The Effect of Dexmedetomidine on the Acute Pain After Cardiothoracic Surgeries: A Systematic Review

Introduction Acute post-operative pain remains a troublesome complication of cardiothoracic surgeries. Several randomized controlled trials have examined the efficacy of dexmedetomidine as a single or as an adjuvant agent before, during and after surgery. However, no evidence-based conclusion has been reached regarding the advantages of dexmedetomidine over the other analgesics. Objective To review the effect of dexmedetomidine on acute post-thoracotomy/sternotomy pain. Methods Medline, SCOPUS, Web of Science, and Cochrane databases were used to search for randomized controlled trials that investigated the analgesia effect of dexmedetomidine on post-thoracotomy/sternotomy pain in adults' patients. The outcomes were postoperative pain intensity or incidence, postoperative analgesia duration, and the number of postoperative analgesic requirements. Results From 1789 citations, 12 trials including 804 subjects met the inclusion criteria. Most studies showed that pain score was significantly lower in the dexmedetomidine group up to 24 hours after surgery. Two studies reported the significant lower postoperative analgesia requirements and one study reported the significant lower incidence of acute pain after surgery in dexmedetomidine group. Ten studies found that the total consumption of narcotics was significantly lower in the dexmedetomidine group. The most reported complications of dexmedetomidine were nausea/vomiting, bradycardia and hypotension. Conclusion Dexmedetomidine can be used as a safe and efficient analgesic agent for reducing the postoperative pain and analgesic requirements up to 24 hours after cardiothoracic surgeries. However, further well-designed trials are needed to find the optimal dosage, route, time, and duration of dexmedetomidine administration.


INTRODUCTION
Acute pain is one of the intense complications after cardiothoracic surgeries (CTS), which can delay patients' recovery and may increase patients' morbidity and mortality [1] . Acute pain after CTS has been determined as a main risk factor in the pathogenesis of numerous postoperative side effects such as respiratory failure [2,3] . Inadequately controlling the postoperative pain (POP) increase the risk of pulmonary complications due to the diaphragmatic dysfunction and incapability of patients to take large-volume breaths [4] . Consequently, effective pain management can play a vital role in reducing patients discomfort and, therefore, it should be a prerequisite for promoting respiratory and cardiac function of patients undergoing CTS [5,6] .
In last decades, several pharmacological and nonpharmacological interventions have been developed

Study Selection and Data Collection Process
Two authors (FHK-HS) searched the databases using search strategy (n=1789). They independently screened the titles and abstracts of retrieved studies against the predetermined inclusion criteria for selecting relevant articles (1221 title rejected straightaway because of duplicate or irrelevant study. Reasons for excluding an article were documented. The full-text of potentially relevant articles, which met the inclusion criteria, was reviewed for comprehensive assessment against the inclusion criteria. Disagreement about study selection was resolved by discussion and consensus with the third author (VH). In cases that additional data was required, the corresponding author of the study was contacted. Each included study was independently evaluated by three authors (VH-FHK-HS) for content. Then, data extraction table was completed by relevant data of studies that met the inclusion criteria. None of the review authors (VH-FHK-HS) was blinded to reference details during the study selection process.

Assessing Risk of Bias
The methodological quality of the selected studies was independently evaluated by two authors (VH-FHK) using the Cochrane Collaboration's tool for assessing risk of bias ( Table 1). As recommended by tool developer [26] , we did not determine the total quality score for each domain, however, in interpreting the results, the limitations of each study were considered.
to reduce acute POP including opioids, paravertebral and epidural infusion of local anesthetics, sedatives, nerve blockades, intrapleural analgesia, nerve stimulation, ketamine, gabapentinoids, selective COX-2 inhibitors, nonsteroidal antiinflammatory drugs, alpha2-agonists, and aromatherapy [2,7] . However, the effectiveness and efficacy of those interventions are variable among studies. Many of those interventions, particularly opioids, have several side effects that can impair cardiac and respiratory function following surgery [1,2] . In addition, the benefits of thoracic epidural analgesia as a gold standard for controlling POP have been questioned because of higher risk of severe cardiovascular complications [8] . Hence, acute pain management continues to be a challenge in CTS.
Recently, some opioid-sparing analgesics such as dexmedetomidine (DEX) have demonstrated a promising opportunity to decrease the postoperative complications particularly impairment of respiratory function [9,10] . DEX has been recommended for sedating agitated patients in the intensive care unit (ICU) [11] , because it does not depress the respiratory and cognitive dysfunctions [9,12] .
Several randomized controlled trials (RCTs) have examined the efficacy of DEX on POP after CTS. However, a clear advantage of DEX over other analgesics has not been evident so far. Therefore, the aim of this study was to review the effectiveness of DEX for reducing the acute post-thoracotomy/sternotomy pain in comparison with other analgesics.

