Complementary Therapies in Medicine (2015) 23, 265—274

Available online at www.sciencedirect.com

ScienceDirect

jo ur nal home p ag e: www.elsev ierhea l th .com/ journa ls /c t im

Traditional herbal medicine for cancer pain:A systematic review and meta-analysis

Jung-Woo Leea,1, Won Bock Leea,1, Woojin Kima,Byung-Il Mina,b, HyangSook Leec, Seung-Hun Chod,∗

a Department of East-West Medicine, Graduate School, Kyung Hee University, Seoul 130-701, South Koreab Department of Physiology, College of Medicine, Kyung Hee University, Seoul 130-701, South Koreac Acupuncture and Meridian Science Research Center, College of Korean Medicine, Kyung Hee University,Seoul, South Koread Hospital of Korean Medicine, Kyung Hee University Medical Center, #1 Heogi-Dong, Dongdaemun-Gu,Seoul 130-701, South KoreaAvailable online 19 February 2015

KEYWORDSTraditional herbalmedicine;Cancer pain;Systematic review;Meta-analysis

SummaryBackground: The effectiveness of traditional herbal medicine (THM) as an adjunctive therapyfor cancer pain is unclear.Objective: To assess the effectiveness of THM as an adjunctive therapy for cancer pain usingrandomized controlled trials (RCTs).Methods: Five electronic databases, including those from the UK and China, were systematicallysearched for the period before September 2013. All RCTs involving the use of THM in combinationwith conventional cancer therapy for cancer pain were included.Results: Twenty-four RCTs involving 4889 patients with cancer pain were systematicallyreviewed. Among them, nine studies of 952 patients reported a significant decrease in thenumber of patients with cancer pain in the treatment group. Four studies of 1696 patientsreported a significant decrease in the degree of pain in the treatment group.Conclusion: The results of these studies suggest that THM combined with conventional therapy

is efficacious as an adjunctive therapy for patients with cancer pain. However, more research,including well-designed, rigorous, and larger clinical trials, are necessary to address theseissues.© 2015 Elsevier Ltd. All rights reserved.

∗ Corresponding author. Tel.: +81 02-958-9184; fax: +81 02 958 9183.E-mail address: chosh@khmc.or.kr (S.-H. Cho).

1 Both authors contributed equally to this work.

http://dx.doi.org/10.1016/j.ctim.2015.02.0030965-2299/© 2015 Elsevier Ltd. All rights reserved.

2

C

I

TirmuwtbpmocoirhavAlhe4aTHAc

taabsihectr

66 J.-W. Lee et al.

ontents

Introduction.............................................................................................................. 266Materials and methods ................................................................................................... 266

Search strategy...................................................................................................... 266Study selection...................................................................................................... 266Quality assessment .................................................................................................. 267Statistical analysis................................................................................................... 267

Results ................................................................................................................... 267Study description.................................................................................................... 267Traditional herbal medicine used for pain management ............................................................. 267VAS improvement by number of persons ............................................................................. 267VAS improvement by degree......................................................................................... 272

Discussion................................................................................................................ 272Conclusions .............................................................................................................. 273Conflict of interest statement............................................................................................ 273

Acknowledgements..................................................................................................... 273References ............................................................................................................. 273

ntroduction

he International Association for the Study of Pain (IASP)s an international professional organization promotingesearch, education, and policies related to pain manage-ent. The often-quoted IASP definition of pain as ‘‘an

npleasant sensory and emotional experience associatedith actual or potential tissue damage, or described in

erms of such damage’’ is derived from a 1964 definitiony Harold Merskey. Although cancer encompasses multiplehysical symptoms, the symptom of pain is often cited asost critical. Pain is one of the most common symptoms

f cancer, and its intensity increases as the stage of can-er advances. A meta-analysis reported cancer pain in 64%f patients with metastatic disease, 59% of patients receiv-ng antineoplastic therapy, and 33% of patients who hadeceived curative cancer treatment.1 In Asia, traditionalerbal medicine (THM) is frequently combined with Westernpproaches to treat cancer, usually in regimens that combinearious traditional Asian herbs into one treatment strategy.lternative medicine has been used to meet patient needs in

ieu of or as an adjunct to conventional medicine.2 More thanalf of patients treated with traditional Asian herbs reportffective relief of their symptoms, including pain.3 About1—62% of patients with cancer use traditional Asian herbss an alternative therapy.4,5 Clinical trials suggest that AsianHM may alleviate cancer pain with no adverse effects.owever, a scientific evaluation of the effect of traditionalsian herbs on pain is lacking, and safety and toxicity areoncerns.

