Efficacy and safety of perioperative dexmedetomidine infusion in thyroid surgery: a systematic review and meta-analysis of randomized controlled trials Authors: Ebraheem Albazee, Abdulrahman Hamad Aldousari, Bader Qasem, Yousef Almutairi, Hussain Almaqtouf, Ahmad Aldhaen, Abdullah Almaqtouf, Salah Termos Published: 2026 ABSTRACT AbstractBackgroundDexmedetomidine is increasingly used to improve perioperative stability and analgesia in thyroid surgery; however, the overall evidence regarding its efficacy and safety remains uncertain. This study evaluates whether perioperative dexmedetomidine infusion is effective and safe in patients undergoing thyroid surgery.MethodsA systematic review and meta-analysis of randomized controlled trials (RCTs) was conducted using PubMed, Scopus, Web of Science, Embase, and CENTRAL databases. Eligible RCTs were assessed for quality using the RoB-2 tool. The evaluated outcomes included perioperative hemodynamic parameters (i.e., mean arterial pressure (MAP), heart rate (HR)), recovery characteristics (i.e., extubation, awakening times), postoperative pain intensity, rescue analgesic requirements, length of stay (i.e., hospital, post-anesthesia care unit (PACU)), and adverse events. Outcomes were pooled as mean differences (MDs), standardized mean differences (SMDs), or risk ratios (RRs) with 95% confidence intervals (CIs).ResultsA total of 10 RCTs involving 863 patients were analyzed. No significant differences were observed in anesthesia or surgical duration. Dexmedetomidine was associated with longer extubation and awakening times. Moreover, dexmedetomidine significantly reduced MAP and HR before, during, and after extubation. Postoperative pain scores were reduced at early postoperative time points (1–2, 3–4, 6–8, and 24 h), with reduced rescue analgesia requirements. No significant differences were found in PACU or hospital length of stay. Dexmedetomidine also reduced the incidence of cough and PONV. In terms of safety, dexmedetomidine increased the risk of hypotension, while the incidence of bradycardia was not significantly different between groups.ConclusionDexmedetomidine may improve intraoperative hemodynamic stability and reduce the incidence of emergence coughing and PONV. However, findings regarding its early postoperative analgesic benefits must be interpreted cautiously due to the low certainty of current evidence. These potential benefits should be carefully weighed against observed trade-offs, including mild delays in early clinical recovery and an increased risk of transient hypotension.PROSPERO registrationID: CRD420261308862.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12871-026-03944-3.Keywords: Dexmedetomidine, Hemodynamic stability, Hypotension, Meta-analysis, Pain, Thyroid surgery Introduction Thyroid surgery is a frequently performed procedure and is typically associated with mild to moderate postoperative pain [1, 2]. However, airway manipulation, tracheal intubation, and emergence from general anesthesia may provoke marked physiological responses, including hypertension, tachycardia, coughing, agitation, and discomfort [3]. These reactions may increase the risk of postoperative complications such as hemorrhage, delay recovery, and increase analgesic requirements [4]. Opioids remain the cornerstone of perioperative analgesia in thyroid surgery and are commonly used to attenuate stress responses [1, 5]. Nevertheless, opioid use is associated with adverse effects, including respiratory depression, postoperative nausea and vomiting, hyperalgesia, and delayed recovery [6]. Consequently, increasing attention has been directed toward opioid-sparing and multimodal anesthetic strategies aimed at optimizing analgesia and hemodynamic stability while minimizing opioid-related complications [7, 8]. Dexmedetomidine, a highly selective α2-adrenergic receptor agonist, has emerged as a promising anesthetic adjunct due to its sedative, anxiolytic, sympatholytic, and analgesic properties with minimal respiratory depression [9]. Several randomized controlled trials (RCTs) have evaluated perioperative dexmedetomidine in thyroid surgery, reporting potential benefits in postoperative pain control, attenuation of hemodynamic responses, reduction of emergence cough, and improved recovery profiles [10–19]. However, small sample sizes, variability in methodological quality, differences in dosing regimens, and heterogeneity in outcome assessment have limited the strength and generalizability of the evidence, with conclusions mainly dependent on individual studies [10–19]. Therefore, we conducted a systematic review and meta-analysis of RCTs to provide high-level evidence regarding the efficacy and safety of perioperative dexmedetomidine infusion in terms of perioperative hemodynamic parameters (i.e., mean arterial pressure (MAP) and heart rate (HR)), recovery characteristics (i.e., extubation and awakening times), postoperative pain intensity, rescue analgesic requirements, length of stay (i.e., hospital and post-anesthesia care unit (PACU)), and adverse events in patients undergoing thyroid surgery. We anticipate that the synthesized evidence from this study will inform clinical practice and guide decision-making among anesthesiologists and surgeons regarding the perioperative use of dexmedetomidine in thyroid surgery. Methods and materials Protocol registration and reporting standards The review protocol was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO) under registration number [CRD420261308862]. The study was conducted in accordance with the methodological guidance outlined in the Cochrane Handbook for Systematic Reviews of Interventions [20] and was prepared and reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Supplementary Material 1) [21]. Ethical approval was not required, as this study was based exclusively on data derived from previously published