| Home | E-Submission | Sitemap | Contact Us |  
Search
Clin Exp Emerg Med Search

CLOSE

Clin Exp Emerg Med > Volume 11(3); 2024 > Article
Lee, Ryu, Jeon, Park, You, Jeong, Cho, Ahn, Kang, and Oh: Comparison of modified quick Sequential Organ Failure Assessment models as triage tools for febrile patients

Abstract

Objective

Effective triage of febrile patients in the emergency department is crucial during times of overcrowding to prioritize care and allocate resources, especially during pandemics. However, available triage tools often require laboratory data and lack accuracy. We aimed to develop a simple and accurate triage tool for febrile patients by modifying the quick Sequential Organ Failure Assessment (qSOFA) score.

Methods

We retrospectively analyzed data from 7,303 febrile patients and created modified versions of qSOFA using factors identified through multivariable analysis. The performance of these modified qSOFAs in predicting in-hospital mortality and intensive care unit (ICU) admission was compared using the area under the receiver operating characteristic curve (AUROC).

Results

Through multivariable analysis, the identified factors were age (“A” factor), male sex (“M” factor), oxygen saturation measured by pulse oximetry (SpO2; “S” factor), and lactate level (“L” factor). The AUROCs of ASqSOFA (in-hospital mortality: 0.812 [95% confidence interval, 0.789–0.835]; ICU admission: 0.794 [95% confidence interval, 0.771–0.817]) were simple and not inferior to those of other more complex models (e.g., ASMqSOFA, ASLqSOFA, and ASMLqSOFA). ASqSOFA also displayed significantly higher AUROC than other triage scales, such as the Modified Early Warning Score and Korean Triage and Acuity Scale. The optimal cutoff score of ASqSOFA for the outcome was 2, and the score for redistribution to a lower level emergency department was 0.

Conclusion

We demonstrated that ASqSOFA can be employed as a simple and efficient triage tool for emergency febrile patients to aid in resource distribution during overcrowding. It also may be applicable in prehospital settings for febrile patient triage.

INTRODUCTION

Fever is a principle reason for patients to visit the emergency department (ED) [1,2]. Recently, during the COVID-19 pandemic, febrile patients underwent screening and isolation in the ED to prevent the transmission of infection. Consequently, patients with fever received slower ED treatment than in the pre-pandemic period, which resulted in an increase in the mortality and morbidity rates of these patients [3,4]. This situation can be attributed to a limited medical capacity and high demand for treatment. Therefore, patient classification is crucial for effectively distributing limited resources.
However, studies assessing triage tools for febrile patients are lacking. Most scoring tools for febrile patients in the ED require laboratory data and were developed primarily for sepsis prognosis [57]. These tools have been used for support of patients after emergency management, such as application of vasopressors and mechanical ventilation. For this reason, these studies have many limitations in their application in triage of patients. Consequently, the development of simple triage tools for febrile patients is required.
The quick Sequential Organ Failure Assessment (qSOFA) score has been employed for the early detection of sepsis based on simple vital signs. However, it has demonstrated low accuracy and inadequacies as a prognostic or triage tool [8,9]. The best prognostic tool for infections is controversial [1015]. Previously, it has been demonstrated that modified versions of qSOFA can serve as useful tools for predicting mortality in patients with infections [1517]. In particular, Liu et al. [15] utilized lactate level with qSOFA in patients with sepsis and observed an increased area under the receiver operating characteristic curve (AUROC) value for mortality. However, this approach has limitations in triage because of the need for laboratory data.
Therefore, we conducted this study to investigate the potential effectiveness of modified versions of qSOFA as simple triage tools for febrile patients in the ED and to identify the most suitable version of the modified qSOFA for this application.

METHODS

Ethics statement

This study was approved by the Institutional Review Board of Chungnam National University Hospital (No. 2021-03-024). The extracted data included in the study only clinical data and did not include any personally identifiable information. Therefore, the need for informed consent was waived.

Study design

In this retrospective study, we reviewed secondary data extracted from electronic medical records (EMRs). The study included patients aged 19 years or older who visited the ED with fever (body temperature, ≥37.5 °C) during the period from April 2021 to December 2022. Body temperature was measured in the triage area, and tympanic temperature was measured with the Thermoscan ExacTemp (model IRT 6510, Braun). Chungnam National University Hospital (Daejeon, Korea) is a tertiary care university hospital with 1,350 beds, and approximately 50,000 patients visit the ED per year.

