Acute Bronchitis

Meet the patient

A 28-year-old walks into your outpatient clinic with a raw, productive cough for ten days, having started with a streaming cold a fortnight ago. She is afebrile, saturating 98 per cent, chest clear apart from a few coarse crackles that clear on coughing. She asks, quite reasonably, for "the antibiotic that worked last time".[1][5]

Your job over the next five minutes is to answer one question for her and one for yourself: how long should this cough last? (one to three weeks, sometimes eight — and that is normal), and is this pneumonia or just a cold that moved south? (her vitals and chest exam answer that at the bedside). Hold those two questions and the whole consultation slots into place.[5]

One anatomical fact that decides everything

Acute bronchitis inflames the conducting airways; pneumonia consolidates the alveoli. That single anatomical fact is the pivot on which the entire decision turns — and it is why a normal chest exam and normal vitals are, in this illness, a diagnosis rather than a reassurance.[2]

The lung parenchyma is spared. The trachea and large bronchi are acutely inflamed and swollen after a viral upper-respiratory infection, which is why the cough lingers long after the cold has gone. The parenchyma staying clean is what separates bronchitis from pneumonia, and it is what lets you send a well patient home without a chest X-ray.[2]

Etymology for viva gold: bronchitis is from the Greek bronkhos, "windpipe", plus -itis, "inflammation" — the windpipe and its main branches are inflamed, nothing more. The word has outlived a century of antibiotic over-prescribing for exactly the reason it was coined: the airway is inflamed, the alveolus is not.[2]

Name the gang — nearly all of them are viral

Over 90 per cent of acute bronchitis is viral, and the examiner expects you to name the gang in order. Rhinovirus is the commonest; then influenza A and B, respiratory syncytial virus (RSV), parainfluenza, coronavirus (including SARS-CoV-2), adenovirus and human metapneumovirus.[1]

The bacterial causes are uncommon but they matter, because two of them change management. Bordetella pertussis is the single most important treatable cause of a prolonged acute cough. Mycoplasma pneumoniae and Chlamydophila pneumoniae are the atypicals that blur the line between bronchitis and atypical pneumonia. Bordetella parapertussis causes a milder pertussis-like illness.[1]

The classic trap: the patient with a viral bronchitis who develops a secondary bacterial infection — Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis — especially in COPD. That is a reason to treat the underlying exacerbation by Anthonisen criteria, not a reason to antibiotic every simple bronchitis that walks in.[1]

The cough that outstays its welcome

Patients expect a cough to last days; the data say weeks. The single most therapeutic sentence in this consultation is telling her the cough will run one to three weeks and may stretch to eight — that expectation alone halves re-attendance.[5]

The mechanism is post-viral airway hyper-reactivity. A respiratory virus infects the ciliated epithelium, the cells desquamate, the cilia are lost, mucus and debris pool, and the exposed irritant receptors and C-fibres fire at a lower threshold. The virus clears, the cough does not — because the mucociliary repair and the airway hyper-reactivity lag behind, sometimes for up to eight weeks.[1][6]

Everyone forgets: green or yellow sputum does not mean bacterial. It is inflammatory cells — neutrophils and their myeloperoxidase — not bacteria. Reaching for an antibiotic because the sputum turned green is the recurring trainee error.[1]

Exclude pneumonia — the bedside fork

The diagnosis of uncomplicated acute bronchitis is made when the vitals are normal, the chest has no focal signs, and the oxygenation is normal — and at that point no chest X-ray is needed. Any abnormal vital sign or any focal finding redirects the diagnosis to pneumonia, and the CXR follows.[1][2]

The four vital-sign redirects, in the order that matters:[1]

• Respiratory rate — the single most sensitive marker of a lower-respiratory infection; a raised rate redirects to pneumonia or sepsis.
• Oxygen saturation — normal in bronchitis; hypoxia mandates a CXR and a new diagnosis.
• Temperature — low-grade in bronchitis; a fever over 38.5°C points to influenza, pneumonia, or another infection.
• Heart rate and blood pressure — tachycardia or hypotension suggests sepsis, dehydration, pulmonary embolism, or significant illness.[1]