METHODS
This systematic review was accomplished in accordance to the PRISMA: the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines [13] . Our PICOS research question was formulated as follows: (P) patients undergoing thoracotomy or sternotomy; (I) dexmedetomidine; (C) placebo or other analgesic drug; (O) postoperative pain; (S) trial.

Eligibility Criteria
Inclusion criteria were: (1) Study designed with RCT; (2) Patients undergoing thoracotomies or sternotomy; (3) Study with at least two groups that compared perioperative (preoperative, intraoperative, or postoperative) administration of DEX with other analgesic agents or placebo; (4) DEX with different routes, dosage, frequency, and duration of administration; (5) POP should be one of the study outcomes.
Conference proceedings, abstracts, letters, and commentaries were excluded. In addition, quasi-randomized trials, nonrandomized trials, studies not published in English and animal trials were excluded.

Outcomes Measurement
Primary outcomes were (1) POP intensity measured by visual analogue scale (VAS) or verbal or numerical rating scales (VRS or NRS) or POP incidence; (2) number of postoperative narcotic and/ or analgesic requirements; (3) postoperative analgesia duration. Secondary outcomes were: (1) number of DEX-associated major adverse events.  [14,23] . The duration of DEX administration among the trials was varied and ranged from one to 72 hours.

DEX versus Placebo
Nine trials compared DEX with placebo. Intraoperative administration of DEX was compared with placebo (normal saline) in four trials [15,16,17,21] , while postoperative administration of DEX was compared with placebo (normal saline) in five trials [14,18,20,24,25] . All of these nine trials showed significant lower POP scores in the DEX group. In general, intra-and postoperative administration of DEX could reduce the pain intensity score after surgery in comparison with placebo.

DEX versus Morphine
In comparison with morphine (0.2 mg via PCIA), administration of DEX 4 mcg/kg/h via PCIA could improve the pain control during the first 12 hours after surgery and decrease intravenous morphine consumption during ICU stay [23] .

DEX Addition to Bupivacaine
One study compared the addition of DEX (1 mcg/kg) to epidural bupivacaine 0.5% with epidural bupivacaine 0.5% and found that epidural use of DEX could decrease the anesthetic requirements and improve postoperative analgesia [22] .

Post-Operative Pain Intensity or Incidence
Nine trials reported the POP scores at different time points [14,15,[17][18][19][20][21][22][23] . Table 6 shows the POP scores at different time points, which were significantly lower in the DEX group. Only one trial [14] showed a significant lower pain intensity 36 hours after surgery in the DEX group. The incidence of POP in the DEX group was significantly lower in the DEX group when DEX was administered intraoperatively via IV route [16] . The median of POP was significantly lower at all time points up to 24 hours in DEX group when DEX was administered intraoperatively via IV route [21] . The POP scores and morphine consumption were significantly lower in the DEX group when DEX was used intraoperatively via epidural catheter [15] . In all of the trials, no significant difference was found between groups 48 and 72 hours after surgery in terms of POP scores. In general, DEX probably is able to reduce the pain intensity score after CTS up to 24 hours.

Study Characteristics
All twelve trials were RCT with two parallel groups, except for one study that consist of two groups with different dosages of DEX and a third control group [19] . Among the 12 RCTs, the VAS was the most frequently used scale to determine the intensity of POP. Eight studies used the VAS [14,15,18,20,21,[23][24][25] , three used the NRS [16,17,19] and one of them used the VRS [22] .
Patients' mean age among all trials was 55.89 years (range between 34.4 and 67.7 years). Trials included a total number of 566 (70.4%) male and 228 (28.4%) female, for the last 110 subjects. An error in reported data in one study was found [14] (Table 1) and one study did not report the male/female ratio. Generally, the number of male patients was greater than female.
No statistically significant difference was found between the DEX and control groups regarding the baseline characteristics of patients in all included studies. Table 2 depicts the details for perioperative data and anesthesia techniques.

Other Outcomes
The clinical efficacy of DEX on the ICU length of stay was only reported by one study, which showed that ICU stay was significantly shorter in the DEX group than in the control group (2 and 3 days, respectively) [22] . DEX efficacy on the time spent on the ventilator was not reported by any of the included trials. In addition, the information regarding the number of patients who admitted to the ICU after surgery and the duration of ICU stay were not clearly reported across the reviewed trials.

