Intraoperative dexmedetomidine for prevention of postoperative delirium in pediatric patients after orthopedic surgery: a single-center retrospective study

Abstract

Purpose

To evaluate the effectiveness of intraoperative dexmedetomidine in preventing postoperative delirium (POD) in pediatric patients undergoing lower extremity orthopedic surgery.

Methods

This retrospective study included pediatric patients aged 3–12 years who underwent lower extremity orthopedic surgery and received single-shot caudal block combined with intravenous patient-controlled analgesia (IV-PCA) from January 2021 to December 2023. Patients were divided into two groups according to intraoperative adjuvant medication: dexmedetomidine (1 µg/kg) or sufentanil (0.1 µg/kg). Propensity score matching (PSM) was used to balance baseline characteristics. The primary outcome was the incidence of POD, assessed by the Pediatric Anesthesia Emergence Delirium (PAED) scale. Secondary outcomes included postoperative pain scores (r-FLACC scale) and adverse events.

Results

After PSM, 128 patients (64 pairs) were analyzed. The incidence of POD was significantly lower in the dexmedetomidine group compared to the sufentanil group (14.1% vs. 34.4%, p = 0.01). Multivariate logistic regression revealed that intraoperative dexmedetomidine was independently associated with a reduced risk of POD (OR = 0.29, 95% CI: 0.12–0.74, p < 0.01). Dexmedetomidine also resulted in lower r-FLACC pain scores in the PACU and during the first 6 h postoperatively (all p < 0.01). No significant differences in hemodynamic changes or adverse events were observed between the groups.

Conclusion

Intraoperative dexmedetomidine significantly reduces the incidence of POD and improves early postoperative pain control in pediatric patients undergoing lower extremity orthopedic surgery, without increasing adverse events.

Introduction

Postoperative delirium (POD) is a serious but often overlooked complication in children after surgery [1]. This transient neuropsychiatric disorder may prolong hospital stays, change behavior, and impair cognition [2]. In pediatric populations, POD may manifest as agitation, sleep disturbances, emotional dysregulation, and challenges in social interaction and school reintegration. These consequences are particularly concerning given the vulnerability of the developing brain. Studies have shown that POD affects 20–30% of pediatric patients after major surgeries [3–5], leading to longer hospital stays, behavioral disturbances, and heightened parental stress. Despite heightened clinical awareness, effective preventive strategies for POD in children remain limited, particularly in the context of orthopedic procedures, which often involve prolonged operative times and significant postoperative pain.

For orthopedic surgery involving lower extremities, effective postoperative pain management is crucial. At our institution, single-shot caudal block combined with intravenous patient-controlled analgesia (IV-PCA) is commonly used to achieve this goal. While opioids such as sufentanil are frequently employed as adjuvants to enhance analgesia, their side effects—including respiratory depression, cognitive impairment, and increased POD risk—pose challenges, particularly in children with heightened opioid sensitivity [6, 7].

Dexmedetomidine, an alpha-2 adrenergic agonist, has gained increasing attention in pediatric anesthesia due to its unique pharmacological profile, which includes sedative, analgesic, and anxiolytic properties. Unlike opioids, it does not cause respiratory depression and may offer neuroprotection. It induces a sleep-like state by acting on the locus coeruleus, mimicking natural nonrapid eye movement (non-REM) sleep rather than suppressing cortical activity like GABAergic agents such as benzodiazepines [8, 9]. This more physiological sedation may explain its lower risk of POD and better cognitive outcomes. In adults, a meta-analysis showed dexmedetomidine reduced POD by 65% [10]. However, its role in pediatric patients, particularly in orthopedic surgery, has not been fully elucidated. Given the unique developmental physiology and pharmacodynamics in children, extrapolating adult findings may not be appropriate. Furthermore, existing pediatric studies often focus on non-orthopedic procedures, highlighting a critical gap in knowledge.

Orthopedic surgery was chosen for this study because it is commonly performed in pediatric patients and is frequently associated with prolonged surgical time, substantial postoperative pain, and higher emotional stress—all of which are risk factors for postoperative delirium and behavioral disturbances. Moreover, dexmedetomidine and sufentanil are both routinely used during these procedures at our institution, providing a consistent clinical context for comparison. Therefore, this study aims to evaluate the effectiveness of intraoperative dexmedetomidine in preventing POD in pediatric patients undergoing orthopedic surgery.

