Prediction of Noninvasive Ventilation Failure in a Mixed Population Visiting the Emergency Department in a Tertiary Care Center in India

Mathen, Prannoy George; Kumar, Krishna; Mohan, Naveen; Sreekrishnan, TP; Nair, Sabarish B; Krishnan, Arun K.; Prasad, S Bharath; Ahamed, D Riaz; Theresa, Manna Maria; Kathyayini, VR; Vivek, U

doi:10.5005/jp-journals-10071-24338

Cited by 6 publications

(10 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Heart failure mortality has been looked at in patients requiring different levels of care from step-down care, progressive care, and intensive care to different stages and with many different covariates [ 35 – 38 ]. Researchers have utilized methods from logistic regression to machine learning [ 39 – 44 ]. Within machine learning researchers are starting to utilize transparent methods for visualization [ 24 , 45 ].…”

Section: Discussionmentioning

confidence: 99%

Dendrogram of transparent feature importance machine learning statistics to classify associations for heart failure: A reanalysis of a retrospective cohort study of the Medical Information Mart for Intensive Care III (MIMIC-III) database

Huang

2023

PLoS ONE

View full text Add to dashboard Cite

Background There is a continual push for developing accurate predictors for Intensive Care Unit (ICU) admitted heart failure (HF) patients and in-hospital mortality. Objective The study aimed to utilize transparent machine learning and create hierarchical clustering of key predictors based off of model importance statistics gain, cover, and frequency. Methods Inclusion criteria of complete patient information for in-hospital mortality in the ICU with HF from the MIMIC-III database were randomly divided into a training (n = 941, 80%) and test (n = 235, 20%). A grid search was set to find hyperparameters. Machine Learning with XGBoost were used to predict mortality followed by feature importance with Shapely Additive Explanations (SHAP) and hierarchical clustering of model metrics with a dendrogram and heat map. Results Of the 1,176 heart failure ICU patients that met inclusion criteria for the study, 558 (47.5%) were males. The mean age was 74.05 (SD = 12.85). XGBoost model had an area under the receiver operator curve of 0.662. The highest overall SHAP explanations were urine output, leukocytes, bicarbonate, and platelets. Average urine output was 1899.28 (SD = 1272.36) mL/day with the hospital mortality group having 1345.97 (SD = 1136.58) mL/day and the group without hospital mortality having 1986.91 (SD = 1271.16) mL/day. The average leukocyte count in the cohort was 10.72 (SD = 5.23) cells per microliter. For the hospital mortality group the leukocyte count was 13.47 (SD = 7.42) cells per microliter and for the group without hospital mortality the leukocyte count was 10.28 (SD = 4.66) cells per microliter. The average bicarbonate value was 26.91 (SD = 5.17) mEq/L. Amongst the group with hospital mortality the average bicarbonate value was 24.00 (SD = 5.42) mEq/L. Amongst the group without hospital mortality the average bicarbonate value was 27.37 (SD = 4.98) mEq/L. The average platelet value was 241.52 platelets per microliter. For the group with hospital mortality the average platelet value was 216.21 platelets per microliter. For the group without hospital mortality the average platelet value was 245.47 platelets per microliter. Cluster 1 of the dendrogram grouped the temperature, platelets, urine output, Saturation of partial pressure of Oxygen (SPO2), Leukocyte count, lymphocyte count, bicarbonate, anion gap, respiratory rate, PCO2, BMI, and age as most similar in having the highest aggregate gain, cover, and frequency metrics. Conclusion Machine Learning models that incorporate dendrograms and heat maps can offer additional summaries of model statistics in differentiating factors between in patient ICU mortality in heart failure patients.

show abstract

Section: Discussionmentioning

confidence: 99%

Dendrogram of transparent feature importance machine learning statistics to classify associations for heart failure: A reanalysis of a retrospective cohort study of the Medical Information Mart for Intensive Care III (MIMIC-III) database

Huang

2023

PLoS ONE

View full text Add to dashboard Cite

show abstract

Section: Dear Editormentioning

confidence: 99%

“…We are curious to know whether they used simple bedside parameters or scores predictive of NIV failure such as initial high respiratory rate, low PaO 2 /FiO 2 ratio, HACOR score >5 at the end of 1 hour of NIV, and initial hs-CRP. 4 Individuals with comorbidities and decreased functional capabilities are at higher risk for NIV failure. Patients with poor baseline performance status and having higher DECAF (dyspnea, eosinopenia, consolidation, acidemia, atrial fibrillation) score on admission can predict severity of acute clinical deterioration.…”

mentioning

confidence: 99%

Neutrophil/Lymphocyte Ratio as a Predictive Biomarker of NIV Failure in AECOPD: The Unveiled Facts [Letter]