Data collection

To identify the effective factors of the modified qSOFA that could be used as a triage tool for febrile patients, we extracted data from the EMR collected during the triage process. The extracted data were age, sex, vital signs, mental status, oxygen saturation measured by pulse oximetry (SpO2), medical history, level of medical facility of patient transfer, ED length of stay (LOS; the time interval between the arrival to and departure from the ED), and the Korean Triage and Acuity Scale (KTAS) (Supplementary Table 1). Furthermore, to compare the performances of the modified qSOFAs with and without the incorporation of lactate level, we extracted data on initial serum lactate level (6,620 of 7,303 [90.6%]) from the laboratory data. Using the extracted data, we calculated the values of qSOFA and the Modified Early Warning Score (MEWS). Patients who were transferred to other medical facilities and those with missing data that could affect outcome determination were excluded from the analysis.
The qSOFA score was modified by incorporating factors such as age, male sex, SpO2, and lactate level. Each modified version of qSOFA was given a name based on the initials of the added factor: “A” for age, “L” for lactate level, “M” for male sex, and “S” for SpO2. The modified qSOFAs were compared to determine the most suitable version for triage. To ensure an accurate comparison, patients whose lactate values were not available were excluded from the analysis of the modified versions of qSOFA. Furthermore, we compared the efficacy of other triage tools, such as MEWS and KTAS, with the most appropriately modified version of the qSOFA. The outcome was the overall cause of in-hospital mortality and intensive care unit (ICU) admission.

Statistical analysis

Categorical variables were analyzed using the chi-square test or Fisher exact test, and continuous variables were analyzed using the Mann-Whitney U-test. Categorical variables are expressed as numbers (%), and continuous variables are expressed as medians (interquartile ranges). After univariable analysis, multivariable logistic regression analysis was performed to identify independent prognostic factors. The variables with a P-value of <0.1 in the univariable analysis that were not included in qSOFA were selected for the multivariable logistic regression model. Among these variables, continuous variables were categorized based on commonly used cutoff values [18] or the Youden index. Receiver operating characteristic curve analysis was performed for the outcome variables. The AUROC and cutoff values (Youden index) were obtained for individual variables. DeLong test was performed to confirm the differences between the AUROCs of the variables (modified qSOFAs, MEWS, and KTAS). AUROCs are expressed as numbers and 95% confidence intervals (CIs). All statistical analyses were performed using R ver. 4.2.2 (R Foundation for Statistical Computing). P-values of <0.05 were considered statistically significant.

RESULTS

Patient demographics

In total, 7,303 patients were enrolled. The overall mortality and ICU admission rates were 4.5% (325 patients) and 4.8% (351 patients), respectively. The median age was 64 years (range, 43–77 years). The basic characteristics of the patients and a comparison of variables based on in-hospital mortality and ICU admission are presented in Table 1 and Supplementary Table 2.

Multivariable logistic regression analysis

The variables included in the multivariable analysis were age, sex, diastolic blood pressure, heart rate, SpO2, lactate level, qSOFA, and transfer from a long-term care facility. As systolic blood pressure, respiration rate, and mental status were already included in the qSOFA, these factors and MEWS were not included in the multivariable analysis. The results demonstrate that age ≥65 years, lactate level ≥2 mmol/L, male sex, and SpO2 <95% had a significant effect on both outcomes (Table 2). Therefore, these four factors were used to modify the qSOFA score, and each additional factor was assigned one point for easy calculation, despite the variable adjusted odds ratios.

