End-of-Life Care in an Acute Neurological Ward: A Retrospective Analysis of Clinical Practices During the Last Week of Life

BACKGROUND: The care of patients with neurological diseases near the end of life (EOL) is challenging and often complicated by prognostic uncertainty. Difficulty recognizing the terminal phase may delay transitions to EOL care, leading to continued use of diagnostic procedures and medical interventions with a limited focus on comfort. METHODS: This retrospective observational study included medical records of 209 deceased patients with neurological diseases who died in an acute hospital ward. Diagnostic procedures and medical and supportive interventions delivered during the final 7 days of life were extracted. Patients were categorized into 2 groups: acute neurological diseases (ischemic stroke and hemorrhagic stroke) and progressive neurological diseases (Parkinson disease or extrapyramidal and movement disorders, and amyotrophic lateral sclerosis). Frequencies of interventions were compared between diagnostic groups and according to documented EOL decision status. RESULTS: In both groups, documented EOL decisions were associated with reduced use of diagnostic procedures and increased use of support measures. Medication administration and supportive care patterns differed between the acute and progressive disease groups and were related to documented EOL decisions. The most frequently administered interventions were oral suctioning (97%), intravenous fluids (70%), antibiotics (45%), pain relief medications (98%), and benzodiazepines (93%). Patients with progressive neurological diseases received more respiratory support interventions than those with acute neurological diseases. CONCLUSIONS: Our findings suggest that documenting EOL decisions was related to comfort-focused care, although the frequency varied by disease group. Neurological disease-specific EOL guidelines should be implemented to identify the terminal phase in alignment with the goals of comfort-focused care in a timely manner.