The efficacy of acupuncture in smoking cessation, and its effect on the urge to smoke are unclear. We evaluated the effect of a standardized protocol of transcutaneous electric acupoint stimulations (TEAS) on alleviating the urge to smoke. Ninety-eight smokers were recruited in two double-blind studies. Participants abstained from smoking for 26 h, and were randomized to receive TEAS alternating between 2 and 100 Hz at four acupoints (LI4 and PC8, PC6 and TE5) at four different intensities (10, 5, Intermittent 5 or 0 mA). The urge to smoke was assessed by the Questionnaire of Smoking Urges (QSU-Brief). In Experiment 1, the 10 mA group (
Several types of interventions, such as counseling [
Acupuncture can be applied in different ways. Manual needling is the classic method, and electro-acupuncture (EA), that is, electrical stimulation is introduced to the body via metallic needles inserted into the acupoints is the most commonly used modality today [
A thorough study has been made to compare the effectiveness of EA and TEAS in producing antinociceptive effect in rats [
While the effect of acupuncture has been the subject of increasing amount of research, the central activation pattern induced by acupuncture (increased production and release of morphine-like substances such as enkephalins, endorphins, dynorphins and other neurotransmitters) has only recently been clarified in animal model [
Acute cravings for cigarettes (i.e., the urge to smoke) have repeatedly been associated with relapses, and are identified as having a critical role in promoting smoking relapse [
The aim of the present experiments was to evaluate the efficacy of a standardized protocol of TEAS in alleviating the urge to smoke in nicotine-dependent individuals, during a 26-h abstinence period. Two exploratory studies were performed at Singapore's National University Hospital (NUH). The first (Experiment 1) was a pilot study to investigate the efficacy of TEAS as a method for relieving acute cravings in a controlled situation, where smokers were not allowed to smoke [
Two studies were performed consecutively. Both were single-center, randomized and double-blind pilot studies evaluating smoking urges in smokers with a Fagerström score higher than 4 (Fagerström Test for Nicotine Dependence, FTND [
In the first study (Experiment 1), a total of 40 participants (38 males and 2 females) were recruited through the NUH clinical trial unit database (
For the second study (Experiment 2), a total of 58 participants (49 males and 9 females) were recruited through street interviews and randomized (via
Subjects in both groups received four TEAS sessions: one session was performed on Day 1 (18:00 h in the non-abstinent state and three sessions on Day 2 (08:30, 12:30 and 16:30 h) while abstaining from smoking.
Participants were prescreened, to ensure that they had an FTND score equal to or >4, having smoked at least 15 cigarettes a day 12 months prior to enrollment, and had no intention to quit during the study. Participants were not included if they had any earlier experience with transcutaneous electric nerve stimulation (TENS), TEAS, EA or manual acupuncture or if they used any nicotine replacement therapies or “stop smoking” therapies during the 3-month period prior to enrollment in the present study. Subjects with pacemakers, a history of chronic medical conditions, any acute illness or reported alcohol/drug abuse were also not included. Participants were required to sign a written informed consent to participate and to stop smoking from the Friday evening 18:00 h until Saturday evening 20:00 h, for exactly 26 h. Participants were paid SGD100 for their participation in each experiment. Both study protocols were approved by the NUH Ethics Committee and implemented in the NUH Clinical Trial Unit (CTU).
The TEAS device (HANS LH-202, Huawei Co. Beijing, China) is a dual-channel acupoint nerve stimulator with two pairs of constant current electric output. The frequency of the output stimulation is an alternating dense-and-disperse mode, where the 2 Hz (0.6-ms pulse width) stimulation is alternated with 100 Hz stimulation (0.2-ms pulse width) automatically, each lasting for 3 s. The device is connected to the subject via four carbon-pad adhesive electrodes at four acupoints. They are Hegu (LI4, located at the mid-point of the second metacarpus on the radial side), and Laogong (PC8, mid-way between the second and third metacarpus on the palmal side, where one's middle finger falls when one makes a fist) in the same hand to form an electric circuit. The other two points are: Nei Guan (PC6, 2
In the first study (Experiment 1), a specially designed control-TEAS device was used for the control group, which capped the output intensity to 5 mA, even if the investigator selected the protocol-required intensity of 10 mA.
In the second study (Experiment 2), the same control-TEAS device (capped at 5 mA) was used but was combined with a specific-switching device programmed to switch the stimulation on for 3 min and off for 7 min (the i5 mA group), or totally blocking the stimulation from reaching the subject (the 0 mA group). A flashing signal indicated that the treatment was in progress; this was added to enhance the blinding further.
