, future in- the removal of the offending exposure and vestigations are needed to refine the criteria for systemic corticosteroid therapy in those with suspecting TB and initiating a diagnostic evalu- evidence of physiologic impairment. Level of ation, to utilize resources in a cost-effective evidence, low; benefit, substantial; grade of recom- manner and to improve patient and caregiver mendation, B adherence to diagnostic recommendations. Level of evidence, expert opinion; benefit, substan- tial; grade of recommendation, E/A Cough: Occupational and Environmental 4. In populations at increased risk of becom- Considerations25 ing infected with TB and transmitting it to 1. In every patient with cough, when taking a others by cough (eg, those persons in prisons medical history, ask about occupational and and nursing homes), special measures to pre- environmental causes. Level of evidence, expert vent outbreaks must be made by public health opinion; benefit, substantial; grade of recommenda- agencies to screen for new cases, maintain sur- tions, E/A veillance of existing populations, and establish 2. In every patient with cough who has po- effective diagnostic and treatment programs tentially significant exposures to suspicious en- early in the evaluation. Level of evidence, good; vironmental or occupational causes, determine benefit, substantial; grade of recommendation, A the relationship of these occupational and envi- 5. In patients with unexplained chronic cough ronmental factors to confirm or refute their who have resided in areas of endemic infection role in cough and to modify or eliminate expo- with fungi or parasites, a diagnostic evaluation for sure to the relevant agents. Level of evidence, these pathogens should be undertaken when expert opinion; benefit, substantial; grade of recom- more common causes of cough have been ruled mendations, E/A out. Level of evidence, low; benefit, substantial; grade 3. Because outdoor environmental pollution of recommendation, B and occupational exposures can be important Peritoneal Dialysis and Cough27 factors in causing cough, physicians should play a role in developing and supporting enforce- 1. In patients receiving long-term peritoneal able standards for safe workplace and outdoor dialysis with cough, evaluate the patient for the air pollution exposure limits. Level of evidence, potential causes with increased prevalence in expert opinion; benefit, substantial; grade of recom- this population such as GERD, ACE inhibitors, mendations, E/A pulmonary edema, asthma that may be exacer- ␤ 4. In patients with a high suspicion of cough bated by -adrenergic-blocking medications, due to environmental or occupational expo- and infection. Level of evidence, expert opinion; sures, consider referring the patient to a benefit, substantial; grade of recommendation, E/A specialist in this area or consult evidence- Cough in the Immunocompromised Host28 based guidelines. Level of evidence, expert opin- ion; net benefit, substantial; grade of recommen- 1. In patients with immune deficiency, the dation, E/A initial diagnostic algorithm for patients with www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 17S acute, subacute, and chronic cough is the same Unexplained (Idiopathic) Cough30 as that for immunocompetent persons, taking 1. The diagnosis of unexplained (idiopathic) into account an expanded list of differential cough is a diagnosis of exclusion. It should not diagnoses that considers the type and severity be made until a thorough diagnostic evaluation of immune defect and geographic factors. Level is performed, specific and appropriate treat- of evidence, expert opinion; benefit, substantial; ment (according to the management protocols grade of recommendation, E/A that have performed the best in the literature) -In HIV-infected patients, CD4؉ lympho .2 has been tried and has failed, and uncommon cyte counts should be used in constructing the causes have been ruled out. Level of evidence, list of differential diagnostic possibilities poten- expert opinion; benefit, substantial; grade of recom- tially causing cough. Level of evidence, low; ben- mendation, E/A efit, substantial; grade of recommendation, B ؉ 3. HIV-infected patients with CD4 lympho- An Empiric Integrative