Sepsis-related mortality with SOFA and qSOFA in emergency department patients

Abstract

AimPrediction of sepsis-related mortality in the emergency department (ED) is important. In this study, we aimed to assess the predictive power of the newly defined scoring systems in sepsis-related mortality and reduce it in the ED.MethodsA prospective cohort study was conducted on a sample of patients who presented to the ED with sepsis. Patients aged <18 years and those with shock from non-septic causes were excluded. Age, vital signs, laboratory findings on admission, culture time, time of empiric antibiotic therapy, results of scoring systems, duration of ED stay and hospitalization, focus of infection and clinical outcome were recorded.ResultsA total of 48 patients were enrolled in the study. SOFA scores were higher in patients who died (p = 0.001, 95% CI, 0.639–0.902). The best cut-off point for diagnostic performance was a SOFA score of 4.5. At this point, sensitivity was 82.61%, specificity was 56.0%, positive predictive value was 63.3% and negative predictive value was 77.8%.DiscussionqSOFA and SIRS cannot provide adequate prognostic information in the ED, whereas, SOFA reliably predicted mortality. Our results indicate that vital signs are more flexible and efficient data sources. Although it is presently not precisely understood how RDW is associated with clinical outcomes but patients with increased RDW levels should be more aggressively treated and admission RDW could also be used for prognostic purposes, particularly in busy EDs. Also, lactate levels were correlated with SOFA and qSOFA scores and that the former could predict mortality (p= 0.012) is consistent with previous studies of infectionConclusionIn conclusion, qSOFA had poor performance for the prediction of sepsis-related mortality in the ED. SOFA had the best performance.

Keywords

Introduction

Sepsis refers to organ dysfunction threatening the host’s life that results from an impaired host response elicited by the culprit infection and remains a major concern in ED as a result of determining the mortality in one-hour-therapeutic management, so the emergency physicians play a major role.^1-2 Delay in adequate treatment, subsequently impacts mortality and cost in sepsis.³ According to Sepsis 3, SOFA is mainly a clinical diagnostic criterion; qSOFA is a screening tool to predict sepsis-related outcomes.^3-5 Recently many studies have tried to figure out the diagnostic and prognostic value of these newly defined scoring tools.^6-10 No consensus has been reached, because some studies are based on the ED septic patients and others on patients seen outside. Because emergency physicians are the first to encounter patients with sepsis, tests and tools that can be used for bedside diagnostic and prognostic purposes are desirable. We conducted a study to assess the predictive power of the newly defined scoring systems SOFA and qSOFA compared with traditional SIRS criteria for sepsis-related mortality in the ED patients.

