Abstract
Background. Moxibustion has been traditionally used to manage pain related to chronic diseases, including cancer. This study aims to investigate the efficacy and safety of moxibustion for relieving cancer pain in patients with metastatic cancer. Methods. A total of 16 patients were randomly divided into a true moxibustion (TM) group or a sham moxibustion (SM) group. In both groups, moxibustion was applied for 10 minutes, once daily for 7 consecutive days. In the SM group, the moxa cone was removed earlier than in the TM group, so as not to deliver heat stimulation completely into the skin. The changes of pain severity using the Brief Pain Inventory (BPI) and quality of life measured by the Functional Assessment of Cancer Therapy–General (FACT-G) were observed. A blinding credibility test was done to validate the sham moxibustion procedure. Results. The total BPI score significantly decreased in the TM group compared with the SM group (TM vs SM: −0.97 ± 1.05 vs 0.35 ± 0.60, P = .025). The 2 subsets of BPI, pain intensity score and pain interference score, also significantly decreased in the TM group (TM vs SM: intensity, −0.82 ± 0.93 vs 0.46 ± 0.87, P = .020; interference, −1.12 ± 1.31 vs 0.24 ± 0.61, P = .047). Even after adjusting for the values of opioid consumption, these results remained significant. FACT-G did not significantly improve in the TM group. The blinding to sham moxibustion was credible and no serious adverse events occurred. Conclusion. We suggest that moxibustion could be a safe and potential modality for cancer-related pain in patients with metastatic cancer. With the limitation of small sample size, a larger and long-term follow-up study is necessary to determine more definitely the efficacy of moxibustion.
Introduction
Pain is one of the most common and dreaded symptoms associated with cancer. Overall, 75% of patients with advanced stages of cancer experience pain severe enough to require treatment with opioids during their illness.1-3 Unrelieved pain greatly affects their activities, motivations, interactions with family and friends, and overall quality of life. Thus, the management of pain is important for improving the quality of life of patients with cancer.
Cancer pain is usually managed by opioids according to the World Health Organization (WHO) analgesic ladder, which is an algorithm developed by the WHO.4,5 Although opioids are the cornerstone of cancer pain management, they do not always completely relieve pain, because cancer pain is complex and multidimensional. 6 Moreover, opioids are associated with various side effects, including effects on neurologic, cardiopulmonary, gastrointestinal, urologic, endocrinologic, dermatologic, and immunologic systems. 7 Therefore, multidisciplinary approaches are needed to control cancer pain effectively. The optimal use of nonpharmacological interventions may serve as valuable additions to pharmacological interventions.
Moxibustion is a therapeutic method in which burning moxa produces thermal stimulation to the human body, and it affects the function of the meridians and acupuncture points to treat the disease. 8 Moxibustion has been adopted as an analgesic method in various diseases, including cancer, for thousands of years in Eastern countries and it is still frequently used in the present clinical practice.9-11 The analgesic mechanism of moxibustion is associated with local blood flow, endogenous opioids, diffuse noxious inhibitory controls, and the afferent fibers in the spinal cord.11,12 Although the effectiveness was not clearly demonstrated, some clinical studies have reported the potential effects of moxibustion in pain conditions, including osteoarthritis, scleroma, and herpes zoster, 13 and in supportive cancer care including chemotherapy-induced nausea and vomiting. 14 However, a clinical trial of moxibustion for cancer pain has not yet been explored.
In this study, we aimed to investigate whether moxibustion could reduce pain in patients with metastatic cancer compared with sham moxibustion treatment.
Patients and Methods
Study Design
The primary objective of the study was to determine whether moxibustion reduces pain in patients with metastatic cancer. The secondary objective was to determine whether moxibustion improves the quality of life in this population.
This study was a prospective, randomized, single-blinded, sham-controlled trial. The random assignment was stratified according to visual analog scale for pain severity (4-6 vs 7-10) using blocks of random length. Random assignment was implemented via a secure, computerized database, ensuring full allocation concealment. Patients’ accrual and treatment occurred from December 2009 to July 2010. This trial was conducted by a single institute, the Kyung Hee University Hospital at Gangdong, and was approved by the Institutional Review Board of the Kyung Hee University Hospital at Gangdong. All patients signed a written informed consent to participate in the study.
Study Subjects
Inclusion criteria for enrollment included age 18 years or older, histologically or cytologically documented cancer, the ability to communicate pain intensity using a numerical pain intensity scale, pain due to cancer with a baseline pain level of 4 or greater as measured using a numerical pain intensity scale, and the use of opioids (eg, oral morphine, oxycodone, hydrocodone, codeine, transdermal fentanyl, etc) for cancer pain more than once per month.
