Patients with osteoarthritis scheduled for orthopaedic surgery tend to have chronic pain and many are treated with opioids. Preoperative chronic pain and opioid consumption are risk factors for severe postoperative pain and persistent postoperative pain.1–4 The demonstration of diffuse hyperalgesia in a nonsurgical context has provided evidence of central sensitisation, both in patients with chronic osteoarthritis pain5 and in patients on long-term morphine treatment.6–9 This long-term opioid-induced hyperalgesia (OIH) has been detected by changes in heat tolerance thresholds and temporal summation test results.8 Acute OIH after high doses of remifentanil is well documented,10–13 but preoperative long-term OIH has not been reported previously in patients scheduled for surgery. However, experimental studies have suggested that sensitivity to pain is greater in animals exposed to opioids before surgery than in unexposed animals.14–18
OIH decreases the analgesic effect of opioids, resulting in a need for higher doses to achieve the same effect. OIH and opioid tolerance during the perioperative period are related clinical phenomena. The primary aim of this study was to assess the occurrence of preoperative OIH in patients scheduled for orthopaedic surgery, using quantitative sensory testing. The secondary aims were to assess its clinical consequences in terms of pain and morphine consumption after surgery in patients treated with opioids before surgery.
Materials and methods
We carried out a single-centre (Raymond Poincaré Hospital in Garches), observational, case–control study, with no direct individual benefit, on orthopaedic surgery patients. This case–control study was performed according to the STROBE statements.19 Ethical approval for the study was obtained from the appropriate institutional review board (Comité pour la Protection des Personnes, Ile de France VIII, Ambroise Paré Hospital, Boulogne 92100 No. 110435) in February 2011. The inclusion criteria were adults of American Society of Anesthesiologists’ (ASA) physical status 1 to 3 with osteoarthritis, scheduled for hip, knee or ankle arthroplasty or spine arthrodesis, taking any type of opioid daily for at least 1 month before surgery (opioid-treated group) or without opioid treatment during the month before surgery (control group). Patients were enrolled in this prospective case–control study between February and June 2011. We included consecutive patients attending anaesthesia consultations until the planned sample size was reached for the opioid-treated group. The criteria for noninclusion were occasional opioid use during the month before surgery, alcoholism or known psychiatric disease (severe depression, psychosis). The choice of type of anaesthesia was left to the discretion of the anaesthetist. For general anaesthesia, we used a balanced general anaesthesia protocol comprising propofol, sufentanil, atracurium and sevoflurane, together with a 1 : 1 mixture of nitrous oxide and oxygen. Hyperbaric bupivacaine was used for spinal anaesthesia. The anaesthetist in charge of anaesthesia was blind to the results of preoperative quantitative sensory tests (QSTs). Analgesic management followed the recommendations of the relevant scientific society:20 multimodal analgesia, including a combination of paracetamol, nefopam/tramadol and ketoprofen, unless contraindicated; continuous peripheral nerve block for 48 h for patients undergoing lower limb surgery; administration of a bolus of ketamine (0.3 mg kg−1) after induction of analgesia; systematic morphine titration when the patient regained consciousness, according to the following protocol: 3 mg every 5 min until a visual analogue score (VAS) of 3 or less, a Ramsay sedation score of more than 2 or a respiratory frequency less than 10 breaths per min was reached; and prescription of patient-controlled analgesia (PCA) with morphine (1 mg ml−1 dilution, with the programmed delivery of 1 mg every 5 min and with a maximum of 20 mg over 4 h).
On the day before surgery, we recorded the age, sex, weight, height and ASA status of the patients. We asked each patient to evaluate mean pain intensity at the intervention site during the preceding week, on an 11-point numerical rating scale (NRS; 0, no pain, 10, worst possible pain). We asked patients to specify the duration of the pain in months, and their daily dose of opioids. Daily opioid consumption was converted into morphine sulphate equivalents as follows: 1 mg morphine = 0.5 mg oxycodone = 6 mg codeine = 5 mg tramadol.
We assessed preoperative hyperalgesia on the day before surgery, using QSTs. The QSTs were selected on the basis of their feasibility at the patient's bedside and their reported ability to detect hyperalgesia.8,21 These tests were carried out on a nonpainful area of healthy skin on the right arm, in a quiet, isolated room. All tests were carried out by the same investigator (N.H.).
