This website is intended for healthcare professionals only.
Take a look at a selection of our recent media coverage:
27th February 2024
A clinical trial at Amsterdam UMC is aiming to show that an artificial intelligence (AI) algorithm can reduce the number of blood cultures taken in emergency departments (EDs) worldwide by almost a third.
The algorithm takes data from the electronic health record of patients with suspected sepsis, such as body temperature and other standard lab results that provide information about possible infections, to identify patients with a low risk of infection or inflammation.
The clinician uses this prediction, together with their own observation, to then consider whether a blood culture is required.
Dr Prabath Nanayakkara, professor of internal medicine at Amsterdam UMC, whose work focuses on increasing efficiency in acute care, said: ‘Of all the blood cultures that are done in the emergency room, only 15% indicate that something is wrong, and it’s even more interesting if you take a closer look at those positive results. Then about half of the positive blood cultures turn out to be false positives. We’re used to doing it this way but that doesn’t mean the analysis is necessary. Certainly not when you look at the numbers.’
The new clinical trial – known as the ABC study – will test the algorithm in practice and is one of the first randomised controlled trials worldwide to consider such an algorithm in acute care.
It was previously developed by Amsterdam UMC’s acute AI team and extensively tested and validated nationally and internationally, the team said.
The AI’s prediction is now built into the electronic patient record at Amsterdam UMC, so the healthcare provider is shown this prediction on the computer screen when requesting a blood culture in the intervention group, which the researchers say is a ‘unique’ approach.
‘With tailor-made advice for each patient, we expect to drastically reduce unnecessary cultures, hopefully by up to 30%,’ said Professor Nanayakkara.
‘If this study is successful, we will not only reduce the number of blood cultures – and therefore the costs – but also ensure that the number of false positive blood tests decreases and thus, no patients will be treated unnecessarily.
‘False positive results lead to uncertainty among patients and doctors, extra procedures, excessive treatment with antibiotics and, longer hospital admissions and even higher mortality in that group.’
He added: ‘We shouldn’t do things in healthcare that aren’t necessary [and] this is pre-eminently an issue in which AI can be used. In doing so, we are showing that AI can really change healthcare.’
In October 2023, a study presented at the European Society of Emergency Medicine congress led to calls for increased sepsis awareness as two of the four internationally recommended sepsis screening tools used by emergency medical services were deemed inadequate for recognising the condition.
And in February 2024, updated guidance from the National Institute for Health and Care Excellence on identifying and managing sepsis in over-16s recommended better targeting of antibiotics for suspected sepsis.
2nd February 2024
Updated guidance from the National Institute for Health and Care Excellence (NICE) on identifying and managing sepsis in over-16s recommends better targeting of antibiotics for suspected sepsis.
The updates specify that secondary care teams should target antibiotic use as more is learned about a patient’s condition to ensure the right people receive treatment as soon as possible but the medicines are not overused, which can lead to antibiotic resistance.
Clinicians should use the NEWS2 risk score to assess those with suspected sepsis in those aged over 16 who are not or have not been recently pregnant and are in an acute setting or ambulance, the guidelines state.
NICE said use of NEWS2 should lead to more people with suspected sepsis graded at a lower risk level where treatment should begin within one to three hours and the diagnosis clarified before antibiotics are given, targeted at a specific infection if possible.
People graded by NEWS2 as being the most severely ill should be prioritised and continue to receive broad-spectrum antibiotics within an hour, the guideline adds.
The new NICE recommendations also aim to ensure better communication and inclusivity for patients with suspected sepsis if they have learning disabilities, autism or where English is not their first language.
This includes assessing these individuals at every stage of the healthcare journey with extra care if they cannot give a good history and tailoring the timing, content and delivery of information to the person’s needs and preferences.
Detail on identifying the source of infection and involving surgical teams have also been broadened to cover risk of sepsis in all parts of the body and a wider range of interventions.
