Original article
Confusion, Urea, Respiratory Rate and Shock Index or Adjusted Shock Index (CURSI or CURASI) criteria predict mortality in community-acquired pneumonia

https://doi.org/10.1016/j.ejim.2010.07.005Get rights and content

Abstract

Background

Community-acquired pneumonia (CAP) is common and associated with a significant mortality. Shock index, heart rate divided by systolic blood pressure, has been shown to be associated with outcome in sepsis.

Objective

To examine the usefulness of two new criteria CURSI (confusion, urea, respiratory rate and shock index), and CURASI where shock index is replaced by temperature adjusted shock index in mortality assessment of CAP.

Methods

A prospective study was conducted in Norfolk and Suffolk, UK. We explored the usefulness of CURSI and CURASI which we derived and performed mapping exercise using a different cohort. In this study we compared these new indices with the CURB-65 criteria in correctly predicting mortality in CAP.

Results

A total of 190 patients were included (males = 53%). The age range was 18–101 years (median = 76 years). There were a total of 54 deaths during a six-week follow-up. All died within 30-days. Sixty-five (34%) had severe pneumonia by CURB-65. Using CURSI and CURASI, 71(37%) and 69(36%) had severe pneumonia, respectively. The sensitivity, specificity, positive and negative predictive values in predicting death during six-week follow-up were comparable among three indices examined. The Receiver Operating Characteristic curve values (95%CI) for the criteria were 0.67(0.60–0.75) for CURB-65, 0.67(0.59–0.74) for CURSI and 0.66(0.58–0.74) for CURASI (p > 0.05). There were strong agreements between these three indices (Kappa values ≥0.75 for all). Repeating analyses in those who were aged 65 years and over (n = 135) did not alter the results.

Conclusions

Both CURSI and CURASI are similarly useful to CURB-65 in predicting deaths associated with CAP including older patients.

Introduction

Community-acquired pneumonia (CAP) is common and associated with significant mortality [1], [2], [3]. It is therefore essential to identify those with CAP who are most likely to die by using validated severity assessment scores. This approach helps clinician in deciding choice of appropriate route and dose of antibiotics and effective communication with patient and the family as well as closer monitoring of the progress in severe CAP cases.

The best accepted tools to discriminate patients with CAP into high or low risk are the CURB-65 score (confusion, serum urea nitrogen level > 19.6 mg/dL [to convert to millimoles per liter, multiply by 0.357 i.e. urea > 7 mmol/L], respiratory rate ≥30/min, low blood pressure (< 90 mm Hg systolic or ≤60 mm Hg diastolic, and age ≥65 years) (severe pneumonia: ≥3 score)which is recommended in the UK and the Pneumonia Severity Index (PSI), which is widely used in the North America [4], [5]. CURB-65 is easy to perform and the recently suggested variations of CURB-65 [5], [6], [7], [8] also have great potential for use in clinical practice without requiring additional resources.

However, blood pressure is dependent on the cardiac output and the peripheral resistance. Using the blood pressure criterion in the severity assessment of CAP may lead to false negativity in the older people due to high prevalence of systolic hypertension with increasing age. While advanced age is associated with poor outcome, biological ageing is a complex issue and it is not necessarily correlated with chronological age. Therefore, CAP severity assessment criteria which include chronological age could lead to false positivity. Furthermore, age of the patient may not be easily available in emergency setting (e.g. found collapsed).

The Shock Index (SI) [9], the ratio of the heart rate to systolic blood pressure, may be useful in risk stratification of CAP as it may be a more valid measure of physiological changes which reflect the severity of CAP. We hypothesise that a severity assessment rule for CAP which include shock index instead of age and blood pressure may be as good as or superior to CURB-65. We have also shown previously that two new scores called CURSI and CURASI that are modification of CURB-65, where age and blood pressure criteria in CURB-65 were replaced with SI have good correlation to CURB-65 in our pilot mapping exercise described in details in Methods section [10].

