Keywords
benign prostatic hyperplasia - acupuncture - meta-analysis - systematic review
Introduction
Benign prostatic hyperplasia (BPH) is one of the common urological diseases in middle-aged
and elderly males.[1] Studies have shown that the pathogenesis of BPH is closely related to abnormal hormone
levels, inflammatory reactions, and prostatic interstitial cell proliferation.[2] In recent decades, increased modifiable risk factors, such as metabolic diseases
and obesity, have led to an increased incidence of BPH.[3]
[4]
[5] In 2010, there were more than 210 million BPH patients worldwide, accounting for
6.05% of the total male population.[6] In the United States, the incidence of BPH increases with age, increasing by 10%
for every 10 years of age, and by approximately 80% for those older than 80 years.[7]
[8] Clinically, BPH often presents with lower urinary tract symptoms of varying degrees,
including frequent urination, urgent urination, increased nocturia, incomplete urination,
and dysuria. If timely and effective treatment is not available, it may cause complications
such as urinary retention, bladder stones, erectile dysfunction, and even renal insufficiency,
seriously affecting the quality of life (QOL) of patients.[9]
[10]
[11] Current therapies can first be divided into medical or surgical intervention. Medical
therapy for BPH includes 5-α-reductase inhibitors and α-blockers, or a combination
of both. Surgical interventions include a conventional transurethral resection of
the prostate (TURP), as well as newer modalities such as bipolar TURP, holmium laser
enucleation of the prostate (HoLEP), Greenlight and thulium laser, and prostatic urethral
lift (PUL).[12]
[13] These treatments are prone to cause dizziness, headache, fatigue, drowsiness, orthostatic
hypotension, abnormal ejaculation, bladder sphincter injury, and other adverse reactions
(ARs).[14] Therefore, more and more urologists are noticing that complementary and alternative
therapy may be an effective and personalized treatment option for some patients with
drug and surgical intolerance.[15]
Complementary and alternative medicine is an option for reducing symptoms and enhancing
efficiency, including acupuncture, music therapy, acupoint massage, homeopathy, traditional
Chinese medicine (TCM), and exercise therapy.[16]
[17] As an external method of TCM, acupuncture therapy has been used in treatment of
a variety of diseases.[18] Acupuncture therapy includes manual acupuncture (MA), electroacupuncture (EA) and
others, of which, both MA and EA have been proposed to treat BPH, but to date, evidence
for immediate effects on BPH is limited. Therefore, a systematic study of randomized
controlled trials (RCTs) at home and abroad was conducted in this study to further
evaluate the efficacy of acupuncture treatment for BPH, and to provide certain reference
significance for the application of acupuncture for BPH in subsequent clinical work.
Methods
Inclusion and Exclusion Criteria
The inclusion criteria were as follows: (a) RCTs with regard to the effectiveness
of acupuncture therapy for BPH; (b) the age of the research subjects and the source
of the cases were not limited; (c) in the experimental groups (EGs), EA or acupuncture
was adopted, with or without Western medication (WM) treatment. In the control groups
(CGs), WM was used independently; (d) EGs and CGs provided complete data of treated
patients; (e) the primary outcome indicators that we considered were therapeutic effects
(TEs), International Prostate Score Scale (IPSS), maximum urinary flow rate (Qmax),
post-void residual (PVR), and QOL; and (f) the secondary outcome indicators were prostate
volume (PV) and ARs.
The following situations were excluded: (a) Reviews, conferences, case reports, animal
studies, letters, editorials and qualitative studies; (b) articles with incomplete
data and without full-text access; (c) duplicate publication; (d) EGs contained other
TCM therapies; and (e) CGs referring to the other therapies involved except for WM.
Search Strategy and Data Extraction
Articles search was reported through February 1, 2022, including the following online
databases, that is, PubMed, EMBASE, Cochrane Library, Chinese Biomedical Literature
Database, Chinese National Knowledge Infrastructure, Wanfang Data, and China Science
and Technology Journal Database. Thirteen search components were applied: “benign
prostatic hyperplasia,” “prostatic hyperplasia,” “acupuncture,” “electroacupuncture,”
“acupuncture therapy,” “scalp acupuncture,” “ear acupuncture,” “abdominal acupuncture,”
“filiform needle,” “acupoints,” “manual acupuncture,” “randomized controlled trial,”
and “clinical trial.”
