We evaluated the utility of the Korean Modified Early Warning Score (KMEWS), which combines the Modified Early Warning Score (MEWS) and the Korean Triage and Acuity Scale (KTAS), as a triage tool to screen for infection in patients who visit the emergency department.
We retrospectively reviewed data extracted from electronic medical records. Patients aged ≥18 years with an infection who were admitted to the hospital via the emergency department between January 2018 and December 2019 were eligible for inclusion. The KMEWS score was calculated as the sum of the KTAS level and the MEWS score. We generated receiver operating characteristic curves and determined the area under the receiver operating characteristic curve (AUC) for the KMEWS, KTAS, MEWS, and Mortality in Emergency Department Sepsis (MEDS) scales. The primary outcome was septic shock, and secondary outcomes were intensive care unit admission and in-hospital mortality.
The AUC values (95% confidence interval) for predicting septic shock were as follows: KMEWS, 0.910 (0.902–0.918); MEWS, 0.896 (0.887–0.904); KTAS score, 0.809 (0.798–0.819); and MEDS, 0.927 (0.919–0.934). The AUC values (95% confidence interval) for predicting in-hospital mortality were as follows: KMEWS, 0.752 (0.740–0.764); MEWS, 0.717 (0.704–0.729); KTAS score, 0.764 (0.752–0.776); and MEDS, 0.844 (0.834–0.854). The AUC values (95% confidence interval) for predicting intensive care unit admission were as follows: KMEWS, 0.826 (0.816–0.837); MEWS, 0.782 (0.770–0.793); KTAS score, 0.821 (0.810–0.831); and MEDS, 0.839 (0.829–0.849).
The KMEWS, which is a combination of the MEWS and the KTAS scores, might be a useful triage tool in emergency department patients who present with infection, particularly for predicting septic shock.
Several scores for predicting septic shock have been identified, but most of them require laboratory tests that are time intensive. Therefore, a tool is required that can predict septic shock early in the triage stage.
The Korean Modified Early Warning Score, which is a combination of the Modified Early Warning Score and the Korean Triage and Acuity Scale, might be a useful triage tool in emergency department patients with infection, particularly for predicting septic shock.
Sepsis is an inflammatory disease caused by a reaction of the immune system, which can be life-threatening and is responsible for 20% of all hospital deaths each year [
The Modified Early Warning Score (MEWS) is a simple physiological score used to screen patients at risk of clinical deterioration using body temperature, blood pressure, pulse rate, respiratory rate, and level of consciousness values and to allow for the early detection of clinical deterioration and the potential need for a higher level of care (
The study was approved by the Institutional Review Board of Chungnam National University Hospital (No. 2020-10-059). The need for informed consent was waived because of the retrospective study design and the use of anonymized data. Only clinical data were extracted, and no personal or identifiable information was recorded.
We retrospectively reviewed data extracted from electronic medical records. The study sample included patients aged ≥18 years with infections who were admitted to the hospital via the ED between January 2018 and December 2019 at a tertiary care university hospital with 1,350 beds in Daejeon, Korea. The ED provides medical care to approximately 55,000 patients per year. Patients with missing data were excluded.
The diagnosis of infection was confirmed using the relevant International Classification of Diseases, 10th Revision (ICD-10) codes in the medical records. Patients with any of the following infection-related ICD-10 codes were eligible for enrollment: A00–B99, G00–09, I00–02, I30–33, I38–41, J00–22, J36, J37, J40–J43, J68, J69, J80, J85–J86, K11–12, K35–37, K57, K61, K63, K65, K67, K75, K77.0, K80–81, K83.0, K85, L00–08, M00–03, M86, N10, N12, N13.6, N16.0, N28.84–28.86, N30, N34, N39.0, N41, N45, N61, N70–74, and O91.
