Prospective Evaluation of Acupuncture

as Treatment for Glaucoma

SIMON K. LAW, STARRIE LOWE, SAMUEL M. LAW, JOANN A. GIACONI, ANNE L. COLEMAN, AND
JOSEPH CAPRIOLI

G

PURPOSE: To evaluate acupuncture as treatment for LAUCOMA MANAGEMENT IN WESTERN MEDICINE
glaucoma. is well established and its efficacy has been

DESIGN: Prospective double-masked randomized cross- demonstrated by a number of randomized clinical
over study. trials.1 However, despite medical and surgical therapies, a

METHODS: SETTING: Clinical practice. POPULATION: significant proportion of glaucoma patients continue to
One eye per patient with primary open-angle glaucoma go blind.2 Alternative therapies to reduce the progression
and stable intraocular pressure (IOP). INTERVENTION: rate of glaucoma have been actively sought, but as yet
Patients were randomized to receive 1 acupuncture series without much success. Acupuncture, a branch of Chinese
(12 sessions with either eye-related [eye-points] or non- traditional medicine used for over 2000 years in the treat-
eye-related [non-eye-points] acupoints) and then crossed ment of various illnesses, is one alternative therapy that has
over to receive the other series. OUTCOME MEASURES: IOP, garnered the attention of many glaucoma patients.3
blood pressure (BP), heart rate (HR), best-corrected Ocular effects associated with acupuncture have been
visual acuity (BCVA), visual field (VF), optic disc and studied in animal models and in small samples of human
peripapillary retinal nerve fiber layer (RNFL) measure- subjects. Some studies report beneficial effects such as
ments, compliance, and adverse reactions. Probability to intraocular pressure (IOP) reduction,4–10 improvement of
detect 3 mm Hg IOP difference between series was 90%. central visual acuity,10,11 and alterations of the visual

RESULTS: Twenty-two patients volunteered and 11 field.12 However, most of the clinical results are inconclu-
(50.0%) completed the study; 8 (36.4%) did not com- sive because of inadequate research design or incompletely
plete treatment owing to changes of health, moving described methodologies.13 In addition, variable locations
away, lack of transportation, or family crisis; and of acupoints and different acupuncture intensities, dura-
3(13.6%) were withdrawn owing to needle sensitivity tions, and frequencies of stimulation were studied. This
or IOP elevation (8 mm Hg) in the contralateral eye. Af- clinical heterogeneity makes scientific analyses or useful
ter an acupuncture session, mean IOP increased slightly comparisons of these acupuncture study results difficult, if
with both eye-points (from 12.9 ± 1.8 mm Hg to 13.6 not impossible.13,14
± 2.0 mm Hg, P [ .019) and non-eye-points (from The purpose of this investigation was to prospectively
13.0 ± 1.5 mm Hg to 13.5 ± 1.7 mm Hg, P [ .073) evaluate the effects of acupuncture as glaucoma treatment.
series. HR, diurnal IOP, and BCVA showed no statisti- In particular, IOP, central visual acuity, visual field status,
cally significant changes after 12 sessions of either series. and optic nerve and retinal nerve fiber structure were stud-
Systolic and diastolic BP were reduced after 12 sessions of ied in patients with primary open-angle glaucoma.
non-eye-points series (P [ .040, P [ .002, respec-
tively). Optic disc, RNFL, and VF showed no statistically
significant changes.

CONCLUSIONS: Acupuncture has no overall effect on METHODS
diurnal IOP or BCVA but may temporally increase the
IOP immediately after a treatment session. BP is lowered THIS WAS A PROSPECTIVE, DOUBLE-MASKED, RANDOMIZED
by acupuncture with non-eye-points, but not with eye- crossover comparative study approved by the Institutional
points. Compliance and adverse event rates were Review Board of the University of California Los Angeles.
low. (Am J Ophthalmol 2015;160(2):256–265. The study was conducted according to the tenets of the
Ó 2015 by Elsevier Inc. All rights reserved.) Declaration of Helsinki and all applicable HIPAA rules.
Informed consent was obtained from subjects prior to
screening and study enrollment. Flyers explaining the na-
ture of the study were available to patients who visited
Accepted for publication Apr 21, 2015. the Glaucoma Division of the Stein Eye Institute. Patients
From the Glaucoma Division, Stein Eye Institute, David Geffen School who expressed interest in the study underwent a screening
of Medicine, University of California Los Angeles, Los Angeles, visit for eligibility. Eligibility criteria were primary open-
California.
Inquiries to Simon K. Law, 100 Stein Plaza #2-235, Stein Eye Institute, angle glaucoma, age 18 years or older, stable intraocular
Los Angeles, CA 90095; e-mail: [email protected] pressures (less than 2 mm Hg variation in the last 2

256 Ó 2015 BY ELSEVIER INC. ALL RIGHTS RESERVED. 0002-9394/$36.00

http://dx.doi.org/10.1016/j.ajo.2015.04.033
FIGURE 1. The 4 eye-related acupuncture points (UB2, GB37, Liv3, LI4) used in the Acupuncture as Treatment for Glaucoma
study.

