Impact of intraoperative dexmedetomidine on postoperative outcomes in shoulder arthroscopy: a systematic review and meta-analysis

Abstract

Objective

To evaluate the effectiveness and safety of dexmedetomidine in adult patients undergoing shoulder arthroscopy.

Methods

A systematic search was conducted in PubMed, Embase, and the Cochrane Library along with a manual search of reference lists from previous systematic reviews and meta-analyses. We applied random effects and fixed effects models for the analysis of outcomes, depending on the level of heterogeneity observed. We assessed the risk of bias using version 2 of the Cochrane Risk of Bias tool (RoB 2) and conducted sensitivity analyses to examine the robustness of our results. Additionally, evidence quality was rated using the GRADE system, with studies categorized as high quality (level I), moderate quality (level II or III), or low quality (level IV).

Results

Eighteen randomized controlled trials involving 1,203 patients were included. Dexmedetomidine significantly reduced postoperative pain scores at 6 hours (SMD = –0.44, 95% CI –0.63 to –0.25) and 12 hours (SMD = –0.47, 95% CI –0.71 to –0.23) compared with those of the control groups. The incidence of PONV was also lower in the dexmedetomidine group (RR = 0.62, 95% CI 0.46–0.84). However, dexmedetomidine was associated with an increased risk of bradycardia (RR = 2.58, 95% CI 1.71–3.88) and hypotension (RR = 1.94, 95% CI 1.32–2.86). Although these findings should be interpreted with caution, as most studies reported these events without standardized definitions or severity grading.

Conclusion

Dexmedetomidine appears to be effective at reducing postoperative pain and the incidence of nausea and vomiting following shoulder arthroscopy. Although the magnitude of pain reduction was modest, the effect was statistically significant and may still contribute to improved patient comfort and reduced opioid consumption when used as part of a multimodal analgesic regimen. Careful monitoring is recommended because of its potential hemodynamic effects, such as bradycardia and hypotension.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12871-026-03798-9.

Introduction

Shoulder arthroscopy is typically used in clinical practice to treat chronic shoulder disorders or acute injuries. Compared with traditional open surgery, shoulder arthroscopy offers several advantages, including minimal invasiveness, less soft tissue damage, fewer surgical complications, and faster functional recovery [1]. Despite these benefits, postoperative complications remain a concern. Common postoperative complications include pain [2], restricted shoulder mobility, and neurovascular injury [3]. Studies have shown that 74.3% of patients experience moderate to severe pain within the first 3 days after shoulder surgery, with this percentage decreasing to 66.2% at 6 weeks after surgery [4]. Key risk factors for early postoperative pain after shoulder arthroscopy include female sex [5, 6], nonfull-thickness rotator cuff tears [7], and concomitant fat infiltration [8]. The management of postoperative pain is multifaceted and involves pharmacological treatment, physical therapy, psychological support, and individualized rehabilitation plans.

Dexmedetomidine is a highly selective α2-adrenergic receptor agonist with notable sedative and analgesic properties. Studies have shown that when combined with other sedative and analgesic agents, dexmedetomidine significantly reduces the need for other anesthetic drugs [9–11], such as opioids [12], and improves patient satisfaction. Owing to the unique positioning required during shoulder arthroscopy, controlled hypotension [13] and reduced intraoperative blood loss are necessary. Research has indicated that dexmedetomidine provides stable hemodynamics in elderly patients, reduces intraoperative blood loss, and facilitates surgical maneuvers [14]. Additionally, dexmedetomidine has neuroprotective effects [15]. Dexmedetomidine can reduce postoperative delirium [16], stress [17], and inflammatory responses [18] ; shorten the duration of postoperative delirium [19]; and enhance postoperative cognitive function [20]. In addition to its sedative and analgesic properties, dexmedetomidine has been shown to prolong the duration of sensory and motor blockade when used as an adjuvant in regional anesthesia, thereby enhancing postoperative analgesia and reducing opioid consumption [21]. This property highlights the potential clinical value of dexmedetomidine in extending block duration and improving recovery quality in patients undergoing shoulder arthroscopy.

To date, no comprehensive meta-analysis has specifically synthesized evidence on both the analgesic efficacy and the hemodynamic safety of dexmedetomidine in patients undergoing shoulder arthroscopy. This lack of a focused quantitative evaluation limits clinical understanding of its overall benefit–risk balance in this population. Therefore, the present systematic review and meta-analysis was conducted to clarify the effectiveness and safety profile of dexmedetomidine in shoulder arthroscopy, providing evidence-based insights to guide perioperative management.

Methods

Standard protocol approval, registration, and patient consent

Our research proposal has been registered on PROSPERO, an international systematic review registration website (registration number: CRD420251039059). This study was conducted and reported in accordance with the Cochrane handbook and the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 statement [22]. Given the nature of this study, informed consent and institutional review board approval are not required.

Search strategy

A comprehensive literature search was conducted to assess the impact of dexmedetomidine on postoperative outcomes following shoulder arthroscopy, covering the period from inception to July 24, 2025. A combination of keywords and MeSH terms was used, including “Shoulder Joint”, “Rotator Cuff Injury”, “Anesthesia”, “Dexmedetomidine”, and “Postoperative Complications” (eTable 1). We performed searches in PubMed, Embase, and the Cochrane Library using both free-text terms (including synonyms and closely related terms) and MeSH terms. In cases where multiple studies reported on the same cohort, we included only the most recent study or the one with the largest sample size for analysis. Notably, no restrictions were applied regarding language, sex distribution, publication year, or sample size during the search process. In addition to this comprehensive search strategy, we meticulously reviewed the reference lists of relevant systematic reviews and included studies that were cited.

