| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Department of Anesthesiology and General Intensive Care Medicine (B); Core Unit for Medical Statistics and Informatics; and Bernhard Gottlieb Department of Oral Surgery, Medical University of Vienna, Austria.
Address correspondence and reprint requests to Andrea Michalek-Sauberer, MD, Department of Anesthesiology and General Intensive Care Medicine (B), Medical University of Vienna, Vienna General Hospital, Waehringer Guertel 18-20, A-1090 Vienna, Austria. Address e-mail to andrea.michalek-sauberer{at}meduniwien.ac.at.
 |
Abstract |
|---|
 Top Abstract IntroductionMETHODSRESULTSDISCUSSIONAPPENDIX: PATIENT DIARY-...REFERENCES |
|---|
METHODS: Patients received either AA with electrical stimulation (AE, n = 76) or without (AA, n = 37) electrical stimulation at an alternating frequency of 2/100 Hz or a sham AE with metal plates instead of needles and no electrical stimulation, no-needle (NN, n = 36) at the AA points 1 (tooth), 55 (Shen men) and 84 (mouth) during the entire study period. Regularly rated pain intensity (five-point verbal rating scale), consumption of acetaminophen 500 mg tablets and additional rescue medication with mefenamic acid 500 mg were assessed.
RESULTS: The median fraction of time when pain was rated as moderate or worse (upper and lower quartile): AE: 33% (12%, 64%), AA: 22% (6%, 56%), NN: 30% (7%, 53%) did not differ significantly among the treatment groups. There were no significant differences in mean number of acetaminophen 500 mg tablets (range): AE: 5.2 (0–12), AA: 4.6 (0–11), NN: 5.4 (0–10) or percentage of patients requiring additional mefenamic acid: AE: 19%, AA: 18%, NN: 19%.
CONCLUSION: We conclude that neither AE nor AA alone reduce either pain intensity or analgesic consumption in a molar tooth extraction model of acute pain.
|
Introduction |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONAPPENDIX: PATIENT DIARY-...REFERENCES |
|---|
Auricular acupuncture (AA) is a distinct form of acupuncture based on a somatotopic relation of the external ear to other body regions, which was demonstrated by Oleson et al. (6). Electrical stimulation of acupuncture points produces analgesic effects which are mediated by the release of different neuropeptides, depending on the stimulation frequency (7). Auricular electrostimulation had an anesthetic-sparing effect after acute noxious stimulation (8) and enhanced the effects of conventional AA in the treatment of chronic musculoskeletal pain (9,10). The aim of the present study was to test whether auricular electroacupuncture (AE) would also be more effective than AA for the relief of pain and reduce analgesic consumption after mandibular third molar surgery.
|
METHODS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONAPPENDIX: PATIENT DIARY-...REFERENCES |
|---|
After approval by the Ethics Committee at the University of Vienna, 149 patients (ASA I–II, aged 18–35 yr), scheduled for elective extraction of one impacted mandibular third molar, were enrolled after written informed consent between October 2002 and June 2004. Exclusion criteria were previous experience with acupuncture, language difficulties, history of drug abuse, chronic pain therapy, lesions at the external ear, immunosuppression, the presence of a pacemaker, asthma, coagulation disorders, neurological or psychiatric disorders, pregnancy, and allergy to acetaminophen or mefenamic acid.
Patients were randomized to treatment with AE (n = 76), AA (n = 37), or sham AE, no-needle (NN) (n = 36). The randomization procedure was random permuted blocks within strata (block size four). The strata were formed by the participating surgeons.
Each patient received a disposable P-StimTM AE device (Biegler Medizinelektronik GmbH, Mauerbach, Austria), which costs approximately 63 US$. The P-Stim consists of a battery-powered stimulator worn behind the ear, which was connected to sterile, single-use permanent AA needles (titanium, 27 gauge, length 3 mm) or to small metal plates (NN). Each P-Stim was programmed by an independent technician for electrical stimulation or no stimulation before the study and was packed together with the needles or metal plates in a nontransparent case, numbered consecutively according to a computer-generated randomization list.
