Pain Medicine

Section Editor: Spencer S. Liu

Ultrasound-Guided Pulsed Radiofrequency Stimulation

of the Suprascapular Nerve for Adhesive Capsulitis:
A Prospective, Randomized, Controlled Trial
Yung-Tsan Wu, MD,* Cheng-Wen Ho, MD, PhD,† Yi-Ling Chen, MD,* Tsung-Ying Li, MD,*
Kuei-Chen Lee, PT, MS,* and Liang-Cheng Chen, MD, MS*

BACKGROUND: The treatment of adhesive capsulitis (AC) is a well-known, complicated, and

long process. Recent studies have shown that pulsed radiofrequency (PRF) lesioning of the
suprascapular nerve (SSN) using a fluoroscopy- or computed tomography-guided technique can
alleviate shoulder pain. However, there are no studies of PRF lesioning of the SSN in patients
with AC using ultrasound-guided (UG) techniques, except for 2 case reports. In this study, we
compared the effect of physical therapy alone with physical therapy and PRF lesioning of the
SSN using a UG technique.
METHODS: Sixty patients with AC were included in the study. Patients were randomized into
the following 2 groups: the intervention group containing patients who received 12 weeks of
physical therapy after 1 treatment of PRF lesioning of the SSN, and the control group containing
patients who received 12 weeks of physical therapy alone. All outcome measurements includ-
ing visual analog scale (VAS), shoulder pain and disability index, and passive range of motion
(PROM) were performed at 1, 4, 8, and 12 weeks after treatment.
RESULTS: Forty-two patients (21 patients in each group) completed the study. The intervention
group had a notably shorter time to onset of significant pain relief (6.1 ± 3.4 vs 28.1 ± 9.2
days; P < 0.001) and noticeable reduction of VAS score at week 1 (40% vs 4.7%) than the con-
trol group (P < 0.001). All measured variables in the intervention group and most variables in
the control group showed significant improvement from the baseline (P < 0.05). A comparison
of the 2 groups indicated significantly greater improvement in the intervention group at all times
in VAS and shoulder pain and disability index scores (all P < 0.05), and for most gain of PROM
(P < 0.05). There were no serious adverse effects or complications in either group.
CONCLUSIONS: This study indicates that the application of PRF lesioning of the SSN using a UG
technique combined with physical therapy provided better and faster relief from pain, reduced
disability, and improved PROM when compared with physical therapy alone in patients with AC,
an effect that persisted for at least 12 weeks.  (Anesth Analg 2014;119:686–92)

A
dhesive capsulitis (AC), more commonly referred are many treatments including oral analgesics, intraar-
to as frozen shoulder, affects 2% to 5% of the gen- ticular steroid injections, and physical therapy, there is
eral United Kingdom population at any given disagreement in the literature on the best treatment at dif-
time.1 This condition is characterized by shoulder pain and ferent stages of the disease.
a gradual onset of significantly diminished active and pas- The suprascapular nerve (SSN) originates from the
sive (PROM) range of motion (ROM) of the glenohumeral upper trunk of the brachial plexus (C5–C6), and it inner-
joint.1 The prognosis of AC varies depending on the study, vates approximately 70% of the superior and posterior
and 20% to 50% of patients with AC can have continued regions of the shoulder joint, capsule, and acromioclavicu-
pain and limited ROM beyond 3 years.2,3 Although there lar joint.4 This nerve also supplies motor innervation to the
supra- and infraspinatus muscles.5 An SSN block (SSNB)
From the *Department of Physical Medicine and Rehabilitation, Tri-Service can effectively relieve the pain and increase the ROM of
General Hospital and †Institute of Aerospace Medicine, School of Medicine,
National Defense Medical Center, Taipei, Taiwan, Republic of China.
patients with long histories of shoulder pain including AC,
Accepted for publication May 9, 2014. rotator cuff injuries, osteoarthritis, and rheumatoid arthri-
Funding: The study is supported, in part, by the National Science Council tis.4,6–9 Jones and Chattopadhyay performed a randomized
(grant no. 102-2314-B-016-013) and the Tri-Service General Hospital (grant trial that compared SSNB with intraarticular injections in
no. TSGH-C100-096), Taipei, Taiwan, Republic of China.
patients with AC and concluded that SSNB provided faster
The authors declare no conflicts of interest.
acting and more long-lasting pain relief for up to 3 months.10
Reprints will not be available from the authors.
Previous research indicated that pulsed radiofrequency
Address correspondence to Liang-Cheng Chen, MD, MS, Department of
Physical Medicine and Rehabilitation, Tri-Service General Hospital, School (PRF) lesioning can alleviate chronic pain by delivering an
of Medicine, National Defense Medical Center, No. 325, Sec. 2, Cheng-Kung electrical field and heat bursts (temperature of <42°C, in
Rd., Neihu District, Taipei, Taiwan, Republic of China. Address e-mail to
[email protected]. contrast to conventional radiofrequency applications, which
Copyright © 2014 International Anesthesia Research Society deliver a constant temperature of 60°C–80°C) to neural tis-
DOI: 10.1213/ANE.0000000000000354 sue without causing neural injury.11 There is also evidence

