Tissue injury produces a barrage of nociceptive input to the nervous system, producing sensory changes characterized by prolonged pain, increased sensitivity to painful stimuli (hyperalgesia), and pain after innocuous stimuli (allodynia) (1). These changes persist long after the initial injury and appear to occur within the central nervous system (CNS) at the level of the spinal cord and possibly elsewhere (1,2). The response properties of spinal dorsal horn neurons are changed after tissue injury to result in enlarged receptive fields and increased excitability (3,4). The increase in excitability involves the activation of N-methyl-d-aspartate receptors by excitatory amino acids, such as glutamate and aspartate (5,6), and neuropeptides, substance P, and calcitonin gene-related peptide (7). It has been proposed that the release of excitatory amino acids at sites within the CNS leads to activation of N-methyl-d-aspartate receptor sites and excessive depolarization, contributing to the expansion of receptive fields and hyperexcitability and thereby leading to an amplification of pain and an increase in its duration (8). These changes in the CNS initiated by afferent nociceptive barrage are characterized as central sensitization or central hyperexcitability and contribute to postoperative pain.
Demonstration of central sensitization in animal models after tissue injury and its reduction by the administration of opioids or local anesthetics administered before tissue injury (2,9,10) led to clinical studies evaluating this phenomenon in humans. Prospective clinical studies demonstrated that preincisional administration of a local anesthetic reduces pain in comparison to surgery performed without local anesthesia (11) or postoperative infiltration of a local anesthetic (12). We previously demonstrated that the administration of the long-acting local anesthetic, bupivacaine, in comparison to a saline placebo before oral surgery suppressed the intraoperative release of pituitary β-endorphin, an index of central nociceptive input, and resulted in reduced spontaneous pain reporting at 48 h (13). Pain and plasma β-endorphin levels were increased at 1 h after surgery in the placebo group, indicating a continued afferent nociceptive barrage after surgery that may have contributed to the development of central sensitization. However, that study did not differentiate between the preemptive effect of the long-acting local anesthetic, bupivacaine, and its carryover into the postoperative period. This present study was designed to selectively block intraoperative nociceptive input, postoperative pain, or both by using a before-and-after factorial design as recommended by Katz (14) and Kissin (15). The results demonstrate that reducing pain in the immediate postoperative period is more effective for minimizing the establishment of central sensitization than is blocking the intraoperative afferent barrage after oral surgery.
Subjects were oral surgery outpatients undergoing the surgical removal of two to four third molars and who had expressed a preference for general anesthesia at the time of their initial screening visit. Inclusion criteria included the presence of two partial or full bony impacted third molars and missing, erupted, or soft-tissue impacted maxillary third molars, to maximize pain generated from the mandible and minimize that from the maxilla. Subjects were free of systemic disease, were not taking any concomitant analgesic medications, and provided informed consent to the risks of the surgical procedure, general anesthesia, and participation in the study. The clinical protocol and informed consent document were approved by the IRB of the National Institute of Dental and Craniofacial Research, National Institutes of Health.
On the day of surgery, a blood sample was drawn for baseline measurement of plasma β-endorphin 20 min after venipuncture. Patients were then premedicated with a sedative dose of midazolam (mean dose, 3.0 ± 1.3 mg) to alleviate anxiety (Fig. 1). Local anesthetic or saline placebo was randomly allocated and administered before and at the end of surgery in a double-blinded, parallel-groups factorial design to result in four treatment groups: preoperative administration of 2% lidocaine with 1:200,000 epinephrine and postoperative injection of saline with epinephrine 1:200,000 (preoperative local anesthesia group); preoperative saline placebo and postoperative 0.5% bupivacaine, both with 1:200,000 epinephrine (postoperative local anesthesia group); preoperative 2% lidocaine and postoperative 0.5% bupivacaine, both with 1:200,000 epinephrine (pre- and postoperative local anesthesia group); or preoperative and postoperative saline with 1:200,000 epinephrine injections (no local anesthesia group). After 5 min, with the oral surgeon outside the room, an unblinded observer assessed the efficacy of mandibular block for all subjects by probing the retromolar area with a sharp dental instrument and questioning the patient for the presence of the normal signs of local anesthesia (paresthesia of the lower lip and absence of pain upon noxious stimulation). If the regional anesthesia was not complete in subjects administered lidocaine, the anesthetic was readministered and the efficacy assessed again after 5 min.
