- Question: What are the effects of the adenosine triphosphate-binding cassette, subfamily B, member 1 (ABCB1) polymorphism c.3435C>T (rs10454642) on heat pain (HP) perception in a cohort of opioid-free adults with chronic pain?
- Findings: Under a recessive model of allele effects, a significant association between HP threshold and genotype was observed where HP threshold was significantly greater among individuals with the TT genotype than individuals with the CC and CT genotypes (P = .005).
- Meaning: These results posit that the efflux of endogenous opioid peptides is reduced in individuals with the TT genotype due to lower expression of P-glycoprotein (P-gp), which, in turn, results in higher HP threshold.
Adenosine triphosphate-binding cassette transporters are membrane proteins that utilize energy to transport solutes across cellular membranes.1 These transporters comprise a superfamily of genes that are involved in a multitude of cellular processes. The adenosine triphosphate-binding cassette, subfamily B, member 1 gene (ABCB1), also known as the multidrug resistance 1 (MDR1) gene, is located on the chromosomal region 7q21.12.1 This gene encodes the amino acid P-glycoprotein (P-gp).2 P-gp function influences intracellular drug elimination, drug absorption, and the efflux of synthetic and endogenous opioid peptides.2,3 A widely investigated ABCB1 polymorphism is c.3435C>T (rs1045642), which is a synonymous polymorphism (cysteine to thymine) at nucleotide 3435.4 Individuals with the thymine-thymine (TT) genotype have a 4-fold lower expression of P-gp compared to individuals with the cytosine-cytosine (CC) genotype.1
A widely recognized function of P-gp is the efflux of drugs5,6 including opioids.7 However, a lesser known function is the efflux of endogenous opioid peptides across the blood-brain barrier in the brain to blood direction.7,8 Alterations in the efflux of endogenous opioids associated with the c.3435C>T polymorphism could influence pain perception in humans. This is clinically relevant because opioid-free adults with chronic pain are reported to have significantly lower standardized values of heat pain (HP) threshold and tolerance compared to established normative values.9 Thus, the c.3435C>T polymorphism could be an unrecognized contributor to lower pain thresholds and tolerances in this patient population. The primary objective of this study was to determine the effects of the ABCB1 polymorphism c.3435C>T (rs10454642) on HP perception in a group of opioid-free adults with chronic pain. Portions of this data set have been previously published.10–13
The Mayo Foundation Institutional Review Board approved this study; all patients provided written informed consent before study participation. As reported previously,11,13 adults with chronic pain consecutively admitted to the Mayo Comprehensive Pain Rehabilitation Center from March 2009 to March 2010 were eligible for study participation. During this period, 524 adults with chronic pain were admitted. Inclusion criteria included: (1) admission to the pain treatment program; (2) noncancer pain >3-month duration; and (3) age ≥18 years. Exclusion criteria included major medical (eg, severe cardiovascular disease), surgical (eg, major surgery within 6 months of admission), or psychiatric conditions that would limit full participation in the 3-week outpatient treatment program. Inclusion and exclusion criteria were met by 300 patients.
Quantitative sensory testing (QST) was not performed in 22 subjects due to lack of study personnel, the c.3435C>T genotype was not determined in 1 subject, and 143 subjects were using prescription opioids for chronic pain. Thus, the study cohort was comprised of 134 opioid-free adults with chronic pain.
The clinical setting has been described.14 The outpatient program is for 3 weeks, and patients attend daily for 8 hours. Admission criteria include noncancer pain of 6 months and willingness to participate in all aspects of the program. A cognitive behavioral model is the basis of treatment. The treatment goals of the group-based program are rehabilitation of physical functioning and improvement in emotional functioning. Specific components of the program included: (1) psychoeducational group sessions to improve management of depression and pain-related anxiety; (2) group-based relaxation training sessions; (3) techniques to manage pain-related stress; (4) daily group-based physical and occupational therapy sessions; and (5) techniques to ameliorate pain behaviors.
Demographic and Clinical Factors
Demographic and clinical factors collected on admission included age, sex, race, ethnicity, pain diagnosis, and duration of pain.
