Effect of Dexmedetomidine Dose on Reducing Delirium in Adults: An Indirect Comparison of Controlled Trials

Introduction

Postoperative delirium is defined as delirium that occurs after a surgical procedure, and is characterized by disturbed consciousness and cognitive dysfunction [1,2]. It is most common in the early postoperative period, particularly in the first 3 days postoperatively. Delirium markedly impacts postoperative recovery, including prolonged ICU and hospital stays [3,4], increased incidence of perioperative complications [5], higher mortality [6], and a long-term lower quality of life for survivors [7].

Dexmedetomidine is a highly selective α2-adrenergic receptor agonist that induces anxiolysis, drowsiness, and mild analgesia while causing minor respiratory depression [8]. In laboratory trials, dexmedetomidine has been shown to provide protection against ischemia/reperfusion (I/R) and other types of damage to the heart, brain, kidneys, liver, and lungs. The anti-inflammatory and anti-delirium effects of dexmedetomidine have also been demonstrated [9–11].

According to accumulating evidence, dexmedetomidine appears to minimize postoperative delirium. However, the ideal dose of dexmedetomidine for reducing the incidence of postoperative delirium remains uncertain. The efficacy and safety of dexmedetomidine may differ depending on the dose used. In the absence of direct comparisons, indirect comparisons can provide useful information regarding the relative efficacy of different doses of dexmedetomidine. According to the loading dose, the 2 patients were divided into low-dose (≤0.5 μg/kg) and high-dose groups (>0.5 μg/kg). The goal of this study was to compare the efficacy of low-dose and high-dose dexmedetomidine treatments for postoperative delirium using an indirect comparison.

Material and Methods

PROTOCOL AND REGISTRATION:

This meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO, CRD42022377835).

SEARCH STRATEGY:

Two of the authors (J.Z. and K.H.) independently retrieved relevant studies published before November 18, 2022, from 4 electronic databases: PubMed, Embase, Web of Science, and the Cochrane Library. The search term “dexmedetomidine and delirium” was used in all databases. The full search strategy for PubMed was: (“dexmedetomidine” [MeSH Terms] OR “dexmedetomidine” [All Fields]) AND (“delirium” [MeSH Terms] OR “delirium” [All Fields]).

STUDY SELECTION:

We considered all randomized controlled trials that examined the effectiveness of high or low doses of dexmedetomidine in preventing postoperative delirium. Randomized, single-blind, or double-blind trials with adults (aged 18 years and over) published in full manuscript form were examined. Exclusion criteria were the use of dexmedetomidine after surgery, studies involving patients with pre-existing cognitive impairment, trials that did not use a loading dose of dexmedetomidine, nonrandomized clinical trials, case studies, and systematic reviews. Non-English language studies were excluded from the search. The papers in this meta-analysis were divided into 2 groups based on dexmedetomidine dosage: low-dose (0.5 μg/kg) and high-dose (>0.5 μg/kg). No benzodiazepines were used during the perioperative period.

DATA EXTRACTION AND OUTCOME MEASURES:

Two investigators (J.Z. and K.H.) independently extracted the data, and any disagreements were resolved by consulting a third reviewer (L.P. and X.J.). The study endpoints were defined as (1) incidence of delirium and (2) major adverse events (hypotension or bradycardia). Intraoperative hypotension was defined as a <20% decrease in systolic blood pressure relative to preoperative values. Bradycardia was defined as heart rate (HR) <50 bpm. We also evaluated the 30-day mortality rate, length of ICU stay, and length of hospital stay. However, given the lack of available data on these aspects after data extraction, these parameters could not be evaluated in this meta-analysis.

ASSESSMENT OF BIAS:

Using the Cochrane risk-of-bias instrument [12], 2 reviewers independently assessed the risk of bias in each study, including random sequence generation, allocation concealment, performance bias, detection bias, and attribution bias. Each bias was subsequently graded as “high,” “uncertain,” or “low” risk of bias.

STATISTICAL METHODS:

Delirium incidence, bradycardia incidence, and hypotension incidence were categorical variables, which were reported as a risk ratio (RR) with 95% CI. In the absence of randomized trials making head-to-head comparisons, an indirect comparison is possible using a common comparator [13], in which in population P, the relative effect of BC(P) of treatment C versus treatment B is determined from the difference in the relative effects of B and C against treatment A [14,15]. We performed a meta-analysis combining trials of low-dose dexmedetomidine versus placebo to obtain the estimated relative risk. A separate meta-analysis of trials comparing high-dose dexmedetomidine with placebo yielded an estimated relative risk. We obtained the estimated difference between the effect of low- and high-dose dexmedetomidine by combining the 2 estimated log-relative risks. Using these values, we calculated the 95% confidence interval for the logRR. All values were back-transformed to estimate the RR with a 95% confidence interval. The indirect comparison method compares the magnitude of the treatment effect in each trial by evaluating differences between the treatment and placebo arms. This method preserves the original trial randomization, so the potential intra-trial impact of effect-modifying, non-disease characteristics is removed. All analyses were performed using Stata version 15.0 (Texas, S.T.A.T.A. Center, Inc).