We used ENDNOTE 21 software to import titles and abstracts and initially screened for duplicates and citations that did not meet the inclusion criteria. Two researchers (X.B. and H.Y.) independently screened the titles and abstracts for eligibility and extracted the relevant data. Discrepancies were resolved by discussion with a third researcher (Y.L.). Risk-of-bias assessments were independently performed by L.Y. and Y.W., with a consensus reached through consultation with H.Z. A flowchart detailing the screening process is provided in Fig. 1.

Fig. 1.

Flowchart of the study selection process

Inclusion and exclusion criteria:

The selection criteria for this study adhered to the PICOS principles (population, intervention, comparison, outcome, and study design).

Inclusion criteria:

1. Population: Patients who underwent shoulder arthroscopy.

2. Intervention: Administration of dexmedetomidine during the perioperative period.

3. Control Group: Standard care, placebo or opioid use were considered acceptable as control group conditions.

Exclusion criteria:

1. Case reports, reviews, expert opinions, meta-analyses, and conference abstracts were excluded.

2. Studies with insufficient experimental data in the original literature (e.g., unclear or incomplete sample data interpretation) were excluded.

Data extraction

Two independent authors (X.B. and H.Y.) extracted the following information from each study: first author, year of publication, study location, sample size, mean age, mean weight, intervention and control drug dosages, injection volumes, injection timing, and outcome data (including the number of patients in each group, means and standard deviations for continuous outcomes, and the number of events for dichotomous outcomes).

Quality assessment

Given that only RCTs were included, the risk of bias was assessed using the second version of the Cochrane Risk of Bias tool (RoB 2). The risk assessment was conducted by two independent authors, with any discrepancies resolved through discussion of the factors contributing to the disagreement. The level of evidence for each study was also evaluated (Fig. 2 Table 1).

Fig. 2.

Methodological quality score of the included studies based on version 2 of the Cochrane Risk of Bias tool (RoB 2)

Table 1.

Risk of Bias 2 (RoB 2) — Domain-Level Judgments for 18 Randomized Controlled Trials

Study	D1 Randomization process	D2 Deviations from intended interventions	D3 Missing outcome data	D4 Measurement of the outcome	D5 Selection of the reported result	Overall risk of bias	Rationale summary
Maghawry et al., 2015 [26]	Low	Low	Low	Low	Low	Low	Adequate randomization; objective physiologic measures.
Jang et al., 2016 [39]	Low	Low	Low	Low	Some concerns	Some concerns	Randomization clear; registry not reported.
Lu et al., 2017 [27]	Low	Low	Low	Low	Low	Low	Blinded; complete data; consistent outcomes.
Kataria et al., 2019 [28]	Low	Low	Low	Low	Low	Low	Double-blind; adequate concealment.
Hwang et al., 2020 [29]	Low	Low	Low	Low	Low	Low	Robust randomization; clear outcome reporting.
Wang et al., 2021 [31]	Low	Low	Low	Low	Some concerns	Some concerns	Randomization and reporting partly unclear.
Lee et al., 2021 [30]	Low	Low	Low	Low	Low	Low	Well-conducted; consistent reporting.
Kanakalakshmi et al., 2022 [32]	Low	Low	Low	Low	Low	Low	Methods clear; open-label justified; objective outcomes.
Albrecht et al., 2023 [33]	Low	Low	Low	Low	Low	Low	Registered RCT; fully reported results.
Hamed et al., 2023 [34]	Low	Low	Low	Low	Low	Low	Double-blind; pre-specified outcomes.
Luan et al., 2023 [36]	Low	Low	Low	Low	Some concerns	Some concerns	Registry available but incomplete reporting.
Zhao et al., 2023 [41]	Low	Low	Low	Low	Some concerns	Some concerns	Randomization clear; registry absent.
Khaled et al., 2023 [35]	Low	Low	Low	Low	Low	Low	Objective outcomes; consistent reporting.
Hapugoda et al., 2024 [42]	Low	Low	Low	Low	Low	Low	Clear outcomes; objective endpoints.
Helwa et al., 2024 [43]	Low	Low	Low	Low	Low	Low	Assessor-blind; full data available.
Xue et al., 2024 [40]	Low	Low	Low	Low	Low	Low	Registered; low attrition; good transparency.
Nashaat et al., 2024 [38]	Low	Low	Low	Low	Some concerns	Some concerns	Reporting selective; no registration found.
Huan et al., 2025 [37]	Low	Low	Low	Low	Low	Low	Randomized, blinded; adequate reporting.

Publication bias was evaluated by funnel graph analysis and Egger’s test to determine whether trials with comparable weights were symmetrically distributed across all end points.

