The postoperative respiratory and analgesic effects of dexmedetomidine and morphine have not been compared in children with sleep apnoea having adenotonsillectomy. In a randomised double-blind study we recruited 60 children, aged 2–13 years, who received either intravenous dexmedetomidine 1 μg.kg−1 or morphine 100 μg.kg−1 on anaesthetic induction. End-tidal carbon dioxide, Children’s Hospital of Eastern Ontario Pain Scale score and supplementary morphine administration were recorded every 15 min for 60 min postoperatively. Over 60 min, mean (SD) end-tidal carbon dioxide was consistently lower with dexmedetomidine compared with morphine (5.4 (0.7) kPa vs 6.0 (0.6) kPa, respectively; p = 0.001). Mean (SD) pain scores were higher with dexmedetomidine (8.1 (2.0) immediately postoperatively and 6.7 (1.0) at 60 min vs 7.6 (1.8) and 6.3 (0.7), respectively, with morphine (p = 0.023)). More patients required supplementary morphine with dexmedetomidine (13/30 (43%) vs 21/30 (70%); p = 0.037). Postoperatively, dexmedetomidine produced less respiratory depression than morphine, but less effective analgesia.
Adenotonsillectomy is a common surgical procedure in the paediatric population . The main indication for the procedure is obstructive sleep apnoea syndrome . The most common serious postoperative complication following this procedure is desaturation and airway obstruction . The surgery can cause moderate to severe pain . Opioids such as morphine have been widely used in such procedures. However, opioids may depress respiration and contribute to airway obstruction after adenotonsillcetomy for obstructive sleep apnoea syndrome. Since repeated hypoxaemia can increase a patient’s sensitivity to morphine , the risks of respiratory complications in these patients may increase even with a regular dose.
Dexmedetomidine is a highly selective α2-adrenoceptor agonist, and increasingly used because of its combined sedative, analgesic and anti-sympathetic effects [6, 7]. As dexmedetomidine has an analgesic effect with little respiratory depression , we decided to explore whether or not dexmedetomidine could be used as a substitute for opioids in this population, potentially reducing postoperative respiratory depression. We hypothesised that the postoperative end-tidal carbon dioxide of children receiving dexmedetomidine would be lower than that of children receiving morphine.
With Institutional Review Board approval and written, informed consent from each child’s parent or guardian, we recruited 60 children, aged 2–13 years, of ASA physical status 1–2, scheduled to undergo adenotonsillectomy due to obstructive sleep apnoea. The diagnosis of sleep apnoea was made by the ear, nose and throat surgeon. A randomisation code (generated by Microsoft Excel 2003) was used to assign patients to the two groups. The allocation sequence was placed in sealed envelopes with an envelope for each child opened before each anaesthetic induction. Children were not studied if they had craniofacial abnormalities, previous upper airway trauma, failure to thrive, cor pulmonale, asthma or a recent history of upper respiratory tract infection.
Treatment drugs were prepared for each patient by an assistant who was not involved in the management of anaesthesia or postoperative care. For the dexmedetomidine group, 5 ml saline 0.9% was prepared in a 5-ml syringe, and dexmedetomidine 1 μg.kg−1 was diluted to 20 ml with saline in a 20-ml syringe. For the morphine group, morphine 100 μg.kg−1 was diluted to 5 ml with saline, and 20 ml saline was prepared in another 20 ml syringe. Effective blinding was ensured by giving each patient in each group both of their syringes.
All children received paracetamol 15 mg.kg−1 orally 30 min before surgery, and came into the operating room with a peripheral intravenous catheter in place. After standard monitoring was established, anaesthesia was induced with propofol 2 mg.kg−1 intravenously, followed by 8% sevoflurane in 100% oxygen, and respiration was assisted manually as necessary. Other intravenous agents were administrated subsequently: ondansetron 0.1 mg.kg−1; dexamethasone 0.5 mg.kg−1 (maximum 10 mg); the study drug in the 5-ml syringe, and the study drug in the 20-ml syringe (each injected slowly over a 10-min period). Vecuronium 0.1 mg.kg−1 was used to facilitate intubation of the patient’s trachea 4 min after the commencement of induction. Anaesthesia was maintained with sevoflurane and nitrous oxide in oxygen. End-tidal concentrations of sevoflurane and nitrous oxide were maintained at 1.5–2% and 70%, respectively, in accordance with intra-operative haemodynamic changes. Inhalation anaesthesia was discontinued at the end of surgery and neostigmine 0.02 mg.kg−1 and atropine 0.02 mg.kg−1 were administered to reverse residual neuromuscular blockade. The patient’s trachea was extubated when they were awake as defined by eye opening, purposeful movement or response to command. The time to extubation (defined as time from end of surgery to time to extubation of the patient’s trachea), intra-operative heart rate, arterial blood pressure and oxygen saturation were recorded by an anaesthetist who was blinded to patient assignment.
Patients were taken to the recovery room immediately after extubation of their trachea. All children received oxygen via a mask (2 min.l−1). End-tidal carbon dioxide (the values of the plateau phase were measured by inserting a catheter into the nasal cavity), Children’s Hospital of Eastern Ontario Pain Scale (CHEOPS) score , Ramsay sedation score, heart rate, arterial blood pressure and oxygen saturation were recorded by a nurse who was unaware of patient assignment in recovery room, on arrival and at 15, 30, 45 and 60 min after arrival. The CHEOPS is a behavioural scale for evaluating postoperative pain in young children. This tool incorporates six categories of behaviour that are each scored individually (range of 0–2 or 1–3) and then totalled for a pain score ranging from 4 to 13. Patients with a CHEOPS score > 8 received intravenous morphine 10 μg.kg−1 at 5-min intervals until the score was < 8 (maximum 50 μg.kg−1). The number of doses of supplementary morphine and the time of administration were recorded. If a patient vomited in the recovery room, ondansetron 0.1 mg.kg−1 was administered intravenously.
