Uridine 5′-diphosphate-glucuronosyltransfera ses (UGTs) are phase II enzymes that have a prominent role in biotransformation of endogenous and exogenous substrates.1 UGTs primarily facilitate glucuronidation. Glucuronidation is the process in which glucuronic acid is linked to a substrate by means of a catalyzing UGT enzyme.2 Consequently, the substrate can be eliminated.2 UGTs generally are grouped into the UGT1 and UGT2 families.3UGT2B15 is one of the UGT enzymes. The major drug substrate for UGT2B15 is lorazepam.2 Lorazepam clearance is executed primarily by direct glucuronidation followed by mainly renal excreation.4
It is known that the UGT2B15 genotype has a strong influence on the pharmacokinetics of lorazepam: 61% of the total variance in lorazepam clearance is explained by this UGT2B15 genotype in healthy subjects.5 Lévesque et al6 showed that a guanine-to-thymine change in UGT2B15 genotype leads to an amino acid substitution at position 85 from aspartic acid (D) to tyrosine (Y), resulting in the UGT2B15(D85Y) genetic polymorphism (ie, UGT2B15*2). As a consequence, homozygous subjects have reduced glucuronidation compared with wild-type and heterozygous carriers. Therefore, homozygous subjects have been shown to have significantly greater lorazepam concentrations.5 Without a change in the general pharmacodynamics (which is the S-shaped dosage-effect relationship) of lorazepam, these greater plasma levels will induce more pronounced clinical effects in patients with UGT2B15*2 polymorphism.
Clinical effects of lorazepam (sedation, psychomotor performance, alertness, motor function, and coordination) show a large variability in humans according to the UGT2B15 genotype.5 However, the effect of lorazepam on anxiety according to genotype is of clinical importance—because it is widely used for its anxiolytic properties,7 such as anxiolytic premedication8—but it is still unknown.
If the anxiolytic effect of lorazepam differs significantly according to the UGT2B15 genotype, this may in part explain patient variability in anxiolytic effect. Therefore, we evaluated interactions between lorazepam and UGT2B15 genotype. In addition, we assessed interactions with patient sex and preoperative anxiety score because these determinants are relevant in studies with lorazepam and perioperative patient anxiety.9,10
This study is a derivative of a larger completed randomized controlled trial (RCT) conducted and completed at the Erasmus University Medical Center with the same study population, study design, procedure, and assessments of outcomes. Therefore, these sections of the methods in this study are in line with our previous publication.9 The primary and secondary objectives of the original RCT were to evaluate the effects of lorazepam (versus placebo) on the quality of recovery and psychologic phenomena, including anxiety, in same-day surgery, respectively.9 The RCT showed that, among others, lorazepam did not improve quality of recovery and that the expected decrease in postoperative anxiety was lower in the lorazepam group.9
This study protocol was approved by the Medical Ethical Committee of Erasmus MC and by the Netherlands Central Committee on Research involving Human Subjects and registered with EudraCT number 2010-020332-19. The trial also is registered under identification number NCT01441843 in the ClinicalTrials.gov protocol registration system. Signed written informed consent was obtained from all patients, including whether they consented to genotyping (Table 1).
Between October 2010 and September 2011, 400 patients were recruited from our same-day surgery department.9 Inclusion criteria were as follows: all patients who were referred for same-day surgery and at least 18 years of age. Patients were excluded if they met one or more of the following criteria: clearly insufficient command of the Dutch language, mental retardation, undergoing eye surgery, extracorporeal shock wave lithotripsy, endoscopy, Botox treatment, abortion, or chronic pain treatment. The latter procedures generally are considered as low invasive, some cooperation of the patient is required, and most practitioners agree that these do not require premedication. Finally, preceding use of psychopharmaceuticals and contraindication to lorazepam use were exclusion criteria as well.
