| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea.
Correspondence to: Dong-Chull Choi, MD, PhD, Associate Professor, Division of Allergy, Department of Medicine, Samsung Medical Center, 50 Ilwon-dong, Kangnam-Gu, Seoul, Korea 135–710; e-mail: dcchoi{at}smc.samsung.co.kr
Abstract
Study objectives: Cough lasting > 3 weeks has been defined as chronic cough. However, it might be more persuasive to divide cough into subacute, lasting 3 to 8 weeks, and chronic, lasting > 8 weeks. We evaluated the causes and clinical courses of subacute cough, and the value of the bronchoprovocation test and induced sputum examination (IS).
Methods: Nonsmoking patients with cough of 3 to 8 weeks duration were enrolled into the study. Patients with dyspnea, basal FEV1 of < 70% predicted, abnormal findings on a plain chest radiograph or physical examination were excluded. We prescribed an antihistamine-decongestant for patients who were suspected to have postinfectious cough or postnasal drip. If patients had positive results on a bronchoprovocation test or IS, therapy with inhaled corticosteroids was substituted according to an algorithmic approach.
Results: One hundred eighty-four patients (77 men and 107 women) were evaluated; the mean age of the study group was 47.5 years. Eighty-nine of 184 patients had postinfectious cough. Cough resolved without treatment in 62 patients. Twenty-nine of 43 patients with positive bronchoprovocation test results had cough-variant asthma.
Conclusions: Postinfectious cough was the most common cause of subacute cough. The spontaneous resolution of cough was frequent in patients with subacute cough. Unless asthma was strongly suspected, the performance of the bronchoprovocation test could be delayed until after empirical treatment had been administered.
Key Words: cough • diagnosis • treatment
A cough lasting for > 3 weeks has generally been accepted as a chronic cough.1 It has been suggested that some patients with a chronic cough should not be treated symptomatically, but first should be evaluated for the causes of the cough. It has been widely accepted for > 20 years that the causes of chronic cough can be proven in 88 to 100% of patients with a diagnostic algorithm regarding the causes of chronic cough.23 The common causes of chronic cough are postnasal drip (PND), cough-variant asthma, and gastroesophageal reflux disease (GERD).456 The recent report7 regarding the algorithmic approach in patients with chronic cough lasting for > 8 weeks confirmed the common causes of chronic cough.
We recognized that many patients whose chief complaint was a cough lasting for > 3 weeks spontaneously improved without treatment. Spontaneously resolving cough was exceedingly rare in patients who experienced a long duration of cough, such as several months or years. Thus, patients with cough of a relatively short duration must be regarded separately from patients with a cough of a longer duration.
Irwin and Madison3 subdivided cough in terms of duration into subacute cough, lasting from 3 to 8 weeks, and chronic cough, lasting for > 8 weeks. They listed postinfectious cough, sinusitis, PND, and bronchial asthma as common causes of subacute cough, and recommended that patients with subacute cough should first be treated symptomatically. Pratter and colleagues8 reported that when patients were prescribed antihistamines and decongestants, significant improvements were observed in > 55% of the patients with chronic cough. However, the data regarding the causes and clinical courses of chronic cough with shorter duration remained scarce. Therefore, our intention was to prospectively evaluate the causes and clinical courses of subacute cough, and to determine whether any clinical differences exist between subacute and chronic cough.
Materials and Methods
Patient Selection
In this study, we defined subacute cough as cough lasting 3 to 8 weeks, and chronic cough as cough lasting > 8 weeks. From April 2001 to August 2003, patients with a chief complaint of subacute cough were enrolled in the study. The following were exclusion criteria: (1) current or past dyspnea; (2) wheezing or rales on physical examination; (3) decreased lung function (FEV1 of < 70% predicted; (4) abnormal findings on plain chest radiograph; (5) current smoking; and (6) use of an angiotensin-converting-enzyme inhibitor within the past 2 months. The study was approved by the research ethics committee, and all subjects gave a written informed consent.
Methacholine Responsiveness
A methacholine bronchial provocation test (MBPT) was performed according to American Thoracic Society guidelines,9 using a spirometer (
max 2130; SensorMedics; Yorba Linda, CA). Methacholine chloride (Sigma-Aldrich; St. Louis, MO) diluted with physiologic saline solution was inhaled using a nebulizer (Pari-Master 084G7300; PulmoMed; Nanterre Cedex, France) as a five-breath dosimeter protocol. The provocative concentration of methacholine causing a 20% fall in FEV1 (PC20) was adopted as the marker for bronchial hyperresponsiveness (BHR). BHR was defined as a PC20 of 
8 mg/mL.
