Acupuncture Combined with Three-Step Analgesic Drug Therapy for Treatment of Cancer Pain: A Systematic Review and Meta-Analysis of Randomised Clinical Trials

Objective The purpose of this study was to systematically evaluate the efficacy and safety of acupuncture combined with the WHO three-step analgesic drug ladder for cancer pain. Methods The Cochrane Library, PubMed, and CNKI Database of Systematic Reviews were searched. Using the Cochrane Register for Randomized Controlled Trials, the quality of the included literature was evaluated, and the meta-analysis was carried out with RevMan 5.3 software. Results Compared with three-step analgesia alone, acupuncture combined with three-step analgesia for cancer pain increased pain relief response rates (RR = 1.12, 95% CI: 1.08∼1.17, P < 0.00001), reduced NRS score (SMD = −1.10, 95% CI: −1.86∼−0.35, P=0.004), reduced the rate of side effects (RR = 0.45, 95% CI: 0.38∼0.53, P < 0.00001), including nausea (P < 0.00001), vomiting (P=0.008), constipation (P < 0.00001), and dizziness (P=0.010), reduced the burst pain rate (SMD = −1.38; 95% CI: −2.44∼−0.32, P=0.01), shortened analgesia effect onset time (P=0.004), and extended the duration of response (P < 0.0001). Conclusion For the treatment of cancer pain, acupuncture combined with three-step analgesic drugs is better than using only three-step analgesic drugs.


Introduction
Pain is one of the most debilitating symptoms experienced by patients with advanced cancer. According to WHO statistics, 70% of cancer patients worldwide have some degree of pain in the advanced stages of cancer [1]. Because the pain is intense and easily aggravated, it directly affects the appetite, sleep, psychological status, and treatment effect of patients, reduces their quality of life, and increases their psychological stress [2,3]. Cancer pain has become a medical, psychological, and social issue of great concern. At present, the treatment of cancer pain mostly utilizes the three-step "ladder" treatment principle proposed by the WHO, where mild, moderate, and severe pain are treated with nonsteroidal anti-inflammatory drugs (NSAIDs), weak opioids, and strong opioids, respectively [4]. Although the analgesic effect of three-step drugs is good, their side effects, such as liver and kidney function damage, risk of dependency and addiction, respiratory inhibition, and gastrointestinal side effects, limit their clinical application [5]. One primary reason why cancer pain is difficult to control is because cancer pain patients cannot tolerate the side effects of analgesics. erefore, there is a consensus to seek other effective and safe analgesic methods [6]. Acupuncture is an important part of TCM. Acupuncture is to insert a needle at one of the patient's acupoints and use a specific manipulation to stimulate the patient's acupoints to achieve an effect (DE QI). Acupuncture has outstanding performance in the treatment of all kinds of pain through its principle of moving qi, dredging qi channels and collaterals, and activating blood. Various acupuncture treatments can be combined, with no risk of addiction, no side effects, convenient application and at a low cost, demonstrating the unique advantages of TCM in the treatment of cancer pain [7]. In the last 20 years, there have been many clinical reports on the utility and safety of acupuncture for the treatment of cancer pain, and acupuncture therapy is a widely recognised alternative measure for the treatment of cancer pain. erefore, it is necessary to use a systematic evaluation method to rigorously evaluate the randomised controlled study of acupuncture combined with three-step analgesic drugs to treat cancer pain, to assess its exact effect in the treatment of cancer pain.

Data Sources.
e following databases were searched from their inception to January 10, 2021: the Cochrane Library, PubMed, Embase, CNKI, China Biology Medicine disc (CBMdisc), Chinese Journal of Science and Technology database (VIP), and Wanfang database. We searched MeSH (Medical Subject Headings) term trees for "acupuncture" and "pain" in PubMed, and the keywords searched included "acupuncture", "needling", "tumour", "cancer", "neoplasm", "ache", "pain", and "randomised controlled trial". e keywords were translated into Chinese and searched in the above-mentioned Chinese databases. Search terms were combined with the Boolean "AND" and "OR" terms in search strategies, for example, ("acupuncture" OR "needling") AND ("cancer" OR "tumour" OR "neoplasm" OR "ache" OR "pain") AND ("randomised controlled trial"). Comprehensive retrieval was carried out according to the characteristics of different databases. en, the literature mentioning "randomised controlled" and "randomised grouped" was screened. In addition, we manually searched our own personal literature files. After reading the full text of the included literature and related articles, we collected the documents together in hard copy format for preservation.

