ORIGINAL ARTICLE

C-Reactive Protein Bedside Testing in Febrile Children Lowers Length of Stay at the Emergency Department Ruud G. Nijman, MD,* Henriëtte A. Moll, MD, PhD,* Yvonne Vergouwe, PhD,† Yolanda B. de Rijke, PhD,‡ and Rianne Oostenbrink, MD, PhD* Background: C-Reactive protein (CRP) is an important diagnostic marker for serious bacterial infections in febrile children. C-Reactive protein bedside testing could potentially accelerate the diagnostic evaluation and shorten length of stay (LOS). Objective: The aim of the study was to study the effect of introducing CRP bedside testing on the LOS of febrile children at the emergency department (ED). Design and Intervention: A prospective observational study with a preimplementation cohort (2008) with traditional CRP testing and a postimplementation cohort (2009–2011) in which CRP bedside testing was introduced. Patients and Setting: All previously healthy children with fever, aged 1 month to 16 years, attending the ED of a university hospital were included; non–ill-appearing children with an upper airway infection were not eligible for CRP bedside testing. Analysis and Main Outcome Measure: Multivariable linear regression and propensity score analyses were used to determine the effect of CRP bedside testing on the logarithmic transformation length of stay [(log)LOS]. Results: The preimplementation cohort included 609 children of whom 286 (47%) had traditional CRP. The postimplementation cohort included the following 1330 children: 728 (55%) children had bedside CRP and 156 (12%) children had traditional CRP. Bedside CRP significantly lowered the median LOS of children in whom an additional diagnostic CRP test was performed, from 178 minutes (interquartile range, 135–232 minutes) to 148 minutes (interquartile range, 108–200 minutes) (30 minutes, 19% of total LOS). A significant reduction of 15% of the (log)LOS remained after adjusting for other determinants of (log)LOS; propensity score analysis showed a 16% reduction. Conclusions: C-Reactive protein bedside testing substantially lowered the LOS of children with fever at the ED in whom an additional diagnostic CRP test was performed. From the *Department of General Paediatrics, Erasmus MC-Sophia Children's Hospital; †Center of Medical Decision Making, Erasmus MC; and ‡Department of Clinical Chemistry, Erasmus MC-Sophia Children's Hospital, Rotterdam, the Netherlands. Disclosure: The authors declare no conflict of interest. Reprints: Rianne Oostenbrink, MD, PhD, Department of General Paediatrics, Erasmus MC-Sophia Children's Hospital, Room SP 1549, PO Box 2060, 3000 CB, Rotterdam, the Netherlands (e‐mail: [email protected]). R.N. is supported by ZonMW, a Dutch organization for health research and development, and Erasmus MC Doelmatigheid; R.O. is supported by an unrestricted grant of Europe Container Terminals B.V and by a fellowship grant of the European Society of Paediatric Infectious Diseases in 2010. The Afinion AS100 analyzer and Afinion AS100 C-Reactive Protein bedside kits were provided by Axis-Shield PoC AS, Norway, and distributed by Clindia Benelux BV. Both Axis-Shield PoC AS and Clindia Benelux BV were not involved in any other aspect of the study or the manuscript preparation. All authors have completed the copyright assignment and the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available upon request from the corresponding author). Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.pec-online.com). Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0749-5161

Key Words: C-reactive protein, bedside test, length of stay, fever (Pediatr Emer Care 2015;31: 633–639)

E

mergency departments (EDs) are becoming increasingly overcrowded.1 Dealing with overcrowding of EDs requires options that improve patient flow at the ED and that shorten the length of stay (LOS).1–4 However, many of the factors contributing to an increased LOS at the ED, such as need for hospitalization, season, and time of arrival, can hardly be influenced.2–4 Therefore, selective and rapid diagnostic tools that have the potential to handle specific patient populations at the ED, such as febrile children, more effectively need to be explored.5 Several clinical signs and symptoms have been described to assist physicians in the diagnostic evaluation of febrile children.6,7 None, however, can be used as solitary predictors for identifying those children who have a serious bacterial infection (SBI) among the large number of febrile children with a self-limiting infectious disease. As a consequence, additional diagnostic tests are ordered frequently for ruling in or ruling out SBIs in febrile children. C-Reactive protein (CRP) is a potent biomarker adding considerable diagnostic value to clinical signs alone.8 Recently, a prediction model was developed and validated to combine clinical signs, clinician's judgment, and CRP to detect SBI in febrile children.9 This model outperformed a model including only clinical signs and clinician's judgment and advocated the use of CRP at an early stage of the diagnostic evaluation. Moreover, CRP outperforms other more traditional biomarkers such as white blood cell count.10–12 One of the limitations of CRP testing performed in a central laboratory is that the turnaround time can substantially prolong the LOS. Nowadays, CRP bedside tests are available, and their ability to predict the presence of SBI in febrile children has been validated for both primary and secondary care settings.13–17 Thus, introducing CRP bedside testing for children with fever at the ED seems attractive, both because of the high prevalence of children with fever presenting to an ED18 and the diagnostic uncertainty of identifying those children with serious illnesses. Because the effect of CRP bedside testing on LOS has not been studied before, we aimed to assess the effect of introducing CRP bedside testing on the LOS of febrile children at an ED.

