t
Pain Medicine Section Editor: Spencer S. Liu
Catechol-O-Methyltransferase Polymorphisms Predict Opioid Consumption in Postoperative Pain Keith A. Candiotti, MD, Zhe Yang, MD, PhD, David Buric, BS, Kris Arheart, PhD, Yanping Zhang, PhD, Yiliam Rodriguez, MD, Melvin C. Gitlin, MD, Enisa Carvalho, MD, Isabel Jaraba, MD, and Liyong Wang, PhD BACKGROUND: Previous studies have associated the catechol-O-methyltransferase (COMT) enzyme rs4680 polymorphism with opioid consumption in the treatment of chronic cancer pain. In this study, we evaluated the association between COMT rs4680 and rs4818 polymorphisms and opioid consumption in the acute postoperative period after a nephrectomy. METHODS: Opioid consumption and pain scores were evaluated in 152 patients for 48 hours after nephrectomy. The genotype of each patient was determined using polymerase chain reaction on DNA extracted from blood samples. The association between rs4680 and rs4818 genotypes and opioid consumption was evaluated using general linear model regression analysis. All P values and confidence intervals were Bonferroni corrected for the 3 comparisons among genotypes. RESULTS: In the 24-hour period after surgery (COMT rs4680), patients homozygous for the variant Val/Val consumed 36% (95% confidence interval, 31%–41%) more opioids than patients homozygous for the Met/Met group (P = 0.009). No statistically significant differences among the 3 genotype groups were noted for pain scores or emesis medication use in the first 24 hours after surgery. There was a statistically significant increase in emesis medication use in patients possessing the CC genotype of rs4818 when compared to patients carrying the GG genotypes (P = 0.035). In the 6- to 48-hour postsurgery period, there was significantly higher opioid consumption in the high-activity homozygotes Val/Val than in the homozygous Met/Met group for COMT rs4680 (0–6 h: P = 0.005; 0–12 h: P = 0.015; 0–24 h: P = 0.015; and 0–48 h: P = 0.023). Patients in the homozygous GG group COMT rs4818 single nucleotide polymorphism showed statistically significant differences in opioid consumption in the first 6 hours after nephrectomy compared with heterozygous CG patients (P = 0.02). CONCLUSIONS: The genetic variant of the COMT rs4680 single nucleotide polymorphism is associated with variability in opioid consumption in postoperative nephrectomy patients. The COMT rs4818 polymorphism may prove useful in predicting emesis medication use postoperatively. (Anesth Analg 2014;119:1194–200)
O
pioids, such as morphine and oxycodone, are used as analgesics in the postoperative period. However, opioid consumption varies widely among individuals.1 Catechol-O-methyltransferase (COMT) metabolizes biogenic amines including catecholamines such as dopamine, epinephrine, and norepinephrine, thereby acting as a key modulator of dopaminergic and adrenergic/ noradrenergic neurotransmission. The dopamine and norepinephrine systems also influence opioid behaviors. Thus, variability in the COMT gene may contribute to the variability of opioid consumption.2 COMT contains a common From the Department of Anesthesiology, Perioperative Medicine and Pain Management, University of Miami School of Medicine, Miami, Florida. Accepted for publication July 14, 2014. Funding: This work was supported by the Department of Anesthesiology, Perioperative Management and Pain Medicine, University of Miami, Miller School of Medicine. The authors declare no conflicts of interest. This report was previously presented, in part, at the International Anesthesia Research Society’s 2013 Annual Meeting. Drs. Yang and Candiotti contributed equally to this manuscript. Reprints will not be available from the authors. Address correspondence to Keith A. Candiotti, MD, Department of Anesthesiology, Perioperative Medicine and Pain Management, University of Miami School of Medicine/Jackson Memorial Hospital, 1611 NW 12th Ave., Central 130(R-370), Miami, FL 33101. Address e-mail to [email protected]. Copyright © 2014 International Anesthesia Research Society DOI: 10.1213/ANE.0000000000000411
1194 www.anesthesia-analgesia.org
functional coding polymorphism rs4680, also known as COMT Val158Met, which substitutes valine with methionine at amino acid position 158, leading to a 3- to 4-fold reduction in the activity of the COMT enzyme.3 Zubieta et al.4 examined this polymorphism’s effect on pain duration in humans. Individuals with a homozygous Met158 genotype reported greater sensory and affective ratings of pain and showed a reduced activation of the endogenous opioid system on experimental pain stimuli and a higher regional density of µ opioid receptors in the brain. Additionally, it has been demonstrated that the rs4680 (Val158Met) polymorphism of the human COMT gene contributed to differences in opioid consumption in patients treated for chronic cancer pain.5,6 A meta-analysis performed by Tammimaki and Mannisto7 showed that the COMT single nucleotide polymorphism (SNP) rs4680 was associated with chronic pain only in patients suffering from fibromyalgia or chronic widespread pain. However, the results of studies examining the interaction of COMT rs4680 polymorphism and its effect on opioid consumption in the presence of acute pain have been negative.8–10 Furthermore, it has been shown that there is no association between the COMT rs4680 polymorphism and opioid consumption in acute pain therapy.10 Two studies performed by Henker et al.9 and Lee et al.11 did not report an association between the COMT rs4680 polymorphism and opioid consumption in acute pain therapy, but did report an association November 2014 • Volume 119 • Number 5
between another COMT polymorphism, rs4818, and opioid consumption in the context of acute pain. Unlike the rs4680 SNP, rs4818 is a synonymous mutation that is thought to affect COMT enzyme translation through alterations in mRNA secondary structure.12 Nevertheless, both studies were performed in relatively small samples groups (79 and 98 patients, respectively), and Henker et al.’s study consisted of only Caucasians and did not reach HardyWeinberg equilibrium, compelling us to further examine these effects. Moreover, no association between the rs4680 polymorphism and acute pain ratings was found in the study performed by Kim et al.; however, a significant association between another COMT SNP, rs740603, and maximal postoperative pain ratings was noted. SNP rs740603 is, however, relatively rare and was omitted from this study in favor of the more common SNPs rs4680 and rs4818. In this study, we evaluated the association between COMT rs4680 and rs4818 polymorphisms and opioid consumption during the acute postoperative period in 152 patients undergoing a nephrectomy, a painful procedure, using an increased sample size more than the previously mentioned studies.
METHODS
After obtaining IRB approval (University of Miami, Miami, FL) and written informed consent, those patients who were scheduled to undergo an elective nephrectomy and were classified as ASA physical status I to III were enrolled. Patients provided medical histories and demographic information including height, weight, age, sex, ethnicity, and smoking history.
Inclusion Criteria
Patients scheduled to undergo a laparoscopic or open nephrectomy using only general anesthesia were eligible. Nephrectomy patients were chosen due to the significant pain experienced after surgery. Patients were excluded if they were currently menstruating females, had undergone a major surgical procedure within the past 30 days, were scheduled to receive an epidural for pain control intraoperatively or postoperatively, were not willing to remain in the study for at least 72 hours, had received any experimental medications within the past 30 days, or had allergies to any of the protocol medications. Patients were premedicated and anesthetized according to the protocol, previously described.13 Briefly, the patients were given midazolam 1 to 2 mg in the preoperative area. Induction of anesthesia was accomplished with propofol 2 to 3 mg/kg, lidocaine 1 mg/ kg and fentanyl 3 µg/kg, and neuromuscular blockade was achieved with cisatracurium 0.2 mg/kg. Endotracheal intubation was used in all patients, and fentanyl and cisatracurium were redosed at the discretion of the anesthesiologist. Anesthesia was maintained using sevoflurane (0%–4.0%) and O2 (50%–70%) titrated to a bispectral index (Covidien, Mansfield, MA) between 45 and 60. Thirty minutes before tracheal extubation, either ondansetron 4 mg or granisetron 0.1 mg was administered. Neuromuscular blockade was reversed with neostigmine 0.05 mg/kg and glycopyrrolate 0.01 mg/kg. All standard vital signs were monitored, and bupivacaine 0.5% was used locally at all incision sites.13 Postoperatively, IV morphine was initially administered on request in the postanesthesia care unit and then by
November 2014 • Volume 119 • Number 5
patient-controlled anesthesia in all patients for the first 72 hours of the postoperative period. A visual analog scale for pain was assessed at 6, 12, 24, 30, 36, 48, and 72 hours postoperatively. For the few patients given oral Percocet 5/325 (oxycodone hydrochloride, 5 mg, and acetaminophen, 325 mg, in each tablet; Endo Pharmaceuticals, Chadds Ford, PA) in addition to morphine, an equivalency factor of oral oxycodone/IV morphine of 1:0.333 was used.10 Laboratory staff members responsible for genotyping were blinded to the clinical results.