The reference lists of selected articles were also manually filtered for possible
relevant new studies. The language was restricted to English and Chinese. Two researchers
independently screened all the retrieved titles and abstracts and then obtained the
full text of all potentially eligible articles. Two researchers independently examined
these full-text articles in accordance with the inclusion criteria and selected eligible
studies for inclusion in the review. The third party was consulted to determine if
there was a divergence. The self-made data extraction table was used for data extraction.
The extracted contents mainly included the author, publication year, sample size,
age, intervention, course of treatment, dropout, acupuncture points, randomization
methods, follow-up periods, quality of methods, and ARs.
Methodological Quality Evaluation
Two researchers independently evaluated each article using the RCT bias risk assessment
tool recommended by Cochrane Handbook 5.1.0.[19]
[20] Once there were differences, we discussed with each other or negotiated with a third
party to resolve them. The evaluation contents included random allocation method,
hidden allocation scheme, blind method, the integrity of result data, selective reporting,
and other biases. According to the research results, the above six items were judged
as “low risk” “high risk” and “unclear risk”.
Statistical Analysis
A meta-analysis was performed using Review Manager version 5.3. Subgroup analysis
was performed according to possible heterogeneity factors among studies. χ
2 tests (P < 0.05) and I
2 statistics of heterogeneity were used for analysis. If the P > 0.1 and the I
2 < 50%, the test of homogeneity was statistically significant, and then a fixed-effects
model was adopted. Otherwise, a random-effects model was adopted, and sensitivity
and subgroup analysis were performed to explore heterogeneity. Dichotomous variables
were expressed as relative risk (RR) and continuous variables as mean difference (MD). Both were expressed by effect value and 95% confidence interval (CI). If the study was not suitable for meta-analysis, descriptive analysis was performed.
If ≥10 articles were included in an outcome index, the publication bias was evaluated
by funnel plots.
Results
Study Selection
The flow diagram of article retrieval process and results are depicted in [Fig. 1]. A total of 716 related articles were acquired from initial retrieval. The remaining
339 articles were individually screened when duplicate articles were removed. After
further reading article titles and abstracts, 75 full-text articles were assessed
for eligibility. Then, this assessment further excluded 53 studies because there were
no CG (n = 9), no data for extraction (n = 4), no accordance with the inclusion criteria (n = 12), and was case report or duplication (n = 6). Eventually, 22 studies[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42] were selected to be incorporated into the meta-analysis.
Fig. 1 PRISMA 2020 flow diagram.
Study Characteristics
[Table 1] summarizes the main characteristics of the included studies. A total of 1,765 patients