Types of Studies.
e included studies were all randomised controlled clinical trials. e published experiments included were mainly in the form of theses and abstracts.
ere were no restrictions on the language of publication.

Types of Participants.
e subjects were patients with malignant tumours confirmed by cytology or histopathology, and all patients had cancer pain. ere were no limits on age, gender, race, and nationality of the patients; however, patients had to be able to clearly describe their pain to medical staff.

Types of Interventions.
In the literature, the intervention treatment group was treated with acupuncture augmented by three-step analgesia, including traditional acupuncture or other acupuncture methods, such as ear acupuncture and electroacupuncture. Acupuncture points included traditional acupuncture points and pain points. e control groups only received three-step analgesic treatment. .  e included materials  had a clear evaluation standard for curative effects and at  least one clinical index related to cancer pain, including the  effective rate of pain relief after treatment, quality of life  score, side effect rates, burst pain rate, onset time to analgesic effect, and duration of response.

Exclusion Criteria.
e exclusion criteria were as follows: if patients had one or more other type(s) of pain in addition to cancer pain; if the study used moxibustion, percutaneous electrical stimulation of nerves, acupoint injection, laser irradiation, cupping, massage, herbal medicines, or other intervention measures; if the experiments were carried out on patients during or a few days after surgical therapy, radiotherapy, chemotherapy, or hyperthermia-therapy on their malignant tumours; if the trial design was not rigorous; if inappropriate statistical methods were used; if the paper was only an abstract, review, or summary of previously published literature; if the study has no result indicators; if the experimental design was unreasonable; or if the literature could not be obtained by contacting the author.

Data Extraction and Bias Risk Assessment.
Two researchers independently evaluated the quality of each study meeting the inclusion criteria and extracted the data, including the baseline situation, intervention measures, and efficacy results, and cross-checked the data. Any disagreements were resolved through discussion or assessment by a third researcher. We used a "Modification of Cochrane Tool to assess the risk of bias in randomised trials," where a decision regarding bias must be made, categorised into "probably no" or "probably yes," for items that are thought to be of unclear risk [8]. We judged trials with more than 2 and more than 4 high-risk components as moderate risk and high risk, respectively [9]. e following criteria were used to assess the risk of bias: whether the study was randomised; how allocation concealment was conducted; whether the study was double-blind or triple-blind; whether the results data were complete; and whether there was selective reporting or other types of bias. e authors categorised studies into "low risk," "unclear risk," and "high risk" categories. For dropout patients, we contacted the authors of the studies twice over four weeks via e-mail for missing or unclear data. If missing data could not be found, they were recorded as high risk; if no response was received, the data were marked as unclear risk. All authors reached a consensus on the results of bias risk assessment.

Data Synthesis.
e effect of acupuncture combined with three-step analgesic drug therapy for treatment of cancer pain was analyzed in terms of response rate, numerical rating scale (NRS), side effect rates, times of burst pain, onset time, and duration of response (DOR). If the information included in the study was insufficient, we communicated with the main author to obtain accurate data. RevMan 5.3 software provided by the Cochrane Collaboration Network was used for the meta-analysis. e relative risk (RR) was used for the enumeration data, the mean difference (MD) was used for the measurement data, and the 95% confidence interval (CI) was used for each effect quantity. When the heterogeneity of test results was not statistically significant (P > 0.05), a fixed effects model was selected; when the heterogeneity of test results was statistically significant (P < 0.05), a random effects model was selected. A funnel plot was used to analyse and detect publication bias.
ree studies used wrist-ankle acupuncture (WA) [17][18][19]. Among them, the two acupuncture methods were all included in Fu Yang et al.'s report [17], in which morphine hydrochloride sustained-release tablets and acupuncture or wrist-ankle acupuncture were used in the treatment of cancer pain. All studies provided patients with a semistandardised acupuncture programme, that is, the use of a predefined set of acupoints combined with a set of acupoints according to the location of the tumour.
e Ashi point, Zusanli (ST36), Hegu (LI4), Sanyinjiao (SP6), and Taichong (LR3) points were most frequently used. For most studies, patients received acupuncture treatment for 1 to 3 weeks, for durations of 20 to 60 min per session. e evaluation criteria for the curative effect were similar across studies. e objective outcome measures were treatment response rate, NRS, side effect rates (nausea, vomiting, constipation, hiccups, dizziness, itching, palpitation, and abdominal distention), times of burst pain, onset time to analgesic effect (min), DOR (h), quality of life (QOL), Karnofsky performance status (KPS), and quality of life questionnaires (QLQ-C30). e minimal important difference (MID) refers to the change in the score of the smallest efficacy evaluation questionnaire recognised by the patient. MID indicates an important improvement in symptoms and signs; the intervention has achieved the minimal important difference.
e total response rate of the treatment group was better than that of the control group, and the difference was statistically significant (n � 1351, RR � 1.12; 95% CI: 1.08∼1.17, P < 0.00001; see Figure 3).

