Highlights
Delusional Memories and PTSD Risk
Delusional memories are independently linked to a significant increase in the prevalence of post-traumatic stress disorder (PTSD) symptoms, rising from 6.7% at three months to 18.1% at twelve months post-ICU.
Clinical Determinants
Female sex and the duration of deep sedation are primary predictors of complete ICU amnesia, while delirium and the length of ICU stay are significantly associated with the development of delusional memories.
The Trajectory of Psychological Morbidity
Unlike factual memories or amnesia, delusional memories drive a worsening trajectory of psychological distress over the first year of recovery, suggesting that memory disturbances are not merely acute symptoms but long-term pathogenic factors.
The Psychological Burden of Critical Care
As mortality rates in intensive care units (ICUs) have declined due to advances in organ support and precision medicine, the focus of clinical research has shifted toward the quality of survival. Post-Intensive Care Syndrome (PICS) encompasses a spectrum of physical, cognitive, and mental health impairments that persist long after hospital discharge. Among these, psychological morbidity—specifically post-traumatic stress disorder (PTSD)—remains a significant barrier to functional recovery.
One of the most complex aspects of the ICU experience is how patients remember it. ICU memories are generally categorized into factual memories (e.g., voices of family, procedures), amnesia (no recollection of the ICU stay), and delusional memories (e.g., hallucinations, nightmares, or paranoid ideations). While it was once hypothesized that amnesia might protect patients from trauma, emerging evidence suggests that the nature of these memories dictates the long-term psychiatric trajectory. The study by Kooken et al. (2025) provides a robust, multicenter longitudinal analysis of these memory determinants and their causal relationship with PTSD symptoms over a one-year period.
Methodological Framework: A Multicenter Longitudinal Analysis
The study utilized a prospective cohort design involving 426 adult patients from two Dutch University ICUs. To capture the nuance of the patient experience, researchers employed the validated ICU-Memory Tool (ICU-MT) during structured telephone interviews three months after ICU discharge. This tool allowed for the differentiation between factual recollections and delusional constructs.
To assess the impact of these memories on mental health, the Impact of Event Scale (IES-6) was administered at two critical time points: 3 months and 12 months post-ICU. This longitudinal approach is essential for understanding whether psychological symptoms resolve spontaneously or worsen over time. The researchers employed multinomial logistic regression to identify clinical determinants of memory types and linear mixed-effects models to adjust for confounders when evaluating the trajectory of PTSD symptoms.
Characterizing ICU Memories: Factual, Amnesic, and Delusional
The distribution of memory types among the cohort was revealing. Nearly half of the patients (47.7%) retained factual memories without delusional interference. Complete amnesia was reported by 13.8%, while 38.5% experienced delusional memories. Notably, among those with delusional memories, 41% had documented clinical delirium during their ICU stay, leaving a significant portion of patients who experienced delusions without a formal diagnosis of delirium.
Determinants of Memory Formation: From Sex to Sedation
Understanding why some patients remember and others do not is vital for preventative care. The study identified several key determinants:
Predictors of ICU Amnesia
Female sex was associated with nearly double the odds of complete ICU amnesia (adjusted Odds Ratio [aOR] 1.99, 95% CI 1.04-3.81). Furthermore, the duration of deep sedation played a dose-dependent role; each additional day of deep sedation increased the likelihood of amnesia (aOR 1.34, 95% CI 1.09-1.65). This suggests that pharmacological interventions intended to provide comfort may inadvertently create a ‘black hole’ in the patient’s narrative of their illness.
Predictors of Delusional Memories
Delirium was a potent predictor of delusional memories (aOR 1.94, 95% CI 1.04-3.61). Additionally, the length of ICU stay (aOR 1.11 per day) independently contributed to the formation of these false memories. The environment of the ICU—characterized by sleep deprivation, constant noise, and lack of natural light—likely synergizes with clinical delirium to weave complex, often terrifying, delusional narratives.
The Trajectory of PTSD: Why Delusional Memories Matter
One of the most striking findings of the research was the evolution of PTSD symptoms over time. In the overall cohort, the prevalence of significant PTSD symptoms increased from 4.5% at 3 months to 10.0% at 12 months. This doubling of symptomatic patients was driven almost entirely by the group with delusional memories.
Fig. 2. Demographic and clinical determinants of different ICU memory types (multinomial logistic model).
Fig3. Prevalence of PTSD symptoms per timepoint and memory group.
Specifically, patients with delusional memories saw their PTSD symptom prevalence jump from 6.7% to 18.1% between the two time points. In contrast, those with only factual memories or amnesia remained relatively stable. The linear mixed-effects model confirmed that delusional memories were independently linked to higher IES scores at both 3 months (12.9% mean difference vs. factuals) and 12 months (18.1% mean difference vs. factuals). These data suggest that delusional memories do not fade into the background; rather, they may serve as a constant source of psychological re-traumatization.
Fig 4. The relationship between different ICU memories and the development and trajectory of PTSD symptoms (linear mixed-effects model).
Expert Commentary: The Sedation Paradox and Clinical Implications
These findings challenge the traditional ‘protective’ view of deep sedation. While sedation is necessary for safety and comfort in many cases, its association with amnesia and the subsequent risk of delusional formation suggests a ‘sedation paradox.’ By wiping away factual anchors of the ICU experience, heavy sedation may leave the brain to fill the gaps with hallucinations and paranoid constructions, which later manifest as PTSD.
Clinicians should consider the following takeaways:
1. Targeted Screening: Patients who experience delirium or long ICU stays should be flagged for early psychological follow-up, as they are at high risk for delusional memories.
2. Sedation Stewardship: Minimizing deep sedation and prioritizing ‘light’ sedation protocols (when clinically feasible) may help preserve factual memories, which appear to act as a protective factor against PTSD.
3. ICU Diaries: Although not directly studied here, the use of ICU diaries—where nurses and family members record daily events—could help bridge the gap for amnesic patients and provide a factual reality to counter delusional memories.
Limitations of the study include the potential for attrition bias (only 250 patients completed both follow-ups) and the reliance on self-reported scales rather than clinical psychiatric interviews. However, the multicenter nature and the use of validated tools provide high internal and external validity for these results.
Summary and Future Directions
Kooken et al. have demonstrated that the psychological aftermath of the ICU is profoundly influenced by the nature of the patient’s memory. Delusional memories are not merely transient side effects of critical illness; they are independent drivers of long-term PTSD. By identifying the determinants of these memories—such as sedation duration and delirium—clinicians have a roadmap for intervention. Future research should focus on whether early ‘reality-testing’ interventions or structured memory-reconstruction therapy can mitigate the toxic effects of delusional memories and alter the trajectory of post-ICU mental health.
References
Kooken RWJ, Tilburgs B, Slooter AJC, van den Boogaard M. Determinants of ICU memories and the impact on the development and trajectory of post-traumatic stress symptoms: a multicenter longitudinal cohort study. Intensive Care Med. 2025 Nov;51(11):2021-2030. doi: 10.1007/s00134-025-08132-4 IF: 21.2 Q1 .