Esmolol is a β1-adrenoreceptor antagonist with a rapid onset and short duration.1–3 The drug is used to control hypertension and tachyarrhythmia,1 and in anaesthesia, to attenuate the sympathetic response to laryngoscopy and tracheal intubation.4–6 Furthermore, new areas of use have emerged7 including a reduction in volatile anaesthetic,8–11 and opioid requirements during surgery.12 The intraoperative administration of esmolol may also reduce postoperative analgesic requirements,9–11,13,14 suggesting it could possibly have an important perioperative role.
Several animal studies have provided results suggesting that esmolol may possess inherent analgesic properties.15,16 However, only a few trials have addressed the possibility of a direct analgesic effect of esmolol in men.17–19 In these trials, pretreatment with esmolol was shown to reduce the incidence of injection pain with propofol and subparalysing doses of rocuronium, respectively. It has also been suggested that the opioid-sparing effect may in fact be related to secondary effects, such as synergism with, or altered pharmacokinetics of coadministered anaesthetics and opioids, or possibly through the avoidance of opioid-induced hyperalgesia.9,12,20–25
Should esmolol have a specific analgesic action, it could be used to complement existing multimodal analgesia strategies to reduce the perioperative use of opioids and possibly opioid-related complications.
The main objective of this study was to evaluate whether esmolol, in the absence of opioids or anaesthetics, has an analgesic effect during experimental pain testing using the cold pressor test (CPT). We also aimed to evaluate the effect of esmolol in attenuating the sympathetic stress response during CPT, and whether or not esmolol produces any of the side-effects associated with remifentanil treatment (swallowing difficulty, nausea, respiratory depression and desaturation).
The study protocol was approved by the Regional Ethics Committee in Uppsala, Stockholm, Sweden (Dnr 2012/070) on 7 March 2012 (Chairperson Erik Lempert) and by the Swedish Medical Products Agency on 23 March 2012 (No. 151 : 2012/12484) and registered in the European Clinical Trials database (EudraCT no. 2011-005780-24, https://eudract.ema.europa.eu/) prior to the start of enrolment. It was conducted at the Department of Anaesthesia and Intensive Care, University Hospital in Örebro, Örebro, Sweden between November 2013 and February 2014. Volunteer recruitment was by advertisements on the Örebro University notice boards, and the study participants were included consecutively. The study participants were fully informed of the details of the study protocol prior to obtaining written informed consent. Financial remuneration was provided.
Inclusion criteria were ASA classification I; men or women; BMI less than 30 kg m−2 and age 18 to 40 years. Exclusion criteria were: ongoing treatment with cardiovascular medication, benzodiazepines or analgesics; allergy to drugs used in the trial; pregnancy and breastfeeding and participation in other medical trials.
After a brief physical examination and interview where exclusion criteria were ruled out, study participants were familiarised with the outline of the protocol, the CPT and the numeric pain rating scale (NRS) used (range 0 to 10, 0 = no pain, 10 = worst pain imaginable).26,27 An intravenous line for fluid and drug administration was established and monitoring equipment applied [ECG, pulse oximetry and noninvasive blood pressure (BP)]. Monitoring of the above parameters was performed throughout each study session and documented at 5-min intervals. All study participants went through three intervention sessions, each session being separated by at least 3 days. During each study session, two infusion pumps (Alaris CC syringe pump, Alaris Medical Nordic AB, Solna, Stockholm County, Sweden) were used for intravenous administration of the study drugs. In session A, esmolol (Brevibloc; Baxter Healthcare Ltd., Thetford, Great Britain) was administered by one of the pumps as a bolus dose of 0.7 mg kg−1 over 1 min followed by an infusion of 10 μg kg−1 min−1 over 30 min administered by the second pump. In session B, remifentanil (Remifentanil Teva; TEVA Pharmaceuticals Works Private Limited Company, Gödöllö, Pest County, Hungary) was administered as an infusion of 0.2 μg kg−1 min−1 that was terminated without gradual withdrawal after 30 min. The infusion of remifentanil was preceded by a bolus of saline over 1 min to resemble the bolus administration of esmolol. During session C, saline was administered by both syringe pumps as a 1-min bolus followed by a 30-min infusion.
At the first study session, the study participants were randomised to the sequence of intervention (A-B-C, A-C-B, B-A-C, B-C-A, C-A-B or C-B-A) using sealed opaque envelopes. Envelopes were prepared by departmental staff who had no other part in the study. In preparation, a note containing one of the six possible intervention sequences was placed in each envelope, which were sealed, mixed and numbered. The envelopes were then used consecutively. The study participants were blinded to the intervention sequence.
