Sepsis criteria versus clinical diagnosis of sepsis in burn patients: A validation of current sepsis scores

Yan, Juming; Hill, William F.; Rehou, Sarah; Pinto, Ruxandra; Shahrokhi, Shahriar; Jeschke, Marc G.

doi:10.1016/j.surg.2018.05.053

Cited by 49 publications

(41 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Sepsis-3 has also been investigated, however its heavy reliance on the SOFA score remains controversial in multiple populations including in burned patients. It must be noted that a new study recently suggested Sepsis-3 may be superior to ABA Consensus Guidelines 5 , however, performance appears to still lag behind AI/ML models produced in this study.…”

Section: Discussioncontrasting

confidence: 54%

Novel application of an automated-machine learning development tool for predicting burn sepsis: proof of concept

Tran

Albahra

Pham

et al. 2020

Sci Rep

View full text Add to dashboard Cite

Sepsis is the primary cause of burn-related mortality and morbidity. Traditional indicators of sepsis exhibit poor performance when used in this unique population due to their underlying hypermetabolic and inflammatory response following burn injury. To address this challenge, we developed the Machine intelligence Learning optimizer (MiLo), an automated machine learning (ML) platform, to automatically produce ML models for predicting burn sepsis. We conducted a retrospective analysis of 211 adult patients (age ≥ 18 years) with severe burn injury (≥ 20% total body surface area) to generate training and test datasets for ML applications. the MiLo approach was compared against an exhaustive "non-automated" ML approach as well as standard statistical methods. For this study, traditional multivariate logistic regression (LR) identified seven predictors of burn sepsis when controlled for age and burn size (OR 2.8, 95% CI 1.99-4.04, P = 0.032). The area under the ROC (ROC-AUC) when using these seven predictors was 0.88. Next, the non-automated ML approach produced an optimal model based on LR using 16 out of the 23 features from the study dataset. Model accuracy was 86% with ROC-AUC of 0.96. In contrast, MILO identified a k-nearest neighbor-based model using only five features to be the best performer with an accuracy of 90% and a ROC-AUC of 0.96. Machine learning augments burn sepsis prediction. MILO identified models more quickly, with less required features, and found to be analytically superior to traditional ML approaches. Future studies are needed to clinically validate the performance of MILO-derived ML models for sepsis prediction. Burn patients are at high risk for infections, with sepsis being the most common cause of morbidity and mortality 1. Traditional indicators of sepsis defined previously by the Surviving Sepsis Campaign 2 and other organizations exhibit poor performance when used in this unique population due to their underlying hypermetabolic and inflammatory response to burn injury. For example, the systemic inflammatory response syndrome 2,3 lacks clinical sensitivity and specificity when applied to severely burned patients 1 , while the newer 2016 "Sepsis-3" criteria remain controversial in both burned and non-burned patients 4-7. To this end, early and accurate recognition of sepsis represents a significant clinical knowledge gap in burn critical care. The American Burn Association (ABA) Consensus Guidelines published in 2007 was intended to better differentiate burn sepsis from the natural host-response to injury (Table 1) 1. These guidelines recognized deficiencies of traditional indications of sepsis and removed less specific parameters such as white blood cell count (WBC). Fever was redefined as temperatures > 39 °C to improve specificity and at the cost of sensitivity. Glycemic variability and thrombocytopenia were also included in the ABA Consensus Guidelines; however, measurement of glycemic variability is challenging without continuous glucose monitoring technology and platelet count aids ...

show abstract

Section: Discussioncontrasting

confidence: 54%

Novel application of an automated-machine learning development tool for predicting burn sepsis: proof of concept

Tran

Albahra

Pham

et al. 2020

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Burns patients are generally more susceptible to infection due to impaired immune function. Risk factors for sepsis in burns include >20% of total body surface area (TBSA), inhalation injury, delayed burn wound excision, increased length of hospital stay, use of artificial medical devices and ICU admission [24][25][26][27][28]. Patients in burn ICUs are more vulnerable to septicaemia than general ICU patients [29].…”