All DNA was extracted from whole blood. Genotyping was performed using ABI TaqMan reagents (Applied Biosystems, Foster City, CA, now ThermoFisher Scientific, Waltham, MA), according to manufacturer’s instructions (10-ng DNA). Primers and probes were specific for ABCB1 (rs10454642). End reactions were read on an ABI Prism 7900ht using the ABI Sequence Detection Software; allelic discrimination and genotype data were generated in an electronic data set. Controls included a normal Centre d'Etude Polymorphisme Humain DNA (Coriell Institute for Medical Research, Camden, NJ) and a no-template negative control.
Self-report measures of depressive symptoms and pain intensity were completed on the day of admission, and QST was conducted 1 day after admission.
The Center for Epidemiologic Studies-Depression (CESD) scale was used to measure depressive symptoms.15 The CESD is scored on a 4-point Likert scale, and scores range from 0 to 60. Higher scores denote greater levels of depression symptoms. The reliability and validity of the CESD have been established in adults with chronic pain.16
The short version of The Pain Anxiety Symptoms Scale (PASS-20) measures pain-related anxiety.17 The PASS-20 is a 20-item self-report questionnaire, and scores range from 0 to 100. Higher scores indicate greater levels of pain-related anxiety.
Pain intensity was assessed using the 11-point verbal rating scale (VRS). A VRS score of 0 indicates no pain, and a score of 10 indicates the worse possible pain. The validity of the VRS has been established.18
QST was conducted using the Computer Aided Sensory Evaluator IV (CASE IV; WR Medical Electronics, Stillwater, MN) device based on the method of levels.19–21 A random series of heat stimuli of various magnitudes with intervening null stimuli is delivered via a ceramic thermode that has a surface area of 10 cm2. The entire testing protocol has been standardized.19–22 During testing, the individual is masked to the magnitude of the heat stimulus. After delivery of each stimulus, the individual is asked to rate the intensity on an 11-point scale, where 0 indicates no pain and 10 indicates the greatest possible pain. The test is completed when either the individual rates pain intensity ≥5 or the maximum stimulus has been delivered.
Twenty-five magnitudes of heat stimuli are delivered by the CASE IV device. The magnitudes of heat stimuli are expressed in units termed just noticeable difference (JND).19,20,23 Temperature and duration of exposure comprise the magnitude of each level of heat stimuli. The lowest temperature (level 1) is 34 °C, and the rate of temperature rise is 4 °C per second. The temperature increases exponentially until 48 °C (level 21). The temperature then remains at 48 °C for 1.5 seconds (level 22), 5 seconds (level 23), and 10 seconds (level 24). The maximum temperature is 49 °C for a duration of 10 seconds (level 25). Thermal injury is avoided by not using temperatures >49 °C.
Pain threshold and tolerance are calculated by the CASE IV software (WR TestWorks, version 2.0, WR Electronics). Pain threshold, termed HP 0.5, is the midpoint between a nonpainful stimulus and smallest stimulus needed to produce a pain sensation. Pain tolerance, termed HP 5, is the stimulus magnitude needed to produce a pain rating of 5. The stimulus response slop, defined as the slope of the line connecting HP 0.5 and HP 5, is termed HP 5-0.5. Raw sensory data are recorded in JND units and automatically adjusted by the software for anthropometric factors (age, sex, height, weight, body surface area, body mass index, and body region of testing) that are known to influence pain sensation in humans.24–26 The adjusted sensory values are standardized using data from a normative population24 and converted into a unit of measure termed normal deviate (ND). An ND value equal to zero represents the 50th percentile (standard deviation [SD], 1). An ND value <0 indicates a continuous tendency toward worsening pain sensitivity (ND = −2.33 corresponds to the first percentile), and an ND >0 indicates a continuous tendency toward pain insensitivity (ND = 2.33 corresponds to the 99th percentile). ND values of HP perception have been established for the feet, calf region of the lower extremities, anterior thigh region of the lower extremities, arms, shoulders, and dorsum of the hands.20,24
All sensory tests were performed in a single designated room, and the testing environment was similar for all subjects. The QST procedure takes 5 minutes to complete,22 and the dorsum of the nondominant hand was the test site.
