Stroke mortality prediction using machine learning: systematic review.

BACKGROUND AND AIMS: Accurate prognostication of stroke may help in appropriate therapy and rehabilitation planning. In the past few years, several machine learning (ML) algorithms were applied for prediction of stroke outcomes. We aimed to examine the performance of machine learning-based models for the prediction of mortality after stroke, as well as to identify the most prominent factors for mortality. MATERIALS AND METHODS: We searched MEDLINE/PubMed and Web of Science databases for original publications on machine learning applications in stroke mortality prediction, published between January 1, 2011, and October 27, 2022. Risk of bias and applicability were evaluated using the tailored QUADAS-2 tool. RESULTS: Of the 1015 studies retrieved, 28 studies were included. Twenty-Five studies were retrospective. The ML models demonstrated a favorable range of AUC for mortality prediction (0.67-0.98). In most of the articles, the models were applied for short-term post stroke mortality. The number of explanatory features used in the models to predict mortality ranged from 5 to 200, with substantial overlap in the variables included. Age, high BMI and high NIHSS score were identified as important predictors for mortality. Almost all studies had a high risk of bias in at least one category and concerns regarding applicability. CONCLUSION: Using machine learning, data available at the time of admission may aid in stroke mortality prediction. Notwithstanding, current research is based on few preliminary works with high risk of bias and high heterogeneity. Thus, future prospective, multicenter studies with standardized reports are crucial to firmly establish the usefulness of the algorithms in stroke prognostication.

Mortality And Prognostic Factors In Patients Operated For Acute Epidural Hematoma.

Background: Epidural hematoma is one of the most common surgical emergencies encountered in neurosurgery. This study was conducted to determine the mortality and prognostic factors in patients operated for traumatic intracranial epidural hematoma in a resource-constrained setting from a developing country. Methods: This retrospective study was conducted in the Department of Neurosurgery at Ayub Teaching Hospital Abbottabad from 1st January 2019 to 31st Dec 2021. Inclusion and exclusion criteria were created. The medical record of 116 patients admitted and operated on for traumatic extradural hematoma was retrospectively reviewed. Information was recorded using a standardized structured questionnaire. The outcome was measured in terms of the Glasgow coma outcome score. Results: Out of 116 patients, 93 were male and 23 were female.19 (16.4%) patients were in the age range 0-5 years, 42 (36.2%) were in the age range 6-15, 35(31.0%) were in the age range 16-30, 11 (9.5%) were in the age range 31-45 years, 7 (6.0%) were in the age range 46-60 years while only one patient included in this study was above 60 years. Overall mortality was 4.3%. Mortality was higher in females and in those aged less than 5 years (3.4%). 4 out of 16 patients died with GCS less than 8, while none of the patients died when GCS was above 12. Mortality was significantly higher in the presence of associated lesions (4.4% vs. 0%) and anisocoria (2.6% vs.1.7%). Patients who operated within 6 hours of trauma resulted in better outcomes with a mortality rate of 0.0% and functional recovery of 57.8 % while for those who operated after 6 hours, mortality was significantly higher (4.3%) and functional recovery was significantly low (15.5%). Conclusion: Good surgical outcomes can be achieved with early operative intervention if indicated. Female gender, low preoperative GCS score, presence of pupillary dilatation, presence of associated lesions, delayed surgical intervention and age less than 5 years are significant predictors for poor outcomes.

Déficit de base à admissão na unidade de terapia intensiva: um indicador de mortalidade precoce / Base deficit at intensive care unit admission: an early mortality indicator

JUSTIFICATIVA E OBJETIVOS: O déficit de base é considerado um indicador de lesão tissular, choque e reanimação. O objetivo deste estudo foi estabelecer uma associação entre o déficit de base na admissão dos pacientes internados em unidade de terapia intensiva (UTI) e seu prognóstico. MÉTODO: Estudo retrospectivo com análise de 110 pacientes admitidos consecutivamente na UTI, durante o período de 01 de junho a 31 de dezembro de 2006. RESULTADOS: Houve predomínio do sexo feminino, com idade média de 54,2 ± 18,7 anos. O tempo médio de permanência foi 6,5 ± 7,4 dias e o APACHE médio foi de 21 ± 8,1 pontos. A razão de mortalidade padronizada foi 0,715. A mortalidade dos pacientes com déficit de base superior a 6 mEq/L foi maior (38,9 por cento) que a daqueles com déficit menor (ou excesso) (20,6 por cento); p < 0,05. Os pacientes com mortalidade precoce (primeiras 48h de UTI) tiveram maior déficit de base (7,75 ± 8,33 mEq/L) que os sobreviventes (3,17 ± 5,43 mEq/L); p < 0,05. Não se documentou associação entre o tempo de permanência e o déficit de base. CONCLUSÕES: O déficit de base revelou-se significativamente maior entre os pacientes que faleceram precocemente, principalmente quando superior a 6 mEq/L.

BACKGROUND AND OBJECTIVES: Base deficit is considered an indicator of tissue injury, shock and resuscitation. The objective of this study was to establish an association between base deficit obtained on the admission of patients in intensive care unit (ICU) and their prognosis. METHODS: A retrospective study with analysis of 110 patients admitted consecutively in the ICU, during the period of June to December 2006. RESULTS: There was a predominance of women, with age mean 54.2 ± 18.7 years old. Length of stay in ICU was 6.5 ± 7.4 days and the mean APACHE II score was 21 ± 8.1 points. The standardized mortality ratio was 0.715. Mortality was higher in patients with base deficit > 6 mEq/L (38.9 percent) than in those with base deficit < 6 mEq/L (20.6 percent); p < 0.05. Patients with early mortality had lower base deficit (7.75 ± 8.33 mEq/L) than survivors (3.17 ± 5.43 mEq/L); p < 0.05. Patients with permanence in ICU until 7 days and patients that stayed in this unit for more than 7 days had similar base deficit. CONCLUSIONS: Base deficit had been associated with early mortality during ICU internment. Base deficit > 6 mEq/L is a marker of significant mortality.