follow-up visits over a period of at least 3 months and reaction evaluation, (3) central corneal thickness measure-
without alteration of glaucoma therapy for 6 months), no ment, (4) ocular motility evaluation, (5) external and peri-
use of any alternative therapy including cannabis, no his- orbital evaluation, (6) anterior segment examination with
tory of surgical or laser intervention for glaucoma or failed slit-lamp biomicroscopy, (7) gonioscopic evaluation of
interventions within the prior 3 years, no cardiovascular or anterior chamber drainage angle, (8) optic disc evaluation,
neurologic diseases or bleeding disorders, and no retinal and (9) dilated retinal examination. The visual field was
abnormalities or nonglaucomatous optic neuropathy. One tested with achromatic automated perimetry (Humphrey
eye per patient was enrolled; if both eyes qualified for Field Analyzer [HFA] II; Carl Zeiss Meditec, Dublin, Cal-
enrollment, only the right eye was enrolled. Diagnosis of ifornia, USA) with the 24-2 Standard Swedish Interactive
primary open-angle glaucoma was made when glaucoma- Threshold Algorithm (SITA-standard). Optic disc and
tous visual field defects and glaucomatous optic neuropathy RNFL measurements were obtained with confocal scan-
were diagnosed by the treating physician (a glaucoma ning laser ophthalmoscopy (Heidelberg Retinal Tomogra-
specialist) in the setting of an open angle. Secondary glau- phy [HRT]; Heidelberg Engineering, Carlsbad, California,
comas were excluded, including pigmentary dispersion, USA) and spectral-domain OCT (Cirrus SD OCT; Carl
pseudoexfoliation, uveitic, steroid-induced, or traumatic Zeiss Meditec), respectively.
glaucomas. After the baseline examination, patients were scheduled
A separate baseline examination was scheduled after to undergo acupuncture treatment. Acupuncture was
screening to obtain baseline measurements, which performed by an experienced, licensed acupuncturist
included blood pressure (BP) and heart rate (HR), a com- (S.L.). Two groups of acupuncture points were evaluated.
plete ophthalmic examination, visual field testing, optic One group of points was related to the eye or used to treat
disc and peripapillary retinal nerve fiber layer (RNFL) eval- eye diseases according to traditional Chinese acupuncture
uation, and diurnal IOP measurements (measured with theory (eye-points). The 4 eye-points were UB2, GB37,
Goldmann tonometry every 2 hours from 8 AM to 4 Liv3, and LI415,16 (Figure 1). The second group of
PM). A complete ophthalmic examination consisted acupuncture points was not eye related and is traditionally
of (1) Snellen visual acuity measurement, (2) pupillary used to treat nonocular disease (non-eye-points). The 4

VOL. 160, NO. 2 ACUPUNCTURE FOR GLAUCOMA 257

FIGURE 2. The 4 non-eye-related acupuncture points (St36, St38, Sp6, UB60) used in the Acupuncture as Treatment for Glau-
coma study.

non-eye-points were St36, St38, Sp6, and UB60. These acupuncture needles left in place for 20 minutes while
non-eye-points are typically used for pain control, stress the patient rested on a 45-degree-reclined examination
management, fatigue, and treatment of insomnia and diges- chair, without electric or laser stimulation of the needle
tive illnesses15,16 (Figure 2). There is no consensus among or moxibustion. (Moxibustion is the application of heat
acupuncturists on a set of acupuncture points for treating resulting from the burning of herbs, or moxa, to acupoints
any particular ocular disease. Literature on acupuncture to facilitate the acupuncture effect.) The examination door
and Chinese medicine suggests that one cause of ocular dis- was closed to avoid noise or disturbance; no music was
ease is an imbalance of body energy flow along the gall- played.
bladder (GB) and liver (Liv) meridians. Two eye-points Patients were randomly assigned (by flipping a coin) to
chosen in this study (GB37 and Liv3) belong to these me- receive 1 acupuncture series, either the eye-points or
ridians. Eye-points GB37, Liv3, and LI4 were studied in a non-eye-points, for 12 sessions over 6–12 weeks (1–2 ses-
prospective case series of acupuncture treatment in patients sions per week) and then were crossed over to receive the
with glaucoma or ocular hypertension.8 Eye-point UB2, other acupuncture series (12 sessions over 6–12 weeks)
located on the medial end of the eyebrow, belongs to a with a washout period of 4 weeks in between. Glaucoma
group of acupuncture points surrounding the orbit that is medical therapy was not altered at any time during the
believed to have a stronger therapeutic effect in treating study.
ocular diseases than distal acupuncture points.16 Acupunc- Patients and research personnel, except the licensed
ture needles were placed on points on both sides of the acupuncturist, were masked to the acupuncture points
body. Sterile stainless steel disposable acupuncture needles used, the therapeutic effects of the acupuncture points,
(1 inch 36 gauge needles and 1/2 inch 38 gauge needles) and the order of the acupuncture series (whether a patient
were used. A sensation of numbness and mild aching at was receiving the eye-points therapy or the non-eye-points
the acupoint (traditionally termed ‘‘De Qi’’) was elicited therapy).
as an indication of proper needle placement. Each BP, HR, and IOP were measured 15 minutes before and
acupuncture treatment session was conducted in an after each treatment session. Diurnal IOP measurement
assigned examination room with light dimmed and and automated visual field measurements were obtained

258 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2015

 TABLE 1. Demographic Characteristics of Patients Who Completed and Did Not Complete the Acupuncture as Treatment for
Glaucoma Study

Patients who Completed the Study Patients who Did Not Complete the Study

Number of patients 11 11
Mean age, y (range) 63.7 6 14.3 (35.6–87.6) 64.5 6 10.0 (45.0–76.7)
Sex: female/male 8/3 8/3
Race: white/black/Hispanic/Asian 5/1/2/3 10/0/0/1
On systemic beta-blockers None None
Mean central corneal thickness (mm) 528.8 6 39.1 533.1 6 43.4
Lens status: phakic/pseudophakic 10/1 7/4
Previous laser trabeculoplasty 0 0
Previous trabeculectomy 2 1
Previous cataract surgery 1 4

TABLE 2. Blood Pressure, Heart Rate, and Intraocular Pressure Immediately Before and After Each Session of Eye-Related and Non-
Eye-Related Acupuncture Treatment in the Acupuncture as Treatment for Glaucoma Study

Before Individual Treatment Session After Individual Treatment Session P Value (Paired Student t Test)