The P-Stim was applied approximately 30 min before the start of the surgical procedure by the same investigator (Diploma of Acupuncture, Austrian Medical Chamber; 5 yr of clinical experience) in all patients. AA points 1 (tooth), 55 (Shen Men) and 84 (mouth) (Fig. 1) were identified by means of a point-finding device (multipoint selection pen, Biegler Medizinelektronik, Mauerbach, Austria) at the ear ipsilateral to the site of surgery. AA points 1 and 84 are specific for the dental region and the oral mucosa respectively, whereas point 55 is nonspecific, with strong analgesic properties (12). After disinfection of the skin, the acupuncture needles were inserted into the auricular acupoints of the patients in the AE and AA groups, secured with adhesive tape and connected to the P-Stim. In patients of group NN, small metal plates were fixed to the auricular acupoints with adhesive tape and connected to the P-Stim. In the AE-group, electrical stimulation consisted of constant current pulses of 2 and 100 Hz every 3 s (3 h of stimulation followed by a pause of 3 h). Patients in groups AA and NN had no electrical stimulation. Patients were told that with AE they may or may not feel the electrical stimulation. The P-Stim was worn for 48 h and thereafter removed by the patients.
|
All procedures were scheduled between 8 am and 1 pm, as circadian rhythms influence pain and analgesic consumption after oral surgery (13). The surgical procedure was performed under local anesthesia. Articaine 4% with epinephrine 0.01 mg/mL (Ultracain dental forte®, Aventis Pharma, Frankfurt/Main, Germany) was used for an inferior alveolar nerve block and gingival infiltration to achieve adequate analgesia. Gingival incision was followed by a mucoperiostal flap dissection and bone removal using a drill cooled with physiological saline. Surgeons rated the tissue trauma as minor, moderate or extensive. No sedatives were used perioperatively.
Patients were asked to keep a diary, which was returned at a surgical follow-up visit after conclusion of the study. Patients rated postoperative pain hourly until 10 pm on the day of the operation. Thereafter maximum pain intensity was rated in 4-h intervals (except for an 8-h interval between 10 pm and 6 am) for the rest of the study period using a 5-point verbal rating scale (0 = no pain, 1 = minimal pain, 2 = moderate pain, 3 = severe pain, 4 = worst pain). Each patient was provided with 12 tablets of acetaminophen 500 mg (Mexalen®, Goedecke, Freiburg, Germany). Patients were instructed to take one tablet every time the pain self-rating equaled or exceeded 2. The maximum daily allowance was 3 g acetaminophen. Mefenamic acid 500 mg (Parkemed®, Merckle, Blaubeuren, Germany) was permitted as rescue medication (maximum 1500 mg per day) in case of insufficient pain relief after two tablets of acetaminophen. Time and amount of analgesic intake were recorded in the diary. Patients noted the duration of local anesthetic effect, which was defined as numbness of the tongue and lower lip, and any problems that occurred when using the P-Stim and when the P-Stim was removed. The diary included questions concerning blinding success, the patient’s evaluation of AE, and willingness to use a P-Stim again (Appendix).
Statistical Analysis
The primary outcome variable was the number of acetaminophen tablets consumed during the first 48 h after surgery. Mean analgesic consumption with placebo treatment was estimated to be four tablets and a Poisson distribution was assumed. A decrease of at least 50% in tablet consumption with AE was considered clinically relevant. A sample size of n = 88 was computed (AE:AA:NN = 44:22:22) for a power of 90% and a two-sided test. A Bonferroni-adjusted significance level of 0.05/3 = 0.0167 was used to account for multiple testing of the intended pairwise group comparisons. An assumed dropout rate of 25% increased the sample size to n = 120. Finally, the inclusion of up to n = 150 patients was projected on the basis of a possible lower actual mean analgesic consumption, the possibility that analgesic consumption would follow a zero-inflated Poisson-distribution, and an improved evaluation of secondary outcome variables. The study was analyzed according to the intention-to-treat principle and the missing-at-random-assumption. A generalized linear mixed regression model with Poisson-distributed outcome was used for comparison of daily tablet intake among the groups (using PROC GLIMMIX of SAS). This model directly allowed both the assessment of potential confounding factor effects and a more powerful adjustment for multiple comparisons than the Bonferroni adjustment.