DNA Extraction
Blood samples were collected in tubes containing EDTA 1 hour before surgery and were stored at −80°C. DNA was extracted upon sample arrival from whole-blood samples, with QIAamp DNA Blood Midi Kit (QIAGEN, Valencia, CA). DNA samples were then stored at −20°C.
Genotyping
Two SNPs, rs4680 and rs4818, were chosen to evaluate COMT based on previously identified haplotypes.5,6,9,10 We genotyped all COMT SNPs using the TaqMan assay (Applied Biosystems, Foster City, CA). We conducted amplification and genotype assignments using the ABI7000 and SDS 2.0 software (Applied Biosystems). For all genotyping conducted during this study, double-masked genotyping assignments were made for each variant, which were compared, and each discrepancy was addressed using raw data or by regenotyping.
Statistical Analysis
The study population was a convenience sample of patients treated at our facility collected between July 2005 and January 2013. All patients who met our inclusion criteria were offered the opportunity to enroll in this study. One hundred seventy-two patients were enrolled, but 20 patients electively withdrew for unspecified reasons. Consequently, we were left with 152 study subjects for data analysis. Of the 152 study subjects, body mass index (BMI) data were missing for 2, pain score was missing from 1, and the procedure information was missing from 1. Power analysis was done using QUANTO (University of Southern California, Los Angeles, CA). With the sample size of 152, we had >80% power to detect an effect size of 10 (for 24-hour accumulative opium consumption), at an α level of 0.05. The 3 genotypes of COMT rs4680 and rs4818 were treated as independent variables. We calculated summary statistics for all variables in genotypes rendered by the SNP of COMT. General linear model analysis was used to assess whether there were significant differences between genotypes for numerical variables. Linear models allow a flexible analysis that uses linear regression techniques to analyze a host of different analysis of variance and regression problems for continuous data. This technique can be generalized to accommodate other distributions such as binomial, Poisson, and negative binomial. Clinical and demographic continuous data were summarized as mean ± SE. Generalized linear models were used to analyze the association between the genotypes and binary variables. Binary clinical data were summarized as percent and SE. Some categorical data were analyzed with a Pearson χ2 test for equality
www.anesthesia-analgesia.org 1195
COMT Polymorphisms Predict Opioid Consumption in Acute Pain
Table 1. Demographic and Clinical Characteristics of COMT rs4680 Genotype Groups Variable Age (y)a Maleb Race/ethnicityb Hispanic Caucasian African American BMI (kg/m2)a Smokerb Open procedureb Fentanyl (µg/kg/h) Surgical duration (h)
COMT GG 52.6 ± 3.4 21 (67.7%)
COMT AG 54.3 ± 2.5 38 (63.3%)
COMT AA 50.7 ± 2.6 35 (57.4%)
8 (25.8%) 20 (64.5%) 3 (9.7%) 29.7 ± 1.1 16 (51.6%) 24 (77.4%) 1.39 ± 0.11 3.90 ± 1.22
23 (38.3%) 31 (51.7%) 6 (10.0%) 27.5 ± 0.8 21 (35.6%) 44 (73.3%) 1.41 ± 0.15 3.97 ± 1.28
31 (50.8%) 21 (34.4%) 9 (14.8%) 28.2 ± 0.8 21 (34.4%) 41 (68.3%) 1.50 ± 1.20 3.87 ± 1.15
P 0.577 0.597 0.080
Statistical methods General linear χ2 χ2
0.256 0.237 0.636 0.52 0.4
General linear χ2 χ2 General linear General linear
Continuous variables expressed as amean ± SE and bfrequency (%). Discrete variables are given as counts. BMI = body mass index; open = open nephrectomy; LSC = laparoscopically assisted nephrectomy.
Table 2. Twenty-Four-Hour Cumulative Opioid Consumption, Pain Score, and Emesis Medication Usage Among COMT rs4680 Different Genotypes Cumulative opioid consumption (mg) Pain score Emesis medicine usage
Met/Met n = 61 31.9 ± 27.5 3.13 ± 2.77 13.6%
Val/Met n = 60 37.5 ± 27.3 3.56 ± 2.83 15.3%
Val/Val n = 31 49.4 ± 38.9 4.16 ± 2.82 16.1%
P value Met/Met versus Val/Val 0.020* 0.249 0.939
Adjusted mean ± SD. Opioid model includes a significant covariate for smoking status (P = 0.030); there were no significant interactions between genotype and any covariate. Pain score model includes significant covariates for ethnicity (P = 0.016) and age (P = 0.030); there were no significant interactions between genotype and any covariate. Emesis model includes a significant covariate for gender (P = 0.0070); there were no significant interactions between genotype and any covariate. *P < 0.05
of proportions to assess the significance of differences among genotypes. This type of data was summarized as absolute frequency and percent. To assess potential confounding of genotype effects and clinical characteristics, analyses with and without covariates (i.e., age, sex, smoking, ethnic origin, BMI, and surgery type), and interactions between genotype and covariates were performed. Interactions not significant at the 0.050 level were removed first followed by any covariate that was not significant at the 0.050 level. All P values and confidence intervals were Bonferroni corrected for the 3 comparisons among genotypes. SAS 9.3 (SAS Institute Inc., Cary, NC) was used for all analyses.
RESULTS
One hundred fifty-two patients undergoing a nephrectomy, who met inclusion criteria, were included in the study. COMT rs4680 and rs4818 genotype frequencies are listed as follows. For rs4680, all the subjects were distributed into 3 genotype groups: Met/Met homozygotes (n = 61), Val/Met (n = 60), and Val/Val (n = 31). This polymorphism was in HardyWeinberg equilibrium; no distortion of genotype distribution compared with the expected distribution from the allele frequencies was noted. For rs4818, all the subjects were distributed into 3 genotype groups: CC homozygotes (n = 63), CG (n = 68), and GG (n = 21). This polymorphism was also found to be in HardyWeinberg equilibrium. The clinical background characteristics of the 3 genotype groups for cumulative 24-hour opioid consumption for rs4680 were compared (Table 1), with no statistically significant differences noted among them. Twenty-four-hour opioid consumption among the 3 rs4680 COMT genotype groups reached statistical significance (P = 0.020) with the Val/Val group demonstrating
1196 www.anesthesia-analgesia.org
a higher 24-hour opioid consumption when compared with that of the Met/Met group (P = 0.009) (Table 2). After adjusting for age, sex, ethnicity, BMI, smoking, and surgical approach using the general linear model, the association between 24-hour opioid consumption and the 3 COMT rs4680 genotypes still reached statistical significance (P = 0.032). There were no statistically significant differences among the 3 genotype groups for 24-hour postoperative pain scores or requirements for medication to treat nausea or vomiting (Table 2). The emesis model includes a significant covariate for gender status (P = 0.007); there were no significant interactions between genotype and any covariate. Cumulative 24-hour opioid consumption was compared based on ethnicity, age, sex, BMI, smoking history, and surgical type (Table 3). Smokers had a higher 24-hour morphine consumption compared to nonsmokers (P = 0.014). The ethnicity, age, sex, BMI, and surgical type did not show a significant effect on morphine consumption (all P ≥ 0.067). Differences in cumulative 24-hour opioid consumption among the 3 COMT rs4680 genotypes with respect to several clinically relevant variables were compared post hoc and are summarized. Twenty-four-hour opioid consumption among the 3 rs4818 COMT genotype groups did not reach statistical significance (P = 0.223) (Table 4). There were no statistically significant differences among the 3 genotype groups for 24-hour postoperative pain scores (P = 0.250). There were statistically significant differences among the 3 genotype groups (P = 0.030) for 24-hour postoperative requirements for medication to treat nausea or vomiting with the CC group consuming more antiemetic drugs over a 24-hour period when compared with that of the GG group (P = 0.035) (Table 4). The emesis model includes a significant
anesthesia & analgesia
Table 3. Twenty-Four-Hour Cumulative Opioid Consumption Based on Different Clinical Variables Variable Race/ethnicity
Age (y)
Gender BMI
Smoker Surgery Type
Value Hispanic (n = 62) Caucasian (n = 72) African American (n = 18) ≤40 (n = 30) 41–50 (n = 26) 51–60 (n = 45) 61–70 (n = 30) ≥71 (n = 21) Male (n = 94) Female (n = 58) ≤25 (n = 49) 26–29 (n = 57) 30–35 (n = 33) ≥36 (n = 9) Yes (n = 58) No (n = 93) Open (n = 109) LSC (n = 42)
24-h opioid consumption (mg) 34.5 ± 3.9 40.8 ± 3.6 28.7 ± 7.2 37.7 ± 5.6 42.2 ± 6.0 42.0 ± 4.5 28.5 ± 5.6 29.6 ± 6.7 38.2 ± 3.2 34.5 ± 4.0 35.6 ± 4.4 40.0 ± 4.1 31.0 ± 5.3 54.7 ± 10.2 44.6 ± 4.0 32.0 ± 3.1 37.4 ± 3.0 35.2 ± 4.8
P 0.244
0.242
0.478 0.183
0.014* 0.687
All variables analyzed with a general linear model. BMI = body mass index; LSC = laparoscopically assisted nephrectomy. *P < 0.05
covariate for gender status (P = 0.007); there were no significant interactions between genotype and any covariate. During the 6- to 48-hour postoperative period, there was significantly higher opioid consumption in the 6-, 12-, 24-, and 48-hour time intervals in patients in the COMT rs4680 Val/Val group (high enzymatic activity) when compared with patients in the Met/Met group (low enzymatic activity) (0–6 h: P = 0.005; 0–12 h: P = 0.015; 0–24 h: P = 0.015; and 0–48 h: P = 0.023) (Fig.1). The analysis model included covariates for smoking status (P = 0.043), and there were no significant interactions between genotype and any covariate. In the heterozygous Val/Met group, the opioid consumption was higher than the Met/Met group and lower than the Val/ Val group in the 48-hour postoperative time interval, with the difference between Val/Met and Val/Val reaching statistical significance only in the first 6 hours postoperatively. During the 48-hour postoperative period, patients in the COMT rs4818 GG group consumed significantly more opioids in the first 6 hours after the operation when compared to patients in the heterozygous CG group (P = 0.020). Other comparisons between rs4818 genotypes did not show statistically significant differences in opioid consumption (Fig. 2). The analysis model included covariates for smoking status (P = 0.043); there were no significant interactions between genotype and any covariate.