For each genotype, baseline demographic and clinical factors were summarized. Mean (± SD) values were calculated for continuous variables, and total counts were calculated for categorical variables. For all HP measures, 1-sample Kolmogorov-Smirnov tests were used to assess the distribution of ND values, and the median and 25th to 75th interquartile range (IQR) were calculated for each measure. The distribution of the ABCB1 c.3435C>T genotype was evaluated for departure from Hardy-Weinberg equilibrium using a χ2 test.27 Genotype differences in demographic variables, clinical factors, and HP perception were evaluated with nonparametric tests (Kruskal-Wallis) for continuous variables and χ2 tests for categorical variables. The main effect of genotype on HP perception was further evaluated with 2-sample nonparametric tests (Mann-Whitney U).
A recessive (CC+cytosine-thymine [CT] versus TT) and dominant (CC versus CT+TT) genotype model was used to evaluate the associations between the ABCB1 polymorphism and ND values of HP perception. These genotype models have been used to study the influence of the ABCB1 c.3435C>T polymorphism on a wide range of drugs.5,6,28 Two-sample nonparametric tests (Mann-Whitney U) were used to evaluate the associations between the ND values of HP perception and genotype for the recessive and dominate models. The median difference and 95% confidence intervals (CIs) of HP perception across the recessive and dominate models were estimated using the Hodges-Lehman estimator. A nonparametric test was used to estimate the effect size of the significant association between the recessive genotype model and HP perception (Cliff delta).29
Two genetic models were tested; thus, the significance criterion for all tests was set at P ≤ .025. All analyses were completed using SPSS (IBM, Inc; version 21.0, Chicago, IL) except for Cliff delta (R version 3.6.2; R Foundation for Statistical Computing, Vienna, Austria). This article adheres to the Strengthening the Reporting of Genetic Association Studies (STREGA)—an extension of the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement guidelines for reporting genetic association studies.30
Based on our previous work,9 a sample size of 134 will provide 80% power to detect group differences of: (1) 0.56 ND in HP 5; (2) 0.49 ND in HP 0.5; and (3) 0.55 ND in HP 5-0.5 at a 2-sided 0.05 significance level. These differences in HP perception are consistent with a small to medium effect size.
Demographic and Clinical Factors
Table 1. -
Demographic and Clinical Characteristics
||Total (N = 134)
ABCB1 (rs1045642) genotype
|CC (N = 29)
||CT (N = 60)
||TT (N = 45)
|Age, y (mean ± SD)
||43.6 ± 13.0
||45.9 ± 13.4
||41.0 ± 12.2
||45.5 ± 13.6
|Sex, female, n (%)
|Pain duration, y
||8.7 ± 7.6
||9.4 ± 7.6
||8.9 ± 7.7
||8.1 ± 7.7
| Current smoker
|Primary pain diagnosis
| Fibromyalgia, headache, facial, and generalized pain
| Low back, neck pain
| Abdominal, pelvic, and chest
| Upper and lower extremities
||27.8 ± 13.6
||28.1 ± 14.5
||29.8 ± 13.3
||24.8 ± 13.2
||47.0 ± 20.8
||46.6 ± 17.7
||47.9 ± 21.0
||46.0 ± 22.6
||5.8 ± 2.0
||5.7 ± 2.3
||5.9 ± 1.9
||5.9 ± 2.1
Abbreviations: CC, cytosine-cytosine; CT, cytosine-thymine; SD, standard deviation; TT, thymine-thymine.
aKruskal-Wallis test for continuous variables and χ2 for categorical variables.
bCenter for Epidemiologic Studies-Depression scale.
cPain Anxiety Symptoms Scale-short version.
dVerbal Pain Rating Scale.
Table 1 presents a summary of the demographic and clinical factors of the study cohort. The distribution of ABCB1 c.3435C>T genotypes was 22% (N = 29) for CC, 45% (N = 60) for CT, and 33% (N = 45) for TT. There was no departure from Hardy-Weinberg equilibrium (χ2 = 1.558, P > .1). The frequency of the minor T allele was 0.34, which was similar to the 0.36 frequency reported in the HapMap project.31 The majority of patients were Caucasian women with a mean age of 43.6 years. The mean pain duration was 8.7 years. No significant group differences based on genotype were observed for depression, pain-related anxiety, or pain intensity. Supplemental Digital Content 1, Table 1, https://links.lww.com/AA/D576, and Supplemental Digital Content 2, Table 2, https://links.lww.com/AA/D577, present the demographic and clinical factors based on the recessive and dominant models, respectively.
