CPD | Coughs (of the acute kind) and antimicrobial prescribing

This is a new guideline  from NICE on when to prescribe antibiotics for coughs. It only covers those with acute coughs and doesn't include COPD patients (they are covered in a separate guideline).

I was surprised by some of the evidence base here and what may work and what may not work, so I will cover some of this.

The only real learning points for me otherwise were:

  • Coughs normally last 3-4/52 (I would probably have given a shorter estimate than this).
  • A 5 day course of antibiotics is recommended. In adults, doxycycline should be used first line, though amoxicillin, clarithromycin and erythromycin can be used too.

Advice to give patients

The cough is likely to last 3-4/52. Most are viral and self-limiting.

Seek further advice if the cough gets worse rapidly or significantly, if they become systemically seriously unwell or if they aren't better by 3-4/52.

There is very limited evidence on what self-care treatments work. Some have limited evidence of efficacy, but it is hard to know what clinical benefit this would bring. Some drugs also don't have enough information on their safety profile to recommend them.

What may work? The following have limited evidence of some benefit:

  • honey (if > 1).
  • pelargonium (a herbal medicine for the over 12s).
  • OTC cough medicines with the expectorant guaifenesin (in the over 12s - the clinical benefit of this is unclear).
  • OTC cough medicines with cough supressants other than codeine. Pholcodine can't be used in the under 6s and dextromethorphan can't be used in the under 12s.

What shouldn't we advise or offer?

  • inhaled or oral steroids (unless they have an underlying disorder like asthma).
  • inhaled or oral bronchodilators (unless they have an underlying disorder like asthma).
  • mucolytics (there isn't enough evidence of benefit).
  • antihistamines, decongestants or codeine. There is limited evidence that these aren't helpful in coughs.

Acute Bronchitis

Most acute bronchitis is viral, but about 15% will be bacterial.

We shouldn't routinely offer antibiotics to people with acute bronchitis unless they are systemically very unwell or have a higher risk of complications (see below).

Why shouldn't we be giving antibiotics routinely in acute bronchitis?

  • Antibiotics don't improve the 'overall clinical condition'. They have a beneficial effect on some outcomes, but the results are difficult to interpret and may not be clinically meaningful for many people.
  • They make little difference to the duration of illness (about 1/2 a day in a 3-4/52 illness).
  • They have side-effects.

Acute cough in people who are systemically very unwell

Offer an immediate antibiotic prescription.

This should only be done after a face to face assessment.

Acute cough in people with a higher risk of complications

Consider offering an immediate or back-up prescription to people at higher risk as below.

This should only be done after a face to face assessment.

Patients should be advised to use the back-up prescription if they worsen rapidly or significantly at any time. Of note, back-up prescriptions make no difference to the length of cough when compared to an immediate prescription or no prescription.

Who is at higher risk?

Young children born prematurely.

People with co-morbidities (eg cardiac, pulmonary, renal, neuromuscular, immunosuppression or cystic fibrosis).

People > 65 with 2 of the following, or people over 80 with 1 of the following risk factors:

  • hospitalisation in the previous year.
  • diabetes (type 1 or 2).
  • chronic heart failure.
  • current use of oral steroids.

What antibiotics should we use and for how long?

Adults over 18:

  • 1st line - doxycycline (200mg OD, then 100mg OD for 5 days in total - avoid if pregnant).
  • Or amoxicillin, clarithromycin or erythromycin for 5 days.

Young people under 18:

  • 1st line - amoxicillin for 5 days.
  • Or clarithromycin, erythromycin or doxycycline (if over 12).

There are no studies on what course length is best, but expert opinion is that 5d is likely to be adequate.

No Comments Yet.

Leave your comments