Hospital-acquired pneumonia

01OverviewDefinition, clinical context and the essential points that orientate the chapter.

Hospital onset alters the microbiological prior but does not make every infiltrate infectious. Fever, secretions and inflammatory markers are common after surgery and during ventilation, while portable radiographs are imperfect. The working diagnosis should therefore explain the timing, new gas-exchange problem, imaging and host response better than a competing process.

The infectious-disease decision is exposure-led. A recent ESBL-producing isolate, carbapenemase carriage, prolonged ICU stay or broad antibiotic course has more prescribing value than the label 'late onset' alone. Conversely, a patient with early non-severe disease and no resistance markers should not receive an ICU-spectrum regimen by habit.

Response review is a second diagnostic test. Falling oxygen need and improving physiology support the chosen pathway; persistent shock or infiltrates prompt checks of drug delivery, renal exposure, resistance, pleural infection, abscess, obstruction and non-pulmonary sepsis before additional agents are stacked.

Key points

Hospital-acquired pneumonia is considered when symptoms begin at least 48 hours after admission and infection was not incubating on arrival; disease appearing on days 3 to 5 without resistance risk may still follow the community-acquired regimen.
Diagnosis requires a new lower-respiratory syndrome and new or progressive imaging change after considering oedema, atelectasis, aspiration pneumonitis, pulmonary embolism, ARDS and alveolar haemorrhage.
Endotracheal tubes become colonised quickly. A resistant organism in sputum or tracheal aspirate changes probability but does not prove that it is invading lung tissue.
Before empirical treatment, review recent antibiotics, earlier cultures, known MRSA or multidrug-resistant Gram-negative carriage, structural lung disease, immune suppression and the current ward or ICU antibiogram.
NICE advises antibiotics as soon as possible after HAP is suspected and within 4 hours; suspected sepsis may require the shorter sepsis-specific timeframe.
Use oral treatment first for non-severe low-resistance-risk disease when absorption is reliable; severe illness, VAP or resistance risk requires a locally selected intravenous regimen and microbiology input.
At 48 hours revisit whether pneumonia remains credible, then stop, narrow, change route or define an end date. Review after 5 total days and consider stopping when clinically stable.
Prevention is cumulative: remove avoidable devices, reduce unnecessary ventilation and sedation, maintain safe oral and airway care, mobilise, prevent aspiration and avoid antimicrobial exposure without an indication.

02AetiologyUnderlying causes, associations and risk factors, with why each one matters.

Oropharyngeal hospital flora

Antibiotic pressure and healthcare contact reshape upper-airway colonisation, allowing Gram-negative bacilli or resistant Gram-positive organisms to reach distal lung through microaspiration.

Artificial airway biofilm

Endotracheal and tracheostomy tubes bypass normal defence, retain secretions and develop biofilm that repeatedly seeds the lower airway without making every recovered organism invasive.

Aspiration and impaired clearance

Dysphagia, reduced consciousness, supine care, weak cough and immobility increase inoculation and prevent clearance of contaminated secretions.

Compromised host defence

Critical illness, immune suppression, frailty and structural lung disease reduce mucociliary, cellular and systemic control of organisms acquired during admission.

03PathophysiologyThe causal sequence from the underlying abnormality to symptoms and harm.

1
Colonisation precedes invasion
Hospital organisms establish in the mouth or artificial airway; repeated microaspiration introduces them beyond the glottis into susceptible distal air spaces.
2
Alveolar inflammatory exudate
Replicating bacteria recruit neutrophils and protein-rich fluid, filling alveoli and creating the new opacity and purulent secretion pattern.
3
Ventilation perfusion mismatch
Perfusion continues through poorly ventilated consolidated lung, producing shunt, hypoxaemia and increased respiratory effort that may destabilise ventilatory support.
4
Systemic organ injury
When local containment fails, bloodstream spread and dysregulated inflammation cause vasodilatation, kidney injury, encephalopathy and septic shock.

04Clinical features and red flagsSymptoms, examination findings, patterns of presentation and time-critical warnings.

Credible non-ventilated HAP

A new cough or purulent sputum, fever or hypothermia, focal chest findings and new opacity arise after the 48-hour boundary with no stronger alternative explanation.

Ventilator-associated syndromeRed flag

After more than 48 hours of intubation, new infiltrates accompany worsening oxygenation or ventilator settings, altered secretion burden and systemic inflammatory change; several findings together are required.

Resistance-enriched presentationRed flag

Prior broad-spectrum therapy, previous resistant isolates, known colonisation, bronchiectasis, prolonged critical care or overseas healthcare increases the chance that a narrow default regimen will be inactive.

Culture without invasion

Stable physiology, unchanged imaging and no inflammatory deterioration despite a positive airway culture favours colonisation, particularly in a patient with an endotracheal tube or tracheostomy.

Necrotising or pleural complicationRed flag

Cavitation, haemoptysis, persistent bacteraemia, pleuritic pain or a loculated effusion suggests tissue destruction or an infected pleural space requiring urgent imaging, sampling and source control.

