Li · BMC anesthesiology 2023 · prospective cohort study · n=76

Circadian rhythm disturbance and delirium in ICU patients: a prospective cohort study.

Cited 27 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective observational cohort study

PubMed 37312021 · doi:10.1186/s12871-023-02163-4 · record verified 2026-08-27

What was done

A prospective cohort study enrolled 76 post-surgical patients admitted to a surgical ICU of a tertiary teaching hospital who were conscious and expected to stay longer than 24 hours. Serum melatonin and plasma cortisol levels were measured three times daily via arterial blood sampling during the first three post-admission days. Sleep quality was evaluated daily using the Richard-Campbell Sleep Questionnaire (RCSQ), and delirium was assessed twice daily with the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU).

What was found

Seventeen of 76 patients developed delirium. Melatonin levels significantly differed between delirium and non-delirium groups at 08:00 on day 1 (p = 0.048), at 03:00 (p = 0.002) and 08:00 (p = 0.009) on day 2, and at all three daily sampling points on day 3 (p = 0.032, 0.014, and 0.047). Plasma cortisol was significantly lower in delirium patients at 16:00 on day 1 (p = 0.025). Non-delirious patients maintained significant circadian secretion rhythms (melatonin p < 0.001, cortisol p = 0.026), whereas delirious patients lacked significant rhythmicity (melatonin p = 0.064, cortisol p = 0.454). RCSQ sleep scores did not differ significantly between groups across the first three days.

Why it matters

This study provides prospective evidence directly linking circadian endocrine disruption (melatonin and cortisol) to the development of delirium in surgical critical care patients.

Limits

The study is limited by a small sample size (76 patients, 17 delirium cases) and a single-center surgical population. Blood sampling three times daily provides low temporal resolution for circadian curves, subjective sleep scoring was used instead of objective polysomnography, and the observational design cannot establish causality.

Cited by