European Journal of Cardiovascular Medicine

Sepsis is a global healthcare issue and continues to be the leading cause of death from infection. The mortality seen in patients with sepsis is high ^[1]. It is a cause of more than 2,00,000 USA deaths reported per year and an in-hospital mortality of more than 30% despite advances in critical care ^[2]. It is associated with more than $24 billion in costs in the United States annually ^[3,4]

Sepsis is defined as a “life-threatening organ dysfunction due to a dysregulated host response to infection”. Despite high treatment expense, sepsis is often fatal ^[5,6]. Many a times, Sepsis is diagnosed late, and proper treatment is delayed. When Sepsis is identified early in emergency department (ED) and aggressive therapy is initiated early, the mortality and morbidity rates can be significantly reduced because most cases of sepsis present in the ED and in the wards rather than the intensive care unit (ICU) ^[7].

For early diagnosis and predict the outcome of sepsis many scoring systems are available. In present study we aimed to compare between quick sepsis-related organ failure assessment (qSOFA), modified shock index (MSI), and national early warning score2 (NEWS2) in sepsis and it's outcome in emergency department

Present study was Observational, Prospective, Unicentric Study, conducted in department of Department of Emergency Medicine, Max Super Specialty Hospital, Vaishali, Ghaziabad India. Study duration was of 1 Year. (December 2018 to December 2019). Study was approved by institutional ethical committee.

Inclusion criteria

• Patients of age ≥ 18yrs. both male and female, who met Suspected infection definition, willing to participate in present study.
• Patients were diagnosed Sepsis based on an infectious source (whether clinical, radiological, or microbiological) or an equivocal presentation (for example, a febrile patient with inflammatory syndrome)⁷

Exclusion criteria

• Subject is unwilling or unable to give consent to participate in the study.
• Age<18yrs
• All types of Shock except Septic Shock.
• patients who received mechanical ventilation or vasopressor medications before the first suspicion of sepsis.
• Trauma Patients.
• Pregnant Patients.

Study was explained to participants in local language & written informed consent was taken. Data was filled at the time of presentation in ED, which includes details of patients – such as age, details of patient’s history, qSOFA score, MSI, NEWS2 score, associated co-morbidities and outcome.

All the parameters and vitals in the data collection form were taken by the pretrained ED triage doctors at the time of presentation in ED. To reduce interobserver variability, training was given to ED triage doctors prior to the study. Blood pressure was recorded manually by pretrained ED doctor. After the recruitment, during follow up 2 independent doctors reviewed all the patients and judged whether the acute presentation to the ED was related to an infection. Evidence of infection was sought through the analysis of radiological studies, microbiological findings, or clinical context. In cases of disagreement, consensus was sought between the 2 independent doctors. Patients in whom infection was not confirmed were then excluded from analysis. Pretrained ED triage doctors calculated the following scores while filling out the data collection form.

·        The qSOFA criteria contains Systolic blood pressure ≤100 mm Hg, Respiratory rate ≥22 breaths/min & Altered mentation (Glasgow coma scale<15).

·        Modified Shock Index contains Heart rate & Mean arterial pressure.

·        National Early Warning Score2 (NEWS2) contains respiration rate, oxygen saturation, systolic blood pressure, pulse rate, level of consciousness or new confusion & temperature.

Statistical analysis was performed by the SPSS program for Windows, version 17.0(SPSS, Chicago, Illinois). Continuous variables are presented as mean ± SD, and categorical variables are presented as absolute numbers and percentage. Data were checked for normality before statistical analysis. Normally distributed continuous variables were compared using the unpaired t test, whereas the Mann-Whitney U test was used for those variables that were not normally distributed. Categorical variables were analysed using either the chi square test or Fisher’s exact test. A receiver operating characteristics (ROC) analysis was calculated to determine optimal cut-off values for qSOFA, MSI and NEWS2 predicting mortality and ICU stay>3 days. The area under the curve and its standard deviation (AUC _ SD), the sensitivity, and the specificity was calculated to analyse the diagnostic value of all these markers. For all statistical tests, a p value less than 0.05 was taken to indicate a significant difference.

Out of 187 patients, majority were from 61-70 years age group (40.6%), followed by 51-60 years (31%) & 71-80 years age group (14.4%). 125 (66.8%) were male and 62 (33.2%) were female. Majority of the patient have respiratory infection42.8% (80 cases) followed by urinary tract infection 21.4%, Intra-abdominal infection38%, CNS infection 5.9%, skin / soft tissue infection 5.3 % and others 4.3%.