Patients were excluded from the study if they intended to receive chemotherapy or radiotherapy during the study period or had received chemotherapy or radiotherapy in the 4 weeks prior to the study. Patients were also excluded if they had more than 3 dose increases of opioids within 2 weeks of the trial because rapid increase of opioids during the study period could be confused with the effect of moxibustion. Also patients that had local skin problems at or near the moxibustion sites, any significant comorbidity, poor oral intake, and severe insomnia or depression were also excluded.
General supportive care, except for surgery, chemotherapy, and radiotherapy, was permitted during the trial. Escalation or reduction of the opioid as determined by the WHO analgesic ladder was allowed during the study period, as was administration of rescue medication for breakthrough pain. Continuous use of nonsteroidal anti-inflammatory drugs (NSAIDS), acetaminophen, and co-analgesics was permitted. Co-analgesics included antidepressants, anticonvulsants, topical agents, and corticosteroids. Conversely, neither alteration of the doses of NSAIDS, acetaminophen, and co-analgesics nor their addition was permitted during the study period.
Study Intervention
On enrollment in the study, patients were randomly assigned to either the true moxibustion (TM) group or the sham moxibustion (SM) group. The random allocation sequence was stratified according to visual analog scale for pain intensity (4-6 vs 7-10), using a table in which an “A” or a “B” was assigned based on the order in which patients were registered. While patients were blind as to which group they belonged, the clinician was aware of which patients were allocated to the TM or SM groups.
A Korean Medical Doctor (KMD) skilled in moxibustion treatment delivered the intervention during the trial. The KMD is authorized to perform moxibustion, acupuncture, and herbal medicine as determined by passing the national license examination after 6 years of education in colleges of traditional Korean medicine. Patients in both groups were treated with moxibustion 5 times at each point in a single session, once daily for 7 consecutive days. Moxibustion (Dong-bang Acupuncture Inc, Seoul, South Korea) consisting of herb preparation, Artemisia vulgaris, in a cone shape of 1.4 cm diameter, 1.5 cm height, and 150 mg mass was applied directly to the skin surface.
A total of 5 points, including Guan-won (CV4), Jung-wan (CV12), and three ah shi acupoints were selected by the KMD for moxibustion treatment. Ah shi points were determined as the tenderness on the areas where patients complained of pain the most. CV4 and CV12 points were selected because they invigorate Qi at the middle and lower burners. 15
The method of selecting moxibustion points was the same in both groups. After selecting the points, the KMD placed a moxa cone on each point and ignited it. After the cone was burnt, it was removed and another one was placed on the same point repeatedly for 10 minutes at each session, once daily for 7 consecutive days
In the TM group, heat stimulation of moxa was delivered directly and completely to the treatment points, and the moxa cone was removed just before the skin was burnt (less than 0.5 cm of unburned moxa remained). In the SM group, the moxa cone was removed earlier so as not to deliver heat stimulation completely to the treatment points (more than 1 cm of unburned moxa remained). Patients were not informed of the SM method at the beginning of the trial to ensure the blindness of the study, and patients were not allowed to view the moxa cone during treatments. Patients remained blinded to their treatment throughout the trial period.
Outcome Measures
The primary endpoint of the study was the difference in the Brief Pain Inventory (BPI) total scores between day 0 and day 7. The BPI is a reliable and validated tool for assessing cancer pain that measures both the intensity of pain and the degree to which the pain interferes with patient life.16,17 The BPI was first developed in English and has been validated in the Korean language. 18 The BPI pain intensity score is the average score of 4 questions—the current rate of patient pain, the pain rating at its worst and at its least, and the average over the last 24 hours—using an 11-point numeric rating scale. Likewise, the BPI pain interference score, which uses a similar numeric rating scale as that for the pain intensity score, is the average score of 7 questions regarding the degree to which pain interferes with general activities including mood, walking, working, sleeping, relationships with others, and enjoyment of life. The BPI total score is the average of the BPI pain intensity score and the BPI pain interference score, with higher scores indicating more severe pain.
The secondary end points included the differences in the Functional Assessment of Cancer Therapy–General (FACT-G) score between day 0 and day 7. The FACT-G scale is widely used to measure health-related quality of life in cancer patients, and the Korean version has been validated using Korean breast cancer patients.19,20 The FACT-G questionnaire includes 27 items with 4 distinct subscales: physical well-being, social well-being, emotional well-being, and functional well-being, with higher scores indicating more favorable health.
