In Reply

We agree with Dr. Goucher et al. that low-dose dexmedetomidine infusion did not restore the normal sleep architecture because stage 3 non–rapid eye movement sleep and rapid eye movement (REM) sleep remained significantly decreased or absent in our patients.¹  This is also the case when dexmedetomidine was administered for sedation in mechanically ventilated patients.^2,3  It should be noted that the target subjects were patients in the intensive care unit (ICU) after major surgery in our study¹  or receiving mechanical ventilation in another previous study.³  It is well known that significant sleep disturbances such as fragmented sleep, decreased sleep efficiency, increased stage 1 non-REM sleep, and decreased or absent stage 3 non-REM and REM sleep are often present in those patients. Dexmedetomidine partially improved “sleep architecture” through increasing the percentage of stage 2 non-REM sleep (and decreasing the percentage of stage 1 non-REM sleep), a unique property that has also been demonstrated in other clinical studies previously.⁴ 

Considering the importance of sleep for ICU recovery and the lack of effective pharmacologic interventions to improve sleep,⁵  prophylactic low-dose dexmedetomidine may be a choice, although not the best. Clinical effectiveness was demonstrated in our previous trial in 700 patients admitted to the ICU after noncardiac surgery, in which low-dose dexmedetomidine infusion reduced the incidence of delirium during the first 7 days after surgery (9% compared with 23% with placebo) and also decreased the incidence of nondelirium complications and increased early hospital discharge.⁶  We cannot establish a causal relationship between delirium reduction and improved sleep in patients from these two independent trials.^1,6 

It is true that hemodynamic disturbances are major concerns when using dexmedetomidine in ICU patients. Indeed, dexmedetomidine at such a low dose (0.1 µg kg⁻¹ h⁻¹) slightly increased the occurrence of hypotension, although not statistically significantly,⁶  indicating that close monitoring is necessary whenever dexmedetomidine is on-board. Last but not least, whether delirium prevention by dexmedetomidine ultimately improves patients’ long-term outcome remains unknown and warrants further study.

Regarding the question of Dr. Reade, herein we confirm that these two studies^1,6  are completely independent trials in which each has an individual registration (Chinese Clinical Trial Registry [Chengdu, Sichuan, China; www.chictr.org.cn] Nos. ChiCTR-TRC-10000802 and ChiCTR-TRC-12002567). Patients who were recruited in one study were not enrolled the other one.