This partial update to the 2016 guidelines also urges immediate hospital transfer for patients who meet any high-risk criteria for sepsis.
GPs or ambulance services must pre-alert secondary care when a patient with suspected sepsis is on their way, the new recommendations state.
And in remote and rural locations where transfer time to emergency department is likely to be more than an hour, GPs and ambulance services should have mechanisms in place to start antibiotics straightaway while ensuring due consideration is given to both patient safety and antimicrobial stewardship, NICE said.
Altered mental state, raised respiratory rate, new need for oxygen, low blood pressure, raised heart rate or not passing urine, a mottled or ashen appearance, cyanosis or a non-blanching rash are all listed as high-risk factors that primary healthcare professionals should look for.
Figures suggest there are at least 245,000 sepsis cases diagnosed in the UK every year.
Professor Jonathan Benger, NICE chief medical officer said the guidance would help ensure antibiotics are targeted to those at the greatest risk of severe sepsis, so they get rapid and effective treatment.
He said: ‘We know that sepsis can be difficult to diagnose so it is vital there is clear guidance on the updated NEWS2 so it can be used to identify illness, ensure people receive the right treatment in the right clinical setting and save lives.
‘This update is the latest part of the process to ensure NICE guidance is as current as possible. We recognise this is a vital and rapidly evolving area, so this is the latest in a series of planned updates to our guidance.’
UK Sepsis Trust founder Dr Ron Daniels welcomed the update and said: ‘We particularly support that the update continues to recommend the identification of high-risk factors, whilst reinforcing the importance of clinical judgement to prevent injudicious use of antibiotics.
‘The recommendation for GPs and ambulance services to consider how they give antibiotics to people that are at high risk of sepsis is increasingly relevant as transit times increase, and could be potentially transformational in terms of patient outcomes.
He added: ‘We’re now presented with an opportunity to deliver a coordinated and cohesive approach to the recognition and management of sepsis across the NHS.’
3rd October 2023
Two of the four internationally recommended sepsis screening tools used by emergency medical services (EMS) are inadequate for recognising the condition, according to a new study presented at the European Society of Emergency Medicine (EUSEM)‘s recent congress.
The study also highlighted that both EMS and emergency physicians should be more vigilant in documenting a suspicion of sepsis so that further clinical patient assessment and evaluation can take place. This could help the condition to be diagnosed and treated earlier, which could help to save lives and maintain patients‘ quality of life.
Researchers from the Charité – Universitätsmedizin Berlin and the universities of Magdeburg and Jena in Germany, aimed to assess which of four screening tools recommended in the Surviving Sepsis Campaign guidelines was best for emergency medical services (EMS) to predict the condition.
They found only the National Early Warning Score (NEWS-2) screening tool to have a reasonably accurate prediction rate for sepsis. A second tool – the quick Sequential Organ Failure Assessment (qSOFA) – had a high level of accuracy in predicting patients who did not have it.
However, the two other standard screening tools, Systemic Inflammatory Response Syndrome (SIRS) and the Modified Early Warning Score (MEWS), performed poorly in both sensitivity and specificity of sepsis detection.
Dr Wolfgang Bauer, senior physician at the Charité, said: ‘In emergency care, there are good and long-established standards for the detection and treatment of heart attacks and stroke that have improved patients’ chances of survival. Unfortunately, a lot less attention is paid to sepsis and standards to improve early sepsis recognition and survival.
‘Our study found there was a similar incidence for sepsis (1.6%) as for heart attacks (2.6%) and stroke (2.7%) in cases seen by emergency medical services. However, in terms of both percentages and absolute numbers, more patients died from sepsis than from heart attacks or stroke.
‘Out of all cases with sepsis, 31.4% died within 30 days after being seen by emergency services, versus 13.4% and 11.8% respectively for heart attacks and stroke. These findings emphasise the need for better sepsis awareness and more frequent use of effective screening tools.‘
The retrospective cohort study linked data on 221,429 emergency care cases, with follow-up between 2016 and 2017 from 10 health insurance companies, with information from documentation by paramedics and emergency doctors. In doing so, it was possible to calculate the ability of the four tools to predict a sepsis diagnosis.