Secondary outcome variables were analyzed in an exploratory manner by applying appropriate statistical tests. The programs DSTPLAN (Version 4.2, April 2000, The University of Texas, MD Anderson Cancer Center, Houston, TX) and SAS (Version 9.1, SAS Institute, Cary, NC) were used for sample size calculations and statistical analyses.
|
RESULTS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONAPPENDIX: PATIENT DIARY-...REFERENCES |
|---|
|
|
Analgesic consumption, acetaminophen as well as mefenamic acid, did not differ among the groups. Twenty-four patients used mefenamic acid as additional analgesic medication (Table 2). Varying the statistical approach from intention-to-treat analysis to analysis per protocol (i.e., evaluating only those patients who fully adhered to the protocol) did not affect the results.
|
The duration of pain requiring analgesic treatment (i.e., 
2 on the verbal rating scale) did not differ among the groups (Table 3). Time to first perception of postoperative pain and first analgesic intake did not differ among the groups. Contrary to the protocol, some patients in every group did take analgesic medication before they felt pain. Numbness of lip and tongue persisted longer than the analgesic effect of local anesthesia in all groups.
|
Most patients believed that they had been treated with AE, even in the AA and NN groups. Thirty-eight percentage of patients reported side effects of AA, mainly tiredness (16%) and pain at the ear (10%) (Table 4). Forty-four percentage of patients believed that the P-Stim effectively reduced postoperative pain, 15% reported no pain reduction. Two patients in the AE group and one in the NN group reported increased pain after removal of the P-Stim. The comfort of wearing the P-Stim was rated good to moderate by 92% of patients. Twenty-three percentage of the patients would not use an AA device again, mostly because of inadequate comfort (75%) or limited analgesic effects (36%). Seventy-five percentage of patients would use AE for future surgical procedures even in the AA and NN groups.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONAPPENDIX: PATIENT DIARY-...REFERENCES |
|---|
In volunteers, AA or AE at nonspecific auricular acupoints reduced anesthetic requirement significantly (8,14), but to an extent that would not be useful in clinical practice. In our study, specific auricular acupuncture points and an auxiliary analgesic acupoint were chosen according to recent recommendations (4). Although acupuncture is usually customized to the individual patient, a fixed combination of auricular acupoints was used in our patient population as the nociceptive field after 3rd molar extraction is well defined. Using an analog regimen, a relevant reduction of postoperative analgesic requirement was found after hip (15) and knee (16) surgery. Different specific auricular acupoints were used because pain after these procedures projects to other regions of the auricle. Whether auricular acupoints, which represent different body regions, differ in their analgesic properties as well has not been investigated. If so, a stronger analgesic effect of auricular acupoints representing the musculoskeletal system would explain the significant analgesic efficacy of AA after orthopedic surgery.
The nature of pain varies after different types of surgery and the efficacy of acetaminophen and NSAIDs for pain relief is different after dental and orthopedic surgery (17,18). The underlying pathophysiological mechanisms are not yet understood. Whether similar mechanisms could explain the differing analgesic effects of AA after orthopedic and dental surgery awaits further investigation.
The optimal timing of AA for postoperative pain relief has not been investigated. In our study, the P-Stim was applied 30 min before the start of the surgical procedure because peak analgesic effects of electroacupuncture are mediated by humoral mechanisms (7) and occur 20–40 min after needle insertion (19). In animal experiments, tolerance to the analgesic effects of electroacupuncture develops after repeated or prolonged electrostimulation (19). In our study, as AE was not effective initially and decreased in efficiency later on, it is unlikely that tolerance occurred.
A stimulation pattern of alternating frequencies between 2 and 100 Hz produces maximum analgesic effects in acute (20) and chronic pain conditions, presumably through the simultaneous release of four different endogenous opioid peptides (7). However, AE with a constant frequency of 1 Hz was effective in chronic musculoskeletal pain (9,10). It remains unclear why the alternating stimulation pattern in our study, instead of being potentially more effective than a constant low frequency, failed to reduce postoperative pain.
