Ting Ting Wu, BCCCP, BCPS
Critical Care Pharmacy Research Fellow
Northeastern University
Boston, Massachusetts
Disclosure information not submitted.
Marieke Zegers, PhD
Senior researcher
Department of Intensive Care Medicine, Radboud Institute Health Science, Radboud University Nijmegen Medical Center, United States
Disclosure information not submitted.
Sally Ko, BSc, n/a
PharmD candidate
Northeastern University, United States
Disclosure information not submitted.
Rens Kooken, BSc
Master student in Medicine
Department of Intensive Care, Radboud Institute for Health Science, Radboud University Medical Center, United States
Disclosure information not submitted.
John Devlin, BCCCP, PharmD, MCCM
Professor
Northeastern University
Boston, Massachusetts, United States
Disclosure information not submitted.
Mark Van Den Boogaard, RN, PhD
Assistant professor
Department of Intensive Care, Radboud Institute for Health Science, Radboud University Medical Center, United States
Disclosure information not submitted.
Title: Pre-ICU anxiety and depression and the risk of delirium in critically ill adults
Introduction: Anxiety (A) and depression (D) may increase postoperative delirium risk in non-ICU patients but the association between A/D and delirium in critically ill adults remains unclear. We hypothesized pre-ICU A/D is a risk for greater ICU delirium given A, D and delirium share common pathophysiologic pathways.
Methods: This secondary, IRB-approved, subgroup analysis of the MONITOR-IC study included adults admitted to the 36-bed mixed ICU at RadboudUMC (Nijmegen, NL) from 7/2016-2/2020. Patients admitted after elective surgery, not expected to survive ≥48 hr, or who did not undergo admission Hospital Anxiety and Depression Scale (HADS) evaluation were excluded. A score ≥8/21 on the HADS-A and HADS-D subscales denoted baseline A and D, respectively. All patients without coma (RASS=-4/-5) were evaluated for delirium using the CAM-ICU q8h. A delirium day was defined by ≥1 +CAM-ICU and/or scheduled antipsychotic use. Multivariable logistic regression models adjusting for ICU days prior to delirium or discharge and 9 risk variables [Pre-ICU:age, Charlson Comorbidity, cognitive impairment; ICU baseline: APACHE-IV, admission type (medical vs. urgent surgery); Daily ICU (until delirium or discharge): opioid (≥10 mg IV MME/day) or benzodiazepine (≥5 mg IV midazolam/day) use, mean SOFAscore, and coma] were included to study the association between pre-ICU A+D (or A, D alone) and ICU delirium occurrence and duration.
Results: Among 991 patients, 145(14.6%) had both A+D, 78(7.9%) had A, 91(9.2%) had D, and 677(68.3%) had neither. Delirium occurred in 406/991(41.0%) patients for 2(1-5) days. Among the A+D group, delirium occurred in 78/145(53.8%) for 3(1-5) days; in the A group in 37/77(47.4%) for 2(1-5) days, in the D group in 39/91(42.9%) for 2(1-5) days, and in the neither A/D group in 252/677(37.2%) for 2(1-5) days. Presence of AD at ICU baseline was associated with significantly increased delirium (aOR 1.92, 95% CI 1.06-3.41, P=0.03) but was not associated with ICU delirium days (aRR 0.97, 95% CI 0.73-1.26, P=0.80). Presence of A or D alone were not associated with either greater ICU delirium or duration. Adjustment for pre-ICU benzodiazepine and antidepressant use did not change the results.
Conclusions: Presence of both A+D before ICU admission may increase delirium risk. Further prospective research is warranted.