On auscultation the chest is usually clear, though coarse crackles and a mild wheeze may appear and clear or shift with coughing. The signs of consolidation are absent — no dullness to percussion, no bronchial breath sounds, no increased vocal resonance. Their absence is the diagnosis.[1]

Why antibiotics fail — and the one time they do not

Antibiotics do not work because there is no bacterial target. The Cochrane review (Smith 2017) put a number on it: antibiotics shortened the cough by less than a day, with no effect on illness limitation, while adverse effects — nausea, vomiting, diarrhoea, rash — rose significantly. The conclusion is unambiguous: small, clinically insignificant benefit, real harm.[1]

The GRACE trial (Little 2013) tested amoxicillin against placebo in over 2000 primary-care adults with lower-respiratory symptoms in whom pneumonia was not suspected — and found no benefit in symptom duration or severity, with one pneumonia prevented for every 30 patients treated, a number too small to justify routine use. The ACP and CDC stewardship line follows directly: do not prescribe antibiotics for uncomplicated acute bronchitis.[2][4]

The exception that earns an antibiotic is pertussis. A paroxysmal cough past three weeks, an inspiratory whoop, post-tussive vomiting, or a known exposure — send a nasopharyngeal swab for PCR and treat with a macrolide. That is the one bacterial acute-cough cause worth catching, and the only one that changes the no-antibiotic rule.[1][8]

[1]

The discriminator line: normal vitals plus a clear chest equals bronchitis; any abnormal vital or any focal sign equals pneumonia until the CXR says otherwise.[1]

The differential of an acute cough — CHOPS

The bedside task is not to confirm bronchitis but to exclude the serious and treatable mimics. The mnemonic CHOPS collapses a wide differential into five bites:[1]

[1]

Three of these hide as a "bad cold with a cough" and bite: pulmonary embolism (pleuritic pain and dyspnoea out of proportion, risk factors), heart failure (orthopnoea, raised JVP, bibasal crackles), and early sepsis. None is excluded by reassurance — check the vitals and examine the chest.[1]

The chronic-cough pathway opens at eight weeks: the triad is asthma, reflux, post-nasal drip, with ACE inhibitors and non-asthmatic eosinophilic bronchitis close behind. A cough that has not settled by eight weeks has left acute bronchitis behind and enters that workup.[6]

What to send, what to skip

Acute bronchitis is a clinical diagnosis, and in the typical uncomplicated case no investigations are required. The role of testing is to exclude the alternatives, not to confirm bronchitis — so a routine CXR, full blood count, CRP, and sputum culture in a well adult with classic features waste resources and trigger false-positive cascades.[2]

Order a chest X-ray only when a red flag has surfaced:[1]

• Abnormal vital signs — tachypnoea, hypoxia, high fever, tachycardia, hypotension.
• Focal chest signs on auscultation or percussion.
• Diagnostic uncertainty — you cannot confidently exclude pneumonia.
• Haemoptysis, pleuritic pain, or significant dyspnoea.
• Comorbidity — COPD, heart failure, immunocompromise, malignancy.
• Failure to improve, or deterioration after the first assessment.
• Older smokers (over 50) with persistent symptoms — exclude underlying malignancy.[1]

Test for pertussis when the cough has run past three weeks, is paroxysmal, or comes with a whoop or post-tussive vomiting — or after a known exposure. A nasopharyngeal swab for PCR is the test of choice in the first three to four weeks; culture is specific but less sensitive and only positive early; serology (anti-pertussis-toxin IgG) is useful later. A marked absolute lymphocytosis supports the call.[1][8]

Two point-of-care tests earn a mention. CRP and procalcitonin — low levels support withholding antibiotics, and procalcitonin-guided algorithms cut antibiotic use in respiratory tract infections without worsening outcomes, though they are not routine in simple acute bronchitis. Rapid influenza and SARS-CoV-2 testing matters when those viruses are prevalent and a treatment or isolation decision hinges on the result (pregnancy, immunocompromise, the frail elderly, an outbreak).[3]

The four pillars — and the delayed prescription

[1]

The four pillars are reassure, relieve, withhold antibiotics, and catch pertussis. The most therapeutic intervention is the first one — setting the expectation that the cough runs one to three weeks and may stretch to eight, and that an antibiotic will not help and may harm.[1][2]