with BPH [889 (50.4%) in EGs and 876 (49.6%) in CGs] were included in this study.
All participants were Chinese and studies were published between 2005 and 2021. All
studies reported comparable or no significant differences between intervention groups
with respect to general information such as age and course of disease. EGs in the
included studies utilized MA[21]
[22]
[24]
[25]
[26]
[28]
[29]
[32]
[33]
[34]
[40]
[41]
[42] or EA[23]
[27]
[30]
[31]
[35]
[36]
[37]
[38]
[39] as intervening measurse; seven studies mixed MA with WM,[24]
[26]
[29]
[34]
[40]
[41]
[42] three studies mixed EA with CWM.[27]
[30]
[31] CGs were WM as intervening measures, including the following drugs: tamsulosin hydrochloride
sustained release capsules (THSRC),[21]
[22]
[23]
[27]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[39]
[40]
[41] terazosin hydrochloride capsules (THC),[25]
[37] finasteride tablets (FT),[24]
[26]
[28]
[38]
[42] doxazosin mesylate extended-release tablets (DMERT).[38]
[42] The duration of treatment ranged from 10 days to 12 weeks. The intervention in 2
studies lasted <4 weeks[30]
[42] and those in the remaining 20 studies lasted ≥4 weeks. Only one study had follow-up
records.[34] Seventeen studies evaluated TEs,[22]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[32]
[33]
[34]
[35]
[36]
[38]
[39]
[41]
[42] 19 studies reported IPSS,[21]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[31]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42] 15 studies reported Qmax,[21]
[23]
[25]
[27]
[28]
[31]
[32]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41] 12 studies presented PVR,[21]
[23]
[24]
[25]
[27]
[29]
[31]
[32]
[38]
[39]
[40]
[41] 14 studies tested QOL,[23]
[25]
[26]
[27]
[29]
[31]
[33]
[34]
[35]
[36]
[37]
[38]
[40]
[41] 12 studies assessed PV,[24]
[25]
[28]
[29]
[31]
[34]
[35]
[37]
[38]
[39]
[40]
[41] and 5 studies mentioned AR.[22]
[23]
[33]
[37]
[39]
Table 1
Characteristics of included studies
|
Study
|
Number of participants
|
Finished number
|
Mean age (years)
|
Course of disease
|
Intervening measure
|
Stimulus parameters
|
|
E/C
|
E/C
|
E
|
C
|
E
|
C
|
E
|
C
|
Stimulus
|
Time (minutes)
|
Frequency (n/day)
|
Relevant outcomes
|
|
Wang (2021)[22]
|
30/30
|
30/30
|
65.6 ± 11.2
|
59.3 ± 11.2
|
18.3 ± 6.1(m)
|
21.7 ± 5.3 (m)
|
MA, 4 w, 6 times a week
|
THSRC, 0.2 mg, qn
|
Acupuncture manipulation
|
30
|
1
|
①⑦
|
|
Hu et al. (2020)[34]
|
33/34
|
30/30
|
61.3 ± 5.1
|
61.7 ± 4.9
|
6.1 ± 2.1 (y)
|
6.3 ± 2.1 (y)
|
MA + THSRC, 4 w, qod
|
THSRC, 0.2 mg, qn
|
Acupuncture manipulation
|
30
|
1
|
①②③⑤⑥
|
|
Jiang (2020)[24]
|
31/31
|
30/30
|
60.4 ± 6.4
|
59.3 ± 6.6
|
14.9 ± 7.0 (m)
|
16.5 ± 5.2 (m)
|
MA + FT, 12 w, qod
|
FT, 5 mg, qd
|
Acupuncture manipulation
|
30
|
1
|
①②④⑥
|
|
Yuan et al. (2009)[25]
|
35/30
|
35/30
|
64.5
|
1–13 (y)
|
MA, 4 w, qod
|
THC, 2 mg, qn
|
Acupuncture manipulation
|
30
|
1
|
①②③④⑤⑥
|
|
Wang et al. (2020)[26]
|
31/31
|
30/30
|
66.3
|
67
|
1–22 (y)
|
10 (m)–2 0 (y)
|
MA + FT, 6 w, 5 times a week
|
FT, 5 mg, qd
|
Acupuncture manipulation
|
20
|
1
|
①②⑤
|
|
Huang et al. (2016)[40]
|
30/30
|
30/30
|
60 ± 14
|
–
|
MA + THSRC, 4 w, take 3 days off every 10 days
|
THSRC, 0.2 mg, qn