Dealing with Missing Data
In four cases, we contacted the corresponding author to request further information regarding random sequence generation, allocation concealment, additional blinding details, and type of surgery without success and in one case the contact address was not retrievable.

DISCUSSION
Pain management after CTS is an important issue for clinicians because POP can significantly impair the cardiovascular and respiratory function. The present study, including 12 RCTs, reviewed the effectiveness of DEX in reducing POP. Regardless of the methodological quality of included studies, the overall results are relatively consistent among studies. Approximately all included studies were methodologically homogenous; however, they were different in the sample size, use of analgesic and anesthetic agent, number of measured outcomes, study population, route and timing of DEX administration and type of surgery.
DEX group was significantly lower when DEX was administered postoperatively via intravenous route [24,25] . One study compared the addition of different dosage of DEX (0.02 and 0.04 mcg/kg/h) to SUF with SUF 0.02 mcg/kg/h. The addition of DEX 0.04 mcg/kg/h to SUF could improve the analgesic effect of SUF and decrease the total dosage of SUF during the first 72 hours after surgery [19] . In general, DEX administration probably is able to reduce the requirements for supplemental narcotic, rescue sedation and analgesia in the postoperative period for up to 24 hours.

DEX Adverse Events
Only six trials (n=206) have reported the adverse events of DEX administration. In all of those trials, DEX was administered postoperatively through intravenous injection or using PCIA. As depicted in Table 7, the differences between two groups regarding the adverse events were not statistically significant, except for the occurrence of atelectasis, which was significantly higher in the control group (OR 0.400, CI 95%: 0.177-0.904). Because of incomplete report of some trials, the adverse events rate was not comparable among patients who received DEX intraoperatively and postoperatively. Therefore, the duration and timing of DEX administration (short vs. prolonged) on the incident of adverse events was not evaluated.
• DEX is associated with lower postoperative pain scores or incidence after cardiothoracic surgeries in comparison with placebo (normal saline) • DEX is probability able to reduce the analgesia requirement during and after cardiothoracic surgeries • DEX is unable to reduce the postoperative pain score or incidence after 36 hours from the start of surgery Note: Due to the limited number of available trials regarding the effectiveness of DEX, these findings are preliminary; hence, confirmation or rejection of any of these findings warrants further research. Table 5. Protocol for DEX administration in the DEX group.