The effectiveness of traditional Asian herbs is con-roversial among current practitioners of complementarylternative medicine. To date, no systemic review of the oraldministration of traditional Asian herbs for cancer pain haseen conducted. Thus, we conducted this systemic review toummarize and critically assess the evidence from random-zed controlled trials (RCTs) showing that traditional Asianerbs are effective for reducing cancer pain. Indeed, sev-

Materials and methods

Search strategy

The following sources were searched from their incep-tion to September 2013: The Cochrane Central Register ofControlled Trials, MEDLINE, EMBASE, the Allied and Comple-mentary Medicine Database, and the Cumulative Index toNursing and Allied Health Literature.

The reference lists of the articles were checked forfurther relevant publications, and experts were asked forinformation concerning any additional trials. An additionalmanual search for relevant journals, symposia, and con-ference proceedings was performed, and all identifiedpublications were cross-referenced. When necessary, per-sonal contact was made with the authors of the publishedstudies to request additional data.

The search terms used were cancer pain [TIAB]OR cancer patient [TIAB] OR cancer patients [TIAB]OR ((Neoplasms[MeSH] OR Neoplasms*[TI] OR Cancer*[TI]OR Tumor*[TI] OR Tumor*[TI] OR Carcinoma[MeSH] ORCarcinoma*[TI] OR Adenocarcinoma[MeSH] OR Adenocar-cinoma*[TI] OR adenomatous[TI] OR Lymphoma[MeSH] ORlymphom*[TI] OR lymphedema*[TI] OR Sarcoma[MeSH]OR Sarcoma*[TI] OR ‘‘Antineoplastic agents’’[MeSH] ORantineoplas*[TI] OR ((adenom*[TI] OR adenopath*[TI])AND malignant*[TI])) AND (PAIN[MeSh] OR PAIN MEA-SUREMENT[MeSh] OR PAIN CLINICS[MeSh] OR ANALGE-SIA’’[MeSh] OR ANALGESICS’’[MeSh] OR pain*[TIAB] ORanalges*[TIAB]))). Each database was searched indepen-dently, as each database searched for this review possessedits own subject headings. No language restrictions wereimposed.

Study selection

ral RCTs have reported that Asian THM is effective againstancer pain. Thus, we conducted this study as a follow-upo the systematic review conducted by Ling et al.6 to updateesearch in this area.

Wron

e selected only articles on RCTs; quasi-randomized or non-andomized trials were excluded. We also excluded articlesn animal or in vivo experiments. Studies in which THM wasot orally administered were also excluded.

tet1aemnrpwq

sudod

awah

Mmorahisc

Tm

Tu

1

2

3

m

V

Traditional herbal medicine for cancer pain

Studies using THM combined with conventional cancertherapy for the treatment group were included. We selectedpatients receiving conventional treatment such as radiothe-rapy or chemotherapy, placebos or microwave treatment, orno treatment as the control group.

Quality assessment

We (JWL and WBL) independently selected the endpoint dataon the major outcome measures from each trial in stud-ies evaluating THM. Details of this procedure have beenexplained elsewhere.7 We prefer continuous to binary databecause most of the eligible trials reported continuous out-comes. We referenced the previous reviews8,9 in an opendiscussion. Summaries and quality assessments of all stud-ies were completed by two reviewers following the detaileddescriptions of these categories provided in the CochraneHandbook for Systematic Reviews of Interventions.10 The fol-lowing questions were assessed and answered by reviewers:(a) Was the allocation sequence adequately generated? (b)Was allocation adequately concealed? (c) Was knowledge ofthe allocated interventions adequately prevented during thestudy? (d) Were procedures to ensure blindness regardingoutcome assessment adequate? (e) Were incomplete out-come data adequately addressed? (f) Were the results ofthe study free of selective outcome reporting? (g) Was thestudy apparently free of other problems that could put it ata risk for bias? This review used ‘‘Y,’’ ‘‘U,’’ and ‘‘N’’ to coderesponses to these questions; ‘‘yes’’ (Y) indicated a low riskof bias, ‘‘unclear’’ (U) indicated an uncertain risk of bias,and ‘‘no’’ (N) indicated a high risk of bias.

Statistical analysis

Study data were summarized using basic statistics and sim-ple counts and means. The main purpose of the analyseswas to quantify and compare the effects of THM combinedwith conventional cancer therapy (treatment group) withthose of conventional cancer therapy alone (control group)on patients with cancer pain using only RCTs. The statisticalanalysis was performed using Review Manager 5.1 for Win-dows (The Nordic Cochrane Center). The odds ratios (ORs)for improvement on the visual analog scale (VAS) by numberof persons, degree, and number of pain occurrences per day,with their 95% confidence intervals (CI), are presented indi-vidually for each trial. An OR <1 indicates a lower risk forthe treatment than for the control group, and an OR >1 indi-cates a greater risk for the treatment than for the controlgroup.