Assessing the strength of evidence

The Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) grading system [23] was used to grade the strength of the evidence in the identified associations. The GRADE grading system is a scientifically rigorous and transparent method developed by the GRADE Working Group for evaluating the quality of evidence and recommending the strength of clinical practice guidelines. The evaluation encompasses five factors for downgrading evidence: risk of bias, publication bias, imprecision, inconsistency, and indirectness. The classification of evidence in this study is depicted in eTable 2, 3.

Statistical analysis

For dichotomous outcomes, risk ratios (RRs) with 95% confidence intervals (CIs) were calculated, whereas standardized mean differences (SMD) with 95% CIs were used for continuous outcomes. All pooled analyses were performed using a random-effects model to account for potential clinical and methodological heterogeneity among the studies. Heterogeneity between studies was assessed using the Cochran Q test and I² statistic. Heterogeneity was considered to be statistically significant when p < 0.05 or I² ≥ 50% [24]. I² values of 25%-50% indicated low heterogeneity, 50%-75% indicated moderate heterogeneity, and 75%-100% indicated high heterogeneity. Sensitivity analysis was performed to exclude sources of heterogeneity, and a random-effects model was used to compute the final results. To further explore the potential sources of heterogeneity, we conducted additional subgroup analyses. The Galbraith plot was used to assess inter-study heterogeneity and identify potential outliers. The symmetry of the funnel plots was visually assessed, and potential publication bias was evaluated using Egger’s test [25]. Additionally, Duvall & Tweedie’s trim-and-fill method was employed to adjust the pooled results for potential publication bias. All the statistical tests were two-sided, with P < 0.05 considered to indicate statistical significance. All the statistical analyses were performed using STATA software (version 17.0; StataCorp, College Station, TX, USA).

Results

Selection of the included studies

Initially, a total of 26,954 articles were retrieved. After removing duplicates and screening for irrelevant studies, we conducted full-text reviews of the remaining 63 potentially eligible studies. Owing to the inclusion of meta-analyses, reviews, or missing outcome data, 42 studies were ultimately excluded. Therefore, 18 randomized controlled trials [26–40] involving 1,203 participants met the inclusion criteria for this meta-analysis.

Study characteristics

All the studies were published between 2015 and 2025, with 78% of the studies being published in 2020 or later. Approximately 68% of the participants were male, and 602 patients (50%) were randomly assigned to the dexmedetomidine combined with shoulder arthroscopy group. Tables 2 and 3 presents the baseline characteristics of each study along with their respective levels of evidence. This meta-analysis included studies from three continents: 13 from Asia [26–32, 36, 37, 39–42], 4 from Africa (all conducted in Egypt) [34, 35, 38, 43], and 1 from Europe [33]. The mean age of the participants was 54.4 years.

Table 2.

Baseline characteristics of included studies

Authors, year	Country	Population number(C, D)	Female (%)(C, D)	Mean age(C, D)	Mean weight in kg (C, D)	Intervention (Group D)	Control (Group C)	Follow-up	Level of Evidence
Maghawry et al. 2015 [26]	Saudi Arabia	50;25;25	44%;52%	36.9;41.1	83.3;81.6	Normal saline + DEX	Normal saline + Esmolol	7 Days	Level I
Jang et al. 2016 [39]	Korea	50;25;25	36%;24%	61.2;58.1	67.6;68.3	Ropivacaine + DEX	Ropivacaine + Normal saline	NR	Level I
Lu et al. 2017 [27]	China	152;76;76	NR	65.0;65.5	67.2;66.3	Sufentanil + DEX	Sufentanil	7 Days	Level II
Kataria, et al. 2019 [28]	India	60;30;30	20%;16.7%	30.17;30.13	73.2;73.97	Ropivacaine + DEX	Ropivacaine + DXA	24 h	Level I
Hwang et al. 2020 [29]	Korea	50;25;25	36%;24%	61.9;55.6	67.5;68.5	DEX + Ropivacaine	Ropivacaine + Normal saline	2 Days	Level I
Wang et al. 2021 [31]	China	60;30;30	56.7%;53.3%	43.7;46.3	BMI(24.5;25.0)	DEX	Midazolam + Propofol + Sufentanil + Cisatracurium	2 Days	Level II
Lee et al. 2021 [30]	Korea	40;20;20	50%;55%	63.0;58.0	65.5;65.5	Ropivacaine + DEX	Ropivacaine + Normal saline	2 Days	Level I
Kanakalakshmi, et al. 2022 [32]	India	48;24;24	8.33%;8.33%	38.95;38.41	NR	DEX	Bupivacaine	24 h	Level I
Albrecht et al. 2023 [33]	France	122;62;60	64.52%;61.67%	57.1;57.5	78.2;79.7	Dexamethasone + DEX	Dexamethasone + normal saline	4 Days	Level I
Hamed et al. 2023 [34]	Egypt	60;30;30	14.8%;21.4%	44.81;41.11	BMI(32.42,31.79)	Bupivacaine + DEX	Bupivacaine	24 h	Level I
Luan et al. 2023 [36]	China	44;22;22	54.5%;40.9%	49.8;48.9	71.3;72.5	Ropivacaine + DEX	Ropivacaine	24 h	Level II
Zhao et al. 2023 [41]	China	74;37;37	54.05%;35.14%	54.24;53.68	68.47;67.28	DEX + ropivacaine + lidocaine	Normal saline + ropivacaine + lidocaine	24 h	Level II
Khaled et al. 2023 [35]	Egypt	30;15;15	26.7%;40%	70.8;70.1	NR	Dexmedetomidine +lidocaine	Fentanyl	3 Days	Level I
Hapugoda et al. 2024 [42]	Sri Lanka	56;28;28	21.5%;25%	44.07;46.14	BMI(21.99;22.79)	DEX	Propofol	24 h	Level I
Helwa et al. 2024 [43]	Egypt	140;70;70	61.4%;71.4%	65.5;65	BMI(30;29.9)	DEX	Normal saline	14 Days	Level I
Xue et al. 2024 [40]	China	60;30;30	33.3%;40%	59.6;57.9	BMI(24.7;25.5)	DEX	Propofol + Cisatracurium + Fentanyl	24 h	Level I
Nashaat et al. 2024 [38]	Egypt	60;30;30	20%;23.3%	32.9;33.1	BMI(26.8;27.3)	Bupivacaine + DEX	Bupivacaine + normal saline	14 Days	Level II
Huan et al. 2025 [37]	China	47;23;24	38.3%;38.3%	36.6;51.7	BMI(24.9;24.5)	Ropivacaine + DEX	Ropivacaine	24 h	Level I