studies. Literature search and search strategy A comprehensive literature search was conducted in PubMed, Scopus, Web of Science, Embase, and the Cochrane Central Register of Controlled Trials (CENTRAL) from database inception through August 2025 and was updated in February 2026. The search strategy combined terms related to thyroid surgery with terms related to dexmedetomidine, including its generic name and trade names. The following keywords were used: (“Thyroid Surgery” OR “Thyroidectomy” OR “Thyroid Resection” OR “Thyroid Gland Removal”) AND (“Dexmedetomidine” OR “MPV1440” OR “Precedex” OR “Igalmi” OR “Sedadex” OR “Sileo” OR “Cepedex” OR “Dexdor” OR “Dexdomitor” OR “Hydrochloride”). No restrictions were applied regarding country of origin or publication status. The reference lists of all included studies were manually screened to identify additional relevant articles. Trial registries, including ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform (ICTRP), were also searched to capture ongoing or unpublished studies. Additional efforts were made to identify potentially eligible unpublished studies through sources such as ResearchGate. The full search strategies for each database are provided in Table S1. Authors of studies with incomplete or unclear data were contacted for clarification, as needed. Criteria for study eligibility The eligibility criteria were strictly defined according to the PICOS framework: Population (P): Adult patients (≥ 18 years) undergoing thyroid surgery (regardless of surgical type or thyroid pathology) under general anesthesia. Population (P): Adult patients (≥ 18 years) undergoing thyroid surgery (regardless of surgical type or thyroid pathology) under general anesthesia. Intervention (I): Perioperative dexmedetomidine infusion, irrespective of dosage or timing of administration. Intervention (I): Perioperative dexmedetomidine infusion, irrespective of dosage or timing of administration. Comparison (C): A control group receiving either a placebo or no dexmedetomidine. Comparison (C): A control group receiving either a placebo or no dexmedetomidine. Outcomes (O): Hemodynamic parameters (mean arterial pressure and heart rate), recovery characteristics (extubation and awakening times), postoperative pain intensity, rescue analgesic requirements, length of stay (hospital or post-anesthesia care unit), and perioperative adverse events (cough, hypotension, bradycardia, or postoperative nausea and vomiting). Outcomes (O): Hemodynamic parameters (mean arterial pressure and heart rate), recovery characteristics (extubation and awakening times), postoperative pain intensity, rescue analgesic requirements, length of stay (hospital or post-anesthesia care unit), and perioperative adverse events (cough, hypotension, bradycardia, or postoperative nausea and vomiting). Study Design (S): RCTs. Study Design (S): RCTs. Studies were excluded if they evaluated dexmedetomidine in combination with other investigational agents without an isolated effect, compared dexmedetomidine with other active pharmacologic interventions (i.e., metoprolol), were observational or non-randomized in design, involved animal models, were abstracts without full-text availability, were review articles, case reports, or were published in languages other than English. Study selection and screening process All identified records were imported into EndNote software, and duplicate citations were removed. The screening process consisted of two stages: an initial screening of titles and abstracts, followed by a full-text assessment of potentially eligible studies. Two reviewers (AHA & BQ) independently screened all records. The same independent process was applied during the risk of bias assessment (YA & HA) and data analysis (AhA & AbA). Any discrepancies were resolved through discussion with a third reviewer (EA) until consensus was reached. Data extraction and detailed outcomes A standardized data extraction form was used to systematically collect all relevant information from each included study. The extracted data included study identification details, trial registration number, country of origin, study design, total sample size (n), dexmedetomidine dosing regimen, timing of administration, anesthesia technique (i.e., total intravenous anesthesia (TIVA), balanced anesthesia), and type of surgical procedure. Baseline characteristics included mean age, sex distribution, American Society of Anesthesiologists (ASA) physical status classification, body weight, body mass index, and duration of follow-up. Our prespecified outcomes included duration of anesthesia [minutes] (defined as the time from anesthetic induction to discontinuation of anesthetic agents), duration of surgery [minutes] (from skin incision to complete skin closure), extubation time [minutes] (from discontinuation of anesthetics to removal of the endotracheal tube), and time to awakening [minutes] (from discontinuation of anesthetics to spontaneous eye-opening or response to verbal commands). Perioperative hemodynamic responses, including MAP (mmHg) and HR (bpm), were assessed at four predefined time points: (i) baseline, (ii) before extubation, (iii) during extubation, and (iv) after extubation. Additional outcomes included postoperative pain intensity measured using validated 10-point pain scales (Visual Analog Scale (VAS) or Numeric Rating Scale (NRS)) at multiple postoperative time points (i.e., 1–2 h, 3–4 h, 6–8 h, 12 h, and 24 h), incidence of rescue analgesic requirement [%] (defined as the proportion of patients requiring additional analgesia beyond the standard postoperative analgesic regimen), length of PACU stay [minutes] (defined as the time from admission to the PACU until discharge to the ward based on standard recovery criteria), and length of hospital stay [days] (defined as the time from surgery to hospital discharge). Safety outcomes included cough (any and severe), PONV, hypotension, and bradycardia. To account for minor methodological variations across the included trials, definitions of adverse events were accepted as reported by the original study authors. In general, hypotension was defined as a > 20% reduction from baseline blood pressure or the need for vasopressor support, while bradycardia was defined as a heart rate < 50–60 beats per minute or the need for anticholinergic intervention. Dichotomous outcomes were extracted as event and total numbers, while continuous outcomes were extracted as means with corresponding standard deviations. When studies reported medians with interquartile ranges or ranges, established conversion methods described by Wan et al. [22] were applied to estimate the mean and standard deviation. For trials reporting outcomes at multiple time points, data were extracted at each clinically relevant time point according to the predefined outcome framework adopted in this review. Methodological quality and certainty of evidence assessment The methodological quality of the included RCTs was assessed using the revised Cochrane Risk of Bias tool (RoB-2) [23], which evaluates five domains: randomization process, deviations from intended interventions, missing outcome data, outcome measurement, and selective reporting. Each domain was rated as low risk, some concerns, or high risk of bias, and an overall risk of bias judgment was assigned to each study using the tool’s recommended algorithm. In addition, the certainty of evidence for each outcome was evaluated using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach [24, 25]. The certainty of evidence was assessed across five domains: risk of bias, inconsistency, indirectness, imprecision, and publication bias. Based on these domains, the overall certainty of evidence for each outcome was categorized as high, moderate, low, or very low. Effect measure and meta-analysis All statistical analyses were performed using StataBE version 18. For continuous outcomes, pooled estimates were calculated using mean differences (MDs); standardized mean differences (SMDs) were applied when different measurement scales were used. Dichotomous outcomes were analyzed using risk ratios (RRs). All effect estimates were reported with corresponding 95% confidence intervals (CIs), and p-values ≤ 0.05 were considered statistically significant. Given the anticipated clinical heterogeneity across studies, a random-effects model was applied to all analyses. Statistical heterogeneity was assessed using the Chi-square test and the I² statistic, with substantial heterogeneity defined as I² ≥ 50% or a Chi-square p-value < 0.10 [20]. For outcomes with zero-event cells, a continuity correction of 0.5 was applied. Leave-one-out sensitivity analyses were conducted for outcomes with substantial heterogeneity to explore potential sources of variability and to assess the robustness of the pooled estimates. Prespecified subgroup analyses were performed according to anesthesia type (TIVA vs. balanced anesthesia), timing of dexmedetomidine administration (before the start vs. before the end of surgery), and dosing regimen (single vs. two doses). Publication bias was not assessed, as fewer than 10 RCTs were available for each outcome [26]. In studies with multiple intervention arms, each arm was treated as a separate comparison and analyzed against the same control group [20]. Results Summary of the systematic search The systematic search identified 576 records across all databases. After removal of 171 duplicate records, 405 unique records remained and were screened based on titles and abstracts. Of these, 365 records were excluded during the initial screening. Forty full-text articles were sought for retrieval, of which nine could not be retrieved. The remaining 31 reports were assessed for eligibility at the full-text level. During full-text assessment, 21 studies were excluded for failing to meet the predefined eligibility criteria, as detailed in Table S2. Ultimately, 10 RCTs [10–19] were included in the final qualitative and quantitative synthesis. The study selection process is illustrated in Fig. 1. The Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flow chart of the screening process Descriptive summary of the included RCTs and participants A total of 10 RCTs [10–19], including 863 patients (dexmedetomidine = 446; control = 417), were included in this meta-analysis. The included RCTs were conducted in four different countries; most of the studies were carried out in China (n = 6) [12, 13, 16–19], followed by South Korea (n = 2) [14, 15], while one study each was conducted in Iran [10] and India [11]. The pooled mean age was 45.5 ± 10.1 years. Females comprised the majority of participants (n = 643, 74.5%), while males accounted for 220 participants (25.5%). Dexmedetomidine dosing varied across the included studies. Continuous infusion regimens typically ranged from 0.4 to 0.6 µg/kg/h, and a loading dose of 0.5 µg/kg was the most frequently used. One study arm [18] used a lower dose of 0.2 µg/kg, while another [19] employed a higher loading dose of 1 µg/kg. Three RCTs [12, 13, 19] administered dexmedetomidine as both a loading dose followed by continuous infusion, whereas the remaining trials used either a single-dose regimen or continuous infusion alone. The overall summary characteristics of the included RCTs and participants are presented in Tables 1 and 2. Regarding anesthesia technique, six RCTs [10–12, 14, 15, 19] employed balanced anesthesia, while four [13, 16–18] used TIVA. The detailed surgical and anesthetic characteristics are provided in Table S3. Table 1Summary information of the included randomized controlled trialsStudy IDTrial Registration IDStudy DesignCountrySample Size (n)Dexmedetomidine RegimenControlTime of AdministrationDosesAnesthesiaAzar 2024 [10]IR20211209053338N1RCTIrann = 700.5 