Study, year Time and route of injection Protocol for DEX injection in DEX group
Dong et al. [14] , 2017 Start: Postoperatively, after transfer to the general ward End: after 48h; Route: IV using PCIA A PCIA protocol consists of sufentanil 3 mcg/kg and 8 mg ondansetron was started for all patients. The PCIA was programmed to deliver a 2ml bolus with a lockout interval of 10 min, and a background infusion rate of 4 ml/h. DEX 4 mcg/kg was added to the PCIA for DEX group.
Dutta et al. [15] , 2017 Start: Intraoperatively, before induction of anesthesia End: after 72h post-operative period; Route: epidural catheter All patients received the study medications through paravertebral (multipored epidural) catheter. Patients in the DEX group received 15 mL of 0.75% ropivacaine plus DEX, 1 mg/kg bolus over 3-to-5 minutes followed by an infusion of 0.2% ropivacaine plus 0.2 mg/ kg/h of dexmedetomidine at 0.1 mL/kg/h. Paravertebral infusion was stopped and the catheter was removed 72h after surgery.
Jabbary Moghaddam et al. [16] , 2016 Start: Intraoperatively, after induction; End: ?h after extubation in ICU; Route: IV infusion 0.5 mcg/kg/h of DEX was infused from the initiation of anesthesia until extubation in the ICU.
Cai et al. [17] , 2016 Start: Intraoperatively, before the start of anesthesia End: 30min before the end of surgery; Route: IV infusion Before anesthesia, patients were administered a loading dose of 1 mg/kg DEX for 10min, followed by continuous infusion at 0.5 mg/kg/h until 30min before the end of surgery.
Priye et al. [18] , 2015 Start: Post-operative, after transfer to ICU; End: after 12h; Route: IV infusion After surgery, patients were transferred intubated and ventilated to the ICU to receive 12h infusion of DEX 0.4 mcg/ kg/h without a loading dose.
Ren et al. [19] , 2015 Start: Postoperatively, after patients were transferred to PACU; End: after 72h; Route: IV using PCIA All patient received DEX intraoperatively. After surgery, 2 doses of DEX in addition of sufentanil were compared with sufentanil using same PCIA protocol. PCIA was programmed to deliver a bolus dose of 2 mL, with background infusion of 2 mL/h and a lockout of 5min, 4h limit of 40 mL.
Ramsay et al. [20] , 2014 Start: Postoperatively, 18 to 24h after surgery when patients were admitted to the telemetry unit; End: After 24h; Route: IV infusion An intraoperative infusion of DEX at 0.2 to 0.5 mcg/kg/h was started for all patients that continued during their ICU or PACU. 0.1-0.5 mcg/kg/h DEX was started about 18 to 24h after surgery when patients were admitted to the telemetry unit for up to 24h.
Abdel-Meguid [21] , 2013 Start: Intraoperatively, after induction; End: 12h after extubation; Route: IV infusion DEX started by continuous infusion at 0.5 mcg/kg/h after induction of anesthesia; this was reduced to 0.3 mcg/kg/h on admission to the ICU and continued for 12h post extubation.
Ghandi et al. [23] , 2005 Start: Postoperatively, after transfer to ICU; End: after 24h; Route: IV using PCIA After transfer of patients to ICU, they received infusion of DEX 0.2 mcg via a PCIA pump in the first 24 hours after surgery.
Wahlander et al. [24] , 2005 Start: Postoperatively, on ICU arrival; End: after 24h; Route: IV infusion The DEX group received an IV loading dose of DEX of 0.5 mcg/ kg over 20min, followed by continuous IV infusion at 0.4 mcg/ kg/h.
Venn et al. [25] , 1999 Start: Postoperatively, after transfer to ICU; End: 6h-24h after extubation; Route: IV infusion DEX started within 1h of arrival on the ICU with a loading dose of 1 mcg/kg over 10min followed by a maintenance infusion rate of 0.2-0.7 mcg/kg/h to total maximum duration of infusion was 24h.
Note: "?h" means that the end time of medication was not reported by Jabbary Moghaddam.
of DEX for more than 24 hours [27,28] . However, the safe use of this drug has been reported from 24 hours to more than a week [28,29] .
In the present review, the detailed comparison of the results of the included trials was not possible due to differences in intervention protocol and outcomes measurement. Additionally, five of 12 trials [20][21][22][23]25] were likely underpowered for the outcomes, since they did not power the sample size. Therefore, the optimal dosage, timing, and route of DEX administration remain to be elucidated in future studies.
Findings from our review suggest that, compared with normal saline as a placebo, DEX probably is able to reduce the pain intensity score, the number of narcotic consumption and analgesic requirements up to 24 hours. However, due to the low to medium quality of reviewed trials, further studies are warranted to confirm or refute our findings.
Our finding may have noteworthy implications for pain management of adults' patients undergoing CTS, particularly in the first 24 hours after surgery. It is necessary to mention that the use of DEX beyond 24 hours may be associated with a doserelated increase in adverse events and for this reason, the Food and Drug Administration (FDA) has not recommended the use As the findings of our review suggest, several advantages may encourage clinicians to use DEX over other agents for POP reduction. First, DEX does not interfere with respiratory function and has predictable and stable hemodynamic responses. Second, because of its synergistic effects with narcotics and sedatives, DEX can be used to reduce the total dosage of those drugs. Third, DEX has anxiolytic and sedative properties that may improve POP control. Forth, DEX can be used as an adjuvant to local anesthesia; hence, it can improve postoperative analgesia, and reduce the opioid requirement. Fifth, technically, the use of intravenous DEX is easier that paravertebral or peridural route in terms of equipment, skill, and side effects.
These advantages are consistent with the finding of several reviews that have emphasized the analgesic effects of DEX on POP in different sample of patients and surgeries. Schnabel et al. [30] found that the IV administration of DEX compared with placebo or opioids reduces acute POP and opioid consumption, as well as declines the risk of opioid-related adverse events in patients undergoing non-thoracotomy surgeries. Peng et al. [9] found that postoperative PCIA protocols containing opioid-DEX combination have beneficial effects for reducing the POP intensity, postoperative morphine-equivalent consumption and the adverse events. Liu et al. [37] reviewed the efficacy of DEX on perioperative opioid consumption and POP intensity of patients undergoing neurosurgery and found that DEX could reduce opioid consumption and POP intensity. Bellon et al. [38] found that the intraoperative administration of DEX could reduce postoperative opioids consumption and POP intensity in children undergoing surgery.
There are also some studies demonstrating that administration of DEX cannot reduce POP. Jessen Lundorf et al. [39] concluded that perioperative administration of DEX in comparison with placebo seems to have some opioid-sparing effect with no important differences in POP in adult patients undergoing abdominal surgery. Tan and Ho [40] showed that DEX might reduce the length of ICU stay and duration of mechanical ventilation, but increases the risk of bradycardia and hypotension in critically ill adult patients.