BMI Body mass index, C control group, D dexmedetomidine group, DEX dexmedetomidine, DXA dexamethasone, Levobupi levobupivacaine, MAC Minimum Alveolar Concentration, SEV sevoflurane, NR Not Reported

Table 3.

Summary of Dexmedetomidine Dosing Regimens Across Included Randomized Controlled Trials

Study (Year)	Country	Route of Administration	Dexmedetomidine Dose	Local/Concomitant Anesthetic	Timing of Administration	Type of Anesthesia	Follow-up	Main Reported Outcomes
Maghawry et al., 2015 [26]	Saudi Arabia	Intravenous (IV)	0.2 ~ 0.6 µg/kg/h	DEX vs. Esmolol	After anesthesia	General anesthesia	7 days	MAP and HR stability
Jang et al., 2016 [39]	Korea	Perineural	100 µg	Ropivacaine + DEX vs. Ropivacaine + normal saline	After anesthesia	General anesthesia	NR	Duration of operation, BCP time
Lu et al., 2017 [27]	China	Intravenous (IV)	0.5 µg/kg/h	Sufentanil + DEX vs. Sufentanil	30 min before anesthesia	Local anesthesia	7 days	Early cognitive outcomes
Kataria et al., 2019 [28]	India	Perineural	0.5 mcg/kg	Ropivacaine + DEX vs. Ropivacaine + DXA	After anesthesia	General anesthesia	24 h	Duration of analgesia
Hwang et al., 2020 [29]	Korea	Perineural	100 µg	Ropivacaine + DEX vs. Ropivacaine + saline	Simultaneous with local anesthesia	Local anesthesia	2 days	VAS, IL-6, IL-8 levels
Wang et al., 2021 [31]	China	Intravenous (IV)	0.2 ~ 0.7 µg/kg/h	DEX vs. midazolam + propofol + sufentanil + cisatracurium	After anesthesia	General anesthesia	2 days	Pain score, hemodynamics
Lee et al., 2021 [30]	Korea	Perineural	0.5 ml(50 µg)	Ropivacaine + DEX vs. Ropivacaine + saline	NR	Local anesthesia	2 days	Postoperative pain
Kanakalakshmi et al., 2022 [32]	India	Intravenous (IV)	0.5mcg/kg/h	DEX vs. Bupivacaine	After anesthesia	General anesthesia	24 h	Analgesic effect
Albrecht et al., 2023 [33]	France	Intravenous (IV)	1 µg/kg/h	Dexamethasone + DEX vs. Dexamethasone + saline	30 min after induction	General anesthesia	4 days	Duration of block, PONV
Hamed et al., 2023 [34]	Egypt	Perineural	0.5 µg/kg	Bupivacaine + DEX vs. Bupivacaine	After anesthesia	General anesthesia	24 h	ESP block analgesia
Luan et al., 2023 [36]	China	Intravenous (IV)	0.5 µg/kg/h	Ropivacaine + DEX vs. Ropivacaine	After anesthesia	General anesthesia	24 h	Pain, satisfaction
Zhao et al., 2023 [41]	China	Intravenous (IV)	0.5 µg/kg/h	DEX + ropivacaine + lidocaine vs. saline	Before anesthesia	General anesthesia	24 h	Lung compliance, HR
Khaled et al., 2023 [35]	Egypt	Intravenous (IV)	1 µg/kg/h	DEX vs. Fentanyl	Before anesthesia	General anesthesia	3 days	Hemodynamic stability
Hapugoda et al., 2024 [42]	Sri Lanka	Intravenous (IV)	0.1 ~ 0.7 µg/kg/h	DEX vs. Propofol	After anesthesia	General anesthesia	24 h	Hemodynamic changes
Helwa et al., 2024 [43]	Egypt	Intravenous (IV)	0.6 µg/kg/h	DEX vs. saline	After anesthesia	General anesthesia	14 days	Cognitive function
Xue et al., 2024 [40]	China	Intravenous (IV)	0.3–0.5 µg/kg/h	DEX vs. propofol + cisatracurium + fentanyl	Atter endotracheal intubation	General anesthesia	24 h	PONV, hemodynamic profile
Nashaat et al., 2024 [38]	Egypt	Perineural	0.1 µg/kg	Bupivacaine + DEX vs. Bupivacaine + saline	After anesthesia	General anesthesia	14 days	Analgesic duration
Huan et al., 2025 [37]	China	Perineural	50 µg	Ropivacaine + DEX vs. Ropivacaine	After anesthesia	Local anesthesia	24 h	Rebound pain reduction

Pain scores

Pain scores were defined using the Numeric Rating Scale (NRS) or Visual Analog Scale (VAS) on a 0–100 mm or 0–10 cm scale. When the 0–100 mm scale was used, both the mean values and standard deviations were divided by 10 to achieve equivalence at the 0–10 cm scale for clarity and ease of data interpretation.

