Performance of influenza-specific triage tools in an H1N1-positive cohort: P/F ratio better predicts the need for mechanical ventilation and critical care admission

An AI-based multiphase framework for improving the mechanical ventilation availability in emergency departments during respiratory disease seasons: a case study

BACKGROUND

Shortages of mechanical ventilation have become a constant problem in Emergency Departments (EDs), thereby affecting the timely deployment of medical interventions that counteract the severe health complications experienced during respiratory disease seasons. It is then necessary to count on agile and robust methodological approaches predicting the expected demand loads to EDs while supporting the timely allocation of ventilators. In this paper, we propose an integration of Artificial Intelligence (AI) and Discrete-event Simulation (DES) to design effective interventions ensuring the high availability of ventilators for patients needing these devices.

METHODS

First, we applied Random Forest (RF) to estimate the mechanical ventilation probability of respiratory-affected patients entering the emergency wards. Second, we introduced the RF predictions into a DES model to diagnose the response of EDs in terms of mechanical ventilator availability. Lately, we pretested two different interventions suggested by decision-makers to address the scarcity of this resource. A case study in a European hospital group was used to validate the proposed methodology.

RESULTS

The number of patients in the training cohort was 734, while the test group comprised 315. The sensitivity of the AI model was 93.08% (95% confidence interval, [88.46 - 96.26%]), whilst the specificity was 85.45% [77.45 - 91.45%]. On the other hand, the positive and negative predictive values were 91.62% (86.75 - 95.13%) and 87.85% (80.12 - 93.36%). Also, the Receiver Operator Characteristic (ROC) curve plot was 95.00% (89.25 - 100%). Finally, the median waiting time for mechanical ventilation was decreased by 17.48% after implementing a new resource capacity strategy.

CONCLUSIONS

Combining AI and DES helps healthcare decision-makers to elucidate interventions shortening the waiting times for mechanical ventilators in EDs during respiratory disease epidemics and pandemics.

Critical care and pandemic preparedness and response

Critical care was established partially in response to a polio epidemic in the 1950s. In the intervening 70 yr, several epidemics and pandemics have placed critical care and allied services under extreme pressure. Pandemics cause wholesale changes to accepted standards of practice, require reallocation and retargeting of resources and goals of care. In addition to clinical acumen, mounting an effective critical care response to a pandemic requires local, national, and international coordination in a diverse array of fields from research collaboration and governance to organisation of critical care networks and applied biomedical ethics in the eventuality of triage situations. This review provides an introduction to an array of topics that pertain to different states of pandemic acuity: interpandemic preparedness, alert, surge activity, recovery and relapse through the literature and experience of recent pandemics including COVID-19, H1N1, Ebola, and SARS.

Alteration of circulating redox balance in coronavirus disease-19-induced acute respiratory distress syndrome

BACKGROUND

Mechanisms underpinning ARDS induced by COVID-19 are mostly immune-mediated, but need to be completely clarified. This study aimed to investigate redox balance in COVID-19 patients with ARDS, trying to recognize possible differences from typical ARDS related to the pathophysiology of severe disease.

METHODS

Patients affected by ARDS and positive for the SARS-CoV-2 virus (N = 40, COVID-19) were compared to ARDS patients negative to the molecular test (N = 42, No COVID-19). Circulating markers of redox balance were measured in serum and erythrocytes, and related to markers of inflammation and coagulability.

RESULTS

No differences in serum markers of oxidative damage were found between both groups, but a reduction in total antioxidant status and serum ceruloplasmin level was observed in COVID-19 rather than No COVID-19 patients. Redox balance alterations were described in erythrocytes from COVID-19 with respect to No COVID-19 group, characterized by increased lipofuscin and malondialdehyde concentration, and reduced glutathione S-transferase and glutathione reductase activity. These markers were associated with circulating indexes of respiratory disease severity (Horowitz index and alveolar-to-arterial oxygen gradient), inflammation (interleukin-6 and interleukin-10), and hypercoagulability (D-dimer) in COVID-19 patients with ARDS.

CONCLUSIONS

ARDS caused by COVID-19 is sustained by impairment of redox balance, particularly in erythrocytes. This alteration is associated with the pro-inflammatory and pro-coagulant status which characterizes severe COVID-19.