Comparison of the AUROCs of modified qSOFAs and other severity scales

To identify the most appropriate version of the modified qSOFAs for triaging febrile patients, AUROCs of the modified qSOFAs were compared (Fig. 1 and Supplementary Table 3). Among the qSOFA models modified with a single factor, only SqSOFA exhibited superior AUROCs for predicting both outcomes to the AUROCs of qSOFA (in-hospital mortality: 0.793 [95% CI, 0.769–0.817] vs. 0.764 [95% CI, 0.739–0.788], P<0.001; ICU admission: 0.797 [95% CI, 0.774–0.820] vs. 0.784 [95% CI, 0.762–0.806], P<0.001). Among the modified qSOFA models that incorporated complex factors, the AUROC of ASqSOFA for in-hospital mortality (0.812; 95% CI, 0.789–0.835) did not differ significantly from those of more complex models ASMqSOFA (0.810; 95% CI, 0.786–0.834; P=0.695), ASLqSOFA (0.821; 95% CI, 0.798–0.843; P=0.066), and ASMLqSOFA (0.819; 95% CI, 0.797–0.842; P=0.242). Furthermore, the AUROC for ICU admission obtained using ASqSOFA (0.794; 95% CI, 0.771–0.817) was either significantly superior or not significantly different from those of the other modified qSOFAs, except the AUROCs of ASLqSOFA (0.808; 95% CI, 0.786–0.830; P=0.004) and ASMLqSOFA (0.810; 95% CI, 0.789–0.831; P=0.020), both of which incorporated lactate level. Moreover, the AUROC of ASqSOFA demonstrated a significantly superior performance to other severity scales, such as MEWS and KTAS.

Cutoff values

We determined the optimal cutoff value for ASqSOFA (Table 3). The Youden index scores were analyzed, and the optimal ASqSOFA score for in-hospital mortality was 2, even for ICU admission.
When comparing the cutoff values for low mortality between ASqSOFA and KTAS, an ASqSOFA score of 0 showed a mortality rate of 0.6% (14 of 2,495), whereas KTAS 4 and 5 showed mortality rates of 0.8% (14 of 1,649) and 1.0% (1 of 98), respectively (Table 4).