Results

Study description

An initial search identified 331 potentially relevant arti-

cles. Of these, only 2411—34 met our inclusion criteriaand, thus, were subjected to our systematic review.Nine articles11,12,14,19,20,22,23,27,34 were in English, and 15articles13,15—18,21,24—26,28—33 were in Chinese.

Ant(

267

A total of 307 articles were initially excluded becausehey did not meet our inclusion criteria. Of these, 87 werexcluded because they were duplicates of other articles orheir titles clearly reflected their irrelevance. Additionally,96 articles were excluded after an Abstract review. Tendditional articles were included after the review of refer-nces. Thirty-four additional articles were excluded after aore detailed evaluation of each article. Six studies were

ot RCTs, and 28 did not meet our inclusion criteria. As aesult, 24 RCTs11—34 on THM orally administered for cancerain were reviewed. The total number of subjects evaluatedas 4889. Fig. 1 summarizes the search results based on auality of reporting of meta-analyses flow diagram.35

Various cancers (breast, gastric, lung, colorectal, non-mall cell lung, pancreatic, esophageal, stomach, cervical,terine, kidney, leukemia, prostate, ENT, esophagus, blad-er, submaxillary gland, bone, primary hepatic carcinoma,varian, gallbladder, renal, large intestine and mid-le/advanced stage of lung) were included in RCTs.

The intervention varied considerably across the tri-ls. There are four kinds of comparisons. Herbs vsestern therapy,16,21—24,26—28,31,33 herbs + western ther-py vs herbs,13,15,17—20,32,34 herbs vs western therapy vserbs + western therapy,12,14,25,29,30 and herbs vs none.11

All studies based the THM elections on Traditional Chineseedicine theory. According to the Chinese medicine treat-ent method, strengthening Qi and eliminating pathogens

ften use botanicals, promoting blood circulation andemoving stasis often use botanicals and animal productsnd promoting cytotoxic effect often use minerals. Variouserbal medicines were used in the included RCTs; the botan-cal was commonly used, processed animal products wereecond commonly used and processed minerals were the lessommon. Key data are summarized in Table 1.

raditional herbal medicine used for painanagement

hree major kinds of traditional herbal medicine are usuallysed to control pain

. Botanicals such as Corydalis Rhizoma, LigusticumRhizoma, Libanotus, Myrrha, Cynanchum PaniculatumRadix, Clematis Radix, Aconitum Radix, Aconitum Kusne-zoffi Radix, Paeoniae Alba Radix, Carthami Flos, PaeoniaeRubra Radix, and Angelicae Pubescentis Radix.

. Processed animal products such as Venenum Bufonis,Scorpion, Scolopendra, Eupolyphaga Seu Steleophag, andLumbricus.

. Processed minerals such as Borneolum Syntheticum andRealgar.

Most of these are known in THM to relieve pain by pro-oting blood flow and qi circulation.

AS improvement by number of persons

n analysis of improvement on the VAS based onine13,15,17,18,20,28,30,32,33 of the 23 controlled trials revealedhat a total of 495 (81.8%) patients in the treatment groupn = 605) and 169 (48.7%) members of the control group

268

J.-W.

Lee et

al.

Table 1 Characteristics of randomized controlled trials (RCTs) on the use of traditional herbal medicine (THM) for cancer pain.

Reference Type of cancer pain Participants(n)/mean age(SD)/male(%)/country

Comparisontreatment vscontrol/treatmentfrequency (treatmentperiod)

Drugs Outcome reported Qualityassessmenta

Jeong et al.11 Breast cancer, gastriccancer, lung cancercolorectal cancer, othercancer

40/52.6(11.4)/37.5/Korea

Herbs vs none/3 per day (2weeks)

None VAS p < 0.05 prefer treatment Y-U-N-U-Y-Y

Tian et al.12 Non-small cell lungcancer

60/over 60 were35%/50/China

Herbs vsherbs + chemotherapy vschemotherapy/herbs: 2 perday (2 moths);chemotherapy: vinorelbineor gemcitabine 2 permonth, cisplatin 1 permonth (2 months)

Vinorelbine pluscisplatin,gemcitabine pluscisplatin

No significant difference U-U-N-U-Y-Y

Chen et al.13 Lung, liver, breast,gastric, pancreatic,esophageal, colorectalcancer

50/Treat: 54.6(11.35), Control:53.2(10.28)/66/China

Western medicine + herbs vswestern medicine/herbs: 3per day (2 weeks); westernmedicine: 2—3 per day (2weeks)

