Share this article on:

Clonidine Does Not Reduce Pain or Opioid Consumption After Noncardiac Surgery

Turan, Alparslan MD; Babazade, Rovnat MD; Kurz, Andrea MD; Devereaux, Phillip J. MD, PhD; Zimmerman, Nicole M. MS; Hutcherson, Matthew T. BS; Naylor, Amanda J. BA; Ali Sakr Esa, Wael MD; Parlow, Joel MD, FRCPC; Gilron, Ian MD; Honar, Hooman MD; Salmasi, Vafi MD; Sessler, Daniel I. MD

doi: 10.1213/ANE.0000000000001356
Regional Anesthesia and Acute Pain Medicine: Original Clinical Research Report

BACKGROUND: Clonidine is an α2-adrenoceptor agonist, which has analgesic properties. However, the analgesic efficacy of perioperative clonidine remains unclear. We, therefore, tested the hypothesis that clonidine reduces both pain scores and cumulative opioid consumption during the initial 72 hours after noncardiac surgery.

METHODS: Six hundred twenty-four patients undergoing elective noncardiac surgery under general and spinal anesthesia were included in this substudy of the PeriOperative ISchemia Evaluation-2 trial. Patients were randomly assigned to 0.2 mg oral clonidine or placebo 2 to 4 hours before surgery, followed by 0.2 mg/d transdermal clonidine patch or placebo patch, which was maintained until 72 hours after surgery. Postoperative pain scores and opioid consumption were assessed for 72 hours after surgery.

RESULTS: Clonidine had no effect on opioid consumption compared with placebo, with an estimated ratio of means of 0.98 (95% confidence interval, 0.70–1.38); P = 0.92. Median (Q1, Q3) opioid consumption was 63 (30, 154) mg morphine equivalents in the clonidine group, which was similar to 60 (30, 128) mg morphine equivalents in the placebo group. Furthermore, there was no significant effect on pain scores, with an estimated difference in means of 0.12 (95% confidence interval, −0.02 to 0.26); 11-point scale; P = 0.10. Mean pain scores per patient were 3.6 ± 1.8 for clonidine patients and 3.6 ± 1.8 for placebo patients.

CONCLUSIONS: Clonidine does not reduce opioid consumption or pain scores in patients recovering from noncardiac surgery.

From the *Department of Outcomes Research, Anesthesiology Institute, Cleveland Clinic, Cleveland, Ohio; Department of General Anesthesiology, Anesthesiology Institute, Cleveland Clinic, Cleveland, Ohio; Departments of Cardiology and Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada; §Departments of Outcomes Research and Quantitative Health Sciences, Cleveland Clinic, Cleveland, Ohio; Department of Anesthesiology, Queen’s University and Kingston Hospitals, Kingston, Ontario, Canada; Departments of Anesthesiology and Biomedical and Molecular Sciences, Queen’s University, Kingston General Hospital, Kingston, Ontario, Canada; and #Department of Outcomes Research, Anesthesiology Institute, Cleveland Clinic, Cleveland, Ohio.

Rovnat Babazade, MD, is currently affiliated with Department of Anesthesiology, University of Texas Medical Branch, Galveston, Texas.

Accepted for publication March 17, 2016.

Funding: Supported by the Canadian Institutes of Health Research.

The authors declare no conflicts of interest.

Reprints will be available from the authors.

Address correspondence to Alparslan Turan, MD, Department of Outcomes Research, Cleveland Clinic, 9500 Euclid Ave, P-77, Cleveland, OH 44195. Address e-mail to turana@ccf.org.

Multimodal analgesia combines various classes of medications with the principal goal of interrupting pain transmission and moderating the response to noxious stimulation at various levels to improve efficacy while simultaneously diminishing side effects.1,2 Commonly used nonopioid multimodal analgesics include N-Methyl-D-aspartate antagonists, anticonvulsants, nonsteroidal antiinflammatory agents, local anesthetics, and α2-adrenoceptor agonists.3

Clonidine is an α2-adrenoceptor agonist, which is generally prescribed for hypertension; however, the drug has distinct sedative and analgesic properties.4 Clonidine provides analgesia via both peripheral and central mechanisms. The peripheral effect apparently results from blocking C fibers and/or interaction with inhibitory G-proteins.5 Centrally, analgesia appears to result from stimulation of α2-adrenergic receptors located in substantia gelatinosa of the dorsal horn at the central nervous system. This stimulation increases acetylcholine concentrations and inhibits neurotransmission by decreasing the release of substance P and glutamate.2,6 Supraspinal analgesia is mediated through noradrenergic neurons, stimulating neuronal firing in locus coeruleus and norepinephrine release.7

