Results: Overall mortality at 14 days was 22.8%. Models had a high prediction performance, with the best prediction for overall mortality achieved through Naive Bayes (area under the curve = 0.906). The most significant predictors were the GCS at admission and prehospital GCS, age, and pupil reaction. When predicting the length of stay at the intensive care unit, the Conditional Inference Tree model had the best performance (root mean square error = 1.011), with the most important variable across all models being the GCS at scene.Conclusions: Models for early mortality and hospital length of stay using Machine Learning can achieve high performance when based on registry data even in LMICs. These models have the potential to inform treatment decisions and counsel family members.Level of evidence: This observational study provides a level IV evidence on prognosis after TBI.
On January 2013, a disaster at Santa Maria (RS) due to a fire in a confined space
caused 242 deaths, most of them by inhalation injury. On November 2013, four
individuals required intensive care following smoke inhalation from a fire at the
Memorial da América Latina in São Paulo (SP). The
present article reports the clinical progression and management of disaster victims
presenting with inhalation injury. Patients ERL and OC exhibited early respiratory
failure, bronchial aspiration of carbonaceous material, and carbon monoxide
poisoning. Ventilation support was performed with 100% oxygen, the aspirated material
was removed by bronchoscopy, and cyanide poisoning was empirically treated with
sodium nitrite and sodium thiosulfate. Patient RP initially exhibited cough and
retrosternal burning and subsequently progressed to respiratory failure due to upper
airway swelling and early-onset pulmonary infection, which were treated with
protective ventilation and antimicrobial agents. This patient was extubated following
improvement of edema on bronchoscopy. Patient MA, an asthmatic, exhibited carbon
monoxide poisoning and bronchospasm and was treated with normobaric hyperoxia,
bronchodilators, and corticosteroids. The length of stay in the intensive care unit
varied from four to 10 days, and all four patients exhibited satisfactory functional
recovery. To conclude, inhalation injury has a preponderant role in fires in confined
spaces. Invasive ventilation should not be delayed in cases with significant airway
swelling. Hyperoxia should be induced early as a therapeutic means against carbon
monoxide poisoning, in addition to empiric pharmacological treatment in suspected
cases of cyanide poisoning.
BackgroundIntensive care unit (ICU) admission triage is performed routinely and is often based solely on clinical judgment, which could mask biases. A computerized algorithm to aid ICU triage decisions was developed to classify patients into the Society of Critical Care Medicine’s prioritization system. In this study, we sought to evaluate the reliability and validity of this algorithm.MethodsNine senior physicians evaluated forty clinical vignettes based on real patients. The reference standard was defined as the priorities ascribed by two investigators with full access to patients’ records. Agreement of algorithm-based priorities with the reference standard and with intuitive priorities provided by the physicians were evaluated. Correlations between algorithm prioritization and physicians’ judgment of the appropriateness of ICU admissions in scarcity and nonscarcity settings were also evaluated. Validity was further assessed by retrospectively applying this algorithm to 603 patients with requests for ICU admission for association with clinical outcomes.ResultsAgreement between algorithm-based priorities and the reference standard was substantial, with a median κ of 0.72 (interquartile range [IQR] 0.52–0.77). Algorithm-based priorities demonstrated higher interrater reliability (overall κ 0.61, 95 % confidence interval [CI] 0.57–0.65; median percentage agreement 0.64, IQR 0.59–0.70) than physicians’ intuitive prioritization (overall κ 0.51, 95 % CI 0.47–0.55; median percentage agreement 0.49, IQR 0.44–0.56) (p = 0.001). Algorithm-based priorities were also associated with physicians’ judgment of appropriateness of ICU admission (priorities 1, 2, 3, and 4 vignettes would be admitted to the last ICU bed in 83.7 %, 61.2 %, 45.2 %, and 16.8 % of the scenarios, respectively; p < 0.001) and with actual ICU admission, palliative care consultation, and hospital mortality in the retrospective cohort.ConclusionsThis ICU admission triage algorithm demonstrated good reliability and validity. However, more studies are needed to evaluate a difference in benefit of ICU admission justifying the admission of one priority stratum over the others.Electronic supplementary materialThe online version of this article (doi:10.1186/s13054-016-1262-0) contains supplementary material, which is available to authorized users.
Objective To assess the impact of intracranial pressure monitoring on the short-term
outcomes of traumatic brain injury patients.Methods Retrospective observational study including 299 consecutive patients admitted due
to traumatic brain injury from January 2011 through July 2012 at a Level 1 trauma
center in São Paulo, Brazil. Patients were categorized in two groups
according to the measurement of intracranial pressure (measured intracranial
pressure and non-measured intracranial pressure groups). We applied a
propensity-matched analysis to adjust for possible confounders (variables
contained in the Crash Score prognostic algorithm).Results Global mortality at 14 days (16%) was equal to that observed in high-income
countries in the CRASH Study and was better than expected based on the CRASH
calculator score (20.6%), with a standardized mortality ratio of 0.77. A total of
28 patients received intracranial pressure monitoring (measured intracranial
pressure group), of whom 26 were paired in a 1:1 fashion with patients from the
non-measured intracranial pressure group. There was no improvement in the measured
intracranial pressure group compared to the non-measured intracranial pressure
group regarding hospital mortality, 14-day mortality, or combined hospital and
chronic care facility mortality. Survival up to 14 days was also similar between
groups.Conclusion Patients receiving intracranial pressure monitoring tend to have more severe
traumatic brain injuries. However, after adjusting for multiple confounders using
propensity scoring, no benefits in terms of survival were observed among
intracranial pressure-monitored patients and those managed with a systematic
clinical protocol.
Respiratory-dependent dynamic parameters for predicting fluid responsiveness in ICU may have restricted applicability in daily practice, even in more severe patients, due to low prevalence of required conditions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.