Quick Sequential Organ Failure Assessment Is Not Good for Ruling Sepsis In or Out

Machine Learning Model to Identify Sepsis Patients in the Emergency Department: Algorithm Development and Validation

Accurate stratification of sepsis can effectively guide the triage of patient care and shared decision making in the emergency department (ED). However, previous research on sepsis identification models focused mainly on ICU patients, and discrepancies in model performance between the development and external validation datasets are rarely evaluated. The aim of our study was to develop and externally validate a machine learning model to stratify sepsis patients in the ED. We retrospectively collected clinical data from two geographically separate institutes that provided a different level of care at different time periods. The Sepsis-3 criteria were used as the reference standard in both datasets for identifying true sepsis cases. An eXtreme Gradient Boosting (XGBoost) algorithm was developed to stratify sepsis patients and the performance of the model was compared with traditional clinical sepsis tools; quick Sequential Organ Failure Assessment (qSOFA) and Systemic Inflammatory Response Syndrome (SIRS). There were 8296 patients (1752 (21%) being septic) in the development and 1744 patients (506 (29%) being septic) in the external validation datasets. The mortality of septic patients in the development and validation datasets was 13.5% and 17%, respectively. In the internal validation, XGBoost achieved an area under the receiver operating characteristic curve (AUROC) of 0.86, exceeding SIRS (0.68) and qSOFA (0.56). The performance of XGBoost deteriorated in the external validation (the AUROC of XGBoost, SIRS and qSOFA was 0.75, 0.57 and 0.66, respectively). Heterogeneity in patient characteristics, such as sepsis prevalence, severity, age, comorbidity and infection focus, could reduce model performance. Our model showed good discriminative capabilities for the identification of sepsis patients and outperformed the existing sepsis identification tools. Implementation of the ML model in the ED can facilitate timely sepsis identification and treatment. However, dataset discrepancies should be carefully evaluated before implementing the ML approach in clinical practice. This finding reinforces the necessity for future studies to perform external validation to ensure the generalisability of any developed ML approaches.

Strengthening sepsis care at a tertiary care teaching hospital in New Delhi, India

Introduction

Failure of early identification of sepsis in the emergency department (ED) leads to significant delays in antibiotic administration which adversely affects patient outcomes.

Aim

The primary objective of our Quality Improvement (QI) project was to reduce the door-to-antibiotic time (DTAT) by 30% from the preintervention in patients with suspected sepsis. Secondary objectives were to increase the blood culture collection rate by 30% from preintervention, investigate the predictors of improving DTAT and study the effect of these interventions on 24-hour in-hospital mortality.

Methods

This QI project was conducted in the ED of a tertiary care teaching hospital of North India; the ED receives approximately 400 patients per day. Adult patients with suspected sepsis presenting to our ED were included in the study, between January 2019 and December 2020. The study was divided into three phases; preintervention phase (100 patients), intervention phase (100 patients) and postintervention phase (93 patients). DTAT and blood cultures prior to antibiotic administration was recorded for all patients. Blood culture yield and 24-hour in-hospital mortality were also recorded using standard data templates. Change ideas planned by the Sepsis QI Team were implemented after conducting plan-do-study-act cycles.

Results

The median DTAT reduced from 155 min in preintervention phase to 78 min in postintervention phase. Drawing of blood cultures prior to antibiotic administration improved by 67%. Application of novel screening tool at triage was found to be an independent predictor of reduced DTAT.

Conclusion

Our QI project identified the existing lacunae in implementation of the sepsis bundle which were dealt with in a stepwise manner. The sepsis screening tool and on-site training improved care of patients with sepsis. A similar approach can be used to deal with complex quality issues in other high-volume low-resource settings.

Monocyte distribution width (MDW) performance as an early sepsis indicator in the emergency department: comparison with CRP and procalcitonin in a multicenter international European prospective study

BACKGROUND

Early sepsis diagnosis has emerged as one of the main challenges in the emergency room. Measurement of sepsis biomarkers is largely used in current practice to improve the diagnosis accuracy. Monocyte distribution width (MDW) is a recent new sepsis biomarker, available as part of the complete blood count with differential. The objective was to evaluate the performance of MDW for the detection of sepsis in the emergency department (ED) and to compare to procalcitonin (PCT) and C-reactive protein (CRP).

METHODS

Subjects whose initial evaluation included a complete blood count were enrolled consecutively in 2 EDs in France and Spain and categorized per Sepsis-2 and Sepsis-3 criteria. The performance of MDW for sepsis detection was compared to that of procalcitonin (PCT) and C-reactive protein (CRP).

RESULTS

A total of 1,517 patients were analyzed: 837 men and 680 women, mean age 61 ± 19 years, 260 (17.1%) categorized as Sepsis-2 and 144 patients (9.5%) as Sepsis-3. The AUCs [95% confidence interval] for the diagnosis of Sepsis-2 were 0.81 [0.78-0.84] and 0.86 [0.84-0.88] for MDW and MDW combined with WBC, respectively. For Sepsis-3, MDW performance was 0.82 [0.79-0.85]. The performance of MDW combined with WBC for Sepsis-2 in a subgroup of patients with low sepsis pretest probability was 0.90 [0.84-0.95]. The AUC for sepsis detection using MDW combined with WBC was similar to CRP alone (0.85 [0.83-0.87]) and exceeded that of PCT. Combining the biomarkers did not improve the AUC. Compared to normal MDW, abnormal MDW increased the odds of Sepsis-2 by factor of 5.5 [4.2-7.1, 95% CI] and Sepsis-3 by 7.6 [5.1-11.3, 95% CI].