|
Acupuncture manipulation
|
30
|
1
|
②③④⑤⑥
|
|
Zhao et al. (2020)[28]
|
30/30
|
28/28
|
66 ± 6
|
64 ± 6
|
3.9 ± 1.7 (y)
|
3.9 ± 1.9 (y)
|
MA, 12 w, qod
|
FT, 5 mg, qd
|
Acupuncture manipulation
|
30
|
1
|
①②③⑥
|
|
Liu et al. (2021)[29]
|
34/34
|
30/30
|
63.7 ± 6.7
|
62.8 ± 5.9
|
6.3 ± 2.8 (y)
|
6.7 ± 2.6 (y)
|
MA + THSRC, 4 w, qod
|
THSRC, 0.2 mg, qn
|
Acupuncture manipulation
|
30
|
1
|
①②④⑤⑥
|
|
Gou (2017)[41]
|
40/40
|
40/40
|
66.8 ± 7.8
|
67.9 ± 8.0
|
5.2 ± 1.9 (y)
|
5.5 ± 1.8 (y)
|
MA + THSRC, 4 w, 5 times a week
|
THSRC, 0.2 mg, qn
|
Acupuncture manipulation
|
20
|
1
|
①②③④⑤⑥
|
|
Zhou et al. (2018)[42]
|
50/50
|
50/50
|
68.5 ± 5.7
|
68.6 ± 5.5
|
5.4 ± 4.7 (y)
|
5.3 ± 4.7(y)
|
MA+ FT, DMERT, 10 days, bid
|
FT, 5 mg, qd, DMERT, 4 mg, qn
|
Acupuncture manipulation
|
30
|
2
|
①②
|
|
Xu et al. (2020)[32]
|
35/35
|
35/35
|
61.6 ± 5.8
|
62.1 ± 5.4
|
3.0 ± 1.2 (y)
|
3.0 ± 1.1 (y)
|
MA, 4 w, 5 times a week
|
THSRC, 0.2 mg, qn
|
Acupuncture manipulation
|
30
|
1
|
①③④
|
|
Chen et al. (2018)[33]
|
32/32
|
32/32
|
63.3 ± 9.8
|
65.5 ± 7.6
|
54.6 ± 38.6 (m)
|
43.1 ± 31.4 (m)
|
MA, 4 w, biw
|
THSRC, 0.2 mg, qn
|
Acupuncture manipulation
|
20
|
1
|
①②⑤⑦
|
|
Kong et al. (2017)[21]
|
80/80
|
80/80
|
50–75
|
1–12 (y)
|
MA, 4 w, bid
|
THSRC, 0.2 mg, qn
|
Acupuncture manipulation
|
15
|
2
|
②③④
|
|
Du (2020)[35]
|
30/30
|
30/30
|
63.1 ± 4.2
|
64.3 ± 4.1
|
20.3 ± 8.0 (m)
|
20.3 ± 8.6 (m)
|
EA, 4 w, 6 times a week
|
THSRC, 0.2 mg, qn
|
2 Hz
|
30
|
1
|
①②③⑤⑥
|
|
Wang et al. (2018)[36]
|
22/22
|
22/22
|
61.6 ± 6.4
|
62.3 ± 6.2
|
3.2 ± 1.9 (y)
|
3.0 ± 1.9 (y)
|
EA, 4 w, 6 times a week
|
THSRC, 0.2 mg, qn
|
2–10 Hz
|
30
|
1
|
①②③⑤
|
|
Yang et al. (2008)[37]
|
47/46
|
46/45
|
67.4 ± 8.4
|
69.7 ± 8.7
|
4.4 ± 4.0 (y)
|
3.8 ± 3.3 (y)
|
EA, 4 w, qod
|
THC, 2 mg, qn
|
20 Hz
|
30
|
1
|
②③⑥⑦
|
|
Wang (2010)[38]
|
33/32
|
33/32
|
61.5 ± 5.1
|
62.1 ± 5.5
|
5.6 ± 3.8 (y)
|
5.3 ± 3.4 (y)
|
EA, 4 w, qod
|
FT, 5 mg, qd, DMERT, 4 mg, qn
|
20 Hz
|
30
|
1
|
①②③④⑤⑥
|
|
Yu (2005)[39]
|
50/44
|
48/44
|
—
|
—
|
EA, 4 w, 6 times a week
|
THSRC, 0.2 mg, qn
|
Patient tolerance
|
20–30
|
1
|
①②③④⑥⑦
|
|
Zheng et al. (2017)[23]
|
30/30
|
26/24
|
65.4 ± 10.0
|
68.0 ± 8.4
|
110.3 ± 102.8 (m)
|
85.6 ± 61.6 (m)
|
EA, 6 w, 5 times a week
|
THSRC, 0.2 mg, qn
|
5 Hz
|
20
|
1
|
②③④⑤⑥⑦
|
|
Zhang et al. (2011)[27]
|
42/42
|
42/42
|
61 ± 9
|
5.6 ± 2.4 (y)
|
5.5 ± 2.3 (y)
|
EA + THSRC, 4 w, 6 times a week
|
THSRC, 0.2 mg, qn
|
Patient tolerance
|
30
|
1
|
①②③④⑤
|
|
Liu (2013)[30]
|
96/96
|
96/96
|
56.5
|
5.4 (y)
|
EA + THSRC, 14 days, qd
|
THSRC, 0.2 mg, qn
|
100 Hz
|
30
|
1
|
①
|
|
Wu (2020)[31]
|
48/47
|
48/47
|
67.6 ± 13.2
|
66.8 ± 12.0
|
9.6 ± 6.5 (y)
|
9.1 ± 6.4 (y)
|
EA + THSRC, 4 w, 5 times a week
|
THSRC, 0.2 mg, qn
|
Patient tolerance
|
20
|
1
|
②③④⑤⑥
|
Abbreviations: bid, bis in die; biw, bid times a week; C, control group; DMERT, doxazosin
mesylate extended-release tablet; E, experimental group; EA, electroacupuncture; FT,
finasteride tablet; MA, manual acupuncture; min, minutes; qd, quaque die; qn, quaque
nocte; qod, quaque omni die; THC, terazosin hydrochloride capsule; THSRC, tamsulosin
hydrochloride sustained release capsule; w, week.