Experimental pain testing was performed using the CPT.28,29 The device used for the CPT consisted of a 10-l plastic container that was two-thirds filled with tap water and one-third crushed ice. Constant manual stirring and temperature measurement ensured an even target temperature of 0.0 to 1.0°C. During each CPT, the right hand was immersed in the ice water mixture above the wrist. Pain intensity levels (NRS) and heart rate (HR) were measured immediately before, and at 15 s intervals during immersion. Noninvasive BP measurements were performed as frequently as possible. Study participants were informed that voluntary hand withdrawal was possible at any time during the test. The test was terminated after 2 min if withdrawal had not already taken place.
Measures of cold pain intensity [NRS-max, NRS-area under curve (AUC)]30,31 and cold pain tolerance were used. NRS-max was defined as the maximum NRS score during the CPT.31 The area under the NRS pain curve was calculated using trapezoidal approximation using the NRS-max to extrapolate the curve if withdrawal had taken place before 2 min. Cold pain tolerance was defined as the time(s) from immersion to spontaneous withdrawal,32 or termination of the test whichever was the case.
During each study session, a cold pressure test was performed prior to intervention, at the very end of the 30-min intervention period, and 20 min after the infusion had been stopped.
The primary end point was the perceived maximum pain intensity score (NRS-max). Secondary end points were the pain intensity score NRS-AUC and pain tolerance. Haemodynamic changes (BP, HR) during CPT, and the occurrence of side-effects defined as oxygen saturation 92% or less, respiratory rate 8 min−1 or less, subjective swallowing difficulty and nausea were also measured. Initially, we aimed to evaluate the degree of swallowing difficulty, and nausea each on a four-point scale. However, as few volunteers had difficulty with swallowing or nausea, only the occurrence of swallowing difficulty or nausea, regardless of severity, are reported in the results section. Comparisons were made between esmolol and placebo and between remifentanil and placebo.
As no study similar to the present one, on the analgesic effect of esmolol, had been performed before, it was difficult to perform a sample size estimation. In theory, during intervention with esmolol compared to placebo, with a ±SD of 2, a sample size of 10 paired comparisons should have been sufficient to demonstrate a reduction in NRS-max of 2 points on the NRS scale given the power of 0.8 and a type 1 error probability of 0.05. In the present study 14 study participants were enrolled.
Pain intensity scores (NRS-max, NRS-AUC) data were evaluated using a linear mixed model with unstructured covariance structure, which showed the best model fit from Akaike Information Criteria. Intervention (esmolol/remifentanil/placebo), time of CPT (before, during and after drug administration), sequence of intervention and the interaction variable intervention by time, were categorical independent variables in the mixed model. Analysis of haemodynamic data was performed in a similar manner, except that it was adjusted for baseline values. The Shapiro–Wilk test was used to verify the normality assumption of the residuals from the mixed model. The correlation between NRS-max and NRS-AUC was measured using Pearson's correlation coefficient (r).
The Kaplan–Meier method was used to visualise cold pain tolerance, and the log-rank test was used to compare times to hand withdrawal between placebo and esmolol and remifentanil, respectively.
Fisher's exact test was used to analyse categorical side-effect data (presence of swallowing difficulty, oxygen saturation ≤92%, respiratory rate ≤8 and nausea). As two interventions were compared with placebo (esmolol vs. placebo and remifentanil vs. placebo), correction for multiple testing was performed using the Bonferroni–Holm method.33 Corrected P values less than 0.05 were considered statistically significant. Mean differences with 95% confidence intervals (CI) were used as association measures. Statistical analyses were performed using SPSS version 22.0 (IBM Corp., Armonk, New York, USA) and STATA release 14 (College Station, Texas, USA: StataCorp LP).
The manuscript adheres to the applicable equator guidelines (Fig. 1, CONSORT flow diagram) and the Helsinki declaration.
All 14 study participants completed the study sessions without any unexpected incidents. Demographic data are presented in Table 1.
The mean pain intensity scores, measured as NRS-max, were similar with esmolol 8.5 (±1.4) and placebo 8.4 (±1.3); mean difference 0.1 [95% CI (−1.2 to 1.4)], P = 0.83 (Table 2). Remifentanil, on the other hand, significantly reduced the pain intensity score, 5.4 (±2.1), compared to placebo; mean difference −3.1 [95% CI (−4.4 to −1.8)], P < 0.001. Similar findings were observed when the pain intensity score was measured as NRS-AUC (Table 2). NRS-max and NRS-AUC values showed a very high correlation (r = 0.9). Furthermore, mean pain intensity scores (NRS-max and NRS-AUC) were similar in the CTPs performed after interventions, as in the CPTs performed before (Table 2).