Additionally, daily opioid intake for 7 consecutive days was recorded, and the difference between day 0 and day 7 was calculated. Assessment of blinding credibility was also performed to validate the sham moxibustion procedure using a questionnaire that asked patients “Do you think you received true moxibustion?” on day 7. Choices included “true moxibustion” or “sham moxibustion.”
Statistical Analysis
All results are expressed as mean ± standard deviation unless otherwise stated. The principal analysis of this study consisted of testing for differences in the evaluation criteria between the TM and SM groups before and after intervention using 1-way analysis of covariance. In each analysis, the baseline level of each patient was used as a covariate. We assumed that improvement in BPI could have been affected by change in opioid usage, and thus BPI was also subjected to analysis of covariance by converting the daily dose of opioid to an oral morphine equivalent dose and using opioid consumption as a covariate. The Mann–Whitney test and Fisher’s exact test were also used to analyze the differences in the data between day 0 and day 7 in the TM and SM groups. P values were 2-sided, and those less than .05 were considered statistically significant. Data were analyzed using the Statistical Package for Social Sciences (SPSS) Release 15.0 (SPSS Inc, Chicago, IL).
Results
Patients’ Characteristics
A total of 16 patients who met all aforementioned inclusion criteria and provided consent were enrolled in the trial. Two patients were dropped during the study period; one was failed at day 3 due to aggravation of ascites, and the other patient was failed at day 2 due to aggravation of pancreatic cancer. Therefore, 14 of 16 patients completed the trial.
Baseline characteristics were not significantly different between the TM and SM groups (Table 1). The mean patient age (standard deviation) was 50.8 (12.3) years and 6 (42.9%) of the patients were male. Patients had an Eastern Cooperative Oncology Group (ECOG) performance status range of 1 to 3 with an ECOG of 3 in 9 (64.3%) patients. Eleven (78.6%) patients had progressive diseases and 3 (21.4%) had stable diseases at baseline. The cancer origins of participants were colon in 5 (35.7%), breast in 2 (14.3%), stomach in 2 (14.3%), and other cancer types in 5 (35.7%) patients. All patients had metastatic sites: the liver in 7 (50%), lymph nodes in 7 (50%), lung in 6 (42.9%), peritoneum in 4 (28.6%), and other sites in 6 (42.9%) patients. Opioids were given to all patients as oxycodone in 7 (50%), morphine in 2 (14.3%), and other forms in 5 (35.7%) patients. There were no differences between the TM and SM groups in total BPI score, BPI intensity, BPI interference, FACT-G, and opioid consumption at baseline (Table 2). During the trial, none of the patients received chemotherapy, radiotherapy, or surgery; they only received the best supportive care.
Baseline Characteristics.
Abbreviations: SD, standard deviation; ECOG, Eastern Cooperative Oncology Group.
P values from Fisher’s exact test or Mann–Whitney U test.
Opioid was calculated in milligrams of oral morphine equivalents per day.
Changes in BPI, FACT-G, and Opioid Consumption.
Abbreviations: BPI, Brief Pain Inventory; FACT-G = Functional Assessment of Cancer Therapy–General; SD, standard deviation; Δ, score at day 7 minus score at day 0.
P values from 1-way analysis of covariance, the baseline level as a covariate.
P values from 1-way analysis of covariance, adjusted to the difference values of opioid consumption.
P values from Mann–Whitney U test.
P values <.05.
Changes of BPI, FACT-G, and Opioids
After 7 days of moxibustion treatment, there was a significant decrease in the total BPI score, BPI intensity score, and BPI interference score compared with baseline in the TM group, but no such decreases were observed in the SM group (Table 2, Figure 1). The differences in total BPI, BPI intensity, and BPI interference between day 0 and day 7 were −0.97 ± 1.05 versus 0.35 ± 0.60 (P = .025), −0.82 ± 0.93 versus 0.46 ± 0.87 (P = .020), and −1.12 ± 1.31 versus 0.24 ± 0.61 (P = .047) in TM and SM groups, respectively. These statistical significances remained unchanged after adjusting for differences in opioid consumption.
Brief Pain Inventory Score in true moxibustion group (n = 7) and sham moxibustion group (n = 7). BPI, Brief Pain Inventory. *Indicates significant changes by 1-way analysis of covariance (P < .05).
Even though it was not significant, there was an increase in FACT-G compared with baseline in the TM group but a decrease in the SM group. The differences in FACT-G between day 0 and day 7 were 2.71 ± 10.42 versus −5.14 ± 3.85 (P = .140) in the TM and SM groups, respectively.