Methods

From January 2021 to December 2023, patients aged 3 to 12 years who underwent orthopedic surgery of the lower extremities and received single-shot caudal block combined with IV-PCA at our hospital were included. Exclusion criteria were as follows: (1) known or suspected coagulopathy, (2) cognitive impairment or developmental delay diagnosed prior to surgery, (3) pre-existing neurological or psychiatric disorders (e.g., epilepsy, autism spectrum disorder, attention deficit/hyperactivity disorder [ADHD]), and (4) incomplete clinical data.

Anesthetic protocol and group allocation

Standard intraoperative monitoring was applied. Anesthesia induction was achieved with propofol (1–2 mg/kg) and cisatracurium (0.1–0.15 mg/kg). Maintenance was carried out using sevoflurane (2–3% in 50% oxygen/air mixture) and remifentanil infusion (0.05–0.1 µg/kg/min), titrated to maintain both hemodynamic stability and a BIS score of 40–60. Under sterile conditions, experienced pediatric anesthesiologists performed a single-shot caudal block using the standard landmark technique. The injection consisted of 0.2% ropivacaine (1 mL/kg; maximum dose 30 mL) administered prior to surgical incision.

Approximately 30 min before the end of surgery, patients received either dexmedetomidine (1 µg/kg) or sufentanil (0.1 µg/kg), with the choice determined by the attending anesthesiologist based on individual clinical judgment, patient characteristics (such as age, baseline hemodynamics, and potential for agitation), and institutional practice patterns at the time. These decisions were recorded in the electronic medical records. The intraoperative administration of dexmedetomidine or sufentanil was intended to serve as a single-dose adjunct for smooth emergence and early postoperative comfort. Sufentanil was selected as the comparator because it is a standard intraoperative opioid at our institution, widely used to facilitate emergence and early postoperative analgesia in pediatric orthopedic surgeries.

Remifentanil infusion was then discontinued, and a uniform postoperative IV-PCA protocol with sufentanil was initiated for all patients, regardless of their intraoperative assignment. The IV-PCA settings were standardized across groups, consisting of a 0.5 µg/kg bolus dose, a 15-minute lockout interval, and a continuous infusion of 0.02 µg/kg/h sufentanil for 48 h postoperatively. This standardized postoperative analgesia ensured consistent pain management across both groups and minimized potential confounding from opioid variability.

This multimodal analgesic approach, combining a single intraoperative dose of either dexmedetomidine or sufentanil, caudal ropivacaine block, and standardized postoperative IV PCA sufentanil, aims to optimize pain control while minimizing opioid exposure. The PCA opioid dose was carefully controlled and well within safe limits, with no observed excessive sedation or respiratory complications.

Primary and secondary outcomes

The primary outcomes of this study were the incidences of POD, assessed using validated pediatric-specific neurocognitive scales. POD was evaluated using the Pediatric Anesthesia Emergence Delirium (PAED) Scale, a validated tool specifically designed for assessing emergence delirium in pediatric patients. The PAED scale comprises five behavioral items: eye contact, purposeful actions, awareness of surroundings, restlessness, and inconsolability. Each item is scored on a scale from 0 to 4, resulting in a total score ranging from 0 to 20. A PAED score of > 12 was considered indicative of emergence delirium. The PAED scale has been validated for use in children and is particularly suited for the immediate postoperative setting due to its sensitivity to behavioral changes commonly seen in emergence delirium [11]. PAED scores were extracted from anesthesia and nursing records documented in the post-anesthesia care unit (PACU) at 10-min intervals for the first 40 min after awakening from anesthesia, as this period represents the peak time window for emergence delirium.

Secondary outcomes were postoperative pain intensity and adverse events. Postoperative pain was assessed using the revised Face, Legs, Activity, Cry, and Consolability (r-FLACC) pain scale, a validated behavioral pain assessment tool widely used in children who are unable to reliably self-report pain. The r-FLACC scale evaluates five behavioral domains: facial expression, leg movement, activity level, crying, and consolability, each scored from 0 to 2, resulting in a total score ranging from 0 (no pain) to 10 (severe pain). Trained PACU nurses assessed and documented r-FLACC scores at predetermined postoperative time points, allowing for objective and consistent pain evaluation across patients.

