MANAGEMENT OF DELIRIUM IN THE ELDERLY

Authors

  • Lucas Xavier dos Santos
  • Yngryson Almeida Diniz
  • Maria Larissa do Nascimento Melo
  • Renan Schmitz Marcheti
  • Nailon de Morais Kois
  • Maria Clara Teixeira da Silva
  • Pedro Augusto Godinho de Castilho
  • Miquéia Aurélia Vieira Diniz Dantas
  • Mike Draiher da Silva
  • Gustavo Stanislaski Bazana
  • Hercia Simone Palhano Oliveira Pereira
  • Rubia Martinez Santos

DOI:

https://doi.org/10.56238/sevened2026.016-001

Keywords:

Delirium, Elderly, Postoperative Complications, Prevention, Machine Learning

Abstract

Postoperative delirium (POD) is one of the most frequent and serious complications in elderly patients undergoing surgical procedures, characterized by acute and fluctuating disturbances of attention, consciousness, and cognition. Its occurrence is associated with significant adverse outcomes, including increased morbidity and mortality, prolonged hospital stay, functional and cognitive decline, institutionalization, and increased healthcare costs. The pathophysiology of POD is complex and multifactorial, involving the interaction between underlying brain vulnerability (predisposing factors) and acute perioperative stressors (precipitating factors). Predisposing factors include advanced age, frailty, malnutrition, multimorbidity, previous cognitive impairment, and depression. Precipitating factors include major surgeries, systemic inflammatory response, intraoperative hypotension, poorly controlled pain, and the use of medications such as benzodiazepines. Prevention of postoperative delirium (POD) is based on multicomponent non-pharmacological interventions, such as those proposed by the Hospital Elder Life Program (HELP), which have demonstrated effectiveness in reducing the incidence of the condition. Comprehensive Geriatric Assessment (CGA) in the preoperative period allows for the systematic optimization of modifiable risk factors. Evidence for pharmacological prophylaxis is limited, with dexmedetomidine standing out as a promising option in selected populations, while benzodiazepines should be avoided. Machine learning-based prediction models have demonstrated superior accuracy to traditional logistic regression in identifying high-risk patients, incorporating biomarkers such as brain natriuretic peptide (BNP), troponin T, and C-reactive protein. The relationship between delirium and dementia is intimate and bidirectional, with delirium frequently underdiagnosed in patients with pre-existing dementia. The clinical approach should prioritize the identification and management of triggering factors in all confused elderly individuals, regardless of suspicion of underlying dementia.

Published

2026-04-06

How to Cite

dos Santos, L. X., Diniz, Y. A., Melo, M. L. do N., Marcheti, R. S., Kois, N. de M., da Silva, M. C. T., de Castilho, P. A. G., Dantas, M. A. V. D., da Silva, M. D., Bazana, G. S., Pereira, H. S. P. O., & Santos, R. M. (2026). MANAGEMENT OF DELIRIUM IN THE ELDERLY. Seven Editora, 1-11. https://doi.org/10.56238/sevened2026.016-001