February 2017 FAQs
February 2017 FAQs
Can direct oral anticoagulants be prescribed to patients taking carbamazepine?
Can direct oral anticoagulants be prescribed to patients taking carbamazepine?
Currently, 4 direct oral anticoagulants (DOACs) are approved and marketed in the US: apixaban (Eliquis), dabigatran etexilate (Pradaxa), edoxaban (Savaysa), and rivaroxaban (Xarelto). 1-5 Apixaban, edoxaban, and rivaroxaban are factor Xa inhibitors while dabigatran is a direct thrombin inhibitor. These agents differ in their Food and Drug Administration (FDA)-approved indications, which are shown in Table 1.
Table 1. Indications approved by FDA for DOACs. 2-5
Nonvalvular atrial fibrillation
DVT prophylaxis after hip replacement surgery
DVT prophylaxis after knee replacement surgery
Treatment of DVT and PE
Prevention of recurrent DVT and PE
Abbreviations: DOACs=direct oral anticoagulants; DVT=deep venous thrombosis; FDA=Food and Drug Administration; PE=pulmonary embolism.
Since introduction of DOACs to the market, their utilization has significantly increased. 6 Dabigatran was approved by the FDA in 2010, and then rivaroxaban followed in 2011 and apixaban in 2012 . 7 Edoxaban is the most recent addition to the list of available DOACs with an FDA-approval date in 2015. An analysis of a national database, the IMS Health National Disease and Therapeutic Index, containing information on office-based physician visits showed decreased utilization of warfarin and increased utilization of DOACs from 2009 to 201 4 . 6 Between 2010 and 2012, the most common DOAC noted during physician visits was dabigatran. However, starting in 2013, rivaroxaban became the most common DOAC observed during the visits.
At the same time, the Centers for Disease Control and Prevention estimated that in 2010 about 1% of adults experienced active epilepsy in the US. 8 The lifetime prevalence for epilepsy is about 1.8% of adults in the US. Carbamazepine (Tegretol) is one of the available anticonvulsants for epilepsy and carries indications for partial seizures, generalized tonic-clonic seizures, and mixed seizures. 9 Carbamazepine is also used to treat pain from trigeminal neuralgia.
This article reviews evidence on whether DOACs can be prescribed to patients who concurrently take carbamazepine.
Core drug information resources such as Clinical Pharmacology, Facts and Comparisons, LexiComp, and Micromedex provide interaction checkers that identify drug-drug interactions. 1,10-12 Table 2 shows the results from these drug information resources when checked for an interaction between carbamazepine and a DOAC. All drug information resources agree that combinations of carbamazepine with apixaban or rivaroxaban may result in a severe interaction, but the results for use of carbamazepine with dabigatran or edoxaban are inconsistent. LexiComp fails to note that carbamazepine interacts with dabigatran while 3 other references place this interaction into a severe category. Only Clinical Pharmacology notes a moderate interaction between carbamazepine and edoxaban while 3 other resources do not identify this drug interaction.
Table 2. Drug interaction results between carbamazepine and DOACs from core drug information resources (interactions checked on January 16, 2017). 1,10-12
Facts and Comparisons
Potentially severe or life-threatening
Potentially severe or life-threatening
Potentially severe or life-threatening
aClinical Pharmacology defines major drug interaction when the use of 2 agents is contraindicated in some patients and alternative therapy is advised. Moderate drug interaction may result in an unintended clinical outcome, which requires patient monitoring.
bMicromedex defines major drug interaction as an interaction that is life-threatening and may need intervention.
Abbreviation: DOAC=direct oral anticoagulant.
Besides identifying drug interactions, the drug interaction checkers describe potential mechanisms of interactions, clinical management, and any previous studies that explored these interactions. 1,10-12 Carbamazepine is a strong inducer of the cytochrome P450 (CYP)3A4 enzyme and P-glycoprotein (P-gp). Apixaban and rivaroxaban, both substrates of CYP3A4 and P-gp, should not be given in combination with strong inducers of CYP3A4 and P-gp due to potential for decreased blood levels of apixaban or rivaroxaban. Clinical Pharmacology suggests an administration of an increased rivaroxaban dose when given concurrently with carbamazepine but the recommended dose is not list ed. 11 The resource does not mention increasing the dose for apixaban in a similar concurrent administration case. Other resources mention pharmacokinetic studies and a patient case describing co-administration of rifampin or rifampicin with rivaroxaban or apixaban as the basis for the interac tion. 1,1 0,12 LexiComp points out that the Canadian product monograph does not list to avoid the combination of rivaroxaban and strong CYP3A4 inducers but instead to use the combination with c aut ion. 1
Most references recommend to avoid concomitant use of dabigatran and carbamazepine since carbamazepine is a strong P-gp inducer. 10-12 The induction of P-gp may result in a decreased blood concentration of dabigatran. Facts and Comparisons provides the most detail regarding the interaction by describing that the evidence consists of a pharmacokinetic study with rifampin, an observational study involving a patient receiving dabigatran with phenytoin and phenobarbitone, and a case report of a patient taking dabigatran and phenytoin. 10
Only Clinical Pharmacology warns that the concomitant use of carbamazepine with edoxaban may result in decreased blood concentrations of edoxaban. 11 The interaction is due to carbamazepine being a strong P-gp inducer and edoxaban being a substrate for P-gp.