As might thus be expected, clonidine prolongs analgesia and decreases opioid requirements when added as an adjuvant to local anesthetics in intrathecal or epidural blocks.8–10 However, the analgesic efficacy of perioperative systemic clonidine remains unclear. Furthermore, the few published studies included only a small number of patients and often evaluated a single perioperative dose that was unlikely to last long enough to provide substantive postoperative analgesia.2,7 The analgesic effect of transdermal clonidine patches, which provide prolonged drug delivery with more reliable bioavailability than oral dosing, has yet to be evaluated.

We, therefore, conducted a preplanned subanalysis of patients who participated in the Perioperative Ischemia Evaluation-2 (POISE-2) trial.11 Specifically, we tested the hypothesis that clonidine (an oral dose followed by a patch) reduces both pain scores and cumulative opioid consumption during the initial 72 hours of hospitalization after noncardiac surgery.

Back to Top | Article Outline

METHODS

With approval of the Cleveland Clinic’s IRB, patients undergoing elective noncardiac surgery who had given written consent and were enrolled in the POISE-2 trial (ClinicalTrials.gov, NCT01082874; registration date: March 8, 2010; principal investigator: PJD) were included in this substudy. No new patients were enrolled beyond those already included in the POISE-2 trial. The methodology of the POISE-2 trial has been published,12 as were the primary results.11 The POISE-2 trial included patients older than or equal to 45 years of age who were scheduled for inpatient noncardiac surgery and had, or were at risk of, vascular disease. This substudy was restricted to patients enrolled in the POISE-2 at 3 Cleveland Clinic centers where postoperative pain scores were recorded in the electronic medical records. We excluded POISE-2 patients whose procedures were done under local anesthesia or who were given epidural analgesia and nerve blocks. Patients with spinal anesthesia were included, and per Cleveland Clinic standards, no long-acting opioids are added to local anesthetics.

Back to Top | Article Outline

Protocol

In brief, the POISE-2 trial used a 2 × 2 factorial design to evaluate the effects of low-dose aspirin versus placebo and clonidine versus placebo on myocardial infarction and death in patients with, or at risk of, vascular disease who were undergoing noncardiac surgery. Patients were randomly allocated to: (a) 200 mg aspirin or placebo 2 to 4 hours before surgery and then 100 mg aspirin or matched placebo daily after surgery for either 7 days or 30 days depending on whether patients were in the continuation or initiation strata, respectively; (b) 0.2 mg oral clonidine or placebo 2 to 4 hours before surgery and then 0.2 mg/d transdermal clonidine patch or placebo patch that was maintained until 72 hours after surgery.

Back to Top | Article Outline

Measurements

Patients, clinicians, research fellows, data collectors, and outcome adjudicators were blinded to the treatment allocation. For this substudy, information was collected from the electronic records of patients enrolled in the POISE-2 trial at the Cleveland Clinic Main Campus, Hillcrest Hospital, and Fairview Hospital. Additional data were obtained from the POISE-2 database, the Cleveland Clinic Perioperative Health Documentation System, and chart review.

Postoperative pain was evaluated using an 11-point verbal response scale (0 = no pain and 10 = worst pain imaginable) 2 hours after surgery in the postanesthesia recovery unit and then every 4 hours after surgery for 72 hours while patients remained hospitalized. Pain scores were recorded every 4 hours by nursing staff per routine. Opioid use was determined by converting total amounts of various drugs to morphine sulfate equivalents.13,14

Back to Top | Article Outline

Statistical Analysis

This study assessed whether clonidine improved acute pain management among surgical patients using a joint hypothesis framework. A priori, we defined clonidine as more effective than placebo if it proved superior on both cumulative opioid consumption (defined in milligram morphine equivalents) and pain scores within 72 hours after surgery.

The main POISE-2 study used a 2-by-2 factorial design, randomizing each patient to receive clonidine/acetylsalicylic acid (ASA), clonidine/ASA placebo, clonidine placebo/ASA, or clonidine placebo/ASA placebo in a 1:1:1:1 fashion. Because we were not interested in the effect of aspirin on pain management, we primarily compared clonidine versus nonclonidine patients regardless of ASA randomization. However, we considered whether the clonidine effect differed in patients receiving ASA by testing for the clonidine-by-ASA interaction in all analyses.