After calculating oral morphine equivalents per day because of the heterogeneous opioid types and amounts used by the patients, there was no statistically significant difference in opioid consumption between the 2 groups (P = .694, Table 2). Changes in daily opioid intake were largely associated with rescue medication for management of breakthrough pain.
Blinding Credibility Test
The blinding credibility test was based on patients’ conjecture regarding whether they had actually received treatment. As shown in Table 3, there were no statistically significant differences between the 2 groups (P = .559). Therefore, the blinding to sham moxibustion in this trial can be considered a credible method.
Blinding Credibility Test.
p-Values from Fisher’s exact test.
Safety
No treatment-related adverse events, including moxibustion-induced skin damage, were reported by either group, although 1 patient complained about the smell from the moxa as it burned.
Discussion
This randomized, single-blinded, sham-controlled pilot study with small population demonstrated the significant analgesic efficacy of TM in comparison with SM in patients with metastatic cancer. Moxibustion reduced the total BPI score, as well as both the BPI pain intensity and BPI pain interference scores. Importantly, the observed pain intensity and pain relief were not associated with opioid consumption.
Cancer pain shares the same neuropathophysiological pathways as noncancer pain. It is a mixed mechanism pain, rarely presenting as purely neuropathic, visceral, or somatic. Rather, it may involve inflammatory, neuropathic, ischemic, and compressive mechanisms at multiple sites. 21
Moxibustion and acupuncture have traditionally been used to alleviate pain symptoms related to many diseases and are similar in achieving therapeutic effects by stimulating acupuncture points. However, according to the theory of traditional Korean medicine, moxibustion is a preferable method to treat chronic diseases, whereas acupuncture has been preferably used to treat acute diseases. 15 Clinical studies of acupuncture showed positive effects on relieving cancer-related pain22,23 and it is recommended as a complementary therapy when pain is poorly controlled. 24 Moxibustion has been relatively less studied compared with acupuncture because of the risks of fire and burn injury 25 and difficulties in applying it to animal models. Clinical studies on the analgesic effects of moxibustion were reported in patients with osteoarthritis of the knee, herpes zoster, and scleroma. 13 Murase and Kawakita 26 reported that diffuse noxious inhibitory controls were closely related to the analgesic mechanism of moxibustion. Okada and Kawakita 11 and Noguchi et al 12 indicated that moxibustion is related to the action of endogenous opioids and induced local inflammatory responses, such as vasodilatation and extravasation, by releasing neuropeptides. Fukuda et al 27 reported that moxibustion decreased serotonin levels in rat nucleus accumbens, and suggested that moxibustion affects the brain reward system. Taken together, clinical and preclinical studies of the analgesic effect of moxibustion strongly support the positive result of moxibustion on cancer pain reported in this study.
We observed no significant improvements in FACT-G, a measurement of the quality of life. Contrary to this result, Bian et al 28 showed favorable effects of moxibustion on quality of life compared with morphine injection, and de Valois et al 29 reported that applying moxibustion with acupuncture improved quality of life after 4 weeks of intervention for cancer survivors with lymphedema. In our study, of all patients with metastatic cancer, most had progressive cancer disease and very poor performance status with ECOG 3. One limitation to this study is that the study period was only 1 week. A long-term study of cancer patients with good performance status is needed to confirm these results.
There are several reported sham moxibustion devices and these devices were used in the blinding of studies and are therefore feasible for use in clinical trials.30,31 These sham devices were made to prevent the transfer of heat from the moxa to the patient by inserting insulating materials between the moxa and skin. In this study, we also varied the amount of heat that reached the skin between sham moxibustion and true moxibustion by manipulating the procedure. Though this method has some limitations, including that the procedure requires skillful manipulation and is not blinded to the practitioner, the blinding credibility of this method was statistically validated.
No serious adverse events related to moxibustion were reported in this study, except that one patient complained about the smell of burning moxa. Park et al 25 reported several potential adverse events including burning, allergic reactions, and infections associated with moxibustion. Skillful manipulation of moxibustion is necessary because skin damage at the moxibustion site is one of the most common adverse events.
Even though this study has a limitation of a small sample size, this is the first randomized pilot study demonstrating the efficacy of moxibustion for cancer pain in metastatic cancer patients. Based on this study, more methodologically rigorous studies should be performed. Specifically, a large population study with a sham moxibustion device that is blind to both participants and practitioners and long-term follow-up must be considered.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Clinical Research Institute, Kyung Hee University Hospital at Gangdong in 2010 (KHNMCCRI-2010-03).