Table 1: General characteristics

Out of 187 cases of in our study, 117(62.6%) patients discharged while 70 (37.4%) patients died. Mean age for the discharge population was 60.91 years. Mean age for died population was 69.39 years, difference was statistically significant. Among discharged population 60.7% were Male and 39.3% were Female. Among died population 77.1% were male and 22.9% were female. A statistically significant correlation was found between Male sex and mortality (P <0.021)

Table 2: Correlation between age, gender with outcome

Majority patients presented with History of fever 79.7% (149 cases) and also presented with hemoptysis 5.3% (10 cases) and dizziness 5.3% (10 cases) history. Shortness of breath associated with 88.6% mortality in dead population and 47% discharge outcome in the total discharged patients. Altered mental status associated with mortality of 41.4% in dead population and only 12.8% discharge outcome in the total discharged patients. Cough with expectoration, shortness of breath, and altered mental status statistically significant (P value<0.001) with mortality.

Table 3: Comparison between HISTORY and OUTCOME

Diabetes Mellitus (DM) was found in 85.6% (160 patients), hypertension was found in 47.1% (88 patients), COPD in 26.2% (49 patients), coronary artery disease (CAD) in 31.6% (59 patients), chronic liver disease (CLD) in 13.4% (25 patients), cerebrovascular disease (CVA) in 11.8% (22 patients) and Chronic Kidney disease (CKD) in 12.8% (24 patients) of the patients. A statistically significant correlation was found between Hypertension and mortality (P value <0.002).

Table 4: Comparison between PAST HISTORY and OUTCOME

67.9% (127 cases) were admitted in ICU more than 3 days and 27.3% (51 cases) admitted in ICU ≤3 days. Not applicable indicates 9 patients died <3 day from the time of admission . A statistically significant correlation was found between ICU stay > 3 days and mortality (p value <0.001).

Table 5: Correlation between ICU stay >3 days & outcome

*9 cases did not stay in ICU

In qSOFA, RR >22 is associated with 91.4% mortality in the total dead patients and 51.3 % discharge outcome in the total discharged patients. Also vice versa RR <22 is associated with 48.7% discharge outcome among discharged patients and only 8.6% mortality among dead patients. Similarly, presence of altered mental status is associated with 48.6% mortality in the total dead patients and 12.8% % discharge outcome in the total discharged patients. Also vice versa patient with normal mental status associated with 87.2% discharge outcome among discharged patients and 51.4% mortality among dead patients. Presence of RR> 22 & altered mental status statistically associated with mortality. (P value <0.001).

Table 6: Correlation between qSOFA parameters & OUTCOME

Mean value of Heart rate (HR), Systolic blood pressure (SBP) and diastolic blood pressure (DBP), the MSI scores respectively are 113.56 ± 12.39, 89.14 ± 31.29, 46.43 ± 18.02, 2.18 ± 1.02and are associated with mortality. While it is statistically significant, p values respectively are as follows. (P value <0.001, <0.001, <0.005, <0.001).

Table 7: Comparison between MSI parameters & OUTCOME

Presence of hypercapnic respiratory failure and oxygen support other than alert mental status is statistically associated with mortality. (P value respectively <0.014, 0.001. <0.001).

Mean value of Respiratory rate (RR) 22.93 ± 2.97), Spo2 Scale 1&Spo2 Scale 2(If Target 88- 92%) 94.82 ± 2.62, Systolic blood pressure (SBP) 103.93 ± 19.87and pulse rate (PR) 106.83 ± 9.50, temperature (TEMP) 99.82 ± 1.07and result 6.86 ± 2.81 is associated with discharge outcome which is statistically significant respectively (P value <0.001, <0.001, <0.001, <0.001<0.036 &<0.001).

Mean value of Respiratory rate (RR) 26.93 ± 3.59, Spo2 Scale 1&Spo2 Scale 2(If Target 88- 92%) 90.44 ± 3.62, Systolic blood pressure (SBP) 89.43 ± 31.30 and pulse rate (PR) 112.73 ± 14.61, temperature (TEMP) 99.47 ± 1.18 and result 11.50 ± 2.68 associated with death outcome which is statistically significant respectively (P value<0.001, <0.001, <0.001, <0.001<0.036 &<0.001).

Table 8: Correlation between NEWS2 score parameters & outcome

qSOFA had an AUC of 0.767 against a standard error of 0.036 whereas for MSI it was 0.682 against a standard error of 0.044. And finally, NEWS2 had an AUC of 0.879 against a standard error of 0.025.

Table 9: Area under the curve cut off value for qSOFA, MSI and NEWS2 score for predicting outcome.

