Subsyndromal Delirium and Institutionalization Among Patients With Critical Illness

Abstract

Background: The prognostic importance of subsyndromal delirium is unknown.

Objective: To test whether duration of subsyndromal delirium is independently associated with institutionalization.

Methods: The Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) was used twice daily to assess for subsyndromal delirium in patients with respiratory failure or shock. Delirium was considered present if the assessment was positive. Subsyndromal delirium was considered present if the assessment was negative but the patient exhibited any CAM-ICU features. Multivariable regression was used to determine the association between duration of subsyndromal delirium and institutionalization, adjusting for age, education, baseline cognition and disability, comorbidities, severity of illness, delirium, coma, sepsis, and doses of sedatives and opiates.

Results: Subsyndromal delirium, lasting a median of 3 days, developed in 702 of 821 patients (86%). After adjusting for covariates, duration of subsyndromal delirium was an independent predictor of increased odds of institutionalization (P = .007). This association was greatest in patients with less delirium (P for interaction = .01). Specifically, of patients who were never delirious, those with 5 days of subsyndromal delirium (upper interquartile range [IQR]) were 4.2 times more likely to be institutionalized than those with 1.5 days of subsyndromal delirium (lower IQR).

Conclusions: Subsyndromal delirium occurred in most critically ill patients, and its duration was an independent predictor of institutionalization. Routine monitoring of all delirium symptoms may enable detection of full and subsyndromal forms of delirium.

Conflict of interest statement

Conflicts of interest:

The remaining authors report no financial conflicts of interest.

©2017 American Association of Critical-Care Nurses.

Figures

Figure 1. Assessing Subsyndromal Delirium Using the Confusion Assessment Method for the ICU (CAM-ICU) framework

We determined the presence of subsyndromal delirium (SSD) using the Confusion Assessment Method for the ICU (CAM-ICU) assessed twice each day while patients were in the ICU and once per day thereafter. At each assessment each of the 4 CAM-ICU features were assessed. According to the CAM-ICU framework, if a patient demonstrated an acute change in mental status or a fluctuating course of mental status (Feature 1) and inattention (Feature 2) plus either Altered Consciousness (Feature 3) or Disorganized Thinking (Feature 4), the CAM-ICU was considered positive and delirium was present (dark blue box). If the CAM-ICU was negative, but the patient exhibited any of the CAM-ICU features, subsyndromal delirium was considered present (light blue box). If no CAM-ICU features were present, was the patient considered to have normal mental status (white box).

Figure 2. The relationship between the duration of subsyndromal delirium and odds of institutionalization stratified by the duration of delirium

The overall association between duration of subsyndromal delirium (SSD) and adjusted probability of institutionalization was significant (p=0.007). We hypothesized a priori that this association would be modified by the delirium duration and found this interaction to be significant (p=0.01), indicating a greater independent risk of disability from SSD in patients with shorter durations of overt delirium. Therefore, we present the association stratified by delirium duration. The panels display the step-wise reduction in the strength of the association among patients with 0 days of delirium (no delirium), 1 day of delirium (25th percentile), 2 days of delirium (50th percentile) and 6 days of delirium (75th percentile), respectively, with all other covariates adjusted to their respective median or mode. Black lines represent the independent association and shading represents the 95% confidence intervals. The rug plots illustrate the distribution of patients with a given duration of SSD.