Eye-Related Acupuncture Treatment

Mean systolic blood pressure (mm Hg) 118.6 6 17.3 121.6 6 18.0 .040
Mean diastolic blood pressure (mm Hg) 72.0 6 6.1 72.0 6 7.2 .976
Mean heart rate (beats per minute) 66.8 6 13.1 64.1 6 9.8 .082
Mean intraocular pressure (mm Hg) 12.9 6 1.8 13.6 6 2.1 .019
Non-Eye-Related Acupuncture Treatment
Mean systolic blood pressure (mm Hg) 120.5 6 13.2 118.1 6 13.8 .015
Mean diastolic blood pressure (mm Hg) 71.1 6 7.4 71.2 6 9.7 .901
Mean heart rate (beats per minute) 67.6 6 10.5 64.3 6 8.5 .006
Mean intraocular pressure (mm Hg) 13.0 6 1.5 13.5 6 1.7 .073

TABLE 3. Blood Pressure, Heart Rate, Best-Corrected Visual Acuity, and Diurnal Intraocular Pressure Before and After 12 Sessions of
Eye-Related and Non-Eye-Related Acupuncture Treatments in the Acupuncture as Treatment for Glaucoma Study

Before 12 Treatments After 12 Treatments P Value (Paired Student t Test)

Eye-Related Acupuncture Treatment

Mean systolic blood pressure (mm Hg) 122.5 6 17.7 119.0 6 15.3 .291
Mean diastolic blood pressure (mm Hg) 72.5 6 8.5 71.8 6 8.6 .771
Mean heart rate (beats per minute) 60.0 6 11.6 64.4 6 11.7 .067
Mean best-corrected visual acuity (logMAR) 0.084 6 0.138 0.084 6 0.138 >.999
Mean diurnal intraocular pressure (mm Hg) 12.9 6 2.0 12.7 6 2.0 .711
Non-Eye-Related Acupuncture Treatment
Mean systolic blood pressure (mm Hg) 126.5 6 19.7 116.2 6 16.8 .040
Mean diastolic blood pressure (mm Hg) 78.4 6 7.4 69.5 6 9.7 .002
Mean heart rate (beats per minute) 61.1 6 9.2 62.4 6 10.4 .565
Mean best-corrected visual acuity (logMAR) 0.102 6 0.125 0.097 6 0.140 .895
Mean diurnal intraocular pressure (mm Hg) 13.1 6 1.8 12.8 6 1.8 .381

at the baseline visit, within 1 week of completion of the first acupuncture series. Adverse reactions were recorded
series of 12 sessions, and within 1 week before and after throughout the study period.
the second series of 12 sessions. Optic disc (HRT) and In this pilot study, a sample size calculation was based on
RNFL measurements (SD OCT) were obtained at the base- IOP, which is the most important risk factor for glaucoma
line visit and within 1 week of completing the second and the primary parameter monitored in glaucoma therapy.

VOL. 160, NO. 2 ACUPUNCTURE FOR GLAUCOMA 259

 TABLE 4. Visual Field Indices (Mean Deviation and Pattern Standard Deviation of 24-2 Achromatic Automated Perimetry Standard
Swedish Interactive Threshold Algorithms) Before and After 12 Sessions of Eye-Related and Non-Eye-Related Acupuncture
Treatments in the Acupuncture as Treatment for Glaucoma Study

Before 12 Treatments After 12 Treatments P Value (Paired Student t Test)

Eye-Related Acupuncture Treatment

Average mean deviation (dB) 5.44 6 7.58 5.75 6 5.22 .409
Average pattern standard deviation (dB) 5.47 6 4.46 7.28 6 4.34 .259
Non-Eye-Related Acupuncture Treatment
Average mean deviation (dB) 4.98 6 6.11 4.99 6 5.78 .997
Average pattern standard deviation (dB) 5.61 6 4.77 5.78 6 4.83 .554