Approach to the < ␮ cyte counts of 200 cells/ L or those patients Management of Cough3 with counts of > 200 cells/␮L with unexplained fever, weight loss, or thrush who have unex- 1. In patients with cough, the starting point plained cough should be suspected of having is the medical history and physical examination. Pneumocystis pneumonia, tuberculosis, and Although the timing and characteristics of the other opportunistic infections, and should be cough are of little diagnostic value, the medical evaluated accordingly. Level of evidence, low; history is important to determine whether the benefit, substantial; grade of recommendation, B patient is receiving an ACE inhibitor, is a smoker, or has evidence of a serious life-threat- ening or systemic disease. Level of evidence, Uncommon Causes of Cough29 expert opinion; benefit, substantial; grade of recom- 1. In patients with chronic cough, uncommon mendation, E/A causes should be considered when cough per- 2. In patients with an acute cough, first de- sists after evaluation for common causes and termine whether the acute cough is a reflection when the diagnostic evaluation suggests that an of a serious illness such as pneumonia or pul- uncommon cause, pulmonary as well as ex- monary embolism, or, as is usually the case, a trapulmonary (see Table 1 in section 28), may manifestation of a non-life-threatening disease be contributing. Level of evidence, low; benefit, such as a respiratory tract infection (eg, com- substantial; grade of recommendation, B mon cold or lower respiratory tract infection), 2. In patients with chronic cough, until un- an exacerbation of a preexisting condition (eg, common causes that potentially may be contrib- COPD, UACS, asthma, or bronchiectasis), or an uting to the patient’s cough have been ruled environmental or occupational exposure to out, the diagnosis of unexplained cough should some noxious or irritating agent (eg, allergic or not be made. Level of evidence, low; benefit, irritant-induced rhinitis). Level of evidence, ex- substantial; grade of recommendation, B pert opinion; benefit, substantial; grade of recom- 3. If cough persists after consideration of the mendation, E/A most common causes, perform a chest CT scan 3. In patients with a subacute cough, first and, if necessary, a bronchoscopic evaluation. determine whether it is a postinfectious cough Level of evidence, low; benefit, substantial; grade of or not. If it is postinfectious, determine whether recommendation, B it is a result of UACS, transient bronchial hy- 4. In patients who present with abrupt onset of perresponsiveness, asthma, pertussis, or an cough, consider the possibility of an airway for- acute exacerbation of chronic bronchitis. If it is eign body. Level of evidence, low; benefit, substantial; noninfectious, manage the cough the same way grade of recommendation, B as chronic cough. Level of evidence, expert opin- 5. In patients with unexplained cough, eval- ion; benefit, substantial; grade of recommendation, uate the possibility of drug-induced cough. E/A Level of evidence, low; benefit, substantial; grade of 4a. In patients with chronic cough, systemat- recommendation, B ically direct empiric treatment at the most com- 6. In patients with unexplained cough, con- mon causes of cough (ie, UACS, asthma, NAEB, sider a therapeutic trial of withdrawing the and GERD). Level of evidence, low; benefit, sub- drug that is suspected to cause the cough. Level stantial; grade of recommendation, B of evidence, low; benefit, substantial; grade of rec- 4b. In patients with chronic cough, therapy ommendation, B should be given in sequential and additive steps