General anesthesia was then induced with propofol and succinylcholine. A blood sample was obtained after intubation to examine the changes in circulating β-endorphin due to the anesthetic drugs and the stimulus of intubation. Anesthesia was maintained with propofol and 60% nitrous oxide while the third molars were surgically extracted. Blood samples were obtained immediately after the last tooth extraction to examine the effects of the surgical stimulus on circulating β-endorphin levels. Local anesthesia or placebo was administered at the end of surgery, according to the randomization scheme, consisting of either 0.5% bupivacaine or saline injections with 1:200,000 epinephrine. Additional blood samples were collected at 1, 2, 3, and 4 h after surgery.
Subjects remained at the clinic to permit collection of postoperative blood samples and to ensure compliance with instructions for the administration of the initial dose of analgesic. Pain medication (acetaminophen 975 mg) was administered in the postoperative period if requested for the relief of moderate or severe pain. The duration of bupivacaine with epinephrine for mandibular nerve block is 6–8 h, as compared with a 2- to 3-h duration of nerve block for lidocaine with epinephrine (16). Patients were dispensed acetaminophen with instructions to take three tablets (975 mg) every 6 h by the clock, and they were given codeine 30 mg to be taken only if needed for unrelieved pain.
Subjects recorded analgesic drug intake in a medication diary and completed pain questionnaires consisting of a 100-mm visual analog scale (VAS) and 200-mm verbal descriptor scales (14) for pain intensity and the affective component of pain over the postoperative observation period and on awakening at 24 and 48 h, before the ingestion of any analgesics. The 100-mm VAS was anchored by the words “none” and “worst possible pain,” and subjects were instructed to “rate their pain intensity at this time.” The verbal descriptor scale for pain intensity consists of a 200-mm vertical bar with 12 verbal descriptors ranging from “weak,” “mild,” and “moderate” to “strong,” “intense,” and “very intense” spaced along the scale at intervals based on previous psychophysical rankings of their relative magnitude (17,18). 1 The verbal descriptor scale for the affective component of pain substituted similarly spaced verbal descriptors, including “unpleasant,” “annoying,” “distressing,” and “intolerable.” Subjects were instructed to mark the point on the scales that best corresponded to the intensity and unpleasantness of the pain that they were experiencing. Verbal descriptor scales are sensitive to small differences in nociceptive stimuli (18) and are useful for measuring pain in sedated outpatients (19), similar to subjects emerging from general anesthesia. Subjects were contacted by phone at 24 h after surgery and returned to the clinic at 48 h after surgery to submit the pain ratings and medication diaries and to assess compliance with the medication regimen through a pill count.
Blood samples were collected into chilled tubes containing 0.1 mL of 15% EDTA, centrifuged under refrigeration, frozen on dry ice, and stored at −80°C. Plasma samples (200 μL) were analyzed in duplicate by immunoradiometric assay (Nichols Institute Diagnostics B.V., Wijchen, The Netherlands), a method whereby the sample is incubated with antibody and 125I-labeled antibody to form a solid phase antibody (β-endorphin)-labeled antibody complex. After unbound material is removed, the radioactivity is measured with a γ counter. The concentration of β-endorphin is directly proportional to the radioactivity measured and is quantified by comparing the samples with the standard curve obtained in the same assay with known human β-endorphin standards. The limit of detection for the assay was 12.5 pg/mL.
A total of 110 subjects were enrolled; 6 were not randomized after enrollment because they did not return to the clinic for surgery. Of the 104 subjects randomized to a treatment group, 5 reported paresthesia of the inferior alveolar nerve after surgery, consistent with surgical trauma. An additional nine subjects who were administered bupivacaine at the conclusion of surgery while still under general anesthesia did not display signs of mandibular anesthesia at any of the postoperative observations. These data were not analyzed because of the ineffective intervention. The remaining 90 subjects did not differ for the mean demographic, surgical, and anesthetic variables across the four groups (Tables 1 and 2).