Notes: ① Therapeutic effects. ② International Prostate Score Scale. ③ Maximum urinary
flow rate. ④ Postvoid residual urine volume. ⑤ Quality of life. ⑥ Prostate volume.
⑦ Adverse reactions; tamsulosin hydrochloride sustained release capsule.
Bias Risk Assessment
All studies were randomized controlled studies, 10 studies[21]
[25]
[28]
[30]
[32]
[34]
[38]
[39]
[40]
[41] did not introduce specific randomization methods, and 12 studies were grouped by
low-risk random number table. None of the studies mentioned the allocation of hidden
information, nor did they describe the blind method of patients and evaluators. Of
all the studies, 14[21]
[22]
[25]
[27]
[30]
[31]
[32]
[33]
[35]
[36]
[38]
[40]
[41]
[42] had complete data, and 8 reported missing data and were judged to be at high risk.
Because there was no prior registration of the studies, it was not clear whether there
was reporting bias and other risks. Overall, the methodological quality of the included
studies was generally low, as shown in [Figs. 2] and [3].
Fig. 2 Risk for bias graph.
Fig. 3 Risk for bias summary.
Effects of the Interventions
Therapeutic Effects
Seventeen studies[22]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[32]
[33]
[34]
[35]
[36]
[38]
[39]
[41]
[42] reported TEs, including 1,212 participants. The meta-analysis revealed that the
comparison of TEs between EGs and CGs had statistical significance (RR: 1.23, 95% CI: 1.16, 1.3; P < 0.00001, I
2 = 10%). As there was no heterogeneity among studies, subgroup analysis was not conducted,
as shown in [Fig. 4].
Fig. 4 Forest plot of TEs. TE, therapeutic effect.
International Prostate Score Scale
Nineteen studies[21]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[31]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42] assessed IPSS as an outcome measure, including 1,406 participants. In the meta-analysis,
MA or EA combined with or without WM was superior to WM in terms of reducing IPSS
(MD: −2.06, 95% CI: −3.17, −0.96; P = 0.0002, I
2 = 95%). We provided the subgroup analysis based on interventions to explore the sources
of heterogeneity. Heterogeneity was not reduced when the interventions were divided
into four groups. For MA versus WM, the results of four studies[21]
[25]
[28]
[33] revealed no statistical significance (MD: 0.2, 95% CI: −2.82, 3.22; P = 0.9, I
2 = 96%). Seven studies[24]
[26]
[29]
[34]
[40]
[41]
[42] that compared MA plus WM versus WM (MD: −2.72, 95% CI: −3.66, −1.79; P < 0.00001, I
2 = 85%) demonstrated a lower IPSS in the EGs. In addition, six studies[23]
[35]
[36]
[37]
[38]
[39] involving EA versus WM showed a significant difference (MD: −2.76, 95% CI: −4.19, −1.33; P = 0.0002, I
2 = 75%). Two studies[27]
[31] discovered that EA combined with WM showed a better effect than WM alone (MD: −2.26, 95% CI: −2.83, −1.68; P < 0.00001, I
2 = 0%). Sensitivity analysis of IPSS outcome indicators showed that changing the effect
model had no significant effect on the results. Nevertheless, when we deleted the
study of Kong et al.,[21] the heterogeneity reduced to 84%. We considered that the small sample size of included
studies may have contributed to the significant heterogeneity. The result was relatively
stable, as shown in [Fig. 5].