Pain and tolerance threshold in a mechanical pressure test
A hand-held electronic pressure algometer (Somedic AB, Stockholm, Sweden) with a 0.5 cm2 probe area was used. Pressure pain and tolerance thresholds were measured on the pulp of the third finger of the right hand, as described by Hsu et al.22 The patients were asked to press a button freezing the digital display as soon as they perceived a pain (pain detection threshold) and when the pain became unbearable (tolerance threshold). The mean of two measurements with an inter-stimulus interval of 60 s was defined as the pressure pain threshold value, expressed in kPa.
Temporal summation tests with Von Frey filaments
This test evaluates the pain triggered by the application of a 180 g Von Frey filament (Bioseb, in Vivo Research Instruments, http://www.bioseb.com) to the inner surface of the right upper arm until the filament curves. The pain induced was evaluated on an 11-point NRS after the first stimulation, and then again after a tenth consecutive application to an area of 1 cm2, at a frequency of 1 Hz, as described by Weissman-Fogel et al.23 The difference between the NRS scores obtained after the first and tenth stimulations was calculated (ΔNRS10–1).
Warm detection, heat pain and tolerance thresholds
We used a thermal test (MSA thermal stimulator; Somedic AB) for quantitative sensory measurements. The warm detection, heat pain and heat tolerance thresholds were determined by the method of limits, by gradually increasing the temperature of one thermode with a surface area of 7 cm2 from 32°C to a maximum of 52°C. Three consecutive measurements were carried out and the mean value was recorded.
Tolerance to 47°C
We determined tolerance to the application of a 7 cm2 thermode at a temperature of 47°C for a maximum of 60 s, as described by Chen et al.8 Patients were told to stop the stimulation when it became unbearable. In temporal summation tests, four identical stimuli at 47°C were applied manually to the forearm, with an inter-stimulus interval of 2 s. The patients were asked to evaluate pain immediately after the fourth stimulus. This test is a manual adaptation of the method described by Chen et al.8
After surgery, the total dose of morphine received during titration and the daily cumulative dose of morphine delivered by the pump during the first 72 h were determined. Maximal pain at rest was evaluated using a NRS in the recovery room, and then daily during the first 3 days of hospitalisation.
Sample size was calculated on the basis of the heat-induced pain detection threshold test.8 We determined that a minimum of 21 patients per group would be required to detect a 1.5°C decrease in the heat-induced pain threshold with a power of 90% and an α risk of 5%. We used Kolmogorov–Smirnov tests to check that the data were normally distributed.
Quantitative data were then compared using Student's t-tests for normally distributed data or Mann–Whitney U-tests for nonnormal data. Hyperalgesia was considered to be present in QSTs in opioid-treated patients with pain and tolerance thresholds below the 10th centile or NRS scores for pain above the 90th centile in comparison with the control group. Qualitative data were compared using χ2 tests. We considered P values less than 0.05 to be significant. SPSS version 18 software (SPSS, Chicago, IL) was used for all statistical analyses.
Seventy-three patients were asked to participate in this study (Fig. 1). Two patients were not included because of uncontrolled psychiatric disease, one patient was ineligible for surgery (skin infection) and two patients refused to participate. Thus, 68 patients in total were included in the study. All the patients had chronic pain with a mean ± SD duration of 44 ± 68 months and scored 7.6 ± 1.5 on the NRS for pain intensity. Forty patients were included in the control group and 28 in the opioid-treated group. Mean daily opioid consumption in the opioid-treated group was 42.0 ± 25.4 mg of morphine sulphate equivalent. Five of the patients were taking strong opioids (morphine sulphate n = 4, oxycodone n = 1), and the other 23 patients were taking weak opioids (tramadol n = 14, codeine n = 6, dextropropoxyphene n = 3). The two groups were comparable in terms of general characteristics and the duration and intensity of presurgical pain (Table 1). The use of regional analgesia for the control group was (21/40; 52.5%) and for the opioid-treated group was (10/28; 35.7%) (P = 0.19).
In the control group, the median (10th to 90th centile) values for pressure pain and tolerance thresholds, and mechanical temporal summation test results were 330 (181 to 584), 471 (296 to 814) kPa, and 0 (0 to 2), respectively. The patients in the opioid-treated group presented significant preoperative mechanical hyperalgesia in mechanical temporal summation tests with a Von Frey filament (P = 0.036; Table 2). In this test, 50% (14/28) of opioid-treated patients had mechanical hyperalgesia. No significant difference in the pain threshold or in tolerance to standardised pressure was found (Table 2).