Materials and Methods

This single-center, prospective cohort study was assessed between November 1, 2017 and March 31, 2018 in the ED of an academic tertiary care hospital. The study was approved by the local ethics committee. Consent for participation in the study was obtained from the patients or from the relatives of patients who could not give consent because of unconsciousness, mental retardation, psychiatric illness, or Alzheimer’s or other dementia. Data were obtained from hospital electronic records and patient follow-up forms. The subjects were examined by ED physicians for assessment and fulfillment of the clinical criteria for severe sepsis or septic shock according to the guidelines of the Surviving Sepsis Campaign and were subsequently admitted to the hospital between the dates indicated (n = 187). Patients who were diagnosed with sepsis and hospitalized or discharged from the ED were included in the study. The extracted data included the presence of SIRS criteria, the qSOFA and SOFA scores and the time required to meet the criteria. We excluded patients younger than 18 years, patients referred from outside facilities and patients with shock from non-septic causes, such as cardiogenic shock, left heart failure, right heart failure, arrhythmia, acute coronary syndrome, pulmonary embolism, tension pneumothorax, cardiac tamponade, hypovolemic shock, vasodilatory (distributive) shock, neurogenic (spinal) shock, adrenal shock and anaphylaxis. A study form was used that included age, vital signs (blood pressure, heart rate, body temperature, oxygen saturation and shock index) and laboratory findings (complete blood count, biochemistry and blood gas measurements) on admission and culture time, time of empiric antibiotic therapy, results of scoring systems (SOFA, qSOFA, and SIRS), duration of ED stay and hospitalization, focus of infection and in-hospital mortality. Chart reviews were completed by trained emergency physician researchers (following predetermined guidelines defining abstraction criteria) to determine the presence and timing of the various components of the SIRS, qSOFA, and SOFA criteria. The qSOFA criteria were altered mental status, respiratory rate (RR)>22/min and systolic blood pressure (SBP)>100 mmHg. The SIRS criteria were heart rate (HR)>90 bpm, white blood cell (WBC) count >12,000/dL or<4000/dL, RR>20/min, temperature >38.5°C or <36°C, and a five-degree SOFA score consisting of respiratory (PaO2/FiO2[mmHg]), coagulation (platelets [*103mm3]), hepatic (bilirubin [mg/dL]), cardiovascular (hypotension, defined as mean arterial pressure ≤65 mmHg and need for vasopressor support), central nervous system (Glasgow Coma Scale) and renal (creatinine [mg/dL]) functions. The sample size required to achieve our objectives was primarily determined by the accuracy (width of the confidence interval around the point estimate of sensitivity) of SOFA for the primary outcome. Before the study, we estimated that the mean SOFA score would be approximately 4±2 in the surviving group and 6 in the deceased group. Based on these values, the required sample size was 16 for each group, with a type1 error of 0.05 and a power of 80%. The mean SOFA score was 4.56±2.66 in the surviving group and 8.61±4.43 in the deceased group. According to these scores, the minimum sample size was 7 for each group, with a type1 error of 0.05 and a power of 80%.The patient flow is shown in Figure 1. Statistical analysis was performed with SPSS version 23.0 software (SPSS Inc., Chicago,IL, USA). For the variables, a normal distribution was determined using the one-sample KolmogorovSmirnov test; continuous variables that were not normally distributed were expressed as medians (min–max), and categorical variables were expressed as numbers and percentages. The Mann-Whitney U test was used to compare continuous variables between two groups. The significance of differences between categorical variables was calculated using the ChiSquare test. Correlations between two continuous variables were calculated by the Spearman’s rank correlation coefficient (rho). Coefficients between 0 and 0.3 indicated weak correlation, coefficients between 0.3 and 0.7 indicated moderate correlation and coefficients between 0.7 and 1 indicated strong correlation. Comparison of prognostic performances of SOFA and qSOFA with SIRS and the influence of the continuous SOFA score on mortality was performed by receiver operating characteristic (ROC) analyses; the best cut-off point was determined as the point at which the sum of sensitivity and specificity was the greatest. P-values <0.05 were considered to indicate statistical significance.

Results

A total of 48 patients presenting to our ED with sepsis were eligible for this study. The mean age was 69.25±15.38 years (minimum, 22; maximum, 94). The most common site of infection was the pulmonary system (n=20, 41.7%), and the least common site was the central nervous system (n=1, 2.1%). qSOFA was ≥2 in 25 patients (52.1%), SOFA was ≥2 in 45 patients (93.7%) and SIRS was ≥ in 38 patients (79.1%). Tables 1 and 2 show the predictive values of variables affecting mortality. Initial red cell distribution width (RDW), aspartate aminotransferase (AST), lactate and base deficit (BD) were greater in patients who died (p = 0.001, 0.014, 0.012 and 0.003, respectively), whereas bicarbonate (HCO3) and fever were lower (p=0.003 and 0.002, respectively) (Table 1). Among patients who died, SOFA scores were greater (p=0.001) and hospital stay was shorter (0.016) (Table 2). Correlations of variables with the scoring systems are shown in Table 3. SBP, fever, glomerular filtration rate (GFR) and BD had moderate negative correlations with SOFA score, while the Glasgow Coma Scale (GCS), mean platelet volume (MPV), blood urea nitrogen (BUN), creatinine and lactate had moderate positive correlations with SOFA score. Age and GCS had moderate positive correlations and MPV and lactate had weak positive correlations with qSOFA. HR, RR, fever and platelet count had moderate positive correlations and SO2 had a moderate negative correlation with SIRS. Overall, in-hospital mortality was 47.9% (n= 23). The ROC curve of the SOFA score for predicting mortality is shown in Figure 2 (AUC=0.770; p=0.001; 95% CI, 0.639–0.902). Based on diagnostic performance, the best cut-off point was a SOFA score of 4.5. However, we cannot use a SOFA score of 4.5 as a cut-off because it is not an integer. We calculated the mortality rates above and below a SOFA score of 4. At this point, the mortality rate was 62.5%, the sensitivity was 82.61%, the specificity was 56.0%, the positive predictive value was 63.3% and the negative predictive value was77.8% (Figure 2). According to our results, qSOFA and SIRS cannot provide adequate diagnostic and prognostic information in the ED (Table 2). In contrast, SOFA reliably predicted mortality (Figure 2).