Indomethacin,tramal, morphine

VRS: p < 0.05 prefer treatmentVAS: p < 0.05 prefer treatmentKarnofsky: p > 0.05 nosignificant difference

U-U-N-N-Y-Y

Wu et al.14 Lung, liver, colorectal,stomach, cervical,uterine, kidney,leukemia, prostate, ENT,breast, pancreas,esophagus, bladder,others cancer

2466/40—79 were87%/50.7/China

Herbs + Taiwan herbal tonicvs hospital meals vs Taiwanherbal tonic/3 per day (1week)

None Hospital meals group, Taiwanherbal tonic group: painp < 0.05 significantly reducedherbs + Taiwan herbal tonicgroup: pain p < 0.01significantlyreduced. 10 days aftertreatment herbs + Taiwanherbal tonic group, hospitalmeals group: pain p < 0.01reduced, Taiwan herbal tonicgroup: pain p < 0.05 reduced.

U-U-N-U-Y-Y

Lin et al.15 Esophageal, gastric,liver, lung, breast, rectalcancer

60/Treat: 57.2(14.6), Control: 55.8(15.7)/61.7/China

Western medicine + herbs vswestern medicine/4 per day(1 month)

Aspirin, tramaltablet, pethidine

Analgesic effect p < 0.05 prefertreatment duration ofanalgesia p < 0.01 prefertreatment Side effect p < 0.01prefer treatment

U-U-N-N-Y-Y

Shi et al.16 Liver, lung, gastriccancer

180/Treat: 48.9(15.5), Control: 46.7(14.8)/65/China

Herbs vs westernmedicine/2 per day (2weeks)

Tramal Time to pain relief p < 0.01prefer treatment

U-U-U-U-Y-Y

Traditional herbal

medicine

for cancer

pain

269

Zhang et al.17 Stomach, liver, lung,breast, colon,pancreatic, andsubmaxillary glandcancer (most metastatic)

110/62.4/NA/China Herbs + western medicine vswestern medicine/herbs: 3per day; western medicine:2 per day (3 weeks)

Indomethacin,pethidine

Effect of pain relief p > 0.05 nosignificant difference

U-U-U-U-Y-Y

Cao et al.18 Bone cancer 82/Treat: 59.5,Control:57.4/62.2/China

Western medicine + herbs vswestern medicine/4 per day(1 month)

Aspirin, tramaltablet, pethidine

Experimental group’spercentage of reduction ofpain 97.56%, control group’spercentage of reduction ofpain 92.68% p < 0.05 prefertreatment.

U-U-N-N-Y-Y

Tian et al.19 Primary hepaticcarcinoma

97/Treat: 51.4(10.5), Control: 52.4(10.8)/83.5/China

Herbs + HACE VSHACE + chemotherapy + westernmedicine/herbs: 1 per day(4 weeks)

MMC, THP, 5-Fu Pain p < 0.05 significantdifference

U-U-N-Y-Y-Y

Lin et al.20 Primary hepatocellularcarcinoma

72/Treat: 51.3 (4.5)Control: 49.1(5.3)/66.7/China

Microwave + herbs vsmicrowave/herbs: 3 per day(1 month), microwave: 1per week (2 weeks)

None Liver pain p < 0.01 prefertreatment

Y-U-N-U-Y-Y

Li et al.21 Liver, colon, head,pancreas, prostate,stomach, ovarian,gallbladder, renal,bladder cancer

84/46—84/71.4/China Herbs vs westernmedicine/3 per day

Indomethacin Effect of pain relief, durationof pain relief p > 0.05 nosignificant difference

U-U-U-U-Y-Y

Wan et al.22 Cancer 65/Treat: 59.6,Control:59.7/52.3/China

Herbs + acupuncture vswestern medicine/easternmedicine: 1 per day (1week), western medicine:i.m. 3 per day (1 week)

Bucinnazine Percentage of reduction ofpain no difference. But painrelief p < 0.05 significantdifference effect about strongpain p < 0.05 prefer treatment.Time to pain relief nodifference. Time to be curedp < 0.05 prefer treatment

U-U-N-U-Y-Y

Zhang et al.23 Lung, gastric, liver,esophagus, largeintestine cancer

84/Treat: 56.2(8.4), Control: 52.7(9.5)/63.1/China

Herbs + opioid analgesics vsopioid analgesics/2 per day(2 weeks)

Morphinehydrochloridesustainedreleasetablets

Frequency of pain, period ofpain relief, time to pain reliefp < 0.05 prefer treatment

Y-U-N-U-Y-Y

270

J.-W.

Lee et

al.