Statistical analysis

To reduce the potential bias caused by baseline differences between dexmedetomidine and sufentanil groups, propensity score matching (PSM) was performed. Propensity scores were calculated using a logistic regression model that included age, gender, BMI, and type of surgery as covariates. Patients in the dexmedetomidine group were matched 1:1 with those in the sufentanil group based on their propensity scores, using a nearest-neighbor matching algorithm without replacement and a caliper width of 0.2 of the standard deviation of the logit of the propensity score. Balance between the matched groups was assessed using standardized mean differences, with a threshold of < 0.1 indicating adequate balance.

Continuous variables were expressed as mean ± standard deviation (SD) or median (IQR) based on normality (Shapiro-Wilk test), and categorical variables (e.g., POD incidence) as frequencies (%). Significant differences were assessed using Student’s t-tests for continuous data and Chi-square tests for categorical data. Multivariate logistic regression analysis was performed to explore the association of intraoperative dexmedetomidine with POD, after controlling for potential confounding factors.

Statistical significance was defined as two-tailed p < 0.05. Statistical analysis was performed using the SPSS 22.0 for Windows (SPSS, IBM, USA).

Results

A total of 343 patients were initially screened. After excluding 47 patients, 296 were included in the study, with 112 receiving dexmedetomidine and 184 treated with sufentanil. Following PSM, 64 matched pairs were generated, resulting in a final cohort of 128 patients for analysis (Fig. 1).

Fig. 1.

Flowchart of patient selection and enrollment. Of 343 initially screened pediatric patients undergoing lower extremity orthopedic surgery, 47 were excluded (reasons specified). After propensity score matching (PSM), 64 matched pairs (n = 128) were included in the final analysis

Table 1 presents the demographic characteristics and operative data before and after PSM. Before matching, there were significant differences in age and body mass index (BMI) (all p < 0.01). After PSM, all matching variables were well balanced (all p > 0.05), and no significant differences were observed in other perioperative parameters including extubation time, emergence time, intraoperative blood loss, or PACU duration, indicating comparable intraoperative and early postoperative conditions between the matched cohorts.

Table 1.

Demographic characteristics and operative data

	Pre-PSM			Post-PSM
	Dexmedetomidine group (n = 112)	Sufentanil group (n = 184)	p value	Dexmedetomidine group (n = 64)	Sufentanil group (n = 64)	p value
Age	8.3 ± 2.8	9.3 ± 2.9	< 0.01	8.8 ± 2.9	9.1 ± 2.8	0.56
Gender			0.09			0.86
Male	62 (55.4%)	121 (65.8%)		37 (57.8%)	38 (59.4%)
Female	50 (44.6%)	63 (34.2%)		27 (42.2%)	26 (40.6%)
BMI, kg/m2	16.2 ± 2.2	17.4 ± 2.2	< 0.01	16.8 ± 2.4	16.9 ± 2.4	0.90
HR, beats/min	97.0 ± 13.6	93.1 ± 10.5	< 0.01	96.1 ± 8.3	94.0 ± 9.0	0.17
MAP, mm/Hg	72.9 ± 11.3	68.4 ± 14.5	< 0.01	71.4 ± 5.8	70.0 ± 6.6	0.20
Type of surgery			0.22			0.35
Femur Surgery	52 (46.4%)	101 (54.9%)		32 (50.0%)	36 (56.3%)
Tibia and Fibula Surgery	27 (24.1%)	41 (22.3%)		16 (25.0%)	15 (23.4%)
Foot Surgery	17 (15.2%)	23 (12.5%)		9 (14.1%)	7 (10.9%)
Knee Surgery	7 (6.3%)	14 (7.6%)		2 (3.1%)	5 (7.8%)
Others	9 (8.0%)	5 (2.7%)		5 (7.8%)	1 (1.6%)
Duration of anesthesia, min	344.3 ± 65.5	337.8 ± 67.4	0.42	344.0 ± 75.1	330.8 ± 70.1	0.31
Duration of operation, min	275.4 ± 52.4	270.2 ± 53.9	0.42	275.1 ± 60.1	264.7 ± 56.1	0.31
Extubation time, min	9.3 ± 1.8	8.9 ± 1.9	0.08	9.2 ± 1.9	8.7 ± 1.9	0.12
Intraoperative blood loss, mL	251.4 ± 59.7	263.0 ± 51.9	0.08	246.1 ± 76.0	250.3 ± 75.5	0.76
Emergence time, min	10.1 ± 2.8	9.7 ± 2.6	0.17	10.0 ± 3.0	9.4 ± 3.2	0.28
Duration of PACU stay, min	51.6 ± 13.7	49.9 ± 13.1	0.29	49.0 ± 15.9	52.1 ± 14.2	0.24