Associations Between HP Perception and ABCB1 c.3435C>T Genotype
Table 2 presents median ND values and IQR of HP 0.5, HP 5, and HP 5-0.5 for each genotype. The ND values of HP 0.5 and HP 5 were not normally distributed (Kolmogorov-Smirnov Z, P < .05), but the ND values of HP 5-0.5 were normally distributed (Kolmogorov-Smirnov Z, P > .05) (Figure 1). A main effect of genotype was identified for HP 0.5 (P = .017). Focused contrast analysis identified that the ND values of HP 0.5 for the TT group were significantly greater than the CC (median difference, −1.00; 95% CI, −1.79 to −0.28; P = .007) and CT (median difference, −0.68; 95% CI, −1.45 to 0.00; P = .024) genotype groups. However, no significant group difference was identified between the CC and CT (median difference, 0.00; 95% CI, −0.74 to 0.45; P = .647) genotype groups.
Table 2. -
Median Values of Heat Pain Perception Based on ABCB1
|CC (n = 29)
||CT (n = 60)
||TT (n = 45)
|HP 0.5, median (IQR)
||−1.13 (−2.19 to −0.26)
||−1.20 (−2.05 to 0.46)
||0.05 (−1.76 to 1.48)
|HP 5, median (IQR)
||−0.64 (−2.05 to 0.24)
||−1.04 (−2.33 to 0.88)
||−0.08 (−1.70 to 1.68)
|HP 5-0.5, median (IQR)
||0.52 (−0.76 to 1.85)
||0.67 (−0.12 to 1.34)
||0.10 (−0.92 to 1.22)
Abbreviations: CC, cytosine-cytosine; CT, cytosine-thymine; HP, heat pain; IQR, 25th to 75th interquartile range; ND, normal deviate; TT, thymine-thymine.
A significant association was observed between the recessive model and HP 0.5 (Figure 2). The ND values of HP 0.5 were significantly greater in the TT group than the CC+CT group (median difference, −0.77; 95% CI, −1.49 to −0.23; P = .005). The effect size estimate was small (Cliff delta, 0.30; 95% CI, −0.08 to 0.48). In the dominant model, no significant difference in HP 0.5 was observed between the CC and CT + TT groups (median difference, −0.45; 95% CI, −1.15 to 0.00; P = .108).
The main finding of this study that involved opioid-free adults with chronic pain was that a standardized measure of HP threshold was significantly associated with the ABCB1 c.3435C>T genotype in a recessive model of allele effects. More specifically, individuals with the TT genotype had significantly greater values of HP threshold than the group comprised of the CC and CT genotypes, and the effect size estimate was small. No significant association was observed between HP threshold and the dominant model.
The influence of the ABCB1 c.3435C>T polymorphism on pain perception in opioid-free adults with chronic pain has not been reported. However, the associations between a diplotype of 3 ABCB1 polymorphisms (c.3435C>T, c.1236C>T, and c.2677G>T/A) and nonstandardized values of cold-pressor pain threshold and tolerance were reported for a cohort of 152 opioid naïve Malay men.32 In a subgroup of 18 individuals with a diplotype that contained the c.3435C>T CC genotype, cold pain threshold and tolerance were significantly greater than individuals (n = 134) without this diplotype.32 This counters the observations from our study and possible reasons include: (1) disparate effects of the other diplotype alleles on the c.3435C>T polymorphism; (2) use of nonstandardized scores of pain threshold and tolerance25,33; (3) ethnic, racial, and sex differences in pain perception34,35; and (4) differences in pain perception in healthy controls versus adults with chronic pain.36
The mechanisms responsible for the findings of this study warrant further consideration. P-gp is localized to the luminal side of endothelial cells comprising the blood-brain barrier.3 An important function of P-gp is the efflux of endogenous opioid peptides7,8 across the blood-brain barrier in the brain to blood direction. In preclinical studies, inhibition of P-gp resulted in a 75% reduction in serum ß-endorphin levels7 and enhanced antinociception following administration of the synthetic opioid peptide [D-penicillamine2,5] enkephalin.37,38 However, P-gp is not required for the efflux of other opioid peptides including endomorphin 1, endomorphin 2, and met-enkephalin.39,40 These preclinical studies support the observations from our study where individuals with the TT genotype had significantly greater values of HP threshold than the CC and CT genotype groups. This is relevant because, as previously noted, individuals homozygous for the T allele have a 4-fold lower expression of P-gp compared to individuals homozygous for the C allele.1 Thus, it can be posited that the efflux of endogenous opioid peptides is reduced in individuals with the TT genotype due to lower expression of P-gp, which, in turn, results in higher values of HP threshold.