Since core drug information resources do not provide specific and consistent information regarding interactions between carbamazepine and DOACs, search for primary literature is warranted.
There is a limited amount of primary literature discussing drug interactions between carbamazepine and DOACs that has been published. A recent systematic review on interactions between DOACs and antiepileptic medications highlights the lack of literature on this topic. 13 Case reports comprise the majority of information regarding these potential interactions. The authors of the review identified only 6 case reports/series that discuss interactions between DOACs and anticonvulsants with 1 case report related to rivaroxaban and carbamazepine.
A case report discussing an interaction between rivaroxaban and carbamazepine was published in 2013. 14 A 53-year old male patient with epilepsy received a prescription for rivaroxaban 10 mg daily after partial knee arthroplasty. At the same time, the patient was taking carbamazepine 600 mg twice daily to manage his epilepsy. A day later, the patient started to experience shortness of breath and chest pain. He was diagnosed with pulmonary embolism, which the authors attributed to carbamazepine causing decreased blood levels of rivaroxaban.
Another case report, available only as an abstract, involved concomitant administration of dabigatran and 3 enzyme-inducing anticonvulsant medications – carbamazepine, phenytoin, phenobarbital. 15 An 80-year old male patient received a prescription for dabigatran 110 mg twice daily after being diagnosed with atrial fibrillation. He also used carbamazepine, phenytoin, and phenobarbital. He eventually developed dizziness and instability and was diagnosed with cardioembolic stroke at hospitalization. The authors concluded that the cardioembolic stroke was due to insufficient anticoagulation caused by the 3 enzyme-inducing anticonvulsants.
Recommendations for clinical practice
The literature and evidence regarding clinical outcomes of drug interactions between carbamazepine and DOACs remain limited. Drug information resources such as Clinical Pharmacology, Facts and Comparisons, LexiComp, and Micromedex provide inconsistent information regarding carbamazepine interactions with DOACs and highly theoretical explanations for these interactions. Studies on drug interactions for these agents are lacking, and most information on clinical outcomes with the concomitant use is obtained through case reports. Two available case reports revealed that interactions of carbamazepine with rivaroxaban or dabigatran lead to the development of pulmonary embolism or cardioembolic stroke, respectively. 14,15 Therefore, the authors of the systematic review article on drug interactions between anticonvulsants and DOACs recommend avoiding use of DOACs with carbamazepi ne. 13 Vitamin K antagonist, warfarin, should be the preferred therapy in patients who take anticonvulsants that induce P-gp and/or CYP3A4 enzyme such as carbamazepine, levetiracetam, phenobarbital, phenytoin, topiramate, and valproic acid. The utilization of warfarin would allow clinicians to monitor International Normalized Ratio (INR) and adjust warfarin doses as necessary.
1. LexiComp Online [database online]. Hudson, OH: Lexicomp; 2017. http://online.lexi.com/lco/action/home. Accessed January 16, 2017.
2. Eliquis [package insert]. Princeton, NJ: Bristol-Myers Squibb; 2016.
3. Pradaxa [package insert]. Ridgefield, CT: Boehringer, Ingelheim; 2015.
4. Savaysa [package insert]. Parsippany, NJ: Daiichi Sankyo; 2016.
5. Xarelto [package insert]. Titusville, NJ: Janssen Pharmaceuticals; 2016.
6. Barnes GD, Lucas E, Alexander GC, Goldberger ZD. National trends in ambulatory oral anticoagulant use. Am J Med. 2015;128(12):1300-1305.e1302.
7. Drugs@FDA [database online]. Silver Spring, MD: Food and Drug Administration; 2016. http://www.accessdata.fda.gov/scripts/cder/daf/. Accessed January 19, 2017.
