Clinical characteristics and survival outcomes of terminally ill patients undergoing withdrawal of mechanical ventilation

Prediction of survival time after terminal extubation: the balance between critical care unit utilization and hospice medicine in the COVID-19 pandemic era

BACKGROUND

We established 1-h and 1-day survival models after terminal extubation to optimize ventilator use and achieve a balance between critical care for COVID-19 and hospice medicine.

METHODS

Data were obtained from patients with end-of-life status at terminal extubation from 2015 to 2020. The associations between APACHE II scores and parameters with survival time were analyzed. Parameters with a p-value ≤ 0.2 in univariate analysis were included in multivariate models. Cox proportional hazards regression analysis was used for the multivariate analysis of survival time at 1 h and 1 day.

RESULTS

Of the 140 enrolled patients, 76 (54.3%) died within 1 h and 35 (25%) survived beyond 24 h. No spontaneous breathing trial (SBT) within the past 24 h, minute ventilation (MV) ≥ 12 L/min, and APACHE II score ≥ 25 were associated with shorter survival in the 1 h regression model. Lower MV, SpO2 ≥ 96% and SBT were related to longer survival in the 1-day model. Hospice medications did not influence survival time.

CONCLUSION

An APACHE II score of ≥ 25 at 1 h and SpO2 ≥ 96% at 1 day were strong predictors of disposition of patients to intensivists. These factors can help to objectively tailor pathways for post-extubation transition and rapidly allocate intensive care unit resources without sacrificing the quality of palliative care in the era of COVID-19. Trial registration They study was retrospectively registered. IRB No.: 202101929B0.

Trend analysis of palliative care consultation service for terminally ill non-cancer patients in Taiwan: a 9-year observational study

Backgrounds

Early integration of palliative care for terminally ill non-cancer patients improves quality of life. However, there are scanty data on Palliative Care Consultation Service (PCCS) among non-cancer patients.

Methods

In this 9-year observational study Data were collected from the Hospice-Palliative Clinical Database (HPCD) of Taichung Veterans General Hospital (TCVGH). Terminally ill non-cancer patients with 9 categories of diagnoses who received PCCS during 2011 to 2019 were enrolled. Trend analysis was performed to evaluate differences in categories of diagnosis throughout study period, duration of PCCS, patient outcomes, DNR declaration, awareness of disease by patients and families before and after PCCS.

Results

In total, 536 non-cancer patients received PCCS from 2011 to 2019 with an average age of 70.7 years. The average duration of PCCS was 18.4 days. The distributions of age, gender, patient outcomes, family’s awareness of disease before PCCS, and patient’s awareness of disease after PCCS were significantly different among the diagnoses. Organic brain disease and Chronic kidney disease (CKD) were the most prevalent diagnoses in patients receiving PCCS in 2019. For DNR declaration, the percentage of patients signing DNR before PCCS remained high throughout the study period (92.8% in 2019). Patient outcomes varied according to the disease diagnoses.

Conclusion

This 9-year observational study showed that the trend of PCCS among non-cancer patients had changed over the duration of the study. An increasing number of terminally ill non-cancer patients received PCCS during late life, thereby increasing the awareness of disease for both patients and families, which would tend to better prepare terminally ill patients for end-of-life as they may consider DNR consent. Early integration of PCCS into ordinary care for terminally non-cancer patients is essential for better quality of life.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12904-021-00879-z.

Palliative Care Consultation Services on Terminally Ill Cancer Patients and Non-Cancer Patients: Trend Analysis from a 9-Year-Long Observational Study in Taiwan