We assessed the balance of clonidine versus placebo patients on potentially confounding baseline variables using absolute standardized difference, defined as the absolute difference in proportions, means, or mean ranks divided by the pooled SD. All variables with absolute standardized difference

CV

CV

were defined as imbalanced and adjusted for in the analyses.15

We assessed the effect of clonidine on log-transformed opioid consumption by using a multivariable linear regression model adjusting for site, aspirin allocation, and imbalanced potential confounders. The ratio of geometric means opioid consumption was estimated by exponentiating the treatment effect parameter.

We estimated the clonidine effect on all pain scores recorded for a patient by using a repeated-measures linear regression model with an autoregressive correlation structure, adjusting for site, ASA, and time after surgery. Pain score data were somewhat zero-inflated, but mean pain score per patient appeared normally distributed. Therefore, we also performed a sensitivity analysis estimating clonidine effect on mean pain score per patient by using a multivariable linear regression model, adjusting for site, ASA allocation, and imbalanced potential confounders.

The heterogeneity of the treatment effect across site and ASA randomization was assessed in both primary analyses by testing for both the treatment-by-site and treatment-by-aspirin interactions. We also assessed the clonidine-by-time interaction in the repeated-measures pain score analysis. All interaction tests used a significance criterion of 0.10. Because of an observed clonidine-by-aspirin interaction in the opioid consumption model, we assessed the effect of clonidine on opioid consumption within each aspirin group.

The overall significance level was 0.05 for the primary analyses. Because superiority was required on both primary outcomes (ie, an intersection-union test), no adjustment to the significance criterion was necessary. Thus, we assessed each outcome by using 2-tailed tests at the 0.05 significance level (0.025 in each direction).

Back to Top | Article Outline

Sample Size and Power

Six hundred twenty-four patients enrolled in the POISE-2 trial met the inclusion criteria for this substudy. Given the observed pain score SD of 3 in the control group, we had >99% power to detect a difference in mean pain score of 1.5 at the 0.05 significance level. Furthermore, with the observed coefficient of variation of 1.23 in the control group, we had 83% power to detect a 20% reduction in opioid consumption at the 0.05 significance level. Thus, we had about 82% power to detect the differences on both outcomes, assuming low within-subject correlation across outcomes (ie, 0.83 × 0.99 × 100%).

Back to Top | Article Outline

Meta-Analysis

We updated the 2012 meta-analysis by Blaudszun et al16 that evaluated randomized controlled trials comparing clonidine to placebo by adding our findings and 2 additional randomized controlled trials published between March 2011 and June 2015. Our search included published full reports of randomized, controlled trials that compared perioperative systemic clonidine to placebo administration. We searched for the terms “clonidine,” “α2 agonist,” “pain,” “opioid,” “an(a)esthesia,” and their combinations in the MEDLINE, EMBASE, and CENTRAL.

We checked the full text of retrieved articles. The authors were contacted to obtain additional information if necessary. All articles identified through the literature search were reviewed for inclusion by one author. Then queries were solved with 2 other authors. Extracted information included the number of patients in the clonidine and placebo groups and mean ± SD pain scores.

We reviewed the difference in pain scores in patients given clonidine and placebo at 24 and 48 hours after surgery by using mean ± SD pain scores reported in each study. The overall effect size was estimated from random-effect models based on estimated means, SDs, and sample sizes for each study. Between-study variance was estimated by using the restricted maximum-likelihood approach. We assessed the heterogeneity of the effect sizes using the Q-statistic and its associated P value for each analysis.

In addition, we reported Higgins I 2,17 a measure of the proportion of total variation in the study estimates because of between-study heterogeneity. Effect sizes were presented using forest plots. Publication bias was visually evaluated by using funnel plots. R statistical software version 3.2.0 (The R Foundation for Statistical Computing, Vienna, Austria) and the metafor R package18 were used for the meta-analyses and associated graphs.

Because few studies examined cumulative opioid consumption at 72 hours, we did not perform a meta-analysis on this outcome.