05InvestigationsWhat to request, why it matters and how to interpret it.

Investigation order

Read from the initial assessment onwards. Tests may run in parallel in urgent care; first-line, preferred, confirmatory, definitive and gold-standard labels appear only when the chapter explicitly states them.

01
Serial observations and respiratory-support recordFirst step
Why
Define the onset and severity of the new pulmonary deterioration.
Interpretation and limitations
Chart oxygen or ventilator change, respiratory rate, haemodynamics, temperature, consciousness and secretion burden. A trajectory is more discriminating than one fever or one saturation.
02
Chest radiograph compared with earlier imaging
Why
Identify a genuinely new infiltrate and screen for effusion, collapse or oedema.
Interpretation and limitations
Portable films are nonspecific. Rapid change after recruitment or diuresis weakens infection; cavitation, unilateral fluid or unresolved uncertainty may justify ultrasound or CT.
03
Quality lower-respiratory specimen before treatment
Why
Recover a plausible pathogen and susceptibility profile that can support narrowing.
Interpretation and limitations
Use sputum when representative and tracheal aspirate or the locally agreed VAP sample when intubated. Interpret microscopy, specimen quality, abundance and earlier colonisation together.
04
Blood cultures and source search in severe illness
Why
Detect bacteraemia and ensure another hospital-acquired focus is not being missed.
Interpretation and limitations
Collect before antibiotics when this is safe. Line infection, urinary obstruction, abdominal infection and C difficile can coexist with nonspecific chest change.
05
Resistance and exposure ledger
Why
Translate previous microbiology and antimicrobial exposure into empirical-risk decisions.
Interpretation and limitations
Record the organism, site, susceptibility, collection date, recent drug classes and local antibiogram. Carriage informs risk but still does not establish pneumonia.
06
Pleural ultrasound or CT for non-response
Why
Find empyema, abscess, obstruction, embolism or an alternative process that changes treatment.
Interpretation and limitations
Use advanced imaging for a defined unresolved question. Drain purulent or biochemically complicated pleural infection through the pleural pathway rather than relying on broader antibiotics.

06Differential diagnosisRealistic alternatives and the features that help distinguish them.

Atelectasis

Dependent collapse after surgery or ventilation can cause opacity and hypoxaemia, often changing rapidly with recruitment, secretion clearance and mobilisation.

Pulmonary oedema

Positive fluid balance, cardiac dysfunction and bilateral congestion favour hydrostatic oedema, although infection and oedema may coexist in critical illness.

Aspiration pneumonitis

A witnessed large-volume aspiration followed by abrupt hypoxaemia may be sterile chemical injury initially; later bacterial pneumonia requires a new infectious pattern.

Embolic or inflammatory lung injury

Pulmonary embolism, ARDS, alveolar haemorrhage and drug or transfusion reactions can create fever, hypoxaemia and infiltrates without bacterial invasion.

07ManagementImmediate care, first-line treatment, alternatives and escalation.

01SuspectEstablish infection and immediate riskFirst stepA hospital inpatient develops a new lower-respiratory deterioration after the admission boundary.

1EscalationPerform ABCDE, prescribe oxygen to an individual target and involve critical care for shock, exhaustion or escalating respiratory support.
2Compare current and previous imaging, review aspiration and fluid balance, and examine for pulmonary and non-pulmonary alternatives.
3Collect a representative respiratory specimen and blood cultures when indicated, then meet the NICE or sepsis treatment timeframe without waiting for final results.

02SelectBuild an exposure-informed empirical regimenBacterial HAP remains sufficiently likely to justify treatment before susceptibilities are available.

1For non-severe disease without higher resistance risk, use the current NICE or local narrower oral option if swallowing and absorption are dependable.
2For severe disease, VAP or a resistance-enriched history, select intravenous Gram-negative cover from the live local policy and prior susceptibilities.
3Add MRSA activity only for a defined signal such as previous MRSA, current carriage, compatible necrotising disease or a high local probability; document why every drug is present.

03ReviewUse 48-hour evidence to reduce treatmentEarly cultures, imaging comparison and physiological response are available.

1Ask again whether bacterial pneumonia is the best explanation and stop treatment if the syndrome has dissolved or a non-infectious cause is established.
2When infection remains likely, choose the narrowest active agent, adjust to current renal function and convert to oral therapy when severity and absorption allow.
3Count every active inpatient dose, set the 5-day review and name the microbiological or clinical reason if a longer course will be required.

04EscalateInvestigate failure before adding spectrumEscalationShock, gas exchange, fever or radiology worsens despite apparently active treatment.

1Verify administration times, line function, dose, renal clearance, organism susceptibility and whether the sampled organism was actually causal.
2Image for empyema, necrosis, obstruction and embolism and re-examine lines, abdomen, urine, skin and medicines for another fever source.
3Obtain respiratory, infection, microbiology and procedural input; alter antibiotics only for a stated resistance or coverage hypothesis and arrange source control promptly.