BMI Body Mass Index, HR Heart Rate, MAP Mean Arterial Pressure, PACU Post-Anesthesia Care Unit, PSM Propensity-score match

Dexmedetomidine showed lower POD rates than sufentanil (14.1% vs. 34.4%, p = 0.01). Multivariate logistic regression analysis confirmed that the use of dexmedetomidine was associated with a reduced risk of POD (OR = 0.29, 95% CI: 0.12–0.74, p < 0.01) (Table 2).

Table 2.

Multivariate logistic regression for exploring the association of Dexmedetomidine with the risk of postoperative delirium

	POD	Non-POD	p	Multivariate logistic regression
				OR	95% CI	p
Dexmedetomidine group	9 (14.1%)	55 (85.9%)	0.01	0.29	0.12–0.74	< 0.01
Sufentanil group	22 (34.4%)	42 (65.6%)		Ref.

In terms of secondary outcomes, the dexmedetomidine group had lower r-FLACC pain scores both in the PACU (1 [0–2] vs. 2 [1–4], p < 0.01) and at 1–6 h postoperatively (0 [0–1] vs. 1 [0–3], p < 0.01) (Table 3). These lower scores reflect reduced postoperative pain and improved early analgesia in the dexmedetomidine group.

Table 3.

Comparison of revised face, legs, activity, cry, consolability (r-FLACC) scores between Dexmedetomidine and sufentanil groups

	Dexmedetomidine group (n = 64)	Sufentanil group (n = 64)	p value
r-FLACC score
PACU	1 (0–2)	2 (1–4)	< 0.01
1–6 h	0 (0–1)	1 (0–3)	< 0.01
6–12 h	3 (1–4)	3 (1–4)	0.51
12–24 h	2 (1–4)	2 (1–3)	0.27
24–48 h	1 (0–3)	1 (1–3)	0.64

Although not statistically significant, a gradual decrease in heart rate (HR) was observed in the dexmedetomidine group, compared to a slight increase in the sufentanil group (Fig. 2). Mean arterial pressure (MAP) was also comparable between the two groups throughout the intraoperative periods (Fig. 2). There were no significant differences in adverse events between the two groups (all p > 0.05; Table 4). Specifically, the incidences of bradycardia, hypotension, and respiratory depression were low and comparable, likely due to moderate intraoperative dosing and careful monitoring. Other events such as emergence agitation, nausea and vomiting, urinary retention, oversedation, and pruritus also occurred at similar rates, supporting the overall safety of dexmedetomidine in this setting.

Fig. 2.

Hemodynamic changes after administration of the study drug

Table 4.

Comparison of adverse events between Dexmedetomidine and sufentanil groups

	Dexmedetomidine group (n = 64)	Sufentanil group (n = 64)	p value
Emergence agitation	12 (18.8%)	8 (12.5%)	0.47
Nausea and vomiting	14 (21.9%)	18 (28.1%)	0.54
Urinary retention	6 (9.4%)	10 (15.6%)	0.42
Respiratory depression	0	0	-
Bradycardia	0	0	-
Hypotension	0	0	-
Oversedation	0	0	-
Pruritus	0	0	-

Discussion

In this retrospective, propensity score-matched study, we found that intraoperative administration of dexmedetomidine significantly reduced the incidence of POD in pediatric patients undergoing lower extremity orthopedic surgery compared to sufentanil. Additionally, dexmedetomidine improved early postoperative pain control without increasing the incidence of adverse events. These findings suggest that dexmedetomidine is an effective and safe adjuvant in pediatric orthopedic anesthesia.