8. Epilepsy in adults and access to care–United States, 2010. MMWR Morb Mortal Wkly Rep. 2012;61(45):909-913.
9. Tegretol [package insert]. East Hanover, NJ: Novartis; 2015.
10. Facts & Comparisons eAnswers [database online]. St. Louis, MO: Wolters Kluwer Health, Inc.; 2017. http://online.factsandcomparisons.com/index.aspx. Accessed July 16, 2017.
11. Clinical Pharmacology [database online]. Tampa, FL: Gold Standard, Inc.; 2017. http://clinicalpharmacology-ip.com/default.aspx. Accessed January 16, 2017.
12. Micromedex Solutions [database online]. Greenwood Village, CO: Truven Health Analytics; 2017. http://www.micromedexsolutions.com/micromedex2/librarian/. Accessed January 16, 2017.
13. Stollberger C, Finsterer J. Interactions between non-vitamin K oral anticoagulants and antiepileptic drugs. Epilepsy Res. 2016;126:98-101.
14. Risselada AJ, Visser MJ, van Roon E. [Pulmonary embolism due to interaction between rivaroxaban and carbamazepine]. Ned Tijdschr Geneeskd. 2013;157(52):A6568.
15. Manso G, Jimeno FJ, Ordonez L, Salgueiro ME. Drug interactions of dabigatran: report of one case [Abstract P56]. Basic Clin Pharmacol Toxicol. 2012;111(suppl.1):27.
The information presented is current as January 16, 2017. This information is intended as an educational piece and should not be used as the sole source for clinical decision-making.
Are there data to support the treatment of Pseudomonas infections with cefepime 1 gram every 6 hours vs. 2 grams every 8 hours?
Are there data to support the treatment of Pseudomonas infections with cefepime 1 gram every 6 hours vs. 2 grams every 8 hours?
Pseudomonas species are an aerobic gram-negative rod bacteria that are a common cause of many hospital (and some community) acquired infections.1 Pseudomonas aeruginosa is the fifth most common pathogen causing nosocomial infection and the second most common pathogen causing ventilator-associated pneumonia and catheter-associated urinary tract infection. Other common sites of infection with P. aeruginosa include the bloodstream and skin and soft tissue. Patients who are admitted to an intensive care unit have an especially high risk of contracting this bacteria.
Agents from the beta-lactam, fluoroquinolone, and aminoglycoside drug class all contain one or more agents with antipseudomonal activity.1 Antipseudomonal beta-lactams include piperacillin-tazobactam, ceftazidime, cefepime, imipenem, meropenem, and doripenem. Antipseudomonal fluoroquinolones include ciprofloxacin and levofloxacin, and antipseudomonal aminoglycosides include tobramycin, gentamicin, and amikacin.
Several national guidelines provide recommendations for the treatment of P. aeruginosa at various sites. For example, the American Thoracic Society and the Infectious Diseases Society of America (ATS/IDSA) provide treatment recommendations for specific pneumonia infections, including hospital-acquired pneumonia, ventilator-associated pneumonia, and healthcare-associated pneumonia.2 The IDSA also has guidelines that focus on bloodstream, intraabdominal, skin and soft tissue, and urinary tract infections. In general, one or more antimicrobial agents that have antipseudomonal activity are recommended when treating P. aeruginosa at one of these sites. However, effective treatment of this bacterial infection has become increasingly difficult due to the development of numerous mechanisms of antimicrobial resistance.1
Cefepime is a fourth generation cephalosporin antimicrobial with activity against several gram-positive and gram-negative organisms, including P. aeruginosa.3 The usual dose of cefepime for P. aeruginosa pneumonia, bacteremia associated with an intravascular line, and complicated intraabdominal infection (moderate to severe) is 2 grams intravenously (IV) every 8 hours. Dosing adjustments are recommended for patients who have a creatinine clearance (CrCl) ≤ 60 mL/min or receiving hemodialysis.