DISCUSSION

Patients that present at the ED with fever are sometimes diagnosed with an infectious disease. Predicting the possibility of sepsis and septic shock in these infected patients is crucial for employing intensive treatments, such as the 1-hour bundle [10]. The COVID-19 pandemic resulted in the overcrowding of EDs, leading to increased mortality among infected patients [3]. In such a situation, it is crucial to ensure that febrile patients are directed to an appropriate medical facility based on the severity of their condition. Therefore, in this study, we focused on developing a tool that can effectively classify febrile patients by predicting mortality during triage. Table 2 demonstrate that variables of age, male sex, SpO2, and lactate level exhibited significant odds ratios for in-hospital mortality and ICU admissions. Consequently, the modified qSOFAs were constructed using these four variables. Our findings reveal that ASqSOFA, which includes age and SpO2 level in addition to qSOFA, exhibited superior predictive performance for outcomes, except to those of ASLqSOFA and ASMLqSOFA
Numerous studies have consistently identified age as an independent prognostic variable for infection [19,20]. Furthermore, several prognostic tools have used age as a key factor. In our study, age also was identified as an independent prognostic factor in febrile patients. When age was included as a modification factor in qSOFA, it led to a significant improvement in the AUROC, indicating its enhanced ability to predict the evaluated outcome.
The influence of sex on sepsis prognosis remains a debated topic. While ongoing discussions and conflicting research findings exist, certain studies have suggested potential differences in outcomes between male and female patients [21,22]. In our study, we observed sex-based differences in the prognoses of febrile patients (Tables 1, 2). However, when sex was included as a factor in the modification of qSOFA, it did not result in improved predictions (Fig. 1). This discrepancy might be caused by the low adjusted odds ratio of sex (in-hospital mortality, 1.41 [95% CI, 1.10–1.81]; ICU admission, 1.35 [95% CI, 1.06–1.71]) compared with other factors. Therefore, modified qSOFAs that do not include sex as a factor may be a more appropriate tool.
SpO2 alone has not been used as a prognostic factor as its measurement can be influenced by factors such as poor perfusion, changes in oxygen supply, and other respiratory or cardiovascular conditions. However, SpO2 can be valuable when combined with other clinical indicators to assess a patient's overall clinical profile. In several studies, the usefulness of SpO2-related parameters, such as the SpO2 to fraction of inspired oxygen (FiO2) ratio and respiratory rate-oxygenation (ROX) index, has been analyzed [2325]. However, calculation of the SpO2/FiO2 ratio and ROX index and division of patients into subgroups according to the severity of the index values is complex and requires memorization of complex values. Additionally, most patients arriving at triage do not receive oxygen, except in cases transferred by the Emergency Medical Service system. Therefore, SpO2 alone, without FiO2, could be a reliable indicator of the respiratory status of patients during triage. Complex indices, such as the SpO2/FiO2 ratio and ROX index, may be more appropriate than SpO2 alone for use in critical care areas of the ED, where advanced monitoring and interventions are available. In this study, we evaluated SpO2 as a prognostic factor for triage. Despite the limitations associated with SpO2 measurements, its incorporation into the modified qSOFA resulted in a significant improvement in AUROC, indicating its added value in predicting the evaluated outcome (Fig. 1). In addition, compared to the findings of previous studies [16], our results show that the AUROC of ASqSOFA for in-hospital mortality (0.812; 95% CI, 0.789–0.835) may not be inferior to that of the modified qSOFA using SpO2/FiO2 ratio (0.805; 95% CI, 0.776–0.833).
Of the four prognostic factors examined in our study, age, sex, and SpO2 were commonly used in the triage data and can be incorporated efficiently into a prognostic triage tool for ED use. However, lactate level is not employed widely in ED triage due to its reliance on laboratory machines and blood samples, rendering it impractical for routine use. Moreover, considering the cost of lactate testing, conducting this test for all febrile patients in triage may not be feasible. Other inflammatory markers such as procalcitonin or interleukin-6 were excluded because they have additional limitations to lactate. Our study AUROC showed significant improvement when lactate level was included in the modified qSOFA score. This finding suggests that incorporating lactate level as an additional factor in the assessment of febrile patients can enhance the predictive accuracy