Data were analyzed with the BMDP statistical software package (SPSS Inc., Chicago, IL). Statistical differences among the four groups were determined by two-way analysis of variance for the results of the VAS and the verbal descriptor scales. Surgical variables, the doses of anesthetic drugs, demographic variables (age, height, and weight), and β-endorphin levels were analyzed among groups by one-way analysis of variance and Duncan’s multiple range test. For all statistical tests, differences were accepted as significant if the probability of occurrence by chance alone was <5% (P < 0.05) in a two-tailed test.
Plasma β-endorphin concentrations increased significantly during surgery in the subjects receiving the saline injections before surgery (Fig. 2), which is indicative of a nociceptive barrage sufficient to activate pituitary β-endorphin release. Plasma β-endorphin remained significantly increased at 60 min postsurgery in the placebo group, which is consistent with postoperative pain that was blocked in the other three groups by local anesthesia. β-Endorphin decreased in the group receiving bupivacaine at the end of surgery in the sample collected at 1 h postsurgery, which is consistent with blockade of postoperative pain as subjects recovered from the effects of the general anesthesia. β-Endorphin did not increase during surgery in the two groups that received lidocaine before surgery (Fig. 2). Levels remained significantly lower in the 60-min postoperative sample in the three groups receiving local anesthetic in comparison to the placebo group. The plasma concentrations of β-endorphin increased in individual patients at varying times over the remaining 3 h of the observation period as the local anesthetic effects dissipated and subjects reported postoperative pain. The administration of rescue analgesics subsequently decreased pain reports and plasma β-endorphin levels in individual subjects, confounding any mean differences among the groups at the 2- to 4-h time points (Table 3).
Acute pain over the first 4 h postsurgery (Fig. 3, upper panel) was significantly less in the two groups receiving bupivacaine after surgery (F = 60.0, P < 0.001) compared with the saline/saline treatment and the lidocaine/saline treatment (F = 2.8). Pain intensity was also lower at 48 h (Fig. 3, lower panel) in the two groups receiving bupivacaine after surgery (F = 6.8, P < 0.05), whereas no effect was demonstrated for preoperative lidocaine (F = 0.3) when analyzed by two-way analysis of variance. The affective component of pain was also significantly reduced by bupivacaine (F = 8.7, P < 0.01), but not lidocaine (F = 1.4), at 48 h postsurgery (Table 4). Similar results were seen for the VAS for pain intensity (Table 4).
The time to request for analgesics in the immediate postoperative period varied relative to the duration of the local anesthetic (Table 3). Most subjects in the placebo and lidocaine groups requested postoperative analgesia in the immediate postoperative period (87.0% and 90.9%, respectively), in comparison to the bupivacaine or lidocaine plus bupivacaine groups (41.2% and 74.1%). No significant difference was noted in the consumption of acetaminophen (325-mg tablets) over the first 24 h after surgery or from 24 to 48 h after surgery. There was a nonsignificant trend for subjects in the placebo and lidocaine preoperative groups to self-administer more codeine tablets for unrelieved pain (Table 3).
The experimental design of this study permitted differentiation between the effects of intraoperative nociceptive input and postoperative inflammatory pain on the development of sensitization. The blockade of the intraoperative afferent barrage by preoperative lidocaine did not result in a detectable effect on pain at 24 and 48 hours, suggesting that the intensity and duration of nociceptive input during oral surgery is insufficient to produce central sensitization manifesting as increased pain at later time points. There was no significant difference between groups in the consumption of analgesics on Days 1 and 2, indicating that differences in pain reporting at 24 and 48 hours were not confounded by analgesic intake that might attenuate pain or the inflammatory process. In contrast to previous reports (11–13), preoperative blockade of intraoperative nociceptive input alone did not have an effect on pain at 48 hours in this model, suggesting that the relatively brief duration of nociceptive input during oral surgery is a less important stimulus than the more prolonged postoperative pain attributed to inflammation in this model. Other studies using a long-duration anesthetic have failed to take into account the carryover of preoperative interventions into the postoperative period, thereby also blocking postoperative pain input and contributing to the development of sensitization (13,20).