This study was a randomized, double-blinded, placebo-controlled clinical trial with a parallel group design, with varying block sizes (8-10-12) across time. The ratio of allocation to the treatment conditions (lorazepam or placebo) was 1:1. Randomization was performed by a computer-generated table, and the patients were assigned subsequent numbers on inclusion. Health care professionals, patients, and researchers were blinded to the treatment condition; however, nurses who were not involved directly in the care of these patients prepared the study medication according to the randomization table. A flowchart from this study has been published.9
Figure 1 provides a timeline of the study procedure. All patients scheduled for same-day surgery received written information about the trial at least 1 week before surgery. A member of the research group enrolled patients after admission to the day surgery center and sought written informed consent. Patients who consented to participate completed a set of online questionnaires when waiting for surgery (T0). Next, in the preoperative holding, the independent recovery nurses, who had access to the group assignment document, prepared the medication. Blinding was achieved by preparing the transparent fluids in identical syringes. Depending on the patient’s body weight, <75 and ≥75 kg, another nurse blinded to treatment condition injected the single dosage of 1 or 1.5 mg lorazepam by peripheral infusion 30 minutes before induction of anesthesia. The placebo group received an equal volume of 0.9% NaCl. The recovery nurses obtained 8 mL venous blood from the peripheral infusion. After the surgical procedure, patients completed an online questionnaire before discharge (T1). Because of the long duration of action with known therapeutic concentrations even after a single intravenous dosage,11 the preoperative injected lorazepam was considered to be active at the moment of postoperative anxiety testing.
UGT2B15 genotyping was done with the Taqman allelic discrimination assay on an ABI PRISM 7500 FAST sequence detection system. The assay consisted of a master mix (TaqMan GTXpress Master Mix, Applied Biosystems, Nieuwerkerk a/d IJssel, The Netherlands) and 2 specific minor groove binding probes, labeled with the fluorescent dyes VIC and FAM. For the UGT2B15*2 (253G>T) variant, we used a commercial available Drug Metabolizing Enzyme assay mix: C__27028164, rs1902023 (Applied Biosystems). Each reaction was performed in a 10-μL reaction mix containing 2 μL of DNA template (10 ng/μL). The thermal profile consisted of an initial preread step at 60°C for 1 minute, denaturation at 95°C for 20 seconds, followed by 45 cycles of denaturation at 92°C for 3 seconds, and an annealing and extension step at 60°C for 30 seconds, followed by a postread step at 60°C for 1 minute. Genotypes were scored by measuring allelic-specific fluorescence using the 7500 software version 2.0.5 for allelic discrimination (Applied Biosystems).
Anxiety was measured by the Dutch version of the State-Trait Anxiety Inventory (STAI).12 The STAI consists of 2 scales, each containing 20 items. We used the State scale (STAI-State) in this study because this scale measures how the patient feels at the moment of completing the questionnaire.12 Conversely, the Trait scale measures how one generally feels.12 Theoretically, the latter is not expected to be affected by a stressful situation like surgery. We calculated the sum score by summing the scores on the items, theoretically ranging from 20 to 80. Greater scores indicate a greater level of anxiety. In this study, differences in STAI-State levels relative to the preoperative score served as outcome. Thus, anxiety difference was calculated by subtracting the postoperative anxiety score from the preoperative anxiety score. Consequently, a positive difference indicates that the postoperative anxiety score is lower than the preoperative anxiety scores, and a negative difference indicates an increase in anxiety in the postoperative period. The median score of these anxiety differences are called “median differences” in this work.
We used a recessive genetic model for analyzing UGT2B15 genotype because previous literature suggests pharmacodynamic differences according to such a model.13 This means that only patients who were homozygous for the UGT2B15*2 variant were scored as deviant UGT2B15 genotype. As a result, wild-type UGT2B15 and heterozygous UGT2B15*2 were scored as normal UGT2B15 genotype. In addition, treatment condition (lorazepam or placebo), patient sex, and preoperative anxiety score were also used as determinants.
Four-way analysis of variance was used to assess anxiety scores for interaction between the following factors: UGT2B15 genotype status, treatment condition, patient sex, and preoperative anxiety score. The effects were evaluated by F-tests and the corresponding P values. The model was adjusted for type of surgery (ie, surgical specialty) and type of anesthesia (ie, general anesthesia, peripheral regional, and neuraxial). The distribution of the constructed outcome (ie, anxiety difference) was considered normal according to the obtained histogram and Q–Q plot. The Kolmogorov–Smirnov test and Shapiro–Wilk test turned out to be statistical significant nonetheless. Neither the square root transformation nor the log transformation was adequate to achieve a normal distribution for the anxiety score including the residuals. Therefore, analysis of variance was still performed because this method has been shown to be resilient for nonnormality.14–17 It was checked whether there were influential observations (standardized residuals ≥±3.0). There was 1 influential observation. The equality of error variances was evaluated by the Levene’s test, which was nonsignificant (P = .34). The percentage of variance explained by the model was based on the coefficient of determination (R2).