Fig. 5 Forest plot of IPSS. CI, confidence interval; EA, electroacupuncture; IPSS, International
Prostate Score Scale; MA, manual acupuncture; SD, standard deviation; WM, Western
medicine.
Maximum Urinary Flow Rate
Fifteen studies[21]
[23]
[25]
[27]
[28]
[31]
[32]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41] measured Qmax, including 1,132 participants. There was evidence of a better improvement
in Qmax in EGs than in CGs. The difference was statistically significant (MD: 1.7, 95% CI: 0.89, 2.52; P < 0.0001, I
2 = 92%). There was considerable study heterogeneity. Consequently, we conducted subgroup
analyses to quantify the heterogeneity. Four studies[21]
[25]
[28]
[32] discovered that MA showed a better effect compared with WM (MD: 2.03, 95% CI: 0.14, 3.93; P = 0.04, I
2 = 93%). In addition, studies that compared EA with WM (MD: 1.83, 95% CI: 0.56, 3.09; P = 0.005, I
2 = 90%),[23]
[35]
[36]
[37]
[38]
[39] and EA plus WM with WM (MD: 2.83, 95% CI: 1.02, 4.63; P = 0.002, I
2 = 62%),[27]
[31] demonstrated a higher Qmax rate in the EGs. Merely three studies[34]
[40]
[41] presented MA plus WM versus WM, but no significant difference was found for Qmax
between these two groups (MD: 0.3, 95% CI: −1.05, 1.64; P = 0.67, I
2 = 83%). when we removed the studies of xu et al. and Hu et al.,[32]
[34] the heterogeneity just reduced to 82%. The outcome was steady, as shown in [Fig. 6].
Fig. 6 Forest plot of Qmax. CI, confidence interval; EA, electroacupuncture; IPSS, International
Prostate Score Scale; MA, manual acupuncture; Qmax, SD, standard deviation; WM, Western
medication.
Post-void Residual
Twelve studies[21]
[23]
[24]
[25]
[27]
[29]
[31]
[32]
[38]
[39]
[40]
[41] were included in the analysis of the reduction in PVR, including 1,032 participants.
The meta-analysis results demonstrated marked divergence between the two groups in
reducing PVR (MD: −8.25, 95% CI: −12.14, −4.36; P < 0.0001, I
2 = 95%). With obvious heterogeneity being seen, subgroup analyses were run for analysis.
For MA versus WM, three studies[21]
[25]
[32] presented no remarkable difference (MD: −8.89, 95% CI: −18.33, 0.54; P = 0.06, I
2 = 99%). For MA plus WM versus WM, four studies[24]
[29]
[40]
[41] indicated a better effect that favor MA plus WM (MD: −7.67, 95% CI: −13.62, −1.71; P = 0.01, I
2 = 65%). For EA versus WM, three studies[36]
[38]
[39] exhibited no marked divergence between the two groups in reducing PVR (MD: −4.75, 95% CI: −9.63, 0.12; P = 0.06, I
2 = 65%). The remaining two studies[27]
[31] revealed that PVR in the EA plus WM group was less than WM group, and the variance
between both groups was statistically significant (MD: −14.42, 95% CI: −21.29, −7.55; P < 0.0001, I
2 = 80%). By excluding the studies one by one, we found that after removing Xu et al.,[32] the heterogeneity dropped from 95% to 81%. There was no significant change, as shown
in [Fig. 7].
Fig. 7 Forest plot of PVR. CI, confidence interval; EA, electroacupuncture; IPSS, International
Prostate Score Scale; MA, manual acupuncture; PVR, postvoid residual urine volume;
SD, standard deviation; WM, Western medication.