The mean values of HP threshold for the CC and CT genotypes are comparable to previously reported values. Units of ND are standardized values derived from a normative population where a value of 0 corresponds to the 50th percentile and the SD is 1.24 In a study that involved a subgroup of 102 opioid-free adults with chronic pain, mean ND values of HP threshold were −0.64.41 In the current study, the mean ND value of HP threshold for the CC genotype was −1.00, and the mean value for the CT genotype was −0.72. Alternatively, the mean ND value of HP threshold for the TT genotype was −0.01 indicating that the HP threshold of individuals with the TT genotype approximated normative values. Collectively, these data extend the findings of previous work 2-fold and suggest that: (1) the C allele of c.3435C>T may be a significant contributor of lower HP threshold in opioid-free adults with chronic pain and (2) the TT genotype may mitigate the risk of developing lower HP threshold in adults with chronic pain.
The observations of this study have implications for future research. The findings need to replicate in more diverse cohorts of opioid-free adults with chronic pain. This line of research should also: (1) incorporate a control group to ascertain if the c.3435C>T polymorphism is associated with altered HP threshold in adults without chronic pain and (2) investigate the effects of chronic pain duration on the associations between the c.3435C>T polymorphism and HP perception. Prospective studies are also needed to investigate how lower values of HP threshold in individuals with the CC and CT genotypes influence rehabilitative, pharmacological, interventional, and surgical outcomes. Prospective studies should also incorporate adults using opioids for chronic pain, which could help delineate the effects that the c.3435C>T polymorphism has on opioid-related analgesia.
This study has limitations. The effects of a single ABCB1 polymorphism on HP perception were investigated, which does not fully represent the breadth of ABCB1 activity.4 Hence, the study findings may have been affected by other polymorphisms and other unrecognized confounding factors including individual psychosocial, racial, and ethnic influences. In the absence of a control group, it is not possible to ascertain how the c.3435C>T polymorphism influences the development or maintenance of chronic pain. A single sensory modality was tested, and the findings may not be applicable to other sensory modalities or methods for determining HP threshold. The study findings could have been influenced by prior interventional procedures and use of nonopioid medications; however, a previous study suggests that CASE IV results are not altered by simple analgesics,42 but the potential effects of interventional procedures have not been reported. Finally, as well documented in our previous work,10–13 referral bias could have influenced the study findings. At our tertiary medical center, patients with chronic pain who are referred to our pain rehabilitation program may not be representative of community dwelling adults with chronic pain. However, in a population-based sample of adults with chronic pain recruited from the catchment area of our tertiary referral medical center, 96% were Caucasian and 56% were women.43 Despite these demographic similarities, the risk of referral bias cannot be entirely dismissed.
In summary, among opioid-free adults with chronic pain, the TT genotype of ABCB1 c.3435C>T was associated with significantly higher HP threshold in a recessive model of allele effects. The HP threshold of the TT genotype approximated normative values, while the CC and CT genotypes were 1.00 to 0.72 SD below normative values. This study contributes to the emerging understanding of how the ABCB1 c.3435C>T polymorphism contributes to pain perception in opioid-free adults with chronic pain and provides the foundation for investigating the potential effects of this polymorphism on the clinical course of chronic pain.
Name: W. Michael Hooten, MD.
Contribution: This author helped with study conception and design, data analysis and interpretation, data acquisition, and manuscript development.
Conflicts of Interest: W. M. Hooten receives funding from the National Institutes of Health and US WorldMeds.
Name: Danqing Hu, MD, PhD.
Contribution: This author helped with study design, data acquisition, data analysis and interpretation, and manuscript development.
Conflicts of Interest: None.
Name: Julie M. Cunningham, PhD.
Contribution: This author helped with study design, data acquisition, data analysis and interpretation, and manuscript development.
Conflicts of Interest: None.
This manuscript was handled by: Jianren Mao, MD, PhD.
1. Hoffmeyer S, Burk O, von Richter O, et al. Functional polymorphisms of the human multidrug-resistance gene: multiple sequence variations and correlation of one allele with P-glycoprotein expression and activity in vivo. Proc Natl Acad Sci U S A. 2000;97:3473–3478.