Only the NEWS-2 had a reasonably accurate prediction rate for sepsis, with it able to correctly predict 72.2% of all cases (sensitivity) and correctly identify 81.4% of negative, non-septic cases (specificity).
In contrast, the qSOFA had a much lower sensitivity (24.1%) but a high specificity (96.6%) for identifying those without sepsis.
For SIRS, the corresponding sensitivity and specificity were 30.4% and 93.8%, respectively, and for MEWS they were 46.8% and 88.4%.
In total, 24.3% of cases were predicted to have sepsis by at least one of the screening tools, and only 0.9% of cases were predicted to have the conditon by all four tools.
The predictions were confirmed or rejected during subsequent hospital investigations after contact with EMS.
Silke Piedmont, a health scientist for the Department of Emergency Medicine, Campus Benjamin Franklin at the Charité, and first author on the abstract said: ‘We found that paramedics never documented a suspicion of sepsis, and emergency services physicians rarely did so, only documenting a suspicion in 0.1% of cases. The screening tools recommended in the Surviving Sepsis Campaign guidelines differed greatly in terms of which and how many patients were identified as possibly having sepsis.’
Emphasising the need for better awareness and more frequent use of screening tools, Ms Piedmont added: ‘No screening tool provides ideal performance. NEWS-2 best supports EMS in identifying most patients with sepsis. EMS patients that are NEWS-2 positive should be flagged up as potentially having sepsis and referred for special attention and assessment by emergency doctors who are expert in sepsis. If EMS insist on using the qSOFA, they should be aware that a positive qSOFA makes sepsis likely, but also that a negative qSOFA cannot rule out sepsis conclusively.
‘A rule of thumb for EMS staff could be that NEWS-2 negative patients are the most likely not to have sepsis, and qSOFA positive patients are the most likely to have sepsis – and also that qSOFA misses many patients with sepsis.‘
The researchers hope their findings will help to inform new guidelines on the condition, which are being drawn up in some countries to give more specific recommendations for its screening. In particular, Ms Piedmont said: ‘Future sepsis guidelines should be more precise and omit recommendations for MEWS and SIRS for EMS since they were inferior in all the measures for accuracy.’
These findings could also apply to other countries, the researchers suggest, especially as previous studies performed in Canada and the UK support aspects of their results.
‘Ours is the first study comparing all four screening tools and showing the predictive usefulness of applying the screening tools to all adult patients independently of any presumptions or preliminary diagnoses by EMS,‘ said Mrs Piedmont.
10th March 2023
A prospective study by researchers at the Karolinska University Hospital, Stockholm, Sweden, has revealed that elevated plasma calprotectin levels was superior to other biomarkers for the identification of emergency department (ED) patients with sepsis who required urgent transfer to either an intensive care (ICU) or high dependency unit (HDU).
In data for 2017, it was reported that globally, there were an estimated 48.9 million incident cases of sepsis and which led to 11 million deaths. Moreover, it is also recognised that the early detection of sepsis results in greater patient survival. Calprotectin is a protein present in the cytoplasm of neutrophils and levels have been found to increase early in response to a bacterial challenge and in fact, other work demonstrates how serum calprotectin levels are a reliable biomarker in patients with bacterial sepsis. Nevertheless, while measurement of serum calprotectin levels is of value in sepsis, whether the biomarker serves to identify ED sepsis patients who required direct transfer to an ICU or HDU is less clear.
In the current study, the Swedish team turned to patients ED patients identified as likely to have sepsis based on an alert system used within the department, i.e., those with signs of organ dysfunction and symptoms of infection. Among such patients, blood samples were analysed for several biomarkers including C-reactive protein, procalcitonin, neutrophils, lymphocytes and calprotectin which was analysed from frozen plasma samples. The researchers set the primary outcome as direct transfer to an ICU or HDU from the ED, based on the decision of the multidisciplinary team.