18S Diagnosis and Management of Cough: ACCP Guidelines because more than one cause of cough may be and asthma or NAEB, treatment for GERD present. Level of evidence, low; benefit, substantial; should be instituted next. Level of evidence, low; grade of recommendation, B benefit, substantial; grade of recommendation, B 5. Patients with a chronic cough who smoke 11. In patients with cough whose condition should be counseled and assisted with smoking remains undiagnosed after all of the above has cessation. Level of evidence, low; benefit, substan- been done, referral to a cough specialist is indi- tial; grade of recommendation, B cated. Level of evidence, expert opinion; benefit, 6. In a patient with cough who is receiving an substantial; grade of recommendation, E/A ACE inhibitor, therapy with the drug should be stopped and the drug should be replaced. Level Assessing Cough Severity and Efficacy of Therapy of evidence, low; benefit, substantial; grade of rec- in Clinical Research31 ommendation, B 7. In patients with chronic cough, initial em- 1. In patients with chronic cough, to opti- piric treatment should begin with an oral first- mally evaluate the efficacy of cough-modifying generation A/D. Level of evidence, low; benefit, agents, investigators should use both subjective substantial; grade of recommendation, B and objective methods because they have the 8a. In patients whose chronic cough persists potential to measure different things. A pa- after treatment for UACS, the possibility that tient’s subjective response is likely to be the asthma is the cause of cough should be worked only one that measures the impact of the inten- up next. The medical history is sometimes sug- sity of cough. Level of evidence, expert opinion; gestive, but is not reliable in either ruling in or benefit, substantial; grade of recommendation, E/A ruling out asthma. Therefore, ideally, broncho- 2. In patients with chronic cough, with re- provocation challenge (BPC), if spirometry does spect to subjective methods, it is recommended not indicate reversible airflow obstruction, that a valid and reliable cough-specific health- should be performed in the evaluation for related quality-of-life instrument be utilized. asthma as a cause of cough. In the absence of Level of evidence, fair; benefit, substantial; grade of the availability of BPC, an empiric trial of recommendation, A antiasthma therapy should be administered (see 3. When assessing patients with chronic section on the treatment of asthma in this cough, even though visual analog scales have guideline). Level of evidence, low; benefit, substan- not been psychometrically tested, they are rec- tial; grade of recommendation, B ommended because they are commonly used 8b. In patients with chronic cough, in whom and valid, and they are likely to yield different the diagnoses of UACS and asthma have been but complementary results to cough-specific eliminated or treated without the elimination of health-related quality-of-life instruments. Level cough, NAEB should be considered next with a of evidence, low; benefit, intermediate; grade of properly performed induced sputum test for recommendation, C eosinophils. If a properly performed induced 4. When assessing patients with chronic cough, sputum test to determine whether eosinophilic because health-related quality-of-life instruments bronchitis is present cannot be performed, an have been psychometrically tested and visual an- empiric trial of corticosteroids should be the alog scales have not, the cough-specific health- next step. Level of evidence, low; benefit, substan- related quality-of-life instruments are recom- tial; grade of recommendation, B mended as the primary subjective outcome 9. In the majority of patients with suspected measure. Level of evidence, fair; benefit, intermedi- cough due to asthma, ideally, before starting an ate; grade of recommendation, B oral corticosteroid regimen, a BPC should be 5. In patients with chronic cough, with re- performed and, if the result is positive, some spect to objective methods, tussigenic chal- combination therapy of ICSs, inhaled ␤-ago- lenges should be used before and after the nists, or oral leukotriene inhibitors should be intervention to assess the effect of therapy on administered. A limited trial of oral corticoste- cough sensitivity only in disease states in which roids, however, should be administered in some cough reflex sensitivity is known to be height- patients who are suspected of having asthma- ened. Level of evidence, low; benefit, small/weak; induced cough before eliminating the diagnosis grade of recommendation, C from further consideration. Level of evidence, 6. In patients with chronic cough, because low; benefit, substantial; grade of recommendation, B the act of coughing has the potential to trauma- 10. In patients whose cough responds only tize the upper airway (eg, the vocal cords), partially or not at all to interventions for UACS assessing the presence of upper airway edema www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 19S before and after therapy with flow-volume coughing. Level of evidence, good; benefit, sub- loops is useful. Level of evidence, low; benefit, stantial; grade of recommendation, A intermediate; grade of recommendation, C 9. In patients with acute cough due to the 7. In patients undergoing treatment for common cold, preparations containing zinc are chronic cough, cough counting over 24 h is not recommended. Level of evidence, good; ben- recommended with a computerized methodol- efit, none; grade of recommendation, D ogy that is reliable and accurate, noninvasive 10. In patients with acute cough due to the and portable, and easy to use in unattended, common cold, over the counter combination ambulatory, real-life settings within a patient’s cold medications, with the exception of an older home environment. Level of evidence, low; bene- antihistamine-decongestant, are not recom- fit, intermediate; grade of recommendation, C mended until randomized controlled trials prove they are effective cough suppressants. Cough Suppressant and Pharmacologic Protussive Level of evidence, fair; benefit, none; grade of Therapy32 recommendation, D 11. In patients with acute or chronic cough 1. In patients with chronic bronchitis, not due to asthma, albuterol is not recom- agents that have been shown to alter mucus mended. Level of evidence, good; benefit, none; characteristics are not recommended for cough grade of recommendation, D suppression. Level of evidence, good; benefit, 12. In patients with neuromuscular impair- none; grade of recommendation, D ment, protussive pharmacologic agents are in- 2. In patients with cough due to upper respi- effective and should not be prescribed. Level of ratory infection (URI) or chronic bronchitis, the evidence, good; benefit, none; grade of recommen- only inhaled anticholinergic agent that is rec- dation, D ommended for cough suppression is ipratro- 13. In patients with bronchitis, hypertonic pium bromide. Level of evidence, fair; benefit, saline solution and erdosteine are recom- substantial; grade of recommendation, A mended on a short-term basis to increase cough 3. In patients with chronic or acute bronchi- clearance. Level of evidence, good; benefit, sub- tis, peripheral cough suppressants, such as levo- stantial; grade of recommendation, A dropropizine and moguisteine, are recom- 14. In adult patients with CF, amiloride is mended for the short-term symptomatic relief recommended to increase cough clearance. of coughing. Level of evidence, good; benefit, Level of evidence, good; benefit, substantial; grade substantial; grade of recommendation, A of recommendation, A 4. In patients with cough due to URI, periph- 15. In adult patients with CF, while recombi- eral cough suppressants have limited efficacy nant DNase does improve spirometry it is not and are not recommended for this use. Level of recommended to increase cough clearance. evidence, good; benefit, none; grade of recommen- Level of evidence, good; benefit, none; grade of dation, D recommendation, D 5. In patients with chronic bronchitis, central cough suppressants, such as codeine and dex- Nonpharmacologic Airway Clearance Therapies33 tromethorphan, are recommended for the short-term symptomatic relief of coughing. 1. In patients with CF, chest physiotherapy is Level of evidence, fair; benefit, intermediate; grade recommended as an effective technique to in- of recommendation, B crease mucus clearance, but the effects of each 6. In patients with cough due to URI, central treatment are relatively modest and the long- cough suppressants have limited efficacy for term benefits unproven. Level of evidence, fair; symptomatic relief and are not recommended benefit, small; grade of recommendation, C for this use. Level of evidence, good; benefit, none; 2. In patients with expiratory muscle weak- grade of recommendation, D ness, manually assisted cough should be consid- 7. In patients with chronic or acute cough ered to reduce the incidence of respiratory requiring symptomatic relief, drugs that affect complications. Level of evidence, low; benefit, the efferent limb of the cough reflex are not small; grade of recommendation, C recommended. Level of evidence, low; benefit, 3. In persons with airflow obstruction caused none; grade of recommendation, D by disorders like COPD, manually assisted cough 8. In patients requiring intubation during may be detrimental and should not be used. Level general anesthesia, the use of neuromuscular of evidence, low; benefit, negative; grade of recommen- blocking agents is recommended to suppress dation, D

