Chronic Cough: A practical, Cost Effective Approach
Likely one of the most frustrating clinical problems for both the patient and physician is an unrelenting, undiagnosed cough. There are 4 main reasons why the management is so difficult:
- 1 Chronic cough can be caused by several different pathologies.
- 2 A given patient may have 2 or more pathologies simultaneously.
- 3 Diagnostic tests can be difficult and/or uncomfortable for the patient.
- 4 Even if the pathologies are found, they can be difficult to treat.
A reasonable approach is empirical (and aggressive) treatment of the most likely pathology while observing the patient's response. If the patient does not respond completely, treat the next most likely pathology, and so on. Accompanying signs and symptoms of these disorders may not be obvious or even present - this is a common theme in chronic cough patients. Below are descriptions of these pathologies and their treatments.
Allergic rhinitis and or chronic sinusitis commonly cause cough via post-nasal drip or the naso-bronchil reflex. Uncomplicated Allergic rhinitis should respond well to the combination of antihistamines and nasal steroids. If chronic sinusitis is a consideration, then a minimum three-week course of a broad-spectrum antibiotic (e.g., Augmentin™ or Biaxin™) with good anaerobic coverage should be used along with nasal steroids and decongestants. A short burst of prednisone (40mg qD x 5d) may also be used when antibiotics are initiated. An option is to obtain a screening sinus CT to make the diagnosis - plain sinus films are not very sensitive and may be an obsolete procedure.
Chronic cough may be the only manifestation of gastroesophageal reflux (GERD). When treating empirically, it is necessary to use an ample dose of a proton-pump inhibitor (e.g. Prevacid™ 30mg qD or BID) for a minimum of 6 weeks. H2-blockers and/or antacids are frequently ineffective in treating the respiratory symptoms of GERD, and thus are not appropriate for an empiric trial. If the patient responds, the dose should be titrated down if possible. Another approach is to do a 24-hour gastric pH probe looking for drops in pH in coinciding with episodes of cough. This is an uncomfortable and expensive procedure.
Cough-variant asthma manifests as a cough without the typical wheezing, shortness of breath, etc. This is sometimes a diagnosis of exclusion, because lung function testing does not always show abnormalities. It is worth performing spirometry, a quick and inexpensive test, because finding significant obstruction helps seal this particular diagnosis. Empiric therapy may include a trial of prednisone (e.g. 40mg qD x 7 - 10 days, then taper) or a trial with a very potent inhled steroid (Flovent™ 220 2 puffs BID x 2-3 weeks). If they patient responds they should be maintained on the lowest effective inhaled steroid dose, perhaps with the addition of a long acting bronchodilator (e.g., Serevent™) or Leukotriene modifier (e.g., Singulair™).
A Post-viral cough can sometimes last for months after infection resulting in prolonged bronchial hyperreactivity. This type of cough, which can be thought of as "temporary asthma," can usually be managed with inhaled steroids and/or Atrovent.™
Of course this empiric approach assumes that there is little suspicion of chronic lung disease (TB, bronchiectasis, interstitial disease, smoking), endobronchial lesions, CHF, or a cough related to ACE inhibitor therapy. These entities should be assessed by history, PE and radiological studies - at least a CXR should be done at some point in all these patients. Lastly, a different approach might be used in patients who do not tolerate medications well (e.g., an elderly patient or one with multiple drug allergies). In these patients it might be advisable to use diagnostic tests first before beginning empiric therapy.