Evaluation of a hypothetical decision-support tool for intensive care triage of patients with coronavirus disease 2019 (COVID-19)

Background: At the start of the coronavirus disease 2019 (COVID-19) pandemic there was widespread concern about potentially overwhelming demand for intensive care and the need for intensive care unit (ICU) triage. In March 2020, a draft United Kingdom (UK) guideline proposed a decision-support tool (DST). We sought to evaluate the accuracy of the tool in patients with COVID-19. Methods: We retrospectively identified patients in two groups: referred and not referred to intensive care in a single UK national health service (NHS) trust in April 2020. Age, Clinical Frailty Scale score (CFS), and co-morbidities were collected from patients’ records and recorded, along with ceilings of treatment and outcome. We compared the DST, CFS, and age alone as predictors of mortality, and treatment ceiling decisions. Results: In total, 151 patients were included in the analysis, with 75 in the ICU and 76 in the non-ICU-reviewed groups. Age, clinical frailty and DST score were each associated with increased mortality and higher likelihood of treatment limitation (p-values all <.001). A DST cut-off score of >8 had 65% (95% confidence interval (CI) 51%-79%) sensitivity and 63% (95% CI 54%-72%) specificity for predicting mortality. It had a sensitivity of 80% (70%-88%) and specificity of 96% (95% CI 90%-100%) for predicting treatment limitation. The DST was more discriminative than age alone (p<0.001), and potentially more discriminative than CFS (p=0.08) for predicting treatment ceiling decisions. Conclusions: During the first wave of the COVID-19 pandemic, in a hospital without severe resource limitations, a hypothetical decision support tool was limited in its predictive value for mortality, but appeared to be sensitive and specific for predicting treatment limitation.

Performance of the Pandemic Medical Early Warning Score (PMEWS), Simple Triage Scoring System (STSS) and Confusion, Uremia, Respiratory rate, Blood pressure and age ≥ 65 (CURB-65) score among patients with COVID-19 pneumonia in an emergency department triage setting: a retrospective study

BACKGROUND

Healthcare institutions are confronted with large numbers of patient admissions during large-scale or long-term public health emergencies like pandemics. Appropriate and effective triage is needed for effective resource use.

OBJECTIVES

To evaluate the effectiveness of the Pandemic Medical Early Warning Score (PMEWS), Simple Triage Scoring System (STSS) and Confusion, Uremia, Respiratory rate, Blood pressure and age ≥ 65 years (CURB-65) score in an emergency department (ED) triage setting.

DESIGN AND SETTING

Retrospective study in the ED of a tertiary-care university hospital in Düzce, Turkey.

METHODS

PMEWS, STSS and CURB-65 scores of patients diagnosed with COVID-19 pneumonia were calculated. Thirty-day mortality, intensive care unit (ICU) admission, mechanical ventilation (MV) need and outcomes were recorded. The predictive accuracy of the scores was assessed using receiver operating characteristic curve analysis.

RESULTS

One hundred patients with COVID-19 pneumonia were included. The 30-day mortality was 6%. PMEWS, STSS and CURB-65 showed high performance for predicting 30-day mortality (area under the curve: 0.968, 0.962 and 0.942, respectively). Age > 65 years, respiratory rate > 20/minute, oxygen saturation (SpO2) < 90% and ED length of stay > 4 hours showed associations with 30-day mortality (P < 0.05).

CONCLUSIONS

CURB-65, STSS and PMEWS scores are useful for predicting mortality, ICU admission and MV need among patients diagnosed with COVID-19 pneumonia. Advanced age, increased respiratory rate, low SpO2 and prolonged ED length of stay may increase mortality. Further studies are needed for developing the triage scoring systems, to ensure effective long-term use of healthcare service capacity during pandemics.

Allocation of intensive care resources during an infectious disease outbreak: a rapid review to inform practice

BACKGROUND

The COVID-19 pandemic has placed sustained demand on health systems globally, and the capacity to provide critical care has been overwhelmed in some jurisdictions. It is unknown which triage criteria for allocation of resources perform best to inform health system decision-making. We sought to summarize and describe existing triage tools and ethical frameworks to aid healthcare decision-making during infectious disease outbreaks.