20S Diagnosis and Management of Cough: ACCP Guidelines 4. In patients with COPD and CF, huffing spirometry (if age appropriate). Level of evi- should be taught as an adjunct to other methods dence, expert opinion; benefit, intermediate; grade of sputum clearance. Level of evidence, low; ben- of recommendation, E/B efit, small; grade of recommendation, C 3. In children with specific cough, further 5. In patients with CF, autogenic drainage investigations may be warranted, except when should be taught as an adjunct to postural asthma is the etiologic factor. Level of evidence, drainage as a method to clear sputum because it expert opinion; benefit, intermediate; grade of rec- has the advantage of being performed without ommendation, E/B assistance and in one position. Level of evidence, 4. Children with chronic productive purulent low; benefit, small; grade of recommendation, C cough should always be investigated to docu- 6. In patients with neuromuscular weakness ment the presence or absence of bronchiectasis and impaired cough, expiratory muscle training and to identify underlying and treatable causes is recommended to improve peak expiratory such as cystic fibrosis and immune deficiency. pressure, which may have a beneficial effect on Level of evidence, low; benefit, substantial; grade of cough. Level of evidence, expert opinion; benefit, recommendation, B small; grade of recommendation, E/C 5. In children with chronic cough, the etiol- 7. In patients with CF, positive expiratory ogy should be defined and treatment should be pressure is recommended over conventional etiologically based. Level of evidence, expert opin- chest physiotherapy because it is approximately ion; benefit, substantial; grade of recommendation, as effective as chest physiotherapy, and is inex- E/A pensive, safe, and can be self-administered. 6. In children with nonspecific cough, cough Level of evidence, fair; benefit, intermediate; grade may spontaneously resolve, but children should of recommendation, B be reevaluated for the emergence of specific 8. In patients with CF, devices designed to etiologic pointers (see Table 1 in Chang and oscillate gas in the airway, either directly or by Glomb4). Level of evidence, low; benefit, substan- compressing the chest wall, can be considered tial; grade of recommendation, B as an alternative to chest physiotherapy. Level of 7. In children with nonspecific cough and risk evidence, low; benefit, conflicting; grade of recom- factors for asthma, a short trial (ie, 2 to 4 weeks) of mendation, I beclomethasone, 400 ␮g/d, or the equivalent dos- 9. In patients with neuromuscular disease age with budesonide may be warranted. How- with impaired cough, mechanical cough assist ever, most children with nonspecific cough do not devices are recommended to prevent respira- have asthma. In any case, these children should tory complications. Level of evidence, low; benefit, always be reevaluated in 2 to 4 weeks. Level of intermediate; grade of recommendation, C evidence, fair; benefit, intermediate; grade of recom- 10. The effect of nonpharmacologic airway mendation, B clearance techniques on long-term outcomes 8. In children who have started therapy with a such as health-related quality of life and rates of medication, if the cough does not resolve during exacerbations, hospitalizations, and mortality is the medication trial within the expected response not known at this time. The committee recom- time, the medication should be withdrawn and mends that future investigations measure these other diagnoses considered. Level of evidence, low; outcomes in patients with CF, and in other benefit, intermediate; grade of recommendation, C populations with bronchiectasis, COPD, and 9. In children with cough, cough suppressants neuromuscular