Pillar 1 — Reassure, with a delayed prescription in your pocket. A no-antibiotic strategy, or a delayed prescription (post-dated, or "collect from reception if not better in 10 to 14 days"), cuts antibiotic use without worsening outcomes and is recommended by NICE and the ACP and CDC stewardship principles. Tell her the green sputum is neutrophils, not bacteria.[1][7]

Pillar 2 — Symptomatic relief, with named doses.[1]

• Rest and adequate oral fluids.
• Antipyretic and analgesia: paracetamol 1 g orally every 6 hours (maximum 4 g per 24 hours in an adult), or ibuprofen 400 mg orally every 8 hours with food.
• Cough: honey, one to two teaspoons, in adults and children over one year — modestly reduces cough frequency and severity (never give honey to an infant under one year — infant botulism). An antitussive such as dextromethorphan 10 to 20 mg orally every 4 hours (maximum 120 mg per 24 hours) may help a troublesome dry cough; codeine 15 to 30 mg up to four times daily is an alternative but is contraindicated in children under 12 and in breastfeeding.
• Smoking-cessation advice — accelerates mucociliary recovery and cuts recurrence.[1]

Pillar 3 — Withhold antibiotics, except in three named situations. Give a macrolide for suspected or confirmed pertussis (see below); give antibiotics for a COPD exacerbation meeting Anthonisen criteria; treat a clear secondary bacterial infection (pneumonia, sinusitis, otitis media) for what it is — not for the bronchitis.[1]

Pillar 4 — Safety-net and follow-up. Advise her to return urgently with breathlessness, chest pain, haemoptysis, high fever, confusion, or failure to improve by three weeks. Review at three weeks for a persistent cough and enter the chronic-cough pathway if it has not settled.[2]

Pertussis — the hundred-day cough

Pertussis is the one bacterial acute-cough cause you must not miss. Bordetella pertussis adheres to ciliated epithelium, evades clearance, and secretes pertussis toxin, adenylate cyclase toxin and tracheal cytotoxin — these paralyse and destroy the cilia and drive a marked lymphocytosis, producing the paroxysmal cough, the inspiratory whoop (forced inspiration against a narrowed glottis), and the post-tussive vomiting.[8]

The disease runs in three stages: a catarrhal stage (one to two weeks, indistinguishable from a common cold), a paroxysmal stage (two to six weeks or longer), and a convalescent stage (weeks to months). In adolescents and adults the whoop is often absent and the presentation is simply a prolonged paroxysmal cough — which is why pertussis hides inside "a virus that will not go".[8]

Treat with a macrolide, given early to shorten the course and cut transmission:[1]

• Azithromycin 500 mg on day 1, then 250 mg daily on days 2 to 5 — preferred in pregnancy and in neonates and children.
• Clarithromycin 500 mg twice daily for 7 days — alternative in non-pregnant adults.
• Erythromycin 500 mg four times daily for 7 days — traditional, more gastrointestinal adverse effects; avoid in late pregnancy and in long-QT.
• For macrolide allergy, co-trimoxazole for 14 days is an alternative (limited evidence).[1]

Then notify public health, give chemoprophylaxis to household and high-risk contacts, and review vaccination status. The public-health lever is maternal Tdap vaccination in every pregnancy, which protects the neonate through transplacental antibody — the infant who dies of pertussis is the infant whose mother was not vaccinated.[8]

Acute bronchitis in COPD and asthma — Anthonisen holds the pen

A viral infection is the commonest trigger of an acute exacerbation of COPD, and here the no-antibiotic rule bends. The Anthonisen criteria decide: increased sputum volume, increased sputum purulence, increased dyspnoea. Antibiotics are indicated when at least two of three are present — especially all three, or purulence plus one other.[1]

Manage the exacerbation with short-acting bronchodilators (salbutamol plus ipratropium via nebuliser or spacer), systemic corticosteroids (prednisolone 30 to 40 mg orally for 5 to 7 days), oxygen to target, and an antibiotic covering H. influenzae, S. pneumoniae and M. catarrhalis — amoxicillin-clavulanate 500/125 mg three times daily, doxycycline 100 mg twice daily, or a macrolide for 5 to 7 days.[1]

