Recommendations
Urgent
Think 'Could this be sepsis?' based on acute deterioration in a patient in whom there is clinical evidence or strong suspicion of infection.[28][29][30] See Sepsis in adults.
Use a systematic approach, alongside your clinical judgement, for assessment; urgently consult a senior clinical decision-maker (e.g., ST4 level doctor in the UK) if you suspect sepsis.[28][30][31][32]
Refer to local guidelines for the recommended approach at your institution for assessment and management of the patient with suspected sepsis.
Pneumonia is one of the main sources of sepsis.[33]
During the COVID-19 pandemic, consider all patients with cough, fever, or other suggestive symptoms to have COVID-19 until proven otherwise. For management of patients with suspected or confirmed COVID-19 pneumonia, see Coronavirus disease 2019 (COVID-19).
Pneumonia due to COVID-19 is not covered in this topic.
Start antibiotics within 4 hours of diagnosing HAP (if antibiotics have not already been started [e.g., for suspected sepsis]).[7]
Seek advice from a microbiologist on selection of antibiotic treatment for HAP and consider local and ward-based resistance data. Follow your local protocol.[6]
Assess oxygen requirements. In practice, this is done by measuring oxygen saturations and blood gases (a venous blood gas is preferred to an arterial blood gas). Prescribe oxygen if saturations are <94% (or <88% in patients at risk of hypercapnia).[6] Monitor controlled oxygen therapy. An upper SpO2 limit of 96% is reasonable when administering supplemental oxygen to most patients with acute illness who are not at risk of hypercapnia. Maintain oxygen saturations at target range of 88% to 92% in patients at risk of hypercapnia.[53]
Key Recommendations
Use your clinical judgement to assess the severity of symptoms or signs in patients with HAP.[7] Consult early with a senior colleague for an accurate assessment.
There is no validated tool available to assess severity in HAP.[7]
Prescribe antibiotics according to the severity of signs and symptoms and other impacting factors, such as time in hospital before symptom onset, risk of developing complications, and local and ward-based resistance data.[7] The UK National Institute for Health and Care Excellence (NICE) recommends:[7]
Severe symptoms/signs or those at higher risk of resistance: give intravenous broad-spectrum antibiotics (e.g., piperacillin/tazobactam, ceftazidime, ceftriaxone, cefuroxime, meropenem, ceftazidime/avibactam, levofloxacin).[7]
In patients with suspected or confirmed MRSA infection, add intravenous vancomycin or teicoplanin.
If vancomycin cannot be used, seek advice from a specialist. An option is to give linezolid instead, but only if recommended by the specialist.[7]
Review all patients within 48 hours and if possible switch to the oral route.[7]
Mild or moderate symptoms/signs and not at higher risk of resistance: give oral amoxicillin/clavulanate.[7]
Symptoms/signs developing on days 3 to 5 after hospital admission and not at higher risk of resistance: consider amoxicillin.[7]
Review antibiotic choice as soon as microbiological results are available, switching to a narrower-spectrum antibiotic if appropriate.[7] Seek advice from microbiology and follow your local protocol.[7]
Provide supportive care, which may include the following measures:
Give intravenous fluids to patients with volume depletion[34]
Start vasopressors if the patient is hypotensive during or after fluid resuscitation to maintain mean arterial pressure level ≥65 mmHg[54]
If intensive care unit (ICU) admission is indicated, arrange for the patient to be transferred to ICU and managed by ICU specialists together with respiratory physicians[34]
Patients with respiratory failure despite appropriate oxygen therapy require urgent airway management and possible intubation
Give simple analgesia (e.g., paracetamol) as appropriate (e.g., for chest pain).[34]
Monitor in all patients with HAP:[34]
Pulse
Blood pressure
Respiratory rate
Temperature
Oxygen saturation (including inspired oxygen saturation)
Mental status.
Consider measuring C-reactive protein concentration if the clinical progress is uncertain after 48 to 72 hours.[34]
Consider referring patients with dysphagia, particularly frail patients and those with neurological disorders (including cognitive impairment), to a speech and language therapist.
During the COVID-19 pandemic, consider all patients with cough, fever, or other suggestive symptoms to have COVID-19 until proven otherwise. For management of patients with suspected or confirmed COVID-19 pneumonia, see Coronavirus disease 2019 (COVID-19).