In both the studies, the hands of the subject were placed in an opaque box when receiving stimulation, so that the participants and the investigators could not observe the amplitude of the finger movements and guess the relative intensity he/she was receiving.
Moreover, to ascertain the validity of the double-blind, all the participants were asked to complete a guess test at the end of each session which comprised of the following questions:
(1) “Do you believe that your TEAS unit was functioning properly”? The answers ranged from “I am certain it was working properly” to “I am certain it was not working”, and scored from 0 to 4.
(2) “If your answer is that you had absolutely no idea whether the unit was functioning or not, what would you guess”? The answer could be either “Functioning properly” or “Not functioning properly”.
At the end of each session, the investigators also completed an observer guess test containing the same aforementioned two questions.
The primary outcome measure was the total score of smoking urges as assessed by the 10-item Questionnaire of Smoking Urges-Brief (QSU-Brief) [
Systolic blood pressure (SBP), diastolic blood pressure (DBP) and pulse rate (PR) were recorded by using GE Marquette Dash 3000 Bedside Patient Monitoring System. Body temperature (°C), (measured orally using the Terumo Digital Clinical Thermometer, Model C0402), and respiration rate (breaths per minute, BPM) were recorded each time just before the participants completed the QSU questionnaire.
Carbon monoxide (CO) levels in the expired air using a CO monitor (Minismokerlyzer, Bedfont Scientific Limited, UK) were measured before the participants stopped smoking, and during abstinence at 14, 18, 22, 24 and 26 h after the last cigarette was smoked. Blood samples for plasma cotinine determinations were drawn at baseline and at 14 and 22 h after smoking was stopped. Plasma cotinine concentrations were determined by chemiluminescence using Immulite 2000 (Diagnostic Products Corporation, USA). The reference limit is <25 ng ml−1 for non-smokers. The assay sensitivity is 5 ng ml−1 with a within-run precision of 4.0–9.9%.
The screening/enrollment was performed for not >10 days before the participants started the study. The study was initiated on a Friday evening at 18:00 h, and lasted until Saturday evening 20:00 h.
For both the studies, 18–20 participants were assessed during each session. All participants were instructed to smoke their last cigarettes upon their arrival at the study center, and then not to smoke again until the end of the study. Participants were allowed to go back home for the night.
The categorical variables of demographics and clinical characteristics were summarized by frequency, whereas the continuous variables were summarized by arithmetic mean and standard deviation (SD). The frequency of categorical parameters between the groups was compared with Pearson's chi-squared test, and Fisher's exact test was used if the expected count by cell was less than five. Continuous variables between the two independent groups were compared with the Mann-Whitney non-parametric test.
All the responses from the 10-question QSU-Brief were summed up as a single score (minimum score, 10; maximum score, 70) for each participant at each predefined time points, and was compared between treatment arms as a function of time. On separate factor analysis of QSU scores, according to Cox et al. [
The variations in the mean QSU scores across time points by each treatment arm were plotted. Exploratory graphical presentation suggested that a growth curve quadratic model might be used to fit the observed QSU scores as a function of time. Each interaction term was tested and the covariate,
All analyses were performed on SAS (Version 9.1, SAS Institute Inc., Cary, NC, USA). A two-sided
The relevant data in Experiments 1 and 2 are depicted in Table
Demographic and smoking characteristics of subjects participating in Experiment 1 and 2.
Experiment 1 | Experiment 2 | ||||
---|---|---|---|---|---|
Treatment condition | 5 mA | 10 mA | 0 mA | i5 mA | 10 mA |
( | ( | ( | ( | ( | |
Gender ( | |||||
Female | 0 (0) | 2 (10) | 2 (12) | 7 (35) | 6 (29) |
Male | 20 (100) | 18 (90) | 15 (88) | 13 (65) | 15 (71) |
Age (years), Mean ± SD | 30.8 ± 6.47 | 30.5 ± 8.78 | 25.9 ± 4.75 | 23.5 ± 3.58 | 24.7 ± 4.85 |
Number of cigarettes per day Median (range) | 20 (15–35) | 18.5 (15–28) | 19 (15–28) | 18.3 (15–25) | 17.0 (15–23) |
Years of smoking Median (range) | 14 (5–31) | 13.5 (6–38) | 9.0 (1–23) | 7.5 (4–13) | 8.0 (5–28) |
FTND Median (range) | 5.5 (4–8) | 5.5 (4–8) | 7.0 (4–8) | 7.0 (4–9) | 6.0 (4–8) |
Three participants in the 5 mA arm, one participant in the 10 mA arm of Experiment 1 decided to stop the study. Two participants from each arm of Experiment 2 did not return to the center overnight after the second QSU, and one participant of the 0 mA group did not continue the treatment after the fourth QSU assessment. A total of nine minor adverse events were reported by the participants. In the 10 mA arm, one patient reported a runny nose and coughing 14 h after the last cigarette, and one patient experienced giddiness after blood drawn at time 0, and numbness in a finger after 2 h, and this subject decided not to complete the study. In the i5 mA group, three participants reported a runny nose and one reported insomnia. In the 0 mA group, one participant reported a runny nose after 14 h and did not complete the study, one reported giddiness, and one reported numbness in the left hand.