In the control group, the median (10th to 90th centile) values for heat detection, pain and tolerance thresholds, duration of tolerance to stimulation at 47°C and temporal summation test results were 35.7°C (34.3°C to 37.4°C), 43.7°C (37.9°C to 48.9°C), 48.7°C (46.6°C to 50.8°C), 60.0 (20.8 to 60.0) s and 0.5 (0 to 3), respectively. Patients in the opioid-treated group had significant preoperative heat-induced hyperalgesia in two of the five thermal tests. They had a significantly lower heat tolerance threshold (P = 0.045) and they tolerated a stimulus of 47° C for a shorter period than patients in the control group (P = 0.03; Table 2). In these two tests, 50% (14/28) of opioid-treated patients had thermal hyperalgesia.
We found no differences between the groups in the other mechanical and thermal tests (Table 2).
Significant differences were found in the type of analgesia offered by the anaesthetist. Ketamine and tramadol were prescribed more frequently to patients in the opioid-treated group than to patients in the control group (Table 3). General anaesthesia was also significantly more frequent in the opioid-treated group. However, the mean ± SD pain intensity score in the recovery room was 38% higher in the opioid-treated group than in the control group (5.5 ± 2.7 vs. 7.6 ± 2.9, P < 0.001). Morphine titration doses were almost twice as high in the opioid-treated group than in the control group (9.4 ± 8.8 vs. 19.1 ± 9.5 mg, P < 0.001). Cumulative morphine consumption over the first 72 h after surgery was greater in the opioid-treated group than in the control group, and this difference was significant at 48 h (35.9 ± 24.7 vs. 22.6 ± 22.2 mg, P = 0.024) and at 72 h (39.8 ± 25.7 vs. 25.6 ± 25.8 mg, P = 0.024) (Fig. 2). After the recovery room assessments, no significant difference in pain intensity was found between the two groups at any of the time points considered during the first 72 h after surgery (Fig. 3).
Our findings indicate that OIH is a phenomenon encountered in routine clinical practice in the presurgical context. OIH in surgical patients was initially studied in the intraoperative context by assessing the experimental variation of remifentanil or sufentanil doses. Clinical evidence for OIH resulting from the long-term use of opioids has recently been reported for patients with chronic pain,7,8,18 and for individuals without pain, such as drug addicts abusing opioids.7,24,25 Our results confirm those of Chen et al.,8 who clearly demonstrated hyperalgesia induced by long-term opioid use in patients with chronic pain. However, our study also includes new elements. First, we observed hyperalgesia following the use of low doses of opioids – a mean daily dose of 42 mg of morphine sulphate equivalent – contrasting with previous studies, which reported the occurrence of OIH following the use of a daily dose of more than 75 mg of morphine.6,8 Second, most of our patients (80%) were treated with weak opioids such as tramadol, confirming that such opioids can also trigger OIH, as suggested by recent clinical cases.26 Finally, we were able to demonstrate the occurrence of OIH in a specific group, the orthopaedic surgical patients.
We found that significant changes in quantitative sensory testing response were detectable in patients with chronic pain treated with oral opioids before surgery. The changes observed included a decrease in heat tolerance threshold and heat tolerance duration, and an exacerbation of temporal mechanical summation. The suprathreshold heat hyperalgesia and mechanical temporal summation hyperalgesia in an extra-segmental territory distant from the painful site strongly suggest that diffuse central sensitisation is more intense in patients with chronic pain treated with opioids than in those with chronic pain not treated with opioids. Consistent with the findings of Chen et al.,8 thermal heat hyperalgesia differentiated between chronic pain patients treated with and without opioids. These findings also appear to be consistent with the results of animal studies based on paw withdrawal from radiant heat stimuli.27 However, unlike Chen et al.,8 we found that the mechanical temporal summation test was also useful for detecting hyperalgesia. This is an interesting finding because mechanical stimulation is easy to implement in clinical conditions. This test appeared to be more sensitive than thermal tests for the detection of OIH. Indeed, mechanical hyperalgesia was detected in a larger proportion (50%) of opioid-treated patients than thermal hyperalgesia (28%). According to our definitions and results, a difference of two points on the NRS scale between the first and tenth applications can identify patients with preoperative hyperalgesia. Further evaluations of this test are now required.