of the modified qSOFA for the evaluated outcomes, as in a previous study [15]. Moreover, lactate level is an important indicator of septic shock and can be used as an indicator of mortality [26,27]. Liu et al. [15] utilized LqSOFA in patients with sepsis and observed an AUROC value for mortality of 0.751. However, comparison of the modified qSOFAs, in which lactate level was incorporated, with ASqSOFA revealed that the modified qSOFAs had a higher AUROC than ASqSOFA for ICU admissions and did not significantly outperform ASqSOFA in terms of in-hospital mortality (Fig. 1 and Supplementary Table 3). Consequently, it may be more appropriate to use lactate as a prognostic factor for patients with a high triage score than to apply it to all febrile patients
We compared ASqSOFA with the existing severity scores, such as MEWS and KTAS. These established scoring systems had lower AUROCs than the ASqSOFA in predicting both in-hospital mortality and ICU admissions. Moreover, these scoring systems are more complex, requiring a greater number of variables and calculations than those for the ASqSOFA. Overall, the ASqSOFA may be a suitable triage tool for febrile patients in the ED because of its simplicity (involving only five components: age, SpO2, mental status, systolic blood pressure, and respiration rate) and ease of use.
When the ASqSOFA score reached 2 or higher, the probabilities of in-hospital mortality and ICU admission increased (Table 3); therefore, these patients should be monitored closely in the ED. In contrast, patients with ASqSOFA scores of 0 or 1 may be considered for referral to low-level hospitals rather than high-level hospitals in their region. The number of patients with an ASqSOFA score of 0 was 2,495 of 7,303 (34.2%), and the in-hospital mortality rate was only 0.6% (14 of 2,495). In comparison, 98 of 7,303 patients (1.3%) were classified as KTAS level 5, with an in-hospital mortality rate of 1.0% (1 of 97) (Table 4). Moreover, 1,747 of 7,303 patients (23.9%) were classified as KTAS level 4 or 5, with an in-hospital mortality rate of 0.9% (15 of 1,747) (Table 4). These findings indicate that the ASqSOFA may be a more appropriate tool than the KTAS for redistributing low-risk febrile patients to low-level hospitals. Similarly, ASqSOFA may be more appropriate than other modified qSOFAs. Additionally, the positive predictive value increased for both in-hospital mortality and ICU admission when the ASqSOFA score was 4 or higher, emphasizing the need for prompt and intensive management of such patients. Evidently, ASqSOFA enables appropriate triage of febrile patients considering symptom severity, easing the strain on overcrowded EDs and ensuring effective utilization of healthcare resources.
This study has some limitations. First, this was a retrospective study constrained by limited triage data. Consequently, detailed past medical histories, including specific factors such as immunosuppression, were not included in the scoring. Second, vasopressors or oxygen administered during transportation can potentially affect vital signs. However, in the case of vasopressors, infusion pumps were not always applied; thus, the actual application status remains uncertain. In cases of oxygen supply, although oxygen was administered for a short period at triage, SpO2 was measured in the absence of oxygen because of a lack of portable oxygen. Therefore, we believe that errors related to these two aspects would be minimal. Third, there may have been a selection bias. In particular, data collection was limited to individuals aged 19 years or older, raising concerns about the applicability of the triage scoring system for children and adolescents. However, a study focused on age-dependent modified qSOFA scores, specifically for children, was conducted previously [28]. Although the current study did not directly address triage scoring for children, it can offer valuable insights into the prediction of outcomes in the adult population. Fourth, the generalizability of the results may be limited because the study was conducted at a single center. Fifth, many patients with do-not-attempt resuscitation orders were not admitted to the ICU, which could have introduced potential confounding variables. If patients with do-not-attempt resuscitation were included in the analysis, the sensitivity and specificity of the modified qSOFA for predicting ICU admissions could have been improved.
In summary, ASqSOFA demonstrated its potential as a valuable tool in triage for predicting mortality and efficiently classifying febrile patients. Its simplicity and lack of requirement for additional laboratory tests make it a practical option in triage settings. Furthermore, in overcrowded situations, ASqSOFA can aid in the redistribution of febrile patients to appropriate levels of care. However, further investigation and validation in other settings are needed.