Back to Top | Article Outline

RESULTS

Nine hundred ninety-one POISE-2 trial patients were randomly assigned at the Cleveland Clinic Main Campus, Hillcrest Hospital, and Fairview Hospital; however, 365 were excluded from this analysis because they received either epidural or regional anesthesia. Of the patients excluded for epidural or regional anesthesia, 39% were randomly assigned to clonidine group and 34% were randomly assigned to placebo group. In addition, we excluded 2 patients who had no pain scores because they were intubated for >72 hours, leaving 624 patients available for this analysis (Figure 1; CONsolidated Standards Of Reporting Trials [CONSORT] diagram). In the final study population, 158 (52%) clonidine patients and 159 (50%) clonidine-placebo patients were randomly assigned ASA. Fourteen percent of the patients remained in the hospital for 24 to 48 hours, 14% remained 48 to 72 hours, and 71% remained >72 hours.

Figure 1

Figure 1

Table 1 presents baseline characteristics of the randomized groups. Patients were imperfectly balanced on procedure type, paralysis, psychoses, and solid tumors without metastasis, so these variables were adjusted for in all analyses.

Table 1

Table 1

Clonidine had no effect on opioid consumption compared with placebo, with an estimated ratio of means of 0.98 (95% confidence interval [CI], 0.70–1.38); P = 0.92 (Table 2). The clonidine-by-aspirin interaction was nearly significant (P = 0.06). However, there was no significant clonidine effect regardless of whether or not patients took aspirin (Table 2). Median (Q1, Q3) opioid consumption was 63 (30, 154) mg morphine equivalents in the clonidine group and 60 (30, 128) mg morphine equivalents in the placebo group. Opioid consumption was greater at the Main Campus than at the other Cleveland Clinic sites, presumably reflecting higher case severity; however, there was no significant treatment-by-site interaction (P = 0.34), suggesting that the clonidine effect was consistent across sites.

Table 2

Table 2

Figure 2

Figure 2

Clonidine had no significant effect on pain scores, with an estimated difference in means of 0.12 (95% CI, −0.02 to 0.26) points; P = 0.10. Means ± SDs of mean pain scores per patient were 3.6 ± 1.8 for clonidine patients and 3.6 ± 1.8 for placebo patients. There was no significant treatment-by-site interaction (P = 0.72), treatment-by-aspirin interaction (P = 0.69), or treatment-by-time interaction (P = 0.35). In addition, clonidine had no effect on mean pain score per patient, with an estimated mean difference of 0.07 (95% CI, −0.20 to 0.34); P = 0.59 (Figure 2).

Back to Top | Article Outline

Meta-Analysis

Among 400 potentially relevant articles, 5 appeared eligible.7,19–22 However, 3 did not report means or SDs of pain scores or were otherwise unsuitable for our analysis.7,19,21 Two new articles were thus added to those in the previous meta-analysis, along with our current results, for a total of 7 studies representing 802 patients with 24-hour pain results. Our current study thus contributed more than twice as many patients as all previous studies combined.20,22

Our meta-analysis suggests that clonidine has no effect on pain at 24 or 48 hours after surgery. The estimated overall difference in mean pain was −0.19 (−0.86 to 0.48; P = 0.58; clonidine-placebo) at 24 hours after surgery and −0.03 (−0.40 to 0.34; P = 0.87) at 48 hours after surgery (Tables 3 and 4; Figure 3, A and B). The Q-statistic of 34 and Higgins I 2 17 of 81% suggested heterogeneity among studies 24 hours after surgery (P < 0.001). At 48 hours, Higgins I 2 of 0% indicated little between-study variability, and the Q-statistic of 1.5 did not suggest strong heterogeneity (P = 0.69). However, the Q-statistic is more accurate for meta-analyses with more patients and a larger number of studies.

Table 3

Table 3

Table 4

Table 4

Figure 3

Figure 3

Figure 4

Figure 4

Funnel plots of the clonidine effect against study size are presented in Figure 4, A and B. Several differences in mean pain scores were outside of the 95% confidence margin at 24 hours for clonidine versus placebo patients, suggesting heterogeneity among studies. All points were in the 95% confidence margin at 48 hours, suggesting little heterogeneity among studies. Data were symmetrically distributed at 24 and 48 hours, suggesting little publication bias.