To get a full understanding of the anxiety difference, we analyzed the effect of UGT2B15 genotype in combination with treatment condition, patient sex, and preoperative anxiety score. Therefore, we had to test for higher order interactions. Significant statistical interaction implies that the used variables act dependently on the outcome variable.18 All possible interactions with either UGT2B15 genotype or treatment condition were tested simultaneously for significance and all main terms. Interaction between variables means that the effect of variable x1 is modified by the value of variable x2, and x3, etc. When a higher-order interaction appeared to be statistically significant, the outcome is explained by the combination of multiple variables. Consequently, possible significant effects of lower order interactions are less relevant if a higher-order interaction turned out to be significant.19 Therefore, we only presented and described the results of the statistically significant highest order interaction for STAI-State.
All statistical testing took place at .05 level of statistical significance (2 sided). All analyses were done with SPSS software 20.0 (IBM Corp. Armonk, NY).
In the original study, 400 eligible patients were randomized to either lorazepam or placebo. Three hundred ninety-eight patients were analyzed.9 The number of missing values on the outcome variable did not appear to be significantly different between the lorazepam or placebo group (Fisher exact test, 0.22). Furthermore, the number of missing values of UGT2B15 genotype did not differ between these 2 groups (Fisher exact test, 0.84).
The study population included 224 men (56%) and had a mean age of 39.4 years (standard deviation, 13.6). The groups made by randomization have an equally distributed number of patients. For UGT2B15*2 polymorphism, 185 patients (50%) were heterozygous and 102 patients (27%) were homozygous deviant. The wild-type group comprised 85 patients (23%). These frequencies are consistent with Hardy-Weinberg equilibrium (P = .95, Pearson χ2 test). Overall, the level of anxiety was reduced postoperatively (Tables 1 and 2).
All main terms together with all possible interactions including UGT2B15 genotype and treatment condition are presented in Table 3. The highest-order interaction that was statistically significant comprised the following variables: UGT2B15 genotype, treatment condition, patient sex, and preoperative anxiety score (F = 7.15; df = 1, 339; P = .008).
Interpretation of higher-order interactions can be complex.19 For a more detailed understanding of this interaction and its clinical meaning and significance, the median differences and other descriptive statistics of patient categories on STAI-State according to the used determinants are presented and visualized in Appendix 1 and Figure 2, respectively. Because the preoperative measurement of STAI-State is continuous, we dichotomized the scores at the overall mean level of STAI-State to make visualization feasible. Anyway, this did not affect the results.
According to STAI-State descriptive statistics (Appendix 1), median differences range from −0.50 to 18.00. As mentioned previously, positive median differences represent a decrease of anxiety postoperatively.
Patients With Low Preoperative Anxiety Scores
Median differences of patients with low preoperative anxiety scores showed a range from −0.50 to 7.00. All female groups together with males in the placebo condition showed anxiety reduction in the postoperative period independently of UGT2B15 genotype. This was not observed for males in the lorazepam condition: males with normal UGT2B15 genotype showed no anxiety reduction, whereas males with deviant UGT2B15 genotype did show anxiety reduction (Figure 2).
Patients With High Preoperative Anxiety Scores
Median differences of patients with high preoperative anxiety scores showed a range from 9.00 to 18.00. Thus, all patient categories showed anxiety reduction in the postoperative period. Males in the lorazepam condition with normal UGT2B15 genotype showed greater anxiety reduction than males with deviant UGT2B15 genotype. In contrast, females in the lorazepam condition showed opposite results (Figure 2, Appendix 1).
Males in the placebo condition with normal UGT2B15 genotype showed less anxiety reduction than males with deviant UGT2B15 genotype. In contrast, females in the placebo condition showed opposite results (Figure 2, Appendix 1).
The principal finding of this study is that the UGT2B15 genotype has an influence on postoperative anxiety levels in same-day surgery patients receiving a single dosage of lorazepam as premedication. This clinical effect also depends on patient sex and preoperative anxiety score. Our finding supports data from previous research showing a large variability in clinical effects of lorazepam in healthy volunteers.5
For the anxiety median differences, most clinical significant results were seen in patients with high preoperative anxiety scores (Figure 2). They showed more reduction in anxiety compared with patients with low preoperative anxiety scores. Therefore, we believe that the significant highest-order interaction is predominately driven by patients with high preoperative anxiety scores. In light of clinical significance (ie, median differences ≥9.0 on a 20–80 point scale),20 we discuss these patient categories only.