NOTES

Conflicts of interest
The authors have no conflicts of interest to declare.
Funding
The authors received no financial support for this study.
Data availability
Data analyzed in this study are available from the corresponding author upon reasonable request.
Author contributions
Conceptualization: SR, DYL, SYJ; Data curation: SR, DYL, SYJ, SKO; Formal analysis: SR, DYL; Methodology: all authors; Visualization: SR, DYL; Writing–original draft: SR, DYL; Writing–review & editing: all authors. All authors read and approved the final manuscript.

Supplementary materials

Supplementary materials are available from https://doi.org/10.15441/ceem.23.125.

Supplementary Table 1.

Definitions, related conditions, and corresponding medical actions of the KTAS
ceem-23-125-supplementary-Table-1.pdf

Supplementary Table 2.

Baseline characteristics according to the ICU admission
ceem-23-125-supplementary-Table-2.pdf

Supplementary Table 3.

Comparison of the AUROCs for the in-hospital mortality
ceem-23-125-supplementary-Table-3.pdf

REFERENCES

1. Grandey KA. Fever; In: Sherman SC, Weber JM, Schindlbeck MA, Rahul GP, editors. Clinical emergency medicine. McGraw-Hill Education; 2014.

2. Knott JC, Tan SL, Street AC, Bailey M, Cameron P. Febrile adults presenting to the emergency department: outcomes and markers of serious illness. Emerg Med J 2004; 21:170-4.
crossref pmid pmc
3. Eswaran V, Wang RC, Vashi AA, Kanzaria HK, Fahimi J, Raven MC. Patient reported delays in obtaining emergency care during COVID19. Am J Emerg Med 2022; 56:306-9.
crossref pmid
4. Lee SJ, Choi A, Ryoo HW, Pak YS, Kim HC, Kim JH. Changes in clinical characteristics among febrile patients visiting the emergency department before and after the COVID-19 outbreak. Yonsei Med J 2021; 62:1136-44.
crossref pmid pmc pdf
5. Shapiro NI, Wolfe RE, Moore RB, Smith E, Burdick E, Bates DW. Mortality in emergency department sepsis (MEDS) score: a prospectively derived and validated clinical prediction rule. Crit Care Med 2003; 31:670-5.
crossref pmid
6. Macdonald SP, Arendts G, Fatovich DM, Brown SG. Comparison of PIRO, SOFA, and MEDS scores for predicting mortality in emergency department patients with severe sepsis and septic shock. Acad Emerg Med 2014; 21:1257-63.
pmid
7. Garbero RF, Simoes AA, Martins GA, Cruz LVD, von Zuben VG. SOFA and qSOFA at admission to the emergency department: Diagnostic sensitivity and relation with prognosis in patients with suspected infection. Turk J Emerg Med 2019; 19:106-10.
crossref pmid pmc
8. Sabir L, Ramlakhan S, Goodacre S. Comparison of qSOFA and Hospital Early Warning Scores for prognosis in suspected sepsis in emergency department patients: a systematic review. Emerg Med J 2022; 39:284-94.
crossref pmid
9. Jiang J, Yang J, Jin Y, Cao J, Lu Y. Role of qSOFA in predicting mortality of pneumonia: a systematic review and meta-analysis. Medicine (Baltimore) 2018; 97:e12634.
pmid pmc
10. Rhodes A, Evans LE, Alhazzani W, et al. Surviving Sepsis Campaign: international guidelines for management of sepsis and septic shock. 2016. Intensive Care Med 2017; 43:304-77.
pmid
11. Azijli K, Minderhoud T, Mohammadi P, et al. A prospective, observational study of the performance of MEWS, NEWS, SIRS and qSOFA for early risk stratification for adverse outcomes in patients with suspected infections at the emergency department. Acute Med 2021; 20:116-24.
crossref pmid
12. Raith EP, Udy AA, Bailey M, et al. Prognostic accuracy of the SOFA score, SIRS criteria, and qSOFA score for in-hospital mortality among adults with suspected infection admitted to the intensive care unit. JAMA 2017; 317:290-300.
crossref pmid
13. Wang C, Xu R, Zeng Y, Zhao Y, Hu X. A comparison of qSOFA, SIRS and NEWS in predicting the accuracy of mortality in patients with suspected sepsis: a meta-analysis. PLoS One 2022; 17:e0266755.
crossref pmid pmc
14. Wattanasit P, Khwannimit B. Comparison the accuracy of early warning scores with qSOFA and SIRS for predicting sepsis in the emergency department. Am J Emerg Med 2021; 46:284-8.
crossref pmid
15. Liu S, He C, He W, Jiang T. Lactate-enhanced-qSOFA (LqSOFA) score is superior to the other four rapid scoring tools in predicting in-hospital mortality rate of the sepsis patients. Ann Transl Med 2020; 8:1013.
crossref pmid pmc
16. Guarino M, Gambuti E, Alfano F, et al. Predicting in-hospital mortality for sepsis: a comparison between qSOFA and modified qSOFA in a 2-year single-centre retrospective analysis. Eur J Clin Microbiol Infect Dis 2021; 40:825-31.
crossref pmid pmc pdf
17. Xia Y, Zou L, Li D, et al. The ability of an improved qSOFA score to predict acute sepsis severity and prognosis among adult patients. Medicine (Baltimore) 2020; 99:e18942.
crossref pmid pmc
18. American Thoracic Society; American College of Chest Physicians. ATS/ACCP statement on cardiopulmonary exercise testing. Am J Respir Crit Care Med 2003; 167:211-77.
crossref pmid
19. Martin GS, Mannino DM, Moss M. The effect of age on the development and outcome of adult sepsis. Crit Care Med 2006; 34:15-21.
crossref pmid
20. Innocenti F, Tozzi C, Donnini C, et al. SOFA score in septic patients: incremental prognostic value over age, comorbidities, and parameters of sepsis severity. Intern Emerg Med 2018; 13:405-12.
crossref pmid pdf
21. Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med 2003; 348:1546-54.
crossref pmid
22. Thompson KJ, Finfer SR, Woodward M, Leong RNF, Liu B. Sex differences in sepsis hospitalisations and outcomes in older women and men: a prospective cohort study. J Infect 2022; 84:770-6.
crossref pmid
23. Fukuda Y, Tanaka A, Homma T, et al. Utility of SpO2/FiO2 ratio for acute hypoxemic respiratory failure with bilateral opacities in the ICU. PLoS One 2021; 16:e0245927.
crossref pmid pmc
24. Alberdi-Iglesias A, Martin-Rodriguez F, Ortega Rabbione G, et al. Role of SpO2/FiO2 ratio and ROX index in predicting early invasive mechanical ventilation in COVID-19. A pragmatic, retrospective, multi-center study. Biomedicines 2021; 9:1036.
crossref pmid pmc
25. Roca O, Caralt B, Messika J, et al. An index combining respiratory rate and oxygenation to predict outcome of nasal high-flow therapy. Am J Respir Crit Care Med 2019; 199:1368-76.
crossref pmid
26. Singer M, Deutschman CS, Seymour CW, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA 2016; 315:801-10.
pmid pmc
27. Lee SM, An WS. New clinical criteria for septic shock: serum lactate level as new emerging vital sign. J Thorac Dis 2016; 8:1388-90.
crossref pmid pmc
28. Romaine ST, Potter J, Khanijau A, et al. Accuracy of a modified qSOFA score for predicting critical care admission in febrile children. Pediatrics 2020; 146:e20200782.
crossref pmid pdf