Analysis performed in Experiments 1 and 2 showed that the participants were unable to distinguish which stimulations they received (Fisher's exact test:
The average baseline QSU scores were 42.6 ± 15.20 and 36.8 ± 14.96 (mean ± SD) in the 10 and 5 mA group, respectively, showing no significant difference. The fitted quadratic model showed that there was no significant condition by time interaction (
Mean QSU changes from baseline in Experiments 1 (a) and 2 (b). Arrows indicate TEAS. Error bars indicate SEM.
The average QSU scores for the baseline of 10, i5 and 0 mA groups were 43.62 ± 17.37, 42.5 ± 14.64 and 39.29 ± 16.44, respectively, showing no statistically significant difference. There was no main effect of condition, and the fitted quadratic model showed that there was no significant condition by time interaction (
In the exploratory analyses of Experiment 2, interventions were dichotomized into control arm consisting of 0 and i5 mA (
There was no significant condition by time interaction for Factor 1 (
When only subjects with FTND scores ≥5 were selected (
Mean QSU changes from baseline in Experience 2. The two placebo conditions (0 mA and i5 mA) were merged into control condition and compared to the 10 mA condition. Subjects with FTND* ≥5 were included. Arrows indicate TEAS. Error bars indicate SEM.
The difficulty of designing a control group for acupuncture lies in the fact that acupuncture by itself produces sensations and subjective feelings, the so-called “Deqi” sensation [
Eliminating and reducing any bias is a major goal of controlled studies [
These features include the use of “placebo” TEAS devices that are visually identical to active devices, exclusion of patients with earlier TEAS/acupuncture/TENS experience, use of identical protocol/visit schedule, avoidance of treatment discussion between participants and clinicians, and the use of visual signals. In addition, the subjects’ hands were hidden in a box so that neither the investigators nor the subjects could assess the amplitude of the subjects’ finger movements due to electrical stimulations, and finally each participant was explained that “each of the three treatment modalities was set with different levels of intensity, frequency, duration or administration mode, and that some of these parameters such as intensity are more stimulatory than others”. Hence, they would probably feel some sensation with high-intensity stimulations and no sensation with high-frequency stimulations which could nevertheless be as effective. This design proved successful, as the guess test showed that the double-blinding was not compromised.
Considering the future potential application of the technique by the smokers at home, we chose the TEAS instead of manual needling.
Concerning the criteria of choosing the four acupoints, the Hegu point is known to be the most powerful point to produce an analgesic effect via the increase of release of opioid peptides, which are released in the CNS [
Participants for these two studies were included only if they had FTND scores >4, because the urge to smoke (or craving) increases with higher tobacco dependence [
The effect of TEAS to antagonize the urge to smoke concerned only Factor 2 of the QSU, suggesting that TEAS is more effective in reducing an anticipation of relief from negative effect with an urgent desire to smoke during smoking abstinence. Based on our systemic study of the neurochemical mechanisms of acupuncture therapy in animal models [
As to the lack of significant effect of TEAS on Factor 1 (strong desire and intention to smoke, with smoking perceived as rewarding) of the QSU, one has to consider that the subjects were not allowed to smoke during the 26 h of the study, and lack of smoking cannot induce positive reinforcement, so it is not surprising that the TEAS intervention did not affect the positive reinforcement.
The subjects enrolled in this study were mainly of Asian origin. Thus, the present results may not be generalized to smokers of other ethnic origins.
The exploratory analysis showed that only the subgroup of more dependent smokers showed significant antagonism in the urge to smoke by TEAS, but this analysis was not planned, which weakens its strength.
Any earlier alcohol or any illicit drug consumption was not assessed (and controlled for) by structured interview, which would potentially lead to lower homogeneity of the population studied.
Finally, we assessed only a surrogate of smoking cessation (urge to smoke). The efficacy of TEAS on smoking cessation should be assessed in therapeutic trials.
The results obtained in the present study indicate that application of TEAS on the skin is capable of antagonizing the urge to smoke in dependent smokers. Based on these preliminary findings, it seems warranted to assess the efficacy of TEAS in larger populations of dependent smokers, and to plan further for clinical trials for smoking cessation.
Moleac Pte Ltd.