In asthma, viral infections are the commonest trigger of an attack. Treat with an inhaled short-acting beta-2 agonist via spacer or nebuliser, ipratropium if severe, systemic corticosteroids (prednisolone 40 to 50 mg for 5 days in adults), and oxygen to target 94 to 98 per cent; consider magnesium sulphate 2 g intravenously over 20 minutes in life-threatening asthma. Antibiotics are not indicated unless there is clear bacterial infection.[1]

A consultant confession: Mycoplasma lacks a cell wall, so beta-lactams are useless against it — a perennial exam favourite. When an atypical bronchitis or pneumonia is suspected in a young adult, reach for a macrolide or doxycycline, not amoxicillin.[1]

When it goes wrong — the preventable list

• The unnecessary antibiotic — the commonest error; it breeds resistance, causes C. difficile, and teaches the patient that green sputum earns a prescription.[1]
• The missed pneumonia — reassuring a patient who actually has consolidation, because nobody checked the respiratory rate or examined the chest.[1]
• The missed pertussis — attributing a three-week paroxysmal cough to "just a virus" and never sending the swab.[8]
• The over-investigated cold — a CXR, full blood count and sputum culture for a well adult with classic bronchitis, triggering a cascade of incidental findings.[2]
• The red flag ignored — haemoptysis, weight loss or night sweats in a smoker attributed to a viral illness when the CXR was the one test that mattered.[1]
• The masquerader — a pulmonary embolism, heart failure or early sepsis read as "a bad cold with a cough".[1]

Prognosis, disposition, and the safety-net

Acute bronchitis is self-limiting. The cough settles within one to three weeks in most adults; a minority carry a post-bronchitic cough to eight weeks. Prognosis is excellent in the otherwise-well adult — comorbidity (COPD, immunocompromise, heart failure, the extremes of age) is what raises the risk and the threshold to admit.[1]

Community management suits the vast majority — symptomatic treatment, a safety-net, and a delayed prescription if she wants one. Admit when a red flag is present (tachypnoea, hypoxia, confusion, hypotension, significant comorbidity, immunocompromise, social factors), or when the diagnosis is uncertain and a serious mimic cannot be excluded. A follow-up CXR at six to eight weeks is not routine — reserve it for persistent symptoms, smokers over 50, or a non-resolving opacity.[1]

Special populations — where the thresholds move

In pregnancy the mainstay is still symptomatic: paracetamol is the preferred antipyretic throughout; NSAIDs are avoided from 20 weeks and contraindicated in the third trimester (premature closure of the ductus arteriosus). Honey is safe in the pregnant adult; for pertussis, azithromycin is the macrolide of choice — and maternal Tdap in every pregnancy is what protects the neonate.[1][8]

In children, honey is contraindicated under one year (infant botulism) and codeine is contraindicated under 12 years and in breastfeeding. Use weight-based paracetamol or ibuprofen. Watch for bronchiolitis in infants, foreign-body aspiration (sudden onset, monophonic wheeze), and pertussis — apnoea in an infant is a danger sign.[1]

In the elderly, the febrile and inflammatory responses are blunted; the presentation may be confusion, a fall, functional decline, or anorexia rather than a cough. Lower the threshold to investigate (CXR) and to admit, and avoid over-sedating antitussives that risk delirium and secretion retention.[1]

In the immunocompromised, viral lower-respiratory infections can be severe (influenza, RSV, CMV, parainfluenza in transplant recipients). Lower the threshold to test with a multiplex respiratory PCR, to image, and to start antivirals early — oseltamivir empirically for influenza in high-risk patients within 48 hours of symptom onset.[1]

The evidence — Cochrane and the stewardship line

The Cochrane review (Smith 2017) is the single most-cited paper in this topic: antibiotics shortened the cough by less than a day with no effect on illness limitation, while adverse effects rose significantly — small, clinically insignificant benefit, real harm.[1]

The ACP and ASIM principles (Snow and Gonzales 2001) made the stewardship line explicit: clinicians should not prescribe antibiotics for uncomplicated acute bronchitis, and the consultation should focus on patient communication about expected duration and the harms of antibiotics. The GRACE trial (Little 2013) then tested amoxicillin in over 2000 primary-care adults and found no benefit in symptom duration or severity.[2][4]