Fig. 1.
Comparison of the area under the receiver operating characteristic curves (AUROCs) for predicting the outcome. (A–C) In-hospital mortality. (D–F) Intensive care unit admission. (A, D) Modified with single factor. (B, E) Modified with complex factors. (C, F) Severity scores. The AUROCs of the variables were calculated and tested mutually for significance by DeLong tests. Variables are expressed as AUROC (95% confidence internal). Each modified version of the quick Sequential Organ Failure Assessment (qSOFA) was given a name based on the initials of the added factors: “A” for age, “L” for lactate level, “M” for male sex, and “S” for oxygen saturation measured by pulse oximetry. MEWS, Modified Early Warning Score; KTAS, Korean Triage and Acuity Scale.
ceem-23-125f1.jpg
Table 1.
Baseline characteristics according to the in-hospital mortality
Characteristic Survivor (n=6,978) Nonsurvivor (n=325) Total (n=7,303) P-value
Age (yr) 63 (41–77) 77 (68–82) 64 (43–77) <0.001
Male sex 3,308 (47.4) 203 (62.5) 3,511 (48.1) <0.001
Vital sign
 Systolic blood pressure (mmHg) 129.0 (114.0–147.0) 119.0 (100.0–139.0) 129.0 (113.0–147.0) <0.001
 Diastolic blood pressure (mmHg) 77.0 (69.0–86.0) 70.0 (59.0–81.0) 77.0 (68.0–86.0) <0.001
 Mean arterial pressure (mmHg) 95.0 (85.0–105.0) 86.7 (73.7–100.3) 94.7 (84.7–104.8) <0.001
 Heart rate (beats/min) 102 (89–115) 108 (93–125) 102 (89–116) <0.001
 Respiration rate (breaths/min) 20 (20–22) 26 (20–32) 20 (20–22) <0.001
 Body temperature (°C) 38.0 (37.7–38.6) 38.1 (37.8–38.6) 38.0 (37.7–38.6) 0.117
 SpO2 (%) 97 (96–98) 96 (93–98) 97 (96–98) <0.001
 Glasgow Coma Scale 15 (15–15) 15 (12–15) 15 (15–15) <0.001
Laboratory data (lactate level, mmol/L) 1.8 (1.4–2.5) 3.0 (1.9–4.7) 1.8 (1.4–2.6) <0.001
Severity scale
 KTAS 3 (3–3) 3 (2–3) 3 (3–3) <0.001
 MEWS 3 (2–5) 5 (3–6) 3 (2–5) <0.001
 qSOFA 0 (0–1) 1 (1–2) 0 (0–1) <0.001
Long-term care facility 179 (2.6) 24 (7.4) 203 (2.8) <0.001
Admission 3,531 (50.6) 296 (91.1) 3,827 (52.4) <0.001
 Intensive care unit 283 (4.1) 68 (20.9) 351 (4.8) <0.001
ED LOS (hr) 5.3 (3.5–8.1) 7.0 (5.0–10.4) 5.4 (3.5–8.2) <0.001

Values are presented as median (interquartile range) or number (%).

SpO2, oxygen saturation measured by pulse oximetry; KTAS, Korean Triage and Acuity Scale; MEWS, Modified Early Warning Score; qSOFA, quick Sequential Organ Failure Assessment; ED, emergency department; LOS, length of stay.

Table 2.
Multivariable logistic regression analysis for the outcome
Variable In-hospital mortality
ICU admission
Adjusted OR (95% CI) P-value Adjusted OR (95% CI) P-value
Age ≥65 yr 2.73 (2.02–3.75) <0.001 1.43 (1.11–1.86) 0.007
Male sex 1.41 (1.10–1.81) 0.005 1.35 (1.06–1.71) 0.015
DBP <50 mmHg 1.41 (0.87–2.24) 0.156 2.64 (1.73–3.98) <0.001
Heart rate >90 beats/min 0.92 (0.69–1.23) 0.552 0.92 (0.70–1.24) 0.586
SpO2 <95% 3.32 (2.52–4.35) <0.001 2.05 (1.54–2.72) <0.001
qSOFA
 0 Reference - Reference -
 1 3.68 (2.59–5.33) <0.001 5.46 (3.80–8.06) <0.001
 2 7.01 (4.71–10.57) <0.001 12.32 (8.23–18.81) <0.001
 3 9.65 (5.04–18.19) <0.001 11.21 (5.85–21.20) <0.001
Long-term care facility 1.26 (0.76–2.03) 0.352 0.86 (0.49–1.43) 0.579
Lactate level ≥2 mmol/L 1.99 (1.52–2.62) <0.001 1.90 (1.47–2.48) <0.001

Adjusted ORs were the result of the multivariable logistic regression analysis by using the stepwise selection method.