Sepsis patients were defined by the presence of two or three of the three quick SOFA (qSOFA) clinical criteria (altered mentation, respiratory rate ≥22 breaths/min, and systolic blood pressure ≤100 mmHg) [
We collected clinical data from the patients’ electronic medical records. The information included age, sex, systolic arterial pressure (mmHg), respiratory rate (breaths/min), body temperature (°C), and mental status. We calculated the Charlson Comorbidity Index, which categorizes the comorbidities of patients based on the ICD diagnosis codes found in the administrative data [
Continuous variables are expressed by mean±standard deviation or median (interquartile range, IQR). Continuous variables were analyzed using the Student t-test or the Mann-Whitney U-test, and categorical variables were analyzed using chi-square or Fisher exact tests. A multivariable logistic regression was performed to identify predictive factors for septic shock using variables that had previously been reported to be significantly associated with septic shock.
We generated receiver operating characteristic (ROC) curves and determined the area under the ROC curve (AUC) for individual measures (KMEWS, KTAS, MEWS, and MEDS) that was associated with septic shock, ICU admission, and in-hospital mortality. The AUCs of the models were calculated and tested mutually for significance using DeLong equality tests. In addition, the cutoff value was calculated using the Youden index (Youden’s J statistic). All statistical analyses were performed using the IBM SPSS ver. 19.0 (IBM Corp., Armonk, NY, USA) and MedCalc ver. 14.8.1 (MedCalc software, Ostend, Belgium). P-values less than 0.05 were considered statistically significant.
We enrolled 19,228 patients during the study period, of which 7,907 patients had infection-related diagnosis at discharge, and 2,814 patients with missing data were excluded. If data such as blood test items, history, and state of consciousness were collected inadequately, they were treated as missing data. Thus, 5,093 patients of these were included in the analysis data set (
A multivariable logistic regression revealed that age, transfer from a long-term care facility, KTAS score, and MEWS score were significantly associated with septic shock (
In the ROC analysis, the AUC values (95% CI) of factors associated with septic shock were as follows: KMEWS, 0.910 (0.902–0.918); MEWS, 0.896 (0.887–0.904); KTAS score, 0.809 (0.798–0.819); and MEDS score, 0.927 (0.919–0.934) (
In this study, the KMEWS, a combination of the MEWS and KTAS scores, was a useful prognostic marker for patients with infection, particularly for predicting septic shock. In the ED, triage helps predict the severity of conditions and determines the priority of patient treatment [
Patients visit the ED with a wide variety of complaints, but the proportion of patients with infection is substantial [
The qSOFA score is an established screening tool for sepsis [
The Early Warning Score is a simple physiological scoring system that can be easily applied at the bedside [
However, as the MEWS is somewhat nonspecific and does not contain factors related to the chief complaint, there is a limitation in its use for patients with infection. In contrast, the KTAS includes the chief complaint and the vital signs, but the hemodynamic criteria are not subdivided. The KTAS and MEWS can be applied to ED triage because laboratory results are not required. Therefore, we hypothesized that supplementing the physiological data with the MEWS and KTAS scores could help determine a prompt prognosis in infected patients.
As the KTAS includes both vital signs and chief complaints, and the MEWS includes vital signs, the KMEWS (the sum of the KTAS and the MEWS scores) has a weighting value for the initial vital signs. In the multivariable logistic regression analysis of this study, the KTAS and KMEWS scores were independently associated with septic shock and showed similar odds ratios for septic shock. Therefore, the KMEWS was calculated by combining the two scores. As a result, the KMEWS showed similar or higher AUC values for septic shock, ICU admission, and mortality compared to either the KTAS score or the MEWS score alone. The MEDS had a slightly higher AUC value than the KMEWS, but it is unsuitable for use as a septic shock screening tool in the ED. Therefore, the KMEWS could be a useful prognostic tool for triaging patients with septic shock in the ED.