Pneumonia due to COVID-19 is not covered in this topic.
Use your clinical judgement to assess the severity of symptoms or signs in patients with HAP.[7] Consult early with a senior colleague for an accurate assessment.
There is no validated tool available to assess severity in HAP.[7]
Some doctors use the CURB-65 score that is recommended by NICE and the British Thoracic Society for people with community-acquired pneumonia by extrapolation, but the evidence for its use in people with HAP is weak.[3][34]
Practical tip
Signs of severe infection together with laboratory studies (i.e., full blood count, urea and electrolytes, liver function, and C-reactive protein) can point to the presence of multiple organ dysfunction and help to define the severity of illness.
Practical tip
Consider dysphagia as a potential cause of HAP, particularly in frail patients and those with neurological disorders (including cognitive impairment). Consider referring these patients to a speech and language therapist. Managing swallowing problems reduces the subsequent risk of aspiration pneumonia.[42]
Antibiotics
Start antibiotics within 4 hours of diagnosing HAP (if antibiotics have not already been started [e.g., for suspected sepsis] - see Sepsis in adults).[7]
Seek advice from a microbiologist on selection of antibiotic treatment for HAP, and consider local and ward-based resistance data. Follow your local protocol.[6]
Choose antibiotic therapy in accordance with specialist microbiological advice and local and ward-based resistance data.[6][7] Also consider:[7]
Severity of symptoms and signs
Number of days in hospital before onset of symptoms
Risk of developing complications (e.g., if the person has a relevant comorbidity such as severe lung disease or immunosuppression)
Recent antibiotic use
Recent microbiological results, including colonisation with multidrug-resistant bacteria
Recent contact with a health or social care setting before current admission
Risk of adverse effects with broad-spectrum antibiotics, such as Clostridium difficile infection.
Practical tip
In practice, when choosing antibiotics the cause of HAP is often uncertain because many people do not have a microbiological diagnosis.[7]
Evidence: Antibiotic prescribing strategies
Evidence shows no difference in clinical outcomes (i.e., mortality, duration of antibiotic use, and intensive care stay length) between very-broad-spectrum antibiotics with de-escalation (i.e., targeted-pathogen therapy) compared with empirical antibiotics in people with HAP. However, a broad-spectrum antibiotics strategy is associated with a higher incidence of emergence of multidrug resistant organisms.
The evidence review for antibiotic prescribing strategies (broad-spectrum vs. empirical) in adults with HAP is based on one randomised controlled trial (RCT).[55]
An antibiotic prescribing strategy of using antibiotics with very broad antimicrobial cover (imipenem/cilastatin plus vancomycin) followed by de-escalation to a broad-spectrum antibiotic based on culture results was significantly better than empirical antibiotics for achieving adequate initial antimicrobial cover. However, there were no significant differences in clinical outcomes, including mortality up to 28 days.[55] This study included non-ventilated adults with HAP and a small proportion of adults with ventilator-associated pneumonia (8.3%).
The overall incidence of multidrug-resistant bacteria was significantly higher with very broad antimicrobial cover followed by de-escalation, compared with empirical antibiotics.
Antibiotic prescribing strategies for HAP should be designed to minimise the emergence of multidrug resistant organisms.