DISCUSSION
Our current study has shown a strong contribution of the COMT rs4680 polymorphism to interindividual variations in opioid consumption in the acute postoperative period after nephrectomy. Subjects with the Val/Val genotype consumed 36% more opioids compared to subjects with Met/ Met genotype at 24 hours after the operation. In one previous study, Kolesnikov et al.10 reported that carriers of COMT rs4680 Met/Met consumed similar morphine doses compared with wild-type homozygous carriers
November 2014 • Volume 119 • Number 5
of Val/Val who underwent abdominal surgery for either prostatectomy or hysterectomy. Henker et al.9 also demonstrated that the genotypes of the COMT rs4680 polymorphism were not associated with morphine consumption in the postoperative period, but they found that the rs4818 genotype polymorphism was associated with opioid consumption with the GG genotype group consuming more opioids compared to those patients in the heterozygous CG and homozygous CC groups. However, in an interesting contrast, Rakvåg et al.5 found that the COMT rs4680 polymorphism had a statistically significant effect on morphine consumption in the treatment of chronic pain. Patients carrying the Val/Val genotype had significantly higher morphine consumption compared with those possessing the Met/Met genotype. Our results support the findings of Rakvåg et al. We found that there was significantly higher opioid consumption in the first 6 hours postoperatively in patients who were homozygous GG when compared with patients who were heterozygous CG for the rs4818 polymorphism (P = 0.015). However, while there was no statistically significant difference between the CC and GG groups, the homozygous GG group consumed a numerically higher total amount of opioids compared to the homozygous CC group (P = 0.120). Consequently, the decrease in opioid consumption in the CC and CG genotypes may possibly be attributed to C carriage. The difference between our result and Kolesnikov et al.’s10 and Henker et al.’s9 results may be attributable to different surgery types and sample size. Previously, genetically modified mice have been used to study the impact of altered COMT activity on pain.14,15 Mouse lines modified to overexpress the Val158 (Val/Val) variant, specifically in the prefrontal cortex, were studied and shown to have diminished pain sensitivity in both hotplate and tail flick testing.15 Concordant with these findings, COMT knockout mice were shown to have increased pain sensitivity.16 Zubieta et al.4 demonstrated that variations in Val158Met genotypes, and corresponding variation in levels of COMT activity, impact downstream functions regulated by monoamines in humans. They examined µ-opioid receptor density in different brain regions using positron emission tomography and found that individuals expressing the low-activity enzyme, corresponding to the Met/ Met genotype, possessed increased receptor density, while those expressing the high-activity enzyme, corresponding to the Val/Val genotype, possessed lower receptor density. Clinically, this suggests that patients expressing the Met/Met genotype and having higher opioid receptors in critical regions might require less opioid to produce adequate pain control, while those possessing the genotype Val/Val may require relatively more opioid. This idea is supported by a recent study which found that the rs4680 Met/Met genotype and the rs4818 CC genotype were both associated with lower preoperative pain scores and pain intensity 1 year after lumbar discectomy.17 Our data, demonstrating a lower requirement for opioids among patients carrying the Met/Met genotype of the COMT enzyme, are in agreement with these results. In general, our data support the concept that the Met/Met genotype of COMT rs4680 is associated with a low enzymatic activity
www.anesthesia-analgesia.org 1197
COMT Polymorphisms Predict Opioid Consumption in Acute Pain
Table 4. Twenty-Four-Hour Cumulative Opioid Consumption, Pain Score, and Emesis Medication Usage Among COMT rs4818 Different Genotypes Cumulative opioid consumption (mg) Pain score Emesis medicine usage
CC n = 61 36.3 ± 25.5 3.83 ± 2.80 15.8% ± 4.6%
CG n = 60 34.2 ± 24.3 3.28 ± 2.95 13.2% ± 4.9%
GG n = 31 46.9 ± 37.5 3.33 ± 2.65 4.5% ± 3.5%
P value CC versus GG 0.20 0.254 0.035*
Adjusted mean ± SD. Opioid model includes a significant covariate for smoking status (P = 0.030); there were no significant interactions between genotype and any covariate. Pain score model includes significant covariates for ethnicity (P = 0.016) and age (P = 0.030); there were no significant interactions between genotype and any covariate. Emesis model includes a significant covariate for gender status (P = 0.0070); there were no significant interactions between genotype and any covariate. *P < 0.05
Figure 1. Mean opioid consumption in 6, 12, 24 and 48 h for 3 genotype classes. The analysis model included covariates for smoking status (P = 0.043); there were no significant interactions between genotype and any covariate. Statistically significant differences between Val/Val and Val/Met at 0–6 h (P = 0.025) and between Met/Met and Val/Val occurred at 0–6 h (* = 0.0050), 0–12 h (** = 0.015), 0–24 h (# = 0.016), and 0–48 h (## = 0.023). Other comparisons between Met/Met and Val/Met and Val/Met and Val/Val genotypes were not significant. Bar represents 1 SE.
Figure 2. Mean opioid consumption in 6, 12, 24 and 48 h for 3 genotype classes. The analysis model included covariates for smoking status (P = 0.043); there were no significant interactions between genotype and any covariate. Statistically significant differences between CG and GG occurred only at 0–6 h (* = 0.020), but not at 0–12 h (** = 0.050), 0–24 h = 0.060), or 0–48 h (## = 0.090). Other comparisons between CC versus GG and CC versus CG genotypes were not significant. Bar represents 1 SE.
