Original article
Outcome prediction using the Mortality in Emergency Department Sepsis score combined with procalcitonin for influenza patientsPredicción del resultado utilizando la puntuación de mortalidad por sepsis en la unidad de urgencias combinada con procalcitonina en los pacientes con gripe

https://doi.org/10.1016/j.medcle.2019.03.022Get rights and content

Abstract

Background

Severe influenza is often associated with bacterial coinfection and can trigger sepsis, which increases the severity, complexity and mortality of the disease. To determine an effective method for predicting 28-day mortality of emergency department (ED) patients with influenza, we investigated the Mortality in Emergency Department Sepsis (MEDS) score, procalcitonin (PCT) and other relevant biomarkers.

Methods

We conducted a retrospective, observational, monocentric study, and the endpoint was 28-day mortality. Independent predictors were identified and a new combination predictive model was created both by logistic regression, and the model was evaluated by a receiver operating characteristic (ROC) curve.

Results

A total of 364 consecutive ED admitted patients with influenza were enrolled and 45 patients died within 28 days. For predicting 28-day mortality, the MEDS score and PCT were independent predictors with adjusted odds ratio of 1.318 (95% CI 1.206–1.439) and 1.038 (95% CI 1.010–1.065), and with AUCs of 0.817 (95% CI 0.756–0.878) and 0.793 (95% CI 0.725–0.861), respectively. The new combination of the MEDS score with PCT significantly improved the efficacy for predicting 28-day mortality with an AUC of 0.857 (95% CI 0.809–0.905), and was superior to the SOFA score with an AUC of 0.837 (95% CI 0.779–0.894).

Conclusion

The MEDS score and PCT, especially when combined, perform well for predicting mortality of ED admitted patients with influenza.

Resumen

Antecedentes

La gripe severa se asocia a menudo a la coinfección bacteriana, pudiendo desencadenar sepsis, lo cual incrementa la gravedad, la complejidad y la mortalidad de la enfermedad. Para determinar un método efectivo de predecir la mortalidad a 28 días de los pacientes con gripe en la unidad de urgencias (ED), investigamos la puntuación de mortalidad por sepsis en la unidad de urgencias (MEDS), procalcitonina (PCT) y otros biomarcadores relevantes.

Métodos

Realizamos un estudio retrospectivo, observacional y unicéntrico, cuya evaluación clínica fue la mortalidad a 28 días. Identificamos factores predictivos independientes, creamos un nuevo modelo predictivo combinado por regresión logística, y evaluamos el modelo mediante una curva ROC.

Resultados

Incluimos a un total de 364 pacientes consecutivos ingresados en la ED, de los cuales 45 fallecieron en el plazo de 28 días. Para predecir la mortalidad a 28 días, la puntuación MEDS y PCT fueron factores predictivos independientes con odds ratio ajustados de 1,318 (IC 95%: 1,206-1,439) y 1,038 (IC 95%: 1,010-1;065), y ABC de 0,817 (IC 95%: 0,756-0,878) y 0,793 (IC 95%: 0,725-0,861), respectivamente. La nueva combinación de la puntuación MEDS y PCT mejoró significativamente la eficacia para predecir la mortalidad a 28 días con ABC de 0,857 (IC 95%: 0,809-0,905), siendo superior a la puntuación SOFA con ABC de 0,837 (IC 95%: 0,779-0,894).

Conclusión

La puntuación MEDS y PCT, especialmente cuanto se combinan, constituyen una buena predicción de la mortalidad de los pacientes ingresados en la ED con gripe.

Introduction

Influenza epidemics result in approximately 300,000–650,000 deaths worldwide annually and account for heavy social burdens.1, 2 Although these new estimates indicate that influenza-associated mortality has increased, they might underestimate the true mortality burden.3, 4 Current prevention and treatment strategies are seemingly not adequate to reduce this burden.

At the time of epidemics or a sporadic pandemic, it is of vital importance to stratify severity and predict prognosis, which need not only physician's clinical judgment, but also objective severity scores and indicators, especially in the emergency department (ED). Most but not all patients with severe influenza infections present with pneumonia.5 Even if in patients with influenza pneumonia, the most common pneumonia severity scores, including pneumonia severity index (PSI)6; confusion, urea, respiratory rate, blood pressure, age ≥65 years (CURB-65)7; confusion, respiratory rate, blood pressure, age ≥65 years (CRB-65),8 are powerless in predicting mortality.9, 10, 11 The Mortality in Emergency Department Sepsis (MEDS) score is used to assess the severity of disease and to predict the mortality of patients with suspected infection in the ED.12 But fewer studies have investigated the clinical implication of MEDS for evaluating influenza patients.

