The trajectory of sleep after critical illness: a 24-month follow-up study

Abstract

Background: Survivors of critical illness endure long-lasting physical and mental challenges. Despite the persistence of poor sleep quality in a considerable proportion of patients at the 12-month follow-up, studies with assessments exceeding this period are limited. We aimed to investigate the trajectory of sleep over the 24 months following critical illness.

Methods: Observational, prospective study. Patients diagnosed with SARS-CoV-2 infection were recruited during the intensive care unit stay. Evaluations of sleep (Pittsburgh Sleep Quality Index [PSQI]), mental health (Hospital Anxiety and Depression Scale [HADS]), quality of life (12-item Short Form Survey [SF-12]), and other factors were performed in the short-term, and at 12 and 24 months after hospital discharge. Good sleep quality was defined as a PSQI score of ≤ 5. Minimal clinically important improvement (MCII) was defined as a decrease of ≥ 4 points in the PSQI score between the short-term assessment and the 24-month follow-up.

Results: The cohort included 196 patients (69.9% males), with a median [p₂₅;p₇₅] age of 62.0 [53.0;67.2] years. The global population showed a mean (95% CI) change of - 0.91 ( - 1.50 to - 0.31) points in the PSQI score from the short-term assessment to the 24-month follow-up. Based on PSQI score trajectories, three distinct groups of patients were identified: (i) the healthy group, consisting of patients with good sleep quality in the short-term that was maintained throughout the follow-up period; (ii) the MCII group, consisting of patients with poor sleep quality in the short-term, but with improvement over time, ultimately reaching levels comparable to the healthy group; (iii) the non-MCII group, consisting of those with consistently poor sleep quality across the entire follow-up. Further analyses revealed that PSQI score trajectories were closely aligned with those of the HADS and SF-12 mental scores.

Conclusions: Our findings reveal that a subset of critical illness survivors requires up to 24 months after the acute phase to fully restore their sleep quality, while a significant proportion does not experience a clinically significant improvement in sleep quality over this period. These distinct sleep trajectories are strongly correlated with mental health status, highlighting the importance of addressing sleep alongside mental health within the framework of post-intensive care syndrome.

Keywords: Critical survivors; Mental health; Pittsburgh sleep quality index; SARS-CoV-2; Sleep.

Fig. 1

Trajectories of sleep. Data are represented as least square means (± SEM). The p-value threshold defining statistical significance was set at less than 0.05. ESS, Epworth Sleepiness Scale; MCII, minimal clinically important improvement; PSQI, Pittsburgh Sleep Quality Index; SATED, Satisfaction Alertness Timing Efficiency Duration; SEM, standard error of the mean

Fig. 2

Trajectories of other sequelae based on the observed sleep trajectories. Data are represented as least square means (± SEM). The p-value threshold defining statistical significance was set at less than 0.05. HADS, Hospital Anxiety and Depression Scale; MCII, minimal clinically important improvement; SEM, standard error of the mean; SF-12, 12-item Short Form Survey