Postoperative pain scores at 6 h

Five studies [27, 29–31, 40] assessed the relationship between dexmedetomidine and postoperative pain scores at 6 h following shoulder arthroscopy. The results demonstrated that dexmedetomidine significantly reduced postoperative pain scores at 6 h (SMD, -0.68; 95% CI, -1.07 to -0.29), with moderate heterogeneity observed (I² = 66.6%; P = 0.017) (eFigures 1). Sensitivity analysis revealed that the study by Wang et al. [31] was the main source of heterogeneity. After excluding this study, the heterogeneity was reduced to low levels (I² = 42.2%, P = 0.158), and the remaining results remained statistically significant (SMD, -0.82; 95% CI, -1.16 to -0.49) (Fig. 3).

Fig. 3.

Random-effects model Forest Plot for meta-analysis of association between pain scores (PO 6 h) and dexmedetomidine after shoulder arthroscopy, after excluding studies contributing to heterogeneity

Postoperative pain scores at 12 h

Nine studies [27, 29–31, 34, 36, 38, 40, 43] examined the relationship between dexmedetomidine and postoperative pain scores at 12 h following shoulder arthroscopy. The results demonstrated that dexmedetomidine significantly reduced postoperative pain scores at 12 h (SMD, -1.75; 95% CI, -2.69 to -0.82), with high heterogeneity observed (I² = 96.1%; P = 0.00). Unfortunately, sensitivity analysis did not identify the source of heterogeneity for this outcome (Fig. 4). To further explore potential sources of heterogeneity, we performed additional subgroup analyses stratified by: Route of administration (intravenous vs. perineural); Type of anesthesia (general vs. Local anesthesia); Follow-up duration (< 1 day; 1–7 days; >7 days); Sample size (< 50; 50–100; >100 participants). However, despite stratification by these variables, heterogeneity remained high within most subgroups (Fig. 5). Meta-regression was considered but not conducted due to the limited number of studies per covariate and insufficient statistical power.

Fig. 4.

Random-effects model Forest Plot for meta-analysis of association between pain scores (PO 12 h) and dexmedetomidine after shoulder arthroscopy

Fig. 5.

Subgroup analysis forest plot for meta-analysis of association between pain scores (PO 12 h) and dexmedetomidine after shoulder arthroscopy

Postoperative pain scores at 24 h

Eight studies [27, 29–31, 34, 36, 38, 40] assessed the relationship between dexmedetomidine and postoperative pain scores at 24 h following shoulder arthroscopy. The results demonstrated that dexmedetomidine significantly reduced postoperative pain scores at 24 h (SMD, -0.77; 95% CI, -1.28 to -0.26), with moderate heterogeneity observed (I² = 86.5%; P = 0.000) (eFigure 2). Sensitivity analysis revealed that the study by Nashaat et al. [38] was the main source of heterogeneity. After excluding this study, the heterogeneity was reduced to low levels (I² = 0.0%, P = 0.782), and the remaining results remained statistically significant (SMD, -0.44; 95% CI, -0.63 to -0.26) (Figs. 6).

Fig. 6.

Random-effects model Forest Plot for meta-analysis of association between pain scores (PO 24 h) and dexmedetomidine after shoulder arthroscopy, after excluding studies contributing to heterogeneity

Postoperative heart rate

Heart rate after intubation

Four studies [26, 35, 41, 43] investigated the effect of dexmedetomidine on heart rate following intubation in patients undergoing shoulder arthroscopy. The results indicated that dexmedetomidine significantly reduced heart rate postintubation (SMD, -0.40; 95% CI, -0.63 to -0.17), with no significant heterogeneity observed (I² = 0.00%; P = 0.569) (eFigure 3).

Intraoperative 1 h heart rate

Four studies [26, 35, 41, 43] analyzed the association between dexmedetomidine and heart rate at 1 h intraoperatively in patients who underwent shoulder arthroscopy. The results demonstrated that compared with the control treatment, dexmedetomidine significantly reduced heart rate at 1 h (SMD, -0.72; 95% CI, -1.19 to -0.25), with moderate heterogeneity observed (I² = 70.04%, P = 0.018). Unfortunately, sensitivity analysis failed to identify the source of this heterogeneity (eFigure 4).