Discussion

Recent studies comparing sepsis scores have produced different results.^11-14 According to our results, while maintaining the value of SOFA in predicting mortality, qSOFA scores could not identify patients with the most severe forms of infection early in the course in the ED. The latter result is inconsistent with previous studies on the use of qSOFA to predict patients with increased risk of prolonged stay in the intensive care unit or death.^11,15 Although several scoring systems exist to assess the prognosis in critically ill patients, they are difficult to apply in patients presenting to acute care because of time constraints. Vital signs can provide important prognostic information in patients with acute illness.¹⁶ We found that haemoglobin, fever, and GCS correlated well with SOFA and qSOFA (r = 0.349, 0.412, and 0.692 respectively), indicating that vital signs are more flexible and efficient data sources. The sepsis-induced inflammatory milieu and organ damage ultimately resulting in death have been poorly described. The liver is the laboratory of the human body, which is capable of performing more than 200 functions, including detoxification, storage, energy production, nutrient conversion, hormonal balance and coagulation, all of which render the liver a critical organ in sepsis.¹⁷ The liver plays prominent roles in the septic process, such as removing bacteria, mediating the inflammatory esponse and regulating coagulation, which may play a role in the pathogenesis of renal failure, acute lung injury, acute respiratory distress syndrome, coagulopathy and hepatic encephalopathy. The liver is vulnerable to injury from pathogens, toxins, and inflammatory compounds, which may lead to hepatocellular dysfunction, hepatic injury, and ultimately hepatic failure.¹⁸ Septic injuries to the liver can be broadly classified as hypoxic hepatitis or the jaundice type. The latter is the more common type and is the main component of SOFA, which should be assessed as a part of the overall clinical presentation.¹⁹ Rise in hepatic transferase is not sufficient when the use of SOFA is contemplated in the ED, as shown in our study. It is presently not precisely understood how RDW is pathophysiologically formed and associated with clinical outcomes. However, it is known that RDW is elevated by inflammatory processes that interfere with iron metabolism, augment erythrocyte apoptosis, decrease erythropoietin production and suppress bone marrow.^20-21 We found a significantly higher RDW level in patients who died (p=0.001), suggesting that patients with increased RDW levels should be more aggressively treated and admission RDW could also be used for prognostic purposes, particularly in busy EDs. Although the role of MPV in sepsis is not fully understood, it has been reported to remain at normal levels in localized bacterial infections, but to be significantly elevated in half of patients with sepsis.²² We found that MPV was positively correlated with SOFA and qSOFA scores (r = 0.313 and 0.93, respectively) and according to previous reports.²³ impaired thrombocyte production and function caused by the impact of sepsis on bone marrow may be reflected in MPV as an indirect sign of dysfunction; this parameter can be used in the ED as a quick and reliable sign of sepsis. We found that lactate levels were correlated with SOFA and qSOFA scores and that the former could predict mortality (p= 0.012), this result is consistent with previous studies of infection.²⁴ At present, although the use of three scoring systems cannot be universally recommended, lactate measurements should be combined with them when an infection is suspected.