Recently, there has been an increased interest in using lower total daily doses of this agent in hopes of attaining similar outcomes. To our knowledge, 3 pharmacokinetic studies and 1 retrospective cohort study have evaluated the treatment of P.aeruginosa infections with a novel cefepime dosing regimen of 1 gram every 6 hours (see Table 1).4-7 To summarize, clinical outcomes data for this novel dosing regimen are limited, but available data has shown that cefepime 1 gram every 6 hours yields similar rates of clinical success when compared to more conventional dosing regimens.4 Similarly, pharmacokinetic outcomes data have shown that expectation values for the probability of target attainment (PTA) are generally similar for cefepime 1 gram every 6 hours compared to standard dosing regimens. However, the allowable minimum inhibitory concentration (MIC) for the organism varied from 1 mg/L to 8 mg/L amongst available studies. It is well known that the amount of time in which a free or non-protein bound drug concentration exceeds the MIC of the organism (fT>MIC) is the best predictor of bacterial killing and microbiological resource for beta-lactam antimicrobials.1 Most of the pharmacokinetic studies in Table 1 used free or non-protein-bound cefepime concentration exceeding the MIC for the organism for 65% of the dosing interval (65% ƒT>MIC) as their pharmacodynamic target for evaluating the PTA; one study used 67% ƒT>MIC as their target.4-7
Table 1. Summary of trials evaluating treatment of Pseudomonas infections with cefepime 1 gram every 6 hours vs. 2 grams every 8 hours.4-7
Based on the available literature, the novel dosing regimen of cefepime 1 gram every 6 hours has shown to be a potential alternative to 2 grams every 8 hours for the treatment of P. aeruginosa. Using 65% ƒT>MIC as the pharmacodynamic target, Lodsie et al. identified cefepime 1 gram every 6 hours as a 0.5-hour infusion to be an alternative to conventional cefepime dosing regimens.5 Similar PTA expectation values were seen at MICs up to 8 mg/L with additional time and cost-savings advantages. Using the same pharmacodynamic target, Roos et al. determined that the PTA expectation value remains ≥ 90% for intermittent dosing of cefepime 1 gram every 6 hours and 2 grams every 8 hours only when the MIC is ≤ 1 mg/L; if administering the same doses as a continuous infusion, a MIC ≤ 8 mg/L will provide a PTA expectation ≥ 90%.6 On the other hand, Tam et al. targeted a goal cefepime concentration at 67% of the dosing interval greater than or equal to the MIC (67% ƒT>MIC), and a Cmin greater than or equal to the MIC (Cmin>MIC).7 In patients with a CrCl of 120 mL/min, a > 80% PTA for 67% ƒT>MIC and Cmin>MIC was achieved when the MIC was ≤ 8 mg/L and ≤ 4 mg/L, respectively. For those patients with a CrCl of 60 mL/min, a > 80% PTA for 67% ƒT>MIC and Cmin>MIC was achieved when the MIC was ≤ 16 mg/L and ≤ 8 mg/L, respectively. The only identified clinical outcomes data related to the novel cefepime dosing regimen of 1 gram every 6 hours was demonstrated by Hegelson et al.4 In this study, cefepime 1 gram ever 6 hours showed no difference in comparison to standard cefepime dosing regimens for duration of therapy, clinical success, infection-related length of stay (IR-LOS), or mortality.
Despite limited supportive data, there are several institutions that have implemented this novel dosing into practice.5,8 An advantage to using cefepime 1 gram every 6 hours compared to standard dosing of 2 grams every 8 hours is the potential for cost-savings based on a lower total daily dose.9 Four vials of cefepime 1 gram/50mL cost $127.48 versus 3 vials of 2 grams/100mL at $153.30. Nonetheless, there are insufficient data to support a total daily dose of 4 grams per day to empirically treat P. aeruginosa.4-7 Cefepime 1 gram every 6 hours should also be avoided in patient populations that were excluded from the various aforementioned trials, including immunocompromised patients, those with cystic fibrosis, neutropenic fever, severe burns, or spinal cord injury, pregnant women, and patients who are underweight or overweight patients (+ 40% ideal body weight).
Overall, cefepime 1 gram every 6 hours may be an appropriate alternative to 2 grams every 8 hours for the treatment of P. aeruginosa, although, clinical outcome data for this novel dosing is limited to one retrospective study. In pharmacokinetic studies where clinical outcomes were not evaluated, the maximum allowable MIC for the organism varied from 1 mg/L to 8 mg/L. While the lower total daily dose presents a potential cost-savings when used for eligible patients, there are no clinical or pharmacokinetic data for use of the novel cefepime dosing regimen in several patient populations, most notably, those with neutropenic fever (see Discussion). Therefore, cefepime 2 grams every 8 hours should continue to be used for patient populations that were not included in studies that evaluated novel dosing, as well as for empiric therapy.
D’Agata E. Pseudomonas aeruginosa and other Pseudomonas species. In: Principles and Practice of Infectious Diseases. 8th ed. London, England: Elsevier Inc; 2015: 2518-2531.
Infectious Diseases Society of America (IDSA). IDSA Practice Guidelines. IDSA website. http://www.idsociety.org/IDSA_Practice_Guidelines/. Accessed January 11, 2017.