of dopamine, leading to an increase of µ -opioid receptor density in specific brain regions, resulting in a lower concentration of opioid being required to achieve the same degree of activation. To determine whether the genotype groups were different with regard to factors that may affect opioid consumption, we compared the groups with regard to several characteristics including smoking, operation duration, and surgery types. The different genotype groups showed similar results for these characteristics as for other confounders including age, gender, and ethnicity (Table 1). When opioid consumption during the 24-hour postoperative period was compared among these confounders (Table 3), our data showed that smoking increased opioid consumption during the 24-hour postoperative period. This result is consistent with our previous study,13 even with the addition of more patients. When smoking is added as a covariate, the association between morphine consumption during the 24-hour postoperative period and COMT genotypes remained statistically significant. When analyzing pain scores from the 24-hour postoperative period, we observed that the Val/Val genotype demonstrated the highest numerical pain scores, the Val/Met group an intermediate score, and the Met/Met genotype the lowest. However, these differences did not reach statistical significance (P = 0.249). The same trend was observed in the 6-, 12- and 48-hour postoperative pain scores (data
not shown). Overall, the interindividual variations of pain scores may be more pronounced than that of opioid consumption, so an increase in sample size may be required to see a statistically significant difference in future studies. Additionally, pain scores may be difficult to interpret in general due to the fact that patients had unrestricted access to pain medications. Kolesnikov et al.10 reported that COMT rs4680 was associated with the central side effects of opioids such as nausea and sedation in patients who underwent abdominal surgery for either prostatectomy or hysterectomy. In this study, we also investigated a possible correlation between morphine-related side effects (symptomatic nausea/vomiting) and the COMT rs4680 SNP in these patients. Our data indicated that the Met/Met genotype demonstrated the numerically lowest usage of emesis medication, the Val/ Met genotype demonstrated intermediate usage levels, and the Val/Val genotypes showed the highest usage; however, there were no statistically significant differences among the 3 genotype groups for emesis medication use at 24-hours after surgery. Our results are consistent with the study by Ross et al.18 that did not find any associations between the COMT rs4680 polymorphism and emesis medication usage. The discrepancy between our findings and those in Kolesnikov’s study may be attributable to their smaller sample size. Interestingly, we found that the rs4818 polymorphism is associated with variations in emesis medication
1198 www.anesthesia-analgesia.org
anesthesia & analgesia
use after surgery, with the CC genotype group using more emesis medication compared with the CG and GG genotype groups. This association did not correlate with opioid consumption, suggesting that genetic factors may play an independent role in postoperative nausea and vomiting. Finally, our study has limitations. First, this was strictly a gene association trial and, as such, opioid receptors and opioid concentration were not measured in the brain or blood, and the mechanism of our findings can only be assumed. Second, the study only enrolled patients having a nephrectomy, which is relatively painful. We believe that the notable level of pain caused by this procedure makes it an ideal acute pain model; however, it is possible that our findings may not apply to other less painful surgeries. Finally, we only focused on the most common COMT SNPs, rs4680 and rs4818, because these 2 SNPs had a previously reported association with opioid consumption in chronic pain and acute pain patients. In summary, our study demonstrates that the COMT rs4680 SNP is associated with differences in postoperative opioid consumption in patients who underwent a nephrectomy. The COMT rs4680 polymorphism may predict opioid consumption in postoperative patients, and the COMT rs4818 polymorphism may predict emesis medication use postoperatively. It should be noted that while genetic analysis appears useful in predicting the pain phenotype, practical use is likely years away. E DISCLOSURES
Name: Keith A. Candiotti, MD. Contribution: This author helped design the study and write the manuscript. Attestation: Keith A. Candiotti has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files. Name: Zhe Yang, MD, PhD. Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript. Attestation: Zhe Yang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: David Buric, BS. Contribution: This author helped write the manuscript. Attestation: David Buric has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Kris Arheart, PhD. Contribution: This author helped analyze the data and write the manuscript. Attestation: Kris Arheart has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Yanping Zhang, PhD. Contribution: This author helped conduct the study and write the manuscript. Attestation: Yanping Zhang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Yiliam Rodriguez, MD. Contribution: This author helped analyze the data and write the manuscript.
November 2014 • Volume 119 • Number 5
Attestation: Yiliam Rodriguez has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Melvin C. Gitlin, MD. Contribution: This author helped design the study and write the manuscript. Attestation: Melvin C. Gitlin has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Enisa Carvalho, MD. Contribution: This author helped with patient recruitment and processing of data. Attestation: Enisa Carvalho has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Isabel Jaraba, MD. Contribution: This author helped with patient recruitment and processing of data. Attestation: Isabel Jaraba has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Liyong Wang, PhD. Contribution: This author helped design the study and write the manuscript. Attestation: Liyong Wang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. This manuscript was handled by: Spencer S. Liu, MD. ACKNOWLEDGMENTS
The authors thank Dr. Odalys Rodriguez Bravo (research assistant, The Department of Anesthesiology, Perioperative Management and Pain Medicine, University of Miami, Miller School of Medicine) for expert technical assistance. REFERENCES 1. Coulbault L, Beaussier M, Verstuyft C, Weickmans H, Dubert L, Trégouet D, Descot C, Parc Y, Lienhart A, Jaillon P, Becquemont L. Environmental and genetic factors associated with morphine response in the postoperative period. Clin Pharmacol Ther 2006;79:316–24 2. Lachman HM, Papolos DF, Saito T, Yu YM, Szumlanski CL, Weinshilboum RM. Human catechol-O-methyltransferase pharmacogenetics: description of a functional polymorphism and its potential application to neuropsychiatric disorders. Pharmacogenetics 1996;6:243–50 3. Lotta T, Vidgren J, Tilgmann C, Ulmanen I, Melén K, Julkunen I, Taskinen J. Kinetics of human soluble and membranebound catechol O-methyltransferase: a revised mechanism and description of the thermolabile variant of the enzyme. Biochemistry 1995;34:4202–10 4. Zubieta JK, Heitzeg MM, Smith YR, Bueller JA, Xu K, Xu Y, Koeppe RA, Stohler CS, Goldman D. COMT val158met genotype affects mu-opioid neurotransmitter responses to a pain stressor. Science 2003;299:1240–3 5. Rakvåg TT, Klepstad P, Baar C, Kvam TM, Dale O, Kaasa S, Krokan HE, Skorpen F. The Val158Met polymorphism of the human catechol-O-methyltransferase (COMT) gene may influence morphine requirements in cancer pain patients. Pain 2005;116:73–8 6. Rakvåg TT, Ross JR, Sato H, Skorpen F, Kaasa S, Klepstad P. Genetic variation in the catechol-O-methyltransferase (COMT) gene and morphine requirements in cancer patients with pain. Mol Pain 2008;4:64 7. Tammimäki A, Männistö PT. Catechol-O-methyltransferase gene polymorphism and chronic human pain: a systematic review and meta-analysis. Pharmacogenet Genomics 2012;22:673–91
www.anesthesia-analgesia.org 1199
COMT Polymorphisms Predict Opioid Consumption in Acute Pain
8. Kim H, Lee H, Rowan J, Brahim J, Dionne RA. Genetic polymorphisms in monoamine neurotransmitter systems show only weak association with acute post-surgical pain in humans. Mol Pain 2006;2:24 9. Henker RA, Lewis A, Dai F, Lariviere WR, Meng L, Gruen GS, Sereika SM, Pape H, Tarkin IS, Gowda I, Conley YP. The associations between OPRM 1 and COMT genotypes and postoperative pain, opioid use, and opioid-induced sedation. Biol Res Nurs 2013;15:309–17 10. Kolesnikov Y, Gabovits B, Levin A, Voiko E, Veske A. Combined catechol-O-methyltransferase and mu-opioid receptor gene polymorphisms affect morphine postoperative analgesia and central side effects. Anesth Analg 2011;112:448–53 11. Lee PJ, Delaney P, Keogh J, Sleeman D, Shorten GD. Catecholamine-O-methyltransferase polymorphisms are associated with postoperative pain intensity. Clin J Pain 2011;27:93–101 12. Diatchenko L, Slade GD, Nackley AG, Bhalang K, Sigurdsson A, Belfer I, Goldman D, Xu K, Shabalina SA, Shagin D, Max MB, Makarov SS, Maixner W. Genetic basis for individual variations in pain perception and the development of a chronic pain condition. Hum Mol Genet 2005;14:135–43 13. Candiotti KA, Yang Z, Morris R, Yang J, Crescimone NA, Sanchez GC, Bird V, Leveillee R, Rodriguez Y, Liu H, Zhang YD, Bethea JR, Gitlin MC. Polymorphism in the interleukin-1 receptor antagonist gene is associated with serum interleukin-1
1200 www.anesthesia-analgesia.org
