Article Text


Pleural infection and pneumonia
S61 RAPID score: the development of a validated clinical score in pleural infection, to identify at presentation those at risk of poor outcome
  1. N M Rahman1,
  2. B C Kahan2,
  3. R Miller3,
  4. N A Maskell4
  1. 1Oxford Respiratory Trials Unit, Oxford Centre for Respiratory Medicine, Oxford, UK
  2. 2MRC Clinical Trials Unit, London, UK
  3. 3University College London Hospitals, London, UK
  4. 4Academic Respiratory Unit, University of Bristol, Bristol, UK


Introduction Pleural infection remains common with an increasing incidence. It is associated with a high morbidity and mortality. The development of a validated clinical risk score at presentation to identify those at high risk may help formulate early management strategies.

Methods The RAPID score was constructed based on a large cohort of patients entering a multicentre UK pleural infection trial—MIST1 (n=411). Out of 32 baseline clinical characteristics recorded at trial entry, model selection was undertaken to find variables predictive of poor clinical outcome. Results were obtained by using backwards selection with a p value of 0.05. Multiple imputation was used to account for patients with missing baseline variables. The primary outcome assessed was mortality at 3 months. Total time in hospital was also assessed.

Results Age, urea, albumin, hospital acquired infection, and non-purulence were all found to be clinical predictors or poor outcome. A score was developed using these variables.

  • R - urea <5=0, 5–8=1, >8=2

  • A - age <50=0, 50–70=1, >70=2

  • P - purulence=0, non purulence=1

  • I - iatrogenic/hospital acquired infection =1 community acquired =0

  • D - dietary factors. Albumin >27=0 <27=1

  • Total RAPID score range 0–7

In order to help interpret the RAPID score, we stratified patients into low-risk, medium-risk, and high-risk groups. Patients with a RAPID score of 0–2 are considered low risk, a score of 3–4 indicates a medium risk, and a score of 5–7 indicates high risk. This scoring system was then validated using another large cohort of patients with pleural infection who had been enrolled in a UK multicentre trial –MIST2 (n=210). Abstract S61 table 1 shows the main results. Time in hospital increased with increasing RAPID score. In MIST1 patients with RAPID 0–2 had median stay 10 (7–16) days, compared to RAPID >5 who had a median stay of 18 (9–26) days. This trend was also seen in MIST2.

Abstract S61 Table 1

Parameter estimates (mortality at 3 months)—RAPID score

Conclusion The RAPID score appears to allow for risk-stratification of patients with pleural infection at presentation and could prove useful in clinical practice in guiding initial management.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.