Limitations

The most obvious limitation of this research was that of a small sample size. Although the sample size was small, we adequately addressed the research questions or generalized beyond the context of the study and still, the small population did not negate recognition of importance of SOFA in predicting sepsisrelated mortality in ED, but with a larger sample, including a greater number of culturally different participants any real differences would almost certainly have emerged. Our study produced statistically significant results concerning sepsis scoring systems and encouraged to find results similar to larger and more inclusive studies. Secondly, baseline information on cardiovascular risk factors, comorbidities and concurrent medication, which were important potential confounders in this context, was not collected or available to the data extractors, and any potential influence on vital signs was not controlled. Larger, multisite, prospective studies are needed to control for multiple confounders and find clinically important associations.

Conclusion

In conclusion, this study highlighted the poor performance of qSOFA and the reliable performance of SOFA for the prediction of sepsis-related mortality in the ED. We hope this small study will provoke more investigation into the appropriateness of fully adopting mortality predicting sepsis scores as a screening tool by emergency medicine physicians.

Declarations

References

1. 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(3):290-300. doi:10.1001/jama.2016.20328
   Article · PubMed · Google Scholar
2. Moskowitz A, Patel PV, Grossestreuer AV, et al. Quick sequential organ failure assessment and systemic inflammatory response syndrome criteria as predictors of critical care intervention among patients with suspected infection. Crit Care Med. 2017;45:1813-1819. doi:10.1097/ccm.0000000000002622
   Article · PubMed · Google Scholar
3. de Groot B, Stolwijk F, Warmerdam M, et al. The most commonly used disease severity scores are inappropriate for risk stratification of older emergency department sepsis patients: an observational multicentre study. Scand J Trauma Resusc Emerg Med. 2017;25:91. doi:10.1186/s13049-017-0436-3
   Article · PubMed · Google Scholar
4. Zhou X, Wu F. It is SOFA score rather than quick SOFA score that constitutes the clinical criteria of sepsis. Am J Infect Control. 2018;46:1315. doi:10.1016/j.ajic.2018.06.024
   Article · PubMed · Google Scholar
5. Guirgis FW, Puskarich MA, Smotherman C, et al. Development of a simple sequential organ failure assessment score for risk assessment of emergency department patients with sepsis. J Intensive Care Med. 2017
   PubMed · Google Scholar
6. Maitra S, Som A, Bhattacharjee S. Accuracy of quick Sequential Organ Failure Assessment score and systemic inflammatory response syndrome criteria for predicting mortality in hospitalized patients with suspected infection: a meta-analysis of observational studies. Clin Microbiol Infect. 2018;24:1123-1129. doi:10.1016/j.cmi.2018.03.032
   Article · PubMed · Google Scholar
7. Khwannimit B, Bhurayanontachai R, Vattanavanit V. Comparison of the performance of SOFA, qSOFA and SIRS for predicting mortality and organ failure among sepsis patients admitted to the intensive care unit in a middle-income country. J Crit Care. 2018;44:156-160. doi:10.1016/j.jcrc.2017.10.023
   Article · PubMed · Google Scholar
8. van der Woude SW, van Doormaal FF, Hutten BA, Nellen FJ, Holleman F. Classifying sepsis patients in the emergency department using SIRS, qSOFA or MEWS. Neth J Med. 2018;76:158-166.
   PubMed · Google Scholar
9. Rudd KE, Seymour CW, Aluisio AR, et al. Association of the quick sequential sepsis-related organ failure assessment score with excess hospital mortality in adults with suspected infection in low- and middle-income countries. JAMA. 2018;319:2202-2211. doi:10.1001/jama.2018.6229
   Article · PubMed · Google Scholar