Calprotectin levels and ED to ICU transfer
Among a total of 319 patients with a median age of between 71 and 73 (37.5% female) identified with an infection, 26% were immediately transferred to either ICU or HDU and the remainder to a hospital ward.
The overall median calprotectin level was 2.2 mg/L (IQR 1.2 – 3.9 mg/L) for all infected patients but was significantly higher (3.3 vs 2.1, p = 0.0001) among those transferred to either an ICU or HDU compared to a ward.
Using the receiver operating characteristic curve analysis, calprotectin had the highest area under the curve (AUC) at 0.65 compared to C-reactive protein (AUC = 0.55), procalcitonin (AUC = 0.46) and the neutrophil-lymphocyte ratio (AUC = 0.47).
A cut-off value for calprotectin of 4.0 mg/L, gave the best combination of sensitivity (42.5%) and specificity (83%) for the transfer to the ICU/HDU among patients with infection.
The authors concluded that among sepsis alert patients, plasma calprotectin was significantly elevated in patients requiring immediate transfer to ICU or HDU and proved to be superior to other biomarkers for the prediction of such transfers.
Citation
Parke A et al. Plasma calprotectin as an indicator of need of transfer to intensive care in patients with suspected sepsis at the emergency department BMC Emerg Med 2023
17th October 2022
Digital PCR is an effective technique for the rapid identification of pathogens in patients with sepsis according to a recent meta-analysis.
According to the Third International Consensus definitions, sepsis should be considered as a life-threatening organ dysfunction caused by a dysregulated host response to infection.
It affects a huge number of individuals with data from 2017 estimating that globally, 48.9 million incident cases and 11.0 million sepsis-related deaths were recorded.
Bacteria are the most common cause of sepsis, with 62.2% of patients with positive blood cultures harbouring Gram-negative bacteria and 46.8% infected with Gram-positive bacteria.
The identification of the bacterial species responsible for sepsis is therefore of enormous importance to help guide appropriate antibiotic therapy.
Nevertheless, this is not always straight forward and in a study of 6,843,279 admissions of patients with severe sepsis, 47.1% had culture-negative results.
Polymerase chain reaction (PCR) offering highly multiplexed targeting of bacterial and/or fungal pathogens in whole blood samples, provides the best opportunity for clinical impact with results available within four to eight hours after a blood sample has been taken.
Despite this faster turnaround time, PCR systems are typically associated with lower sensitivity and specificity than post-culture detection methods which rely on microbial growth.
A more recent development has been the introduction of digital PCR which is potentially more accurate and, in the present study, a team of Chinese researchers systematically assessed the accuracy of digital PCR, the detection time and its potential clinical value.
They searched for articles where the technology was used to detect micro-organisms from whole blood samples and collected data to calculate the sensitivity, specificity as well as the positive and negative predictive value and diagnostic odds ratio.
The positive predictive value (PPV) represents the probability of disease given a positive test (hence higher values are better) and negative predictive value (NPV) is the probability of no disease following a negative test.
The diagnostic odds ratio describes the odds of a positive test in those with disease relative to the odds of a positive test in those without disease.
The literature search identified eight eligible studies with a total of 1,278 patients.
The pooled sensitivity was 94% (95% CI 85 – 98%) and the pooled specificity was 87% (95% CI 76 – 94%) for the diagnostic accuracy of digital PCR.
The PPV was 84.7% and the NPV was 89.2% and taken together, these figures indicate a good overall accuracy for digital PCR.
The pre-test probability is the probability that a patient has a condition, in this case sepsis, before testing.
In a US study of 568 patients who died in the hospital or were discharged to hospice, sepsis was present in 52.8% of cases.
Using this as the pre-test probability, the authors calculated that with a positive digital PCR test result, the probability of contracting sepsis increased to 89% but when negative, the probability reduced to 7%.