METHODS

We conducted a rapid review of triage criteria and ethical frameworks for the allocation of critical care resources during epidemics and pandemics. We searched Medline, EMBASE, and SCOPUS from inception to November 3, 2020. Full-text screening and data abstraction were conducted independently and in duplicate by three reviewers. Articles were included if they were primary research, an adult critical care setting, and the framework described was related to an infectious disease outbreak. We summarized each triage tool and ethical guidelines or framework including their elements and operating characteristics using descriptive statistics. We assessed the quality of each article with applicable checklists tailored to each study design.

RESULTS

From 11,539 unique citations, 697 full-text articles were reviewed and 83 articles were included. Fifty-nine described critical care triage protocols and 25 described ethical frameworks. Of these, four articles described both a protocol and ethical framework. Sixty articles described 52 unique triage criteria (29 algorithm-based, 23 point-based). Few algorithmic- or point-based triage protocols were good predictors of mortality with AUCs ranging from 0.51 (PMEWS) to 0.85 (admitting SOFA > 11). Most published triage protocols included the substantive values of duty to provide care, equity, stewardship and trust, and the procedural value of reason.

CONCLUSIONS

This review summarizes available triage protocols and ethical guidelines to provide decision-makers with data to help select and tailor triage tools. Given the uncertainty about how the COVID-19 pandemic will progress and any future pandemics, jurisdictions should prepare by selecting and adapting a triage tool that works best for their circumstances.

Comparison of 2 Triage Scoring Guidelines for Allocation of Mechanical Ventilators

IMPORTANCE

In the current setting of the coronavirus disease 2019 pandemic, there is concern for the possible need for triage criteria for ventilator allocation; to our knowledge, the implications of using specific criteria have never been assessed.

OBJECTIVE

To determine which and how many admissions to intensive care units are identified as having the lowest priority for ventilator allocation using 2 distinct sets of proposed triage criteria.

DESIGN, SETTING, AND PARTICIPANTS

This retrospective cohort study conducted in spring 2020 used data collected from US hospitals and reported in the Philips eICU Collaborative Research Database. Adult admissions (N = 40 439) to 291 intensive care units from 2014 to 2015 who received mechanical ventilation and were not elective surgery patients were included.

EXPOSURES

New York State triage criteria and original triage criteria proposed by White and Lo.

MAIN OUTCOMES AND MEASURES

Sequential Organ Failure Assessment (SOFA) scores were calculated for admissions. The proportion of patients who met initial criteria for the lowest level of priority for mechanical ventilation using each set of criteria and their characteristics and outcomes were assessed. Agreement was compared between the 2 sets of triage criteria, recognizing differences in stated criteria aims.

RESULTS

Among 40 439 intensive care unit admissions of patients who received mechanical ventilation, the mean (SD) age was 62.6 (16.6) years, 54.9% were male, and the mean (SD) SOFA score was 4.5 (3.7). Using the New York State triage criteria, 8.9% (95% CI, 8.7%-9.2%) were in the lowest priority category; these lowest priority admissions had a mean (SD) age of 62.9 (16.6) years, used a median (interquartile range) of 57.3 (20.1-133.5) ventilator hours each, and had a hospital survival rate of 38.6% (95% CI, 37.0%-40.2%). Using the White and Lo triage criteria, 4.3% (95% CI, 4.1%-4.5%) were in the lowest priority category; these admissions had a mean (SD) age of 68.6 (13.2) years, used a median (interquartile range) of 61.7 (24.3-142.8) ventilator hours each, and had a hospital survival rate of 56.2% (95% CI, 53.8%-58.7%). Only 655 admissions (1.6%) were in the lowest priority category for both guidelines, with the κ statistic for agreement equal to 0.20 (95% CI, 0.18-0.21).

CONCLUSIONS AND RELEVANCE

Use of 2 initially proposed ventilator triage guidelines identified approximately 1 in every 10 to 25 admissions as having the lowest priority for ventilator allocation, with little agreement. Clinical assessment of different potential criteria for triage decisions in critically ill populations is important to ensure valid and equitable allocation of resources.

Practices in Triage and Transfer of Critically Ill Patients: A Qualitative Systematic Review of Selection Criteria

OBJECTIVES

To identify and appraise articles describing criteria used to prioritize or withhold a critical care admission.