Section: Discussionmentioning

confidence: 99%

Molecular and epidemiological analysis of a Burkholderia cepacia sepsis outbreak from a tertiary care hospital in Bangladesh

et al. 2020

View full text Add to dashboard Cite

Background Burkholderia cepacia complex (Bcc) is a group of serious pathogens in cystic fibrosis patients and causes life threatening infections in immunocompromised patients. Species within the Bcc are widely distributed within the environment, can survive in the presence of disinfectants and antiseptics, and are inherently multidrug resistant (MDR). Methods Dhaka Medical College Hospital (DMCH) patients with a B. cepacia positive blood culture between 20 October 2016 to 23 rd September 2017 were considered as outbreak cases. Blood stream infections (BSIs) were detected using BacT/ALERT 3D at DMCH. B. cepacia was isolated on chromogenic UTI media followed by MALDI-TOF. Minimum inhibitory concentration (MIC) of clinically relevant antibiotics was determined by agar dilution. Whole genome sequencing was performed on an Illumina MiSeq platform. Patients' demographic and clinical data were collected. Patients' clinical history and genomic data of the outbreak strains were merged to investigate possible outbreaks. Ninety-one B. cepacia genomes were downloaded from 'Burkholderia Genome Database' and the genomic background of the global strains were compared with our outbreak strains. Results Among 236 BSIs, 6.35% (15/236) were B. cepacia. Outbreak cases were confined to the burn critical care unit and, to a lesser extent, the paediatrics department. There was a continuum of overlapping cases at DMCH between 23 October 2016 to 30 August 2017. Core genome SNPs showed that the outbreak strains were confined to a single clade, corresponded to a common clone (ST1578). The strains were shown to be MDR and associated with a mortality of 31% excluding discharge against medical advice. MIC profiles of the PLOS NEGLECTED TROPICAL DISEASES

show abstract

“…Therefore, clear diagnostic criteria for burn sepsis are necessary to minimize and prevent septic complications. The updated criterion in the Sepsis-3 consensus definition established in 2016 focuses more on multiple organ dysfunction than on signs of inflammation [ 15 ], compared with the American Burn Association (ABA) sepsis criteria (2007) [ 60 ] and the Mann-Salinas novel burn-specific sepsis predictors (2013) [ 61 , 62 ] ( Table 3 ). This is particularly discerning, considering that a series of pathophysiological events can lead to sepsis and multiple organ failure, including inflammatory response, hypovolaemic shock and vascular leak, immune dysregulation and hypermetabolism ( Figure 3 ), with inflammation present almost throughout the whole process from initial injury to burn wound healing.…”

Section: Reviewmentioning

confidence: 99%

The pathogenesis and diagnosis of sepsis post burn injury

et al. 2021

View full text Add to dashboard Cite

Burn is an under-appreciated trauma that is associated with unacceptably high morbidity and mortality. Although the survival rate after devastating burn injuries has continued to increase in previous decades due to medical advances in burn wound care, nutritional and fluid resuscitation and improved infection control practices, there are still large numbers of patients at a high risk of death. One of the most common complications of burn is sepsis, which is defined as “severe organ dysfunction attributed to host's disordered response to infection” and is the primary cause of death in burn patients. Indeed, burn injuries are accompanied by a series of events that lead to sepsis and multiple organ dysfunction syndrome, such as a hypovolaemic state, immune and inflammatory responses and metabolic changes. Therefore, clear diagnostic criteria and predictive biomarkers are especially important in the prevention and treatment of sepsis and septic shock. In this review, we focus on the pathogenesis of burn wound infection and the post-burn events leading to sepsis. Moreover, the clinical and promising biomarkers of burn sepsis will also be summarized.

show abstract

Sepsis criteria versus clinical diagnosis of sepsis in burn patients: A validation of current sepsis scores

Cited by 49 publications

References 18 publications

Novel application of an automated-machine learning development tool for predicting burn sepsis: proof of concept

Novel application of an automated-machine learning development tool for predicting burn sepsis: proof of concept

Molecular and epidemiological analysis of a Burkholderia cepacia sepsis outbreak from a tertiary care hospital in Bangladesh

The pathogenesis and diagnosis of sepsis post burn injury

Contact Info

Product

Resources

About