Consistent with accumulating evidence [12, 13], our findings support dexmedetomidine’s efficacy in reducing POD across both pediatric and adult surgical populations. For instance, Liao et al. [14] reported that intranasal dexmedetomidine premedication significantly reduced the risk of emergence delirium in children undergoing tonsillectomy and/or adenoidectomy under sevoflurane anesthesia (relative risk [RR] = 0.25, 95% CI: 0.11–0.57, p < 0.001). Similarly, Huang et al. [15] compared the effects of dexmedetomidine and propofol on emergence delirium in children undergoing cleft palate surgery with sevoflurane-based anesthesia. Their results revealed a markedly lower incidence of emergence delirium in the dexmedetomidine group (20.1%) compared to the propofol (58.6%) and saline (85.7%) groups (p < 0.05). Furthermore, dexmedetomidine was associated with significantly lower pain scores, underscoring its dual benefits as both a sedative and analgesic agent. These findings align with our results and highlight dexmedetomidine’s superiority over traditional sedatives and opioids in mitigating POD and improving postoperative outcomes. Our study extends these findings to pediatric orthopedic surgery, a domain previously underexplored.

While dexmedetomidine has shown efficacy in reducing POD in adults [16], its effects in children may differ due to developmental pharmacodynamics. Pediatric patients exhibit higher clearance rates and shorter elimination half-lives for dexmedetomidine, potentially requiring weight-adjusted dosing to achieve optimal effects [17]. Additionally, the developing brain’s susceptibility to anesthesia-induced neurotoxicity may amplify dexmedetomidine’s benefits. Unlike GABAergic agents linked to neuroapoptosis in preclinical models [18], dexmedetomidine’s alpha-2 agonism preserves synaptic plasticity, which may be critical in mitigating delirium risk in children [19]. Furthermore, pediatric POD is often triggered by pain and anxiety—factors more salient than in adults—highlighting dexmedetomidine’s dual advantage as an anxiolytic and analgesic in this population.

Dexmedetomidine’s ability to reduce POD may be attributed to its sedative, analgesic, and neuroprotective properties. As a selective alpha-2 adrenergic agonist, dexmedetomidine modulates the locus coeruleus, a key structure in regulating arousal and cognition [20]. Unlike opioids and benzodiazepines, dexmedetomidine preserves natural sleep-like sedation, minimizing abrupt emergence from anesthesia, a known risk factor for POD [21]. Furthermore, its anti-inflammatory effects and potential neuroprotective properties may attenuate anesthesia-induced neurotoxicity, particularly in the developing pediatric brain [22]. In contrast, opioids like sufentanil may contribute to POD through mechanisms such as µ-opioid receptor overstimulation, disruption of cholinergic neurotransmission, and increased neuroinflammation [23]. Furthermore, sufentanil may occasionally induce opioid-related hyperalgesia or paradoxical agitation, potentially leading to higher early postoperative pain scores [24]. Although both dexmedetomidine and sufentanil were administered via continuous infusion until the end of surgery—ensuring comparable plasma concentrations at emergence—the dexmedetomidine group exhibited lower FLACC scores. These results are consistent with previous pediatric studies [25]. This likely reflects dexmedetomidine’s unique pharmacodynamic profile, including inhibition of nociceptive transmission in the spinal cord and reduction of anxiety and agitation, which can influence behavioral pain assessments such as FLACC [26]. Given that all patients received the same regional anesthesia (caudal block) and standardized postoperative IV-PCA with sufentanil, the lower FLACC scores in the dexmedetomidine group likely represent an additive sedative and anxiolytic benefit beyond baseline analgesia.