We therefore conducted a study to test our hypothesis that replacement of SI or adjusted SI (ASI) defined below to blood pressure and age criteria in CURB-65 by using these two indices may be useful in correctly identifying people with CAP who are at a high risk of dying.

Section snippets

Methods

We compiled data of prospective CAP audits from three UK hospitals: one University Hospital in Norfolk with a catchment population of ~ 750,000 (April–August 2008), one District General Hospital in West Norfolk with a catchment population of ~ 250,000 (December 2006–Jan 2007), and one large District General Hospital in Suffolk with a catchment population of ~ 500,000 (April 2007). The individual project received approval from the respective Clinical Audit Departments. The data were collected for

Results

There were a total of 190 pneumonia cases included in the current report. There were 100 males (53%). The age range was 18–101 years (median 76 years). The number of severe cases were 65 (34%), 71 (37%) and 69 (36%) by CURB-65, CURSI and CURASI criteria, respectively. There were a total of 54 deaths during the 6-week follow-up. There were 32 deaths (49%) in a severe group and 22 deaths (18%) in a non-severe group by CURB-65. The corresponding values for CURSI and CURASI were 33 (47%) and 32 (46%)

Discussion

We found that replacing shock index criterion in CURB-65 in place of blood pressure and age criteria is equally predictive of mortality in community-acquired pneumonia. This is consistent across the analyses (all ages and those who were 65 years or older). The number of patients included in the study is somewhat modest at 190 patients. However, the sample size is comparable with the original validation study of CURB-65 by Lim et al. [16] and has adequate power.

There has been increasing concern

Learning points

  • Pneumonia severity assessment scores are useful in identifying patients with CAP who are likely to die.

  • Shock index can be a useful criterion in severity assessment of CAP.

  • Newly derived CURSI and CURASI indices are as good as CURB-65 in predicting mortality in pneumonia.

  • Using shock index has advantage of easy to remember (compared to BP criteria of CURB-65), can be done readily by visual inspection of observation chart and particularly useful when the patient's age is unknown.

Funding

None.

Disclosure/competing interest

None.

Author contributions

PKM, SMT, AVK and DNS designed the audit pro-forma. DNS, HR, ACS, PP and PS collected the data. PKM, SMT and AVK verified the data. PM analysed the data. PKM and PM prepared the draft manuscript and all co-authors contributed in writing of the manuscript. PKM is the guarantor.

Ethical approval

The current report is the compilation of three pneumonia audits in three hospitals in East Anglia region of the UK. The individual project received approval from the respective Clinical Audit Departments. The results are analysed and presented in the aggregated and anonymised fashion. Therefore, LREC approval was not required.

Acknowledgement

We thank the Clinical Audit Departments of Queen Elizabeth Hospital, King's Lynn; Ipswich Hospital, Ipswich; and Respiratory Medicine Department of Norfolk and Norwich University Hospital, Norwich for their help and assistance with the project. We also thank Mrs Linda Davidson and Niki Day, Department of Medicine for the Elderly Secretaries, Norfolk and Norwich University Hospital for their secretarial support including proof reading.

References (24)

  • A. Capelastegui et al.

    Management of community-acquired pneumonia and secular trends at different hospitals

    Respir Med

    (Mar 2005)
  • J.F. Guest et al.

    Community-acquired pneumonia: the annual cost to the National Health Services in the UK

    Eur Respir J

    (1997)
  • S.J. Wort et al.

    Community acquired pneumonia in elderly people. Current British guidelines need revision

    BMJ

    (1998)
  • W.S. Lim et al.

    Severity prediction rules in community acquired pneumonia: a validation study

    Thorax

    (2000)
  • M.J. Fine et al.

    A prediction rule to identify low-risk patients with community-acquired pneumonia

    N Engl J Med

    (1997)
  • J. Bont et al.

    Predicting death in elderly patients with community-acquired pneumonia: a prospective validation study reevaluating the CRB-65 severity assessment tool

    Arch Intern Med

    (2008)
  • P.K. Myint et al.