We then investigated the postoperative consequences of treatment with opioids before surgery. The patients in the opioid-treated group had a pain intensity 38% greater than that of the patients in the control group on arrival in the recovery room, and the morphine titration dose required was 66% higher in these patients. These findings confirm those of previous studies.4,28 Pain scores remained consistently higher for this group during the postoperative period, although the difference between groups was not significant, and cumulative morphine consumption was 55% higher in the opioid-treated group at 72 h. Rapp et al.28 were the first to demonstrate that patients treated with opioids before surgery had higher levels of surgical pain and higher levels of morphine consumption than patients not treated with opioids before surgery. Chapman et al.4 described a pain trajectory characterised by a higher peak and longer duration of pain in chronic pain patients treated with opioids before surgery than in chronic pain patients not treated with opioids4. Other studies have also reported the preoperative use of opioids to be predictive of the risk of pain becoming chronic.29,30 All these results highlight the concept of vulnerability to pain in experimental studies based on central sensitisation.27,31–33
This observational study was subject to several limitations that must be taken into account when interpreting the results. First, the lack of randomisation resulted in a difference in the number of patients included in each group. However, the sample size met the requirements determined by prior calculation and data from all patients included were analysed. Furthermore, the proportions of patients included in the two groups reflected real life, rendering our results generally applicable to the population of patients undergoing orthopaedic surgery. Second, we cannot exclude the possibility that the groups did not have similar pain intensities before opioid use. The summary evaluation of preoperative pain performed in our study could not evaluate all aspects of pain. Consequently, we do not know the precise respective contributions of OIH and pain intensity before opioid treatment to the development of preoperative hyperalgesia. The hyperalgesia measured probably reflects both sensitisation processes. Finally, this study covered a wide spectrum of anaesthesia and analgesia procedures, which differed between the groups. In particular, ketamine, tramadol and general anaesthesia were used more frequently in the opioid-treated group than in the control group. This had no effect on the primary outcome measured before surgery, but several major confounding factors may have affected the secondary postoperative outcomes. It is therefore important to exercise caution when interpreting the postoperative consequences of preoperative OIH.
Our work also suggests that current management practices, in accordance with national recommendations, are not sufficient to correct the consequences of hyperalgesia due to long-term opioid use.20 We observed that, despite the more frequent prescription of intraoperative ketamine and tramadol in the opioid-treated group, a greater vulnerability to postoperative pain persisted. Our findings contrast with those of previous studies showing that ketamine reverses the hyperalgesia induced by high-dose remifentanil12 or reporting lower opioid requirements and pain scores in opioid-treated patients with chronic back pain undergoing back surgery.34 These differences may be accounted for by the use of a single bolus of ketamine in our study, as opposed to the continuous administration of ketamine over a period of 24 h described in previous trials.
These results indicate that OIH may occur in common preoperative situations and that this condition remains a significant clinical challenge for the anaesthetist.35,36 These results suggest that the management of patients treated with opioids before surgery should be tailored to the individual.
Acknowledgements relating to this article
Assistance with the article: none.
Financial support and sponsorship: we would like to thank the APICIL foundation for financial support.
Conflicts of interest: none.
Presentation: preliminary data from this study were presented as a poster at the French Society of Anaesthesiology and Reanimation (SFAR), 19 to 22 September 2012, Paris.
1. Kehlet H, Jensen TS, Woolf CJ. Persistent postsurgical pain: risk factors and prevention. Lancet
2. Macrae WA. Chronic postsurgical pain: 10 years on. Br J Anaesth
3. Perkins FM, Kehlet H. Chronic pain as an outcome of surgery. A review of predictive factors. Anesthesiology
4. Chapman CR, Davis J, Donaldson GW, et al. Postoperative pain trajectories in chronic pain patients undergoing surgery: the effects of chronic opioid pharmacotherapy on acute pain. J Pain
5. Arendt-Nielsen L, Nie H, Laursen MB, et al. Sensitization in patients with painful knee osteoarthritis. Pain
6. Chu LF, Clark DJ, Angst MS. Opioid tolerance and hyperalgesia in chronic pain patients after one month of oral morphine therapy: a preliminary prospective study. J Pain
7. Compton P, Charuvastra VC, Ling W. Pain intolerance in opioid-maintained former opiate addicts: effect of long-acting maintenance agent. Drug Alcohol Depend
8. Chen L, Malarick C, Seefeld L, et al. Altered quantitative sensory testing outcome in subjects with opioid therapy. Pain
9. Woolf CJ. Central sensitization: implications for the diagnosis and treatment of pain. Pain
2011; 152 (Suppl 3):S2–S15.