Induced Sputum Examination
Sputum was induced by the inhalation of hypertonic saline solution. After spirometric values were recorded and 15 min before sputum induction, subjects inhaled 200 µg of salbutamol via a metered-dose inhaler. Then a 4.5% saline solution was administered at room temperature for 5 min using a nebulizer (Pari-Master 084G7300; PulmoMed). Subjects were encouraged to cough up sputum into a sterile container after rinsing their mouth and throat, and this process was repeated three times. After measuring the weight of the sputum, an aliquot was treated with 0.1% dithiothreitol and was kept at 37°C for 15 min. The mixture was further diluted with phosphate-buffered saline solution and was centrifuged at 2,000 revolutions per minute for 5 min. Cell pellets were resuspended. Slides were prepared by using a Cytospin instrument (Shandon Scientific Ltd; Shandon, UK) and were stained (Diff-Quick; American Scientific Products; Chicago, IL) for differential cell counts. We regarded the prepared sample as being inadequate if squamous cells accounted for > 30% of cells. At least 300 nonsquamous cells were counted in each slide. Differential counts of eosinophils, neutrophils, macrophages, lymphocytes, and epithelial cells were counted. Eosinophilic inflammation was defined as an eosinophil percentage of 
3%.
Study Design
Enrolled patients were evaluated and treated prospectively according to an algorithm (Fig 1
). The initial evaluation was conducted at the first visit; a questionnaire on the duration, frequency, severity, aggravating factors, and other combined symptoms such as PND, nasal itching, sneezing, rhinorrhea, and gastroesophageal reflux-related symptoms. Pulmonary function studies, plain chest radiographs, MBPT, and induced sputum examination (IS) were also performed in all enrolled subjects. We then decided on a series of sequential diagnostic therapeutic steps that evolved into the following algorithm.
|
For patients who exhibited no suggestive symptoms of postinfectous cough or PND, and whose conditions did not improve satisfactorily, the results of MBPT and IS were reviewed. An inhaled corticosteroid (ICS) with or without a long-acting ß2-agonist (Symbicort turbuhaler, 160/4.5 µg/puff, or Pulmicort turbuhaler, 200 µg/puff bid; AstraZeneca; Wilmington, DE) was prescribed for 2 weeks for patients with a positive result. In patients whose cough improved after therapy with an ICS, bronchial asthma was diagnosed in those with positive results from the MBPT, and eosinophilic bronchitis was diagnosed in those with positive results from the IS and negative results from the MBPT.
Patients with persistent cough and negative results from both the MBPT and IS proceeded to the next steps, which included 24-h pH monitoring, CT scan of the lung, or bronchoscopy. Twenty-four–hour pH monitoring was performed using a catheter (Digitrapper MK Gold III; Synectics Medical AB; Stockholm, Sweden). GERD was defined when the DeMeester composite score was > 14.7 or when the percentage of the total time spent with a pH of < 4 was > 4.5%.10
Results
Patient Characteristics
Two hundred forty-three patients visited our clinic due to subacute cough during the study period. Among those patients, 184 patients fulfilled the inclusion criteria. The mean (± SD) age was 47.5 ± 14.4 years (age range, 17 to 76 years). The ratio of men to women was 77:107. The mean basal FEV1 was 100.6 ± 13.% predicted, and the mean FEV1/FVC ratio was 85.8 ± 7.4%.
Frequency of the Causes of Subacute Cough
Eighty-nine of 184 patients (48.4%) had postinfectious cough, 61 of 184 patients (33.2%) had PND, and 29 of 184 patients (15.8%) had cough-variant asthma. No patients had GERD (Fig 2
). If the causes of subacute cough were analyzed only in the 62 patients with spontaneously resolving cough, 46 patients (74.3%) had postinfectious cough.
|
Clinical Courses and Therapeutic Results of Subacute Cough
In 62 of 184 patients (33.7%), cough improved spontaneously without any treatment. We prescribed an antihistamine-decongestant for 3 weeks to 97 of 122 patients whose cough was persistent. Seventy-two patients were suspected to have symptoms or signs of PND, and 25 patients had postinfectious cough. After the 3-week treatment period, 68 patients (70.1%) showed significant improvement
(Fig 3
).
|
Positive Predictive Value of MBPT and IS
Twenty-nine of 43 patients who had BHR and responded to an ICS were diagnosed with cough-variant asthma. Therefore, the positive predictive value (PPV) of the MBPT was 67.4%. When subjects were limited to 122 patients who had persistent cough at the second visit, cough-variant asthma was diagnosed in 26 of 35 patients who had BHR. In this case, the PPV of the MBPT was higher at 74.3%.
Sputum eosinophilia of > 3.0% was present in 43 patients with subacute cough. The PPV of the IS was 55.8% because cough-variant asthma was diagnosed in 24 of 43 patients. When only the patients with persistent cough were analyzed, cough-variant asthma was diagnosed in 22 of 38 patients with eosinophilic inflammation. The PPV was 57.9%. The PPV of cough-variant asthma was as high as 96.0% when we evaluated patients who had undergone both the MBPT and IS at the same time.
Discussion
We analyzed prospectively the causes of subacute cough, and our algorithmic approach with initial use of an antihistamine-decongestant was effective. During the study period, 529 patients complained of subacute cough and chronic cough; 184 patients (34.8%) were regarded as having subacute cough. In patients with subacute cough, 62 improved without medication at the second visit, and the cough was not aggravated again during the study period. The percentage of self-limited cough was higher in the subacute cough group compared to that of the chronic cough group (11.9%; 103 of 529 patients). Sixty percent of spontaneously resolving coughs belonged to patients in the subacute cough group. These data showed that the conditions of many patients, especially those with subacute cough, could improve without treatment.