    Simple modification of the CURB-65 better identifies the patients including the elderly patients with severe community-acquired pneumonia

    Thorax

    (2007)
  • T.T. Bauer et al.

    CRB-65 predicts death from community-acquired pneumonia

    J Intern Med

    (2006)
  • Myint PK, Sankaran P, Musonda P, Subramanian DN, Ruffell H, Smith AC, Prentice P, Tariq SM, Kamath AV.Performance of...
  • D.M. Yealy et al.

    The shock index: all that glitters

    Ann Emerg Med

    (1994)
  • P.K. Myint et al.

    Usefulness of shock index and adjusted shock index in the severity assessment of community-acquired pneumonia

    Respiration

    (2009)
  • J.M. Hoeinig et al.

    The abuse of power: the pervasive fallacy of power calculations for data analysis

    Am Stat

    (2001)
  • Cited by (24)

    • Shock index predicts up to 90-day mortality risk after intracerebral haemorrhage

      2021, Clinical Neurology and Neurosurgery
      Citation Excerpt :

      SI was initially proposed as a measure of severity in hypovolaemic shock [3]. It has since been demonstrated that SI is a useful point-of-care indicator of early sepsis [4], predictor of mortality in community-acquired pneumonia [5,6] and pulmonary embolism [7]. The normal range for SI is 0.5–0.7, with values > 0.7 indicating worsening haemodynamic status [4].

    • Shock index, modified shock index, and age shock index for prediction of mortality in Emergency Severity Index level 3

      2016, American Journal of Emergency Medicine
      Citation Excerpt :

      Shock index (SI), defined by heart rate (HR) divided by systolic blood pressure (SBP), is widely considered as a predictor of mortality in different clinical conditions including trauma [1-4], pneumonia [5,6], acute pulmonary embolism [7], stroke [8], acute myocardial infarction [9], and pediatric sepsis [10].

    • Association of triage time Shock Index, Modified Shock Index, and Age Shock Index with mortality in Emergency Severity Index level 2 patients

      2016, American Journal of Emergency Medicine
      Citation Excerpt :

      The cutoff points reported in the literature to be associated with mortality are as follows: tachycardia as HR > 100, decreased SBP as SBP < 90 mm Hg, decreased DBP as DBP < 60 mm Hg, abnormal MSI as MSI < 0.7 or MSI > 1.3, abnormal SI as SI < 0.5 or SI > 0.9, elevated SI as SI > 0.7, and elevated Age SI as Age SI > 52.1 [7,18,19,26]. Most of the previous studies that assessed the reliability of SI have shown its superiority to HR or SBP measurement alone [7–17], whereas Liu et al [18] concluded that SI is inferior to SBP and other indices such as MSI. In contrast, one study have compared SI with 9 modified SIs and found that standard SI is superior to modified SIs in predicting mortality of patients with gastrointestinal bleeding [27].

    • CUR-65 score for community-acquired pneumonia predicted mortality better than CURB-65 score in low-mortality rate settings

      2015, American Journal of the Medical Sciences
      Citation Excerpt :

      Inclusion of low-blood pressure criterion in the severity assessment of CAP might incur false negativity in older people because of high prevalence of systolic hypertension with increasing age. Myint et al13 discovered that age and blood pressure in the CURB-65 score could be replaced by shock index, heart rate divided by systolic blood pressure, in predicting deaths associated with CAP including older patients to decrease false negativity. Future prospective clinical multicenter studies should be performed to assess the generalizability of the current findings in such settings.

    • Are shock index and adjusted shock index useful in predicting mortality and length of stay in community-acquired pneumonia?

      2011, European Journal of Internal Medicine
      Citation Excerpt :

      They have advantages and limitations thus Loke et al. suggested further advancement in this area is required [18]. In fact, we have previously examined the inclusion of SI/ASI in severity assessment of CAP by modifying CURB-65 [19]. To date the usefulness of SI and ASI on their own in assessing prognosis of CAP has not been examined.

    View all citing articles on Scopus
    View full text