The Kaplan–Meier figure shows the cumulative proportion of hand withdrawals because of pain during CPT (Fig. 2). No significant difference between esmolol and placebo regarding pain tolerance was demonstrated, P = 1.0. When remifentanil was administered, however, all study participants were able to tolerate the 120 s CPT, (P < 0.05 compared to placebo).
Compared to placebo, remifentanil caused significantly more episodes of respiratory depression than esmolol (Table 3). Otherwise, no significant differences in side-effects (Table 3) and haemodynamic changes (Table 4) were demonstrated.
It is well established that the sympathetic system is involved in nociception, and numerous studies on the possible involvement of β-blockers in pain management have been performed.9–11,13,14 With its short onset and duration, esmolol is a β-receptor antagonist suitable for intraoperative use,3 and some reports have indicated that it may possess antinociceptive properties per se.15–19 The possibility of an inherent analgesic effect is, however, controversial, and the exact mechanism of action remains unknown.
It has been suggested that β-receptor antagonists could attenuate stress-related secretion of noradrenaline in the hippocampus,34 a process possibly involved in antinociception. Furthermore, esmolol has been shown to block tetrodotoxin-resistant sodium channels in dorsal root ganglia,35 and to modulate neurotransmission in the trigeminal nucleus of the substantia gelatinosa,36 suggesting attenuating of afferent signals and facilitation of the pain inhibitory system in the spinal cord. Finally, noradrenaline increases heat-induced hyperalgesia in skin that has been sensitised by capsaicin,37 which implies that β-receptor antagonists could be used to modify peripheral inflammatory reactions. The postulated antinociceptive action of esmolol could thus be explained by modulation of pain signals at central, spinal and peripheral levels. The hydrophilic nature of esmolol, however, means that it does not readily cross the blood–brain barrier,3 thus making an action at the central level unlikely.
In addition to the postulated direct action on pain, it has also been suggested that the opioid-sparing effect of esmolol could be related to the secondary effects mentioned in the introduction section.9,12,13,20,23–25,38 As little is known about the mechanisms involved in the opioid-sparing effect of esmolol, the aim of the present study was to establish whether or not the very low dose of esmolol used during anaesthesia for postoperative opioid sparing,10–13 has an analgesic effect when administered alone. Our main finding was that esmolol had no inherent analgesic properties when evaluated with the CPT. Mean NRS-max pain intensity scores during intervention with esmolol and placebo were similar. The CI limit of the primary end point (difference in NRS-max between esmolol and placebo) was nowhere near the hypothesised 2-point reduction used to estimate the sample size, suggesting that the nonsignificant result was not because of the sample size being too small.
Remifentanil on the other hand, produced a significant decrease in the mean NRS-max pain intensity score compared to placebo, which was expected.39 When CPTs were repeated 20 min after termination of the remifentanil infusion, maximum intensity pain scores were similar as in the CPTs performed before initiation of the intervention, implying that development of hyperalgesia did not occur (Table 2).
The experiments in this study were performed in a controlled environment where the CPT was used as a surrogate for perioperative pain. For obvious reasons, the experimental setting cannot be fully compared with the clinical setting of surgery and anaesthesia, where many external and internal factors affect the patient. However, it enabled us to study the postulated analgesic effect of esmolol per se, without the interference of other factors. The failure to demonstrate an analgesic effect may imply that the opioid-sparing effect of esmolol, when used as an adjunct to anaesthesia, could instead be because of one, or several of these secondary factors, rather than a direct analgesic effect. This suggestion is further supported when taking into account the pharmacokinetic properties of esmolol, whose postoperative effect on pain otherwise should reach a point far beyond that where the drug has been fully eliminated.3
Remifentanil-induced episodes of low respiratory rate and desaturation, whereas esmolol, like placebo, had no such negative side-effects. This was expected as remifentanil is known to cause dose-dependent respiratory depression.39 In addition, our research group has demonstrated that remifentanil may impair swallowing function40 and cause pulmonary aspiration in healthy volunteers.41 We, therefore, wanted to explore whether esmolol could cause subjective swallowing difficulties. No such side-effects were demonstrated. In contrast, even though not statistically significant, remifentanil caused more volunteers to experience swallowing difficulties than did placebo in this setting.