Early integration of palliative care for terminally ill cancer and non-cancer patients improves quality of life. However, there are sparse data on results of palliative care consultation services (PCCS) between cancer and non-cancer patients. In this 9-year observational study, data were collected from the Hospice-Palliative Clinical Database (HPCD) of Taichung Veterans General Hospital (TCVGH). Terminally ill cancer and non-cancer patients who received PCCS during 2011 to 2019 were enrolled. Trend analysis was performed to evaluate differences in outcomes of PCCS, including duration of PCCS, the awareness of disease of patients and families before and after PCCS, status of PCCS termination, and DNR declaration before and after PCCS among cancer and non-cancer patients throughout study period. In total, 5223 cancer patients and 536 non-cancer patients received PCCS from 2011 to 2019. The number of people who received PCCS increased stably over the decade, both for cancer and non-cancer patients. The average duration of PCCS for cancer and non-cancer patients was 21.4 days and 18.4 days, respectively. Compared with non-cancer patients, cancer patients had longer duration of PCCS, less DNR declaration (82% vs. 98%, respectively), and more transfers to the palliative care unit (17% vs. 11%, respectively), or for palliative home care (12% vs.8%, respectively). Determinants of late referral to PCCS includes age (OR 0.992, 95% CI 0.987–0.996), DNR declaration after PCCS (OR 1.967, 95% CI 1.574–2.458), patients’ awareness after PCCS (OR 0.754, 95% CI 0.635–0.895), and status of PCCS termination. This 9-year observational study showed that the trend of PCCS among cancer and non-cancer patients had changed over the duration of the study, and early integration of PCCS to all patients is essential for both cancer and non-cancer patients.

Machine Learning Prediction Models for Mechanically Ventilated Patients: Analyses of the MIMIC-III Database

Background: Mechanically ventilated patients in the intensive care unit (ICU) have high mortality rates. There are multiple prediction scores, such as the Simplified Acute Physiology Score II (SAPS II), Oxford Acute Severity of Illness Score (OASIS), and Sequential Organ Failure Assessment (SOFA), widely used in the general ICU population. We aimed to establish prediction scores on mechanically ventilated patients with the combination of these disease severity scores and other features available on the first day of admission.

Methods: A retrospective administrative database study from the Medical Information Mart for Intensive Care (MIMIC-III) database was conducted. The exposures of interest consisted of the demographics, pre-ICU comorbidity, ICU diagnosis, disease severity scores, vital signs, and laboratory test results on the first day of ICU admission. Hospital mortality was used as the outcome. We used the machine learning methods of k-nearest neighbors (KNN), logistic regression, bagging, decision tree, random forest, Extreme Gradient Boosting (XGBoost), and neural network for model establishment. A sample of 70% of the cohort was used for the training set; the remaining 30% was applied for testing. Areas under the receiver operating characteristic curves (AUCs) and calibration plots would be constructed for the evaluation and comparison of the models' performance. The significance of the risk factors was identified through models and the top factors were reported.

Results: A total of 28,530 subjects were enrolled through the screening of the MIMIC-III database. After data preprocessing, 25,659 adult patients with 66 predictors were included in the model analyses. With the training set, the models of KNN, logistic regression, decision tree, random forest, neural network, bagging, and XGBoost were established and the testing set obtained AUCs of 0.806, 0.818, 0.743, 0.819, 0.780, 0.803, and 0.821, respectively. The calibration curves of all the models, except for the neural network, performed well. The XGBoost model performed best among the seven models. The top five predictors were age, respiratory dysfunction, SAPS II score, maximum hemoglobin, and minimum lactate.

Conclusion: The current study indicates that models with the risk of factors on the first day could be successfully established for predicting mortality in ventilated patients. The XGBoost model performs best among the seven machine learning models.

Humanity behind the intention of primary caregiver to choose withdrawing life-sustaining treatment for terminating patients

OBJECTIVE

Mechanical ventilation, a measure of life-sustaining treatment (LST), may not be helpful and can be devastating for patients with terminal illness. We explored the effects of demographic characteristics, attitude, subjective norms, and perceived behavioral control on the behavioral intentions of primary caregivers to withdraw LST of long-term ventilator-dependent patients.

METHODS

Primary caregivers of ventilator-dependent patients in the respiratory care units of six hospitals participated in the study. A cross-sectional design including the domains of attitude, subjective norms, perceived behavioral control, and behavioral intention was adopted.

RESULTS

Valid data for 99 participants were analyzed using logistic regression. Religious belief, a spousal relationship with the patient, item 5 in subjective norms, and item 5 in perceived behavioral control positively influenced the intention to withdraw patient LST.

CONCLUSIONS

Religious beliefs, a spousal relationship, perceived behavioral control (confidence in relieving patient suffering), and the opportunity of current favorable subjective norms are major determinants of the intention to withdraw patients' LST.

PRACTICE IMPLICATIONS

Shared decision-making with the kin and primary caregivers of long-term ventilator-dependent patients at the end of life is crucial.