Pneumonia, a serious infection of the lungs, is a common reason for Intensive Care Unit (ICU) admission. It may also develop as a significant complication of being on a mechanical ventilator. Although the clinical diagnosis is generally straight-forward to make, determining which organism is causing the infection (pathogen) presents a much greater challenge. Existing detection of pathogens relies on growing the organism under specific conditions in a microbiology laboratory. This process is slow, typically taking 48 to 72 hours, and is influenced by factors such as presence of antibiotics and the ease with which specific organisms can be grown. Conventional microbiology may only be positive less than 40% of cases of pneumonia and this means that patients are often treated with 'best guess' antibiotics. These antibiotics are generally broad spectrum, and risk the development of antibiotic resistance. Equally, organisms which are less commonly seen may not be covered by the initial antibiotic selection and may only be started once this organism is grown after 48 to 72 hours leading to delays in appropriate treatment. The aim of this study is to evaluate the performance of a new form of diagnostic test, using detection of pathogens by gene analysis rather than relying on growth. The investigators believe that this approach will be more rapid and more sensitive, and therefore likely to translate into more rapid and appropriate use of antibiotics.

Pneumonia is a common cause of admission to the intensive care unit, and can also develop as a secondary complication of mechanical ventilation. The diagnosis of pneumonia relies on a combination of clinical and radiographic signs, demonstrating an inflammatory infiltrate to the lung parenchyma combined with evidence of infection. The treatment is appropriate antibiotics, together with supportive care as required by the patient's condition.

The selection of appropriate antibiotics presents a significant challenge, as between 60 and 70% of cases of pneumonia do not yield positive results on microbial cultures. This is the case in both community acquired and hospital acquired pneumonia. It can take up to 72 hours for results of conventional cultures to be returned, and these two aspects mean that pneumonia is commonly treated with empiric broad spectrum antibiotics. Rapid, sensitive tests for microbes could lead to a significant reduction in antibiotic use 1 and use of narrower spectrum agents, which will reduce the selective pressure for anti-microbial resistant organisms.

The investigators on this study have previously shown that a multiplex polymerase chain reactions targeting respiratory pathogens can enhance the detection of such organisms in patients with community-acquired pneumonia and immuno-compromised patients developing pneumonia. The use of a TaqMan microarray card allows for large multiplexing of the PCR reactions, which would allow a single card to target a wide range of potential respiratory pathogens including both community-acquired and hospital-acquired organisms.

The aim of this study is to evaluate a new taqMan multiplex PCR array card, which targets common community and hospital-acquired respiratory pathogens. The investigators anticipate that the results from this card will be available more rapidly than conventional culture. The investigators also aim to evaluate the diagnostic performance of the card, compared to conventional cultures, and validate its use in the population of ventilated patients in intensive care. In addition, conventional cultures of blood have a significantly lower yield in pneumonia that respiratory samples, however as they are considerably less invasive to obtain than broncho-alveolar lavage it would be advantageous if a highly sensitive assay for bacteria could detect relevant respiratory pathogen DNA in the blood. Therefore alongside the testing of respiratory samples, the investigators will assess the ability of the taqMan array to detect organisms in a contemporaneously obtained blood sample.

Inclusion Criteria:

• Age >18
• Mechanically ventilated
• Treating clinician clinically suspects pneumonia and is planning to undertake diagnostic bronchoscopy and lavage

Exclusion Criteria:

• Inability to gain advice from a personal or professional consultee. Where a patient has capacity, declining consent for the study.

Mechanically ventilated patients in intensive care with suspected pneumonia, from community acquired, hospital acquired and ventilator-associated sources.