The result of the present study raises some issues that need to be addressed. The dose of esmolol was chosen to resemble the clinical studies demonstrating an opioid-sparing effect of esmolol when administered as adjunct to anaesthesia, rather than a dose that cause HR reduction. In those studies, an initial bolus dose of 0.5 to 1 mg kg−1 in conjunction with induction of anaesthesia, followed by a maintenance dose of between 5 and 15 μg kg−1 min−1 was used.10–13 Given the short elimination half-life of esmolol, the plasma concentration should be very low when the CPT is performed toward the end of the 30-min infusion. This assertion is supported by the failure of esmolol to reduce the HR prior to the CPT, and to attenuate the sympathetic response (rise in BP and HR) during the CPT, compared to placebo. Contrary to the low continuous dose of esmolol used in the present study, esmolol was administered in considerably higher doses in animal studies demonstrating possible direct analgesic properties.35,36 It would, therefore, be interesting to perform a study using a dose titration design to evaluate the possible analgesic effect of esmolol.
Apart from the question of dosage, there are a few other limitations that need to be addressed. The short elimination half-times of esmolol (approximately 7 to 9 min)1,3 and remifentanil (approximately 3 min)42,43 exclude any carryover effect between study sessions. However, even though a very low dose of esmolol was used, the wash out period of 20 min within each study session may be somewhat short to rule out any remaining effect of esmolol during the CPT performed after drug administration. However, as the β-blocking effect of esmolol has been shown to be short lived, with complete recovery of HR 20 min after termination of a 300 μg kg−1 min−1 infusion,1 the wash out period in the present study was considered to be long enough to avoid clinical carryover effects considering the much lower continuous dose that was used (10 μg kg−1 min−1).
In the present study, we managed to enrol only three women (Table 1). The skewed sex distribution may be considered a weakness as men and women perceive noxious stimuli differently.44 However, as the cross-over design allows each volunteer to act as his/her own control, the uneven sex distribution should not have affected the result and is of limited concern.
Finally, various tests of experimentally induced pain sometimes exhibit low intermodality correlation.45,46 The study may, therefore, have benefitted from the addition of another experimental pain modality. The CPT, however, is considered a reliable method for experimental testing of nociceptive pain,30,47,48 and is a standardised well documented procedure that offers good test reliability.29,49–51 Furthermore, the autonomic response to the CPT is well documented.28,52
The study also has several strengths worth mentioning. The two pain intensity scores (NRS-max and NRS-AUC) correlated highly (r = 0.9), suggesting that they truly represented the intended parameter to be measured. The pain tolerance parameter was also in agreement. Furthermore, during the ‘placebo session’ the maximum pain intensity scores were very similar in the before, during and after test sessions, suggesting consistent stability of the test method, and the unlikelihood of test-to-test conditioning,53 temporal summation54 or effect of learning. The cross-over design reduced the risk for confounding issues. To compensate for the slight imbalance in the numbers of study participants randomised to each intervention sequence, the mixed model analysis was adjusted for the sequence of intervention. This adjustment had little impact and did not alter the study findings.
Contrary to our hypothesis, no sign of a direct analgesic effect of esmolol was seen during CPTs when administered as a single drug in a dose similar to that previously shown to cause intra and postoperative opioid sparing. This could possibly be because of the low dose, and could be clarified by a dose titration study.
However, our study indicates that the opioid-sparing effect of esmolol, more likely, is secondary to other factors such as synergy with opioids, or prevention of opioid-induced hyperalgesia, rather than a direct analgesic effect of the drug.
Acknowledgements relating to this article
Assistance with the study: we thank Professor Magnus Wattwil for his contributions to this study and to MD Stefan Enbuske for the help with data acquisition.
Financial support and sponsorship: the work was supported by the Medical Research Fund, Örebro County Council, Örebro, Sweden.
Conflicts of interest: none.
Presentation: preliminary data from this study were presented in a lecture at the American Society of Anesthesiologists Annual Meeting, ‘Anesthesiology 2014’, 11 to 15 October 2014, New Orleans, Louisiana, USA.