Back to Top | Article Outline

DISCUSSION

Our preplanned subanalysis of patients who participated in the POISE-2 trial showed that clonidine has no effect on opioid consumption compared with placebo, with an estimated ratio of means of 0.98 (0.70–1.38); P = 0.92. Furthermore, there was no significant effect on pain scores, with an estimated difference in means of 0.12 (−0.02 to 0.26); 11-point scale; P = 0.10.

The intensity of acute postoperative pain was mild to moderate on the initial postoperative day and then gradually decreased and was mostly rated as low after 72 hours (Figure 2). This pattern reflects the expected progress of acute postoperative pain. Relatively high pain scores on the second and third days likely result from attrition bias. That is, patients with smaller operations and good pain control were more likely to be discharged earlier.

In contrast to our hypothesis, oral clonidine (0.2 mg) combined with transdermal clonidine (0.2 mg/d) applied for 72 hours did not reduce pain scores or opioid consumption when compared with placebo after noncardiac surgery. Our results are consistent with that reported in the study by Dimou et al23 who evaluated the analgesic effect of transdermal clonidine in 40 patients recovering from open hysterectomy; they found no difference in pain or analgesic requirements over the initial 72 hours after surgery. Furthermore, Owen et al24 compared preoperative oral clonidine and transdermal clonidine (0.2 mg/d) with placebo tablet and patch and found that neither reduced morphine consumption or pain scores. In contrast, Segal et al25 reported lower pain scores and a 50% reduction in morphine consumption with 0.2 and 0.3 mg/d clonidine patches combined with oral doses. Others similarly report that oral and IV clonidine substantially reduce opioid use and improve pain scores

The literature was summarized in a 2012 meta-analysis that included 19 randomized controlled trials with a total of almost 600 patients receiving clonidine.16 The analysis included all randomized trials of systemic clonidine with pain and/or opioid consumption as outcomes. There was a slight 4-mg reduction in cumulative morphine equivalent consumption. Pain scores were significantly reduced during the initial 24 hours, but not thereafter. We updated the 2012 meta-analysis to include 2 new publications and our current results. The updated meta-analysis is consistent with our results in showing that clonidine does not reduce postoperative pain. We note, though, that the overall results were heavily weighted by our trial, which is by far the largest. The extent to which there is publication bias against studies with negative results in this situation cannot be reliably assessed.

Although clonidine did not reduce pain scores or opioid consumption in our broad range of patients, it remains possible that clonidine may reduce pain under other circumstances. Dose may matter; 0.2 mg is a moderate dose for hypertension, but it may be insufficient for postoperative pain. However, it is highly unlikely that clonidine would be given at higher doses for analgesia because the dose we used significantly increased the amount of clinically significant hypotension and bradycardia. Dexmedetomidine is another α2-adrenoceptor with analgesic properties that apparently produces fewer hemodynamic complications, possibly making it more suitable for perioperative use.

Transdermal clonidine provides constant and effective dosing. This route is preferred in patients who are unable to tolerate oral use such as those in the postoperative period.23 However, there is a significant delay in the onset of action after initial clonidine patch application, which could have contributed to the negative results in our study. Although we gave an initial oral loading dose, it remains plausible that delayed onset of action contributed to the negative results in our study.

The major results of the underlying POISE-2 trial, previously reported, were that (1) perioperative low-dose aspirin does not reduce the incidence of myocardial infarction or death and promotes major bleeding26 and (2) perioperative moderate-dose clonidine does not reduce the risk of myocardial infarction or death and promotes clinically significant hypotension and bradycardia.11

In summary, clonidine was reported to be analgesic in many previous trials, a conclusion summarized in a 2012 meta-analysis of available trials. Nonetheless, clonidine was not superior for either opioid consumption or pain scores in our patients, which was by far the largest trial in this area. An updated meta-analysis that includes all available data does not support clonidine, providing any important postoperative analgesia. Given that perioperative clonidine does not reduce the risk of myocardial infarction or death, does not provide useful analgesia, and promotes clinically significant hypotension and bradycardia, there appears to be little indication for using the drug prophylactically in patients undergoing noncardiac surgery.

Back to Top | Article Outline

DISCLOSURES

Name: Alparslan Turan, MD.

Contribution: This author helped design the study, conduct the study, and write the manuscript.

Name: Rovnat Babazade, MD.

Contribution: This author helped conduct the study and write the manuscript.

Name: Andrea Kurz, MD.

Contribution: This author helped design the study and write the manuscript.

Name: Phillip J. Devereaux, MD, PhD.