Table 1 (Continued)

Reference Type of cancer pain Participants(n)/mean age(SD)/male(%)/country

Comparisontreatment vscontrol/treatmentfrequency (treatmentperiod)

Drugs Outcome reported Qualityassessmenta

Wu et al.24 Lung, liver, gastric,colon, pancreatic cancer

60/Treat: 58.2(7.3), Control: 58.9(5.2)/58.3/China

Herbs vs westernmedicine/3 per day (1week)

Diclofenac Experimental group:percentage of pain reduction90%, Control group: percentageof pain reduction 83.3% nosignificant difference. Butdegree of pain reduction,duration of pain, time to painrelief, extension of pain relief,tenderness etc. prefertreatment

U-U-N-U-Y-Y

Liu et al.25 Middle-late stagemalignant tumor

81/30—75/37/China Chemotherapy + herbs vschemother-apy + herbs + moxibustion vschemotherapy/herbs: 2 perday (1 week); moxibustion:10 min per day (1 week)

CE-CAP, PVM, FAM,5-FU + CF, CHOP,CMF, CAF

Chemotherapy group painp < 0.05 significantly reduced.chemotherapy + herbs grouppain p > 0.05 no significantdifference.Herbs + moxibustion group painp < 0.05 significantly reduced.When compared with othergroup Chemotherapy groupp < 0.01, chemotherapy + herbsgroup p < 0.05 prefer treatment

U-U-N-U-Y-Y

Ma et al.26 Gastric cancer 62/Treat: 53.1,Control:52.8/79/China

Herbs vs westernmedicine/herbs: 3 per day(15 days); westernmedicine: 2 per day (15days)

Analgesic drug Effect of pain relief, durationof pain relief, VRS, Karnofskyp > 0.05 no significantdifference

U-U-U-U-Y-Y

Bao et al.27 Lung, liver, intestinal,breast, stomach,pancreas, esophagus,kidney, others cancer

124/Treat: 69,Control:67/39.5/China

Herbs compression + 3 Lanalgesia vs 3 Lanalgesia + westernmedicine/herbscompression: 8—12 h perday

Analgesic drug Time to pain relief significantdifference

Y-U-N-U-Y-Y

Zhang et al.28 Advanced pancreaticcancer

65/Treat: 54.6(6.8), Control: 53.2(7.6)/55.4/China

Herbs + arterialperfusion + chemotherapy vschemotherapy/herbs: 1 perday (2 months); arterialperfusion: 1 per month (2months)

Vinorelbine No significant difference U-U-N-U-Y-Y

Traditional herbal

medicine

for cancer

pain

271

Feng et al.29 Primary liver cancer 80/53.8/68.8/China Dexamethasone vsginsenosides vs dexametha-sone + ginsenosides vsplacebo/2 per day

Dexamethasone Both experimental group p < 0.05prefer treatmentdexamethasone + ginsenosidesgroup: effect about (duration ofsymptoms, low bone mineraldensity) reported

U-U-Y-Y-Y-Y

Pan et al.30 Middle/advanced stageof lung cancer

250/59.6/76.4/China Herbs + westernmedicine + Zhiling capsulevs westernmedicine + Zhiling capsuleVS herbs + Zhiling capsule vsPingxiao capsule/3 per day(2 weeks)

Herbs + western medicine + Zhilingcapsule chest pain p < 0.05, 0.01prefer treatment

U-U-Y-Y-Y-Y

Wei et al.31 Liver, lung, gastriccancer

200/Treat: 55.2(10.1), Control: 54.2(10.4)/60.5/China

Herbs vs westernmedicine/3 per day (5 days)

Paracetamolcodeinephosphate,placebo drug

Effect of pain relief, duration ofpain relief p > 0.05 no significantdifference

U-U-Y-Y-Y-Y

Wang et al.32 Middle-advancedpancreatic cancer

58/Treat: 48.2(5.7), Control: 49.7(4.1)/74.1/China

Radiation therapy + herbs vsradiation therapy/radiation:5 per week (6 weeks);herbs: 2 per day (9 weeks)

None Experimental group: abdominalpain 25 people among 30 peoplecured Control group: abdominalpain 16 people among 28 peoplecured p < 0.05 prefer treatment

U-U-N-U-Y-Y

Pan et al.33 Liver, stomach,esophagus, liver, kidney,breast, nasopharyngeal,colorectal

400/Treat: 55.4(12.6), Control: 57.5(15.1)/66.8/China

Zhiling capsule vs Pingxiaocapsule/3 per day (2 weeks)

Experimental group’s percentageof reduction of pain 83.9%,control group’s percentage ofreduction of pain 11.6% p < 0.001significant difference