A treatment difference of 3 mm Hg was used based on the unrelated to glaucoma or acupuncture, 2 of the 8 moved
results of previous studies.4–10,17 Ten subjects were needed away, 3 had problems with transportation during the study,
for a 90% probability to detect a treatment difference of and 1 had a family crisis. An additional 3 patients (13.6%)
3 mm Hg with a 2-sided .05 significance level, assuming a were withdrawn from the study by one of the investigators
within-subject standard deviation of 1.5 mm Hg in a 2- owing to needle sensitivity (2 patients) and to clinically
treatment crossover study. Standard deviation of 1.5 mm significant elevated IOP (8 mm Hg) in the contralateral
Hg was chosen based on the IOP definition of stability in eye that required a change in glaucoma treatment for the
the inclusion criteria. patient. Table 1 summarizes the demographic characteris-
Primary outcome measurements included diurnal IOP, tics of patients who completed and did not complete the
best-corrected visual acuity (BCVA), visual field results, study.
and optic nerve and RNFL measurements. Secondary
outcome measurements included blood pressure, heart
INTRAOCULAR PRESSURE: After each acupuncture ses-
rate, and rates of adverse reactions. BP, HR, and IOP sion, with either eye-points or non-eye-points, mean IOP
were compared before and after each individual acupunc- of the treatment groups exhibited a slight increase. After
ture treatment session. Diurnal IOP, BCVA, and visual treatment of the eye-points, mean IOP increased from
field indices (mean deviation and pattern standard devia- 12.9 6 1.8 mm Hg immediately before to 13.6 6 2.1 mm
tion) were compared before and after each series of 12 ses- Hg immediately after the treatment session (P ¼ .019,
sions. Snellen chart visual acuity was converted to the Table 2). A borderline statistically significant increase of
logMAR scale for analysis. Optic nerve and RNFL mea- mean IOP was noted with the non-eye-point sessions too
surements (linear cup-to-disc area ratio, cup shape mea- (from 13.0 6 1.5 mm Hg to 13.5 6 1.7 mm Hg, P ¼
sure, rim area, rim volume, RNFL height variation .073). This IOP effect appears to be temporary, as no signif-
contour, and mean RNFL thickness were measured with icant change of diurnal IOP was detected after 12 sessions
HRT; average RNFL thickness and RNFL thickness at su- of eye-related or non-eye-related acupuncture treatments
perior, nasal, inferior, and temporal quadrants were (Table 3). Mean diurnal IOPs were 12.9 6 2.0 mm Hg
measured with SD OCT) were compared between the before and 12.7 6 2.0 mm Hg after 12 sessions of eye-
baseline visit and after both 12-session series were related acupuncture treatment, and 13.1 6 1.8 mm Hg
completed. Paired Student t tests were used to compare before and 12.8 6 1.8 mm Hg after 12 sessions of non-
variables before and after interventions. Mean values eye-related acupuncture treatment (P ¼ .711 and P ¼
are reported with their standard deviations (6SD). .381, respectively).
Adverse reactions were reported using descriptive statis-
tics. Microsoft Office Excel 2010 Analysis ToolPak
VITAL SIGNS (BLOOD PRESSURE AND HEART RATE): BP
(Microsoft, Redmond, Washington, USA) was used for and HR responses were different after single eye-point
statistical analysis. and non-eye-point treatment sessions. The mean systolic
BP and HR of the entire group showed a significant reduc-
tion after a single non-eye-point session (Table 2). Mean
systolic BP decreased from 120.5 6 13.2 mm Hg immedi-
RESULTS ately before to 118.1 6 13.8 mm Hg immediately after a
treatment session (P ¼ .015), while mean HR decreased
TWENTY-TWO GLAUCOMA PATIENTS VOLUNTEERED TO from 67.6 6 10.5 beats per minute (bpm) immediately
participate in the study and 11 patients (50.0%) completed before to 64.3 6 8.5 bpm immediately after treatment
the study. Eight patients (36.4%) did not complete (P ¼ .006). In contrast, after an eye-point session, mean
the study; 2 of the 8 patients had changes in their health systolic BP increased from 118.5 6 17.3 mm Hg before

260 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2015

 TABLE 5. Optic Disc and Retinal Nerve Fiber Layer Measurements With Confocal Scanning Laser Ophthalmoscopya in Patients Before
and After Participation in the Eye-Related and Non-Eye-Related Acupuncture Treatment of the Acupuncture as Treatment for
Glaucoma Study (Mean Duration of Study: 8.02 6 2.44 Months)

Before Eye-Related and Non-Eye-Related After Eye-Related and Non-Eye-Related P Value (Paired
Acupuncture Treatment Acupuncture Treatment Student t Test)

Linear cup-to-disc area ratio 0.711 6 0.159 0.704 6 0.150 .560

Cup shape measure 0.086 6 0.105 0.075 6 0.109 .490
Rim area (mm2) 1.019 6 0.399 1.015 6 0.352 .928
Rim volume (mm3) 0.246 6 0.144 0.236 6 0.138 .457
Retinal nerve fiber layer height variation contour (mm) 0.410 6 0.101 0.372 6 0.094 .121
Mean retinal nerve fiber layer thickness (mm) 0.185 6 0.098 0.183 6 0.099 .920

a
Heidelberg Retinal Tomography (HRT); Heidelberg Engineering, Carlsbad, California, USA.

TABLE 6. Retinal Nerve Fiber Layer Measurements With Spectral-Domain Optical Coherence Tomographya in Patients Before and
After Participation in the Eye-Related and Non-Eye-Related Acupuncture Treatment of the Acupuncture as Treatment for Glaucoma
Study (Mean Duration of Study: 8.02 6 2.44 Months)

Before Eye-Related and Non-Eye-Related After Eye-Related and Non-Eye-Related Acupuncture

Acupuncture Treatments Treatments P Value (Paired Student t Test)

Average 71.18 6 14.77 67.73 6 14.55 .124

Superior quadrant 81.36 6 22.77 79.18 6 21.35 .267
Nasal quadrant 67.73 6 9.49 64.55 6 11.06 .169
Inferior quadrant 82.64 6 26.74 81.18 6 26.08 .221
Temporal quadrant 53.00 6 7.69 53.64 6 8.67 .626

a
Cirrus spectral-domain optical coherence tomography (SD OCT); Carl Zeiss Meditec, Dublin, California, USA.

to 121.6 6 18.0 mm Hg after the session (P ¼ .040). HR unchanged after 12 sessions of either treatment (0.084 6
went from 66.8 6 13.1 bpm before to 64.1 6 9.8 bpm after 0.138, 0.097 6 0.140, P >.999 and P ¼ .895, respectively,
an eye-point treatment (P ¼ .082). Mean diastolic BP did Table 3).
not change with either treatment.
The BP response observed after individual non-eye-
HEIDELBERG RETINAL TOMOGRAPHY, SPECTRAL-
point treatment sessions appeared to be sustained. After DOMAIN OPTICAL COHERENCE TOMOGRAPHY, VISUAL
FIELD: The visual field was tested before and after 12 ses-
completion of the series of 12 sessions of non-eye-point
acupuncture, mean systolic BP decreased from 126.5 6 sions of eye-related and before and after 12 sessions of
19.7 mm Hg to 116.2 6 16.8 mm Hg and mean diastolic non-eye-related acupuncture treatments, while optic
BP was lowered from 78.4 6 7.4 mm Hg to 69.5 6 disc measurements with HRT and RNFL thickness with
9.7 mm Hg (P ¼ .040 and P ¼ .002, respectively). Changes SD OCT were obtained at baseline and after the last
in HR were not statistically significant (from 61.1 6 acupuncture treatment in the study. None of the visual
9.2 bpm to 62.4 6 10.4 bpm, P ¼ .565). For eye-point field indices (mean deviation and pattern standard devia-
acupuncture, no statistically significant changes were noted tion), optic disc, or RNFL thickness measurements
in the systolic BP, diastolic BP, or HR after 12 treatment demonstrated any statistically significant changes
sessions (Table 3). (Tables 4–6).