Heart rate after extubation

Four studies [26, 35, 41, 43] examined the relationship between dexmedetomidine and heart rate after extubation in patients who underwent shoulder arthroscopy. The results demonstrated that dexmedetomidine significantly reduced heart rate post extubation (SMD, -0.92; 95% CI, -1.47 to -0.37), with moderate heterogeneity observed (I² = 77.4%; P = 0.004). Sensitivity analysis revealed that the study by Zhao et al. [41] was the primary source of heterogeneity. After excluding this study, the heterogeneity decreased to a mild level (I² = 30.4%, P = 0.238), with the remaining results still showing statistical significance (SMD, -0.63; 95% CI, -0.99 to -0.27) (eFigures 5, 6).

Postoperative bradycardia

Seven studies [27, 33, 34, 39, 40, 42, 43] examined the relationship between dexmedetomidine and bradycardia in patients who underwent shoulder arthroscopy. The results revealed that patients who received dexmedetomidine were more likely to experience this complication (RR, 1.69; 95% CI, 1.19 to 2.40), with no significant heterogeneity observed (I² = 0.00%; P = 0.874) (Fig. 7).

Fig. 7.

Random-effects model Forest Plot for meta-analysis of association between Postoperative Bradycardia and dexmedetomidine after shoulder arthroscopy

Overall, dexmedetomidine was associated with a significant reduction in heart rate in patients who underwent shoulder arthroscopy, with all the results reaching statistical significance. However, it also notably increased the risk of postoperative bradycardia.

Postoperative nausea and vomiting

Six studies [27, 31, 34, 40, 43] evaluated the association between dexmedetomidine use and the incidence of postoperative nausea and vomiting (PONV) in patients who underwent shoulder arthroscopy. The pooled results indicated that dexmedetomidine significantly reduced the risk of PONV (RR, 0.46; 95% CI, 0.31 to 0.68), with no significant heterogeneity observed among the studies (I² = 0.00%; P = 0.899) (Fig. 8).

Fig. 8.

Random-effects model Forest Plot for meta-analysis of association between Postoperative nausea and vomiting and dexmedetomidine after shoulder arthroscopy

Postoperative hypotension

Six studies [31, 33, 34, 39, 42, 43] investigated the association between dexmedetomidine use and the incidence of postoperative hypotension in patients who underwent shoulder arthroscopy. The results of the pooled analysis suggested a higher risk of hypotension in patients who received dexmedetomidine (RR, 1.50; 95% CI, 0.81 to 2.78), with substantial heterogeneity among the studies (I²=78.9%; P = 0.00). Sensitivity analysis identified the study by Wang et al. [31] as the primary source of heterogeneity. After excluding this study, the association remained statistically significant (RR, 2.05; 95% CI, 1.62 to 2.59), and no significant heterogeneity was observed (I² = 0.00%; P = 0.811) (eFigures 7–8).

Intraoperative mean arterial pressure

Five studies [26, 35, 41–43] reported the association between dexmedetomidine use and mean arterial pressure (MAP) at 1 h intraoperatively. The pooled results indicated that patients who received dexmedetomidine had significantly lower intraoperative MAP than controls did (SMD, -0.56; 95% CI, -1.05 to -0.08), with substantial heterogeneity observed (I² = 78.0%; P = 0.001). Sensitivity analysis identified the study by Wang et al. [31] as the primary source of heterogeneity. After excluding this study, heterogeneity was reduced to a mild level (I² = 39.5%; P = 0.175), and the results remained statistically significant (SMD, -0.74; 95% CI, -1.07 to -0.41) (eFigures 9, 10).

Patient satisfaction

Four studies [28–30, 36] evaluated the association between dexmedetomidine use and postoperative satisfaction in patients who underwent shoulder arthroscopy. The pooled analysis indicated a trend toward higher satisfaction among patients who did not receive dexmedetomidine (SMD, 1.30; 95% CI, -0.32 to 2.93). Substantial heterogeneity was observed (I² = 95.8%; P = 0.00), and the source of heterogeneity could not be identified through sensitivity analysis. To further explore potential sources of heterogeneity, we performed additional subgroup analyses stratified by: Route of administration (intravenous vs. perineural); Type of anesthesia (general vs. Local anesthesia); Follow-up duration (< 1 day; 1–7 days; >7 days); Sample size (< 50; 50–100; >100 participants). However, despite stratification by these variables, heterogeneity remained high within most subgroups (Fig. 5). (eFigs. 11 and 12). Meta-regression was considered but not conducted due to the limited number of studies per covariate and insufficient statistical power.

Bias and sensitivity analysis

As previously described in the statistical analysis, publication bias was assessed for the outcomes reported in the 18 included studies. Specifically, bias assessments were conducted for postoperative pain scores, heart rate, postoperative nausea and vomiting, bradycardia, hypotension, patient satisfaction, and intraoperative mean arterial pressure. The results indicated no significant risk of publication bias for most outcomes, except for pain scores at 12 h postoperatively and patient satisfaction (eFigures 13–23). Sensitivity analysis was performed to evaluate the robustness of the results for each study factor combination (eFigures 24–36). All the results are summarized in Table 4.

Table 4.