receptor antagonist concentrations and postoperative opioid consumption. Anesthesiology 2011;114:1162–8 14. Kambur O, Männistö PT, Viljakka K, Reenilä I, Lemberg K, Kontinen VK, Karayiorgou M, Gogos JA, Kalso E. Stressinduced analgesia and morphine responses are changed in catechol-O-methyltransferase-deficient male mice. Basic Clin Pharmacol Toxicol 2008;103:367–73 15. Papaleo F, Crawley JN, Song J, Lipska BK, Pickel J, Weinberger DR, Chen J. Genetic dissection of the role of catechol-O-methyltransferase in cognition and stress reactivity in mice. J Neurosci 2008;28:8709–23 16. Gogos JA, Morgan M, Luine V, Santha M, Ogawa S, Pfaff D, Karayiorgou M. Catechol-O-methyltransferase-deficient mice exhibit sexually dimorphic changes in catecholamine levels and behavior. Proc Natl Acad Sci U S A 1998;95:9991–6 17. Rut M, Machoy-Mokrzyńska A, Ręcławowicz D, Słoniewski P, Kurzawski M, Droździk M, Safranow K, Morawska M, Białecka M. Influence of variation in the catechol-O-methyltransferase gene on the clinical outcome after lumbar spine surgery for one-level symptomatic disc disease: a report on 176 cases. Acta Neurochir (Wien) 2014;156:245–52 18. Ross JR, Riley J, Taegetmeyer AB, Sato H, Gretton S, du Bois RM, Welsh KI. Genetic variation and response to morphine in cancer patients: catechol-O-methyltransferase and multidrug resistance-1 gene polymorphisms are associated with central side effects. Cancer 2008;112:1390–403
anesthesia & analgesia
Pain Medicine Section Editor: Spencer S. Liu
Catechol-O-Methyltransferase Polymorphisms Predict Opioid Consumption in Postoperative Pain Keith A. Candiotti, MD, Zhe Yang, MD, PhD, David Buric, BS, Kris Arheart, PhD, Yanping Zhang, PhD, Yiliam Rodriguez, MD, Melvin C. Gitlin, MD, Enisa Carvalho, MD, Isabel Jaraba, MD, and Liyong Wang, PhD BACKGROUND: Previous studies have associated the catechol-O-methyltransferase (COMT) enzyme rs4680 polymorphism with opioid consumption in the treatment of chronic cancer pain. In this study, we evaluated the association between COMT rs4680 and rs4818 polymorphisms and opioid consumption in the acute postoperative period after a nephrectomy. METHODS: Opioid consumption and pain scores were evaluated in 152 patients for 48 hours after nephrectomy. The genotype of each patient was determined using polymerase chain reaction on DNA extracted from blood samples. The association between rs4680 and rs4818 genotypes and opioid consumption was evaluated using general linear model regression analysis. All P values and confidence intervals were Bonferroni corrected for the 3 comparisons among genotypes. RESULTS: In the 24-hour period after surgery (COMT rs4680), patients homozygous for the variant Val/Val consumed 36% (95% confidence interval, 31%–41%) more opioids than patients homozygous for the Met/Met group (P = 0.009). No statistically significant differences among the 3 genotype groups were noted for pain scores or emesis medication use in the first 24 hours after surgery. There was a statistically significant increase in emesis medication use in patients possessing the CC genotype of rs4818 when compared to patients carrying the GG genotypes (P = 0.035). In the 6- to 48-hour postsurgery period, there was significantly higher opioid consumption in the high-activity homozygotes Val/Val than in the homozygous Met/Met group for COMT rs4680 (0–6 h: P = 0.005; 0–12 h: P = 0.015; 0–24 h: P = 0.015; and 0–48 h: P = 0.023). Patients in the homozygous GG group COMT rs4818 single nucleotide polymorphism showed statistically significant differences in opioid consumption in the first 6 hours after nephrectomy compared with heterozygous CG patients (P = 0.02). CONCLUSIONS: The genetic variant of the COMT rs4680 single nucleotide polymorphism is associated with variability in opioid consumption in postoperative nephrectomy patients. The COMT rs4818 polymorphism may prove useful in predicting emesis medication use postoperatively. (Anesth Analg 2014;119:1194–200)
O
pioids, such as morphine and oxycodone, are used as analgesics in the postoperative period. However, opioid consumption varies widely among individuals.1 Catechol-O-methyltransferase (COMT) metabolizes biogenic amines including catecholamines such as dopamine, epinephrine, and norepinephrine, thereby acting as a key modulator of dopaminergic and adrenergic/ noradrenergic neurotransmission. The dopamine and norepinephrine systems also influence opioid behaviors. Thus, variability in the COMT gene may contribute to the variability of opioid consumption.2 COMT contains a common From the Department of Anesthesiology, Perioperative Medicine and Pain Management, University of Miami School of Medicine, Miami, Florida. Accepted for publication July 14, 2014. Funding: This work was supported by the Department of Anesthesiology, Perioperative Management and Pain Medicine, University of Miami, Miller School of Medicine. The authors declare no conflicts of interest. This report was previously presented, in part, at the International Anesthesia Research Society’s 2013 Annual Meeting. Drs. Yang and Candiotti contributed equally to this manuscript. Reprints will not be available from the authors. Address correspondence to Keith A. Candiotti, MD, Department of Anesthesiology, Perioperative Medicine and Pain Management, University of Miami School of Medicine/Jackson Memorial Hospital, 1611 NW 12th Ave., Central 130(R-370), Miami, FL 33101. Address e-mail to [email protected]. Copyright © 2014 International Anesthesia Research Society DOI: 10.1213/ANE.0000000000000411
1194 www.anesthesia-analgesia.org
functional coding polymorphism rs4680, also known as COMT Val158Met, which substitutes valine with methionine at amino acid position 158, leading to a 3- to 4-fold reduction in the activity of the COMT enzyme.3 Zubieta et al.4 examined this polymorphism’s effect on pain duration in humans. Individuals with a homozygous Met158 genotype reported greater sensory and affective ratings of pain and showed a reduced activation of the endogenous opioid system on experimental pain stimuli and a higher regional density of µ opioid receptors in the brain. Additionally, it has been demonstrated that the rs4680 (Val158Met) polymorphism of the human COMT gene contributed to differences in opioid consumption in patients treated for chronic cancer pain.5,6 A meta-analysis performed by Tammimaki and Mannisto7 showed that the COMT single nucleotide polymorphism (SNP) rs4680 was associated with chronic pain only in patients suffering from fibromyalgia or chronic widespread pain. However, the results of studies examining the interaction of COMT rs4680 polymorphism and its effect on opioid consumption in the presence of acute pain have been negative.8–10 Furthermore, it has been shown that there is no association between the COMT rs4680 polymorphism and opioid consumption in acute pain therapy.10 Two studies performed by Henker et al.9 and Lee et al.11 did not report an association between the COMT rs4680 polymorphism and opioid consumption in acute pain therapy, but did report an association November 2014 • Volume 119 • Number 5
between another COMT polymorphism, rs4818, and opioid consumption in the context of acute pain. Unlike the rs4680 SNP, rs4818 is a synonymous mutation that is thought to affect COMT enzyme translation through alterations in mRNA secondary structure.12 Nevertheless, both studies were performed in relatively small samples groups (79 and 98 patients, respectively), and Henker et al.’s study consisted of only Caucasians and did not reach HardyWeinberg equilibrium, compelling us to further examine these effects. Moreover, no association between the rs4680 polymorphism and acute pain ratings was found in the study performed by Kim et al.; however, a significant association between another COMT SNP, rs740603, and maximal postoperative pain ratings was noted. SNP rs740603 is, however, relatively rare and was omitted from this study in favor of the more common SNPs rs4680 and rs4818. In this study, we evaluated the association between COMT rs4680 and rs4818 polymorphisms and opioid consumption during the acute postoperative period in 152 patients undergoing a nephrectomy, a painful procedure, using an increased sample size more than the previously mentioned studies.
METHODS
After obtaining IRB approval (University of Miami, Miami, FL) and written informed consent, those patients who were scheduled to undergo an elective nephrectomy and were classified as ASA physical status I to III were enrolled. Patients provided medical histories and demographic information including height, weight, age, sex, ethnicity, and smoking history.
Inclusion Criteria
Patients scheduled to undergo a laparoscopic or open nephrectomy using only general anesthesia were eligible. Nephrectomy patients were chosen due to the significant pain experienced after surgery. Patients were excluded if they were currently menstruating females, had undergone a major surgical procedure within the past 30 days, were scheduled to receive an epidural for pain control intraoperatively or postoperatively, were not willing to remain in the study for at least 72 hours, had received any experimental medications within the past 30 days, or had allergies to any of the protocol medications. Patients were premedicated and anesthetized according to the protocol, previously described.13 Briefly, the patients were given midazolam 1 to 2 mg in the preoperative area. Induction of anesthesia was accomplished with propofol 2 to 3 mg/kg, lidocaine 1 mg/ kg and fentanyl 3 µg/kg, and neuromuscular blockade was achieved with cisatracurium 0.2 mg/kg. Endotracheal intubation was used in all patients, and fentanyl and cisatracurium were redosed at the discretion of the anesthesiologist. Anesthesia was maintained using sevoflurane (0%–4.0%) and O2 (50%–70%) titrated to a bispectral index (Covidien, Mansfield, MA) between 45 and 60. Thirty minutes before tracheal extubation, either ondansetron 4 mg or granisetron 0.1 mg was administered. Neuromuscular blockade was reversed with neostigmine 0.05 mg/kg and glycopyrrolate 0.01 mg/kg. All standard vital signs were monitored, and bupivacaine 0.5% was used locally at all incision sites.13 Postoperatively, IV morphine was initially administered on request in the postanesthesia care unit and then by
November 2014 • Volume 119 • Number 5
patient-controlled anesthesia in all patients for the first 72 hours of the postoperative period. A visual analog scale for pain was assessed at 6, 12, 24, 30, 36, 48, and 72 hours postoperatively. For the few patients given oral Percocet 5/325 (oxycodone hydrochloride, 5 mg, and acetaminophen, 325 mg, in each tablet; Endo Pharmaceuticals, Chadds Ford, PA) in addition to morphine, an equivalency factor of oral oxycodone/IV morphine of 1:0.333 was used.10 Laboratory staff members responsible for genotyping were blinded to the clinical results.