Contribution: This author helped design the study and write the manuscript.

Name: Nicole M. Zimmerman, MS.

Contribution: This author helped analyze the data and write the manuscript.

Name: Matthew T. Hutcherson, BS.

Contribution: This author helped conduct the study and write the manuscript.

Name: Amanda J. Naylor, BA.

Contribution: This author helped conduct the study and write the manuscript.

Name: Wael Ali Sakr Esa, MD.

Contribution: This author helped design the study, conduct the study, and write the manuscript.

Name: Joel Parlow, MD, FRCPC.

Contribution: This author helped design the study and write the manuscript.

Name: Ian Gilron, MD.

Contribution: This author helped design the study and write the manuscript.

Name: Hooman Honar, MD.

Contribution: This author helped conduct the study and write the manuscript.

Name: Vafi Salmasi, MD.

Contribution: This author helped conduct the study and write the manuscript.

Name: Daniel I. Sessler, MD.

Contribution: This author helped design the study and write the manuscript.

This manuscript was handled by: Spencer S. Liu, MD.

Back to Top | Article Outline

REFERENCES

1. Wylde V, Lenguerrand E, Gooberman-Hill R, Beswick AD, Marques E, Noble S, Horwood J, Pyke M, Dieppe P, Blom AW. Effect of local anaesthetic infiltration on chronic postsurgical pain after total hip and knee replacement: the APEX randomised controlled trials. Pain 2015;156:1161–70.
2. Zhang X, Bai X. New therapeutic uses for an alpha2 adrenergic receptor agonist—dexmedetomidine in pain management. Neurosci Lett 2014;561:7–12.
3. Kolettas A, Lazaridis G, Baka S, Mpoukovinas I, Karavasilis V, Kioumis I, Pitsiou G, Papaiwannou A, Lampaki S, Karavergou A, Pataka A, Machairiotis N, Katsikogiannis N, Mpakas A, Tsakiridis K, Fassiadis N, Zarogoulidis K, Zarogoulidis P. Postoperative pain management. J Thorac Dis 2015;7:S62–72.
4. Liang Y, Li S, Chen L. The physiological role of drug transporters. Protein Cell 2015;6:334–50.
5. Nasirinezhad F, Hosseini M, Salari S. Anti-allodynic efficacy of NMDA antagonist peptide and noradrenaline alone and in combination in rodent neuropathic pain model. Korean J Pain 2015;28:96–104.
6. Nakamura M, Ferreira SH. Peripheral analgesic action of clonidine: mediation by release of endogenous enkephalin-like substances. Eur J Pharmacol 1988;146:223–8.
7. Bharti N, Dontukurthy S, Bala I, Singh G. Postoperative analgesic effect of intravenous (i.v.) clonidine compared with clonidine administration in wound infiltration for open cholecystectomy. Br J Anaesth 2013;111:656–61.
8. Mohammad W, Mir SA, Mohammad K, Sofi K. A randomized double-blind study to evaluate efficacy and safety of epidural magnesium sulfate and clonidine as adjuvants to bupivacaine for postthoracotomy pain relief. Anesth Essays Res 2015;9:15–20.
9. Moss JR, Cofer S, Hersey S, Goudy S, Werkhaven J, Swanson E, Mantle C, Stowell N, Byrne D, Wang L, Labadie R. Comparison of clonidine, local anesthetics, and placebo for pain reduction in pediatric tonsillectomy. Arch Otolaryngol Head Neck Surg 2011;137:591–7.
10. Ginosar Y, Riley ET, Angst MS. Analgesic and sympatholytic effects of low-dose intrathecal clonidine compared with bupivacaine: a dose-response study in female volunteers. Br J Anaesth 2013;111:256–63.
11. Devereaux PJ, Sessler DI, Leslie K, Kurz A, Mrkobrada M, Alonso-Coello P, Villar JC, Sigamani A, Biccard BM, Meyhoff CS, Parlow JL, Guyatt G, Robinson A, Garg AX, Rodseth RN, Botto F, Lurati Buse G, Xavier D, Chan MT, Tiboni M, Cook D, Kumar PA, Forget P, Malaga G, Fleischmann E, Amir M, Eikelboom J, Mizera R, Torres D, Wang CY, Vanhelder T, Paniagua P, Berwanger O, Srinathan S, Graham M, Pasin L, Le Manach Y, Gao P, Pogue J, Whitlock R, Lamy A, Kearon C, Chow C, Pettit S, Chrolavicius S, Yusuf S; POISE-2 Investigators. Clonidine in patients undergoing noncardiac surgery. N Engl J Med 2014;370:1504–13.