The figures show that anxiety reduction in patients treated with placebo is likely because of the well-known placebo effect, because pharmacologic interaction between placebo and UGT2B15 genotype is not known. Preoperative administration of placebo is known to reduce postoperative anxiety.9
Concerning postoperative anxiety reduction in patients treated with lorazepam, we expected more anxiety reduction in homozygous carriers of the UGT2B15*2 genotype.5 Therefore, of interest was the finding that anxiety reduction in females was almost 2 times greater compared with males with this genotype. Jackson et al10 showed that sex differences after a single dosage of lorazepam is not likely because of pharmacokinetic differences and suggested that endogenous levels of neurosteroid hormones is a feasible explanation for the greater clinical effect of lorazepam in females. Moreover, one must take into account that levels of endogenous neurosteroid hormones, such as allopregnanolone, increase in stressful situations.21 Allopregnanolone—a metabolite of the ovarian hormone progesterone—is one of the most potent endogenous neurosteroid hormones having the ability to positively modulate the γ-aminobutyric acid receptor type A.22 Furthermore, it has been shown that allopregnanolone enhances the effects of benzodiazepine receptor agonists.23–25
Although explorative, this result may be of clinical relevance. It has been reported that almost half (49%) of the same-day surgery patients request preoperative anxiolytic premedication.26 Administration of anxiolytic premedication traditionally focuses on the preoperative stress of a patient, but postoperative recovery has become of increasing importance, which has to be considered when the anesthesiologist decides to prescribe premedication. On the basis of recent randomized clinical trials,9,27 anesthesiologists could negatively counsel patients in their preoperative assessment with respect to administration of premedication with lorazepam. However, our results might indicate that preoperative anxious female patients requesting premedication with lorazepam might benefit from UGT2B15*2 genotyping because homozygous carriers show more postoperative anxiety reduction.
The results and application of this study should be interpreted cautiously and within the context of the study methodology. Although we now have broader evidence that the UGT2B15 genotype is not an independent determinant for the clinical effects of lorazepam, there could be still an explanation for this finding. First, this might be explained by the single dosage of intravenous lorazepam in this study. Second, the dosage of lorazepam administered could be too low to be influenced by UGT2B15 genotype polymorphism. The very reason that we have refrained from administering a greater dosage was that it could provoke unwanted side effects such as drowsiness,4 which would lead to delayed recovery, which is not suitable for patients undergoing same-day surgery. Third, we refrained from significance testing on differences between the patient categories because of an insufficient number in these patient categories and to avoid multiple testing (Appendix 1). Fourth, patients with low preoperative anxiety scores could have biased our results because those patients do not have the ability to show a significant reduction in anxiety. In addition, although the analysis was adjusted for type of surgery and type of anesthesia to control for possible confounding, any medications given perioperatively and postoperative side effects (eg, nausea, drowsiness and pain) could still have biased the results and should be therefore subjected to further research. Fifth, because we do not have pharmacokinetic data in this study, we assume that lorazepam plasma concentration differences have played a role. The rationale for studying the clinical effects of lorazepam on the UGT2B15 genotype was the known significant association between lorazepam pharmacokinetics and the UGT2B15 genotype.5 Furthermore, in clinical practice, lorazepam is administered because of its clinical effects especially from a patient perspective. Sixth, our findings may not be necessarily extrapolated to other ethnic groups because it is known that ethnic origin could influence pharmacogenetics.28 Our results are in line, however, with previous findings on other clinical effects of lorazepam according to UGT2B15 genotype polymorphisms found in Korean people.5 Furthermore, our study population has similar UGT2B15 genotype frequencies compared with previous Caucasian populations studied.29,30
Although it is known that pharmacogenetics influence the clinical response to perioperative drugs, routine screening before pharmacotherapy is still not cost-effective in clinical anesthesiology as we administer drugs to a large number of patients and frequently once only.31 It is recommended, however, that clinical studies evaluating drug response in anesthesiology include pharmacogenetic testing.31,32 Pharmacogenetic data obtained could be used, for example, for diagnostic purposes and risk stratification and could be of help to progress in new clinical studies in anesthesiology.31
In case of premedication with lorazepam, we cannot recommend routine screening of UGT2B15 genotype at the moment. However, this study does define a population of clinical interest for future research. Future studies should be tailored to patients with high preoperative anxiety scores to gain more firm conclusions regarding the interaction between lorazepam and UGT2B15 genotype on postoperative anxiety.