µg/kg infused over 10 min at the end of surgeryPlacebo (normal saline infused over 10 min at the end of surgery)Before the end of surgerySingle doseBalanced anesthesiaDjearadjane et al. 2019 [11]Not reportedRCTIndian = 540.5 µg/kg administered as a 10-minute IV infusion following thyroid gland removalNo dexmedetomidineBefore the end of surgerySingle doseBalanced anesthesiaGuo et al. 2022 [12]ChiCTR1900025761RCTChinan = 600.5 µg/kg administered as a 20 mL IV bolus over 10 min before induction of anesthesiaPlacebo (20 mL normal saline infused over 10 min before induction and continued until surgery completion)Before the start of surgerySingle doseBalanced anesthesiaHu et al. 2019 [13]ChiCTR1800017482RCTChinan = 1200.5 µg/kg IV bolus over 10 min before induction, followed by 0.4 µg/kg/h infusion until 30 min before surgery endPlacebo (20 mL normal saline infused over 10 min before induction and continued until surgery completion)Before the start of surgery and during surgeryTwo dosesTIVAKim et al. 2021 [14]NCT02412150RCTSouth Korean = 1390.6 µg/kg/h starting 15 min before the end of surgeryPlacebo (normal saline 15 min before the end of surgery)Before the end of surgerySingle doseBalanced anesthesiaLee et al. 2015 [15]Not reportedRCTSouth Korean = 1410.5 µg/kg infused over 10 min at the end of surgeryPlacebo (normal saline infused over 10 min at the end of surgery)Before the end of surgerySingle doseBalanced anesthesiaMa et al. 2017 [16]Not reportedRCTChinan = 500.5 µg/kg administered 30 min before the end of surgeryPlacebo (normal saline administered 30 min before the end of surgery)Before the end of surgerySingle doseTIVAShu et al. 2022 [17]ChiCTR2000038442RCTChinan = 800.5 µg/kg administered over 10 min before induction of anesthesia, and continuous infusion of 0.5 µg/kg/h until 30 min before the end of surgeryPlacebo (normal saline)Before the start of surgery and during surgeryTwo dosesTIVAWu et al. 2021 [18]ChiCTR1800018373RCTChinan = 89Group A: 0.5 µg/kg administered before skin closure over > 10 minGroup B: 0.2 µg/kg administered before skin closure over > 10 minPlacebo (normal saline 0.9%)Before the end of surgerySingle doseTIVAXing et al. 2021 [19]ChiCTR1900025222RCTChinan = 601 µg/kg IV loading dose over 10 min before induction, followed by a continuous infusion of 0.4 µg/kg/h until 30 min before the end of surgeryPlacebo (normal saline bolus for 10 min and continous infusion until 30 min before the end of surgery)Before the start of surgery and during surgeryTwo dosesBalanced anesthesiaAbbreviations: RCT randomized controlled trial, TIVA total intravenous anesthesia, IV intravenous Summary information of the included randomized controlled trials Group A: 0.5 µg/kg administered before skin closure over > 10 min Group B: 0.2 µg/kg administered before skin closure over > 10 min Abbreviations: RCT randomized controlled trial, TIVA total intravenous anesthesia, IV intravenous Table 2Baseline characteristics of the included participants and studiesStudy IDGroup N Age (years)Sex (n)[male/female]ASA (n)[I/II]Weight (kg)BMI (kg/m²)Pain ToolSurgical ProcedureFollow UpAzar 2024 [10]Dexmedetomidinen = 3550.7 ± 10.2[8/62]NRNRNRNRThyroidectomyDuring the recovery periodPlacebon = 3546.4 ± 12.3NRNRNRDjearadjane et al. 2019 [11]Dexmedetomidinen = 2743.7 ± 13.4[22/5][11/16]62.3 ± 12.8NRNRTotal thyroidectomy6 min after extubationControln = 2747.9 ± 13.5[18/9][14/13]65.2 ± 11.2NRGuo et al. 2022 [12]Dexmedetomidinen = 3044.5 ± 7[11/19][13/17]60 ± 5.2NR10-point VASThyroidectomy24 hPlacebon = 3045.5 ± 5.5[10/20][19/13]60.5 ± 8.3NRHu et al. 2019 [13]Dexmedetomidinen = 6047.6 ± 7.8[23/37][58/2]57.6 ± 5.723.2 ± 2.710-point VASThyroid surgery24 hPlacebon = 6049.3 ± 7.2[26/34][57/3]60.1 ± 6.423.8 ± 3.7Kim et al. 2021 [14]Dexmedetomidinen = 6944.2 ± 1.4[0/69][49/20]61.7 ± 7.824.1 ± 5.210-point NRSTotal thyroidectomy72 hPlacebon = 7045 ± 4.9[0/70][51/19]61.4 ± 6.423.8 ± 2.1Lee et al. 2015 [15]Dexmedetomidinen = 7045 ± 10[11/59][54/16]NR23.6 ± 310-point NRSUnilateral or total thyroidectomyUntil discharge from PACUPlacebon = 7146 ± 12[18/53][50/21]NR22.9 ± 3.6Ma et al. 2017 [16]Dexmedetomidinen = 2549.28 ± 9.33[12/13]NR61.6 ± 10.26NR10-point VASThyroid surgery72 hPlacebon = 2548.07 ± 9.19[13/12]NR62.8 ± 9.54NRShu et al. 2022 [17]Dexmedetomidinen = 4045.6 ± 11.9[12/28][30/10]63.1 ± 12.724.2 ± 5.810-point VASTotal thyroidectomy with neck dissection24 hPlacebon = 4046.2 ± 11.5[7/33][26/14]65.2 ± 1024.8 ± 4.6Wu et al. 2021 [18]Dexmedetomidine (A)n = 3037.1 ± 10.6[8/22][26/4]61 ± 1422.7 ± 5.610-point NRSThyroidectomy48 hDexmedetomidine (B)n = 3042.6 ± 11.7[11/19][25/5]65 ± 1524.2 ± 6.1Placebon = 2940.4 ± 10.8[9/20][23/6]59 ± 1221.9 ± 5.4Xing et al. 2021 [19]Dexmedetomidinen = 3040.1 ± 12[1/29][25/5]61.1 ± 13.223.6 ± 6.410-point VASThyroidectomy24 hPlacebon = 3041.9 ± 11.1[0/30][23/7]60.1 ± 10.222.8 ± 5.1Abbreviations: ASA American Society of Anesthesiologists, VAS Visual Analog Scale, NRS Numeric Rating Scale, BMI Body Mass Index, PACU Post-Anesthesia Care Unit, NR not reported Baseline characteristics of the included participants and studies Abbreviations: ASA American Society of Anesthesiologists, VAS Visual Analog Scale, NRS Numeric Rating Scale, BMI Body Mass Index, PACU Post-Anesthesia Care Unit, NR not reported Methodological quality and overall certainty of evidence Regarding methodological quality, six RCTs [12–15, 17, 18] demonstrated a low risk of bias, three [10, 11, 19] showed some concerns, and one [16] was judged as having a high risk of bias overall. In the D1 domain (randomization process), Ma et al. [16] and Xing et al. [19] did not provide sufficient details on randomization and allocation concealment. For D2 (deviations from intended interventions), Djeardjane et al. [11] did not use a placebo and did not report blinding, while Ma et al. [16] did not clarify whether the absence of blinding may have influenced the