According to Lerman’s approach , sample size was calculated based on our own pilot study that involved 20 patients. To detect a mean (SD) difference in end-tidal carbon dioxide of 0.55 (0.74) kPa with an alpha error of 0.05 and power of 80%, 30 patients were required in each group. Data were analysed using spss software (version 16; Chicago, IL, USA), and were presented as number (n) or percentage (%) or mean (SD) as appropriate. Two-way repeated measures ANOVA was used for end-tidal carbon dioxide, CHEOPS score, Ramsey sedation score, heart rate, arterial blood pressure and oxygen saturation. The chi-squared test was used for comparison of vomiting and the number of doses of supplementary morphine. Supplementary morphine use was compared by survival analysis (Kaplan–Meier). A value of p = 0.05 or p < 0.05 was considered statistically significant.
Baseline data are presented in Table 1. Frequency of postoperative vomiting was low in both groups and the difference between groups was not significant (Table 1). End-tidal carbon dioxide was consistently lower in the dexmedetomidine group than in the morphine group throughout the period of observation (mean (SD) 5.4 (0.7) vs 6.0 (0.6) kPa, respectively; p = 0.001) (Fig. 1).
Table 1. Characteristics and basic clinical data of patients receiving dexmedetomidine or morphine. Values are mean (SD) or number.
Dexmedetomidine (n =30)
Duration of surgery; min
Time to extubation of the trachea; min
Doses of Supplementary morphine
The CHEOPS score was significantly higher in the dexmedetomidine group than in the morphine group (Fig. 2). A survival analysis of the requirement for supplementary morphine (Fig. 3) revealed a significant difference between groups. A greater percentage of patients did not require morphine in the morphine group (70%, 21/30) than in the dexmedetomidine group (43%, 13/30) (p = 0.037). The actual number of supplementary morphine doses given to those who needed additional morphine was not different between groups (Table 1).
There was no difference in the Ramsey sedation score between the two groups (Fig. 4). Figure 5 shows that the heart rate was lower in the dexmedetomidine group compared with the morphine group, but that there was no difference of mean blood pressure. There was no difference in postoperative oxygen saturation between the two groups, saturation being between 96% and 100% throughout for the morphine group and 95–100% for the dexmedetomidine group (p = 0.85).
In this study, end-tidal carbon dioxide was lower in the dexmedetomidine group than that in the morphine group, even after the administration of supplementary morphine. These data suggest that the respiratory depression effect of intra-operative dexmedetomidine is less than that with morphine. However, more patients receiving dexmedetomidine had poorer analgesia than those receiving morphine, something that may have increased their respiratory drive. Obstructive sleep apnoea syndrome is, however, an important risk factor for peri-operative respiratory complications in children undergoing adenotonsillectomy. Rosen et al.  reported that 27% of patients with obstructive sleep apnoea syndrome experienced severe hypoxia and respiratory depression after tonsillectomy. Consequently, dexmedetomidine may be a useful clinical tool in this situation.
Although fewer patients required analgesia in the morphine group, 43% of the patients of patients receiving dexmedetomidine did not require supplementary morphine within 1 h of surgery. Dexmedetomidine does have some analgesic efficiency. The CHEOPS score in the dexmedetomidine group was slightly higher than that in the morphine group, but importantly, it decreased gradually with time and with the administration of supplementary morphine. When the patients were discharged, the CHEOPS score was similar to that in the morphine group (means 6.8 for dexmedetomidine group and 6.4 for the morphine group, with four to six points typically meaning no pain on the CHEOPS scale). Our results are consistent with those of a recent study , where total postoperative supplementary morphine requirement was similar when dexmedetomidine 1 μg.kg−1 or morphine 100 μg.kg−1 was used intra-operatively.
Recently, Patel et al.  have reported significantly fewer patients requiring postoperative supplementary morphine after intra-operative dexmedetomidine 2 μg.kg−1 compared with fentanyl 1 μg.kg−1 in children undergoing adenotonsillectomy. Our results are contrary to this and this can probably be explained by the different doses and opioids used in the two studies. Fentanyl is a shorter acting opioid than morphine and may have provided worse postoperative analgesia at the dose selected.
Dexmedetomidine decreases heart rate and blood pressure in a dose-dependent fashion . It activates inhibitory neurons in the medullary vasomotor centre, inhibits sympathetic activity and reduces plasma catecholamine concentration [15–17]. Our study demonstrated a decrease in heart rate in the first hour after surgery, something that may be of concern to paediatric anaesthetists. There was, however, no difference in blood pressure between groups. None of the patients in the two groups developed bradycardia or hypotension sufficient enough to require intervention.
We found that intra-operative dexmedetomidine at a dose of 1 μg.kg−1 was associated with less postoperative respiratory depression than morphine 100 μg.kg−1 in children undergoing adenotonsillectomy, though it gave less effective postoperative analgesia. In our study, however, analgesia was easily managed by supplemental morphine which did not counteract any beneficial effect on respiratory function. Dexmedetomidine could be a useful analgesic option for children with sleep apnoea undergoing adenotonsillectomy.
No external funding or competing interests declared.