VISION ACUITY: BCVA (logMAR scale) did not change

after 12 sessions of either eye-point or non-eye-point DISCUSSION
acupuncture treatments. BCVA was 0.084 6 0.138 before
eye-point acupuncture treatment (equivalent to Snellen IN THIS PROSPECTIVE, RANDOMIZED, DOUBLE-MASKED
20/24) and 0.102 6 0.125 before the non-eye-point series crossover study, acupuncture therapy was associated with
(equivalent to Snellen 20/25). It remained essentially a small increase in IOP immediately following 1 treatment

VOL. 160, NO. 2 ACUPUNCTURE FOR GLAUCOMA 261

session. There was, however, no sustained effect on diurnal massaging those points in the first visit. The patients then
IOP from 12 acupuncture treatment sessions (the eye- practiced regular massage using thumb and index fingers
points or non-eye-points). We also found that immediately twice per day for 9 minutes over the next 4 weeks. Patients
after non-eye-point sessions systolic BP and HR decreased, in the sham group received tapping stimulation of sham
and that the effects on systolic BP were sustained after 12 auricular points (wrist [SF2], shoulder [SF4], and jaw
sessions of non-eye-point treatment. No statistically signif- [LO3]) in the first visit and there was no massage of those
icant alterations in BP and HR were observed with eye- acupoints over the 4 weeks of study. Post-treatment IOP
point acupuncture treatments, though a small increase in and visual acuity were compared to baseline, but no com-
systolic BP was detected immediately after single eye- parison was made between the study group and the sham
point treatments. No changes in visual acuity, visual field group by the authors. The post-treatment IOP was reported
indices, or optic disc or RNFL measurements were observed to be statistically significantly lower than baseline in both
over the course of the entire study. eyes at all measured time points. The mean IOP of right
Acupuncture effects on IOP have been reported in a few eyes decreased from 16.7 6 3.9 mm Hg to 14.1 6
small case series. In a study of 18 patients with either glau- 3.8 mm Hg, and mean IOP of left eyes decreased from
coma or ocular hypertension receiving a single session of 17.3 6 4.2 mm Hg to 14.3 6 3.7 mm Hg, over the 4-
acupuncture treatment at 3 eye-related acupoints (GB37, week study. The IOP then returned to baseline levels after
Liv3, LI4), IOP decreased 2.7 6 1.3 mm Hg 15 minutes af- the subjects stopped acupressure treatment for 4 weeks.17
ter treatment and 2.5 6 2.1 mm Hg 24 hours after treat- No statistically significant differences of IOP were found
ment.8 In another case series of 20 glaucomatous eyes of between the study group and the sham group at any
11 patients undergoing acupuncture treatment for 5 weeks, follow-up time point, except at 4 weeks (study group’s
Kurusu and associates reported that the percent IOP reduc- IOP of the right eye was 3.70 mm Hg lower than the
tion (19%) at 15 minutes after acupuncture was statistically sham group’s, 95% confidence interval [CI] 7.11
significant and tended to be lower each week over a 5-week to 0.29; left eye IOP was 4.90 mm Hg lower, 95%
period.10 BCVA was also found by Kurusu and associates10 CI 8.08 to 1.72).14 No statistically significant change
to statistically significantly improve at 5 weeks. Patients of of BCVA was noted in either the study group or the
both studies were on a regimen of topical medications sham group and no statistically significant difference in
before, during, and after their acupuncture treatments.8,10 BCVA was noted between the 2 groups at any time
Improved compliance with conventional treatment by points.14 Although the authors claimed that patients
study participants is part of the regression to the mean were masked to the treatment and the acupressure proce-
seen in clinical trials when studying an intervention that dure, the different manipulation protocols between the 2
is added to ongoing medical treatment. The IOP- groups were obvious, making it hard to maintain the
lowering effect noted after acupuncture in these 2 short- masked condition.18,19 Another limitation of the study
term studies may be due to better compliance with topical was the inclusion of a variety of glaucoma subtypes and 2
medications when the study participants were being moni- patients’ status post filtering surgery, which calls into
tored. In addition, the acupuncture effects may be overesti- question the stability of the IOP in the study sample. It is
mated by Kurusu and associates because both eyes of a difficult to compare the results of our study of
patient were included in the analysis, and variable acupuncture with this report, which used a different
acupoints were used.10 treatment procedure.
Randomized controlled trials on acupuncture for glau- Two major areas of criticism of acupuncture research are
coma have not been performed. In the second extensive the placebo effect and the nonspecific effect of acupunc-
Cochrane Review of acupuncture for glaucoma, only 1 ture. Sham acupuncture with or without a needle may
completed trial on auricular acupressure was identified.14,17 not be an inert placebo and may actually elicit a physiolog-
Auricular acupressure is an alternative method of ical response.13,18–21 Nonspecific effects from counseling
acupuncture based on the theory that different organs of about acupuncture, as well as patient and researcher
the body are represented in different small areas of the expectations, are meaningful and require the controls to
external ear. Similar to the theory of stimulating be treated like the treatment subjects. By virtue of our
acupoints with needles, pressure applied to the evaluating the ocular effects of acupuncture using 2
corresponding points on the ear is believed to be capable acupuncture points groups (1 group traditionally used in
of preventing or treating various disorders.18 That trial by eye diseases and another group for non-eye diseases) and
Her and associates evaluated the effect of auricular acupres- administering the therapy to all subjects in a masked and
sure on IOP and visual acuity by randomly assigning 16 pa- crossover fashion, each subject in this study served as his
tients to receive auricular acupressure and 17 patients to or her own control.18 Thus, we reduced any nonspecific ef-
receive sham treatment (1 patient withdrew).17 The auric- fects of acupuncture. In addition, we limited the study sam-
ular acupressure treatment consisted of tapping stimulation ple to patients with primary open-angle glaucoma with
using a 1-mm alloy ball to the auricular acupoints (kidney controlled and stable IOP to minimize the variable of
[CO10], liver [CO12], and eye [LO5] points on the ear) and IOP fluctuation. Regression to the mean is a significant