Meta-analyses of the impact of Intraoperative Dexmedetomidine on Postoperative Outcomes in Shoulder Arthroscopy

Outcomes	No. of Studies	No. of Patients	SMD/RR (95% CI)	I², %	P value	Egger’s test P value
Pain score
6 h PO	54	302	-0.82(-1.16 to -0.49)	42.2	< 0.001	1.000
12 h PO	9	661	-1.75(-2.69 to -0.82)	96.1	< 0.001	0.063
24 h PO	8	521	-0.44(-0.63 to -0.26)	0.0	< 0.001	1.000
Heart rate
After intubation	4	224	-0.4(-0.63 to -0.17)	0.0	< 0.001	0.734
60-min IO	4	224	-0.72(-1.19 to -0.25)	70.0	< 0.001	0.734
Ater extubation	4	224	-0.63(-0.99 to -0.27)	30.4	< 0.001	0.296
PONV	6	514	0.46(0.31 to 0.68)	0.0	< 0.001	0.707
Bradycardia	7	635	1.69 (1.19 to 2.40)	0.0	0.01	0.548
Hypotension	5	423	2.05(1.62 to 2.59)	0.0	< 0.001	1.000
MAP	4	294	-0.74(-1.07 to -0.41)	39.5	< 0.001	0.308
Patient satisfaction	4	97	1.30 (-0.32 to 2.93)	95.8	0.12	0.308

Abbreviation: IO intraoperative, MAP Mean artery pressure, PO postoperative, PONV Postoperative nausea and vomiting, SMD standardized mean differences, RR relative risk

Discussion

Principal findings

In this meta-analysis, a total of 18 randomized controlled trials involving 1,203 participants were included. The pooled results indicate that dexmedetomidine may effectively reduce postoperative pain scores and the incidence of nausea and vomiting following shoulder arthroscopy. However, this benefit may come at the cost of an increased risk of adverse hemodynamic events, such as hypotension and bradycardia. Our findings underscore the importance of conducting a thorough risk–benefit evaluation before using dexmedetomidine in clinical practice.

Comparisons with previous literature and potential mechanisms

Our findings are consistent with those of two previously published studies [44, 45] that reported that patients in the dexmedetomidine group experienced significantly lower postoperative pain scores compared to controls. However, both studies demonstrated statistically significant differences only in overall pain scores and did not provide a time-specific analysis of postoperative pain. In contrast, our time-point stratification allows for a more nuanced evaluation of analgesic efficacy over time, which is essential for timely adjustment of pain management strategies and for ensuring the continuity and effectiveness of postoperative analgesia.

Although most included RCTs used regional anesthesia with long-acting local anesthetics (such as ropivacaine or bupivacaine), patients receiving dexmedetomidine still had significantly lower pain scores at 6 and 12 h. These results may be explained by the dual mechanisms of dexmedetomidine. Dexmedetomidine, an α2-adrenergic receptor agonist, exerts both analgesic and sedative effects through multifactorial mechanisms that are not yet fully understood. Peripherally, its analgesic effects are believed to stem from the inhibition of norepinephrine release and the nonreceptor-dependent suppression of action potentials in nerve fibers. Centrally, α2-agonists inhibit the release of substance P at the level of dorsal horn neurons and activate α2-adrenergic receptors in the locus coeruleus, contributing to both analgesia and sedation [11, 29, 46–48]. Moreover, dexmedetomidine attenuates serotonergic neurotransmission, potentially by reducing the stimulation of 5-HT2 receptors [49, 50]. Another proposed mechanism involves the inhibition of hyperpolarization-activated cation currents (Ih currents), which suppress peripheral nerve excitability and thereby reduce serotonergic transmission and neuronal hyperexcitability [46]. These mechanisms may account for the observed time-dependent analgesic benefit without implying unequal block efficacy.

In previous acute postoperative pain studies, MCID values for pain VAS on 0–100 mm scales have been reported to range from approximately 8 to 10 mm [51]. In patients with shoulder pain, a study of rotator cuff disease estimated an MCID of 1.4 cm on a 10-cm VAS [52]. These data suggest that although our pooled meta-analysis revealed statistically significant pain reduction, the magnitude of the effect may not always surpass patient-perceived meaningful thresholds in some settings. However, even modest reductions in pain intensity may contribute to greater patient comfort and reduced opioid consumption when dexmedetomidine is used as part of multimodal analgesic regimens.

Dexmedetomidine-induced bradycardia is primarily attributed to a reduction in central sympathetic outflow and an increase in parasympathetic activity, mediated by the activation of α2-adrenergic receptors in the brainstem, particularly within cardiac vagal neurons [53]. This central sympatholytic effect reduces norepinephrine release and produces dose-dependent decreases in arterial blood pressure, heart rate, cardiac output, and norepinephrine levels [54, 55]. Moreover, dexmedetomidine participates in autonomic and cardiovascular regulation, which further contributes to its hemodynamic effects [56]. These mechanisms are supported by the study by Basar et al. [57] that reported that a single dose of dexmedetomidine significantly reduced both the intraoperative and postoperative heart rates. Therefore, caution is warranted when dexmedetomidine is administered to patients with hemodynamic instability or those receiving cardiac medications such as stimulants, β-blockers, or antiarrhythmic agents, as these may exacerbate the risk of bradycardia [58]. Notably, Gong et al. [59] suggested that dexmedetomidine-induced bradycardia and hypotension do not always occur concurrently and that bradycardia without accompanying cardiovascular symptoms may not require clinical intervention.