Key medicines and prescribing safety4 treatments · regimens, roles and cautions

NICE first-line oral treatment when disease is non-severe and resistance risk is not increased.

Co-amoxiclav for low-risk non-severe HAP

Give 500/125 mg orally three times daily for 5 days, then review.

Avoid in immediate penicillin hypersensitivity or previous co-amoxiclav-associated jaundice or hepatic dysfunction; reduce for renal impairment, monitor liver injury and diarrhoea, and review anticoagulant control when clinically relevant.

One NICE-listed broad antipseudomonal option for severe HAP or higher resistance risk when local ecology supports it.

Piperacillin with tazobactam

Give 4.5 g intravenously three times daily, increasing to four times daily for severe infection under the adopted policy.

Avoid after serious beta-lactam hypersensitivity, reduce maintenance in renal impairment and review sodium, potassium, blood counts and liver tests. Concurrent vancomycin and other nephrotoxins increase renal-injury concern; obtain cultures and de-escalate.

Adds MRSA cover to an appropriate HAP backbone when MRSA is suspected or confirmed.

Vancomycin for supported MRSA risk

Use 15 to 20 mg/kg intravenously two or three times daily, concentration-adjusted; a 25 to 30 mg/kg loading dose may be used in serious illness.

Individualise to weight and renal function, use locally adopted concentration or exposure monitoring and slow infusion to reduce infusion reaction. Reassess hearing and renal injury, especially with piperacillin-tazobactam, aminoglycosides or other nephrotoxins.

NICE-listed specialist alternative for suspected or confirmed MRSA HAP when vancomycin is unsuitable.

Linezolid when vancomycin cannot be used

Give 600 mg orally or intravenously twice daily with specialist microbiology advice.

Review serotonergic, adrenergic and monoamine-oxidase-inhibiting medicines for serotonin syndrome or hypertensive interactions; monitor full blood count, particularly with renal impairment or longer use, and watch for neuropathy or lactic acidosis if exposure is prolonged.

08ComplicationsImportant consequences, why they occur and why they matter clinically.

Respiratory failure

Extensive consolidation and secretion burden increase shunt, fatigue and ventilator dependence, sometimes requiring invasive support or delaying liberation from ventilation.

Empyema or necrosis

Pleural extension, virulent organisms or obstruction can produce infected fluid, abscess or cavitation that needs drainage or another procedural plan.

Septic shock

Pulmonary infection can enter the bloodstream and drive vasodilatory shock, acute kidney injury, coagulopathy and multi-organ support needs.

Treatment-associated harm

Broad agents and intravenous access expose patients to nephrotoxicity, cytopenia, C difficile infection, line complications and selection of additional resistance.

09Monitoring and follow-upTreatment response, safety checks and longer-term review.

Record respiratory rate, oxygen or ventilator requirement, haemodynamics, consciousness, urine output and secretion change often enough to detect a failing trajectory.
Review respiratory and blood microbiology daily, including negative results, and document whether each organism is pathogen, coloniser or contaminant.
Recalculate renal dosing during acute kidney injury, renal recovery or replacement therapy and monitor drug concentrations when vancomycin or another level-guided agent is used.
At 48 hours record stop, narrow, switch or continue; at 5 total days apply clinical stability and complication evidence rather than renewing automatically.
For prolonged or broad treatment, track diarrhoea, cytopenia, liver injury, line complications and new resistant or fungal superinfection.
Feed preventable factors such as aspiration, device days, oral care or transmission clusters back to infection prevention and the relevant ward team.

10Special situationsVariants, exceptions and circumstances that change the usual approach.

The antibiogram is local

A nationally listed drug can be a poor empirical choice on a unit where the likely organism is resistant; current local susceptibility and the patient's own isolates take precedence.

A tube changes test specificity

Biofilm makes airway cultures easier to obtain but harder to interpret. The value of the result rises when microscopy, imaging and a new physiological syndrome agree.

Day five is a decision point

Clinical stability supports stopping uncomplicated HAP after review; abscess, empyema, bacteraemia or an unusual pathogen provides a reasoned exception, not an automatic default.

Non-response may be non-infectious

Oedema, collapse, embolism, haemorrhage and drug or transfusion reactions can persist while broad agents accumulate toxicity, so the diagnosis must be reopened.

Source control includes devices

Pleural drainage, relief of obstruction and removal or exchange of a genuinely infected device can determine cure when systemic antibiotics alone cannot control the focus.

11Common pitfallsFrequent interpretation and management errors.

01
Treating an endotracheal isolate in the absence of a new lung syndrome.
02
Using 'hospital acquired' as permission for the broadest formulary regimen without checking severity or resistance evidence.
03
Adding MRSA therapy without carriage, prior infection, necrotising features or local epidemiological support.
04
Failing to adjust broad beta-lactam or vancomycin exposure as renal function changes.
05
Extending treatment because CRP remains raised while the patient is clinically stable and another inflammatory cause is plausible.
06
Broadening repeatedly while overlooking infected pleural fluid, airway obstruction or a non-pulmonary source.