Micromedex [database online]. Greenwood Village, CO: Truven Health Analytics, Inc. 2016. http://micromedex.com/. Accessed January 11, 2017.
Helgsen MR, Rihani DS, Waite RA, Wallace MR, DeRyke CA. Evaluation of cefepime 1 gram every 6 hours for treatment of gram-negative bacteremia. Infectious Diseases Society of America Website. https://idsa.confex.com/idsa/2011/webprogram/Paper32312.html. Updated October 21, 2011. Accessed January 11, 2017.
Lodise TP, Lomaestro BM, Drusano GL, Society of Infectious Diseases P. Application of antimicrobial pharmacodynamic concepts into clinical practice: focus on beta-lactam antibiotics: insights from the Society of Infectious Diseases Pharmacists. Pharmacotherapy. 2006;26(9):1320-1332.
Roos JF, Bulitta J, Lipman J, Kirkpatrick CM. Pharmacokinetic-pharmacodynamic rationale for cefepime dosing regimens in intensive care units. J Antimicrob Chemother. 2006;58(5):987-993.
Tam VH, Mckinnon PS, Akins RL, Drusano GL, Rybak MJ. Pharmacokinetics and pharmacodynamics of cefepime in patients with various degrees of renal function. Antimicrob Agents Chemother. 2003;47(6):1853-1861.
Njoku, J. Supporting evidence for alternate cefepime dosing substitution. NebraskaMed, www.nebraskamed.com. June 2010. Updated March 2012. Accessed January 11, 2017.
Lexicomp [database online]. Hudson, OH: Wolters Kluwer Clinical Drug Information, Inc; 2016. online.lexi.com/lco/action/home. Accessed January 11, 2017.
Kevin Chang, PharmD
PGY-1 Pharmacy Practice Resident
University of Illinois at Chicago College of Pharmacy
The information presented is current as of December 27, 2017. This information is intended as an educational piece and should not be used as the sole source for clinical decision-making.
What are the treatment options for post-operative shivering?
What are the treatment options for post-operative shivering?
Postoperative shivering is an unpleasant side effect of recovery from anesthesia. It is characterized as involuntary mechanical oscillatory muscle movements that can be clonic in nature. 1-3 These tremors can affect one or more groups of skeletal muscles, and their onset may range from 5 to 30 minutes after stopping anesthesia. 3 ,4 It is unclear how many patients experience shivering, but the incidence can vary from 5% to 64% of patients recovering from anesthes ia. 3,5 Several risk factors are thought to increase the incidence of postoperative shivering including male gender, young age, length of surgery and anesthesia, perioperative hypothermia, and type of anesthetic used during s urg ery. 1 Halogenated agents and thiopental appear to increase the incidence of shivering postoperatively, as does the use of neuraxial (both spinal and epidural) an est hesia. 5
While there may be several causes behind postoperative shivering, it can be effectively categorized into one of two classes: thermoregulatory and non-thermoregulatory shivering. 3,5 Thermoregulatory shivering is a response to a drop in core body temperature, typically caused by perioperative hypothermia. The onset of hypothermia is usually a consequence of thermoregulatory inhibition by anesthetics. The causes of non-thermoregulatory shivering are not completely understood, but postoperative pain may play an important role in its onset. Often, it is the combination of anesthetic-induced hypothermia and exposure to a cold environment (such as the operating room) that results in postoperative shivering.
While discomfort and increased postoperative pain (due to surgical incision stretching) are the main clinical implications of shivering, it has been proposed that oxygen consumption and metabolic demand may also be increased. 3,5 However, the extent of this increase and its effect on clinical status or cardiac output and morbidity are still unclear. Shivering may increase metabolic heat production up to 600% above baseline. 3 Practical consequences of postoperative shivering may include greater difficulty obtaining accurate blood pressure, electrocardiogram, and pulse oximetry readin gs. 5
There are several non-pharmacological approaches that can address shivering in recovering patients. Shivering caused by hypothermia can be prevented by skin surface re-warming, either by covering the patient with surgical drapes, using a forced-air warmer prior to anesthesia, raising the temperature in the operating room, or warming intravenous solutions when possible. 1,5 Several pharmacological options can be used for preventing or treating postoperative shiveri ng. 3 These include opiates (buprenorphine, butorphanol, fentanyl, and meperidine), tramadol, ketamine, magnesium sulfate, α2-agonists (eg, clonidine and dexmedetomidine), physostigmine, doxapram, methylphenidate, propofol, and 5-HT3 antagonists (granisetron, ondansetron, and palonosetr on) . 3,5
Despite the number of pharmacological options available for postoperative shivering, no “gold-standard” treatment exists. In fact, the mechanism and relative efficacy of these interventions is unclear. A number of trials have evaluated the role of various parenteral medications in postoperative shivering prophylaxis and treatment. The results of these trials are discussed below.