Although dexmedetomidine significantly increased the risk of bradycardia and hypotension, most of the included RCTs defined bradycardia as an HR < 50–60 bpm and hypotension as an SBP < 90 mmHg or > 20% below baseline. These events were generally mild and transient and responded to conservative management with atropine or ephedrine, and no study reported severe cardiac complications or prolonged hospitalization. Therefore, although the hemodynamic effects were statistically significant, their clinical relevance appears to be limited in stable patients under adequate monitoring. Future studies should quantify dose–response relationships, assess the need for vasopressor support, and examine whether dexmedetomidine use influences postanesthesia care units or hospital length of stay or increases cardiac event risk. Such evidence would help determine the threshold at which these hemodynamic effects become clinically important.

In clinical practice, dexmedetomidine can be considered an adjuvant to improve postoperative pain and reduce nausea and vomiting after shoulder arthroscopy. However, its use should be individualized, with careful monitoring of hemodynamic changes such as bradycardia and hypotension. Further large-scale trials are needed to define optimal dosing strategies that maximize efficacy while minimizing adverse events.

Implications

This meta-analysis revealed that dexmedetomidine effectively reduced postoperative pain and the incidence of postoperative nausea and vomiting (PONV) in patients who underwent shoulder arthroscopy, thereby enhancing postoperative recovery. It is particularly beneficial for patients at high risk of PONV and those requiring intraoperative hemodynamic stability. However, its association with increased risks of bradycardia and hypotension necessitates careful patient selection and continuous hemodynamic monitoring during use. Dexmedetomidine should be incorporated as part of a multimodal analgesia strategy in combination with regional nerve blocks and other analgesic modalities to maximize analgesic efficacy while minimizing adverse effects. Future research should focus on developing individualized administration protocols and optimizing dosage, timing, and coadministration strategies to achieve precise perioperative management. Overall, dexmedetomidine holds significant clinical value in shoulder arthroscopy but must be applied through a balanced evaluation of its benefits and potential risks to ensure maximal patient safety and outcomes.

Strengths

First, to the best of our knowledge, this is the first study to systematically evaluate the effects of dexmedetomidine in patients undergoing shoulder arthroscopy. It provides the most up-to-date evidence regarding the potential benefits of dexmedetomidine in this setting.

Second, a comprehensive and systematic literature search was conducted using both MeSH/Emtree terms and free-text keywords across three major databases—PubMed, the Cochrane Library, and Embase. A well-developed search strategy was employed to maximize the identification of eligible studies and minimize the risk of publication bias, thereby enhancing the reproducibility and credibility of the results.

Third, this study adhered strictly to the PRISMA guidelines and utilized the RoB 2 tool for risk of bias assessment, ensuring transparency, methodological rigor, and interdisciplinary relevance.

Finally, we employed a series of robust statistical methods to assess the stability of our findings, including sensitivity analyses and Egger’s test. The certainty of evidence for each outcome was further evaluated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system, which enhanced the clinical interpretability of the conclusions.

Limitations

Several limitations should be acknowledged. First, although bradycardia and hypotension were analyzed as safety outcomes, most included trials did not report standardized definitions or management strategies, and these events were often presented only as binary outcomes. This limits assessment of severity and warrants cautious clinical interpretation.

Second, substantial heterogeneity was observed in several outcomes. Heterogeneity was assessed for all outcomes, and sensitivity analyses were performed where appropriate, with particular attention to studies rated as having “some concerns” in Domain 5 (selection of the reported result). These studies contributed to increased heterogeneity only in selected outcomes (heart rate after extubation, postoperative hypotension, and pain at 6 h), while the overall pooled estimates remained largely unchanged, indicating that the main conclusions are robust.

Third, variability in dexmedetomidine dosing regimens, administration routes, anesthetic techniques, and outcome timing may have contributed to residual heterogeneity that could not be fully explained. In addition, most included studies were conducted in Asia, which may limit generalizability.

Future randomized controlled trials with standardized outcome definitions, detailed reporting of adverse events, and broader geographic representation are needed to further clarify the clinical effects of dexmedetomidine in shoulder arthroscopy.

Conclusion

The current meta-analysis suggests that, compared with control interventions, dexmedetomidine is frequently associated with reductions in postoperative pain and postoperative nausea and vomiting in patients undergoing shoulder arthroscopy. However, its use is also associated with a higher incidence of postoperative bradycardia and hypotension, warranting careful clinical consideration.

Importantly, the magnitude and consistency of these effects vary across studies, likely reflecting differences in clinical protocols such as dosing regimens, routes of administration, anesthetic techniques, and outcome assessment timing. Therefore, while dexmedetomidine may provide perioperative benefits in selected settings, the pooled estimates should be interpreted with caution. These findings highlight the potential role of dexmedetomidine within multimodal perioperative management, rather than supporting a uniform, standardized application across all shoulder arthroscopy patients.

Authors’ contributions

Study concept and design (XB, HY, YL); Acquisition of data (XB, HY); Analysis and interpretation of data (XB, HY, YL, XH); Drafting of the manuscript (XB, HY); Critical revision of the manuscript for important intellectual content (all authors); Study supervision (XB, HY, XH).

Funding

The authors declare that they have no funding disclosures.

Data availability

This is a summary design study. Data used for meta-analysis was extracted from previously published papers.

Declarations

Ethics approval and consent to participate

Not applicable.Consent for publication: Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.