intervention. In D4 (measurement of outcomes), Xing et al. [19] reported HR and MAP using an unconventional format. For D5 (selection of reported results), Azar 2024 [10] and Ma et al. [16] raised concerns regarding potential selective reporting. Figure 2 presents the risk of bias graph and summary for the included RCTs. Risk of bias summary and graph for the included randomized trials. The upper panel displays the risk of bias assessment for each individual study, categorized as low (green), some concerns (yellow), or high (red). The lower panel summarizes the risk of bias across all included studies for each bias domain Additionally, the overall certainty of evidence for each outcome is summarized in Table S4. To clarify, the certainty of evidence across the evaluated outcomes ranged from moderate to very low. Outcomes such as anesthesia and surgery duration, perioperative hemodynamic parameters (at some time points), and the incidence of cough and postoperative nausea and vomiting were graded as moderate certainty. In contrast, recovery times, postoperative pain scores, and length of stay outcomes were downgraded to low or very low certainty. These downgrades were primarily driven by substantial statistical heterogeneity (I² > 50%) across the included trials, in some cases by risk of bias concerns, and, finally, by imprecision. Efficacy outcomes Duration of anesthesia (minutes) There was no significant difference between the dexmedetomidine and control groups in terms of the duration of anesthesia (n = 6 arms, MD = 4.56 min, 95% CI [− 2.41, 11.53], p = 0.2, moderate-certainty evidence). The heterogeneity across the included studies was low (I² = 35.11%), Fig. 3A. Forest plots of the meta-analysis showing (A) DOA (minutes), (B) DOS (minutes), (C) extubation time (minutes), and (D) time to awakening (minutes). MD = mean difference; CI = confidence interval, DOA = duration of anesthesia; DOS = duration of surgery Duration of surgery (minutes) There was no significant difference between the dexmedetomidine and control groups in terms of the duration of surgery (n = 5 arms, MD = − 0.87 min, 95% CI [− 5.13, 3.39], p = 0.69, moderate-certainty evidence). The heterogeneity across the included studies was low (I² = 0%), Fig. 3B. Extubation time (minutes) Dexmedetomidine was associated with a significantly prolonged extubation time compared with control (n = 6 arms, MD = 3.73, 95% CI [1.64, 5.82], p < 0.001, low-certainty evidence). The heterogeneity across the included studies was high (I² = 86.71%), Fig. 3C. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, Figure S1A. Time to awakening (minutes) Dexmedetomidine was associated with a significantly prolonged time to awakening compared with control (n = 7 arms, MD = 1.07, 95% CI [0.19, 1.94], p = 0.02, low-certainty evidence). The heterogeneity across the included studies was high (I² = 96.17%), Fig. 3D. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, except for Ma et al. [16], Figure S1B. Mean arterial pressure (mmHg) Baseline There was no significant difference between the dexmedetomidine and control groups (n = 7 arms, MD = − 0.7, 95% CI [− 4.42, 3.01], p = 0.71, low-certainty evidence). The heterogeneity across the included studies was high (I² = 86.31%), Fig. 4A. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, except for Kim et al. [14], Figure S2A. Forest plots of the meta-analysis showing the MD in MAP (mmHg) at (A) baseline, (B) before extubation, (C) during extubation, and (D) after extubation. MD = mean difference; CI = confidence interval; MAP = mean arterial pressure Before extubation Dexmedetomidine was associated with a significant reduction in MAP compared with control (n = 3 arms, MD = − 6.58, 95% CI [− 11.69, − 1.48], p = 0.01, low-certainty evidence). The heterogeneity across the included studies was high (I² = 66.40%), Fig. 4B. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, except for Ma et al. [16], Figure S2B. During extubation Dexmedetomidine was associated with a significant reduction in MAP compared with control (n = 7 arms, MD = − 11.22, 95% CI [− 15.16, − 7.29], p < 0.0001, moderate-certainty evidence). The heterogeneity across the included studies was high (I² = 73.18%), Fig. 4C. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, Figure S2C. After extubation Dexmedetomidine was associated with a significant reduction in MAP compared with control (n = 7 arms, MD = − 12.49, 95% CI [− 17.22, − 7.77], p < 0.0001, moderate-certainty evidence). The heterogeneity across the included studies was high (I² = 82.73%), Fig. 4D. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, Figure S2D. Heart rate (bpm) Baseline There was no significant difference between the dexmedetomidine and control groups (n = 5 arms, MD = 1.09, 95% CI [− 1.63, 3.82], p = 0.43, low-certainty evidence). The heterogeneity across the included studies was slightly high (I² = 64.91%), Fig. 5A. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, Figure S3A. Forest plots of the meta-analysis showing the MD in HR (bpm) at (A) baseline, (B) before extubation, (C) during extubation, and (D) after extubation. MD = mean difference; CI = confidence interval; HR = heart rate Before extubation Dexmedetomidine was associated with a significant reduction in HR compared with control (n = 3 arms, MD = − 9.91, 95% CI [− 14.46, − 5.36], p < 0.0001, low-certainty evidence). The heterogeneity across the included studies was high (I² = 72.60%), Fig. 5B. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, Figure S3B. During extubation Dexmedetomidine was associated with a significant reduction in HR compared with control (n = 6 arms, MD = − 13.50, 95% CI [− 18.15, − 8.85], p < 0.0001, moderate-certainty evidence). The heterogeneity across the included studies was high (I² = 83.96%), Fig. 5C. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, Figure S3C. After extubation Dexmedetomidine was associated with a significant reduction in HR compared with control (n = 6 arms, MD = -10.35, 95% CI [-16.21, -4.48], p < 0.0001, low-certainty evidence). The heterogeneity across the included studies was high (I² = 88.30%), Fig. 5D. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, Figure S3D. Postoperative pain score (10-point) At 1–2 h Dexmedetomidine was associated with a significant reduction in pain score compared with control (n = 4 arms, SMD = -1.27, 95% CI [-2.45, -0.09], p = 0.03, very low-certainty evidence). The heterogeneity across the included studies was high (I² = 95.51%), Figure S4A. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, except for Guo et al. [12] and Shu et al. [17], Figure S5A. At 3–4 h Dexmedetomidine was associated with a significant reduction in pain score compared with control (n = 4 arms, SMD = -1.73, 95% CI [-3.02, -0.44], p = 0.01, very low-certainty evidence). The heterogeneity across the included studies was high (I² = 95.66%), Figure S4B. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, except for Ma et al. [16], Figure S5B. At 6–8 h Dexmedetomidine was associated with a significant reduction in pain score compared with control (n = 7 arms, SMD = -1.41, 95% CI [-2.55, -0.27], p = 0.02, very low-certainty evidence). The heterogeneity across the included studies was high (I² = 96.71%), Figure S4C. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, Figure S5C. At 12 h There was no significant difference between the dexmedetomidine and control groups (n = 3 arms, SMD = -1.85, 95% CI [-4.34, 0.65], p = 0.15, very low-certainty evidence). The heterogeneity across the included studies was high (I² = 98.45%), Figure S4D. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, except for Hu et al. [13], Figure S5D. At 24 h Dexmedetomidine was associated with a significant reduction in pain score compared with control (n = 8 arms, SMD = -0.96, 95% CI [-1.77, -0.14], p = 0.02, low-certainty evidence). The heterogeneity across the included studies was high (I² = 95.53%), Figure S4E. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, except for Ma et al. [16], Figure S5E. Rescue analgesia requirement (%) Dexmedetomidine significantly reduced the requirement for rescue analgesia compared with the control group (n = 5 arms, RR = 0.44, 95% CI [0.22, 0.87], p = 0.02, low-certainty evidence). The heterogeneity among the included studies was moderate (I² = 58.33%), Figure S6A. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, except for Shu et al. [17] and Wu et al. (B) [18], Figure S7A. Length of PACU stay (minutes) There was no significant difference between the two groups in terms of length of PACU stay (n = 2 arms, MD = -0.87, 95% CI [-7.14, 5.4], p = 0.79, very low-certainty evidence). The heterogeneity among the included studies was substantial (I² = 97.45%), Figure S6B. Leave-one-out sensitivity analysis was not feasible, as this outcome included a limited number of pooled arms. Length of hospital stay (days) There was no significant difference between the two groups in terms of length of hospital stay (n = 3 arms, MD = -0.51, 95% CI [-1.56, 0.54], p = 0.34, very low-certainty evidence). The heterogeneity among the included studies was substantial (I² = 86.42%), Figure S6C. Leave-one-out sensitivity analysis was not feasible, as this outcome included a limited number of pooled arms. Safety outcomes Cough incidence (%) Any cough Dexmedetomidine significantly reduced the incidence of any cough (n = 5 arms, RR = 0.67, 95% CI [0.54, 0.83], p < 0.001, moderate-certainty evidence). The heterogeneity among the included studies was moderate (I² = 52.62%), Figure S8A. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, Figure S7B. Severe cough Dexmedetomidine significantly reduced the incidence of severe cough (n = 3 arms, RR = 0.33, 95% CI [0.19, 0.59], p < 0.001, moderate-certainty evidence). The heterogeneity among the included studies was nil (I² = 0%), Figure S8B. Post-operative nausea and vomiting (%) Dexmedetomidine significantly reduced the incidence of PONV (n = 5 arms, RR = 0.58, 95% CI [0.37, 0.90], p = 0.02, moderate-certainty evidence). The heterogeneity among the included studies was nil (I² = 0%), Figure S9A. Hypotension (%) Dexmedetomidine significantly increased the risk of hypotension (n = 7 arms, RR = 2.1, 95% CI [1.07, 4.15], p = 0.03, low-certainty evidence). The heterogeneity among the included studies was nil (I² = 0%), Figure S9B. Bradycardia (%) There was no significant difference between the two groups in terms of incidence of bradycardia (n = 7 arms, RR = 3.26, 95% CI [0.97, 10.92], p = 0.06, low-certainty evidence). The heterogeneity among the included studies was moderate (I² = 61.55%), Figure S9C. Leave-one-out sensitivity analysis confirmed the robustness of the results after excluding each study once, except for Xing et al. [19] and Wu et al. [18], Figure S10. Subgroup meta-analysis The detailed subgroup analysis based on type of anesthesia (balanced anesthesia versus TIVA), timing of administration (before the start of surgery versus before the start and during surgery versus before the end of surgery), and dosing (single dose versus two doses) are presented in Figure S11A–Y. Overall, subgroup analysis revealed no significant differences between subgroups across most endpoints (p > 0.05). Exceptions were observed for HR during extubation, where the subgroups receiving dexmedetomidine before the end of surgery and before the start plus during surgery showed a greater reduction in HR compared with administration before the start of surgery (p < 0.001). For HR after extubation, the TIVA subgroup showed a greater reduction in HR compared with the balanced anesthesia subgroup (p = 0.04). Regarding postoperative pain at 3–4 h, the TIVA subgroup demonstrated greater pain reduction compared with balanced anesthesia (p = 0.03), and