Impact of Dexmedetomidine Supplemented Analgesia on Delirium in Patients Recovering From Orthopedic Surgery

A Randomized Controlled Trial

Hong Hong; Da-Zhi Zhang; Mo Li; Geng Wang; Sai-Nan Zhu; Yue Zhang; Dong-Xin Wang; Daniel I. Sessler


BMC Anesthesiol. 2021;21(223) 

In This Article


Results of this blinded randomized trial showed that, in elderly patients following major orthopedic surgery, dexmedetomidine supplemented intravenous analgesia did not reduce delirium within 5 days; however, it improved analgesia and subjective sleep quality without increasing adverse events.

Postoperative pain, opioids, and sleep disruption each potentially contribute to delirium. Patient-controlled sufentanil combined with dexmedetomidine provided better analgesia than sufentanil alone which is consistent with previous reports,[12] although opioid use was similar in each group. Dexmedetomidine also modestly improved subjective sleep quality, again consistent with previous reports.[13,30] However, although dexmedetomidine supplemented analgesia reduced the risk of delirium by about a third, the confidence interval is wide with a potential reduction up to 64% but also with a potential increase of up to 18% and the difference is not statistically significant. The effect of dexmedetomidine supplemented analgesia in preventing delirium deserves further study in high-risk patients.

In the present study, postoperative delirium developed in 7.3% of patients in the placebo group. This incidence was lower than we expected and less than described in some reports,[1–3,5,28] but within some recently described estimates which range from 2.2 to 10.5%.[32,33] There are several factors potentially contributing to the relatively low delirium incidence in our patients. With the exception of the hip fracture patients (3.4% of the trial population),[33] most of our patients did not require ICU admission and were at relatively low risk for postoperative delirium, especially as many had arthroplasty,[1] and 68% of the trial population had regional rather than general anesthesia.[28,30] As expected from previous reports and our own experience,[28,30] delirium was most common on the initial postoperative day.

The largest previous trial of dexmedetomidine supplemented analgesia for prevention of delirium was by Sun and colleagues who randomized 557 non-cardiac surgical patients to analgesia with opioids alone, or opioids combined with dexmedetomidine (0.1 μg·kg−1·h−1) for the initial 48 postoperative hours.[34] Both analgesia and sleep quality improved, but the relative risk reduction for delirium was only 15% which was not statistically significant. Dexmedetomidine supplemented analgesia also failed to reduce delirium in another much smaller and seriously under-powered trial despite improved analgesia.[35] Supplemental dexmedetomidine therefore remains a reasonable sedative and analgesic, but should not be used with the expectation that it will much reduce delirium even in elderly patients recovering from major surgery.

In previous studies, postoperative administration of dexmedetomidine significantly reduced delirium in patients who remained overnight in the intensive care unit,[30,36,37] but not those transferred to the general wards after non-cardiac surgery.[34,35] Power was reduced in patients sent to surgical wards because they were presumably healthier and had less delirium than those who stayed in an intensive care unit (13.0 versus 22.4% in control patients).[30,34–37] Another reason is that for safety, ward studies used lower doses of dexmedetomidine (0.06–0.1 μg·kg−1·h−1) than those observed in monitored settings (0.1–0.7 μg·kg−1·h−1).[30,34–37] In the present study, we adopted a dosing regimen (mean infusion rate 0.026 μ−1.h−1) even lower than in previous studies performed in general wards,[34,35] in order to avoid potential side effects including sedation.[38] To the extent that the drug's effect is dose-dependent, less treatment effect would be expected in ward patients, again reducing trial power. However, prudence is necessary when administering dexmedetomidine in postoperative patients, especially those in the general ward. In a recent trial of 798 participants having cardiac surgery, even moderate dose dexmedetomidine (0.1–0.4 μg·kg−1· h−1 until 24 h postoperatively) increased clinical important hypotension.[39]

Orthopedic surgery is associated with substantial postoperative pain which may impair recovery by increasing complications including delirium.[7] Multimodal analgesia is thought to relieve pain after orthopedic surgery, but often insufficiently so. Most multimodal regimens do not include dexmedetomidine. In our patients, dexmedetomidine reduced NRS pain scores at rest and with movement; the changes reached or surpassed the minimal clinically important difference.[40] Our results are consistent with Shin et al.[41] who reported improved analgesia for up to 48 h. Interestingly, the analgesic benefits provided by dexmedetomidine extended beyond its biological half-life of 2 h. Specifically, analgesia was improved throughout 5 days of recovery even though the drug was always discontinued within 72 h. Supplemental dexmedetomidine thus appears to be a good strategy for relieving postoperative pain although further studies are required to confirm our findings.

Low-dose dexmedetomidine improves subjective sleep quality.[13,30] It also improves sleep architecture by reducing stage N1 sleep, increasing stage N2 sleep, and increasing sleep efficiency.[11,38] As might therefore be expected, we found that dexmedetomidine improved subjective sleep quality during the first 3 postoperative days. Dexmedetomidine might have improved sleep by activating endogenous sleep pathways.[42] Furthermore, good analgesia surely improves postoperative sleep.[43] In our results, dexmedetomidine did not cause excessive sedation or hemodynamic fluctuations, suggesting that the drug in current dose is a safe sedative and analgesic adjuvant.

The major limitation of this study is insufficient power. Our trial was under-powered mostly because the delirium incidence was lower than expected and because the apparent treatment effect was 35% rather than the anticipated 50%. But our results were in line with previous trials, indicating that the effect of dexmedetomidine supplemented analgesia on delirium appears to be relatively small. For safety reasons, the dose we used was relatively low; furthermore, because it was incorporated into patient-controlled analgesia, patients in pain received more of the drug. From a clinical perspective, this approach is reasonable and is often used, but differing doses does complicate interpretation of trial results. Multimodal analgesia was used in our patients, but there was no standardized pain management protocol. This also complicated the interpretation although increased the generalizability of our results. Comparison of outcomes repeated over time, such as pain severity and subjective sleep quality, increased the risk of type I errors. In our results, we did not correct for multiple outcomes. However, the differences of most of these results are robust and statistical compensation for multiplicity would not change our interpretations. Another limitation of our trial is that we did not consider long-term consequences of dexmedetomidine prophylaxis, but given the relatively small effect on delirium, it seems unlikely that any would have been observed.

In summary, supplementing sufentanil intravenous analgesia with low-dose dexmedetomidine did not significantly reduce delirium, but improved analgesia and sleep quality without provoking adverse events. The study was underpowered for the primary outcome. Dexmedetomidine remains a suitable supplement for multimodal analgesia in orthopedic surgical patients but further studies are required.