The procalcitonin Cochrane (Schuetz 2017) provides the modern stewardship tool: procalcitonin-guided algorithms cut antibiotic exposure in respiratory tract infections without worsening outcomes, useful in the undifferentiated acute cough though not routine in simple bronchitis.[3]

The stewardship gap is real and it is measurable. Antibiotic prescribing for acute bronchitis in the United States barely moved between 1996 and 2010 — around 70 per cent of visits still ended in an antibiotic, despite a decade of guidance against it. Closing that gap is the public-health task, and the delayed prescription is the tool.[7]

[1]

The mantra, and the mnemonic

[1]

The mantra: cough for weeks, not days; vitals and chest decide; antibiotics almost never — except pertussis.[1]

Ward-round test — three stems, thirty seconds each

Stem 1 — the patient from the top of the topic (answer)

The 28-year-old with a ten-day productive cough after a cold, afebrile, saturating 98 per cent, chest clear. She wants the antibiotic that worked last time. What do you do? Model: This is uncomplicated acute bronchitis — the vitals and chest exam exclude pneumonia at the bedside, so no CXR and no antibiotic. Set the expectation (cough one to three weeks, sometimes eight; green sputum is neutrophils, not bacteria), treat symptomatically with paracetamol or ibuprofen and honey, and offer a delayed prescription (collect if not better in 10 to 14 days). Safety-net for breathlessness, haemoptysis, high fever, or failure to improve by three weeks.[1]

Stem 2 — the cough that will not stop (answer)

A 42-year-old returns at four weeks with paroxysms of coughing that end in vomiting; her partner had a similar illness two months ago. What is the diagnosis, the test, and the treatment? Model: This is pertussis — paroxysmal cough past three weeks with post-tussive vomiting and a likely exposure. Send a nasopharyngeal swab for PCR, treat with azithromycin 500 mg on day 1 then 250 mg daily on days 2 to 5, notify public health, and arrange chemoprophylaxis for household and high-risk contacts. This is the one bacterial acute-cough cause that earns an antibiotic.[1][8]

Stem 3 — the COPD patient whose cold moved south (answer)

A 68-year-old with known COPD presents with a viral prodrome, now more breathless with more and greener sputum for three days. Does the no-antibiotic rule still hold? Model: No — this is an acute exacerbation of COPD, and the Anthonisen criteria decide: increased sputum volume, increased sputum purulence, increased dyspnoea. He has all three, so antibiotics are indicated. Treat with bronchodilators (salbutamol plus ipratropium), prednisolone 30 to 40 mg for 5 to 7 days, oxygen to target 88 to 92 per cent, and an antibiotic covering H. influenzae, S. pneumoniae and M. catarrhalis — amoxicillin-clavulanate, doxycycline, or a macrolide for 5 to 7 days. The simple-bronchitis rule bends here; the Anthonisen criteria hold the pen.[1]

References

1. [1]Smith SM, Fahey T, Smucny J, et al. Antibiotics for acute bronchitis Cochrane Database Syst Rev, 2017.PMID 28626858
2. [2]Snow V, Mottur-Pilson C, Gonzales R Principles of appropriate antibiotic use for treatment of acute bronchitis in adults Ann Intern Med, 2001.PMID 11255531
3. [3]Schuetz P, Wirz Y, Sager R, et al. Procalcitonin to initiate or discontinue antibiotics in acute respiratory tract infections. Cochrane Database Syst Rev, 2017.PMID 29025194
4. [4]Little P, Stuart B, Moore M, et al. Amoxicillin for acute lower-respiratory-tract infection in primary care when pneumonia is not suspected: a 12-country, randomised, placebo-controlled trial. Lancet Infect Dis, 2013.PMID 23265995
5. [5]Ebell MH, Lundgren J, Youngpairoj S How long does a cough last? Comparing patients' expectations with data from a systematic review of the literature. Ann Fam Med, 2013.PMID 23319500
6. [6]Irwin RS Introduction to the diagnosis and management of cough: ACCP evidence-based clinical practice guidelines. Chest, 2006.PMID 16428688
7. [7]Barnett ML, Linder JA Antibiotic prescribing for adults with acute bronchitis in the United States, 1996-2010. JAMA, 2014.PMID 24846041
8. [8]Crowcroft NS, Pebody RG Recent developments in pertussis. Lancet, 2006.PMID 16765762