administration before the end of surgery showed greater pain reduction than other timing subgroups (p < 0.001). Discussion This systematic review and meta-analysis examined the efficacy and safety of dexmedetomidine infusion among patients undergoing thyroid surgery. A total of 10 RCTs involving 863 patients were analyzed. The pooled analysis showed that perioperative dexmedetomidine use did not prolong anesthesia or surgery duration; however, during the recovery period, it was associated with a slight prolongation of extubation time and time to awakening. Regarding hemodynamics, perioperative dexmedetomidine significantly reduced MAP and HR before, during, and after extubation. Furthermore, postoperative pain scores were reduced at early postoperative time points (1–2, 3–4, 6–8, and 24 h), resulting in lower rescue analgesia requirements. However, these analgesic findings must be interpreted with strict clinical caution. The large effect sizes observed at early time points, the very high heterogeneity, and the very low GRADE certainty suggest that the true clinical magnitude of this pain reduction remains uncertain and may be inflated by methodological differences across the small number of included trials. Also, no significant differences were observed in PACU or hospital length of stay. Dexmedetomidine also reduced the incidence of cough (i.e., any and severe) and PONV. On the other hand, dexmedetomidine increased the risk of hypotension but did not increase the risk of bradycardia. Our meta-analysis findings are in line with dexmedetomidine’s mechanisms of action. The robust sympatholytic properties of dexmedetomidine directly explain the observed undermining of hemodynamic surges; by reducing circulating plasma catecholamines, dexmedetomidine prevents the severe tachycardia and hypertension typically provoked by the endotracheal tube during the extubation sequence [27, 28]. Additionally, the centrally mediated analgesic and sedative properties are directly responsible for the deep opioid-sparing analgesic effects and the suppression of emergence coughing. By ensuring the patient remains in a calm, analgesic state during critical awakening, dexmedetomidine elevates the threshold for stimulation of the rapidly adapting irritant receptors in the tracheal lining [29]. This pharmacological intervention effectively eliminates violent coughing spasms, thereby safeguarding the fragile surgical field from sudden increases in venous pressure and mechanically protecting the hemostasis achieved during the operation [30]. When comparing the pooled findings on hemodynamics and recovery with the existing literature, prolonged extubation and awakening times are consistent with the established pharmacokinetic half-life and sedative profile of α2-agonists [31]. However, the clinical significance of delayed awakening should be highlighted. Pooled analysis revealed that the modest delay in the awakening did not translate into a prolonged length of stay in the post-anesthesia care unit or an extended overall hospital stay. This suggests that although the initial emergence period is intentionally slowed to ensure a seamless transition, the main recovery remains unaffected. A smooth, hemodynamically stable emergence that prevents coughing and bleeding ultimately facilitates timely discharge criteria, proving highly compatible with fast-track ambulatory thyroid surgery protocols [32]. Regarding analgesia, the findings strongly validate the current clinical shift toward multimodal, opioid-sparing anesthesia. Opioids, while traditionally the cornerstone of perioperative pain management, carry a heavy burden of adverse effects that actively delay surgical recovery, including respiratory depression, postoperative ileus, opioid-induced hyperalgesia, and an elevated risk of postoperative nausea and vomiting [33]. Our analysis came in this line, demonstrating a significant reduction in postoperative pain scores and a decreased reliance on rescue opioid analgesia in the dexmedetomidine cohort. Still, the significantly lower incidence of PONV observed in our study is likely not a direct antiemetic effect of dexmedetomidine alone, but an indirect clinical benefit driven by this opioid-sparing effect. By replacing emetogenic opioids with dexmedetomidine’s non-emetic analgesia, practitioners can achieve better pain management while facilitating faster gastrointestinal recovery and reducing opioid-induced PONV [34]. Moreover, subgroup analyses investigating the type of anesthesia and the timing of administration yielded further clinical insights. The total intravenous anesthesia subgroup demonstrated superior outcomes in some outcomes compared to the balanced inhalational anesthesia subgroup. The total intravenous anesthesia subgroup showed greater reductions in heart rate and enhanced pain mitigation at 3–4 h. This effect is likely due to dexmedetomidine’s ability to specifically alleviate remifentanil-induced post-infusion hyperalgesia, as remifentanil frequently induces a paradoxical hyperalgesic state upon cessation, driven by the acute up-regulation of N-methyl-D-aspartate receptors in the central nervous system [35]. Wu et al. confirmed this interaction, demonstrating that dexmedetomidine significantly elevated mechanical pain thresholds and prevented secondary hyperalgesia following high-dose remifentanil [18]. Additionally, the timing of administration affected clinical outcomes. Delivering the infusion before the end of surgery, as used by Kim et al. [14] and Azar [10], proved superior for immediate pain reduction and the targeted suppression of emergence agitation during the critical, high-risk extubation window. Despite these significant clinical benefits, dexmedetomidine significantly increased the risk of hypotension. By inhibiting central sympathetic outflow, dexmedetomidine removes the compensatory catecholamine drive that often