diseases. Level of evidence, expert and other over-the-counter cough medicines opinion; benefit, substantial; grade of recommenda- should not be used as patients, especially young tion, E/A children, may experience significant morbidity and mortality. Level of evidence, good; benefit, none; grade of recommendation, D Guidelines for Evaluating Cough in Pediatrics4 10. In children with nonspecific cough, parental 1. Children with chronic cough require expectations should be determined, and the specific careful and systematic evaluation for the pres- concerns of the parents should be sought and ad- ence of specific diagnostic indicators. Level of dressed. Level of evidence, low; benefit, intermediate; evidence, expert opinion; benefit, substantial; grade grade of recommendation, E/B of recommendation, E/A 11. In all children with cough, exacerbating fac- 2. Children with chronic cough should un- tors such as exposure to tobacco smoke should be dergo, as a minimum, a chest radiograph and determined and interventional options for the ces- www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 21S sation of exposure advised or initiated. Level of evidence-based clinical practice guidelines. Chest 2006; evidence, low; benefit, substantial; grade of recommenda- 129(suppl):222S–231S tion, B 4 Chang AB, Glomb WB. Guidelines for evaluating cough in pediatrics: ACCP evidence-based clinical practice guidelines. 12. Children should be managed according to Chest 2006; 129(suppl):260S–283S the studies and guidelines for children (when 5 McCrory DC, Zelman Lewis S. Methodology and grading of available), because etiologic factors and treat- the evidence for the diagnosis and management of cough: ments in children are sometimes different from ACCP evidence-based clinical practice guidelines. Chest those in adults. Level of evidence, low; benefit, 2006; 129(suppl):28S–32S 6 Canning BJ. Anatomy and neurophysiology of the cough substantial; grade of recommendation, B < reflex: ACCP evidence-based clinical practice guidelines. 13. In children 14 years of age with Chest 2006; 129(suppl):33S–47S chronic cough, when pediatric-specific cough 7 McCool FD. Global physiology and pathophysiology of recommendations are unavailable, adult recom- cough: ACCP evidence-based clinical practice guidelines. mendations should be used with caution. Level Chest 2006; 129(suppl):48S–53S 8 Irwin RS. Complications of cough: ACCP evidence-based of evidence, expert opinion; benefit, intermediate; clinical practice guidelines. Chest 2006; 129(suppl):54S–58S grade of recommendation, E/B 9 Pratter MR. Overview of common causes of chronic cough: ACCP evidence-based clinical practice guidelines. Chest Potential Future Therapies for the Management of 2006; 129(suppl):59S–62S Cough34 10 Pratter MR. Chronic upper airway cough syndrome second- ary to rhinosinus diseases (previously referred to as postnasal • Currently available cough-suppressant therapy is se- drip syndrome): ACCP evidence-based clinical practice verely limited by a dearth of effective agents and their guidelines. Chest 2006; 129(suppl):63S–71S 11 Pratter MR. Cough and the common cold: ACCP evidence- unacceptable side affects. Several classes of pharmaco- based clinical practice guidelines. Chest 2006; 129(suppl): logic agents are currently under investigation in an 72S–74S attempt to develop clinically useful cough suppression. 12 Dicpinigaitis PV. Chronic cough due to asthma: ACCP evidence-based clinical practice guidelines. Chest 2006; 129(suppl):75S–79S Future Directions in the Clinical Management of 13 Irwin RS. Chronic cough due to gastroesophageal reflux 35 Cough disease: ACCP evidence-based clinical practice guidelines. Chest 2006; 129(suppl):80S–94S 1. As suggested in the various sections of this 14 Braman SS. Chronic cough due to acute bronchitis: ACCP guideline, further research should be con- evidence-based clinical practice guidelines. Chest 2006; ducted to elucidate the mechanisms of the pro- 129(suppl):95S–103S duction of cough in various diseases and condi- 15 Braman SS. Chronic cough due to chronic bronchitis: ACCP tions, the optimal methods of assessment, and evidence-based clinical practice guidelines. Chest 2006; 129(suppl):104S–115S treatment, specific to the suspected