A meta-analysis of 27 randomized controlled trials involving over 2200 patients found clonidine to be the most frequently studied drug in the prevention of shivering. 6 Other commonly studied medications included meperidine and tramadol. Methylphenidate, midazolam, dolasetron, ondansetron, physostigmine, and flumazenil were all studied in a smaller number of trials with a limited number of patients. All of the included studies compared single parenteral doses of active drug to placebo or no treatment. Varying doses were investigated; however, all studies showed that pharmacological therapy was more effective than the control intervention at preventing postoperative shivering. One exception was lower doses of meperidine (<24 mg) which were no more effective than controls.
The results of this analysis for the most commonly studied interventions are included in Table 1. 6 For clonidine, the authors of the meta-analysis also looked at the dose of clonidine and its relative benefit on shivering; the dosage categories were arbitrarily set (see Table 1). The authors noted that clonidine, meperidine, and tramadol were each associated with a low number needed to treat to avoid one episode of postoperative shivering. However, there was a high incidence of shivering in the control groups (mean 52%, range ≤20% to 70%) which suggests that the study populations may have had a higher baseline risk for shivering.
Table 1. Comparison of parenteral IV agents in prevention of postoperative shivering. 6
Drug and dosage
Total number of patients enrolled
Relative benefit (95% CI) vs control
Clonidine 65 to 300 mcg
978 (14 trials)
Clonidine 65 to 110 mcg
230 (3 trials)
Clonidine 140 to 150 mcg
440 (5 trials)
Clonidine 220 to 330 mcg
308 (6 trials)
Meperidine 12.5 to 35 mg
250 (5 trials)
Tramadol 35 to 220 mg
250 (4 trials)
Abbreviations: CI=confidence interval; IV=intravenous; NNT=number needed to treat.
Individual meta-analyses support the efficacy of several other agents compared to controls, including tramadol, ondansetron, and the class of 5-HT3 antagonists. 7-9 The comparative studies included in these analyses used a variety of active and non-active controls. Tramadol significantly decreased the incidence of severe postoperative shivering compared to placebo (relative ratio 0.17, 95% confidence interval [CI] 0.12 to 0.23) in a meta-analysis of 17 studies; this finding was maintained in several subgroup analyses. 7 In an analysis of 6 trials, ondansetron significantly reduced the risk of postoperative shivering compared to placebo (risk ratio 0.43, 95% CI 0.27 to 0.7 ). 8 Postoperative shivering was also significanty reduced compared to controls with intravenous 5-HT3 antagonists in an analysis of 16 trials (risk ratio 0.44, 95% CI 0.35 to 0.56); however, the authors concluded that more trials with larger sample sizes may be needed to make definitive conclusions about the efficacy of 5-HT3 antagonists for preventing postoperative shiveri ng. 9
A Cochrane analysis of 20 studies that compared α2-agonists to a control group found a significantly reduced risk of shivering but also a high degree of heterogeneity among the included studies. 10 Dexmedetomidine increased the risk of sedation and bradycardia compared to placebo; these adverse effects were not seen with clonidine. In another meta-analysis of 39 controlled clinical trials, dexmedetomidine was associated with significantly less postoperative shivering compared to placebo (risk ratio 0.26, 95% CI 0.2 to 0.34), but also an increased risk of sedation (risk ratio 2.94, 95% CI 2.18 to 3.98), bradycardia (risk ratio 2.39, 95% CI 1.54 to 3.72), hypotension (risk ratio 1.35, 95% CI 1.04 to 1.75), and dry mouth (risk ratio 7.33, 95% CI 2.28 to 23.58 ). 11 Dexmedetomidine was more effective than propofol in prevening postoperative shivering based on an analysis of 2 studies (risk ratio 0.33, 95% CI 0.11 to 0.98), but all other comparisons with active controls were not significantly different. The authors concluded that dexmedetomidine should not be routinely used for prevention of postoperative shivering.