Quality of Life
Fourteen studies[23]
[25]
[26]
[27]
[29]
[31]
[33]
[34]
[35]
[36]
[37]
[38]
[40]
[41] tested QOL, including 847 participants. The evidence of a reduction in QOL score
was statistically significant (MD: −0.55, 95% CI: −0.8, −0.29; P < 0.0001, I
2 = 91%). There was remarkable heterogeneity, so subgroup analysis was used to perform
further analysis. Two studies[25]
[33] involving MA versus WM showed no significant difference (MD: −0.39, 95% CI: −1.4, 0.63; p = 0.46, I
2 = 90%). Five studies[23]
[35]
[36]
[37]
[38] involving EA versus WM also showed no significant difference (MD: −0.66, 95% CI: −1.37, 0.05; p = 0.07, I
2 = 82%). There were statistical differences in the MA combined with WM groups to WM
using alone in five studies[26]
[29]
[34]
[40]
[41] (MD: −0.57, 95% CI: −1.01, −0.12; P = 0.01, I
2 = 96%). Besides, two studies[27]
[31] that compared EA plus WM with WM also presented marked divergence between the EGs
and CGs. When we removed one study,[26] the heterogeneity was not significantly reduced (81%), suggesting a robust result,
as shown in [Fig. 8].
Fig. 8 Forest plot of QOL. CI, confidence interval; EA, electroacupuncture; IPSS, International
Prostate Score Scale; MA, manual acupuncture; QOL, quality of life; SD, standard deviation;
WM, Western medication.
Prostate Volume
A total of 814 participants were studied in the included 12 studies[24]
[25]
[28]
[29]
[31]
[34]
[35]
[37]
[38]
[39]
[40]
[41] to measure PV. Pooled data showed no conspicuous difference in reducing PV between
the EGs and the CGs (MD: −0.87, 95% CI: −2.66, 0.92; P = 0.34, I
2 = 84%) with obvious heterogeneity. The heterogeneity among the subgroups of MA versus
WM (MD: −0.61, 95% CI: −1.43, 0.22; p = 0.15, I
2 = 0%),[25]
[28] MA plus WM versus WM (MD: −0.13, 95% CI: −2.12, 1.86; p = 0.9, I
2 = 24%),[24]
[29]
[34]
[40]
[41] and EA versus WM (MD: 0.42, 95% CI: −4.12, 4.96; p = 0.86, I
2 = 87%) were not significant. EA plus WM versus WM (MD: −6.42, 95% CI: −8.75, −4.09; P < 0.00001) was included in only one study, so they could not be compared with other
studies.[31] When we removed Du,[35] the heterogeneity was only reduced to 76%, and the results were stable, as shown
in [Fig. 9].
Fig. 9 Forest plot of PV. CI, confidence interval; EA, electroacupuncture; IPSS, International
Prostate Score Scale; MA, manual acupuncture; PV, prostate volume; SD, standard deviation;
WM, Western medication.
Adverse Reactions
Five studies[22]
[23]
[33]
[37]
[39] reported mild ARs, the risks of which were no statistically significant difference
between EGs and CGs (RR: 0.73, 95% CI: 0.26, 2.05; P = 0.55, I2
= 29%) with the following incidence details: a total of five cases[23]
[33] with slight dizziness in the CGs; one case[22] with stuck needle, and one case[22] with fainting; two cases[37] with pain, all occurred in the EGs. Another study[39] reported ARs with no details of THSRC side effects.
Publication Bias
A funnel plot was drawn to evaluate publication bias based on the outcome index, the
TEs. The results testified that the distribution of the studies was asymmetric, suggesting
that there might be bias, as shown in [Fig. 10].
Fig. 10 Publication bias of the TEs. RR, relative risk; TE, therapeutic effect.
Discussions
The objective of this review was to summarize and evaluate the clinical efficacy and
safety of acupuncture therapy for BPH through improving TEs, IPSS, Qmax, PVR, QOL,
and PV. Twenty-two studies were included in this meta-analysis, with a total of 1,765
participants. Among the relevant articles included in this study, only MA and EA were
included in the EGs. The data showed significant heterogeneity among MA or EA interventions
and comparators used to evaluate acupuncture therapy for patients with BPH. Studies
comparing MA with WM, MA plus WM with WM, EA with WM, EA plus WM with WM were included
in our review.
Of the 22 original articles included in this review, 12[22]
[23]
[24]
[26]
[27]
[29]
[31]
[33]
[35]
[36]
[37]
[42] were evaluated by the Cochrane Bias risk assessment tool as a low risk in the randomized
method, and the remaining 10 articles did not give specific random allocation methods.