DNA Extraction
Blood samples were collected in tubes containing EDTA 1 hour before surgery and were stored at −80°C. DNA was extracted upon sample arrival from whole-blood samples, with QIAamp DNA Blood Midi Kit (QIAGEN, Valencia, CA). DNA samples were then stored at −20°C.
Genotyping
Two SNPs, rs4680 and rs4818, were chosen to evaluate COMT based on previously identified haplotypes.5,6,9,10 We genotyped all COMT SNPs using the TaqMan assay (Applied Biosystems, Foster City, CA). We conducted amplification and genotype assignments using the ABI7000 and SDS 2.0 software (Applied Biosystems). For all genotyping conducted during this study, double-masked genotyping assignments were made for each variant, which were compared, and each discrepancy was addressed using raw data or by regenotyping.
Statistical Analysis
The study population was a convenience sample of patients treated at our facility collected between July 2005 and January 2013. All patients who met our inclusion criteria were offered the opportunity to enroll in this study. One hundred seventy-two patients were enrolled, but 20 patients electively withdrew for unspecified reasons. Consequently, we were left with 152 study subjects for data analysis. Of the 152 study subjects, body mass index (BMI) data were missing for 2, pain score was missing from 1, and the procedure information was missing from 1. Power analysis was done using QUANTO (University of Southern California, Los Angeles, CA). With the sample size of 152, we had >80% power to detect an effect size of 10 (for 24-hour accumulative opium consumption), at an α level of 0.05. The 3 genotypes of COMT rs4680 and rs4818 were treated as independent variables. We calculated summary statistics for all variables in genotypes rendered by the SNP of COMT. General linear model analysis was used to assess whether there were significant differences between genotypes for numerical variables. Linear models allow a flexible analysis that uses linear regression techniques to analyze a host of different analysis of variance and regression problems for continuous data. This technique can be generalized to accommodate other distributions such as binomial, Poisson, and negative binomial. Clinical and demographic continuous data were summarized as mean ± SE. Generalized linear models were used to analyze the association between the genotypes and binary variables. Binary clinical data were summarized as percent and SE. Some categorical data were analyzed with a Pearson χ2 test for equality
www.anesthesia-analgesia.org 1195
COMT Polymorphisms Predict Opioid Consumption in Acute Pain
Table 1. Demographic and Clinical Characteristics of COMT rs4680 Genotype Groups Variable Age (y)a Maleb Race/ethnicityb Hispanic Caucasian African American BMI (kg/m2)a Smokerb Open procedureb Fentanyl (µg/kg/h) Surgical duration (h)
COMT GG 52.6 ± 3.4 21 (67.7%)
COMT AG 54.3 ± 2.5 38 (63.3%)
COMT AA 50.7 ± 2.6 35 (57.4%)
8 (25.8%) 20 (64.5%) 3 (9.7%) 29.7 ± 1.1 16 (51.6%) 24 (77.4%) 1.39 ± 0.11 3.90 ± 1.22
23 (38.3%) 31 (51.7%) 6 (10.0%) 27.5 ± 0.8 21 (35.6%) 44 (73.3%) 1.41 ± 0.15 3.97 ± 1.28
31 (50.8%) 21 (34.4%) 9 (14.8%) 28.2 ± 0.8 21 (34.4%) 41 (68.3%) 1.50 ± 1.20 3.87 ± 1.15
P 0.577 0.597 0.080
Statistical methods General linear χ2 χ2
0.256 0.237 0.636 0.52 0.4
General linear χ2 χ2 General linear General linear
Continuous variables expressed as amean ± SE and bfrequency (%). Discrete variables are given as counts. BMI = body mass index; open = open nephrectomy; LSC = laparoscopically assisted nephrectomy.
Table 2. Twenty-Four-Hour Cumulative Opioid Consumption, Pain Score, and Emesis Medication Usage Among COMT rs4680 Different Genotypes Cumulative opioid consumption (mg) Pain score Emesis medicine usage
Met/Met n = 61 31.9 ± 27.5 3.13 ± 2.77 13.6%
Val/Met n = 60 37.5 ± 27.3 3.56 ± 2.83 15.3%
Val/Val n = 31 49.4 ± 38.9 4.16 ± 2.82 16.1%
P value Met/Met versus Val/Val 0.020* 0.249 0.939
Adjusted mean ± SD. Opioid model includes a significant covariate for smoking status (P = 0.030); there were no significant interactions between genotype and any covariate. Pain score model includes significant covariates for ethnicity (P = 0.016) and age (P = 0.030); there were no significant interactions between genotype and any covariate. Emesis model includes a significant covariate for gender (P = 0.0070); there were no significant interactions between genotype and any covariate. *P < 0.05
of proportions to assess the significance of differences among genotypes. This type of data was summarized as absolute frequency and percent. To assess potential confounding of genotype effects and clinical characteristics, analyses with and without covariates (i.e., age, sex, smoking, ethnic origin, BMI, and surgery type), and interactions between genotype and covariates were performed. Interactions not significant at the 0.050 level were removed first followed by any covariate that was not significant at the 0.050 level. All P values and confidence intervals were Bonferroni corrected for the 3 comparisons among genotypes. SAS 9.3 (SAS Institute Inc., Cary, NC) was used for all analyses.
RESULTS
One hundred fifty-two patients undergoing a nephrectomy, who met inclusion criteria, were included in the study. COMT rs4680 and rs4818 genotype frequencies are listed as follows. For rs4680, all the subjects were distributed into 3 genotype groups: Met/Met homozygotes (n = 61), Val/Met (n = 60), and Val/Val (n = 31). This polymorphism was in HardyWeinberg equilibrium; no distortion of genotype distribution compared with the expected distribution from the allele frequencies was noted. For rs4818, all the subjects were distributed into 3 genotype groups: CC homozygotes (n = 63), CG (n = 68), and GG (n = 21). This polymorphism was also found to be in HardyWeinberg equilibrium. The clinical background characteristics of the 3 genotype groups for cumulative 24-hour opioid consumption for rs4680 were compared (Table 1), with no statistically significant differences noted among them. Twenty-four-hour opioid consumption among the 3 rs4680 COMT genotype groups reached statistical significance (P = 0.020) with the Val/Val group demonstrating
1196 www.anesthesia-analgesia.org
a higher 24-hour opioid consumption when compared with that of the Met/Met group (P = 0.009) (Table 2). After adjusting for age, sex, ethnicity, BMI, smoking, and surgical approach using the general linear model, the association between 24-hour opioid consumption and the 3 COMT rs4680 genotypes still reached statistical significance (P = 0.032). There were no statistically significant differences among the 3 genotype groups for 24-hour postoperative pain scores or requirements for medication to treat nausea or vomiting (Table 2). The emesis model includes a significant covariate for gender status (P = 0.007); there were no significant interactions between genotype and any covariate. Cumulative 24-hour opioid consumption was compared based on ethnicity, age, sex, BMI, smoking history, and surgical type (Table 3). Smokers had a higher 24-hour morphine consumption compared to nonsmokers (P = 0.014). The ethnicity, age, sex, BMI, and surgical type did not show a significant effect on morphine consumption (all P ≥ 0.067). Differences in cumulative 24-hour opioid consumption among the 3 COMT rs4680 genotypes with respect to several clinically relevant variables were compared post hoc and are summarized. Twenty-four-hour opioid consumption among the 3 rs4818 COMT genotype groups did not reach statistical significance (P = 0.223) (Table 4). There were no statistically significant differences among the 3 genotype groups for 24-hour postoperative pain scores (P = 0.250). There were statistically significant differences among the 3 genotype groups (P = 0.030) for 24-hour postoperative requirements for medication to treat nausea or vomiting with the CC group consuming more antiemetic drugs over a 24-hour period when compared with that of the GG group (P = 0.035) (Table 4). The emesis model includes a significant
anesthesia & analgesia
Table 3. Twenty-Four-Hour Cumulative Opioid Consumption Based on Different Clinical Variables Variable Race/ethnicity
Age (y)
Gender BMI
Smoker Surgery Type
Value Hispanic (n = 62) Caucasian (n = 72) African American (n = 18) ≤40 (n = 30) 41–50 (n = 26) 51–60 (n = 45) 61–70 (n = 30) ≥71 (n = 21) Male (n = 94) Female (n = 58) ≤25 (n = 49) 26–29 (n = 57) 30–35 (n = 33) ≥36 (n = 9) Yes (n = 58) No (n = 93) Open (n = 109) LSC (n = 42)
24-h opioid consumption (mg) 34.5 ± 3.9 40.8 ± 3.6 28.7 ± 7.2 37.7 ± 5.6 42.2 ± 6.0 42.0 ± 4.5 28.5 ± 5.6 29.6 ± 6.7 38.2 ± 3.2 34.5 ± 4.0 35.6 ± 4.4 40.0 ± 4.1 31.0 ± 5.3 54.7 ± 10.2 44.6 ± 4.0 32.0 ± 3.1 37.4 ± 3.0 35.2 ± 4.8
P 0.244
0.242
0.478 0.183
0.014* 0.687
All variables analyzed with a general linear model. BMI = body mass index; LSC = laparoscopically assisted nephrectomy. *P < 0.05
covariate for gender status (P = 0.007); there were no significant interactions between genotype and any covariate. During the 6- to 48-hour postoperative period, there was significantly higher opioid consumption in the 6-, 12-, 24-, and 48-hour time intervals in patients in the COMT rs4680 Val/Val group (high enzymatic activity) when compared with patients in the Met/Met group (low enzymatic activity) (0–6 h: P = 0.005; 0–12 h: P = 0.015; 0–24 h: P = 0.015; and 0–48 h: P = 0.023) (Fig.1). The analysis model included covariates for smoking status (P = 0.043), and there were no significant interactions between genotype and any covariate. In the heterozygous Val/Met group, the opioid consumption was higher than the Met/Met group and lower than the Val/ Val group in the 48-hour postoperative time interval, with the difference between Val/Met and Val/Val reaching statistical significance only in the first 6 hours postoperatively. During the 48-hour postoperative period, patients in the COMT rs4818 GG group consumed significantly more opioids in the first 6 hours after the operation when compared to patients in the heterozygous CG group (P = 0.020). Other comparisons between rs4818 genotypes did not show statistically significant differences in opioid consumption (Fig. 2). The analysis model included covariates for smoking status (P = 0.043); there were no significant interactions between genotype and any covariate.
