TRIAGE OF SEPSIS PATIENTS: SIRS OR QSOFA – WHICH IS BEST?

Primary Infection Site as a Predictor of Sepsis Development in Emergency Department Patients

BACKGROUND

Sepsis is a life-threatening condition but predicting its development and progression remains a challenge.

OBJECTIVE

This study aimed to assess the impact of infection site on sepsis development among emergency department (ED) patients.

METHODS

Data were collected from a single-center ED between January 2016 and December 2019. Patient encounters with documented infections, as defined by the Systematized Nomenclature of Medicine-Clinical Terms for upper respiratory tract (URI), lower respiratory tract (LRI), urinary tract (UTI), or skin or soft-tissue infections were included. Primary outcome was the development of sepsis or septic shock, as defined by Sepsis-1/2 criteria. Secondary outcomes included hospital disposition and length of stay, blood and urine culture positivity, antibiotic administration, vasopressor use, in-hospital mortality, and 30-day mortality. Analysis of variance and various different logistic regression approaches were used for analysis with URI used as the reference variable.

RESULTS

LRI was most associated with sepsis (relative risk ratio [RRR] 5.63; 95% CI 5.07-6.24) and septic shock (RRR 21.2; 95% CI 17.99-24.98) development, as well as hospital admission rates (odds ratio [OR] 8.23; 95% CI 7.41-9.14), intensive care unit admission (OR 4.27; 95% CI 3.84-4.74), in-hospital mortality (OR 6.93; 95% CI 5.60-8.57), and 30-day mortality (OR 7.34; 95% CI 5.86-9.19). UTIs were also associated with sepsis and septic shock development, but to a lesser degree than LRI.

CONCLUSIONS

Primary infection sites including LRI and UTI were significantly associated with sepsis development, hospitalization, length of stay, and mortality among patients presenting with infections in the ED.

A Case of Severe Acute Respiratory Distress Syndrome Secondary to Atypical Amniotic Fluid Embolism

Acute respiratory distress syndrome (ARDS) is a noncardiogenic pulmonary edema that leads to acute respiratory distress. It remains one of the major diagnoses requiring ICU admission and mechanical ventilation. We present a case of a 25-year-old gravida 3 para 2 female who was admitted for uncomplicated 38-week pregnancy and delivered a healthy male infant but developed acute onset dyspnea six hours after vaginal delivery. She required mechanical ventilation four hours after the onset of respiratory distress and had to be transferred to a higher level facility for extracorporeal membrane oxygenation (ECMO) within 24 hours of the symptom onset. She was diagnosed with severe ARDS. Even though she missed the other typical feature of amniotic fluid embolism, atypical amniotic fluid embolism remained the most likely explanation for her symptoms after the other causes of ARDS were excluded.

Comparison of QSOFA and sirs scores for the prediction of adverse outcomes of secondary peritonitis among patients admitted on the adult surgical ward in a tertiary teaching hospital in Uganda: a prospective cohort study

BACKGROUND

SIRS and qSOFA are two ancillary scoring tools that have been used globally, inside and outside of ICU to predict adverse outcomes of infections such as secondary peritonitis. A tertiary teaching hospital in Uganda uses SIRS outside the ICU to identify patients with secondary peritonitis, who are at risk of adverse outcomes. However, there are associated delays in decision making given SIRS partial reliance on laboratory parameters which are often not quickly available in a resource limited emergency setting. In response to the practical limitations of SIRS, the sepsis-3 task force recommends qSOFA as a better tool. However, its performance in patients with secondary peritonitis in comparison to that of SIRS has not been evaluated in a resource limited setting of a tertiary teaching hospital in a low and middle income country like Uganda.

OBJECTIVE

To compare the performance of qSOFA and SIRS scores in predicting adverse outcomes of secondary peritonitis among patients on the adult surgical wards in a tertiary teaching hospital in Uganda.

METHODS

This was a prospective cohort study of patients with clinically confirmed secondary peritonitis, from March 2018 to January 2019 at the Accident and Emergency unit and the adult surgical wards of a tertiary teaching hospital in Uganda. QSOFA and SIRS scores were generated for each patient, with a score of ≥2 recorded as high risk, while a score of < 2 recorded as low risk for the adverse outcome respectively. After surgery, patients were followed up until discharge or death. In-hospital mortality and prolonged hospital stay were the primary and secondary adverse outcomes, respectively. Sensitivity, specificity, PPV, NPV and accuracy at 95% confidence interval were calculated for each of the scores using STATA v.13.

RESULTS

A total of 153 patients were enrolled. Of these, 151(M: F, 2.4:1) completed follow up and were analysed, 2 were excluded. Mortality rate was 11.9%. Fourty (26.5%) patients had a prolonged hospital stay. QSOFA predicted in-hospital mortality with AUROC of 0.52 versus 0.62, for SIRS. Similarly, qSOFA predicted prolonged hospital stay with AUROC of 0.54 versus 0.57, for SIRS.