UGT2B15 genotype influences postoperative anxiety levels in same-day surgery patients receiving lorazepam premedication. In patients with high preoperative anxiety scores treated with lorazepam, anxiety reduction was greater in females with the homozygous UGT2B15*2 genotype compared with males with the same genotype polymorphism. To gain a deeper understanding in the clinical relevance of the interaction between lorazepam and UGT2B15 genotype on postoperative anxiety levels, future studies—especially RCTs—focusing on patients with high preoperative anxiety scores would be required.
Appendix 1. Descriptive Statistics of STAI-State Distinguished by Treatment Condition, UGT2B15 Genotype Polymorphism, Patient Sex, and Preoperative Anxiety Score
Name: Herjan Mijderwijk, MD, MSc.
Contribution: This author helped design the study, conduct the study, collect the data, analyze the data, and prepare the manuscript.
Name: Markus Klimek, MD, PhD, DEAA, EDIC.
Contribution: This author helped design the study, conduct the study, analyze the data, and prepare the manuscript.
Name: Stefan van Beek, MSc.
Contribution: This author helped conduct the study, collect the data, and prepare the manuscript.
Name: Ron H. N. van Schaik, PhD.
Contribution: This author helped analyze genotype and prepare the manuscript.
Name: Hugo J. Duivenvoorden, PhD.
Contribution: This author helped design the study, conduct the study, analyze the data, and prepare the manuscript.
Name: Robert Jan Stolker, MD, PhD.
Contribution: This author helped design the study, analyze the data, conduct the study, and prepare the manuscript.
This manuscript was handled by: Ken B. Johnson, MD.
1. de Wildt SN, Kearns GL, Leeder JS, van den Anker JN. Glucuronidation in humans. Pharmacogenetic and developmental aspects. Clin Pharmacokinet. 1999;36:439452.
2. Rowland A, Miners JO, Mackenzie PI. The UDP-glucuronosyltransferases: their role in drug metabolism and detoxification. Int J Biochem Cell Biol. 2013;45:11211132.
3. Tukey RH, Strassburg CP. Human UDP-glucuronosyltransferases: metabolism, expression, and disease. Annu Rev Pharmacol Toxicol. 2000;40:581616.
4. Ameer B, Greenblatt DJ. Lorazepam: a review of its clinical pharmacological properties and therapeutic uses. Drugs. 1981;21:162200.
5. Chung JY, Cho JY, Yu KS, et al. Effect of the UGT2B15 genotype on the pharmacokinetics, pharmacodynamics, and drug interactions of intravenous lorazepam in healthy volunteers. Clin Pharmacol Ther. 2005;77:486494.
6. Lévesque E, Beaulieu M, Green MD, Tephly TR, Bélanger A, Hum DW. Isolation and characterization of UGT2B15(Y85): a UDP-glucuronosyltransferase encoded by a polymorphic gene. Pharmacogenetics. 1997;7:317325.
7. Perkins AM, Ettinger U, Davis R, Foster R, Williams SC, Corr PJ. Effects of Lorazepam and citalopram on human defensive reactions: ethopharmacological differentiation of fear and anxiety. J Neurosci. 2009;29:1261712624.
8. Olkkola KT, Ahonen J. Midazolam and other benzodiazepines. Handb Exp Pharmacol
9. Mijderwijk H, van Beek S, Klimek M, Duivenvoorden HJ, Grüne F, Stolker RJ. Lorazepam does not improve the quality of recovery in day-case surgery patients: a randomised placebo-controlled clinical trial. Eur J Anaesthesiol. 2013;30:743751.
10. Jackson A, Stephens D, Duka T. Gender differences in response to lorazepam in a human drug discrimination study. J Psychopharmacol. 2005;19:614619.
11. Greenblatt DJ, Ehrenberg BL, Gunderman J, et al. Kinetic and dynamic study of intravenous lorazepam: comparison with intravenous diazepam. J Pharmacol Exp Ther. 1989;250:134140.