10. Baig MA, Sheikh S, Hussain E, et al. Comparison of qSOFA and SOFA score for predicting mortality in severe sepsis and septic shock patients in the emergency department of a low middle-income country. Turk J Emerg Med. 2018;18:148-151. doi:10.1016/j.tjem.2018.08.002
    Article · PubMed · Google Scholar
11. Tusgul S, Carron PN, Yersin B, Calandra T, Dami F. Low sensitivity of qSOFA, SIRS criteria and sepsis definition to identify infected patients at risk of complication in the prehospital setting and at the emergency department triage. Scand J Trauma Resusc Emerg Med. 2017;25:108. doi:10.1186/s13049-017-0449-y
    Article · PubMed · Google Scholar
12. Kushimoto S, Gando S, Ogura H. The qSOFA requires validation as a promptly applicable clinical criterion. Acute Med Surg. 2017;4:225-226. doi:10.1002/ams2.287
    Article · PubMed · Google Scholar
13. Asiimwe SB, Abdallah A, Ssekitoleko R. A simple prognostic index based on admission vital signs data among patients with sepsis in a resource-limited setting. Crit Care. 2015;19:86. doi:10.1186/s13054-015-0826-8
    Article · PubMed · Google Scholar
14. Askim Å, Moser F, Gustad LT, et al. Poor performance of quick SOFA score in predicting severe sepsis and mortality: a prospective study of patients admitted with infection to the emergency department. Scand J Trauma Resusc Emerg Med. 2017;25:56. doi:10.1186/s13049-017-0399-4
    Article · PubMed · Google Scholar
15. Haydar S, Spanier M, Weems P, Wood S, Strout T. Comparison of qSOFA score and SIRS criteria as screening mechanisms for emergency department sepsis. Am J Emerg Med. 2017;35:1730-1733. doi:10.1016/j.ajem.2017.07.001
    Article · PubMed · Google Scholar
16. Kim KS, Suh GJ, Kim K, et al. Quick Sepsis-related Organ Failure Assessment score is not sensitive enough to predict 28-day mortality in emergency department patients with sepsis: a retrospective review. Clin Exp Emerg Med. 2019;6:77-83. doi:10.15441/ceem.17.294
    Article · PubMed · Google Scholar
17. Canabal JM, Kramer DJ. Management of sepsis in patients with liver failure. Curr Opin Crit Care. 2008;14:189-197. doi:10.1097/mcc.0b013e3282f6a435
    Article · PubMed · Google Scholar
18. Yan J, Li S, Li S. The role of the liver in sepsis. Int Rev Immunol. 2014;33:498-510. doi:10.3109/08830185.2014.889129
    Article · PubMed · Google Scholar
19. Wang D, Yin Y, Yao Y. Advances in sepsis-associated liver dysfunction. Burns Trauma. 2014;2:97-105. doi:10.4103/2321-3868.132689
    Article · PubMed · Google Scholar
20. Wang AY, Ma HP, Kao WF, Tsai SH, Chang CK. Red blood cell distribution width is associated with mortality in elderly patients with sepsis. Am J Emerg Med. 2018;36:949-953. doi:10.1016/j.ajem.2017.10.056
    Article · PubMed · Google Scholar
21. Ju XF, Wang F, Wang L, et al. Dynamic change of red cell distribution width levels in prediction of hospital mortality in Chinese elderly patients with septic shock. Chin Med J (Engl). 2017;130:1189-1195. doi:10.4103/0366-6999.205858
    Article · PubMed · Google Scholar
22. Van der Lelie J, Von dem Borne AK. Increased mean platelet volume in septicaemia. J Clin Pathol. 1983;36:693-696. doi:10.1136/jcp.36.6.693
    Article · PubMed · Google Scholar
23. Becchi C, Al Malyan M, Fabbri LP, Marsili M, Boddi V, Boncinelli S. Mean platelet volume trend in sepsis: is it a useful parameter? Minerva Anestesiol. 2006;72:749-756
    PubMed · Google Scholar
24. Shetty A, MacDonald SP, Williams JM, et al. Lactate ≥2 mmol/L plus qSOFA improves utility over qSOFA alone in emergency department patients presenting with suspected sepsis. Emerg Med Australas. 2017;29:626-634. doi:10.1111/1742-6723.12894
    Article · PubMed · Google Scholar

Tables

Additional Information

Publisher’s Note
Bayrakol MP remains neutral with regard to jurisdictional and institutional claims.

Rights and Permissions

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). To view a copy of the license, visit https://creativecommons.org/licenses/by-nc/4.0/

View full license details →