The Dept of Anaesthesia & Intensive Care, CUHK thanks

for an unrestricted education grant
BASIC instructor/provider course, Hong Kong, July 2nd-4th
Other upcoming courses
Home Feedback Contents


Up CAP Fungal pneumonia Guidelines HAP Mycoplasma Pneumocystis jiroveci

Further reading
Invasive approach
MRSA pneumonia
Risk factors

Healthcare associated pneumonia


Pneumonia developing more than 2-3 days after admission to healthcare facility.


  • 0.5-5% of hospital patients
  • higher incidence in certain groups, eg postoperative patients and patients in ICU.

Risk factors

Click here


  • Thought to result from micro-aspiration of bacteria colonizing the upper respiratory tract and stomach. Microaspiration has been shown to occur in 45% of humans when asleep and the upper airway of 75% of critically ill patients is colonized by enteric gram -ve bacilli.
  • Other routes of infection include:
    • macro-aspiration of gastric contents
    • inhaled aerosols
    • haematogenous spread
    • spread from pleural space
    • direct innoculation from ICU personnel
    • translocation from gut followed by haematogenous spread (unproven)
  • Infected biofilm in endotracheal tube, with subsequent embolization to distal airways, may be important


  • Gram -ve bacilli have been predominant causes but recent data suggest that gram +ve bacteria (particularly S. aureus) are becoming more common
  • Certain factors associated with multi-drug resistant organisms (eg Pseudomonas aeruginosa, Acinetobacter spp and Stenotrophomonas maltophilia):
    • antimicrobial therapy in previous 90 days
    • current hospitalization for 5 days or more
    • high frequency of antibiotic resistance of in the specific hospital unit
    • hospitalization for 2 days or more in previous 90 days
    • residence in nursing home or extended care facility
    • home infusion therapy (including antibiotics)
    • chronic dialysis within 30 days
    • home wound care
    • family member with multi-drug resistant pathogen
    • immunosuppression
    • bronchiectasis
  • NB causative organisms differ from hospital to hospital, ICU to ICU and at different times in the same ICU
  • Legionella should be considered in patients receiving steroids who are not intubated at the time they develop pneumonia
  • Pneumonia caused by multiple organisms common
  • Virus and fungi are uncommon causes


  • Similar to those required for community acquired pneumonia
  • Sampling of lower respiratory secretions
    • non-invasive approach: sputum or tracheal aspirate
    • invasive approach: protected specimen brush or bronchoalveolar lavage specimen
    • some organisms are virtually always pathogens when recovered from respiratory secretions
  • Blood cultures identify aetiological agent in 8-20% of patients
    • Bacteraemia is associated with a worse prognosis
    • A large proportion of patients with severe hospital acquired pneumonia and positive blood cultures have another source of sepsis.


May be difficult:

  • clinical features of pneumonia are nonspecific
  • many non-infectious conditions (eg atelectasis, pulmonary embolus, aspiration, congestive heart failure and cancer) can cause infiltrates on a chest X-ray
  • bacteriological diagnosis is difficult due to the high incidence of colonization of the oropharynx by gram negative bacteria (~70-75% of moribund and chronically ill patients become colonized within 48 hours
    • semi-quantitive culture of sputum increases specificity and may be helpful in diagnosing pneumonia but may not be sufficiently specific in identifying the responsible organisms
    • invasive techniques for collection of distal airway secretions associated with higher specificity but lower sensitivity and considerably greater cost (needs to weighed against cost of unnecessary antibiotic treatment for patients who do not actually have pneumonia).
      • techniques  include aspiration, brushing and bronchoalveolar lavage. Can be performed blindly via an endotracheal tube or under direct vision using a bronchoscope
      • false negative results due to:
        • sampling at an early stage of infection when bacterial load low
        • sampling from an unaffected segment of lung
        • incorrect processing
        • sampling after starting antimicrobial
  • recent randomized controlled trial demonstrated an improved outcome amongst patients investigated by invasive bronchoscopic sampling and quantitative microbiology compared to those investigated with conventional sputum sampling and conventional microbiology. Although these data are used to support  the use of invasive sampling it may have been the quantitative culture that was the key difference
  • although bronchoscopic techniques are probably most accurate non-bronchoscopic (blind) protected specimen brushing and mini-bronchoalveolar lavage associated with good sensitivity and acceptable specificity (~75%)
    • blind techniques have the advantage of being less invasive, available to non-bronchoscopists, available to patients with small diameter endotracheal tubes and of having less effect on gas exchange.