DATA SOURCES

PubMed, Embase, Medline, EBM Reviews, and CINAHL Complete databases. Gray literature searches and a manual review of references were also performed. Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines were followed.

STUDY SELECTION

We sought all articles and abstracts of original research as well as local, provincial, or national policies on the topic of ICU resource allocation. We excluded studies whose population of interest was neonatal, pediatric, trauma, or noncritically ill. Screening of 6,633 citations was conducted.

DATA EXTRACTION

Triage and/or transport criteria were extracted, based on type of article, methodology, publication year, and country. An appraisal scale was developed to assess the quality of identified articles. We also developed a robustness score to further appraise the robustness of the evidence supporting each criterion. Finally, all criteria were extracted, evaluated, and grouped by theme.

DATA SYNTHESIS

One-hundred twenty-nine articles were included. These were mainly original research (34%), guidelines (26%), and reviews (21%). Among them, we identified 200 unique triage and transport criteria. Most articles highlighted an exclusion (71%) rather than a prioritization mechanism (17%). Very few articles pertained to transport of critically ill patients (4%). Criteria were classified in one of four emerging themes: patient, condition, physician, and context. The majority of criteria used were nonspecific. No study prospectively evaluated the implementation of its cited criteria.

CONCLUSIONS

This systematic review identified 200 criteria classified within four themes that may be included when devising triage programs including the coronavirus disease 2019 pandemic. We identified significant knowledge gaps where research would assist in improving existing triage criteria and guidelines, aiming to decrease arbitrary decisions and variability.

Triage

This review provides an overview of triaging critically ill or injured patients during mass casualty incidents due to events such as disasters, pandemics, or terrorist incidents. Questions clinicians commonly have, including "what is triage?," "when to triage?," "what are the types of disaster triage?," "how to triage?," "what are the ethics of triage?," "how to govern triage?," and "what research is required on triage?," are addressed.

Can the emergency department triage category and clinical presentation predict hospitalization of H1N1 patients?

Background

Human H1N1 Influenza A virus was first reported in 2009 when seasonal outbreaks consistently occurred around the world. H1N1 patients present to the emergency departments (ED) with flu-like symptoms extending up to severe respiratory symptoms that require hospital admission. Developing a prediction model for patient outcomes is important to select patients for hospital admission. To date, there is no available data to guide the hospital admission of H1N1 patients based on their initial presentation.

Objective

The aim of this study was to investigate the predictors of hospital admission of H1N1 patients presenting in the ED.

Methods

We conducted a retrospective review of all laboratory-confirmed H1N1 cases presenting to the ED of a tertiary university hospital in the Eastern region of Saudi Arabia within the period from November 2015 to January 2016. We retrieved data of the initial triage category, vital signs, and presenting symptoms. Multivariate logistic regression analysis was performed to evaluate risk factors for hospital admission among H1N1patients presented to the ED.

Results

We identified 333 patients with laboratory-confirmed H1N1. Patients were classified into two groups: admitted group (n=80; 24%) and non-admitted group (n=253; 76%). Sixty patients (75%) were triaged under category IV. Triage category of level III and less were the most predictive for hospital admission. Multivariate regression analysis showed that of all vital signs, tachypnea was a significant risk factor for hospital admission (OR=1.1; 95% CI 1.02 to 1.13, p<0.01). The association between lower triage category and hospital stay was statistically significant (χ² =6.068, p=0.037). Also, patients with dyspnea were 4.5 times more likely to have longer hospital stay (OR=4.5; 95% CI 1.2 to 17.1, p=0.025).

Conclusion

Lower triage category and increased respiratory rate predict the need for hospital admission of H1N1 infected patients; while patients with dyspnea or bronchial asthma are likely to stay longer in the hospital. Further prospective studies are needed to evaluate the accuracy of using the CTAS and other clinical parameters in predicting hospitalization of H1N1 patients during outbreaks.