Figure 1: ROC analysis to determine optimal cut-off values for qSOFA, MSI and NEWS2 predicting outcome.

Cut of value ≥ 2 of qSOFA associated with 70% death out come in dead population. Similarly, < 2 associated with 71.8 % discharge out come in the total discharged patients. A significant correlation was found between qSOFA cut of value ≥ 2 and mortality. (P value <0.001). qSOFA cut off value 2 in predicting outcome: sensitivity =70%, specificity=71.8%, positive predictive value (PPV) = 59.8 %, negative predictive value (NPV) = 80%, accuracy = 71.1 %. Similarly cut of value <1.88 associated with 83.8 % discharge outcome in the total discharged patients. A significant correlation was found between MSI cut off value > 1.88 and mortality. (P value <0.001). MSI cut off value 1.88 predicting outcome(discharge/ death): sensitivity(51.4%), specificity (83.8%), positive predictive value 65.5% (PPV), negative predictive value 74.2% (NPV), accuracy71.7%..

A significant correlation was found between NEWS2 cut off value > 9 and mortality (P value <0.001). NEWS2 cut off value 9 & predicting outcome (discharge/ death) had Sensitivity (85.7%), specificity (76.1%), positive predictive value 68.2% (PPV), negative predictive value 89.9% (NPV), accuracy79.7%.

Table 10: Correlation between Scoring system & Outcome

Cut of value ≥1.88 of MSI associated with 51.4% death outcome in dead population.

The sensitivity for qSOFA was 70%, for MSI 51.4%, and for NEWS2 it was 85.7%. The specificity for qSOFA was 71.8%, for MSI 83.8% and for NEWS2 76.1%. It was observed sensitivity of NEWS2 at cut off value 9 was statistically significantly higher in predicting mortality as compared to qSOFA and MSI. The sensitivity of qSOFA at cut value >= 2 was statistically significantly higher in predicting mortality as compared to MSI (P<0.025).

Table 11: Comparison of sensitivity and specificity of qSOFA cut off value 2, MSI cut off value 1.88 and NEWS2 cut off value 9 for predicting patient mortality and their statistical significance

Table 20:comparision of sensitivity and specificity between qSOFA cut off value 2, MSI cut off value 1.88 and NEWS2 cut off value 9for predicting ICU stay > 3 days) and their statistical significance

The sensitivity for qSOFA was 54.3%, for MSI 30.7%, and for NEWS2 it was 59.7%.

The specificity for qSOFA was 92.2%, for MSI 84.3% and for NEWS2 94.1%.

It was observed sensitivity of NEWS2 at cut off value 9 was statistically significantly higher in predicting ICU stay >3 days as compared to MSI (P value <0.001) but not significantly higher than qSOFA.

The sensitivity of qSOFA at cut value >= 2 was statistically significantly higher than MSI (P<0.001).

Correlation between scoring system and ICU stay >3 days:

Figure19: ROC analysis to determine optimal cut-off values for qSOFA, MSI and NEWS2 predicting ICU stay >3 days.

Table 21: Area under the curve cut off value for qSOFA, MSI and NEWS2 score to predicting ICU stay >3 days

Table 22: Correlation between QSOFA cut off value 2 &ICU stay >3 days.

Above table correlates qSOFA cut off value 2 in predicting ICU stay >3 days. Cut of value ≥ 2 associated with 54.3% of ICU stay >3 days. Similarly cut of value < 2 associated with patient admitted in ICU ≤ 3 days (92.2 %)., A significant correlation was found between qSOFA cut of value ≥ 2 in predicting ICU stay >3 days. (P value <0.001).

Figure 20: qSOFA cut off value 2 for predicting ICU stay >3 days: sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy.

Above figure shows qSOFA cut off value 2 in predicting ICU stay > 3 days: sensitivity =54.3%, specificity=92.2%, positive predictive value (PPV)= 94.5%, negative predictive value (NPV) = 44.8%, accuracy = 65.2 %.

Table 23: Correlation between MSI cut off value 1.585&ICU stay > 3 days

Above table correlate MSI cut off value 1.585 with ICU stay >3 days. Cut off value ≥1.585is associated with (59.1%) ICU stay > 3 days. Similarly cut off value < 1.585 is associated with (60.8%) patient admitted in ICU stay ≤ 3 days. A significant correlation was found between MSI Cut of value ≥1.585in predicting ICU stay >3 days. (P value <0.016).

Figure 21:MSI cut off value 1.88 predicting ICU stay > 3 days: sensitivity(59.1) ,specificity(60.8%), positive predictive value (78.9%)(PPV), negative predictive value(37.3%) (NPV) and accuracy (59.6%) .