  • An outline of management based on an invasive approach is given in figure 1
  • Using a non invasive approach diagnosis is based on:

    • new infiltrate or change in an infiltrate on CXR and
    • growth of pathogenic organisms from sputum plus two of:
      • leucocytosis or leucopaenia
      • core temperature >38ºC
      • purulent sputum

Initial management based on clinical approach

  • Early treatment with antimicrobials that cover all likely pathogens is thought to result in a reduction in morbidity and mortality, although not all studies have shown a difference in survival amongst those who were treated with adequate antibiotic therapy compared with inadequate antibiotic therapy
  • Initial selection of antimicrobials should be made on the basis of:
    • risk factors for multi-drug resistant pathogens
    • local bacterial flora and resistance patterns
    • antimicrobial therapy in past 2 weeks: try to avoid using the same class of antibiotic
    • in the absence of local guidelines the regimes given in tables 1 & 2 may help
  • Do not delay initiation of antibiotic therapy to obtain lower respiratory tract specimens
  • Antibiotic treatment of acute tracheobronchitis may reduce the duration of mechanical ventilation but there is no evidence it alters mortality

Subsequent management based on clinical approach

Reassess management after 2-3 days or sooner if patient deteriorates. Subsequent management should be based on results of microbiological investigations and response to therapy.


Assessing response to treatment

Assessing response is a vital part of the use of an empirical treatment strategy

  • Clinical improvement usually not apparent for 48-72 h and therapy should not be changed in this time unless progressive deterioration occurs or unless dictated by initial microbiological results
    • Abnormal clinical parameters (fever, white cell count, oxygenation and temperature) have usually resolved by 7 days. Persistence of abnormalities should prompt a search for an alternative cause
  • CXR
    • Limited value for assessing response
    • Initial CXR deterioration is common
    • CXR improvement often lags behind clinical response
    • Rapidly deteriorating CXR pattern, >50% increase in size of infiltrate in 48h, new cavitation or significant new pleural effusion should raise concern
  • If patient fails to respond consider:
    • Is it pneumonia?
      • In particular consider the possibility of acute lung injury due to infection at another site
    • Host factors (eg immunosuppressed, debilitated)
      • Consider unrecognized immunosuppression (eg AIDS) and the possibility of Pneumocystis carinii infection
    • Bacterial factors (eg virulent organism)
      • In patients with ventilator associated pneumonia due to Pseudomonas aeruginosa or Acinetobacter species the mortality approaches 90% in some series
    • Unusual organism eg Mycobacterium tuberculosis, virus, fungus
    • Therapeutic factors (eg wrong drug, inadequate dose)

Management of non-responder

  • consider broadening antimicrobial cover while waiting for results of investigations
  • repeat sampling of lower respiratory tract secretions
  • repeat blood cultures
  • consider invasive microbiology: BAL/PSB
  • ultrasound or CT may be helpful to demonstrate a pleural effusion. Diagnostic aspiration of pleural fluid should be carried out to exclude an empyema
  • CT can demonstrate empyema, abscesses, lymphadenopathy and pulmonary masses
  • consider other source of infection
  • open lung biopsy
    • value in non-immunocompromised patient is controversial
    • should only be considered if bronchoscopic cultures and other diagnostic testing is not helpful
    • empiric alteration of antibiotics or initiation of corticosteroid therapy may be preferable if the patient remains haemodynamically stable

Duration of antibiotic therapy

  • Little data to guide duration of therapy but a randomized controlled trial demonstrated that patients who received appropriate initial empiric therapy for ventilator associated pneumonia for 8 days had similar outcomes to those who received treatment for 14 days (Click here to view abstract)
  • Disadvantages of prolonged course of antibiotics:
    • selection of resistant bacteria
    • increased risk of toxicity
    • cost (needs to be balance against cost of relapse)
  • American Thoracic Society recommend shortening the antibiotic course to as short as 7 days provided that the aetiological agent is NOT Pseudomonas aeruginosa and the patient has a good clinical response with resolution of clinical features of infection


CDC guidelines for preventing healthcare associated pneumonia can be downloaded by clicking here

General measures

  • hand washing: the most effective.
  • isolation of patients with multi-drug resistant pathogens
  • nurse patients in semi-recumbent position (30-45º), particularly when receiving enteral feed to prevent aspiration
  • selective decontamination of the digestive tract may be useful in environments where the incidence of multi-drug resistant pathogens is low
  • minimize blood transfusion

Intubation and ventilation

  • Avoid intubation and re-intubation where possible
  • Orotracheal intubation and orogastric tubes are associated with a lower risk of nosocomial sinusitis and may reduce the risk of ventilator associated pneumonia
  • Maintain tracheal cuff pressure >20 cmH2O to minimize leakage of bacterial pathogens around the cuff
  • Careful handling of ventilator tubing and associated equipment
  • Subglottic drainage of secretions. Aspiration of subglottic secretions which pool above the cuff of the endotracheal tube is thought to decrease aspiration of oropharyngeal organisms and has been shown in to be associated with a lower incidence of early-onset ventilator associated pneumonia


  • mortality is higher in patients with ventilator associated pneumonia than those without
    • increased mortality may be due to association of VAP with increased severity of illness rather than due to VAP itself however it is likely that VAP does have a direct effect on outcome
  • mortality higher for VAP due to aerobic gram -ve  bacilli compared to VAP due to gram +ves (when organisms are fully susceptible to antibiotics).

Further reading

Click here

©Charles Gomersall, April, 2014 unless otherwise stated. The author, editor and The Chinese University of Hong Kong take no responsibility for any adverse event resulting from the use of this webpage.
Copyright policy    Contributors