Early decrease of oxidative stress by non-invasive ventilation in patients with acute respiratory failure

Oxidative stress plays an important role in chronic respiratory diseases where the use of non-invasive ventilation seems to reduce the oxidative damage. Data on acute respiratory failure are still lacking. The aim of the study is to investigate the interplay between oxidative stress and acute respiratory failure, and the role of non-invasive ventilation in this setting. We enrolled 60 patients suffering from acute respiratory failure (PaO₂/FiO₂ ratio <300): 30 consecutive patients treated with non-invasive ventilation and 30 consecutive patients treated with conventional oxygen therapy. Serum levels of soluble Nox2-derived peptide (sNOX2-dp), a marker of NADPH-oxidase activation, and 8-iso-PGF2α and H₂O₂, markers of oxidative stress, were evaluated at baseline and after 3 h of treatment. At baseline, higher values of sNOX2-dp, 8-iso-PGF2α and H₂O₂ are associated with lower values of PaO₂/FiO₂ ratio (p < 0.001). After 3 h, serum levels of sNOX2-dp, H₂O₂, and 8-iso-PGF2α significantly decrease in patients treated with non-invasive ventilation, but not in patients treated with conventional oxygen therapy. Delta changes of oxidative stress parameters correlate inversely with the delta changes of PaO₂/FiO₂ (R = -0.623, p < 0.001 for sNOX2-dp; R = -0.428, p < 0.001 for H₂O₂; R = -0.548, p < 0.001 for 8-iso-PGF2α). In the acute respiratory failure setting, treatment with non-invasive ventilation reduces the levels of oxidative stress in the first hours. This reduction is associated with an improvement of PaO₂/FiO₂ ratio as well as in a reduction of NADPH-oxidase activity.

Oxygen exchange and C-reactive protein predict safe discharge in patients with H1N1 influenza

BACKGROUND

: Pandemic influenza has potential to overwhelm healthcare resources. There is uncertainty over performance of existing triage tools for hospital admission and discharge decisions.

AIM

: Our aim was to identify clinical criteria that predict safe discharge from hospital and develop a pragmatic triage tool to guide physician decision-making.

DESIGN

: We retrospectively examined an existing database of patients who presented to the Royal Liverpool University Hospital during the 2010-11 influenza pandemic.

METHODS

Inclusion criteria: patients ≥18 years, with PCR confirmed H1N1 influenza. Exclusion criteria: died in the emergency department or case notes unavailable. Successful discharge was defined as discharge within 24 h of presentation and no readmission within 7 days.

RESULTS

Eighty-six patients were included and 16 were successfully discharged. Estimated P/F ratio and C-reactive protein predicted safe discharge in a multivariable logistic regression model (AUC 0.883). A composite univariate predictor (estimated P/F minus C-reactive protein, AUC 0.877) was created to calculate specific cut off points for sensitivity and specificity. A pragmatic decision tool was created to incorporate these thresholds and relevant guidelines. Discharge: SpO 2 (in air)   ≥ 94% and CRP <50. Observe: SpO 2  ≥   94% and CRP >50 or SpO 2  ≤   93% and CRP <50. Admit: SpO 2  ≤   93% and CRP >50.

CONCLUSIONS

We identified that oxygen exchange and CRP, a marker of acute inflammation, were the most important predictors of safe discharge. Our proposed simple triage model requires validation but has the potential to aid clinical decisions in the event of a future pandemic, and potentially for seasonal influenza.

Triage decisions for ICU admission: Report from the Task Force of the World Federation of Societies of Intensive and Critical Care Medicine

Demand for intensive care unit (ICU) resources often exceeds supply, and shortages of ICU beds and staff are likely to persist. Triage requires careful weighing of the benefits and risks involved in ICU admission while striving to guarantee fair distribution of available resources. We must ensure that the patients who occupy ICU beds are those most likely to benefit from the ICU's specialized technology and professionals. Although prognosticating is not an exact science, preference should be given to patients who are more likely to survive if admitted to the ICU but unlikely to survive or likely to have more significant morbidity if not admitted. To provide general guidance for intensivists in ICU triage decisions, a task force of the World Federation of Societies of Intensive and Critical Care Medicine addressed 4 basic questions regarding this process. The team made recommendations and concluded that triage should be led by intensivists considering input from nurses, emergency medicine professionals, hospitalists, surgeons, and allied professionals. Triage algorithms and protocols can be useful but can never supplant the role of skilled intensivists basing their decisions on input from multidisciplinary teams. Infrastructures need to be organized efficiently both within individual hospitals and at the regional level. When resources are critically limited, patients may be refused ICU admission if others may benefit more on the basis of the principle of distributive justice.