The 2019 antimicrobial prescribing guidelines for HAP from NICE in the UK emphasise minimising the development of resistance and recommend that once microbiology results are available the antibiotic(s) should be changed to a narrower-spectrum option if appropriate.[7][56]
This was based on an open-label RCT of critically ill patients with HAP in Korea. In this trial, 54 patients received initial imipenem/cilastatin (a carbapenem antibiotic) plus vancomycin (very broad coverage) with subsequent de-escalation according to culture results; the other 55 patients received conventional non-carbapenem, non-vancomycin empirical antimicrobials without de-escalation (empirical treatment) with initial and ongoing treatment determined by the attending clinician.[55]
There were no differences between the very broad and empirical groups in the total duration of antimicrobials (mean [SD] 12.5 [5.8] days vs. 14.1 [7.3] days; mean difference 1.6 days less, 95% CI 4.1 days less to 1.0 day more; quality of evidence as assessed by GRADE very low), day-28 mortality (39.6% vs. 25.5%, RR 1.56, 95% CI 0.89 to 2.73; GRADE very low), or duration of ICU stay among survivors (21.1 vs. 14.1 days; P=0.464; GRADE very low).[7]
In their own analysis of the study data, NICE found the number of people who received adequate initial antimicrobial treatment was significantly higher in the broad-spectrum group than in the empirical group (n=54, 75.9% vs. 48.0%, RR 1.58, 95% CI 1.00 to 2.50), although the GRADE for this outcome was very low.[7]
The use of the very-broad-spectrum antibacterials was associated with a higher emergence of multidrug resistant organisms (37.9% vs. 16.7%; unadjusted hazard ratio [HR] 2.76; 95% CI 1.07 to 7.16; P=0.036; HR adjusted for age, sex and APACHE II score: 2.69 [95% CI 1.01 to 7.22]; P=0.049).[55]
The 2016 clinical practice guidelines from the Infectious Diseases Society of America and the American Thoracic Society also suggest that patients treated initially with broad-spectrum antibiotics should have these de-escalated based on culture results.[1]
They made a weak recommendation based on very low-quality evidence from a systematic review including the same RCT as in the NICE guideline above and five additional observational studies.[1]
The observational studies were limited by risk of bias, indirectness (different definitions of de-escalation, different antimicrobial regimens and protocols), and inconsistent results.
The guideline authors therefore relied on expert opinion suggesting that antibiotics should be de-escalated to reduce patient burden, unwanted effects, antimicrobial resistance, and costs.
Evidence: Choice of antibiotic
Evidence shows no differences in clinical effectiveness (clinical cure or mortality) between different broad-spectrum antibiotics or classes of antibiotics, with some small differences in the rates of adverse effects (although high rates of adverse events were reported for many broad-spectrum antibiotics in the studies).
The NICE guideline group considered evidence from six RCTs and one post hoc analysis of an RCT, none of which showed any difference in clinical effectiveness (clinical cure or mortality) so the recommendations were made by consensus based on clinical experience and expert opinion.[7][44][57][58][59][60][61][62]
Patients with severe symptoms or signs or at higher risk of resistance
Give intravenous antibiotics based on advice from microbiology and local resistance data.[7] NICE recommends the following options:[7]
Piperacillin/tazobactam
Ceftazidime
Ceftriaxone
Cefuroxime
Meropenem
Ceftazidime/avibactam
Levofloxacin (consider the safety issues associated with fluoroquinolone use - see More info panel below; consult with a microbiologist and senior colleague first).
Review all patients with severe symptoms or signs or at higher risk of resistance within 48 hours and if possible switch to the oral route.[7]
More info: EMA and MHRA restrictions on the use of fluoroquinolone antibiotics
In November 2018, the European Medicines Agency (EMA) completed a review of serious, disabling, and potentially irreversible adverse effects associated with systemic and inhaled fluoroquinolone antibiotics. These adverse effects include tendonitis, tendon rupture, arthralgia, neuropathies, and other musculoskeletal or nervous system effects.
As a consequence of this review, the EMA now recommends that fluoroquinolone antibiotics be restricted for use in serious, life-threatening bacterial infections only. Furthermore, they recommend that fluoroquinolones should not be used for mild to moderate infections unless other appropriate antibiotics for the specific infection cannot be used, and should not be used in non-severe, non-bacterial, or self-limiting infections. Patients who are older, have renal impairment, or have had a solid organ transplant, and those being treated with a corticosteroid are at a higher risk of tendon damage. Co-administration of a fluoroquinolone and a corticosteroid should be avoided.[63] The UK-based Medicines and Healthcare products Regulatory Agency (MHRA) supports these recommendations.[64]
NICE recommends that for patients with non-severe symptoms or signs of HAP and not at higher risk of resistance, levofloxacin should be prescribed only if switching from intravenous levofloxacin and with specialist advice.[7]
Practical tip
Higher risk of resistance includes people with:[7]
Symptoms or signs starting more than 5 days after hospital admission
A relevant comorbidity, such as severe lung disease or immunosuppression
Recent use of broad-spectrum antibiotics
Colonisation with multidrug-resistant bacteria
Recent contact with a health or social care setting before current admission.