ICU, intensive care unit; OR, odds ratio; CI, confidence interval; DBP, diastolic blood pressure; SpO2, oxygen saturation measured by pulse oximetry; qSOFA, quick Sequential Organ Failure Assessment.

Table 3.
Cutoff values of ASqSOFA
Cutoff Sensitivity (95% CI) Specificity (95% CI) PPV (95% CI) NPV (95% CI)
In-hospital mortality
 0 1.000 (0.989–1.000) 0 (0–0.005) 0.045 (0.040–0.049) -
 1 0.957 (0.929–0.976) 0.356 (0.344–0.367) 0.065 (0.058–0.072) 0.994 (0.991–0.997)
 2 0.803 (0.756–0.845) 0.714 (0.704–0.725) 0.116 (0.103–0130) 0.987 (0.984–0.990)
 3 0.498 (0.443–0.554) 0.905 (0.898–0.912) 0.197 (0.170–0.225) 0.974 (0.971–0.979)
 4 0.209 (0.166–0.258) 0.981 (0.978–0.984) 0.342 (0.276–0.412) 0.963 (0.959–0.968)
 5 0.040 (0.021–0.067) 0.996 (0.997–0.999) 0.565 (0.345–0.768) 0.957 (0.952–0.962)
ICU admission
 0 1.000 (0.990–1.000) 0 (0–0.001) 0.048 (0.043–0.053) -
 1 0.943 (0.913–0.965) 0.356 (0.345–0.367) 0.069 (0.062–0076) 0.992 (0.998–0.995)
 2 0.767 (0.719–0.810) 0.714 (0.704–0.725) 0.119 (0.106–0.133) 0.984 (0.980–0.987)
 3 0.453 (0.400–0.507) 0.904 (0.897–0.911) 0.193 (0.167–0.222) 0.970 (0.966–0.974)
 4 0.162 (0.125–0.205) 0.980 (0.976–0.983) 0.286 (0.225–0.355) 0.939 (0.954–0.963)
 5 0.020 (0.008–0.041) 0.998 (0.996–0.999) 0.304 (0.132–0.529) 0.953 (0.948–0.958)

ASqSOFA, quick Sequential Organ Failure Assessment modified with age and oxygen saturation measured by pulse oximetry; CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value; ICU, intensive care unit.

Table 4.
Mortality proportion of ASqSOFA and KTAS
Severity scale Survivora) (n=6,978) Nonsurvivora) (n=325) Totalb) (n=7,303)
ASqSOFA
 0 2,481 (99.4) 14 (0.6) 2,495 (34.2)
 1 2,504 (98.0) 50 (2.0) 2,554 (35.0)
 2 1,331 (93.1) 99 (6.9) 1,430 (19.6)
 3 531 (84.9) 94 (15.1) 625 (8.5)
 4 121 (68.8) 55 (31.2) 176 (2.4)
 5 10 (43.5) 13 (56.5) 23 (0.3)
KTAS
 1 76 (75.2) 25 (24.8) 101 (1.4)
 2 555 (81.9) 122 (18.0) 677 (9.3)
 3 4,615 (96.6) 163 (3.4) 4,778 (65.4)
 4 1,635 (99.2) 14 (0.8) 1,649 (22.6)
 5 97 (99.0) 1 (1.0) 98 (1.3)

Values are presented as number (%).

ASqSOFA, quick Sequential Organ Failure Assessment modified with age and oxygen saturation measured by pulse oximetry; KTAS, Korean Triage and Acuity Scale.

a)Proportion between survivors and nonsurvivors.

b)Proportion between each scale scores.

Editorial Office
The Korean Society of Emergency Medicine
101-3104, Brownstone Seoul, 464 Cheongpa-ro, Jung-gu, Seoul 04510, Korea
TEL: +82-31-709-0918   E-mail: office@ceemjournal.org
About |  Browse Articles |  Current Issue |  For Authors and Reviewers
Copyright © by The Korean Society of Emergency Medicine.                 Developed in M2PI