Although PND, cough-variant asthma, and GERD have been regarded as frequent causes of chronic cough in previous reports, the most common cause of subacute cough in our study was postinfectious cough. We defined postinfectious cough as a cough that began with symptoms related to the common cold without any abnormal findings on plain chest radiograph and did not meet the criteria for PND. Eighty-nine patients with subacute cough (48.4%) had postinfectious cough, 61 patients had PND (33.2%), and 29 patients had cough-variant asthma (15.8%). In contrast, notably only 4.6% of coughs lasting > 8 weeks (data not shown) were caused by postinfectious cough in our experience. To summarize, postinfectious cough was the most common cause of subacute cough and had a higher rate of spontaneous resolution. Therefore, when the duration of cough is not long, a diagnosis of postinfectious cough should be considered and the cough could improve without the use of medications.
The reason to differentiate postinfectious cough from PND in this study was to reduce the use of unnecessary drugs, since postinfectious cough has a high possibility of spontaneous improvement. If patients had one of the following criteria, we diagnosed PND: (1) current or past PND; (2) cobblestone appearance on posterior pharyngeal wall on physical examination; and (3) if patients had more than two of following symptoms: rhinorrhea; sneezing; nasal itching; and nasal obstruction. If the patients complained of only a sense of a foreign body in the pharynx, and had a clear history of cough starting with a common cold, we assumed it to be a postinfectious cough. We also applied the same criteria in patients who exhibited persistent cough and restrictively diagnosed specific diseases in the patients who responded to therapy with an antihistamine-decongestant. We suggested that the incidence of postinfectious cough was increased proportionally when we limited the subjects to a subacute period and spontaneously resolving cough. Therefore, it is important to acknowledge that the determination of the cause of subacute cough is necessary in terms of effectiveness and cost.
It was interesting that the incidence of GERD was low in our study. Chronic cough due to GERD has been reported as the third most common cause of chronic cough in nonsmokers, at 10 to 20%, and up to 40%.7111213 In our study, no patient had a cough due to GERD when we confined the cough to the subacute period. The following explanations are possible: (1) the mean duration of GERD was much longer than that of other diseases, reportedly 13 to 58 months614; (2) we delayed 12-h pH monitoring if the patients with subacute cough did not complain of esophageal symptoms like acid regurgitation or heartburn; or (3) the incidence of GERD itself might be lower in Asians. It has been reported that the prevalence of GERD was 6.6% in Japan.15
Pertussis is considered an important cause of persistent cough in adults. A prevalence for pertussis of 12.4 to 26% has been reported in studies from the United States.16 In this study, we did not routinely check for Bordetella pertussis infection because there were no standard serologic methods for pertussis detection, and polymerase chain reaction had limitations in practice, even though it was sensitive and specific. Moreover, the incidence of pertussis proven by polymerase chain reaction in Korea was only 2.9%.17 It is possible that some pertussis patients might be included as postinfectious cough patients because B pertussis with chronic cough could disappear spontaneously.
The MBPT showed that 43 patients had BHR, but only 29 of 43 patients were confirmed to have cough-variant asthma. The remaining 14 patients consisted of those with PND and postinfectious cough, which could be explained by transient hyperresponsiveness (eight patients improved without treatment, and 6 patients improved after therapy with only an antihistamine-decongestant). In our study, the PPV of the MBPT was 67.4%. Generally, sensitivity is 100% and the PPV is 60 to 82% when the positive criteria is a PC20 of 8 mg/mL.118 The duration of postinfectious cough was much shorter than that of cough-variant asthma when the comparison included patients with a cough of longer duration (Fig 4
). We could infer that when more patients with cough of a longer duration were included, the incidence of cough-variant asthma increased and the PPV of the MBPT might be higher. In contrast, when the cough duration of enrolled patients was relatively short, the incidence of postinfectious cough increased and the PPV of the MBPT was low. In our study, the reason for this low PPV was due not only to the wide positive criteria in PC20, but also, more importantly, to the confinement of subjects to a subacute period of cough. We assume that the discrepancy regarding PPV in previous reports was due to various durations of cough.
|
Footnotes
Abbreviations: BHR = bronchial hyperresponsiveness; GERD = gastroesophageal reflux disease; ICS = inhaled corticosteroid; IS = induced sputum examination; MBPT = methacholine bronchial provocation test; PC20 = provocative concentration of methacholine causing a 20% fall in FEV1; PND = postnasal drip; PPV = positive predictive value
Received for publication October 25, 2005. Accepted for publication December 13, 2005.
References
This article has been cited by other articles:
|
|
 |
|
  M. R. Pratter and W. Abouzgheib "Make The Cough Go Away" Chest, May 1, 2006; 129(5): 1121 - 1122. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