Suspected or confirmed MRSA infection
In patients with severe symptoms or signs or at higher risk of resistance and with suspected or confirmed MRSA infection, add intravenous vancomycin or teicoplanin (dual therapy with an intravenous antibiotic listed above).[7]
If vancomycin cannot be used, seek advice from a specialist. An option is to give intravenous linezolid instead, but only if recommended by the specialist.[7]
It is standard practice to continue administration of these antibiotics intravenously in most cases with consultation with microbiology. This is because there are limited antibiotics with MRSA coverage available for oral administration.
Patients with mild or moderate symptoms or signs who are not at higher risk of resistance
Give oral antibiotics where possible based on advice from microbiology and local resistance data. NICE recommends amoxicillin/clavulanate first line.[7]
Consider amoxicillin alone for people with symptoms or signs of pneumonia starting on days 3 to 5 after hospital admission who are not at high risk of resistance.[7][65] This recommendation is based on the fact that:
Streptococcus pneumoniae is a common cause of HAP in people developing symptoms and signs within 5 days of hospital admission
Amoxicillin is a narrower-spectrum antibiotic with activity against S pneumoniae.
Give alternative oral antibiotics in patients with penicillin allergy or if amoxicillin/clavulanate is unsuitable (e.g., because of local resistance data).[7] NICE recommends the following options:[7]
Doxycycline
Cefalexin (use caution in penicillin allergy)
Trimethoprim/sulfamethoxazole
Levofloxacin (only if switching from intravenous levofloxacin and following specialist advice; consider safety issues associated with fluoroquinolone use - see More info panel above).
Give intravenous antibiotics to patients unable to take oral antibiotics (e.g., impaired swallowing reflex, impaired consciousness, gastrointestinal malabsorption).[7]
Review within 48 hours and if possible switch to the oral route.[7]
Evidence: Specific treatment for healthcare-associated pneumonia
Evidence indicates that local epidemiology and validated individual patient risk factors, rather than the healthcare-associated pneumonia (HCAP) categorisation, should determine the need for MRSA or antibiotic coverage.[2]
Healthcare-associated pneumonia (HCAP) was previously categorised as a distinct type of pneumonia by the American Thoracic Society (ATS) and Infectious Diseases Society of America (IDSA).[1] Patients with HCAP were considered to be at increased risk of infection with multidrug-resistant organisms and it was recommended that they should receive empirical antibiotics to cover multidrug-resistant organisms.
HCAP was defined as pneumonia occurring in a non-hospitalised patient who had any one of the following potential risk factors for antibiotic-resistant pathogens:[1]
Residence in a nursing home or other long-term care facility
Hospitalisation for ≥2 days in the last 90 days
Receipt of home infusion therapy, chronic dialysis, or home wound care
Family member with a known antibiotic-resistant pathogen.
However, in their latest (2019) guidelines for community-acquired pneumonia the ATS and IDSA recommended abandoning the HCAP category.[2] This is because recent high-quality evidence indicates that:[2]
The recommendation to treat HCAP with empirical broad-spectrum antibiotic therapy against methicillin-resistant Staphylococcus aureus (MRSA) and Pseudomonas aeruginosa has lead to an increased and inappropriate use of broad-spectrum antibiotics[66][67]
The risk of infection with multidrug-resistant organisms is more dependent on individual risk factors than on interaction with the healthcare system.[68][69]
The 2019 ATS/IDSA guidelines emphasise that non-hospitalised patients with pneumonia who have had significant experience with the healthcare system should be managed as for community-acquired pneumonia, using local epidemiology and validated risk factors for MRSA or P. aeruginosa to guide treatment decisions.[2]
Supportive care
Provide supportive care to all patients with HAP, which may include the measures outlined below.
Due to a lack of specific guidance on supportive care for patients with HAP, these general recommendations have been extrapolated from the British Thoracic Society guidelines on community-acquired pneumonia.[34]
Oxygen
Assess oxygen requirements. In practice, this is done by measuring oxygen saturations and blood gases (a venous blood gas is preferred to an arterial blood gas).
Give oxygen if saturations are <94% (or <88% in patients at risk of hypercapnia) and maintain at target range.[6] Early oxygen assessment is associated with improved prognosis.[53]
Monitor controlled oxygen therapy. An upper SpO2 limit of 96% is reasonable when administering supplemental oxygen to most patients with acute illness who are not at risk of hypercapnia.