Nevertheless, this study had some limitations. First, it was a single-center observational study that included only patients admitted to the hospital via the ED. Patients who had been transferred from another hospital or who died in the ED were excluded. Therefore, the generalizability of our results may be limited. Second, it included a collection of retrospective data that could introduce potential information biases and contained much missing data. Third, the sepsis diagnosis process was excluded because it was a retrospective study of patients who had already been diagnosed with an infectious disease. Inclusion criteria in our study were based on ICD-10 codes related to infection, and no blood culture reports were available. The diagnosis of septic shock was defined as sepsis with a serum lactate level >2 mmol/L, which did not reflect the patient’s volume status. Finally, there could have been inter-clinician variability in calculating the KTAS score and the MEWS score during triage.
The KMEWS, which is a combination of the MEWS and the KTAS scores, could be a useful triage tool for screening patients for septic shock in the ED. In addition, it showed acceptable predictive power for mortality or ICU admission in patients with infection. Prospective multicenter studies are necessary to validate these findings.
Supplementary materials are available at
Calculation of the Modified Early Warning Score
Definitions, related conditions, and corresponding medical actions of the Korean Triage and Acuity Scale
Analysis of receiver operating characteristics curves for predicting septic shock.
No potential conflict of interest relevant to this article was reported.
This work was supported by a research fund of Chungnam National University Hospital.
Conceptualization: SR; Data curation: SR, SKO; Formal analysis: SKO, BKL; Funding acquisition: SKO; Investigation: SKO, BKL; Methodology: SR, SJ; Project administration: SR, SJ; Resources: SR, SKO; Software: SR, SJ; Supervision: SR, SKO; Validation: SR, SKO; Visualization: SR, BKL; Writing–original draft: SR, SKO, SJ; Writing–review & editing: all authors.
All authors read and approved the final manuscript.
A flowchart of the study. Infection-related diagnosis was confirmed using the relevant International Classification of Diseases, 10th Revision codes in the medical records. ED, emergency department.
An analysis of the receiver operating characteristic curves for predicting septic shock. The areas under the receiver operating characteristic curve (AUC) of the models were calculated and tested mutually for significance using DeLong equality tests. The Korean Modified Early Warning Score (KMEWS) is the sum of the Korean Triage and Acuity Scale (KTAS) score and the Modified Early Warning Score (MEWS) score. MEDS, Mortality in Emergency Department Sepsis; CI, confidence interval.
An analysis of the receiver operating characteristic curves for predicting in-hospital mortality. The areas under the receiver operating characteristic curve (AUC) of the models were calculated and tested mutually for significance by DeLong equality tests. The Korean Modified Early Warning Score (KMEWS) is the sum of the Korean Triage and Acuity Scale (KTAS) score and the Modified Early Warning Score (MEWS) score. MEDS, Mortality in Emergency Department Sepsis; CI, confidence interval.
An analysis of the receiver operating characteristics curves for predicting intensive care unit admission. The areas under the receiver operating characteristic curve (AUC) of the models were calculated and tested mutually for significance using DeLong equality tests. The Korean Modified Early Warning Score (KMEWS) is the sum of the Korean Triage and Acuity Scale (KTAS) score and the Modified Early Warning Score (MEWS) score. MEDS, Mortality in Emergency Department Sepsis; CI, confidence interval.