DISCUSSION
Our current study has shown a strong contribution of the COMT rs4680 polymorphism to interindividual variations in opioid consumption in the acute postoperative period after nephrectomy. Subjects with the Val/Val genotype consumed 36% more opioids compared to subjects with Met/ Met genotype at 24 hours after the operation. In one previous study, Kolesnikov et al.10 reported that carriers of COMT rs4680 Met/Met consumed similar morphine doses compared with wild-type homozygous carriers
November 2014 • Volume 119 • Number 5
of Val/Val who underwent abdominal surgery for either prostatectomy or hysterectomy. Henker et al.9 also demonstrated that the genotypes of the COMT rs4680 polymorphism were not associated with morphine consumption in the postoperative period, but they found that the rs4818 genotype polymorphism was associated with opioid consumption with the GG genotype group consuming more opioids compared to those patients in the heterozygous CG and homozygous CC groups. However, in an interesting contrast, Rakvåg et al.5 found that the COMT rs4680 polymorphism had a statistically significant effect on morphine consumption in the treatment of chronic pain. Patients carrying the Val/Val genotype had significantly higher morphine consumption compared with those possessing the Met/Met genotype. Our results support the findings of Rakvåg et al. We found that there was significantly higher opioid consumption in the first 6 hours postoperatively in patients who were homozygous GG when compared with patients who were heterozygous CG for the rs4818 polymorphism (P = 0.015). However, while there was no statistically significant difference between the CC and GG groups, the homozygous GG group consumed a numerically higher total amount of opioids compared to the homozygous CC group (P = 0.120). Consequently, the decrease in opioid consumption in the CC and CG genotypes may possibly be attributed to C carriage. The difference between our result and Kolesnikov et al.’s10 and Henker et al.’s9 results may be attributable to different surgery types and sample size. Previously, genetically modified mice have been used to study the impact of altered COMT activity on pain.14,15 Mouse lines modified to overexpress the Val158 (Val/Val) variant, specifically in the prefrontal cortex, were studied and shown to have diminished pain sensitivity in both hotplate and tail flick testing.15 Concordant with these findings, COMT knockout mice were shown to have increased pain sensitivity.16 Zubieta et al.4 demonstrated that variations in Val158Met genotypes, and corresponding variation in levels of COMT activity, impact downstream functions regulated by monoamines in humans. They examined µ-opioid receptor density in different brain regions using positron emission tomography and found that individuals expressing the low-activity enzyme, corresponding to the Met/ Met genotype, possessed increased receptor density, while those expressing the high-activity enzyme, corresponding to the Val/Val genotype, possessed lower receptor density. Clinically, this suggests that patients expressing the Met/Met genotype and having higher opioid receptors in critical regions might require less opioid to produce adequate pain control, while those possessing the genotype Val/Val may require relatively more opioid. This idea is supported by a recent study which found that the rs4680 Met/Met genotype and the rs4818 CC genotype were both associated with lower preoperative pain scores and pain intensity 1 year after lumbar discectomy.17 Our data, demonstrating a lower requirement for opioids among patients carrying the Met/Met genotype of the COMT enzyme, are in agreement with these results. In general, our data support the concept that the Met/Met genotype of COMT rs4680 is associated with a low enzymatic activity
www.anesthesia-analgesia.org 1197
COMT Polymorphisms Predict Opioid Consumption in Acute Pain
Table 4. Twenty-Four-Hour Cumulative Opioid Consumption, Pain Score, and Emesis Medication Usage Among COMT rs4818 Different Genotypes Cumulative opioid consumption (mg) Pain score Emesis medicine usage
CC n = 61 36.3 ± 25.5 3.83 ± 2.80 15.8% ± 4.6%
CG n = 60 34.2 ± 24.3 3.28 ± 2.95 13.2% ± 4.9%
GG n = 31 46.9 ± 37.5 3.33 ± 2.65 4.5% ± 3.5%
P value CC versus GG 0.20 0.254 0.035*
Adjusted mean ± SD. Opioid model includes a significant covariate for smoking status (P = 0.030); there were no significant interactions between genotype and any covariate. Pain score model includes significant covariates for ethnicity (P = 0.016) and age (P = 0.030); there were no significant interactions between genotype and any covariate. Emesis model includes a significant covariate for gender status (P = 0.0070); there were no significant interactions between genotype and any covariate. *P < 0.05
Figure 1. Mean opioid consumption in 6, 12, 24 and 48 h for 3 genotype classes. The analysis model included covariates for smoking status (P = 0.043); there were no significant interactions between genotype and any covariate. Statistically significant differences between Val/Val and Val/Met at 0–6 h (P = 0.025) and between Met/Met and Val/Val occurred at 0–6 h (* = 0.0050), 0–12 h (** = 0.015), 0–24 h (# = 0.016), and 0–48 h (## = 0.023). Other comparisons between Met/Met and Val/Met and Val/Met and Val/Val genotypes were not significant. Bar represents 1 SE.
Figure 2. Mean opioid consumption in 6, 12, 24 and 48 h for 3 genotype classes. The analysis model included covariates for smoking status (P = 0.043); there were no significant interactions between genotype and any covariate. Statistically significant differences between CG and GG occurred only at 0–6 h (* = 0.020), but not at 0–12 h (** = 0.050), 0–24 h = 0.060), or 0–48 h (## = 0.090). Other comparisons between CC versus GG and CC versus CG genotypes were not significant. Bar represents 1 SE.