METHODS Design, Setting, and Participants In this prospective observational study, we compared 2 cohorts of febrile children, aged 1 month to 16 years, who attended the pediatric ED of the Erasmus MC-Sophia Children's Hospital, Rotterdam, the Netherlands. The first cohort consisted of children who attended the ED in 2008, and in this preimplementation cohort, traditional CRP was measured at the discretion of the attending ED physician. Traditional CRP testing was performed in a central laboratory; samples were delivered to the central laboratory by an internal tubing system. The second cohort included children who attended the ED from February 2009 to May

Pediatric Emergency Care • Volume 31, Number 9, September 2015

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

www.pec-online.com

633

Pediatric Emergency Care • Volume 31, Number 9, September 2015

Nijman et al

2011, and in this postimplementation cohort, a study protocol was introduced to guide trained nurses in performing a CRP bedside test in previously healthy children with fever at the moment of triage. Well-looking infants with fever and a clear presentation of viral upper airway infection were included in the study populations but were not eligible for CRP bedside testing. In all febrile children who did not have CRP bedside testing performed, physicians could opt to perform traditional CRP testing. In both cohorts, fever was defined as a body temperature of 38.5°C or greater or “fever” as a positive discriminator of the Manchester Triage System (MTS).19 Body temperature was measured rectally in children aged up to 5 years and by tympanic measurement in older children, at the discretion of the attending nurse. All children with severe underlying chronic disease who were at an increased risk of SBIs were excluded from the study cohorts. We considered only the initial visit for analysis if children revisited the ED within 5 days.

Data Collection Data of all febrile children were registered routinely as part of ongoing prospective observational studies to validate the MTS for children and to develop diagnostic strategies in febrile children.9,20,21 Patient characteristics, referral status, triage data, clinical signs and symptoms, time of arrival at the ED, time of triage and time of ED release, and appointed follow-up after ED release were recorded in a standardized electronic patient form for all children, specifically tailored for the purpose of these studies. All data were collected by trained nurses who were blinded for the outcome of this particular study.

C-Reactive Protein In both the preimplementation and postimplementation cohort, traditional CRP was determined in heparin plasma with an immunoturbidimetric CRP assay using a Roche Modular system (Germany). Detection value of the immunoturbidimetric CRP assay was 1 mg/L. Results were available for the attending physician instantaneously from the electronic patient system; turnaround time was approximately 60 minutes. In the postimplementation cohort, the Afinion AS100 (Axis-Shield PoC AS, Norway, distributed by Clindia Benelux BV) was used for CRP bedside testing. The Afinion AS100 is the new edition of the Nycocard CRP test (distributed by Clindia Benelux BV, the Netherlands), evaluated in earlier days in our hospital.22 The correlation coefficient of the Afinion with the reference method was Afinion = 0.96  CRP (Modular) + 0.38. Whereas the Nycocard CRP test required 5 μL of blood, the Afinion AS100 CRP bedside test requires 1.5 μL of EDTA or heparin blood. Results of the test are available within 4 minutes, with values ranging from 8 to 200 mg/L (values

smaller than 8 are presented as 200 mg/L). The interassay and intra-assay coefficients of variation at a concentration of 12.6 to 17.4 mg/L were 4% and 9.3%, respectively. The CRP bedside test was performed at the ED by trained and certified nurses. Results of the CRP bedside test were entered in the standardized electronic patient form by the nurse and were readily available for the attending physician. Instruction of all involved physicians and nurses included written information by email, plenary research information sessions, and personal instructions by the investigators at the ED. Besides, nurses were trained and certified for the use of the Afinion CRP bedside test and were presented paper cases to train recruitment for the study. Entered CRP bedside tests results were checked by the investigators using the internal memory software of the Afinion AS100.

Main Outcome Measures Main outcome of interest was total LOS at the ED. Length of stay was defined as the time between the moment of registration at the ED front desk and the moment of release from the ED. We also performed a subgroup analysis with LOS defined as the time between ED triage to ED discharge. We transformed LOS to approach a lognormal distribution using the natural logarithm of LOS.

Variables of Interest We created a categorical variable to assess the difference in LOS between the preimplementation and postimplementation cohorts as follows (Table 1): (1) preimplementation cohort including children without CRP, (2) preimplementation cohort including children with traditional CRP, (3) postimplementation cohort including children without CRP, (4) postimplementation cohort including children with traditional CRP, and (5) postimplementation cohort including children with bedside CRP. Next, we considered known determinants of LOS at the ED, such as shift at presentation, season at presentation, and hospitalization (Table 2).1–4 We also included the MTS triage level to describe urgency of care and final diagnoses of SBI and antibiotics prescription as additional descriptors of severity of disease. The MTS is a 5-scale triage system that assesses urgency of care of presenting problems; first, the ED nurse chooses a flowchart describing the presenting problem, after which a discriminator is appointed to assign an urgency level to the presenting problem within the flowchart.19 Serious bacterial infection was defined according to a reference standard and included abnormal radiographic findings and positive cultures from otherwise sterile body sites (urine, blood, spinal fluid).23 A consensus diagnosis would be reached by the investigators if the reference standard

TABLE 1. Multivariable Linear Regression Analysis to Predict Logarithmic Transformation Length of Stay

Cohort

Preimplementation, no. CRP Preimplementation, traditional CRP Postimplementation, no. CRP Postimplementation, traditional CRP Postimplementation, CRP bedside

Intercept

Unadjusted β (SE)

P

Adjusted β (SE)*

P

Reference 0.58 (0.04) −0.02 (0.03) 0.64 (0.05) 0.39 (0.03) 4.59 (0.03)

C-Reactive Protein Bedside Testing in Febrile Children Lowers Length of Stay at the Emergency Department.

C-Reactive protein (CRP) is an important diagnostic marker for serious bacterial infections in febrile children. C-Reactive protein bedside testing co...
245KB Sizes 0 Downloads 12 Views