Several trials have compared ondansetron and meperidine for prevention of postoperative shivering. Several trials have reported a benefit of ondansetron compared to meperidine, but meta-analyses have not. 8,12 One meta-analysis of 5 comparative trials found no difference between ondansetron and meperidine (risk ratio 0.86, 95% CI 0.66 to 1.11) . 12 Similarly, an analysis of 3 trials found similar efficacy between these agents (risk ratio 0.68, 95% CI 0.39 to 1. 1 9). 8 Another meta-analysis of 6 studies reported no difference between meperidine and 5-HT3 antagonists in general (risk ratio 0.89, 95% CI 0.60 t o 1. 34). 13
Several experts have conflicting opinions about the role of pharmacological prevention in clinical practice. It is argued that the best first-line prevention of postoperative shivering is through non-pharmacological methods, such as direct skin pre-warming or re-warming, covering the patients with surgical drapes, and avoiding the administration of cold epidural and intravenous fluids. 5 Some authors have argued that parenteral agents should be reserved for patients who shiver despite non-pharmacological methods. 14
Several interventions including opioids, centrally-acting analgesics, sodium-channel blockers, α2-agonists, methylphenidate, doxapram, magnesium, and ketanserin, were evaluated in a meta-analysis of 20 randomized, placebo-controlled studies on the treatment of postoperative shivering that were published between 1984 and 2000. 2 The most frequently studied agents were meperidine, clonidine, doxapram, alfentanil, and ketanserin (an antihypertensive agent not available in the United States). Meperidine, clonidine, ketanserin, and doxapram had the most evidence in favor of their efficacy when evaluated at 5 minutes. Significance remained in favor of meperidine, clonidine, and alfentanil when shivering was assessed at 10 minutes after administration. There was not enough data about other agents to draw conclusions about their role in postoperative shivering. Some of the efficacy data regarding meperidine, clonidine, and doxapram are presented in Table 2.
Table 2. Comparison of parenteral agents in treatment of postoperative shivering. 2
Percent not shivering after (RR [95% CI]):
Relative Risk (95% CI)a
NNT (95% CI)a
Meperidine 25 mg
43% vs 6%
87% vs 9%
91% vs 23%
Clonidine 150 mcg
87% vs 12%
77% vs 26%
Doxapram 100 mg
33% vs 8%
78% vs 20%
Alfentanil 250 mcg
23% vs 4%
50% vs 9%
54% vs 22%
a At 5 minutes after treatment. Abbreviations: CI=confidence interval; NA=not available: NNT=number needed to treat.
These data suggest that approximately 2 shivering patients need to be treated with meperidine, clonidine, or doxapram to stop shivering in one patient within 5 minutes. 2 Also, these data suggest that meperidine, clonidine, and doxapram may have comparable efficacy at treating postoperative shivering in postsurgical patients. Although significant from the control group, the rates of patients without shivering were lowest with alfentanil.
Since the publication of the previous meta-analysis, several trials have been conducted comparing the efficacy of different agents in the treatment of postoperative shivering. A randomized, prospective, controlled trial evaluated the efficacy of intravenous doxapram 1.5 mg/kg, meperidine 0.35 mg/kg, and saline placebo in 30 postoperative adult patients. 4 The study found no statistical difference between doxapram and meperidine nor between doxapram and placebo. However, meperidine was better than placebo at stopping postoperative shivering. Meperidine stopped shivering in 80% of patients, while saline placebo stopped shivering in 20% of patients (p<0.05). Due to the small sample size, it is difficult to make a meaningful conclusion regarding the efficacy of meperidine. This study also raises the question about the clinical significance of treatment, given that 20% of patients in the placebo group stopped shivering within minutes of administration. Whether this is related to a placebo effect or if it relates to the fact that shivering often self-resolves is not known.
Another prospective, randomized, double-blind study compared the efficacy of meperidine, clonidine, and urapidil (not available in the US) in the treatment of postanesthetic shivering. 14 The trial included 149 postoperative patients. Sixty patients developed shivering after their procedure and were randomized to treatment with intravenous meperidine 25 mg, clonidine 0.15 mg, or urapidil 25 mg. Patients received a second dose after 5 minutes if the shivering did not stop (except for clonidine, which was replaced by saline). A 25-mg dose of meperidine was used as rescue therapy if the second dose of the randomized treatment was not effective after another 5 minutes. The trial found that meperidine and clonidine were both more effective than urapidil (p<0.01). Clonidine was effective in stopping shivering in 16 of 20 patients. The 4 remaining patients required longer than 5 minutes for the shivering to resolve. Shivering ceased within 5 minutes in 18 of 20 patients treated with meperidine. A second dose of meperidine was sufficient to stop shivering in the remaining 2 patients. Treatment was not effective in 8 patients.