NRS Score.
Seven studies reported NRS scores after treatment [10,11,14,17,19,22,24]. Overall, 282 cases were in the treatment group, and 281 were in the control group. e heterogeneity test of the meta-analysis showed that χ 2 � 158.90, P < 0.00001, I 2 � 96%, and the differences between the studies were statistically significant, so a random effects model was used. e NRS score of the treatment group was lower than that of the control group, and the difference was statistically significant (n � 563, SMD � −1.10, 95% CI: −1.86∼−0.35, Z � 2.87, P � 0.004; see Figure 4).

Burst Pain.
Four studies reported the mean number of burst pain events [11,12,17,18]. e heterogeneity test in the meta-analysis showed that χ 2 � 78.30, P < 0.00001, I 2 � 95%, and the differences between the studies were statistically significant, so a random effects model was used. e combined statistical results showed that the incidence of burst pain in the treatment group was lower than that in the control group (n � 244, SMD � −1.38, 95% CI: −2.44∼−0.32, P � 0.01; see Figure 6).

Onset Time to Analgesic Effect and Duration of Response.
Five studies reported the mean onset time [11,17,19,24,28]. e combined statistical results showed that the onset time in the treatment group was shorter than that in the control group (n � 360, SMD � −20.11, 95% CI: −33.90∼−6.33, Evidence-Based Complementary and Alternative Medicine P � 0.004). Six studies reported the mean duration of response [11,17,19,22,24,28]. e combined statistical results showed that the duration of response in the treatment group was longer than that in the control group (n � 440, SMD � 3.22, 95% CI: 1.63∼4.80, P < 0.0001); see Figures 7 and 8.

Publication Bias.
Publication bias, which has always been a problem in meta-analysis, refers to the fact that research with positive results is easier to publish than research with negative results. e funnel chart analysis results of the main outcome indicators of the response rates of pain relief suggested that publication bias might exist and exaggerate the efficacy of acupuncture combined with threestep analgesic drugs in the treatment of cancer pain; see Figure 9.