The increase in intraoperative plasma concentrations of β-endorphin in the two groups receiving placebo local anesthetic injections before surgery suggests activation of an intraoperative nociceptive barrage sufficient to result in descending hypothalamic-pituitary secretion. The observed increase in plasma β-endorphin concentration is similar to changes seen in awake subjects undergoing surgical stress or subjected to moderate to severe postoperative pain (21,22). Local anesthetic blockade of postoperative inflammatory pain input significantly attenuated the nociceptive barrage and β-endorphin release. The findings suggest that the maintenance of central sensitization leading to persistent pain and hyperalgesia is dependent on input from damaged peripheral tissue (23), which is characteristic of the postoperative period. In addition, this maintained input occurring after surgery may be a major contributor to sensitization, leading to increased pain at later time points in the oral surgery model.
The postoperative analgesic effects of presurgical interventions are presumed to depend on their ability to attenuate the central sensitization associated with tissue injury (24). Clinical studies comparing preemptive treatments versus no treatment are overwhelmingly supportive of a beneficial effect in the pretreated patients (11,13,25–28) across a wide variety of clinical models and types of surgery. However, studies comparing preemptive versus postsurgical treatment with regional anesthesia have produced conflicting results that suggest limited (29) or no (30) advantage of presurgical over postsurgical treatment. A possible explanation for these discrepant findings is that the development of sensitization may depend more heavily on the peripheral neural barrage that develops during the postoperative period than on surgical trauma (24,31). The relative roles of surgical trauma and postoperative inflammation on the establishment of central sensitization and hyperalgesia may depend on the site of origin of the surgery and its duration. For example, limb and breast surgery, but not abdominal surgery, are responsive to presurgical epidural morphine (32). This study used a short-duration surgical model that produced a short neural barrage during surgery but prolonged postoperative pain because of the progression of inflammation sufficient to initiate and maintain central sensitization. In addition to duration of pain, the character and intensity of the pain probably also influence the effectiveness of the preemptive approach, because neuropathic pain is not influenced by regional anesthetic block before neuronal injury (20,33). Similarly, a preemptive effect of intrathecal lidocaine in rats administered hind paw injections of 2.5% formalin was overcome by the administration of 3.75% and 5.0% formalin (24), indicating that attenuation of the development of central sensitization is dependent on the magnitude of the nociceptive input.
The importance of postsurgical blockade on the prevention of sensitization leading to increased pain at later time points is illustrated by the blockade of both primary and secondary hyperalgesia from carrageenan in rats administered a prolonged (12- to 16-hour) nerve block with tonicaine (34). The administration of the same anesthetic 5 hours after carrageenan also prevented the development of late hyperalgesia (≥24 hours). However, short-term nerve block with lidocaine produced no significant changes in carrageenan-induced hyperalgesia. These observations suggest that nerve blockade should last until noxious input from the inflamed tissues decreases below the level that can maintain central sensitization (24,31,34).
Our study suggests that the management of postoperative pain after surgery can be optimized not only by administering long-acting local anesthetics to block pain during the postoperative period, but also by combining the local anesthetic with analgesics to attenuate the development of inflammation over the first few days after surgery. Previous studies in the oral surgery model have demonstrated that suppression of pain over the first four to eight hours after surgery by a long-acting local anesthetic (35) is additive with the effects of nonsteroidal antiinflammatory drugs (NSAIDs) (36). Although not directly tested in these studies, the decrease in pain and inflammation over the same time course as in this study suggests that the administration of NSAIDs to suppress postoperative pain also decreases pain at later time points by suppressing the neural barrage, leading to central sensitization.
The continuing controversy over the efficacy of preemptive analgesia (15,37,38) is based on semantic concerns over the use of the term “preemptive,” but it reflects increasing recognition that intervention should be applied before the nociceptive barrage (whether during or after surgery) and provide effective suppression of nociceptive input from the damaged tissues over the time course that normally contributes to the development of sensitization (31). It does not appear to be important whether the intervention is applied before or after incision to be considered preemptive, but rather that postoperative hyperalgesia is attenuated at later time points by limiting the development of central sensitization. This study suggests that postoperative pain contributes to a greater extent than intraoperative nociceptive barrage, at least in the oral surgery model. Given the inflammatory nature of postoperative pain, the administration of antiinflammatory drugs in combination with a long-acting local anesthetic should be additive (36,39) and result in clinically meaningful preemptive analgesia.