Table 24: Correlation between NEWS2 cut off value 7.5 &ICU stay >3 days.

Above table diagram correlates NEWS2 cut off value 7.5 with ICU stay > 3 days. Cut of value ≥7.5 is associated with (73.2%) patient admitted in ICU > 3 days. Similarly cut of value < 7.5 is associated with (90.2%) patients admitted in ICU ≤3 days. A significant correlation was found between NEWS2 cut off value ≥7.5 in predicting ICU stay >3 days. (P value <0.001).

Figure 22: NEWS2 cut off value 9& predicting ICU stay >3 days: sensitivity (73.2%) ,specificity(90.2%), positive predictive value (94.9%)(PPV), negative predictive value (57.5%)(NPV) and accuracy (78.1%).

Sepsis is a world-wide health issue and is also one among the major causes of non- traumatic deaths in the world. It is also a major public health concern as it’s a leading cause of critical illness and mortality in hospital. According to the most recent Centre for Disease Control (CDC) report, it is estimated that sepsis affects around 1.5 million people in the United States of America annually, causing death of nearly 2,50,000 people and has been responsible for 1 out of every 3 hospital deaths. Early diagnosis and treatment of sepsis is essential to prevent high mortality rates.⁹

Of the 187 patients, 40.6% (76 cases) of the patients belonged to 6th to 7th decades with mean age being 64.09 ± 10.29 years. In a similar study by Goulden R et al.,¹⁰ the mean age of the patients included was around 68 years ^[16].

Our study group had males more in number (125 males, 66.8%) than females (62 females,33.2%). Male sex was more associated with morality. But Goulden R et al.,¹⁰ study reported that both sexes have similar mortality. Among study population 37.4 % (70 cases) included in the study succumbed to death indicating a high mortality related to the disease. Contrary to our study, mortality was reported in only 5% of the patients in a study conducted by Churpek MM et al.,¹¹

In this study, it was observed that the Mean age for the discharge population was 60.91 years. Mean age for dead population was 69.39 years. In a similar study mean age for death is high (78 years) ^[17]. It signifies mortality increases with increasing age in sepsis. It correlates with similar study by Goulden R et al.,¹⁰

In qSOFA respiratory rate (RR)>22 is associated with 91.4% mortality among total dead patients and 51.3% discharge outcome among total discharged patients. Similarly altered mental status associated with 48.6% mortality among total dead patients. Respiratory rate>22 and altered mental status parameters in the q SOFA are statistically associated with patient mortality. In MSI, Mean value of Heart rate (HR), Systolic blood pressure (SBP) and diastolic blood pressure (DBP), MSI score respectively 113.56 ± 12.39, 89.14 ± 31.29, 46.43 ± 18.02, 2.18 ± 1.02 significantly associated with mortality. In the NEWS2 score parameters such as presence of hypercapnic respiratory failure, oxygen support other than alert mental status statistically associated with mortality. In the NEWS2 score, Mean value of Respiratory rate (RR) 26.93 ± 3.59, Spo2 Scale 1&Spo2 Scale 2(If Target 88-92%) 90.44 ± 3.62, Systolic blood pressure (SBP) 89.43 ± 31.30 and pulse rate (PR) 112.73 ± 14.61, temperature (TEMP) 99.47 ± 1.18 and result 11.50 ± 2.68significantly associated with mortality.

Our results also consistent with other studies showing qSOFA cut off value ≥ 2 significantly associated with patient morality in sepsis.^11,12 Also from this study we found out that qSOFA cut of value ≥ 2 is significantly associated with patient ICU stay >3 days it is also correlates with other study^.10,11

In our study, qSOFA of value 2 predicts patients’ mortality but the sensitivity is 70% which is lower than(93%) study conducted by Churpek MM et al.,¹¹ .but higher than (37 %) another study conducted by, Goulden R et al.,^{10 .}Our study results were correlates with similar study conducted by Mellhammar L et al.,¹³

From this study, MSI cut off value ≥ 1.88 and there is an increased probability of mortality in sepsis. Similarly other studies also reported significant association with mortality.^14,15 But on contrary to our study they used different cut of value (MSI >1.3). Our results show that in patients with an MSI≥ 1.585 there is an increased probability of ICU admission. It also correlates with another similar study with different cut off value.¹⁴ In our study NEWS2 cut off value of 9 predicts patients’ mortality and the sensitivity is higher than qSOFA. It also correlates with other similar studies but different cut off value.^10,12

Our results show that in patients with an NEWS2 cut off value of 7.5 there is an increased probability of ICU admission more than 3 days. It is also correlates with another similar study with different cut off value.^10,12

In this observational study our findings suggest that for predicting MORTALITY and ICU stay > 3 days among all patients with suspected sepsis, NEWS2 score was most sensitive than qSOFA and MSI score. This finding correlates with similar study conducted by Mellhammar let al.¹³ For predicting mortality qSOFA has higher sensitivity than MSI but lower than NEWS2. The sensitivity of qSOFA for predicting ICU stay > 3 days was near similar to NEWS2 score and higher than MSI.