The authors stated that these figures indicated that digital PCR had both a good accuracy and a high clinical diagnostic value.
The detection time with digital PCR ranged from four to six hours and which was much faster than from blood cultures (ranging from 48 to 120 hours).
The authors concluded that digital PCR had both a high sensitivity and specificity for the clinical diagnosis of sepsis and that the process was much faster the blood cultures.
Taken together, these findings suggested that the technique could ensure timely use of antibiotic treatment for patients.
Citation
Li Y et al. Value of digital PCR in the early diagnosis of sepsis: A systematic review and meta-analysis J Crit Care 2022.
Cystatin C (CC) is a serine protease inhibitor that can be used as a marker of glomerular filtration rate (GFR). In fact, there is a suggestion that measurement of cystatin C should be used for the initial prediction of GFR of a patient and among critically ill patients, serum CC levels significantly outperforms serum creatinine for the detection of an impaired GFR. Moreover, other work has shown that CC-based estimates of GFR in both the elderly and ethnically diverse populations in comparison to serum creatinine, was a better predictor of all-cause mortality. While GFR derived estimates from either CC or creatinine generally agree, a decrease in the CC estimate compared to that of creatinine has been suggested to be due to what has been described as shrunken pore syndrome (SPS). It has since been recognised that SPS has been associated with a substantial increase in mortality or morbidity in all investigated populations.
However, whether cystatin C-based estimates of GFR and SPS are linked to a higher mortality among intensive care unit (ICU) patients with sepsis is uncertain and was the subject of the present study by Swedish researchers. The team undertook a post hoc analysis of data from the FINNAKI study which was a prospective observational study of acute kidney injury patients. For the present analysis, included patients were those with severe sepsis either upon ICU admission or which developed during the period of study. Plasma samples were used to measure both cystatin C and creatinine levels and from which GFR estimates were calculated. The primary outcome was 90-day mortality, whereas secondary outcomes were the development of acute kidney injury (AKI) between 12 and 5 days after ICU admission and renal replacement therapy. CC plasma levels estimated GFR based on CC and creatine were divided into quartiles.
Cystatin C measurements and mortality
A total of 802 patients with a mean age of 65 years (35.9% female) were included. The presence of SPS was present in 9.9% of patients when using an estimated GFRCystatin to GFRcreatinine cut-off ratio of 0.6 and 20% when the cut-off was set at 0.70. A total of 176 patients developed AKI between 12 hours and 5 days after ICU admission.
For plasma CC levels in the highest quartile, there was a positive and significant association with increased 90-day mortality compared to the lowest quartile (hazard ratio, HR = 4.15, 95% CI 2.17 – 7.91, p < 0.001). Similarly, there was a significant association with 90-day mortality for the lowest quartile of CC estimated GFR compared to the highest (HR = 4.45, 95% CI 2.28 – 8.68, p < 0.001). The association with SPS was also significant whether the cut-off was 0.6 or 0.70. In contrast, there was no significant association between 90-mortality and creatinine-based GFR estimates. Even after corticosteroid use in the treatment of septic shock, the associations for serum CC and estimated GFR remained significant although the association for SPS with a cut off of 0.6 was no longer significant (p = 0.14). When the researchers if the association between CC levels and 90-mortality were also linked to the development of AKI within 5 days, the analysis revealed how this association was maintained for the highest quartile of serum CC levels (HR = 4.09, 95% CI 2.14 – 7.80, p < 0.001), as well as CC estimated GFR and SPS.
The authors concluded that higher cystatin C levels together with a reduced CC-based estimate of GFR and the presence of SPS in patients with SPS in ICU was associated with a higher 90-day mortality and that a higher incidence of AKI does not explain this association.