The authors gratefully acknowledge the contribution of the National Institutes of Health (NIH) Pharmaceutical Development Service, the NIH Clinical Center Department of Nursing, and the NIH Department of Anesthesia Services in the conduct of this study.
1 Coghill RC, Gracely RH. Validation of the combined numerical/verbal descriptor scale for pain [abstract]. Am Pain Soc Abstr 1996;15:A86.
1. Woolf CJ. Evidence for a central component of post-injury pain hypersensitivity. Nature 1983; 306: 686–8.
2. Woolf CJ, Wall PD. Morphine-sensitive and morphine-insensitive actions of C-fiber input on the rat spinal cord. Neurosci Lett 1986; 64: 221–5.
3. Cooke AJ, Woolf CJ, Wall PD, McMahon SR. Dynamic receptive field plasticity in rat spinal cord dorsal horn following C-fiber primary afferent input. Nature 1987; 325: 151–3.
4. Hylden JLK, Nahin RL, Traub RJ, Dubner R. Expansion of receptive fields of spinal lamina I projection neurons in rats with unilateral adjuvant-induced inflammation: the contribution of dorsal horn mechanisms. Pain 1989; 37: 229–43.
5. Woolf CJ, Thompson SWM. The induction and maintenance of central sensitization is dependent on N-methyl-D-aspartic acid receptor activation: implications for the treatment of post-injury pain hypersensitivity states. Pain 1991; 44: 293–9.
6. Ren K, Hylden JLK, Williams GM, et al. The effects of a non-competitive NMDA receptor antagonist, MK-801, on behavioral hyperalgesia and dorsal horn neuronal activity in rats with unilateral inflammation. Pain 1992; 50: 331–44.
7. Urban L, Randle M. Slow excitatory transmission in rat dorsal horn: possible mediation by peptides. Brain Res 1984; 290: 336–41.
8. Dubner R. Neuronal plasticity and pain following peripheral tissue inflammation or nerve injury. In: Bond MR, Charlton JE, Woolf CJ, eds. Proceedings of the VIth World Congress on Pain. New York: Elsevier, 1991: 263–76.
9. Dickenson AH, Sullivan AF. Subcutaneous formalin-induced activity of dorsal horn neurons in the rat: differential response to an intrathecal opiate administration pre- or post-formalin. Pain 1987; 30: 349–60.
10. Coderre TJ, Vaccarino AL, Melzack R. Central nervous system plasticity in tonic pain response to subcutaneous formalin injection. Brain Res 1990; 535: 155–8.
11. Tverskoy M, Cozacov C, Ayache M, et al. Postoperative pain after inguinal herniorrhaphy with different types of anesthesia. Anesth Analg 1990; 70: 29–35.
12. Ejlersen E, Andersen HB, Eliasen K, Mogensen T. A comparison between preincisional and postincisional lidocaine infiltration and postoperative pain. Anesth Analg 1992; 74: 495–8.
13. Gordon SM, Dionne RA, Brahim J, et al. Blockade of peripheral neuronal barrage reduces postoperative pain. Pain 1997; 70: 209–15.
14. Katz J. Pre-emptive analgesia: evidence, current status and future directions. Eur J Anaesthesiol 1995; 12: 8–13.
15. Kissin I. Preemptive analgesia: how can we make it work? In: Devor M, Rowbotham MC, Wiesenfeld-Hallin Z, eds. Proceedings of the 9th World Congress on Pain. Seattle: IASP Press, 2000: 973–85.
16. Yagiela JA. Local anesthetics. In: Dionne RA, Phero JC, eds. Management of pain and anxiety in dental practice. New York: Elsevier, 1991: 126.
17. Gracely RH, Dubner R. Reliability and validity of verbal descriptor scales of painfulness. Pain 1987; 29: 175–85.
18. Rosier EM, Iadarola MI, Coghill RC. Reproducibility of pain measurement and pain perception. Pain 2002; 98: 205–16.
19. Dionne RA, Yagiela JM, Moore PA, et al. Comparing efficacy and safety of four intravenous sedation regimens in dental outpatients. J Am Dent Assoc 2001; 132: 740–51.