262 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2015

limitation to prior studies on acupuncture when the base- researchers concluded that active acupuncture provided
line or qualifying visit was not separated from the first inter- no greater benefit than sham acupuncture in reducing sys-
vention visit. By separating the baseline visit IOP from the tolic or diastolic BP.27 However, in our study, we found a
screening measurement in our study, we likely minimized difference in BP effects between acupuncture done at
regression to the mean, but the only way to truly eliminate eye-points and non-eye-points. Both the systolic and dia-
this factor would be to conduct the study in nonmedicated stolic BP showed a significant reduction after 12 sessions
patients. Designing the study so that acupuncture was of non-eye-point therapy (P ¼ .040 and P ¼ .002, respec-
limited to 2 preselected acupoint groups without adjunc- tively), but no significant effect was observed with
tive use of acupressure, massage, heat, or electrical stimula- eye-point therapy. It is possible that certain acupoints are
tion, and without extensive patient counseling, we associated with certain organ systems or autonomic effects.
minimized the number of variables that could confound However, the exact mechanism or physiological process of
the results. acupuncture effects is not clear.
With a rigorous study design, we found that glaucoma In Chinese traditional medicine, the body is viewed as a
patients experienced no sustained IOP changes over a se- delicate balance of 2 opposing and inseparable forces: yin
ries of 12 acupuncture sessions, although there was a small and yang. In brief, yin represents the cold, slow, or passive
immediate rise of IOP after individual treatment sessions; principle, while yang represents the hot, excited, or active
the rise was generally small and clinically insignificant. A principle. An imbalance of these 2 forces is associated with
lack of IOP response has been reported in other case se- blockage in the flow of Qi (vital force or energy) and leads
ries.11,12 In a review of 500 cases of a variety of ocular to various illnesses. Qi flows along pathways known as me-
diseases treated with acupuncture, Wong and Ching ridians, with acupuncture points on the human body that
noted that acupuncture may be associated with connect with them. The underlying philosophy of
improvement of central acuity or symptoms of dim acupuncture is that disorders related to the flow of Qi can
vision, but no improvements in visual field, color vision, be prevented or treated by stimulating the relevant
night blindness, or IOP were observed.11 In healthy indi- acupuncture point on the body surface.14,18,28
viduals, acupuncture also did not produce a statistically sig- Research efforts have focused on explaining acupuncture
nificant short-term effect on the IOP in a randomized within the framework of Western medicine, and different
double-masked controlled trial.22 mechanisms of action have been proposed. The most com-
The effect of acupuncture therapy on visual acuity has mon explanation is that acupuncture stimulates the release
been controversial. There are reports associating acupunc- of neurochemicals such as endogenous opioids, nitric oxide,
ture with improvement of central vision in patients with catecholamines, and serotonin.28–30 It has been suggested
amblyopia, myopia, and retinitis pigmentosa.23-25 that the central nervous system is essential for processing
Although some studies have reported significant the effect of acupuncture by modulating the autonomic
improvement of visual acuity after acupuncture therapy, nervous system, neuroimmune system, and hormonal
most clinicians have serious doubts about the results, regulation.31–33 In addition to having possible effects on
given the possible placebo and nonspecific effects.13,26 In IOP, visual field, and visual acuity, acupuncture has been
the randomized controlled trial of auricular acupressure associated with an increase in ocular blood flow,34–36
by Her and associates, no statistically significant changes preservation of normal waveform characteristics of
in BCVA were noted in either the active or sham multifocal electroretinogram (mfERG),37 alteration of
groups.17 In our study, the BCVA remained the same visual function tested by visual evoked potential (VEP),38
over 12 sessions of acupuncture of eye-points and non- increase of retinal nerve growth factor,39 and protection of
eye-points. However, the BCVA of both our sample and retinal ganglion cells from acute increase of IOP.40
Her’s study were relatively good at baseline, not leaving Although glaucoma is an optic neuropathy with IOP being
much room for improvement. a major risk factor, IOP certainly is not the only factor
The effect of acupuncture on BP has been studied in a involved. Thus, a lack of effect on IOP cannot be taken as
large double-masked, randomized, sham-controlled trial, a lack of effect on the optic neuropathy. This study clarifies
the Stop Hypertension with Acupuncture Research Pro- that acupuncture therapy has little if any effect on IOP.
gram (SHARP), funded by the National Institutes of We did not find any significant changes in visual
Health National Center for Complementary and Alterna- field indices, optic nerve parameters, and peripapillary
tive Medicine.27 It showed that in patients with hyperten- RNFL parameters over the course of the study, though the
sion, although the mean BP decreased from baseline to power to detect any change in these variables is limited by
10 weeks, it did not differ significantly between participants the relatively short study duration and the small sample
randomly assigned to active vs sham acupuncture (systolic size of this pilot study.
BP: 3.56 mm Hg vs 3.84 mm Hg, respectively; 95% CI The major difficulty with any study on acupuncture ther-
for the difference: 4.0 to 4.6 mm Hg; P ¼ .90; diastolic apy or alternative therapy is subject recruitment, since the
BP: 4.32 mm Hg vs 2.81 mm Hg, respectively; 95% therapies studied are not well established and are not part
CI for the difference: 3.6 to 0.6 mm Hg; P ¼ .16). The of the standard of care. As one of the major glaucoma