1. Reilly CS, Wood M, Koshakji RP, et al. Ultra-short-acting beta-blockade: a comparison with conventional beta-blockade. Clin Pharmacol Ther
2. Wiest D. Esmolol. A review of its therapeutic efficacy and pharmacokinetic characteristics. Clin Pharmacokinet
3. Sum CY, Yacobi A, Kartzinel R, et al. Kinetics of esmolol, an ultra-short-acting beta blocker, and of its major metabolite. Clin Pharmacol Ther
4. Helfman SM, Gold MI, DeLisser EA, et al. Which drug prevents tachycardia and hypertension associated with tracheal intubation: lidocaine, fentanyl, or esmolol? Anesth Analg
5. Miller DR, Martineau RJ, Wynands JE, et al. Bolus administration of esmolol for controlling the haemodynamic response to tracheal intubation: the Canadian Multicentre Trial. Can J Anaesth
6. Bensky KP, Donahue-Spencer L, Hertz GE, et al. The dose-related effects of bolus esmolol on heart rate and blood pressure following laryngoscopy and intubation. AANA J
7. Harless M, Depp C, Collins S, et al. Role of esmolol in perioperative analgesia and anesthesia: a literature review. AANA J
8. Johansen JW, Schneider G, Windsor AM, et al. Esmolol potentiates reduction of minimum alveolar isoflurane concentration by alfentanil. Anesth Analg
9. Chia YY, Chan MH, Ko NH, et al. Role of beta-blockade in anaesthesia and postoperative pain management after hysterectomy. Br J Anaesth
10. Moon YE, Hwang WJ, Koh HJ, et al. The sparing effect of low-dose esmolol on sevoflurane during laparoscopic gynaecological surgery. J Int Med Res
11. White PF, Wang B, Tang J, et al. The effect of intraoperative use of esmolol and nicardipine on recovery after ambulatory surgery. Anesth Analg
12. Hwang WJ, Moon YE, Cho SJ, et al. The effect of a continuous infusion of low-dose esmolol on the requirement for remifentanil during laparoscopic gynecologic surgery. J Clin Anesth
13. Collard V, Mistraletti G, Taqi A, et al. Intraoperative esmolol infusion in the absence of opioids spares postoperative fentanyl in patients undergoing ambulatory laparoscopic cholecystectomy. Anesth Analg
14. Lee MH, Chung MH, Han CS, et al. Comparison of effects of intraoperative esmolol and ketamine infusion on acute postoperative pain after remifentanil-based anesthesia in patients undergoing laparoscopic cholecystectomy. Korean J Anesthesiol
15. Davidson EM, Doursout MF, Szmuk P, et al. Antinociceptive and cardiovascular properties of esmolol following formalin injection in rats. Can J Anaesth
16. Ono H, Ohtani N, Matoba A, et al. Efficacy of intrathecal esmolol on heat-evoked responses in a postoperative pain model. Am J Ther
17. Akgun Salman E, Titiz L, Akpek E, et al. Pretreatment with a very low dose of intravenous esmolol reduces propofol injection pain. Agri
18. Yavascaoglu B, Kaya FN, Ozcan B. Esmolol pretreatment reduces the frequency and severity of pain on injection of rocuronium. J Clin Anesth
19. Lee M, Kwon T, Kim S, et al. Comparative evaluation of the effect of remifentanil and 2 different doses of esmolol on pain during propofol injection: a double-blind, randomized clinical consort study. Medicine (Baltimore)
20. Avram MJ, Krejcie TC, Henthorn TK, et al. Beta-adrenergic blockade affects initial drug distribution due to decreased cardiac output and altered blood flow distribution. J Pharmacol Exp Ther
21. Koppert W, Schmelz M. The impact of opioid-induced hyperalgesia for postoperative pain. Best Pract Res Clin Anaesthesiol
22. Angst MS. Intraoperative use of remifentanil for TIVA: postoperative pain, acute tolerance, and opioid-induced hyperalgesia. J Cardiothorac Vasc Anesth
2015; 29 (Suppl 1):S16–S22.