Limitations of present study were single center study & limited data; thus a much larger sample size obviously would be desirable to provide concrete evidence.

MSI though not fairing so well in comparison with qSOFA AND NEWS2 is nevertheless a good scoring tool for predicting sepsis mortality and ICU stays in emergency department. My study does have limitations like a particular geographical locality, limited population etc. Hence, if further studies are made using MSI as a tool in the future, maybe we could garnish more evidence in favor of MSI Scoring tool too and subsequently MSI might even find its role in routine clinical assessment.

We found that NEWS2 is a better score than qSOFA and MSI in predicting sepsis mortality and ICU stay in emergency department. But qSOFA isn’t much inferior to NEWS 2 in predicting mortality and ICU stay and also is easy to use by bed side. NEWS2 is time consuming and needs NEWS2 chart. So qSOFA may be adapted in the busy emergency department.

Conflict of Interest: None to declare

Source of funding: Nil

1. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest. 1992; 101:1644-55.
2. Gaieski DF, Edwards JM, Kallan MJ, Carr BG. Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med. 2013; 41:1167–74.
3. Liu V, Escobar GJ, Greene JD, Soule J, Whippy A, Angus DC, Iwashyna TJ. Hospital deaths in patients with sepsis from 2 independent cohorts. JAMA 2014; 312:90–92.
4. Lagu T, Rothberg MB, Shieh MS, Pekow PS, Steingrub JS, Lindenauer PK. Hospitalizations, costs, and outcomes of severe sepsis in the United States 2003 to 2007. Crit Care Med 2012; 40:754–761.
5. Sepsis-3: International Consensus Definitions for Sepsis and Septic Shock https://www.esicm.org/article-review-sepsis-3-depascale
6. Seymour CW,Iwashyna TJ, BrunkhorstFM,et al. Assessment of clinical criteria for sepsis: for the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) JAMA. 2016; 315:762–774
7. Levy MM, Rhodes A, Phillips GS, Townsend SR, et al. Surviving sepsis campaign: association between performance metrics and outcomes in a 7.5-year study. Crit Care Med. 2015; 43:3–12.
8. Freund Y, Lemachatti N, Krastinova E, et al. Prognostic Accuracy of Sepsis-3 Criteria for In- Hospital Mortality Among Patients with Suspected Infection Presenting to the Emergency Department. JAMA. 2017;317(3):301–308.
9. Bone RC, Grodzin CJ, Balk RA. Sepsis: a new hypothesis for pathogenesis of the disease process. Chest 1997; 112: 235-43.
10. Goulden R, Hoyle MC Monis J, Railton D, Riley V, Martin P et al. qSOFA, SIRS and NEWS for predicting in hospital mortality and ICU admission in emergency admissions treated as sepsis. Emerg Med J. 2018 Jun;35(6):345-349
11. Churpek MM, Snyder A, Han X, Sokol S, Pettit N, Howell MD et al.Quick Sepsis-related Organ Failure Assessment, Systemic Inflammatory Response Syndrome, and Early Warning Scores for Detecting Clinical Deterioration in Infected Patients outside the Intensive Care Unit Am J Respir Crit Care Med. 2017 Apr 1;195(7):906-911.
12. Finkelsztein EJ, Jones DS, Ma KC, et al. Comparison of qSOFA and SIRS for predicting adverse outcomes of patients with suspicion of sepsis outside the intensive care unit.Crit Care 2017;21:73.
13. Mellhammar L, Linder A, Tverring J et al.NEWS2 is Superior to qSOFA in Detecting Sepsis with Organ Dysfunction in the Emergency Department. Clin Med. 2019 Jul 29;8(8):1128.
14. Liu Y, LiuJ,FangZ,ShanG,Xu J, Qi Z et al. Modified shock index and mortality rate of emergency patients. World J Emerg Med. 2012; 3(2): 114–117
15. Jayaprakash N, Smischney N, Kashyap R, et al. modified shock index as a predictor of icu length of stay, sofa score, and mortality. crit care med 2016 • volume 44 • Number 12 (Suppl.)