VOL. 160, NO. 2 ACUPUNCTURE FOR GLAUCOMA 263

centers in the metropolitan area on the West Coast, with a VEP measurement were not studied. It is also important
diverse population rather receptive to Eastern medicine, we to understand that acupuncture therapy is not practiced
receive many inquiries regarding the role of acupuncture in as a static therapy. Acupuncturists tend to alter the main
treating glaucoma. However, owing to the significant time acupoints and add supplemental points during therapy
commitment required for this study, time commitment based on a patient’s response or symptoms. In this study,
along with transportation issues became a major hindrance we limited the acupoints to 2 groups and did not allow
to recruitment and retention. Even with careful screening, alteration or addition during the study. It would be very
the dropout rate was high: 22 glaucoma patients volun- difficult to design a rigorous scientific study allowing a var-
teered to join the study, but only 11 patients (50.0%) iable number of acupoints.24 Acupuncture, in general, also
completed it. Adverse reaction rate was low to moderate includes acupressure, which involves application of digital
(13.6%, 3 patients); 2 patients were withdrawn from the pressure or massage to points on the body or representative
study because of needle sensitivity and 1 experienced points on the ear, electrical stimulation of needles, and
IOP increase of the fellow eye during acupuncture therapy. moxibustion. Our study design did not allow these addi-
There are some notable limitations of this study. tional manipulations of traditional acupuncture, and there-
Although we conducted the study in a prospective cross- fore our study may not be generalizable to the overall
over fashion and with patients masked to the theoretical practice of acupuncture.
treatment effect predicted by traditional Chinese medici- As far as we are aware, this is the first study that has used
nal principles, a placebo effect cannot be eliminated. In a controlled design on acupuncture therapy for glaucoma
addition, the sample size of this pilot study was small, management. It demonstrates the feasibility of studying
although there was 90% statistical power to detect a treat- acupuncture treatment in glaucoma patients in a rigorous
ment difference of 3 mm Hg between groups. Although scientific fashion to evaluate several parameters that are
there was an obvious advantage of including patients important for glaucoma management. Acupuncture ther-
with stable IOPs, the glaucoma medical therapy may apy for glaucoma was not associated with any changes of vi-
have stabilized the IOP enough that further IOP changes sual acuity. Systemic effects, in terms of BP and HR, may be
were difficult to detect. We limited the variables collected achieved by acupuncture of certain acupoints but were not
to IOP, BCVA, BP, HR, visual field indices, and optic disc found with acupuncture of eye-points. Acupuncture has no
and RNFL parameters for practical reasons, but other overall effect on diurnal IOP reduction but may tempo-
possibly important variables such as ocular circulation, rarily cause a small increase in IOP immediately after a
pattern electroretinography, mfERG, VEP, and multifocal treatment session.

ALL AUTHORS HAVE COMPLETED AND SUBMITTED THE ICMJE FORM FOR DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST.
Financial Disclosures: None. Funding Support: Oppenheimer Seed Grants in Complementary, Alternative, and Integrative Medicine, University of Cal-
ifornia Los Angeles, David Geffen School of Medicine, Los Angeles, California. All authors attest that they meet the current ICMJE requirements to
qualify as authors.
The authors acknowledge Fei Yu, PhD, of the Biostatics Department of UCLA for providing statistical analysis.

REFERENCES 6. Liu W, Yang G, Zhao XJ, et al. Impact of acupuncture on 24 h

intraocular pressure of glaucoma. Zhongguo Zhen Jiu 2011;
1. American Academy of Ophthalmology Glaucoma Panel. 31(6):518–520.
Preferred Practice PatternÒ Guidelines. Primary Open- 7. Chu TC, Potter DE. Ocular hypotension induced by electro-
Angle Glaucoma 2010. San Francisco, CA: American Acad- acupuncture. J Ocul Pharmacol Ther 2002;18(4):293–305.
emy of Ophthalmology. Available at, www.aao.org/ppp. 8. Uhrig S, Hummelsberger J, Brinkhaus B. Standardized
Accessed May 14, 2015. acupuncture therapy in patients with ocular hypertension or
2. Peters D, Bengtsson B, Heijl A. Lifetime risk of blindness in glaucoma–results of a prospective observation study. Forsch
open-angle glaucoma. Am J Ophthalmol 2013;156(4):724–730. Komplementarmed Klass Naturheilkd 2003;10(5):256–261.
3. Rhee DJ, Spaeth GL, Myers JS, et al. Prevalence of the use of 9. Kim MS, Seo KM, Nam TC. Effect of acupuncture on intraoc-
complementary and alternative medicine for glaucoma. ular pressure in normal dogs. J Vet Med Sci 2005;67(12):
Ophthalmology 2002;109(3):438–443. 1281–1282.
4. Ralston NS. Successful treatment and management of acute 10. Kurusu M, Watanabe K, Nakazawa T, et al. Acupuncture for
glaucoma using acupuncture. Am J Acupuncture 1977;5(3): patients with glaucoma. Explore (NY) 2005;1(5):372–376.
283–285. 11. Wong S, Ching R. The use of acupuncture in ophthalmology.
5. Dabov S, Goutoranov G, Ivanova R, Petkova N. Clinical Am J Chin Med 1980;8(1-2):104–153.
application of acupuncture in ophthalmology. Acupuncture 12. Woo YM, Lee MS, Nam Y, Cho HJ, Shin BC. Effects of
Electrother Res 1985;10(1-2):79–93. contralateral electroacupuncture on brain function: a