CONCLUSION

SIRS is superior to qSOFA in predicting both mortality and prolonged hospital stay among patients with secondary peritonitis. However, overall, both scores showed a poor discrimination for both adverse outcomes and therefore not ideal tools.

Transatlantic transferability of a new reinforcement learning model for optimizing haemodynamic treatment for critically ill patients with sepsis

INTRODUCTION

In recent years, reinforcement learning (RL) has gained traction in the healthcare domain. In particular, RL methods have been explored for haemodynamic optimization of septic patients in the Intensive Care Unit. Most hospitals however, lack the data and expertise for model development, necessitating transfer of models developed using external datasets. This approach assumes model generalizability across different patient populations, the validity of which has not previously been tested. In addition, there is limited knowledge on safety and reliability. These challenges need to be addressed to further facilitate implementation of RL models in clinical practice.

METHOD

We developed and validated a new reinforcement learning model for hemodynamic optimization in sepsis on the MIMIC intensive care database from the USA using a dueling double deep Q network. We then transferred this model to the European AmsterdamUMCdb intensive care database. T-Distributed Stochastic Neighbor Embedding and Sequential Organ Failure Assessment scores were used to explore the differences between the patient populations. We apply off-policy policy evaluation methods to quantify model performance. In addition, we introduce and apply a novel deep policy inspection to analyse how the optimal policy relates to the different phases of sepsis and sepsis treatment to provide interpretable insight in order to assess model safety and reliability.

RESULTS

The off-policy evaluation revealed that the optimal policy outperformed the physician policy on both datasets despite marked differences between the two patient populations and physician's policies. Our novel deep policy inspection method showed insightful results and unveiled that the model could initiate therapy adequately and adjust therapy intensity to illness severity and disease progression which indicated safe and reliable model behaviour. Compared to current physician behavior, the developed policy prefers a more liberal use of vasopressors with a more restrained use of fluid therapy in line with previous work.

CONCLUSION

We created a reinforcement learning model for optimal bedside hemodynamic management and demonstrated model transferability between populations from the USA and Europe for the first time. We proposed new methods for deep policy inspection integrating expert domain knowledge. This is expected to facilitate progression to bedside clinical decision support for the treatment of critically ill patients.

The challenge of recognising sepsis: Future nanotechnology solutions

The urgent need to start anti-infective therapeutic interventions in suspected sepsis, and the lack of specific time-critical diagnostic information often lead to the widespread administration of broad-spectrum antimicrobial therapies, increasing the risk of unwanted patient harms and contributing to rising pathogen antimicrobial resistance. Nanotechnology, which involves engineering at the nanoscale, allows for the bespoke development of diagnostic solutions with multi-functionality and high sensitivity that has the potential to help provide time-critical information to make more accurate diagnoses and treatment decisions for sepsis. Nanotechnologies also have the potential to improve upon the current strategies used for novel biomarker discovery. Here we describe some of the current limitations to identifying sepsis and explore the potential role for nanotechnology solutions.

Shock index and modified shock index as triage screening tools for sepsis

BACKGROUND

Fever is one of the common conditions encountered in the emergency department, which related to a spectrum of diseases severity. Identifying sepsis patients from uncomplicated febrile patients is challenging in the emergency triage areas and pre-hospital settings.

OBJECTIVES

Assess the triage shock index (SI) and modified shock index (MSI) in febrile patients as predictors for sepsis and sepsis-related outcomes.

DESIGN

A retrospective cohort study.

SETTING

Patients presented to the Emergency Department of King Khalid University Hospital.

PATIENTS AND METHODS

The analysis included all febrile adult patients triaged with a temperature of 38 °C or more from January 2016 to December 2017. Based on triage vital sign we calculate the SI with cut-off levels of ≥0.7 and ≥1 and MSI with cut-off levels of ≥1 and ≥1.3. We report the Relative Risk, Sensitivity, Specificity, Positive and Negative Predictive Values of the predictors.

MAIN OUTCOME MEASURES

Sepsis and sepsis-related outcomes such as hyperlactatemia, ICU admission, and 28 days mortality.

SAMPLE SIZE

274 patients.

RESULTS

274 patients met our inclusion/exclusion criteria. Of the 274 patients, 252 patient (92%) were septic, 62 patients (22%) had hyperlactatemia, 20 patients admitted to the ICU, and 5 patient died within 28 days. An MSI of ≥1 had a sensitivity of 90% for sepsis predication, 85% for ICU admission and 100% for 28 days mortality. MSI of ≥1.3 showed a specificity (59%-100%) for all the outcomes of interest. Non-significant statistical trends of greater accuracy of MSI over SI.

CONCLUSION

MSI and SI were found to be promising predictors in triaging febrile patients. However no single cut-off values of MSI or SI were found to have an optimal accuracy for prediction of sepsis and sepsis-related outcomes. Further studies are required to assess the incorporation of MSI in a multi-item scaling system for the prediction of sepsis and its related outcomes.

LIMITATIONS

Small single center study and the results may not be generalizable.