The baseline characteristics of the study patients
Characteristic | Total (n = 5,093) | Without septic shock (n = 4,698) | With septic shock (n = 395) | P-value |
---|---|---|---|---|
Age (yr) | 63 (43–77) | 61 (42–76) | 78 (68–83) | < 0.001 |
Male sex | 2,528 (49.6) | 2,330 (49.6) | 198 (50.1) | 0.839 |
Initial vital sign | ||||
Systolic blood pressure (mmHg) | 124.0 (109.0–140.0) | 125.0 (111.0–140.0) | 99.0 (85.0–127.0) | < 0.001 |
Pulse rate (beats/min) | 94.0 (81.0–108.0) | 94.0 (81.0–107.0) | 106.0 (88.0–122.0) | < 0.001 |
Respiratory rate (breaths/min) | 20.0 (20.0–22.0) | 20.0 (20.0–22.0) | 24.0 (24.0–28.0) | < 0.001 |
Body temperature (°C) | 37.5 (36.8–38.4) | 37.4 (36.8–38.4) | 37.8 (36.7–38.7) | 0.060 |
Altered level of consciousness | 566 (11.1) | 280 (6.0) | 286 (72.4) | < 0.001 |
Laboratory finding | ||||
White blood cells (× 103 cells/mL) | 11,017 ± 7,179 | 10,782 ± 6,892 | 13,808 ± 9,551 | < 0.001 |
Albumin (g/dL) | 3.5 ± 0.7 | 3.6 ± 0.6 | 2.9 ± 0.7 | < 0.001 |
Blood urea nitrogen (mg/dL) | 17.4 ± 12.8 | 16.5 ± 11.6 | 28.4 ± 19.7 | < 0.001 |
Creatinine (mg/dL) | 1.14 ± 1.08 | 1.10 ± 1.06 | 1.48 ± 1.28 | < 0.001 |
C-reactive protein (mg/dL) | 8.3 ± 8.2 | 8.0 ± 8.1 | 12.0 ± 9.0 | < 0.001 |
Glucose (mg/dL) | 140.4 ± 62.6 | 138.0 ± 59.6 | 169.0 ± 86.6 | < 0.001 |
Charlson Comorbidity Index | 3 (0–5) | 2 (0–5) | 5 (4–6) | < 0.001 |
Long-term care facility | 378 (7.4) | 271 (5.8) | 107 (27.1) | < 0.001 |
MEDS | 4.50 ± 4.09 | 3.91 ± 3.57 | 11.44 ± 3.48 | < 0.001 |
MEWS | 2.96 ± 1.92 | 2.70 ± 1.69 | 6.02 ± 1.90 | < 0.001 |
≥5 | 1,082 (21.2) | 773 (16.5) | 309 (78.2) | < 0.001 |
KTAS |
< 0.001 | |||
Level 1 | 59 (1.2) | 28 (0.6) | 31 (7.8) | |
Level 2 | 507 (10.0) | 317 (6.7) | 190 (48.3) | |
Level 3 | 2,755 (54.1) | 2,596 (55.3) | 159 (40.3) | |
Level 4 | 1,645 (32.3) | 1,632 (34.7) | 13 (3.3) | |
Level 5 | 127 (2.5) | 125 (2.7) | 2 (0.5) | |
Score | 2.75 ± 0.71 | 2.68 ± 0.67 | 3.59 ± 0.70 | < 0.001 |
KMEWS |
5.71 ± 2.37 | 5.38 ± 2.06 | 9.61 ± 2.23 | < 0.001 |
≥7 | 1,691 (33.2) | 1,328 (28.3) | 363 (91.9) | < 0.001 |
Intensive care unit admission | 250 (4.9) | 142 (3.0) | 108 (27.3) | < 0.001 |
In-hospital mortality | 200 (3.9) | 127 (2.7) | 73 (18.5) | < 0.001 |
Values are presented as median (interquartile range) or mean±standard deviation for continuous variables and number (%) for categorical variables.
MEDS, Mortality in Emergency Department Sepsis; MEWS, Modified Early Warning Score; KTAS, Korean Triage Acuity Scale; KMEWS, Korean Modified Early Warning Score.
KTAS scores for levels 1, 2, 3, 4, 5 are 5, 4, 3, 2, 1, respectively.
KMEWS is the sum of the KTAS score and the MEWS score.
A multivariable logistic regression analysis of the factors associated with septic shock in patients in the emergency department
Variable | Odds ratio | 95% Confidence interval | P-value |
---|---|---|---|
Age | 1.04 | 1.03–1.05 | < 0.001 |
Long-term care facility (yes) | 2.97 | 2.11–4.17 | < 0.001 |
Korean Triage Acuity Scale score | 2.18 | 1.76–2.70 | < 0.001 |
Modified Early Warning Score | 2.09 | 1.93–2.27 | < 0.001 |