Electroencephalography and Sleep Quality with Lormetazepam in the Intensive Care Unit
CLIO-ICU, A single-centre prospective pilot study
In Data Analysis
Sleep is not a luxury — it is a biological necessity that supports immune defence, metabolism, memory and the repair of both body and brain. Under healthy conditions, our sleep follows a roughly 24-hour pattern (the circadian rhythm), cycling through lighter and deeper stages in a predictable, well-organised structure that scientists call sleep architecture.
For patients who are critically ill and mechanically ventilated in an intensive care unit (ICU), this delicate structure largely collapses. Continuous light and noise, round-the-clock monitoring and nursing interventions, pain, the underlying illness itself and the medications used to treat it all conspire to fragment sleep and to blur the normal boundary between day and night. Recordings of brain activity in these patients frequently reveal sleep that is severely fragmented, abnormally shallow, or so altered that it no longer resembles normal sleep at all.
From disturbed sleep to a disturbed brain
This matters because disrupted sleep and a broken circadian rhythm are closely linked to delirium — an acute state of confusion, disorientation and fluctuating attention that affects a large proportion of ventilated ICU patients. Delirium is far more than a passing episode of bewilderment. It is associated with longer time on the ventilator, prolonged ICU and hospital stays, higher mortality, and lasting cognitive impairment that can persist for months or even years after discharge, as part of what is known as the post-intensive care syndrome.
In other words, what happens to the brain during critical illness can shape a patient's life long after the acute danger has passed.
Protecting sleep and circadian rhythm is therefore increasingly viewed not merely as a comfort measure, but as a potential lever to improve both short- and long-term recovery.
The benzodiazepine controversy
Among the medications used to keep ventilated patients calm and comfortable, benzodiazepines — a long-established class of sedative and anti-anxiety drugs — have acquired a particularly poor reputation. Numerous studies and current guidelines link their use to a higher risk of delirium and worse outcomes, and many ICUs now avoid them wherever possible.
A more nuanced picture, however, is emerging from the scientific literature. There are plausible data to suggest that the problem may lie less in the molecules themselves than in the way they have traditionally been administered: in excessive doses that drive patients into unnecessarily deep sedation. Deep, prolonged sedation — irrespective of which agent produces it — is itself a well-recognised risk factor for delirium and poor recovery. From this perspective, the decisive question is not simply whether a benzodiazepine is used, but how much, and to what depth of sedation a patient is taken.
A different approach: lormetazepam
Lormetazepam is a benzodiazepine that has recently become available as an intravenous preparation for use in intensive care — at present approved only in Germany and Austria. Early case reports are encouraging: they indicate that lormetazepam can provide very effective relief of anxiety (anxiolysis) and keep patients comfortable, without pushing them into the deep levels of sedation that may themselves be harmful.
This raises a question that has not yet been answered: how does sleep actually behave under lormetazepam? If a sedative can calm a patient without abolishing or distorting natural sleep, it could represent a meaningfully different approach to comfort and safety in the ICU.
“Sleep is strange—“[s]tranger than habit and than obsession,”
Our study
Our research group is conducting a prospective observational study to address precisely this question. We are investigating how a lormetazepam-based, symptom-oriented sedation strategy — that is, treatment guided by each patient's actual individual needs rather than by a rigid fixed protocol — affects the sleep architecture of mechanically ventilated intensive care patients.
Sleep is measured directly through continuous recordings of brain activity, which allow us to characterise not only how much patients sleep, but also the quality and internal structure of that sleep across the full day–night cycle. We plan to enrol a total of 50 patients, with recordings of up to three 24-hour periods per patient.
A special focus: pathological sleep patterns
Beyond the normal sleep stages, we pay particular attention to abnormal patterns that are increasingly recognised in the critically ill. Healthy deep sleep — known as N3 or slow-wave sleep — is the most restorative stage and produces a characteristic signature on brain-activity recordings. In severely ill patients, however, the brain can generate altered or distorted versions of these patterns. One such phenomenon, sometimes termed "atypical N3 sleep," superficially resembles deep sleep but deviates from its normal electrophysiological features — and its appearance is suspected of being a particularly unfavourable prognostic sign.
By systematically searching for these pathological patterns, this pilot study aims to provide a first detailed picture of what sleep under lormetazepam genuinely looks like — and to begin to clarify whether this newer approach to sedation can help to protect, rather than further disturb, the fragile rest of the critically ill brain.