2. Fromm MF. Importance of P-glycoprotein at blood-tissue barriers. Trends Pharmacol Sci. 2004;25:423–429.
3. Beaulieu E, Demeule M, Ghitescu L, Béliveau R. P-glycoprotein is strongly expressed in the luminal membranes of the endothelium of blood vessels in the brain. Biochem J. 1997;326 (Pt 2):539–544.
4. Fung KL, Gottesman MM. A synonymous polymorphism in a common MDR1 (ABCB1) haplotype shapes protein function. Biochim Biophys Acta. 2009;1794:860–871.
5. Chouchi M, Kaabachi W, Klaa H, Tizaoui K, Turki IB, Hila L. Relationship between ABCB1 3435TT genotype and antiepileptic drugs resistance in epilepsy: updated systematic review and meta-analysis. BMC Neurol. 2017;17:32.
6. Ferracini AC, Lopes-Aguiar L, Lourenço GJ, et al. GSTP1 and ABCB1 polymorphisms predicting toxicities and clinical management on carboplatin and paclitaxel-based chemotherapy in ovarian cancer. Clin Transl Sci. 2021;14:720–728.
7. King M, Su W, Chang A, Zuckerman A, Pasternak GW. Transport of opioids from the brain to the periphery by P-glycoprotein: peripheral actions of central drugs. Nat Neurosci. 2001;4:268–274.
8. Oude Elferink RP, Zadina J. MDR1 P-glycoprotein transports endogenous opioid peptides. Peptides. 2001;22:2015–2020.
9. Hooten WM, Sandroni P, Mantilla CB, Townsend CO. Associations between heat pain perception and pain severity among patients with chronic pain. Pain Med. 2010;11:1554–1563.
10. Hooten WM, Biernacka JM, O’Brien TG, Cunningham JM, Black JL. Associations of catechol-O-methyltransferase (rs4680) single nucleotide polymorphisms with opioid use and dose among adults with chronic pain. Pain. 2019;160:263–268.
11. Hooten WM, Hartman WR, Black JL III, Laures HJ, Walker DL. Associations between serotonin transporter gene polymorphisms and heat pain perception in adults with chronic pain. BMC Med Genet. 2013;14:78.
12. Hooten WM, Hu D, Cunningham JM, Black JL III. Effect of catechol-O-methyltransferase (rs4680) single-nucleotide polymorphism on opioid-induced hyperalgesia in adults with chronic pain. Mol Pain. 2019;15:1744806919848929.
13. Hooten WM, Townsend CO, Sletten CD. The triallelic serotonin transporter gene polymorphism is associated with depressive symptoms in adults with chronic pain. J Pain Res. 2017;10:1071–1078.
14. Townsend CO, Bruce BK, Hooten WM, Rome JD. The role of mental health professionals in multidisciplinary pain rehabilitation programs. J Clin Psychol. 2006;62:1433–1443.
15. Radloff L. A self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401.
16. Geisser ME, Roth RS, Robinson ME. Assessing depression among persons with chronic pain using the Center for Epidemiological Studies-Depression Scale and the Beck Depression Inventory: a comparative analysis. Clin J Pain. 1997;13:163–170.
17. McCracken LM, Dhingra L. A short version of the pain anxiety symptoms scale (PASS-20): preliminary development and validity. Pain Res Manag. 2002;7:45–50.
18. Ferreira-Valente MA, Pais-Ribeiro JL, Jensen MP. Validity of four pain intensity rating scales. Pain. 2011;152:2399–2404.
19. Dyck PJ, Zimmerman I, Gillen DA, Johnson D, Karnes JL, O’Brien PC. Cool, warm, and heat-pain detection thresholds: testing methods and inferences about anatomic distribution of receptors. Neurology. 1993;43:1500–1508.
20. Dyck PJ, Zimmerman IR, Johnson DM, et al. A standard test of heat-pain responses using CASE IV. J Neurol Sci. 1996;136:54–63.
21. Dyck PJ, Zimmerman IR, O’Brien PC, et al. Introduction of automated systems to evaluate touch-pressure, vibration, and thermal cutaneous sensation in man. Ann Neurol. 1978;4:502–510.
22. Dyck PJ, O’Brien PC, Kosanke JL, Gillen DA, Karnes JL. A 4, 2, and 1 stepping algorithm for quick and accurate estimation of cutaneous sensation threshold. Neurology. 1993;43:1508–1512.