23. Lee SJ, Lee JN. The effect of perioperative esmolol infusion on the postoperative nausea, vomiting and pain after laparoscopic appendectomy. Korean J Anesthesiol
24. Johansen JW, Flaishon R, Sebel PS. Esmolol reduces anesthetic requirement for skin incision during propofol/nitrous oxide/morphine anesthesia. Anesthesiology
25. Bhawna, Bajwa SJ, Lalitha K, et al. Influence of esmolol on requirement of inhalational agent using entropy and assessment of its effect on immediate postoperative pain score. Indian J Anaesth
26. Williamson A, Hoggart B. Pain: a review of three commonly used pain rating scales. J Clin Nurs
27. Ferreira-Valente MA, Pais-Ribeiro JL, Jensen MP. Validity of four pain intensity rating scales. Pain
28. Wolf S, Hardy JD. Studies on pain. Observations on pain due to local cooling and on factors involved in the ‘cold pressor’ effect. J Clin Invest
29. Compton P, Athanasos P, Elashoff D. Withdrawal hyperalgesia after acute opioid physical dependence in nonaddicted humans: a preliminary study. J Pain
30. Mauermann E, Filitz J, Dolder P, et al. Does fentanyl lead to opioid-induced hyperalgesia in healthy volunteers?: a double-blind, randomized, crossover trial. Anesthesiology
31. Treister R, Nielsen CS, Stubhaug A, et al. Experimental comparison of parametric versus nonparametric analyses of data from the cold pressor test. J Pain
32. von Baeyer CL, Piira T, Chambers CT, et al. Guidelines for the cold pressor task as an experimental pain stimulus for use with children. J Pain
33. Holm S. A simple sequentially rejective multiple test procedure. Scand J Stat
34. Sarvey JM, Burgard EC, Decker G. Long-term potentiation: studies in the hippocampal slice. J Neurosci Methods
35. Tanahashi S, Iida H, Dohi S, et al. Comparative effects of ultra-short-acting beta1-blockers on voltage-gated tetrodotoxin-resistant Na+ channels in rat sensory neurons. Eur J Anaesthesiol
36. Yasui Y, Masaki E, Kato F. Esmolol modulates inhibitory neurotransmission in the substantia gelatinosa of the spinal trigeminal nucleus of the rat. BMC Anesthesiol
37. Drummond PD. Noradrenaline increases hyperalgesia to heat in skin sensitized by capsaicin. Pain
38. Lopez-Alvarez S, Mayo-Moldes M, Zaballos M, et al. Esmolol versus ketamine-remifentanil combination for early postoperative analgesia after laparoscopic cholecystectomy: a randomized controlled trial. Can J Anaesth
39. Noseir RK, Ficke DJ, Kundu A, et al. Sympathetic and vascular consequences from remifentanil in humans. Anesth Analg
40. Savilampi J, Omari T, Magnuson A, et al. Effects of remifentanil on pharyngeal swallowing: a double blind randomised cross-over study in healthy volunteers. Eur J Anaesthesiol
41. Savilampi J, Ahlstrand R, Magnuson A, et al. Aspiration induced by remifentanil: a double-blind, randomized, crossover study in healthy volunteers. Anesthesiology
42. Kapila A, Glass PS, Jacobs JR, et al. Measured context-sensitive half-times of remifentanil and alfentanil. Anesthesiology
43. Westmoreland CL, Hoke JF, Sebel PS, et al. Pharmacokinetics of remifentanil (GI87084B) and its major metabolite (GI90291) in patients undergoing elective inpatient surgery. Anesthesiology
44. Riley JL 3rd, Robinson ME, Wise EA, et al. Sex differences in the perception of noxious experimental stimuli: a meta-analysis. Pain
45. Janal MN, Glusman M, Kuhl JP, et al. On the absence of correlation between responses to noxious heat, cold, electrical and ischemic stimulation. Pain
46. Ruscheweyh R, Stumpenhorst F, Knecht S, et al. Comparison of the cold pressor test and contact thermode-delivered cold stimuli for the assessment of cold pain sensitivity. J Pain
47. Olesen AE, Brock C, Sverrisdottir E, et al. Sensitivity of quantitative sensory models to morphine analgesia in humans. J Pain Res
48. Velasco M, Gomez J, Blanco M, et al. The cold pressor test: pharmacological and therapeutic aspects. Am J Ther
49. Chen AC, Dworkin SF, Haug J, et al. Human pain responsivity in a tonic pain model: psychological determinants. Pain
50. Mitchell LA, MacDonald RA, Brodie EE. Temperature and the cold pressor test. J Pain
51. Garcia de Jalon PD, Harrison FJ, Johnson KI, et al. A modified cold stimulation technique for the evaluation of analgesic activity in human volunteers. Pain
52. Fasano ML, Sand T, Brubakk AO, et al. Reproducibility of the cold pressor test: studies in normal subjects. Clin Auton Res
53. Lewis GN, Heales L, Rice DA, et al. Reliability of the conditioned pain modulation paradigm to assess endogenous inhibitory pain pathways. Pain Res Manag
54. Cathcart S, Winefield AH, Rolan P, et al. Reliability of temporal summation and diffuse noxious inhibitory control. Pain Res Manag