Citation
Linne E et al. Cystatin C and derived measures of renal function as risk factors for mortality and acute kidney injury in sepsis – A post-hoc analysis of the FINNAKI cohort J Crit Care 2022
24th June 2022
The use of intravenous vitamin C for intensive care patients with sepsis who were also in receipt of vasopressor therapy, increased the risk of both death and persistent organ dysfunction according to the findings of LOVIT, a randomised, placebo-controlled trial.
Sepsis represents the body’s inflammatory response to infection and has been found to be the cause of one in every two to three hospital deaths, most of whom had sepsis upon admission to hospital. Moreover, sepsis represents a major cause of death, with an estimated 48·9 million global incident cases and 11·0 million sepsis-related deaths in 2017.
Intravenous vitamin C may be an important adjunctive therapy for critically ill patients. In fact, evidence suggests that critically ill patients have low vitamin C concentrations and that those with septic shock have significantly depleted vitamin C levels probably due to the increased metabolism from the enhanced inflammatory response in these patients.
However, subsequent studies were less consistent, with one network meta-analysis concluding that metabolic resuscitation with vitamin C, glucocorticoids, vitamin B1, or combinations of these drugs was not significantly associated with a decrease in longer-term mortality.
In contrast, another meta-analysis concluded that intravenous vitamin C administration appeared to be safe and may be associated with a trend toward reduction in overall mortality in critically ill patients.
Nevertheless, in another study of 167 patients with sepsis and acute respiratory distress syndrome, a 96-hour infusion of vitamin C compared did not significantly improve organ dysfunction scores or alter markers of inflammation and vascular injury.
Given the potentially conflicting results to date on the value of vitamin C, for the present study, researchers undertook a randomised, placebo-controlled trial, Lessening Organ Dysfunction with Vitamin C (LOVIT).
The trial was designed to test whether intravenous vitamin C would reduce both the risk of death or persistent organ dysfunction in adults with sepsis and who were receiving vasopressor therapy within an intensive care unit (ICU). The team recruited adult patients who had been within an ICU for no longer than 24 hours, with a proven or suspected infection as the main diagnosis and who were receiving a vasopressor.
Participants were randomised 1:1 to receive intravenous vitamin C as a bolus dose of 50 mg/kg, administered over 30 to 60 minutes every 6 hours for a total of 96 hours. The control group received a matching infusion containing either 5% dextrose or normal saline.
The primary outcome of the trial was a composite of death or persistent organ dysfunction, whereas one of the secondary outcomes was 28-day mortality.
Intravenous vitamin C and death or persistent organ dysfunction
A total of 863 patients with a mean age of 65.1 years (37.6% female) were randomised to either intravenous vitamin C or matching placebo. The median stay within the ICU was 6 days and with an overall hospital stay of 16 days.
The primary outcome occurred in 44.5% of those in the vitamin C group and 38.5% of placebo patients and this difference was significant (risk ratio, RR = 1.21, 95% CI 1.04 – 1.40, p = 0.01).
For the secondary outcome of 28-day mortality, death occurred in 35.4% of those using vitamin C and 31.6% of those receiving placebo (RR = 1.17, 95% CI 0.98 – 1.40). Similarly, the proportion of patients with persistent organ dysfunction was present in 9.1% of those given vitamin C and 6.9% of those taking placebo (RR = 1.30, 95% CI 0.83 – 2.05).
The authors concluded that in adults with sepsis who were receiving vasopressor therapy, use of intravenous vitamin C led to a higher risk of death or persistent organ dysfunction compared to placebo.
Citation
Lamontagne F et al. Intravenous Vitamin C in Adults with Sepsis in the Intensive Care Unit N Eng J Med 2022
4th February 2022
Early intravenous fluid use in sepsis patients seen at emergency departments has been found to increase after the introduction of several improvement interventions. This was the conclusion of a study by a team from Western Sydney University, School of Nursing and Midwifery, Australia.
Sepsis is a syndrome of physiologic, pathologic and biochemical abnormalities which is induced by infection. Although the global burden is difficult to ascertain with a high degree of certainty, the World Health Organization reports that in 2017, there were an estimated 48.9 million cases and as many as 11 million sepsis-related deaths globally.
