Get access

Systematic review and meta-analysis of the value of clinical features to exclude radiographic pneumonia in febrile neutropenic episodes in children and young people


  • Conflicts of interest: All authors declare they have no conflicts of interest.

  • Funding: RSP is supported by an MRC Research Training Fellowship G0800472, which also supported RW for this review. AJS and LAS received no external funding for their work in this study. The funder had no role in the design or conduct of the study, nor the production of, or decision to submit, this manuscript.

Dr Bob Phillips, Centre for Reviews and Dissemination, University of York, York YO10 5DD, UK. Fax: +44 (0) 1904 321041; email:


Introduction:  Children and young people who present with febrile neutropenia (FNP) secondary to malignancies or their treatment frequently do not undergo routine chest radiography. With shorter courses of antibiotic therapy, failure to recognise pneumonia and consequent under-treatment could produce significant problems.

Methods:  The review was conducted determine the value of the absence of clinical features of lower respiratory tract infection in excluding radiographic pneumonia at presentation of FNP using Centre for Reviews and Dissemination methods. It was registered with the HTA Registry of systematic reviews, CRD32009100453. Ten bibliographic databases, conference proceedings, reference lists and citations were searched. Cohort studies which compared clinical examination to radiographic findings were included. Results were summarised by random-effects meta-analysis.

Results:  Four studies were included. Synthesis of the three higher-quality studies gave imprecise estimates of the average sensitivity (75%; 95% CI 52% to 89%) and average specificity (69%; 95% CI 57% to 78%) for clinical examination in the detection of radiographic pneumonia. If the prevalence of pneumonia is 5%, these estimates produce a negative predictive value of 98% (95% CI 96% to 99%). Alternatively, there remains a 1.9% probability of pneumonia (95% CI 0.7% to 4.2%).

Conclusion:  Signs and symptoms of lower respiratory infection have only moderate sensitivity and specificity for pneumonia; the low prevalence of the condition justifies the routine withholding of chest radiographs. However, for those with a predisposition to pneumonia, or re-presenting after a short course of antibiotic therapy, a chest X-ray should be performed despite an absence of signs.