Depression predicts long-term cognitive impairment in survivors of critical illness

J Trauma Acute Care Surg. 2021 Jan 1;90(1):79-86. doi: 10.1097/TA.0000000000002955.


Introduction: Intensive care unit (ICU) survivorship is associated with long-term cognitive impairment (LTCI). Our work has found post-ICU depression in up to 30% and posttraumatic stress disorder (PTSD) in up to 10% of ICU survivors. We hypothesized that post-ICU depression and PTSD are independently associated with LTCI in ICU survivors.

Methods: This is a five-center nested prospective cohort of critically ill patients admitted to medical and surgical ICUs who underwent neuropsychological assessments at 3 and 12 months posthospital discharge. Our primary outcome was global cognition using the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) and Trail Making Test, Part B, a test of executive functioning, at 3- and 12-month follow-up. Our independent variables were Beck Depression Inventory II and the PTSD Checklist-Specific Version measured at 3 and 12 months. We performed multivariable linear regression models controlling for covariates such as age, years of education, preexisting cognitive impairment, comorbidities, ventilator days, hypoxemia episodes, and days of delirium or coma.

Results: Of 1,047 patients in the combined cohort, 679 were alive and available for follow-up at 3 months. A total of 590 (87%) ICU survivors completed at least one 3-month assessment, and of the 554 who survived to 12 months, 519 (94%) completed both a 3- and 12-month assessment with a median age of 61 years (52-70 years) and mean daily Sequential Organ Failure Assessment score of 6 (4-8), 520 (88%) were mechanically ventilated, and 420 (71%) were with delirium. Of these, 113 (19%) had PTSD and 187 (32%) had depression at 3 months with similar rates at 12 months. Depression at 3 months was associated with lower 3-month RBANS (coefficient, -2.25; -3.10 to -1.39) and lower Trails B scores at both 3 months (odds ratio, 0.69; 0.56-0.85) and 12 months (odds ratio, 0.66; 0.52-0.84). Posttraumatic stress disorder at 3 months had no association with RBANS or Trails B scores at 3 or 12 months.

Conclusion: Early post-ICU depression, but not PTSD, is independently associated with coexisting LTCI, even when controlling for past ICU delirium. Treatment for early depression represents a novel intervention area for LTCI prevention in ICU survivors.

Level of evidence: Prognostic/epidemiological, level III.

Publication types

  • Multicenter Study
  • Research Support, N.I.H., Extramural
  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Aged
  • Cognitive Dysfunction / etiology*
  • Critical Illness / psychology*
  • Depression / complications*
  • Depression / etiology
  • Female
  • Humans
  • Intensive Care Units / statistics & numerical data
  • Male
  • Middle Aged
  • Neuropsychological Tests
  • Prospective Studies
  • Stress Disorders, Post-Traumatic / complications
  • Stress Disorders, Post-Traumatic / etiology
  • Survivors / psychology*
  • Survivors / statistics & numerical data