Elsevier

Journal of Critical Care

Volume 43, February 2018, Pages 163-168
Journal of Critical Care

Sepsis/Infection
Sepsis mortality score for the prediction of mortality in septic patients

https://doi.org/10.1016/j.jcrc.2017.09.009Get rights and content

Highlights

  • A sepsis mortality score to predict 30-day mortality in critically ill patients is proposed from five biomarkers.

  • Logistic regression was used to create the Sepsis Mortality Score.

  • Leukocytes count, procalcitonin, interleukin-6, arylesterase and paraoxonase activities of paraoxonase-1 were used.

  • This multi-marker approach predicted 30-day mortality with very good performance in our sepsis cohort.

  • This added significant prognostic information to the Sequential Organ Failure Assessment score.

Abstract

Purpose

To derive a prediction equation for 30-day mortality in sepsis using a multi-marker approach and compare its performance to the Sequential Organ Failure Assessment (SOFA) score.

Methods

This study included 159 septic patients admitted to an intensive care unit. Leukocytes count, procalcitonin (PCT), interleukin-6 (IL-6), and paraoxonase (PON) and arylesterase (ARE) activities of PON-1 were assayed from blood obtained on ICU presentation. Logistic regression was used to derive sepsis mortality score (SMS), a prediction equation describing the relationship between biomarkers and 30-day mortality.

Results

The 30-day mortality rate was 28.9%. The SMS was [еlogit(p)/(1 + еlogit(p))] × 100; logit(p) = 0.74 + (0.004 × PCT) + (0.001 × IL-6)  (0.025 × ARE)  (0.059 × leukocytes count). The SMC had higher area under the receiver operating characteristic curve (95% Cl) than SOFA score [0.814 (0.736–0.892) vs. 0.767 (0.677–0.857)], but is not statistically significant. When the SMS was added to the SOFA score, prediction of 30-day mortality improved compared to SOFA score used alone [0.845 (0.777–0.899), p = 0.022].

Conclusions

A sepsis mortality score using baseline leukocytes count, PCT, IL-6 and ARE was derived, which predicted 30-day mortality with very good performance and added significant prognostic information to SOFA score.

Introduction

Sepsis remains an ongoing challenge in intensive care medicine largely because of the high mortality rate despite the provision of optimal care [1]. Initial Sequential Organ Failure Assessment (SOFA) score is one of the scoring systems used for predicting mortality in septic patients [2]. However, this approach has been criticised for its lack of capacity to discriminate outcome [3]. The use of serum biomarkers has significantly improved the clinician's ability to diagnose and predict the outcome of sepsis [4], [5].

In everyday clinical practice, the leukocytes count has been the most widely used biomarker to guide the prognosis of septic patients in addition to other clinical parameters [6]. Over the last two decades, procalcitonin (PCT) and interleukin-6 (IL-6) have emerged as biomarkers for the prediction of sepsis outcome [7], [8], [9], [10], [11], [12]. Recent studies suggested the possible role of decreased PON-1 activity in septic patients [13], [14]. PON-1 is an enzyme that is synthesised primarily in the liver, which neutralizes lipopolysaccharides and inhibits the synthesis of some pro-inflammatory cytokines. PON-1 can be evaluated according to its different activities, such as its paraoxonase (PON) activity and arylesterase (ARE) activity. Few studies have revealed the potential prognostic value of PON-1 activity in septic patients [15], [16], [17].

Because of the complex pathophysiology of sepsis, it is unlikely that a single biomarker would be able to reflect the various host responses to infection. Combining several biomarkers into a single classification rule should help to improve their accuracy and hence, their usefulness. The purpose of the present study was to derive a prediction equation using combination of leukocytes count, PCT, IL-6, and PON and ARE activities of PON-1 to predict 30-day mortality in septic patients, which we call the sepsis mortality score. We then sought to compare the performance of this prediction equation to that of the SOFA score in their ability to predict mortality in sepsis.

Section snippets

Study design and participants

This prospective observational study was performed from July 2011 to June 2014 in a 12-bed intensive care unit (ICU) in a major tertiary hospital in Pahang, Malaysia. The protocol used in this study was approved by the local medical research and ethics committee and registered under the National Medical Research Register (NMRR-13-879-15223). Written informed consent was obtained from either the patients or their legally acceptable representative prior to recruitment. Consecutive adult patients

Biomarker profiles

The medians and interquartile ranges are shown for each of the five biomarkers for the population as a whole, as well as stratified by the outcome of 30-day mortality (Table 2). As a summary measure of predictive accuracy, we determined the AUROC and the ideal cut-off values for the ability of each biomarker to classify patients with 30-day mortality (Table 2).

Development of sepsis mortality score

We used multivariate logistic regression to model the ability of biomarkers to identify patients who have the outcome of 30-day

Discussion

In this prospective study, we assembled a cohort of 159 patients with sepsis and studied five biomarkers on their ICU admission with the overall goal of creating a prediction equation, the “sepsis mortality score”, that would allow discrimination of those who are at increased risk of 30-day mortality. A sepsis mortality score using baseline leukocytes count, PCT, IL-6 and ARE activities of PON-1 predicted 30-day mortality with a very good performance (AUROC 0.814) in our sepsis cohort. Of

Conclusion

The multi-marker approach using the baseline leukocytes count, PCT, IL-6 and ARE activity of PON-1 predicted 30-day mortality with a very good performance in our sepsis cohort. Although not superior to the SOFA score, our sepsis mortality score added significant prognostic information, and is thus a valuable supplement to the SOFA score in predicting mortality in sepsis. Further studies are warranted to validate these findings and to assess whether the sepsis mortality score derived from these

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Funding

This work was supported by the International Islamic University Malaysia Endowment B Research Grant (EDWB11-256-0734).

References (25)

  • M.B. Mat-Nor et al.

    Procalcitonin clearance for early prediction of survival in critically ill patients with severe sepsis

    Crit Care Res Pract

    (2014)
  • J.U. Jensen et al.

    Procalcitonin increase in early identification of critically ill patients at high risk of mortality

    Crit Care Med

    (2006)
  • Cited by (27)

    • Performance of prognostic markers in pediatric sepsis

      2021, Jornal de Pediatria
      Citation Excerpt :

      There is no consensus on the use of leukocytes as a prognostic marker. Unlike in previous studies where a higher or lower leukocyte count has been associated with mortality, in the present study such an association was not observed.21,22 A possible explanation is that leukocytes are altered by the use of medications or by other medical conditions in patients with sepsis, such as corticosteroid use and recent chemotherapy, which could influence the results in these cohorts.

    View all citing articles on Scopus
    View full text