Results

IDENTIFICATION OF ELIGIBLE STUDIES:

We found 3775 potentially relevant studies (PubMed, n=328; Embase, n=1875; Cochrane Library, n=176; and Web of Science, n=1396). After removing duplicate hits, our computerized search yielded 2613 articles, among which 2447 were excluded after screening the titles and abstracts. A total of 166 papers were subjected to complete full-text assessment, of which 151 were removed for the reasons listed in Figure 1. Finally, 15 articles were included in the analysis.

CHARACTERISTICS OF TRIALS:

The study included 2519 patients from 15 trials (9 with low-dose dexmedetomidine [16–23] and 6 with high-dose dexmedetomidine [24–28]). The incidence of delirium was assessed using the Confusion Assessment Method (CAM) for the ICU [20,28] in 2 studies, CAM [17,19,23] in 5 studies, and other assessment methods [18,24,25] in 3 studies, and in the remaining 5 studies the instrument for delirium diagnosis was not disclosed in the manuscripts [16,21,22,26,27]. Surgical types differed among the studies. One trial, which included patients undergoing craniotomy hematoma removal, exhibited high heterogeneity and did not conduct a subgroup analysis.

RISK-OF-BIAS ASSESSMENT:

The results from the risk of bias summary and graphs of randomized controlled trials are presented in Figures 2 and 3. The risk of bias varied significantly among the studies. In the low-dose group, most studies performed random sequence allocation correctly and used appropriate methods of allocation concealment, whereas in the high-dose group, only 1 trial performed random sequence allocation correctly and used appropriate methods of allocation concealment; in the other trials this was unclear.

INCIDENCE OF DELIRIUM: Six studies examined the incidence of delirium using high-dose dexmedetomidine as a placebo, and 9 studies evaluated the frequency of delirium with low-dose dexmedetomidine as a placebo. Except for 1 trial, neither group had delirium. Nine trials comparing low-dose dexmedetomidine with placebo found no statistically significant benefit in reducing delirium (RR, 0.65; 95% CI, 0.40–1.06). The trials were heterogeneous (I2=57.7%) and were evaluated using a random-effects model (Figure 4). After meta-analysis, high-dose dexmedetomidine was found to be more strongly associated with a lower incidence of delirium than low-dose dexmedetomidine (RR, 0.61; 95% CI, 0.38–0.96; P=1.15, Table 1). Almost all confidence intervals (CI) included the value 1, and only 3 studies actually showed a reduction in delirium.

INCIDENCE OF HYPOTENSION: The incidence of hypotension was reported as an outcome in 9 studies [17–20,22,23,26]. Pooled data analysis revealed no significant difference in the incidence of hypotension between the 2 groups (RR, 1.15; 95% CI, 0.71–1.87; P=0.56, Table 1). Compared to placebo, dexmedetomidine had no effect on the occurrence of hypotension (Figure 5).

INCIDENCE OF BRADYCARDIA: In 8 studies [17,19,20,22,23,27] the incidence of bradycardia was examined as an outcome. In 1 study, none of the groups experienced bradycardia. High-dose dexmedetomidine was associated with a significantly higher risk of bradycardia than placebo, whereas low-dose dexmedetomidine and placebo did not significantly differ in terms of the incidence of bradycardia (Figure 6). Interestingly, indirect comparisons revealed no significant difference in the incidence of bradycardia between the low- and high-dose groups (RR, 1.28; 95% CI, 0.61–2.67; P=0.516, Table 1).

Discussion

LIMITATIONS:

This study has some limitations. First, 6 of the included studies had a sample size of 100 patients and can thus be considered as minor studies. Consequently, our study may have been biased by the modest study effect. Second, our study enrolled only patients who received dexmedetomidine intraoperatively. Third, 8 studies did not consider postoperative delirium as their major endpoint. Instead, other outcomes, such as postoperative sleep quality, visual analog scale pain levels, and patient satisfaction, were prioritized. Fourth, not all studies used validated tools to diagnose delirium. Consequently, our retrieved data may have been coincidentally discovered in these articles. Finally, considerable heterogeneity was observed in the low-dose group (I² >50%) in the meta-analysis, and more randomized controlled trials are required to validate these findings.

Conclusions

Our meta-analysis revealed that high-dose dexmedetomidine may be more effective than low-dose regimens in reducing delirium prevalence. Dexmedetomidine is unlikely to cause bradycardia or hypotension.