Early intravenous fluid resuscitation is crucial for stabilisation of sepsis-induced tissue hypo-perfusion or septic shock and has been recommended by the ‘Surviving Sepsis Campaign’ guideline in 2017. In the latest update, it is recommended that crystalloid fluid should be given within the first 3 hours of resuscitation.
The Australian team wondered to what extent any interventions to improve early intravenous fluid administration might improve compliance with the sepsis guidelines and set out to find an examine the value of any such interventions.
The team undertook a literature search for studies in adults presenting at an emergency department with sepsis and included those trials where the purpose of the intervention or strategy was to influence early intravenous fluid administration, for example, through educational programs, sepsis alerts or inclusion in sepsis protocols.
Findings
A total of 31 studies were included in the analysis. In 15 of these studies with a total of 1538 patients assigned to the intervention arm, the summary estimate indicated a 47% improvement in the rate of compliance with early intravenous fluid use (relative risk, RR = 1.47, 95% CI 1.25 – 1.74, p < 0.01).
The mean difference in time to initiation of early fluids was – 24.11 minutes, i.e., an average reduction of 24 minutes in the time to fluid resuscitation between intervention and control groups. Similarly, patients received an average additional 575 ml of fluids within the protocol recommended time as a result of the intervention.
In discussing their findings, the authors reported how early intravenous fluid administration compliance increased by an average 47% and more importantly, this improvement was seen in a variety of emergency departments across the world.
The authors described how there were a range of interventions employed which ranged from educational to process change measures such as a multidisciplinary sepsis program.
They concluded that performance improvement initiatives appeared to significantly improve compliance with early intravenous fluid administration and called for future studies to examine the subjective factors influencing the use of an early fluid bolus.
Citation
Kabil G et al. Early fluid bolus in adults with sepsis in the emergency department: a systematic review, meta-analysis and narrative synthesis BMC Emerg Med 2022
13th September 2021
Sepsis is a life-threatening organ dysfunction response to infection which, according to the World Health Organization, in 2017, affected 48.9 million people and caused 11 million sepsis-related deaths. The early administration of broad-spectrum antibiotics is associated with a reduced progression to severe sepsis and septic shock and a lower mortality. These findings have prompted the development of several automated rule-based sepsis notification systems which have been combined with alerting systems. In a review of studies, it was concluded that digital sepsis alerting systems, reduce both hospital and intensive care unit stays for patients with sepsis. The incorporation of sepsis notification systems within electronic health records (EHRs), providing a real-time alert, could therefore lead to improvements in patient outcomes although recent trial concluded that this was not the case.
Whether sepsis notification systems incorporated into EHRs and monitored by an emergency department pharmacist could improve patient outcomes was the subject of a study by a team from the Division of Pulmonary and Critical Care Medicine, MetroHealth Medical Centre, Cleveland, Ohio, US.
The team incorporated a sepsis alert within with the EHR and randomised patients admitted to hospital with suspected sepsis to either standard care or augmented care (intervention group) in which the sepsis early-warning system (EWS) was visible and monitored by an emergency department pharmacist. Once the sepsis EWS score crossed over the established threshold, an alert occurred, triggering a flag that was displayed on the patient emergency department tracking tool and a message to the EHR system which was monitored by an emergency department pharmacist. Once an alert was raised, the pharmacist ordered appropriate blood tests, antibiotics and fluid boluses for the patient. The primary outcome measure was time to antibiotics from arrival and the primary clinical outcome score was a composite of days alive and out of hospital, 28 days after arrival.