23. Dyck PJ, O’Brien PC, Johnson DM, Klein CJ, Dyck PJB. Dyck PJ, Thomas PK, eds. Quantitative sensation testing. In: Peripheral Neuropathy. 2005:Elsevier Saunders, 1063–1093.
24. Dyck PJ, Litchy WJ, Lehman KA, Hokanson JL, Low PA, O’Brien PC. Variables influencing neuropathic endpoints: the Rochester Diabetic Neuropathy Study of Healthy Subjects. Neurology. 1995;45:1115–1121.
25. Dyck PJ, O’Brien PC, Litchy WJ, Harper CM, Daube JR, Dyck PJ. Use of percentiles and normal deviates to express nerve conduction and other test abnormalities. Muscle Nerve. 2001;24:307–310.
26. O’Brien PC, Dyck PJ. Procedures for setting normal values. Neurology. 1995;45:17–23.
27. Weir BS. Genetic Data Analysis II: Methods for Discrete Population Genetic Data. 1997.Sinauer Associates, Inc.;
28. Roşian AN, Iancu M, Trifa AP, et al. An exploratory association analysis of ABCB1 rs1045642 and ABCB1 rs4148738 with non-major bleeding risk in atrial fibrillation patients treated with dabigatran or apixaban. J Pers Med. 2020;10:E133.
29. Cliff N. Dominance statistics - ordinal analyses to answer ordinal questions. Psychol Bull. 1993;114:494–509.
30. Little J, Higgins JP, Ioannidis JP, et al.; STrengthening the REporting of Genetic Association Studies. STrengthening the REporting of genetic association studies (STREGA): an extension of the STROBE statement. PLoS Med. 2009;6:e22.
31. Thorisson GA, Smith AV, Krishnan L, Stein LD. The International HapMap Project Web site. Genome Res. 2005;15:1592–1593.
32. Zahari Z, Lee CS, Ibrahim MA, et al. Relationship between ABCB1 polymorphisms and cold pain sensitivity among healthy opioid-naive Malay males. Pain Pract. 2017;17:930–940.
33. Rolke R, Baron R, Maier C, et al. Quantitative sensory testing in the German Research Network on Neuropathic Pain (DFNS): standardized protocol and reference values. Pain. 2006;123:231–243.
34. Kim HJ, Yang GS, Greenspan JD, et al. Racial and ethnic differences in experimental pain sensitivity: systematic review and meta-analysis. Pain. 2017;158:194–211.
35. Forsythe LP, Thorn B, Day M, Shelby G. Race and sex differences in primary appraisals, catastrophizing, and experimental pain outcomes. J Pain. 2011;12:563–572.
36. Backonja MM, Attal N, Baron R, et al. Value of quantitative sensory testing in neurological and pain disorders: NeuPSIG consensus. Pain. 2013;154:1807–1819.
37. Chen C, Pollack GM. Altered disposition and antinociception of [D-penicillamine(2,5)] enkephalin in mdr1a-gene-deficient mice. J Pharmacol Exp Ther. 1998;287:545–552.
38. Chen C, Pollack GM. Enhanced antinociception of the model opioid peptide [D-penicillamine] enkephalin by P-glycoprotein modulation. Pharm Res. 1999;16:296–301.
39. Kastin AJ, Fasold MB, Zadina JE. Endomorphins, Met-enkephalin, Tyr-MIF-1, and the P-glycoprotein efflux system. Drug Metab Dispos. 2002;30:231–234.
40. Somogyvari-Vigh A, Kastin AJ, Liao J, Zadina JE, Pan W. Endomorphins exit the brain by a saturable efflux system at the basolateral surface of cerebral endothelial cells. Exp Brain Res. 2004;156:224–230.
41. Hooten WM, Lamer TJ, Twyner C. Opioid-induced hyperalgesia in community-dwelling adults with chronic pain. Pain. 2015;156:1145–1152.
42. Wang AK, Gillen DA, Dyck PJ. Effect of simple analgesics on quantitative sensation test threshold. Neurology. 1999;53:1865–1867.
43. Watkins EA, Wollan PC, Melton LJ III, Yawn BP. A population in pain: report from the Olmsted County health study. Pain Med. 2008;9:166–174.