12. Devereaux PJ; POISE-2 Investigators. Rationale and design of the PeriOperative ISchemic Evaluation-2 (POISE-2) trial: an international 2 × 2 factorial randomized controlled trial of acetyl-salicylic acid vs. placebo and clonidine vs. placebo in patients undergoing noncardiac surgery. Am Heart J 2014;167:804–9.e4.
13. Turan A, Atim A, Dalton JE, Keeyapaj W, Chu W, Bernstein E, Fu A, Jae Ho L, Saager L, Sessler DI. Preoperative angiotensin-converting enzyme inhibitor use is not associated with increased postoperative pain and opioid use. Clin J Pain 2013;29:1050–6.
14. Vissers KC, Besse K, Hans G, Devulder J, Morlion B. Opioid rotation in the management of chronic pain: where is the evidence? Pain Pract 2010;10:85–93.
15. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009;28:3083–107.
16. Blaudszun G, Lysakowski C, Elia N, Tramèr MR. Effect of perioperative systemic α2 agonists on postoperative morphine consumption and pain intensity: systematic review and meta-analysis of randomized controlled trials. Anesthesiology 2012;116:1312–22.
17. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21:1539–58.
18. Viechtbauer W. Conducting meta-analyses in R with the metafor package. J Stat Softw 2010;36:1–48.
19. Salengros JC, Hecquet F, Touihri K, Sekkat J, Barvais L, Engelman E. Low-dose intravenous ketamine and clonidine for poor postoperative opioid responsiveness: a double blind randomized study. Acta Anaesthesiol Belg 2011;62:65–72.
20. Schlimp CJ, Pipam W, Wolrab C, Ohner C, Kager HI, Likar R. Clonidine for remifentanil-induced hyperalgesia: a double-blind randomized, placebo-controlled study of clonidine under intra-operative use of remifentanil in elective surgery of the shoulder. Schmerz 2011;25:290–5.
21. Nitta R, Goyagi T, Nishikawa T. Combination of oral clonidine and intravenous low-dose ketamine reduces the consumption of postoperative patient-controlled analgesia morphine after spine surgery. Acta Anaesthesiol Taiwan 2013;51:14–7.
22. Behdad S, Ayatollahi V, Yazdi AG, Mortazavizadeh A, Niknam F. Effect of oral low dose clonidine premedication on postoperative pain in patients undergoing abdominal hysterectomy: a randomized placebo controlled clinical trial. Rev Med Chir Soc Med Nat Iasi 2013;117:934–41.
23. Dimou P, Paraskeva A, Papilas K, Fassoulaki A. Transdermal clonidine: does it affect pain after abdominal hysterectomy? Acta Anaesthesiol Belg 2003;54:227–32.
24. Owen MD, Fibuch EE, McQuillan R, Millington WR. Postoperative analgesia using a low-dose, oral-transdermal clonidine combination: lack of clinical efficacy. J Clin Anesth 1997;9:8–14.
25. Segal IS, Jarvis DJ, Duncan SR, White PF, Maze M. Clinical efficacy of oral-transdermal clonidine combinations during the perioperative period. Anesthesiology 1991;74:220–5.
26. Devereaux PJ, Mrkobrada M, Sessler DI, Leslie K, Alonso-Coello P, Kurz A, Villar JC, Sigamani A, Biccard BM, Meyhoff CS, Parlow JL, Guyatt G, Robinson A, Garg AX, Rodseth RN, Botto F, Lurati Buse G, Xavier D, Chan MT, Tiboni M, Cook D, Kumar PA, Forget P, Malaga G, Fleischmann E, Amir M, Eikelboom J, Mizera R, Torres D, Wang CY, VanHelder T, Paniagua P, Berwanger O, Srinathan S, Graham M, Pasin L, Le Manach Y, Gao P, Pogue J, Whitlock R, Lamy A, Kearon C, Baigent C, Chow C, Pettit S, Chrolavicius S, Yusuf S; POISE-2 Investigators. Aspirin in patients undergoing noncardiac surgery. N Engl J Med 2014;370:1494–503.
© 2016 International Anesthesia Research Society