U-U-N-U-Y-Y

Chan et al.34 Ovarian cancer 59/Treat: 52.9,Control:51.5/0/China

Chemotherapy + herbs vschemother-apy + placebo/herbs: 2 perday (6 cycles);chemotherapy: 3 weeklycycles (6 cycles)

Placebo,carboplatin,paclitaxel (Taxol)

QOL no difference Y-Y-Y-Y-Y-Y

a (1) Was the allocation sequence adequately generated? (2) Was allocation adequately concealed? (3) Was knowledge of the allocated interventions adequately prevented during thestudy? (4) Were incomplete outcome data adequately addressed? (5) Were reports of the study free of suggestion of selective outcome reporting? (6) Was the study apparently free of otherproblems that could put it at risk of bias? Key: Y, ‘yes’; U, ‘unclear’; N, ‘no’. The English in this document has been checked by at least two professional editors, both native speakers ofEnglish. For a certificate, please see: http://www.textcheck.com/certificate/z9r5ZI.

272 J.-W. Lee et al.

Figure 1 Flow diagram showing the number of studies included and excluded from the systematic review. RCT, randomizedcontrolled trial.

F conv( VAS)

(cbmdspbCb1

V

WiTwoa

Tip

D

Topiaae

T

igure 2 Meta-analysis of combination cancer therapy versusTHM) combined with standard therapy on visual analog scale (

n = 347) reported a complete or partial response. The per-entage (%) of VAS decrease was evaluated by: (differenceetween VAS before and after treatment/VAS before treat-ent) × 100. Complete response and partial response wasefined as scoring 91—100% and 61—90.9%. The resultshowed that concomitant THM therapy was significantly andositively correlated with VAS improvement in terms of num-er of persons reporting positive results (OR, 1.60; 95%I, 1.05—2.43). The randomized-effects model was usedecause of the inter-trial heterogeneity of the results (�2,72.01 with 8 df; p < 0.00001) (Fig. 2).

AS improvement by degree

e identified four trials12,14,29,34 (1696 patients) reportingmprovement on the VAS in terms of degree of change.

he pooled analysis showed that THM in combinationith Western therapy significantly increased the degreef improvement on the VAS compared with Western ther-py alone (MD, −0.51; 95% CI, −0.94 to −0.08; p = 0.02).

p

tn

entional cancer therapy; effect of traditional herbal medicineby number of persons.

he randomized-effects model was used because of thenter-trial heterogeneity of the results (�2, 10.86 with 3 df;

= 0.01) (Fig. 3).

iscussion

his systematic review of 24 RCTs investigated the efficacyf treatment with both THM and conventional therapy foratients with cancer. This is the first systematic reviewncluding only RCTs that investigated the effect of THM asdjunct therapy for cancer pain. We conducted this system-tic review after systematically reviewing the work of Lingt al.6 who included both non-RCTs and RCTs.

This systematic review investigated the effect of adjunctHM therapy on a variety of cancers. Significant evidence has

reviously demonstrated the effect of THM in this regard.

The treatment group differed significantly from the con-rol group when improvement on the VAS was measured byumber of persons reporting decreased pain; this was not

Traditional herbal medicine for cancer pain 273

convAS)

C

N

A

Tt[

R

Figure 3 Meta-analysis of combination cancer therapy versus(THM) combined with standard therapy on visual analog scale (V

clear in the review conducted by Ling et al.6 as it was not ameta-analysis. A similar result was found when improvementon the VAS was measured by degree. We have not added thefunnel plot in this review because the meta-analyzed RCTswere not enough to exclude potential publication bias.

We reviewed two articles that investigated the effectof THM on the number of pain occurrences/day in patientswith cancer. Although the results were positive, the num-ber of trials was too small to draw firm conclusions. Thus,these results should be interpreted with caution, and furtherinvestigation is necessary.

This systematic review had several limitations. First, thequality of the studies was not assessed completely. Weused the ‘‘Risk-of-Bias’’ assessment tool included in theCochrane Handbook10 to assess the quality of the stud-ies included in this systematic review. As it is difficult tojustify the subtle differences in each item using a quali-tative method,36 we tried to assess each trial with regardto seven critical domains: randomization, allocation con-cealment, blindness of participants and personnel, blindnessof outcome assessment, reporting of incomplete outcomedata, selective-outcome reporting, and other biases. How-ever, the risk-of-bias assessment may not be entirelyobjective.37

Second, of 24 articles, only nine11,12,14,19,20,22,23,27,34 werewritten in English. The remaining 1513,15—18,21,24—26,28—33 werewritten in Chinese, making it difficult for other researchersto follow up on the results of these studies.