Discussion
Cancer is a significant global public health issue, and the disease burden is growing. Globally, there are 18.1 million new cancer cases and 9.6 million cancer deaths each year; cancer deaths are expected to exceed 13 million by 2030, and 70% of cancer deaths globally occur in low-income and middle-income countries. In China in 2018, nearly 24% (4.3 million) of global new cases and 30% (2.9 million) of deaths occurred [29,30]. China is the largest developing country. Chinese doctors need to pay attention to promoting cancer prevention for people and treating cancer patients. Cancer pain is severe, intolerable, and intractable pain, and such pain is a main symptom in the advanced stages of malignant tumours. When the tumour body markedly enlarges, tissue necrosis, erosion, and so on result in severe compression of, damage to, and irritation of the nerve sheath, nerve fibres,  29  29  22  97  28  22  25  29  29  33  25  30  58  27  12  39  47  14  26   32  30  23  106  30  25  30  30  30  35  30  31  60  35  16  40  50  16  30   21  25  22  78  24  19  23  25  28  25  24  27  56  33  11  35  38  11  23   30  30  24  101  30  25  30  30  30  34  30  31  60  35    Evidence-Based Complementary and Alternative Medicine and blood vessels. Although there are many ways to treat cancer pain, many years of clinical experience at home and abroad indicate that providers believe that drug therapy is still the most common and effective way to control cancer pain. e WHO three-step cancer pain treatment programme has become an internationally accepted cancer pain drug treatment method that can control most cancer pain; however, three-step pain drugs, especially opioids, are often accompanied by side effects such as nausea, vomiting, constipation, drowsiness, dizziness, and respiratory depression [31].
ere are many ways to treat cancer, but in recent years, TCM has played an increasingly important role in cancer prevention and treatment. As an integral part of TCM, acupuncture has been used to treat pain for thousands of years. e complications of acupuncture in the treatment of pain diseases are fewer than those of drug treatment [  Evidence-Based Complementary and Alternative Medicine 9 pneumothorax, dizziness, pain, needle syncope, infection, and visceral puncture can occur due to improper acupuncture manipulation; however, when doctors master anatomical knowledge, perform acupuncture correctly, and sterilise needles strictly, complications are further reduced. Modern research shows that the mechanism of acupuncture analgesia may be related to regulating the self-healing of the body, changing patients' perceptions of pain, and affecting the conduction of the central nervous system [34]. Another possible acupuncture mechanism is stimulation/excitation of the endogenous pain modulation system, which induces the secretion of endogenous opioids, blocks the transmission of neurotransmitters, and regulates the perception of pain to achieve analgesia [35]. A third possibility is that the pain signals from acupuncture are modulated in the pain receptor areas, and the dorsal root ganglion cells of the outgoing primary neurons transmit the signal to the near end of the secondary neurons. e pain signal produced by acupuncture may then induce the secretion of endogenous opioids and analgesia in the periaqueductal grey matter of the midbrain, or it may induce the penetration of electric ions, stimulate neurons, and exert an inhibitory effect in the intercellular area of the periaqueductal grey matter of the midbrain [36]. e exact mechanism of acupuncture's analgesic effect has not yet been elucidated. However, this review of acupuncture treatment of cancer pain with a large number of RCTexperiments demonstrated that acupuncture treatment of cancer has fewer adverse reactions such as  Figure 8: Forest plot of the duration of response of acupuncture combined with three-step analgesic drugs versus three-step analgesic drugs alone for cancer pain. nausea and vomiting than analgesic drug treatment alone. Acupuncture treatment for cancer pain is considered to have sufficient evidence to determine its effectiveness [37]; these results are encouraging and support further research on acupuncture treatment for cancer.

Conclusion
Based on the meta-analysis of 19 studies, compared with the treatment of cancer pain with three-step analgesic drug treatment alone, the response rates of pain relief from acupuncture combined with three-step analgesic drug treatment were higher, the NRS scores were lower, the incidence of adverse reactions such as nausea and vomiting was less frequent, the incidence of times of burst pain was also less frequent, the onset time to analgesic effect was shorter, and the duration of pain response was longer.
ere were several limitations in this study. e lack of high-quality studies in the literature may limit the validity of the results. Meta-analyses generally face methodological challenges such as insufficient literature retrieval, potential selection bias for which studies are included, and inappropriate evaluation of the quality of the original research.
is study only included published literature and did not search for unpublished literature; in addition, there may be publication bias in the literature.
In conclusion, this study shows that acupuncture combined with three-step analgesic drugs has specific advantages over three-step analgesic drugs alone in the treatment of cancer pain. It is hoped that, in the future, rigorous randomised controlled trials will be carried out with multicentre and large-sample studies to determine acupuncture's exact curative effect and further demonstrate the superiority of acupuncture combined with three-step analgesic drugs over the use of such drugs alone to treat cancer pain.
Data Availability e data can be obtained from the author upon reasonable request.

Conflicts of Interest
All the authors declare no conflicts of interest.

Authors' Contributions
De-hui Li conceived and designed the study, analysed and interpreted the data, and drafted the manuscript. Yi-fan Su and Na Guo contributed to the literature searches, study selection, data extraction, and data synthesis. Huan-fang Fan and Chun-xia Sun assisted in the development of search strategies and critically reviewed the manuscript. All the authors read and approved the final version of the manuscript. Evidence-Based Complementary and Alternative Medicine 11