Bhattacharya,

Esquinas,

Mandal

2024

COPD

View full text Add to dashboard Cite

We appreciate the work of Sun et al, 1 who observed that Neutrophil Lymphocyte Ratio (NLR) can be an effective marker for predicting Non Invasive Ventilation (NIV) failure in Acute Exacerbation of Chronic Obstructive Pulmonary Disease (AECOPD). They also observed that NLR >8.9 had higher risk of NIV failure requiring intubation. However, we raise some points for further clarification.NLR gets elevated in AECOPD compared with stable state. Serial monitoring of NLR can predict NIV failure better than single measurement. 2 NLR is an independent risk factor for 28-day mortality in patients with AECOPD, 2 and the mortality rate increases with NLR ≥4. 3 We are curious to know whether continual monitoring of NLR was done and compared with that of stable state in the same patient. In addition, it would be interesting to know whether they assessed mortality outcome, given that the length of hospital stay was observed already.NLR encompasses both neutrophilia and lymphopenia, the indicators of inflammation and decreased immune response, respectively. Thus, it can reflect the severity of COPD better than when neutrophilia or lymphopenia are monitored alone. Another novel inflammatory marker, PLR (serum absolute count ratio of platelet to lymphocyte) can help predict the severity of clinical condition and mortality in AECOPD patients. 3 The authors observed respiratory rate and arterial blood gas analysis. We are curious to know whether they used simple bedside parameters or scores predictive of NIV failure such as initial high respiratory rate, low PaO 2 /FiO 2 ratio, HACOR score >5 at the end of 1 hour of NIV, and initial hs-CRP. 4 Individuals with comorbidities and decreased functional capabilities are at higher risk for NIV failure. Patients with poor baseline performance status and having higher DECAF (dyspnea, eosinopenia, consolidation, acidemia, atrial fibrillation) score on admission can predict severity of acute clinical deterioration. 5 We wonder whether they had a scope to observe those in their study.

show abstract

“…We read with immense interest the article titled “Prediction of Non-invasive Ventilation Failure in a Mixed Population Visiting the Emergency Department in a Tertiary Care Center in India” by Mathen et al 1 We commend the authors for their research, but we would like to express our views about this.…”

mentioning

confidence: 99%

“…Authors have used the HACOR score along with high-sensitivity C-reactive protein for the prediction of NIV failure in patients presenting to the emergency department (ED). 1 The HACOR score was originally developed from the data obtained from respiratory intensive care units (ICUs) to predict NIV failure in patients with predominantly hypoxemic respiratory failure from respiratory etiology. 2 , 3 Several factors like pneumonia, cardiogenic pulmonary edema, pulmonary ARDS, immunosuppression, septic shock, and the SOFA score prior to NIV impact the success.…”

mentioning

confidence: 99%

Non-invasive Ventilation Failure – Predict and Protect

Jindal¹,

Pratyusha²

2023

Indian Journal of Critical Care Medicine

View full text Add to dashboard Cite

A bstract Modified heart rate, acidosis, consciousness, oxygenation and respiratory rate (HACOR) score takes into consideration pneumonia, cardiogenic pulmonary edema, pulmonary acute respiratory distress syndrome (ARDS), immunosuppression, septic shock, and the sequential organ failure assessment (SOFA) score prior to non-invasive mechanical ventilation (NIV) that would impact the success of NIV and are commonly seen in patients presenting to the emergency. Propensity score matching could have been done for similar distribution of baseline characteristics. Specific objective criteria are needed to define respiratory failure requiring intubation. How to cite this article Pratyusha K, Jindal A. Non-invasive Ventilation Failure – Predict and Protect. Indian J Crit Care Med 2023;27(2):149.

show abstract

Prediction of Noninvasive Ventilation Failure in a Mixed Population Visiting the Emergency Department in a Tertiary Care Center in India

Cited by 6 publications

References 15 publications

Dendrogram of transparent feature importance machine learning statistics to classify associations for heart failure: A reanalysis of a retrospective cohort study of the Medical Information Mart for Intensive Care III (MIMIC-III) database

Dendrogram of transparent feature importance machine learning statistics to classify associations for heart failure: A reanalysis of a retrospective cohort study of the Medical Information Mart for Intensive Care III (MIMIC-III) database

Neutrophil/Lymphocyte Ratio as a Predictive Biomarker of NIV Failure in AECOPD: The Unveiled Facts [Letter]

Non-invasive Ventilation Failure – Predict and Protect

Contact Info

Product

Resources

About