Postoperative shivering can be a disturbing side effect for patients recovering from surgery and may result in increased incision pain and metabolic demand. 3,5 Non-pharmacological methods of preventing shivering include covering the patient with surgical drapes during the procedure, warming intravenous fluids, and raising the temperature in the operating room (>23°C) when possible. Postoperative shivering may still occur if nonpharmacological prophylactic methods are used, which may require the use of pharmacological agents for the prevention and treatment of shivering. A wide variety of agents with different mechanisms of action have been studied. Unfortunately, there is no “gold-standard” drug for the prevention or treatment of postoperative shivering. Numerous trials have studied the efficacy of several agents, either in comparison to placebo or to each other. Clonidine, meperidine, and tramadol all appear to be effective agents in postoperative shivering prophylaxis . 6 Ondansetron and other 5-HT3 antagonists have similar efficacy compared to meperi dine . 8,12 The choice of a specific prophylactic agent should include consideration of the patient’s clinical status and tolerability concerns.
It appears that clonidine and meperidine have the most data to support their efficacy in treatment. 2,4,14 While most studies and one meta-analysis show similar efficacy of both agents in stopping shivering in postoperative patients, the choice depends on the patient’s clinical status and the drug’s side effect profile. Those medications also have additional properties that may determine their usefulness in certain patient populations. For example, meperidine has analgesic effects, while clonidine is an antihypertensive that can be used to lower blood pressure. If pharmacological treatment is to be used in a shivering postoperative patient, the clinical status (including hemodynamic stability) of the patient should be considered along with the medication’s mechanism and potential side effects.
1. Alfonsi P. Postanaesthetic shivering: epidemiology, pathophysiology, and approaches to prevention and management. Drugs. 2001;61(15):2193-2205.
2. Kranke P, Eberhart LH, Roewer N, Tramer MR. Pharmacological treatment of postoperative shivering: a quantitative systematic review of randomized controlled trials. Anesth Analg. 2002;94(2):453-460.
3. De Witte J, Sessler DI. Perioperative shivering: physiology and pharmacology. Anesthesiology. 2002;96(2):467-484.
4. Shrestha AB. Comparative study on effectiveness of doxapram and pethidine for postanaesthetic shivering. JNMA J Nepal Med Assoc. 2009;48(174):116-120.
5. Crowley LJ, Buggy DJ. Shivering and neuraxial anesthesia. Reg Anesth Pain Med. 2008;33(3):241-252.
6. Kranke P, Eberhart LH, Roewer N, Tramer MR. Single-dose parenteral pharmacological interventions for the prevention of postoperative shivering: a quantitative systematic review of randomized controlled trials. Anesth Analg. 2004;99(3):718-727.
7. Shuying L, Ping L, Xuemei L. Efficacy of the prophylactic administration of tramadol against postoperative shivering: a meta-analysis of randomized controlled trials. Minerva Anestesiol. 2016. [Epub ahead of print].
8. Tie HT, Su GZ, He K, Liang SR, Yuan HW, Mou JH. Efficacy and safety of ondansetron in preventing postanesthesia shivering: a meta-analysis of randomized controlled trials. BMC Anesthesiol. 2014;14:12.
9. Wang W, Song X, Wang T, Zhang C, Sun L. 5-HT3 receptor antagonists for the prevention of perioperative shivering: a meta-analysis. J Clin Pharmacol. 2016. [Epub ahead of print].
10. Lewis SR, Nicholson A, Smith AF, Alderson P. Alpha-2 adrenergic agonists for the prevention of shivering following general anaesthesia. Cochrane Database Syst Rev. 2015(8):CD011107.
11. Liu ZX, Xu FY, Liang X, et al. Efficacy of dexmedetomidine on postoperative shivering: a meta-analysis of clinical trials. Can J Anaesth. 2015;62(7):816-829.
12. Li M, Hu X, Tan Y, Yang B, Li K, Tang Z. Meta-analysis of randomized controlled trials on the efficacy and safety of ondansetron in preventing postanesthesia shivering. Int J Surg. 2016;35:34-43.
13. Zhou C, Zhu Y, Liu Z, Ruan L. 5-HT3 receptor antagonists for the prevention of postoperative shivering: a meta-analysis. J Int Med Res. 2016. [Epub ahead of print].
14. Schwarzkopf KR, Hoff H, Hartmann M, Fritz HG. A comparison between meperidine, clonidine and urapidil in the treatment of postanesthetic shivering. Anesth Analg. 2001;92(1):257-260.
The information presented is current as of January 3, 2017. This information is intended as an educational piece and should not be used as the sole source for clinical decision-making.