In our study, a single dose of 1 µg/kg dexmedetomidine was administered approximately 30 min before the end of surgery. This dosage was chosen based on several previous dose-finding studies in pediatric populations. Prior trials have shown that both 1 µg/kg and 2 µg/kg of dexmedetomidine prolong postoperative analgesia compared to local anesthetic alone, with no significant difference in efficacy between the two doses [27]. Similarly, another study comparing 0.5, 1, and 1.5 µg/kg of dexmedetomidine added to 0.2% ropivacaine reported that all doses were equally effective in managing postoperative pain [28]. Although these studies were conducted in children undergoing urologic or lower abdominal surgeries, they provide useful reference points for dosing. Notably, these doses were not associated with delayed PACU discharge despite prolonged sedation. Based on these findings, we selected 1 µg/kg as an effective and safe dose. Dexmedetomidine is known to cause dose-dependent hypotension, bradycardia, and sedation [29]. However, in our study, no patients in the dexmedetomidine group experienced respiratory depression, delayed extubation, or other clinically significant hemodynamic instability. These results support the safety of the selected dose and administration timing in the pediatric orthopedic population. This administration timing was chosen to target the emergence phase, which is the peak period for POD occurrence. Administering dexmedetomidine closer to the end of surgery allows for effective sedation and analgesia during recovery while minimizing intraoperative hemodynamic fluctuations that may arise from earlier use.

This study has several limitations. First, as a retrospective study, the potential for selection bias and residual confounding cannot be eliminated, despite the use of propensity score matching. Second, the assessment of POD relied on the PAED scale, which primarily evaluates emergence delirium rather than late-onset POD. Although PAED is a validated tool for pediatric POD, future studies incorporating longer-term cognitive assessments would provide a more comprehensive evaluation. Third, being a single-center study, generalizability may be limited, especially in settings with different anesthesia protocols or patient populations. Fourth, while PSM effectively balanced measurable baseline characteristics, residual confounding may persist from several sources. Surgical heterogeneity, including unmeasured variations in procedural complexity and duration, could influence outcomes. Similarly, while anesthesia followed standardized protocols, subtle variations in drug titration and regional technique implementation may have existed. Furthermore, clinically relevant but unmeasured factors like preoperative fasting duration and PACU environmental conditions could affect delirium risk. These limitations are inherent to retrospective designs but should be considered when interpreting our findings. Fifth, our choice of administering a single postoperative dose of sufentanil followed by PCA in both groups reflects current clinical practice and was intended to ensure consistent postoperative analgesia. The only difference lies in the intraoperative administration of dexmedetomidine versus sufentanil, allowing us to isolate and evaluate the perioperative effects of dexmedetomidine on POD incidence. While both groups received sufentanil postoperatively, the study design aimed to minimize analgesia-related variability while focusing on the intraoperative sedative and anti-delirium effects of dexmedetomidine. Nonetheless, we acknowledge that the use of sufentanil in both groups may introduce some confounding, and this design choice should be interpreted with caution. Finally, although dexmedetomidine demonstrated short-term benefits in POD prevention and pain control, its long-term cognitive and neurodevelopmental effects in children remain unclear. Future prospective studies with extended follow-up are needed to evaluate any delayed or subtle impacts on neurodevelopmental outcomes.

Conclusion

In conclusion, intraoperative dexmedetomidine was associated with a significantly lower incidence of postoperative delirium in pediatric patients undergoing orthopedic surgery compared to sufentanil. These findings suggest that dexmedetomidine may be a valuable adjunct to current anesthetic practices, offering a safer alternative for postoperative neurocognitive outcomes. While it may not fully replace opioids in clinical practice, its use as part of a multimodal analgesia regimen could enhance recovery and reduce the risk of opioid-related complications in children.

Authors’ contributions

Y.S., S.W. and G.W. conceived the study idea and designed the study; Y.H., W.D., X.Z. and Q.J. performed the study and collected the data; Y.M. performed the data analysis; Y.S. drafted the manuscript. All authors reviewed the manuscript.

Funding

This study is supported by the Bethune Charitable Foundation (EZMR2022-029).

Data availability

The datasets generated and/or analyzed during the current study are not publicly available as it could compromise the privacy of research participants, but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Affiliated Children’s Hospital of Jiangnan University (Wuxi Children’s Hospital) (approval No. 2024012). Informed consent was waived by the Ethics Committee of Affiliated Children’s Hospital of Jiangnan University (Wuxi Children’s Hospital) due to the retrospective nature.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.