Antibiotics don't improve cough in acute bronchitis
Acute bronchitis with cough is overwhelmingly often due to viral infection, but that doesn't stop coughers from seeking antibiotics, or their doctors from obligingly prescribing them. Most patients who ask for antibiotics get them, and the millions of excess antibiotic doses worldwide each year are believed to contribute to rising antibiotic resistance. Doctors seem almost helpless to deny patients a seemingly harmless treatment they're so sure they need, and there's always an arguable chance antibiotics might help.
Discolored sputum is often considered evidence by both patients and doctors that an infection is bacterial, and doctors are 3 times more likely to prescribe antibiotics if yellow or green sputum is described with an acute cough. But although people with chronic bronchitis (e.g., COPD) often have bacteria present in their sputum during acute exacerbations, in people with acute cough without chronic respiratory disease, the "green sputum = I need antibiotics" logic does not apply. In otherwise healthy people, green sputum is often present during viral infections.
That said, a Cochrane analysis of randomized trials concluded that antibiotics did lead to about a half-day faster resolution of cough from acute bronchitis, on average, but one in 17 patients also had a significant antibiotic side effect. Over-the-counter antitussive medicines don't seem to work for acute cough.
Considering its status as the #1 stated reason for primary care visits, cough due to acute bronchitis has been poorly studied overall. Carl Llor et al undertook a large-ish randomized trial to answer the question, does discolored sputum really mean antibiotics will help an acute cough get better?
What They Did
416 adults with acute cough (< 1 week) with discolored sputum but no underlying lung disease (COPD, asthma, etc) or recent antibiotic/prednisone use were randomized to receive either amoxicillin/clavulanate (Augmentin) 500/125 mg, ibuprofen 600 mg, or placebo three times daily in a 1:1:1 ratio. Authors closely observed adherence and followed up at ~3 days and ~12 days. Days of frequent coughing, the primary outcome, were recorded in a symptom diary. The physicians were not 100% blinded, because the pills were different appearances, but they were provided in sealed and obscured packages.
What They Found
There was no difference in number of days with frequent coughing between treatment arms. In other words, neither antibiotics (Augmentin) or ibuprofen improved cough in patients with acute bronchitis, all of whom also had discolored sputum.
Patients coughed for about 2 weeks total. Those taking ibuprofen had 9 days of frequent coughing vs. 11 for antibiotics and placebo, and also fewer total days of cough, but this wasn't statistically significant.
Those taking antibiotics had 16 adverse events, including a GI hemorrhage, compared to 7 with ibuprofen (mostly GI upset, no bleeding) and 4 with placebo.
What It Means
This well-done study casts doubt on the previously mentioned Cochrane analysis suggesting antibiotics improve cough, which included only 275 patients total.
GRACE, the largest trial yet testing antibiotics for acute bronchitis with purulent sputum in 2,060 adults and published in February 2013, showed amoxicillin did not reduce cough symptoms or severity. Antibiotics did prevent "new or worsening" symptoms in an absolute 3% of patients. However, with nausea, rash, and GI upset, the number needed to harm was 21. One patient taking amoxicillin developed anaphylaxis.
Unfortunately, these studies tell us nothing about azithromycin, which may be more commonly prescribed in the U.S. today for acute cough from bronchitis than amoxicillin. Authors say they did not select azithromycin because "we wanted to exclude the possibility of clinical improvement due to an anti-inflammatory effect." But do patients care whether they are getting better from an antibacterial or an anti-inflammatory effect of a medicine? And should we, as their doctors?
Azithromycin probably has fewer GI side effects and anaphylaxis than amoxicillin, as well. However, azithromycin was no better than vitamin C at reducing cough from acute bronchitis among 220 patients in a 2002 randomized trial. If we got a larger trial testing azithromycin to confirm those results, combined with its sudden cardiac death risk data and the poor efficacy here from amoxicillin, there just might be enough evidence-based armor for us pathologically patient-pleasing doctors to feel confident saying "no" -- and I mean it this time -- when patients demand antibiotics for acute cough from acute bronchitis and the common cold.
Clinical Takeaway: Consider offering ibuprofen instead of antibiotics as first-line therapy for patients insisting on treatment for cough due to acute bronchitis. NSAIDs might help -- and at minimum, will allow a few days for natural recovery to reassure patients they're getting better without antibiotics.
Carl Llor et al. Efficacy of anti-inflammatory or antibiotic treatment in patients with non-complicated acute bronchitis and discoloured sputum: randomised placebo controlled trial. BMJ 2013; 347:f5762.