maintains blood pressure during the surgery [36]. Still, regarding bradycardia, the pooled meta-analysis showed that although the overall incidence was higher in the dexmedetomidine group, the difference did not reach statistical significance. Nevertheless, close clinical monitoring should be prioritized, as Shu et al. [17] reported that multiple patients had transient bradycardia that required active pharmacological reversal with anticholinergic agents. This systematic review and meta-analysis has several methodological strengths. The study utilized a highly rigorous methodology, featuring a comprehensive literature search strategy deployed across five major biomedical databases to ensure that a broad, unbiased spectrum of scientific literature was captured. Also, adherence to and application of the PRISMA reporting criteria and GRADE approach provide transparent and reliable reporting of the synthesized evidence, allowing clinical specialists to appropriately weigh the validity of the pooled findings before implementing them in clinical practice. Despite these methodological strengths, some limitations within the evidence base must be acknowledged. First, some outcomes showed substantial statistical heterogeneity, especially in continuous clinical outcomes, such as total extubation time, time to patient awakening, and postoperative pain scores. Second, this statistical heterogeneity mirrors the clinical heterogeneity observed among the included trials, especially regarding dosing protocols, as the regimens evaluated across the 10 studies differed significantly in both total dosage and administration strategy. To clarify, regimens used ranged from solitary bolus doses administered at various surgical phases to heavy loading doses followed by prolonged continuous infusions. However, while the subgroup analyses generally indicated consistent overall pharmacological efficacy across single- and two-dose regimens, the lack of a standardized protocol currently precludes the identification of a single, optimal dosing regimen for routine thyroidectomy practice. Third, this heterogeneity might have stemmed from other clinical differences across the included trials, such as the extent of surgical resection (lobectomy versus total thyroidectomy with neck dissection), institutional pain management protocols, and the inherent subjectivity of pain scales. Fourth, 4 out of 10 trials raised concerns about bias. Fifth, the aforementioned limitations were reflected in the GRADE certainty of evidence assessment, resulting in low or very low certainty and warranting careful interpretation. Finally, because fewer than 10 trials were available for each evaluated outcome, we were unable to formally assess publication bias using visual funnel plots or statistical tests. The routine perioperative use of dexmedetomidine in thyroid surgery should be guided by rigorous, individualized patient selection. Patients presenting with pre-existing cardiovascular diseases, advanced heart block, severe ventricular dysfunction, or those on negative chronotropic medications such as beta-blockers require highly tailored dosing regimens, dose reductions, or alternative multimodal strategies [37]. Also, vigilant intraoperative hemodynamic monitoring is a must, and practitioners must be continually prepared to promptly intervene with anticholinergics, fluid boluses, or vasopressors if clinically significant cardiovascular depression manifests. Finally, for the hemodynamically stable patients, the benefit of a smooth, cough-free emergence that protects against catastrophic neck hematoma generally outweighs the easily identifiable, predictable, and rapidly reversible risks of transient hypotension. Accordingly, future research should prioritize large-scale, multicenter RCTs that directly compare bolus-only administration with continuous infusion strategies, as determining the optimal dosing, concentration, and timing will enable anesthesiologists to maximize airway protection while fully mitigating cardiovascular depression flawlessly. Conclusion Perioperative dexmedetomidine infusion may serve as a beneficial pharmacological adjunct for enhancing the quality of recovery in patients undergoing thyroidectomy. While current evidence suggests it is associated with improved intraoperative hemodynamic stability, reduced emergence coughing, and a lower incidence of PONV, findings regarding its early postoperative analgesia must be interpreted cautiously due to the low certainty of the underlying evidence. These potential perioperative benefits can offer clinical value. Still, they must be carefully weighed against the observed trade-offs, including mild delays in early recovery and an increased risk of transient hypotension. Supplementary Information Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Acknowledgements None. Research involving human participants and/or animals Not applicable. Authors’ contributions Ebraheem Albazee contributed to study conception, study design, data collection, data analysis, write up of original draft of manuscript, and review of manuscript for editorial and intellectual contents. Abdulrahman Hamad Aldousari , Bader Qasem, Yousef Almutairi , Hussain Almaqtouf, Ahmad Aldhaen, and Abdullah Almaqtouf contributed to literature review, data collection, and review of manuscript for editorial and intellectual contents. Salah Termos contributed to supervision and review of manuscript for editorial and intellectual contents. All authors read and approved the final draft of manuscript. Funding None. Data availability All data are available within the manuscript and can be obtained from the corresponding author upon a reasonable request. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Data Availability Statement All data are available within the manuscript and can be obtained from the corresponding author upon a reasonable request.