264 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2015

 double-blind, randomized, pilot clinical trial. J Altern Comple- 28. Kaptchuk TJ. Acupuncture: theory, efficacy, and practice.
ment Med 2006;12(8):813–815. Ann Intern Med 2002;136(5):374–383.
13. [No authors listed]. NIH Consensus Conference. Acupunc- 29. Mayer DJ, Price DD. Endorphin release as mechanism of
ture. JAMA 1998;280(17):1518–1524. acupuncture analgesia. Pain 1981;11(2):273–280.
14. Law SK, Li T. Acupuncture for glaucoma. Cochrane Database 30. Zhao ZQ. Neural mechanism underlying acupuncture anal-
Syst Rev 2013;5:CD006030. gesia. Prog Neurobiol 2008;85(4):355–375.
15. Blackwell R, Macpherson H. ‘‘Bright eyes’’ - the treatment of 31. Cho ZH, Hwang SC, Wong EK, et al. Neural substrates, exper-
eye diseases by acupuncture. J Chinese Med 1992;39:1–8. imental evidences and functional hypothesis of acupuncture
16. Jin G, Jin JX, Jin LL. Contemporary Medical Acupuncture: A mechanisms. Acta Neurol Scand 2006;113(6):370–377.
Systematic Approach. Beijing: Higher Education Press; 32. Kim HW, Kang SY, Yoon SY, et al. Low-frequency electroa-
2007:344–351. cupuncture suppresses zymosan-induced peripheral inflamma-
17. Her JS, Liu PL, Cheng NC, et al. Intraocular pressure- tion via activation of sympathetic post-ganglionic neurons.
lowering effect of auricular acupressure in patients with glau- Brain Res 2007;1148:69–75.
coma: a prospective, single-blinded, randomized controlled 33. Sakai S, Hori E, Umeno K, Kitabayashi N, Ono T, Nishijo H.
trial. J Altern Complement Med 2010;16(11):1177–1184. Specific acupuncture sensation correlates with EEGs and
18. Mayer DJ. Acupuncture: an evidence-based review of the autonomic changes in human subjects. Auton Neurosci
clinical literature. Annu Rev Med 2000;51:49–63. 2007;133(2):158–169.
19. Vase L, Baram S, Takakura N, et al. Specifying the nonspe- 34. Naruse S, Mori K, Kurihara M, et al. Chorioretinal blood
cific components of acupuncture analgesia. Pain 2013; flow changes following acupuncture between thumb and
154(9):1659–1667. forefinger. Nippon Ganka Gakkai Zasshi 2000;104(10):
20. Moffet HH. Sham acupuncture may be as efficacious as true 717–723.
acupuncture: a systematic review of clinical trials. J Altern 35. Mizukami M, Yano T, Yamada J. Effects of ocular circulation
Complement Med 2009;15(3):213–216. by acupuncture stimulation on the crus outside - the compar-
21. Linde K, Niemann K, Schneider A, Meissner K. How large ison of GB36, GB37, GB38, and non-meridian point. The
are the nonspecific effects of acupuncture? A meta-analysis Journal of Japanese Association of Physical Medicine Balneology
of randomized controlled trials. BMC Med 2010;8:75. Climatology 2006;69(3):201–212.
22. Meira-Freitas D, Cariello AJ, Vita RC, Tabosa A, Melo LA Jr. 36. Takayama S, Watanabe M, Kusuyama H, et al. Evaluation of
Short-term effect of acupucture on intraocular pressure in the effects of acupuncture on blood flow in humans with ul-
healthy subjects. Acupunct Med 2010;28(1):25–27. trasound color Doppler imaging. Evid Based Complement
23. Zhao J, Lam DS, Chen LJ, et al. Randomized controlled trial Alternat Med 2012;2012:513638.
of patching vs acupuncture for anisometropic amblyopia in 37. Chan HH, Leung MC, So KF. Electroacupuncture provides a
children aged 7 to 12 years. Arch Ophthalmol 2010;128(12): new approach to neuroprotection in rats with induced glau-
1510–1517. coma. J Altern Complement Med 2005;11(2):315–322.
24. Wei ML, Liu JP, Li N, Liu M. Acupuncture for slowing the 38. Sagara Y, Fuse N, Seimiva M, et al. Visual function with
progression of myopia in children and adolescents. Cochrane acupuncture tested by visual evoked potential. Tohoku J
Database Syst Rev 2011;9:CD007842. Exp Med 2006;209(3):235–241.
25. Bittner AK, Gould JM, Rosenfarb A, Rozanski C, 39. Pagani L, Manni L, Aloe L. Effects of electroacupuncture on
Dagnelie G. A pilot study of an acupuncture protocol to retinal nerve growth factor and brian-derived neurotrophic
improve visual function in retinitis pigmentosa patients. factor expression in a rat model of retinitis pigmentosa. Brain
Clin Exp Optom 2014;97(3):240–247. Res 2006;1092(1):198–206.
26. Metheny J. Acupuncture and the placebo effect. Arch 40. Zhou W, Yang J, Xia Y, Wang H, Guo G, Zhang J. The effect
Ophthalmol 2011;129(8):1107–1108. author reply 1108. of electric acupuncture on the reitnal ganglion cells in rabbits
27. Macklin EA, Wayne PM, Kalish LA, et al. Stop Hypertension with acute high intraocular pressure. In: Peng Y, Weng X,
with the Acupuncture Research Program (SHARP): results of editors. 7th Asian-Pacific Conference on Medical and Bio-
a randomized, controlled clinical trial. Hypertension 2006; logical Engineering IFMBE Proceedings. Beijing: Springer-
48(5):838–845. Verlag Berlin Heidelberg; 2008:711–714.

VOL. 160, NO. 2 ACUPUNCTURE FOR GLAUCOMA 265