Evidence suggests that liberal use of supplemental oxygen (target SpO₂ >96%) in acutely ill adults is associated with higher mortality than more conservative oxygen therapy.[70]
A lower target SpO₂ of 88% to 92% is appropriate if the patient is at risk of hypercapnic respiratory failure.[53]
Evidence: Target oxygen saturation in acutely ill adults
Too much supplemental oxygen increases mortality.
Evidence from a large systematic review and meta-analysis supports conservative/controlled oxygen therapy versus liberal oxygen therapy acutely ill adults who are not at risk of hypercapnia.
Guidelines differ in their recommendations on target oxygen saturation in acutely unwell adults who are receiving supplemental oxygen.
The 2017 British Thoracic Society guideline recommends a target SpO₂ range of 94% to 98% for patients not at risk of hypercapnia, whereas the 2022 Thoracic Society of Australia and New Zealand (TSANZ) guideline recommends 92% to 96%.[53][71]
The 2022 Global Initiative For Asthma (GINA) guidelines recommend a target SpO2 range of 93% to 96% in the context of acute asthma exacerbations.[72]
A systematic review including a meta-analysis of data from 25 RCTs, published in 2018, found that, in adults with acute illness, liberal oxygen therapy (broadly equivalent to a target saturation >96%) is associated with higher mortality than conservative oxygen therapy (broadly equivalent to a target saturation ≤96%).[70]
In-hospital mortality was 11 per 1000 higher for the liberal oxygen therapy group versus the conservative therapy group (95% CI, 2 to 22 per 1000 more).
Mortality at 30 days was also higher in the group who had received liberal oxygen (RR 1.14, 95% CI 1.01 to 1.29).
The trials included adults with sepsis, critical illness, stroke, trauma, myocardial infarction, or cardiac arrest, and patients who had had emergency surgery. The review excluded studies limited to people with chronic respiratory illness or psychiatric illness, or to patients on extracorporeal life support, receiving hyperbaric oxygen therapy, or having elective surgery.
An upper SpO₂ limit of 96% is therefore reasonable when administering supplemental oxygen to patients with acute illness who are not at risk of hypercapnia. However, a higher target may be appropriate for some specific conditions (e.g., pneumothorax, carbon monoxide poisoning, cluster headache, and sickle cell crisis).[73]
In 2019 the BTS reviewed its guidance in response to this systematic review and meta-analysis and decided an interim update was not required.[74]
The committee noted that the systematic review supported the use of controlled oxygen therapy to a target.
While the systematic review showed an association between higher oxygen saturations and higher mortality, the BTS committee felt the review was not definitive on what the optimal target range should be. The suggested range of 94% to 96% in the review was based on the lower 95% confidence interval and the median baseline SpO₂ from the liberal oxygen groups, along with the TSANZ guideline recommendation.
Subsequently, experience during the COVID-19 pandemic has also made clinicians more aware of the feasibility of permissive hypoxaemia.[75]
Management of oxygen therapy in patients in intensive care is specialised and informed by further evidence (not covered in this summary) that is more specific to this setting.[76][77][78]
Standard ICU supportive care
Arrange for patients with an indication for ICU admission to be transferred to ICU and managed by ICU specialists together with respiratory physicians.[34]
Patients with respiratory failure despite appropriate oxygen therapy require urgent airway management and possible intubation.
Intravenous fluids
Assess all patients for volume depletion and give intravenous fluids if required.[34]
Vasopressors
Start vasopressors (according to local protocols) if the patient is hypotensive during or after fluid resuscitation to maintain mean arterial pressure level ≥65 mmHg.[54]
Venous thromboembolism prophylaxis
Consider prophylaxis for venous thromboembolism (VTE) with a low molecular weight heparin according to local protocols for all patients who are not fully mobile.[34] In practice in the UK, prescription of heparin will be prompted if appropriate once you have recorded your VTE risk assessment in the patient’s electronic record.