A credible placebo treatment is difficult to create in acupuncture trials (21,22). Inserting needles into sham points, which do not correspond to defined acupoints, provides analgesia superior to placebo (21,22). We choose a NN placebo treatment to avoid the nonspecific effects of needle placement. This was sufficiently credible for the patients in the NN-group, as evidenced by their strong belief of having received AE. Although the acupuncturist knew whether the patient received a NN treatment, potential bias was minimized by the choice of the primary end-point and statistical analysis by an independent investigator.
A reduction in postoperative analgesic requirement was chosen as the primary outcome parameter (23). Postoperative analgesic medication was mandatory, as it was assumed that patients would require analgesics regardless of their treatment group. With pain ratings as the primary outcome parameter, it would not have been possible to separate the analgesic effects of AE from those of the analgesic medication. Therefore, the study was designed to detect a clinically relevant 50% reduction of postoperative analgesic consumption. A post hoc calculation revealed that even a reduction of at least 34% in the AE group compared to the NN group would have been detectable with 90% power. Acetaminophen is commonly used and is less effective than NSAIDs in relieving pain after dental surgery (17). Even though the protocol allowed an additional rescue medication, this did not affect our results, as the mefenamic acid requirement was similar among the groups.
A five-point verbal rating scale was used for the assessment of postoperative pain (24). Its lower sensitivity compared with a visual analog scale (25) must be acknowledged as a certain weakness of our study design. However, when interpreting the results of our study, both primary and secondary outcome parameters that were used to assess the potential analgesic effect of AA have to be considered together. No trend toward a benefit of AE was found in any of these parameters. It is therefore unlikely that the lack of a positive result in our study was due to the insensitivity of the outcome parameters used.
Missing data and patients’ nonadherence to the protocol were evenly distributed among the groups and thus most probably did not affect the results. There were more female patients and more smokers in the AE group. However, after statistical adjustment for gender and smoking analgesic consumption still did not differ significantly among the treatment groups.
In conclusion, we found no effect of 2/100 Hz AE at the auricular acupoints 1 (tooth), 55 (Shen Men) and 84 (mouth) or AA alone on pain and analgesic consumption after extraction of one impacted mandibular third molar. Patients’ acceptance of nonpharmacological techniques for postoperative pain control was nevertheless high.
|
ACKNOWLEDGMENTS |
|---|
|
APPENDIX: PATIENT DIARY—QUESTIONS REGARDING THE PATIENT’S EVALUATION OF AURICULAR ELECTROACUPUNCTURE |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONAPPENDIX: PATIENT DIARY-...REFERENCES |
|---|
Do you believe that auricular acupuncture will influence the pain after extraction of your wisdom tooth?
yes/no/don’t know
If yes, which effect do you expect?
reduction of pain/pain increase
Patient’s Evaluation of Auricular Acupuncture After the Conclusion of the Study
Which device do you think was used in your treatment?
electrical stimulation/needles only/placebo/don’t know
Why do you suppose so?
because of the talk with or the conduct of the acupuncturist/because the effects of auricular acupuncture (e.g., change in pain intensity)/because of the sensations during the application or while wearing the auricular electroacupuncture device
Do you think that auricular acupuncture did influence the pain after the extraction of your wisdom tooth?
yes/no/don’t know
If yes, which effect did you experience?
reduction of pain/increased pain
Side effects or problems of acupuncture (more than one answer possible):
tiredness/dizziness/pain/inflammation/other (specify)
Comfort of wearing the auricular acupuncture device:
good/moderate/bad (specify reasons)
Would you use an auricular acupuncture device again?
yes/no
If no, specify reasons (more than one answer possible):
problems with comfort/effect/other (specify)
|
Footnotes |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONAPPENDIX: PATIENT DIARY-...REFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Sun, T. J. Gan, J. W. Dubose, and A. S. Habib Acupuncture and related techniques for postoperative pain: a systematic review of randomized controlled trials Br. J. Anaesth., June 2, 2008; (2008) aen146v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
