Home > DARE Reviews > Does this patient have...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet]. York (UK): Centre for Reviews and Dissemination (UK); 1995-.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet].

Does this patient have ventilator-associated pneumonia?

M Klompas.

Review published: 2007.

Link to full article: [Journal publisher]

CRD summary

The review evaluated the precision and accuracy of clinical variables and tests, compared with histology, for diagnosing bacterial ventilator-associated pneumonia. It concluded that radiographic information together with routine bedside evaluations may be indicative of ventilator-associated pneumonia, but additional tests should be considered. The conclusions should be regarded with caution given the shortcomings in review methodology.

Authors' objectives

To evaluate the precision and accuracy of clinical, radiographic and laboratory data for diagnosing bacterial ventilator-associated pneumonia (VAP), compared with histology.


PubMed was searched from inception to October 2006; the search terms were reported. In addition, Google Scholar was searched and references of retrieved articles and reviews were screened. Only studies written in English were eligible for inclusion.

Study selection

Study designs of evaluations included in the review

Studies with at least 25 patients appear to have been eligible for inclusion; other than that no inclusion criteria were specified.

Specific interventions included in the review

Studies evaluating clinical signs and symptoms and results of routine laboratory tests, chest radiographs, Gram stains and cultures of pulmonary secretion were eligible for inclusion. The included studies evaluated a variety of clinical variables (e.g. fever, abnormally high or low leucocyte counts), fluid specimens from blind bronchial aspirates or bronchoalveolar lavage (BAL), cell counts on pulmonary secretions (i.e. for neutrophils and organisms on Gram stain obtained by bronchoscopy), bacteria on Gram stain of pulmonary secretions, cultures of blind bronchial aspirate specimens with more than 1E5 colony forming units (CFU) per mL, BAL fluid of more than 1E4 CFU per mL, radiographic features, combinations of tests, and more than 6 scores on a clinical pulmonary infection scale.

Reference standard test against which the new test was compared

Studies comparing the evaluated indicators with histological examination of pulmonary tissue for VAP by biopsy or autopsy were eligible. The included studies used histology either alone or together with culture results, and all patients underwent full or limited immediate post-mortem lung biopsy.

Participants included in the review

Studies with immunocompetent patients who had undergone mechanical ventilation for at least 1 day were eligible for inclusion. Studies with more than 20% of patients with pneumonia on admission to the intensive care unit were excluded. Half of the included studies were in patients in French hospitals.

Outcomes assessed in the review

Studies had to describe the clinical findings for the included patients; other than that no inclusion criteria were specified. The included studies assessed test data 48 hours or less prior to histological evaluation, or the last clinical values measured before death were used (interval between measurement taken and autopsy unspecified). The review reported sensitivity, specificity, and positive (LR+) and negative (LR-) likelihood ratios.

How were decisions on the relevance of primary studies made?

The author did not state how the papers were selected for the review, or how many reviewers performed the selection.

Assessment of study quality

Studies were assessed using modified criteria from a published review. The studies were classified as level 1 to 4. Level 1 and 2 studies had to enrol consecutive patients, while level 3 studies used non-consecutive patients; level 1 to 3 studies had to report histological evaluations regardless of clinical suspicion of pneumonia, while level 4 studies were non-independent studies on patients already suspected of having pneumonia. The author did not state how the validity assessment was performed.

Data extraction

The author did not state how the data were abstracted for the review, or how many reviewers performed the data extraction. Sensitivity, specificity, LR+ and LR- were calculated for each sign and combination of findings.

Methods of synthesis

How were the studies combined?

Summary estimates (sensitivity, specificity, LR+ and LR-) together with their 95% confidence intervals (CIs) were derived for signs or combinations of findings from selected studies using a random-effects model.

How were differences between studies investigated?

The review grouped the studies according to evaluated test/variables and independence of the samples.

Results of the review

Fourteen studies (n=655) were included.

All included studies used non-consecutive patients and had a quality level of either 3 or 4. The sample sizes ranged from 25 to 141, with the majority of studies including fewer than 40 patients.

Radiographic infiltrate plus two clinical variables (fever, leucocytosis or purulent sputum) increases the likelihood of VAP (LR+ 2.8, 95% CI: 0.97, 7.9; based on 1 study). The absence of a new infiltrate on a plain chest radiograph lowers the likelihood of VAP (LR- 0.35, 95% CI: 0.14, 0.87; based on 3 studies), as do fewer than 50% neutrophils in BAL fluid (reported in 3 studies: LR- values of 0.05, 0.9 and 0.10).

The presence or absence of fever, abnormal white blood cell count, or purulent pulmonary secretions did not substantially alter the probability of VAP.

Authors' conclusions

Radiographic information together with routine bedside evaluations may be indicative of VAP. However, clinicians should consider additional tests to provide further evidence for the presence of VAP or to establish other diagnoses.

CRD commentary

The review addressed a complex research question. The search was limited and restricted to English language publications and there were no specific attempts to locate unpublished material; it is therefore possible that relevant studies were not considered and it has to be suspected that language and publication bias have been introduced into the review. In particular, it should be noted that half of the included studies were conducted in France (a country that the author states enables better access to the data in question), but only reports published in English were eligible for this review. No specific attempts to reduce reviewer bias or errors, such as independent inclusion screening by more than one reviewer, were reported. The validity of the included studies was assessed using a limited number of quality criteria; the results were not reported in detail but partially taken into account in the data synthesis. Individual study results varied and statistical heterogeneity was not assessed, so it is unclear whether statistical pooling was appropriate. The sample sizes of the included studies were small, as was the total number of patients that the evidence for this review was based on. Given the outlined shortcomings the conclusions have to be regarded with caution.

Implications of the review for practice and research

Practice: The author outlined a diagnostic algorithm, taking several clinical variables and tests into account, and presented a specific clinical scenario.

Research: The author did not state any implications for further research.

Bibliographic details

Klompas M. Does this patient have ventilator-associated pneumonia? JAMA 2007; 297(14): 1583-1593. [PubMed: 17426278]

Original Paper URL


Indexing Status

Subject indexing assigned by NLM


Bronchoalveolar Lavage Fluid; Cross Infection /diagnosis /etiology; Gentian Violet; Humans; Phenazines; Pneumonia /diagnosis /etiology; Respiration, Artificial /adverse effects



Database entry date


Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.

CRD has determined that this article meets the DARE scientific quality criteria for a systematic review.

Copyright © 2014 University of York.

PMID: 17426278