of dopamine, leading to an increase of µ -opioid receptor density in specific brain regions, resulting in a lower concentration of opioid being required to achieve the same degree of activation. To determine whether the genotype groups were different with regard to factors that may affect opioid consumption, we compared the groups with regard to several characteristics including smoking, operation duration, and surgery types. The different genotype groups showed similar results for these characteristics as for other confounders including age, gender, and ethnicity (Table 1). When opioid consumption during the 24-hour postoperative period was compared among these confounders (Table 3), our data showed that smoking increased opioid consumption during the 24-hour postoperative period. This result is consistent with our previous study,13 even with the addition of more patients. When smoking is added as a covariate, the association between morphine consumption during the 24-hour postoperative period and COMT genotypes remained statistically significant. When analyzing pain scores from the 24-hour postoperative period, we observed that the Val/Val genotype demonstrated the highest numerical pain scores, the Val/Met group an intermediate score, and the Met/Met genotype the lowest. However, these differences did not reach statistical significance (P = 0.249). The same trend was observed in the 6-, 12- and 48-hour postoperative pain scores (data
not shown). Overall, the interindividual variations of pain scores may be more pronounced than that of opioid consumption, so an increase in sample size may be required to see a statistically significant difference in future studies. Additionally, pain scores may be difficult to interpret in general due to the fact that patients had unrestricted access to pain medications. Kolesnikov et al.10 reported that COMT rs4680 was associated with the central side effects of opioids such as nausea and sedation in patients who underwent abdominal surgery for either prostatectomy or hysterectomy. In this study, we also investigated a possible correlation between morphine-related side effects (symptomatic nausea/vomiting) and the COMT rs4680 SNP in these patients. Our data indicated that the Met/Met genotype demonstrated the numerically lowest usage of emesis medication, the Val/ Met genotype demonstrated intermediate usage levels, and the Val/Val genotypes showed the highest usage; however, there were no statistically significant differences among the 3 genotype groups for emesis medication use at 24-hours after surgery. Our results are consistent with the study by Ross et al.18 that did not find any associations between the COMT rs4680 polymorphism and emesis medication usage. The discrepancy between our findings and those in Kolesnikov’s study may be attributable to their smaller sample size. Interestingly, we found that the rs4818 polymorphism is associated with variations in emesis medication
1198 www.anesthesia-analgesia.org
anesthesia & analgesia
use after surgery, with the CC genotype group using more emesis medication compared with the CG and GG genotype groups. This association did not correlate with opioid consumption, suggesting that genetic factors may play an independent role in postoperative nausea and vomiting. Finally, our study has limitations. First, this was strictly a gene association trial and, as such, opioid receptors and opioid concentration were not measured in the brain or blood, and the mechanism of our findings can only be assumed. Second, the study only enrolled patients having a nephrectomy, which is relatively painful. We believe that the notable level of pain caused by this procedure makes it an ideal acute pain model; however, it is possible that our findings may not apply to other less painful surgeries. Finally, we only focused on the most common COMT SNPs, rs4680 and rs4818, because these 2 SNPs had a previously reported association with opioid consumption in chronic pain and acute pain patients. In summary, our study demonstrates that the COMT rs4680 SNP is associated with differences in postoperative opioid consumption in patients who underwent a nephrectomy. The COMT rs4680 polymorphism may predict opioid consumption in postoperative patients, and the COMT rs4818 polymorphism may predict emesis medication use postoperatively. It should be noted that while genetic analysis appears useful in predicting the pain phenotype, practical use is likely years away. E DISCLOSURES
Name: Keith A. Candiotti, MD. Contribution: This author helped design the study and write the manuscript. Attestation: Keith A. Candiotti has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files. Name: Zhe Yang, MD, PhD. Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript. Attestation: Zhe Yang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: David Buric, BS. Contribution: This author helped write the manuscript. Attestation: David Buric has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Kris Arheart, PhD. Contribution: This author helped analyze the data and write the manuscript. Attestation: Kris Arheart has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Yanping Zhang, PhD. Contribution: This author helped conduct the study and write the manuscript. Attestation: Yanping Zhang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Yiliam Rodriguez, MD. Contribution: This author helped analyze the data and write the manuscript.
November 2014 • Volume 119 • Number 5
Attestation: Yiliam Rodriguez has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Melvin C. Gitlin, MD. Contribution: This author helped design the study and write the manuscript. Attestation: Melvin C. Gitlin has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Enisa Carvalho, MD. Contribution: This author helped with patient recruitment and processing of data. Attestation: Enisa Carvalho has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Isabel Jaraba, MD. Contribution: This author helped with patient recruitment and processing of data. Attestation: Isabel Jaraba has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. Name: Liyong Wang, PhD. Contribution: This author helped design the study and write the manuscript. Attestation: Liyong Wang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript. This manuscript was handled by: Spencer S. Liu, MD. ACKNOWLEDGMENTS
The authors thank Dr. Odalys Rodriguez Bravo (research assistant, The Department of Anesthesiology, Perioperative Management and Pain Medicine, University of Miami, Miller School of Medicine) for expert technical assistance. REFERENCES 1. Coulbault L, Beaussier M, Verstuyft C, Weickmans H, Dubert L, Trégouet D, Descot C, Parc Y, Lienhart A, Jaillon P, Becquemont L. Environmental and genetic factors associated with morphine response in the postoperative period. Clin Pharmacol Ther 2006;79:316–24 2. Lachman HM, Papolos DF, Saito T, Yu YM, Szumlanski CL, Weinshilboum RM. Human catechol-O-methyltransferase pharmacogenetics: description of a functional polymorphism and its potential application to neuropsychiatric disorders. Pharmacogenetics 1996;6:243–50 3. Lotta T, Vidgren J, Tilgmann C, Ulmanen I, Melén K, Julkunen I, Taskinen J. Kinetics of human soluble and membranebound catechol O-methyltransferase: a revised mechanism and description of the thermolabile variant of the enzyme. Biochemistry 1995;34:4202–10 4. Zubieta JK, Heitzeg MM, Smith YR, Bueller JA, Xu K, Xu Y, Koeppe RA, Stohler CS, Goldman D. COMT val158met genotype affects mu-opioid neurotransmitter responses to a pain stressor. Science 2003;299:1240–3 5. Rakvåg TT, Klepstad P, Baar C, Kvam TM, Dale O, Kaasa S, Krokan HE, Skorpen F. The Val158Met polymorphism of the human catechol-O-methyltransferase (COMT) gene may influence morphine requirements in cancer pain patients. Pain 2005;116:73–8 6. Rakvåg TT, Ross JR, Sato H, Skorpen F, Kaasa S, Klepstad P. Genetic variation in the catechol-O-methyltransferase (COMT) gene and morphine requirements in cancer patients with pain. Mol Pain 2008;4:64 7. Tammimäki A, Männistö PT. Catechol-O-methyltransferase gene polymorphism and chronic human pain: a systematic review and meta-analysis. Pharmacogenet Genomics 2012;22:673–91
www.anesthesia-analgesia.org 1199
COMT Polymorphisms Predict Opioid Consumption in Acute Pain
8. Kim H, Lee H, Rowan J, Brahim J, Dionne RA. Genetic polymorphisms in monoamine neurotransmitter systems show only weak association with acute post-surgical pain in humans. Mol Pain 2006;2:24 9. Henker RA, Lewis A, Dai F, Lariviere WR, Meng L, Gruen GS, Sereika SM, Pape H, Tarkin IS, Gowda I, Conley YP. The associations between OPRM 1 and COMT genotypes and postoperative pain, opioid use, and opioid-induced sedation. Biol Res Nurs 2013;15:309–17 10. Kolesnikov Y, Gabovits B, Levin A, Voiko E, Veske A. Combined catechol-O-methyltransferase and mu-opioid receptor gene polymorphisms affect morphine postoperative analgesia and central side effects. Anesth Analg 2011;112:448–53 11. Lee PJ, Delaney P, Keogh J, Sleeman D, Shorten GD. Catecholamine-O-methyltransferase polymorphisms are associated with postoperative pain intensity. Clin J Pain 2011;27:93–101 12. Diatchenko L, Slade GD, Nackley AG, Bhalang K, Sigurdsson A, Belfer I, Goldman D, Xu K, Shabalina SA, Shagin D, Max MB, Makarov SS, Maixner W. Genetic basis for individual variations in pain perception and the development of a chronic pain condition. Hum Mol Genet 2005;14:135–43 13. Candiotti KA, Yang Z, Morris R, Yang J, Crescimone NA, Sanchez GC, Bird V, Leveillee R, Rodriguez Y, Liu H, Zhang YD, Bethea JR, Gitlin MC. Polymorphism in the interleukin-1 receptor antagonist gene is associated with serum interleukin-1
1200 www.anesthesia-analgesia.org
receptor antagonist concentrations and postoperative opioid consumption. Anesthesiology 2011;114:1162–8 14. Kambur O, Männistö PT, Viljakka K, Reenilä I, Lemberg K, Kontinen VK, Karayiorgou M, Gogos JA, Kalso E. Stressinduced analgesia and morphine responses are changed in catechol-O-methyltransferase-deficient male mice. Basic Clin Pharmacol Toxicol 2008;103:367–73 15. Papaleo F, Crawley JN, Song J, Lipska BK, Pickel J, Weinberger DR, Chen J. Genetic dissection of the role of catechol-O-methyltransferase in cognition and stress reactivity in mice. J Neurosci 2008;28:8709–23 16. Gogos JA, Morgan M, Luine V, Santha M, Ogawa S, Pfaff D, Karayiorgou M. Catechol-O-methyltransferase-deficient mice exhibit sexually dimorphic changes in catecholamine levels and behavior. Proc Natl Acad Sci U S A 1998;95:9991–6 17. Rut M, Machoy-Mokrzyńska A, Ręcławowicz D, Słoniewski P, Kurzawski M, Droździk M, Safranow K, Morawska M, Białecka M. Influence of variation in the catechol-O-methyltransferase gene on the clinical outcome after lumbar spine surgery for one-level symptomatic disc disease: a report on 176 cases. Acta Neurochir (Wien) 2014;156:245–52 18. Ross JR, Riley J, Taegetmeyer AB, Sato H, Gretton S, du Bois RM, Welsh KI. Genetic variation and response to morphine in cancer patients: catechol-O-methyltransferase and multidrug resistance-1 gene polymorphisms are associated with central side effects. Cancer 2008;112:1390–403
anesthesia & analgesia