The quality of the trials in this systematic reviewwas generally weak; thus, further high-quality trials areneeded to assess the effectiveness of THM for patientswith cancer pain. And as treatment regimens varied inthe studies included in the review this is an importantlimitation.

Conclusions

In conclusion, this systematic review demonstrated the sig-nificant effect of THM treatment combined with standardtherapy versus standard therapy alone, as the former wasassociated with increased improvement on the VAS as mea-

sured by number of persons reporting improvement, degreeof improvement, and number of pain occurrences per day.Additionally, more RCTs should be conducted to determinethe role of THM in a variety of cancer therapies.

1

entional cancer therapy; effect of traditional herbal medicineby degree.

onflict of interest statement

o authors have any conflict of interest to declare.

cknowledgements

his work was supported by the National Research Founda-ion of Korea (NRF) grant funded by the Korea governmentMEST] (No. 2012-0005755).

eferences

1. van den Beuken-van Everdingen MH, de Rijke JM, Kessels AG,Schouten HC, van Kleef M, Patijn J. Quality of life and non-pain symptoms in patients with cancer. J Pain Symptom Manage2009;38(2):216—33.

2. Eisenberg DM, Davis RB, Ettner SL, Appel S, Wilkey S, VanRompay M. Trends in alternative medicine use in the UnitedStates, 1990—1997: results of a follow-up national survey. JAMA1998;280(18):1569—75.

3. Rao JK, Mihaliak K, Kroenke K, Bradley J, Tierney WM,Weinberger M. Use of complementary therapies for arthri-tis among patients of rheumatologists. Ann Intern Med1999;131(6):409—16.

4. Gozum S, Tezel A, Koc M. Complementary alternative treat-ments used by patients with cancer in eastern Turkey. CancerNurs 2003;26(3):230—6.

5. Richardson MA, Sanders T, Palmer JL, Greisinger A, SingletarySE. Complementary/alternative medicine use in a comprehen-sive cancer center and the implications for oncology. J ClinOncol 2000;18(13):2505—14.

6. Ling X, Lao LX, Ge A, Yu S, Li J, Mansky PJ. Chinese herbalmedicine for cancer pain. Integr Cancer Ther 2007;6(3):208—34.

7. Schuler MS, Durdak C, Hol NM, Klink A, Hauer KA, Oster P,et al. Acupuncture treatment of geriatric patients with ischemicstroke: a randomized, double-controlled, single-blind study. JAm Geriatr Soc 2005;53(3):549—50.

8. Madsen MV, Gotzsche PC, Hrobjartsson A. Acupuncture treat-ment for pain: systematic review of randomised clinical trialswith acupuncture, placebo acupuncture, and no acupunc-ture groups. BMJ 2009;338:a3115, http://dx.doi.org/10.1136/bmj.a3115.

9. Hrobjartsson A, Gotzsche PC. Placebo interventions for all clini-cal conditions. Cochrane Database Syst Rev 2010;(1). CD003974.

0. Higgins J, Green S, editors. Cochrane handbook for systematicreviews of interventions. The cochrane Collaboration; 2008.

2

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

74

1. Jeong JS, Ryu BH, Kim JS, Park JW, Choi WC, Yoon SW.Bojungikki-tang for cancer-related fatigue: a pilot randomizedclinical trial. Integr Cancer Ther 2010;9(4):331—8.

2. Tian J, Liu L, Shi Z, Zhou Z, Wang L. A randomized controlledpilot trial of Feiji recipe on quality of life of non-small cell lungcancer patients. Am J Chin Med 2010;38(1):15—25.

3. Chen GY, Liu YQ, Gao P, Tai YH. Clinical study of Zhongyaozhi-tong capsule on 25 cases with cancer pain. Jiangsu J Tradit ChinMed 2005;26(1):16—7.

4. Wu TH, Chiu TY, Tsai JS, Chen CY, Chen LC, Yang LL.Effectiveness of Taiwanese traditional herbal diet for pain man-agement in terminal cancer patients. Asia Pac J Clin Nutr2008;17(1):17—22.

5. Lin D, Li F, Chen LS, Lin HS, Chen YB. Treatment of carcinoma-tous pain by Jiawei Niantong capsule: a clinical observation of30 cases. J New Chin Med 2001;33(7):18—9.

6. Shi HC, Liu XY, Shi KG, Huang ZG. Clinical observation on thetreatment of 90 cases of pain caused by malignant tumor withAitongning. Guiding J Tradit Chin Med Pharm 2005;11(2):21—2.

7. Zhang M. Observation of therapeutic effect of ‘‘CompoundStrychnos Capsules’’ in treating cancerous. Acta Univ TraditMed Sinensis Pharmacol Shanghai 2001;15(2):31—2.