Findings
A total of 598 patients were included over a 5-month period and randomised to standard care (313) or intervention. The median age of participants ranged from 62.3 (standard care) to 61.5 (intervention group) years with the proportion of females ranging from 46% to 51.2%, standard care and intervention group respectively. Among those assigned to the intervention arm, the time to antibiotic administration from emergency department arrival was 2.3 hours compared with 3 hours in the standard care group (p = 0.039). In addition, the clinical primary outcome score was also higher for the intervention, reflecting better outcomes (median 24.1 vs 22.5 days, p = 0.011). However, the length of stay between groups was not significantly different (p = 0.124) and neither was hospital mortality (p = 0.086).
In their discussion, the authors noted how their results showed only a slightly modest improvement from the intervention. The study was designed as a quality improvement initiative to compare whether visibility of the sepsis notification systems by both clinicians and pharmacists improved outcomes. They concluded that further studies are needed to determine if their approach is generalisable to other healthcare settings.
Citation
Tarabichi Y et al. Improving Timeliness of Antibiotic Administration Using a Provider and Pharmacist Facing Sepsis Early Warning System in the Emergency Department Setting: A randomised Controlled Quality Improvement Initiative. Crit Care Med 2021.
26th July 2021
Sepsis can be defined as is a life-threatening organ dysfunction caused by a dysregulated host response to infection. Furthermore, sepsis is responsible for around 11 million deaths each year, which amounts to approximately 20% of all global deaths. Thus, it is crucial that clinicians have a comprehensive understanding of all the relevant clinical factors that can help with the early identification of those patients for whom a poor outcome is likely. This is particularly important since early use of crystalloid therapy reduces mortality, as does prompt administration of antibiotics. Though several scoring systems for sepsis are available, these are based on the assessment of vital signs but which can sometimes be normal upon admission to an emergency department. While machine learning has been shown to have some level of predictive power for mortality, none of the variables currently used in these models are reflective of the symptoms at first presentation. This led a team from the Department of Medical Sciences, Orebro University, Sweden, to use machine learning in an attempt to identify the variables which were predictive of 7- and 30-day mortality in sepsis patients, based on the clinical presentation at an emergency department. They employed a retrospective design and included patients 18 years and older, admitted to hospital with suspected sepsis. The team input previously identified variables, e.g., abnormal temperature, acute altered mental status, etc into the machine learning algorithm. The sensitivity and specificity of the predictive models generated by the machine learning model, were calculated from the area under the receiver operating curve (AUC).
Findings
A total of 445 patients with sepsis and a median age of 73 years (52.6% male) were included in the retrospective analysis. Overall, 234 (49.7%) had severe sepsis and 63 patients died within 7-days of admission and 98 within 30 days. The accuracy of the 7-day predictive model was maximal after the inclusion of only six variables; fever, abnormal verbal response, low oxygen saturation, arrival by emergency services, abnormal behaviour/level of consciousness and chills. Using these variables, the AUC sensitivity was 0.84 (95 CI 0.78–0.89) and the specificity 0.67 (95% CI 0.64 –0.70). For the prediction of 30-day mortality, again, only 6 variables were significant; abnormal verbal response, fever, chills, arrival by emergency services, low oxygen saturation and breathing difficulties. This model gave a sensitivity of 0.87 (95% CI 0.81–0.93) and a specificity of 0.64 (95% CI 0.61–0.67).
In discussing their findings, the authors highlighted how their results revealed the importance of the using a clinical symptom complex that was representative of what an emergency department clinician would be likely to encounter in practice. They also suggested that the 7-day model might be of more use in practice since it would be of assistance to emergency care staff for the likely short-term outcome for patients. They concluded that given how the clinical presentation of sepsis can often be non-specific, the use of a machine learning algorithm, based on symptoms and observations, would be most helpful to staff and that future work should focus on validating the method in other cohorts.
Citation
Karlsson A et al. Predicting mortality among septic patients presenting to the emergency department– a cross sectional analysis using machine learning. BMC Emerg Med 2021
Download Hospital Healthcare Europe free app
© Cogora 2024
Cogora Limited. 1 Giltspur Street, London EC1A 9DD
Registered in the United Kingdom. Reg. No. 2147432