20. Hu JW, Woda A, Sessle BJ. Effects of pre-emptive local anaesthesia on tooth pulp deafferentation-induced neuroplastic changes in cat trigeminal brainstem neurones. Arch Oral Biol 1999; 44: 287–93.
21. Hargreaves KM, Dionne RA, Mueller GP, et al. Naloxone, fentanyl and diazepam modify beta-endorphin levels during surgery. Clin Pharmacol Ther 1986; 40: 165–71.
22. Dionne RA, McCullough L. Analgesic efficacy of the S(+) isomer of ibuprofen in comparison to racemic ibuprofen. Clin Pharmacol Ther 1998; 63: 694–701.
23. Coderre TJ, Katz J. Peripheral and central hyperexcitability: differential signs and symptoms in persistent pain. Behav Brain Sci 1997; 20: 404–19.
24. Yashpal K, Katz J, Coderre TJ. Effects of preemptive or postinjury intrathecal local anesthesia on persistent nociceptive responses in rats. Anesthesiology 1996; 84: 1119–28.
25. Fletcher D, Kayser V, Guilbaud G. Influence of timing of administration on the analgesic effect of bupivacaine infiltration in carrageenin-injected rats. Anesthesiology 1996; 84: 1129–37.
26. Hannibal K, Galatius H, Hansen A, et al. Preoperative wound infiltration with bupivacaine reduces early and late opioid requirement after hysterectomy. Anesth Analg 1996; 83: 376–81.
27. Pedersen JL, Crawford ME, Dahl JB, et al. Effect of preemptive nerve block on inflammation and hyperalgesia after human thermal injury. Anesthesiology 1996; 84: 1020–6.
28. Salman MA, Yucebas ME, Coskun F, Aypar U. Day-case laparoscopy: a comparison of prophylactic opioid, NSAID or local anesthesia for postoperative analgesia. Acta Anaesthesiol Scand 2000; 44: 536–42.
29. Holthusen H, Irsfeld S, Lipfert P. Effect of pre- or post-traumatically applied i.v. lidocaine on primary and secondary hyperalgesia after experimental heat trauma in humans. Pain 2000; 88: 295–302.
30. Campbell WI, Kendrick RW, Ramsay-Baggs P, McCaughey W. The effect of pre-operative administration of bupivacaine compared with its postoperative use. Anaesthesia 1997; 52: 1212–29.
31. Kissin I. Preemptive analgesia. Anesthesiology 2000; 93: 1138–43.
32. Aida S, Baba H, Yamakura T, et al. The effectiveness of preemptive analgesia varies according to the type of surgery: a randomized double-blind study. Anesth Analg 1999; 89: 711–6.
33. Abdi S, Lee DH, Park SK, Chung JM. Lack of pre-emptive analgesic effects of local anaesthetics on neuropathic pain. Br J Anaesth 2000; 85: 620–3.
34. Kelly DJ, Ahmad M, Brull SJ. Preemptive analgesia. I. Physiological pathways and pharmacological modalities. Can J Anaesth 2001; 48: 1000–10.
35. Sisk AL, Dionne RA, Wirdzek PR. Evaluation of etidocaine hydrochloride for local anesthesia and postoperative pain control in oral surgery. J Oral Maxillofac Surg 1984; 42: 84–8.
36. Dionne RA. Suppression of dental pain by the preoperative administration of flurbiprofen. Am J Med 1986; 80 (Suppl 3A): 41–9.
37. McQuay HJ. Do preemptive treatments provide better pain control? In: Gebhart GF, Hammond DL, Jensen TS, eds. Proceedings of the 7th World Congress on Pain. Seattle: IASP Press, 1994: 709–23.
38. Kehlet H. Controlling acute pain: role of preemptive analgesia, peripheral treatment, and balanced analgesia, and effects on outcome. In: Max M, ed. Pain 1999: an updated review. Seattle: IASP Press, 1999: 459–62.
© 2002 International Anesthesia Research Society
39. Reuben SS, Bhopatkar S, Maciolek H, et al. The preemptive analgesic effect of rofecoxib after ambulatory arthroscopic knee surgery. Anesth Analg 2002; 94: 55–9.