Limitations
This systematic review has some limitations. First, we did not judge regarding the quality of each trials and risk of bias due to the limit number of retrieved trials; however, we used the Cochrane risk assessment tool to demonstrate any risk of bias at each domain. Second, due to the considerable heterogeneity between studies, we could not perform meta-analysis to evaluate statistically the efficacy of DEX over the other analgesic agents. Third, due to the lack of reported data in some of the trials, difference in DEX doses, and different times of administration, we could not synthesis the results based on the subgroups. Forth, the outcome of pain was incompletely measured and reported in some trials; hence, we could not critically appraise the outcome of those trials. Fifth, we cannot compare the DEX group with control group regarding the POP scores based on the type of surgery because the number of patients in subcategories of cardiac surgery was reported incompletely.
1 mcg/kg/h [30] . In our review, the dosage for the intravenous infusion were ranged from 0.02 to 0.7 mcg/kg/h and only two studies [17,25] infused DEX at maximum POP reduction dose. We also found that a limited number of studies suggested a scientific justification of the rationale for choosing a dose.
The common adverse events of DEX are hypotension at low blood concentrations, hypertension at high blood concentrations, bradycardia and nausea [29] . Most of these side effects occur at infusion of 0.2-0.7 mcg/kg/h without a bolus dose [28,31] . In our review, the occurrence of respiratory depression was low and reported only in one trials [14] , which is consistent with previous studies [11,28,31] . Previous study showed that respiratory suppression does not even occur at DEX plasma levels up to 8.0 ng/mL and only there is a risk of over-sedation [32] .
DEX possesses analgesic and opioid-sparing effects in the ICU patients [33] . DEX, a shorter-acting and highly selective presynaptic alpha-2-receptor agonist, also possesses pharmacologic sedative, hypnotic, anti-anxious, sympatholytic and analgesic properties [28] . Its analgesic and opioid-sparing effects are dose-dependent and trigger at spinal cord sites as well as through non-spinal mechanisms [29] . It has been suggested that alpha-2A receptors activation, inhibition of the C and A delta fibers signals conduction, and the local release of encephalin are the underlying non-spinal mechanisms of DEX to provide anti-nociception effects [34] . In terms of pharmacokinetics, its action starts about 15 minutes after intravenous injection and its peak concentration is achieved within an hour of continuous intravenous infusion. Appropriate pharmacodynamic effects of DEX are revealed between the plasma concentration of 0.5 and 1.2 ng/ml. Several strategies have been introduced for POP management [35] . It is believed that multimodal analgesic approaches combining different analgesic agents with different mechanisms of action can maximize pain relief while minimize the opioid consumption and thus can limit the opioid-induced side effects [6,36] . As a method of limiting opioid-induced adverse events, therefore, multimodal POP management has the potential to decrease morbidity and mortality after surgery [35] . Consequently, it is expected that the sedative, anesthetic, analgesic, and cardiorespiratory effects of DEX may enhance with concomitant administration with other anesthetic, sedative and analgesic medications [28] . In our review, regardless of the route of administration, three studies used the multimodal approaches and found a reduction in the narcotic consumption and supplemental analgesics requirements [14,17,19] . In addition, two studies found that the addition of DEX to morphine can reduce the opioid consumption [20,21] , the risk of respiratory depression [20] and the time of extubation [21] . One study found that epidural use of DEX plus bupivacaine 0.5% plus fentanyl can decrease the anesthetic requirements and provides effective post-operative analgesia [22] . It should be noted that the peridural (epidural) form of DEX has not been officially approved by any drug administrations around the world. However, in many clinical practices, the off-label form of DEX has been used in various scenarios in the operating room including thoracic epidural anesthesia, regional anesthesia block, intubation, monitored anesthesia care sedation, cardiothoracic surgery, and neurosurgery. The United States FDA has only approved the form of intravenous injection of DEX [27] .

CONCLUSION
In comparison with placebo or other analgesic agents, the use of DEX after CTS is associated with a lower POP intensity, a lower number of post-operative analgesic requirements and a lower number of adverse events, particularly respiratory depression. Thus, DEX can be used as a safe and efficient analgesic agent for reducing the POP up to 24 hours. Overall, data published to date regarding the use of DEX after CTS suggest a marginal clinical benefit. Further well-designed studies with powered sample size are needed to find the optimal dosage, route, time, and duration of administration as well as the best choice of adjuvant analgesia to DEX for reducing POP.