In view of the limitations of clinical scores, there has been considerable interest in the development of rapid biomarkers for reliable prognosis prediction.13 Among them, procalcitonin (PCT) is widely used in virtue of its higher predictive capacity for bacterial infection,14 and associated with the severity of illness in patients with pneumonia and appears to be a prognostic marker of morbidity and mortality.15 Particularly notable is in recent years the increase in the number of patients with influenza who are admitted to hospitals with bacterial coinfection or superinfection, causing additional hospitalization and mortality.16, 17 These patients should be identified when developing clinical antibiotic strategies, but classical pathogen cultures are slow and have low sensitivity. PCT has been proved as a reliable single predictor to distinguish individual virus infection and virus-bacterial coinfection.18, 19 However, few studies have focused on whether PCT can directly determine prognosis in influenza patients.

Assessment of illness severity and the risk for clinical deterioration at the time of initial diagnosis are essential for optimal management. The purpose of this study is to investigate the prognostic prediction value of MEDS score, PCT and other relevant biomarkers for influenza patients in the ED.

Section snippets

Study design and patient selection

This study was conducted using clinical and laboratory data collected retrospectively at the ED (including fever clinic) of a university-affiliated urban tertiary teaching hospital. From 2010 to 2018, including 8 epidemic influenza seasons, all consecutive ED (including fever clinic) admitted patients with diagnosis of influenza infection were screened. Medical records were independently reviewed by two physicians. This study was approved by the Institutional Review Board and Medical Ethics

Characteristics of the study population

During the study period, 421 admitted patients with diagnosis of influenza were screened. Of these, 57 (13.5%) were excluded: 25 lacked important medical records, 28 were non-ED (or fever clinic) admission, 2 were less than 15 years old, and 2 were pre-existing thyroid disease. Finally 364 (86.5%) were eligible for enrollment, and of them, 230 (63.2%) fulfilled the CAP criteria, 83 (22.8%) were diagnosed with bronchitis, and 51 (14.0%) with respiratory underlying disease aggravated; 335 (92.0%)

Discussion

Severe influenza is a common potential initiating factor for sepsis, and pathogenic bacteria are usually involved in this process, increasing additional mortality and morbidity by coinfection or superinfection.22 These patients undergo a complex interaction process of initial virus, co-pathogen invasion and host immune response, and the disease may deteriorate unexpectedly and rapidly.23 Identifying severe influenza patients at high risk of mortality is crucial to anticipate prognosis, select

Abbreviations

CAP, community-acquired pneumonia; RT-PCR, reverse transcription polymerase chain reaction; PCT, procalcitonin; CRP, C-reaction protein; MEDS, Mortality in Emergency Department Sepsis; SOFA, Sequential Organ Failure Assessment; ED, emergency department; SD, standard deviation; IQR, inter-quartile range; CI, confidence interval; ROC, receiver operating characteristic; AUC, area under the ROC curve; SEN, sensitivity; SPE, specificity; PPV, positive predictive values; NPV, negative predictive

Funding

None.

Conflict of interest

The authors declare that they have no competing interests.

Acknowledgments

The authors would like to thank the staff of the Microbiology Laboratory, the Analysis Department and the Medical Records Department of Beijing Chao-Yang Hospital for their helpful contributions to the study.

References (32)

  • W.S. Lim et al.

    Defining community acquired pneumonia severity on presentation to hospital: an international derivation and validation study

    Thorax

    (2003)
  • W.S. Lim et al.

    BTS guidelines for the management of community acquired pneumonia in adults: update 2009

    Thorax

    (2009)
  • R. Riquelme et al.

    Predicting mortality in hospitalized patients with 2009 H1N1 influenza pneumonia

    Int J Tuberc Lung Dis

    (2011)
  • N.I. Shapiro et al.

    Mortality in Emergency Department Sepsis (MEDS) score: a prospectively derived and validated clinical prediction rule

    Crit Care Med

    (2003)
  • D. Viasus et al.

    Prediction of prognosis by markers in community-acquired pneumonia

    Expert Rev Anti Infect Ther

    (2013)
  • M. Christ-Crain et al.

    Clinical review: the role of biomarkers in the diagnosis and management of community-acquired pneumonia

    Crit Care

    (2010)

    • Cited by (0)

    View full text
    © 2019 Elsevier España, S.L.U. All rights reserved.