Nutritional support
Arrange nutritional support (enteral, parenteral, or via nasogastric feeding) for patients with severe HAP who require a prolonged hospital stay.[34]
Airway clearance
Do not routinely treat people with uncomplicated pneumonia with traditional airway clearance techniques. If needed, offer these patients advice regarding expectoration of sputum.[34]
Consider airway clearance techniques if the patient has difficulty expectorating sputum or if they have a pre-existing lung condition.[34]
Analgesia
Give simple analgesia (e.g., paracetamol) as appropriate (e.g., for chest pain).[34]
Practical tip
Encourage patients with uncomplicated pneumonia (i.e., not complicated by the presence of parapneumonic effusion, empyema, abscess, pneumothorax, necrotising pneumonia [pneumonia associated with cavitation], or bronchopleural fistula), whose medical condition allows them, to sit out of bed. Initially aim for at least 20 minutes in the first 24 hours, and then increase mobility each subsequent day of hospitalisation.
Due to a lack of specific guidance on monitoring for patients with HAP, the following recommendations have been extrapolated from the British Thoracic Society guidelines on community-acquired pneumonia.[34]
Measure observations initially at least twice daily, and more frequently (e.g., every hour) in those admitted to a critical care unit (high-dependency unit or ICU).[34] Monitor:
Pulse
Blood pressure
Respiratory rate
Temperature
Oxygen saturation (with a recording of the inspired oxygen saturation at the same time)
Mental status.
Consider measuring C-reactive protein concentration if clinical progress is uncertain after 48 to 72 hours.[34]
Review all patients with severe symptoms at least every 12 hours until clinical improvement occurs.[34] This should be done by a senior colleague and the medical team.[34]
Review the choice of antibiotic therapy as soon as microbiological results are available to minimise the risk of antimicrobial resistance with broad-spectrum antibiotics.[7]
Change antibiotic(s) according to results (e.g., if bacteria are found to be resistant) using a narrower-spectrum antibiotic, if appropriate.
Seek advice from a microbiologist and follow your local protocol.[7]
Prescribe the shortest course of antibiotics that is likely to be effective to reduce the risk of antimicrobial resistance and minimise the risk of adverse effects.[7]
[ 
]
[Evidence B]
NICE in the UK recommends a total antibiotics course of at least 5 days for people with HAP.[7]
Review treatment after a total of 5 days of antibiotics.[7]
Consider stopping antibiotics on an individual basis if the patient is judged to be clinically stable.[7]
Due to a lack of specific guidance on monitoring for patients with HAP, these recommendations have been extrapolated from the UK NICE guidelines on community-acquired pneumonia (CAP).[34]
Do not routinely discharge patients if they have had 2* or more of the following findings present in the past 24 hours:[34]
Temperature >37.5°C (>99.5°F)
Heart rate >100 beats per minute
Respiratory rate ≥24 breaths per minute
Systolic blood pressure ≤90 mmHg
Oxygen saturation <90% on room air
Inability to maintain oral intake
Abnormal mental status.
(*The British Thoracic Society recommends basing your decision to discharge patients on 1 or more of the findings listed [unless they represent the usual baseline status for that patient] and uses a temperature threshold of 37.8°C [100.04°F].[34] See Evidence panel below.)
Consider delaying the discharge of patients if their temperature is higher than 37.5°C (99.5°F).[34]
Evidence: Cut-off temperature for safe discharge in community-acquired pneumonia
The British Thoracic Society considers a temperature >37.8°C (>100.04°F), rather than the >37.5°C (>99.5°F) threshold recommended by NICE, as a finding that should prompt you to consider delaying discharge in a patient with community-acquired pneumonia (CAP).[3][34]
The >37.5°C (>99.5°F) threshold that is recommended by NICE is based on the conclusions of a prospective cohort study that looked at the value of simple clinical variables for predicting short-term outcomes in patients with pneumonia.[3]
It found a cut-off of 37.5°C (>99.5°F) to be strongly associated with 30-day mortality risk.
For this reason we have based our recommendation on NICE guidance.
At discharge or at follow-up, offer patients access to information about pneumonia, such as a patient information leaflet.[34]
Arrange a follow-up visit at around 6 weeks either with the patient’s general practitioner or in a hospital clinic.[34]
Follow-up after discharge
Request a repeat chest x-ray during recovery after about 6 weeks for patients:[34]
With persisting symptoms or physical signs
Who are at higher risk of underlying malignancy (especially smokers and those aged >50 years).
Use of this content is subject to our disclaimer