In the eight studies[23]
[24]
[26]
[28]
[29]
[34]
[37]
[39] with shedding cases, no intentional analysis was conducted, which may lead to incomplete
outcome reports and loss of follow-up bias. All the articles did not mention allocation
concealment and the implementation of blinding for researchers, and the particularity
of acupuncture therapy increased the difficulty of blinding for patients, which also
caused potential risks, so the overall quality of the literature was low.
The meta-analysis evaluated the TEs of 16 studies, and the results showed that the
TEs of two groups were statistically significant, without heterogeneity. In terms
of other outcomes of IPSS, Qmax, QOL, and PVR, studies manifested statistically significant
benefits of the EGs compared with CGs. However, pooled data showed no conspicuous
difference in reducing PV between the two groups. Due to the high heterogeneity in
this study, we divided the study into four subgroups according to the intervention
used in the EGs. The results indicated that (a) Qmax was significantly higher with
MA alone than WM, but there was no significant difference in IPSS, PVR, QOL, and PV;
(b) IPSS, PVR, and QOL in the treatment of BPH with MA plus WM were significantly
lower than WM, indicating the statistical significance, but without statistical significance
differences in Qmax and PV; (c) for EA versus WM, the results of studies revealed
statistically significant in IPSS and Qmax, nevertheless, in terms of improving PVR,
QOL, and PV, studies presented no remarkable difference; (d) compared with WM, studies
indicated a better effect in IPSS, Qmax, PVR, QOL, and PV that favor EA plus WM. When
the EGs were classified by the type of acupuncture, the heterogeneity was not significantly
reduced. Furthermore, sensitivity analysis also found no heterogeneous sources, which
limited the reliability of the results. Besides, the safety analysis showed that there
may be a certain adverse event in the treatment of BPH, whether acupuncture therapy
or WM, but within an acceptable range. In summary, acupuncture therapy can be used
as a supplement and alternative therapy for BPH.
In addition, after sorting out the acupuncture points commonly used in clinical acupuncture,
this study found that Zhongji (CV3), Guanyuan (CV4), and Sanyinjiao (SP6) were the
most common in the acupuncture treatment of BPH. The inferior nerve at Zhongji (CV3)
and Guanyuan (CV4) points originates from T12-L1 and contains the pelvic nerve that
innervates the bladder detrusor. The nerves at Sanyinjiao (SP6) point enter the spinal
cord segment L4-S2, overlapping with the segment where the nerves innervating the
bladder enter the spinal cord. The study has also found that acupuncture is applied
to these points which can regulate nerve excitability and effectively improve bladder
compensatory function.[43]
This meta-analysis had certain limitations. (a) All of the included studies were published
in China, and most of the literature was single-center and small-sample trials. (b)
In terms of the treatment course, 10 days to 12 weeks of treatment were included in
the study, which had an impact on the difference in results. (c) The quality of the
hidden allocation and blind implementation of most literature was not high, which
may have some bias, so no definitive conclusions could be drawn. (d) acupoints selection,
acupuncture manipulation, the evaluation criteria of curative effect, and different
types of WM resulted in heterogeneity. (e) The follow-up data to evaluate the long-term
efficacy were insufficient, and more long-term studies were needed.
Clinical studies should aim to perfect blinding methods. Acupuncture site selection
and operation methods should be standardized when possible. Outcome measures should
be observed from multiple perspectives and measured multiple times over a long period
of time to facilitate comprehensive analysis and thus improve the reliability of clinical
decision-making. From the included trials, many acupoints selected by the acupuncture
group required good patient compliance. Using the least acupoints, the least stimulation
and the optimized therapy to achieve the best curative effect will be the content
of future research.
Conclusion
In summary, acupuncture therapy has a certain effect on improving TEs, IPSS, Qmax,
PVR, and QOL in patients with BPH, but the effect on reducing PV is not significant
compared with the CGs. However, considering that the ARs of acupuncture therapy are
slight, and acupuncture therapy does not conflict with other standard treatments,
acupuncture therapy can be considered in specific circumstances in clinical practice.
Therefore, a more rigorous design and a large sample of multicenter randomized controlled
trials are needed to verify the conclusions of this study.