cause. Level 16 Brightling CE. Chronic cough due to nonasthmatic eosino- of evidence, expert opinion; benefit, substantial; philic bronchitis: ACCP evidence-based clinical practice strength of recommendation, E/A guidelines. Chest 2006; 129(suppl):116S–121S 2. Research is particularly needed in areas 17 Rosen MJ. Chronic cough due to bronchiectasis: ACCP such as the treatment of postinfectious cough, evidence-based clinical practice guidelines. Chest 2006; 129(suppl):122S–131S the characterization of psychogenic cough, the 18 Brown KK. Chronic cough due to nonbronchiectatic suppu- methods of assessment of cough, and pharma- rative airway disease (bronchiolitis): ACCP evidence-based cotherapy. Level of evidence, expert opinion; ben- clinical practice guidelines. Chest 2006; 129(suppl):132S– efit, substantial; strength of recommendation, E/A 137S 19 Braman SS. Postinfectious cough: ACCP evidence-based ACKNOWLEDGMENT: This clinical practice guideline re- clinical practice guidelines. Chest 2006; 129(suppl):138S– ceived the endorsements of the ATS and the CTS. 146S 20 Kvale PA. Chronic cough due to lung tumors: ACCP evi- dence-based clinical practice guidelines. Chest 2006; 129(suppl):147S–153S References 21 Smith Hammond CA, Goldstein LB. Cough and aspiration of 1 Irwin RS, Boulet LP, Cloutier MM, et al. Managing cough as food and liquids due to oral/pharyngeal dysphagia: ACCP a defense mechanism and as a symptom: a consensus panel evidence-based clinical practice guidelines. Chest 2006; report of the American College of Chest Physicians. Chest 129(suppl):154S–168S 1998; 114(suppl):133S–181S 22 Dicpinigaitis PV. Angiotensin-converting enzyme inhibitor- 2 Rudolph C, Mazur L, Liptak G, et al. Guidelines for evalua- induced cough: ACCP evidence-based clinical practice guide- tion and treatment of gastroesophageal reflux in infants and lines. Chest 2006; 129(suppl):169S–173S children: recommendations of the North American Society 23 Irwin RS, Glomb WB, Chang AB. Habit cough, tic cough, and for Pediatric Gastroenterology and Nutrition. J Pediatr Gas- psychogenic cough in adult and pediatric populations: ACCP troenterol Nutr 2001; 32(suppl):S1–S31 evidence-based clinical practice guidelines. Chest 2006; 3 Pratter MR, Brightling CE, Boulet LP, et al. An empiric 129(suppl):174S–179S integrative approach to the management of cough: ACCP 24 Brown KK. Chronic cough due to chronic interstitial pulmo-

22S Diagnosis and Management of Cough: ACCP Guidelines nary diseases: ACCP evidence-based clinical practice guide- dence-based clinical practice guidelines. Chest 2006; lines. Chest 2006; 129(suppl):180S–185S 129(suppl):220S–221S 25 Tarlo SM. Cough: occupational and environmental consider- 31 Irwin RS. Assessing cough severity and efficacy of therapy in ations: ACCP evidence-based clinical practice guidelines. clinical research: ACCP evidence-based clinical practice Chest 2006; 129(suppl):186S–196S guidelines. (Richard S. Irwin, MD, FCCP) Chest 2006; 26 Rosen MJ. Chronic cough due to tuberculosis and other 129(suppl):232S–237S infections: ACCP evidence-based clinical practice guidelines. 32 Bolser DC. Cough suppressant and pharmacologic protussive Chest 2006; 129(suppl):197S–201S therapy: ACCP evidence-based clinical practice guidelines. 27 Tarlo SM. Peritoneal dialysis and cough: ACCP evidence- Chest 2006; 129(suppl):238S–249S based clinical practice guidelines. Chest 2006; 129(suppl): 33 McCool FD, Rosen MJ. Nonpharmacologic airway clearance 202S–203S therapies: ACCP evidence-based clinical practice guidelines. 28 Rosen MJ. Cough in the immunocompromised host: ACCP Chest 2006; 129(suppl):250S–259S evidence-based clinical practice guidelines. Chest 2006; 34 Dicpinigaitis PV. Potential future therapies for the manage- 129(suppl):204S–205S ment of cough: ACCP evidence-based clinical practice guide- 29 Prakash UBS. Uncommon causes of cough: ACCP evidence- lines. Chest 2006; 129(suppl):284S–286S based clinical practice guidelines. Chest 2006; 129(suppl): 35 Boulet LP Future directions in the clinical management of 206S–219S cough: ACCP evidence-based clinical practice guidelines. 30 Pratter MR. Unexplained (idiopathic) cough: ACCP evi- Chest 2006; 129(suppl):287S–292S

www.chestjournal.org CHEST / 129/1/JANUARY, 2006 SUPPLEMENT 23S