CONCLUSIONS

MDW in combination with WBC has the diagnostic accuracy to detect sepsis, particularly when assessed in patients with lower pretest sepsis probability. We suggest the use of MDW as a systematic screening test, used together with qSOFA score to improve the accuracy of sepsis diagnosis in the emergency department. Trial Registration ClinicalTrials.gov (NCT03588325).

Optimal combination of early biomarkers for infection and sepsis diagnosis in the emergency department: The BIPS study

OBJECTIVE

To define the best combination of biomarkers for the diagnosis of infection and sepsis in the emergency room.

METHODS

In this prospective study, consecutive patients with a suspicion of infection in the emergency room were included. Eighteen different biomarkers measured in plasma, and twelve biomarkers measured on monocytes, neutrophils, B and T-lymphocytes were studied and the best combinations determined by a gradient tree boosting approach.

RESULTS

Overall, 291 patients were included and analysed, 148 with bacterial infection, and 47 with viral infection. The best biomarker combination which first allowed the diagnosis of bacterial infection, included HLA-DR (human leukocyte antigen DR) on monocytes, MerTk (Myeloid-epithelial-reproductive tyrosine kinase) on neutrophils and plasma metaloproteinase-8 (MMP8) with an area under the curve (AUC) = 0.94 [95% confidence interval (IC95): 0.91;0.97]. Among patients in whom a bacterial infection was excluded, the combination of CD64 expression, and CD24 on neutrophils and CX3CR1 on monocytes ended to an AUC = 0.98 [0.96;1] to define those with a viral infection.

CONCLUSION

In a convenient cohort of patients admitted with a suspicion of infection, two different combinations of plasma and cell surface biomarkers were performant to identify bacterial and viral infection.

STAPLAg: a convenient early warning score for use in infected patients in the intensive care unit

Sepsis is a life-threatening disease in the intensive care unit (ICU). The current diagnostic criteria for sequential organ failure assessment (SOFA) scores do not reflect the current understanding of sepsis. We developed a novel and convenient score to aid early prognosis.Retrospective multivariable regression analysis of 185 infected emergency ICU (EICU) patients was conducted to identify independent variables associated with death, to develop the new "STAPLAg" score; STAPLAg was then validated in an internal cohort (n = 106) and an external cohort (n = 78) and its predictive efficacy was compared with that of the initial SOFA score.Age, and initial serum albumin, sodium, PLR, troponin, and lactate tests in the emergency department were independent predictors of death in infected EICU patients, and were used to establish the STAPLAg score (area under the curve [AUC] 0.865). The initial SOFA score on admission was predictive of death (AUC 0.782). Applying the above categories to the derivation cohort yielded mortality risks of 7.7% for grade I, 56.3% for grade II, and 75.0% for grade III. Internal (AUC 0.884) and external (AUC 0.918) cohort validation indicated that the score had good predictive power.The STAPLAg score can be determined early in infected EICU patients, and exhibited better prognostic capacity than the initial SOFA score on admission in both internal and external cohorts. STAPLAg constitutes a new resource for use in the clinical diagnosis of sepsis and can also predict mortality in infected EICU patients. REGISTRATION NUMBER:: ChinCTR-PNC-16010288.

Monocyte distribution width enhances early sepsis detection in the emergency department beyond SIRS and qSOFA

Background

The initial presentation of sepsis in the emergency department (ED) is difficult to distinguish from other acute illnesses based upon similar clinical presentations. A new blood parameter, a measurement of increased monocyte volume distribution width (MDW), may be used in combination with other clinical parameters to improve early sepsis detection. We sought to determine if MDW, when combined with other available clinical parameters at the time of ED presentation, improves the early detection of sepsis.

Methods

A retrospective analysis of prospectively collected clinical data available during the initial ED encounter of 2158 adult patients who were enrolled from emergency departments of three major academic centers, of which 385 fulfilled Sepsis-2 criteria, and 243 fulfilled Sepsis-3 criteria within 12 h of admission. Sepsis probabilities were determined based on MDW values, alone or in combination with components of systemic inflammatory response syndrome (SIRS) or quick sepsis-related organ failure assessment (qSOFA) score obtained during the initial patient presentation (i.e., within 2 h of ED admission).

Results

Abnormal MDW (> 20.0) consistently increased sepsis probability, and normal MDW consistently reduced sepsis probability when used in combination with SIRS criteria (tachycardia, tachypnea, abnormal white blood count, or body temperature) or qSOFA criteria (tachypnea, altered mental status, but not hypotension). Overall, and regardless of other SIRS or qSOFA variables, MDW > 20.0 (vs. MDW ≤ 20.0) at the time of the initial ED encounter was associated with an approximately 6-fold increase in the odds of Sepsis-2, and an approximately 4-fold increase in the odds of Sepsis-3.

Conclusions

MDW improves the early detection of sepsis during the initial ED encounter and is complementary to SIRS and qSOFA parameters that are currently used for this purpose. This study supports the incorporation of MDW with other readily available clinical parameters during the initial ED encounter for the early detection of sepsis.

Trial registration

ClinicalTrials.gov, NCT03145428. First posted May 9, 2017. The first subjects were enrolled June 19, 2017, and the study completion date was January 26, 2018.