10. Guignard B, Bossard AE, Coste C, et al. Acute opioid tolerance: intraoperative remifentanil increases postoperative pain and morphine requirement. Anesthesiology
11. Salengros JC, Huybrechts I, Ducart A, et al. Different anesthetic techniques associated with different incidences of chronic postthoracotomy pain: low-dose remifentanil plus presurgical epidural analgesia is preferable to high-dose remifentanil with postsurgical epidural analgesia. J Cardiothorac Vasc Anesth
12. Joly V, Richebe P, Guignard B, et al. Remifentanil-induced postoperative hyperalgesia and its prevention with small-dose ketamine. Anesthesiology
13. Fletcher D, Martinez V. Opioid-induced hyperalgesia in patients after surgery: a systematic review and a meta-analysis. Br J Anaesth
14. Kim DH, Fields HL, Barbaro NM. Morphine analgesia and acute physical dependence: rapid onset of two opposing, dose-related processes. Brain Res
15. Laulin JP, Maurette P, Corcuff JB, et al. The role of ketamine in preventing fentanyl-induced hyperalgesia and subsequent acute morphine tolerance. Anesth Analg
16. Laboureyras E, Chateauraynaud J, Richebé P, Simmonet G. Long-term pain vulnerability after surgery in rats: prevention by nefopam, an analgesic with antihyperalgesic properties. Anesth Analg
17. Celerier E, Rivat C, Jun Y, et al. Long-lasting hyperalgesia induced by fentanyl in rats: preventive effect of ketamine. Anesthesiology
18. Celerier E, Laulin JP, Corcuff JB, et al. Progressive enhancement of delayed hyperalgesia induced by repeated heroin administration: a sensitization process. J Neurosci
19. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet
20. Fletcher D, Aubrun F. [Long texts for the formalized recommendation of experts on management of postoperative pain]. Ann Fr Anesth Reanim
21. Werner MU, Mjobo HN, Nielsen PR, Rudin A. Prediction of postoperative pain: a systematic review of predictive experimental pain studies. Anesthesiology
22. Hsu YW, Somma J, Hung YC, et al. Predicting postoperative pain by preoperative pressure pain assessment. Anesthesiology
23. Weissman-Fogel I, Granovsky Y, Crispel Y, et al. Enhanced presurgical pain temporal summation response predicts postthoracotomy pain intensity during the acute postoperative phase. J Pain
24. Ho A, Dole VP. Pain perception in drug-free and in methadone-maintained human ex-addicts. Proc Soc Exp Biol Med
25. Angst MS, Clark JD. Opioid-induced hyperalgesia: a qualitative systematic review. Anesthesiology
26. Lee SH, Cho SY, Lee HG, et al. Tramadol induced paradoxical hyperalgesia. Pain Phys
27. Mao J, Price DD, Mayer DJ. Thermal hyperalgesia in association with the development of morphine tolerance in rats: roles of excitatory amino acid receptors and protein kinase C. J Neurosci
28. Rapp SE, Ready LB, Nessly ML. Acute pain management in patients with prior opioid consumption: a case-controlled retrospective review. Pain
29. Zywiel MG, Stroh DA, Lee SY, et al. Chronic opioid use prior to total knee arthroplasty. J Bone Joint Surg Am
30. VanDenKerkhof EG, Hopman WM, Goldstein DH, et al. Impact of perioperative pain intensity, pain qualities, and opioid use on chronic pain after surgery: a prospective cohort study. Reg Anesth Pain Med
31. Richebé P, Rivat C, Laulin JP, et al. Ketamine improves the management of exaggerated postoperative pain observed in perioperative fentanyl-treated rats. Anesthesiology
32. Minville V, Fourcade O, Girolami JP, Tack I. Opioid-induced hyperalgesia in a mice model of orthopaedic pain: preventive effect of ketamine. Br J Anaesth
33. Mao J. Opioid-induced abnormal pain sensitivity: implications in clinical opioid therapy. Pain
34. Loftus RW, Yeager MP, Clark JA, et al. Intraoperative ketamine reduces perioperative opiate consumption in opiate-dependent patients with chronic back pain undergoing back surgery. Anesthesiology
35. Martinez V, Fletcher D. Prevention of opioid-induced hyperalgesia in surgical patients: does it really matter? Br J Anaesth
36. Colvin LA, Fallon MT. Opioid-induced hyperalgesia: a clinical challenge. Br J Anaesth