8. Cao JX, Xu GY. Clinical observation on treatment of 41 casesof pain induced to bone metastasis cancer with Zhuangguzhi-tongsan and western medicine. Chin J Tradit Med Sci Technol2006;13(1):48—9.

9. Tian H, Li H, Wang B, Liang G, Huang X, Huang Z, et al.Treatment of middle/late stage primary hepatic carcinomaby Chinese medicine comprehensive therapy: a prospectiverandomized controlled study. Chin J Integr Med 2010;16(2):102—8.

0. Lin J, Jin C, Zheng M, Ouyang X, Zeng J, Dai X. Clinical studyon treatment of primary hepatocellular carcinoma by shenqimixture combined with microwave coagulation. Chin J IntegrMed 2005;11(2):104—10.

1. Li HL, Hao MA, Wang BT, Bai L. The analgesic effect of traditionalTibetan herb Lamiophlomis rotata on cancer pain. Hebei Med J2002;26(2):146—7.

2. Wan D, Li P. Study on Tongkuaixiao Babu plaster in treatingcancer pain. Chin J Integr Med 2004;10(3):175—80.

3. Zhang T, Ma S, Xie G, Deng Q, Tang Z, Pan X, et al. Clini-cal research on nourishing yin and unblocking meridians recipecombined with opioid analgesics in cancer pain management.

Chin J Integr Med 2006;12(3):180—4.

4. Wu MH, Zhou XP, Cheng HB. Clinical study on aitongping capsulein treating cancerous pain. Zhongguo Zhong Xi Yi Jie He Za Zhi2005;(3):218—21.

3

J.-W. Lee et al.

5. Liu J, Yu RC, Rao XQ. Study on effect of moxibustion and gubenyiliu III combined with chemotherapy in treating middle-latestage malignant tumor. Zhongguo Zhong Xi Yi Jie He Za Zhi2001;21(4):262—4.

6. Ma TX, Li XL, Song ZH, She K. Clinical study about the achecaused by stomach tumor treated with Jiaweibaoankeli. Chin JCancer Prev Treat 2003;10(3):297—9.

7. Bao YJ, Hua BJ, Hou W, Lin HS, Zhang XB, Yang GX. Alleviationof cancerous pain by external compress with Xiaozheng Zhitongpaste. Chin J Integr Med 2010;16(4):309—14.

8. Zhang Q, Wang X-M, Chi H-C. Effects of Guben Yiliu II combinedwith arterial perfusion with chemotherapeutic agent in treatingadvanced pancreatic cancer. Zhongguo Zhong Xi Yi Jie He Za Zhi2007;27(5):400—3.

9. Feng Y-L, Ling C-Q, Li B. Clinical study on integrativemedicine for preventing and treating post-transcatheter arte-rial chemoembolization. Zhongguo Zhong Xi Yi Jie He Za Zhi2005;25(6):534—6.

0. Pan YL, Pan YZ. Observation on the effect of zhiling capsule intreating patients with lung cancer of middle/advanced stage.Zhongguo Zhong Xi Yi Jie He Za Zhi 2006;26(7):604—7.

1. Wei L, Yang CG, Miao WH. Observation of the efficacy and safetyof Tian chan capsule in the treatment of cancer pains. Chin JNew Drug 2003;12(8):663—5.

2. Wang B, Liu X, Wu Z. Effect of qi replenishing and bloodcirculation activating drugs in treatment of middle-advancedpancreatic cancer with radio- and chemotherapy. ZhongguoZhong Xi Yi Jie He Za Zhi 2000;20(10):736—8.

3. Pan Y-L, Pan Y-Z. Observation on effects of Zhiling cap-sule in treating 400 patients with various types of cancerin mid-advanced stage. Zhongguo Zhong Xi Yi Jie He Za Zhi2007;27(9):807—9.

4. Chan KKL, Yao TJ, Jones B, Zhao JF, Ma FK, Leung CY, et al.The use of Chinese herbal medicine to improve quality of life inwomen undergoing chemotherapy for ovarian cancer: a double-blind placebo-controlled randomized trial with immunologicalmonitoring. Ann Oncol 2011;22(10):2241—9.

5. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF.Improving the quality of reports of meta-analyses of randomisedcontrolled trials: the QUOROM statement. QUOROM Group. BrJ Surg 2000;87(11):1448—54.

6. Moher D, Jadad AR, Nichol G, Penman M, Tugwell P, WalshS. Assessing the quality of randomized controlled trials: an

annotated bibliography of scales and checklists. Control ClinTrials 1995;16(1):62—73.

7. Ernst E, Resch KL. Reviewer bias: a blinded experimental study.J Lab Clin Med 1994;124(2):178—82.