12. van der Ploeg HM, Defares PB, Spielberger CD. Handleiding bij de Zelf Beoordelings Vragenlijst, een nederlandstalige bewerking van de Spielberger Stait-Trait Anxiety Inventory, STAI-DY.
1980Lisse: Swets & Zeitlinger.
13. He X, Hesse LM, Hazarika S, et al. Evidence for oxazepam as an in vivo probe of UGT2B15: oxazepam clearance is reduced by UGT2B15 D85Y polymorphism but unaffected by UGT2B17 deletion. Br J Clin Pharmacol. 2009;68:721730.
14. Feir-Walsh BJ, Toothaker LE. An empirical comparison of the ANOVA F-test, normal scores test and Kruskal-Wallis test under violation of assumptions. Educ Psychol Meas
15. Schmider E, Ziegler M, Danay E, Beyer L, Bühner M. Is it really robust? Methodology. 2010;6:147151.
16. Ragen BJ, Maninger N, Mendoza SP, Bales KL. The effects of morphine, naloxone, and κ opioid manipulation on endocrine functioning and social behavior in monogamous titi monkeys (Callicebus cupreus). Neuroscience. 2015;287:3242.
17. Kenkel WM, Yee JR, Porges SW, Ferris CF, Carter CS. Cardioacceleration in alloparents in response to stimuli from prairie vole pups: the significance of thermoregulation. Behav Brain Res. 2015;286:7179.
18. Petrie A, Sabin C. Medical Statistics at a Glance
. 20093rd ed. Oxford: Wiley Blackwell Publising.
20. Kain ZN. The legend of the P value. Anesth Analg. 2005;101:14541456.
21. Droogleever Fortuyn HA, van Broekhoven F, Span PN, Bäckström T, Zitman FG, Verkes RJ. Effects of PhD examination stress on allopregnanolone and cortisol plasma levels and peripheral benzodiazepine receptor density. Psychoneuroendocrinology. 2004;29:13411344.
22. Lephart ED, Lund TD, Horvath TL. Brain androgen and progesterone metabolizing enzymes: biosynthesis, distribution and function. Brain Res Brain Res Rev. 2001;37:2537.
23. Lambert JJ, Belelli D, Harney SC, Peters JA, Frenguelli BG. Modulation of native and recombinant GABA(A) receptors by endogenous and synthetic neuroactive steroids. Brain Res Brain Res Rev. 2001;37:6880.
24. Lambert JJ, Belelli D, Peden DR, Vardy AW, Peters JA. Neurosteroid modulation of GABAA receptors. Prog Neurobiol. 2003;71:6780.
25. Rupprecht R. Neuroactive steroids: mechanisms of action and neuropsychopharmacological properties. Psychoneuroendocrinology. 2003;28:139168.
26. van den Berg AA. Towards needleless induction of anaesthesia. Anaesthesia. 2003;58:806807.
27. Maurice-Szamburski A, Auquier P, Viarre-Oreal V, et al.; PremedX Study Investigators. Effect of sedative premedication on patient experience after general anesthesia: a randomized clinical trial. JAMA. 2015;313:916925.
28. Poolsup N, Li Wan Po A, Knight TL. Pharmacogenetics and psychopharmacotherapy. J Clin Pharm Ther. 2000;25:197220.
29. Lampe JW, Bigler J, Bush AC, Potter JD. Prevalence of polymorphisms in the human UDP-glucuronosyltransferase 2B family: UGT2B4(D458E), UGT2B7(H268Y), and UGT2B15(D85Y). Cancer Epidemiol Biomarkers Prev. 2000;9:329333.
30. Stringer F, Scott G, Valbuena M, Kinley J, Nishihara M, Urquhart R. The effect of genetic polymorphisms in UGT2B15 on the pharmacokinetic profile of sipoglitazar, a novel anti-diabetic agent. Eur J Clin Pharmacol. 2013;69:423430.
31. Iohom G, Fitzgerald D, Cunningham AJ. Principles of pharmacogenetics—implications for the anaesthetist. Br J Anaesth. 2004;93:440450.
Copyright © 2016 International Anesthesia Research Society
32. Landau R, Bollag LA, Kraft JC. Pharmacogenetics and anaesthesia: the value of genetic profiling. Anaesthesia. 2012;67:165179.