686 www.anesthesia-analgesia.org September 2014 • Volume 119 • Number 3

that PRF lesioning can provide pain relief in patients with 54-mm radiofrequency probe (4-mm active tip) was then
diverse chronic conditions.12–14 Recent research reported that advanced under UG toward the scapular notch by using
PRF lesioning of the SSN was effective in treating chronic a medial to lateral approach (Fig. 1). Sensory and motor
shoulder pain, including AC, and it facilitated functional stimulation was tested when the needle was close to the
recovery without risking paralysis of the supra- or infraspi- SSN. Sensory stimulation (50–100 Hz, 1-ms pulse width,
natus muscles.12,15–27 However, most of these studies were up to 0.5 volt) was tested, and the patients reported par-
small and did not include control groups. Furthermore, esthesia in the shoulder joint. After motor stimulation (2
most authors used fluoroscopy- or computed tomography Hz, 1-ms pulse width, up to 1 volt), the contractions of
(CT)-guided techniques, although in 2 case reports, the the infra- and supraspinatus muscles were observed. PRF
authors used ultrasound-guided (UG) techniques.20,24 lesioning was then performed for 180 seconds (2 Hz, 30-ms
The use of a UG technique improves the quality of pulse width, 42°C).19 At 30 minutes after the PRF lesioning
regional nerve blocks because it provides direct visual- procedure, the patients reported a decrease in pain using a
ization of nerves and thereby allows for more rapid onset visual analog scale (VAS). Patients were discharged if there
and effective duration of nerve blocks, with limited neural were no significant complications, such as pain, bleeding,
trauma.28 Several authors of recent reports used UG SSNB, or pneumothorax.
and they showed significant pain relief and functional
improvement in patients with chronic shoulder pain of a Physical Therapy
wide range of etiologies.29–31 Other researchers reported that Initially, hot pack application and transcutaneous elec-
the use of ultrasonography allowed for a more accurate and tronic nerve stimulation were performed for both groups.
closer placement of the needle tip to the SSN.30–32 Therefore, Then, for 12 weeks, stretching exercises, mobilization, and
compared with other SSNB techniques, a UG technique is therapeutic exercises (30 minutes per session, 3 sessions per
more convenient and more effective in targeting the SSN.30 week) were performed under the guidance of a therapist
However, there are only 2 case reports of the use of PRF who was blinded to patient allocation.
lesioning of the SSN under UG in patients with shoulder
pain.20,24 The purpose of this 12-week follow-up study of Outcome Measurements
patients with AC was to compare the effect of physical ther- All outcome measurements were performed by the same
apy alone with physical therapy provided with PRF lesion- physician, who was blinded to patient allocation, at 1, 4, 8,
ing of the SSN under UG. and 12 weeks after the initial treatment. The primary out-
come measure was the reduction of pain intensity as quan-
METHODS tified by the VAS, with a scale ranging from 0 (no pain)
Subjects to 10 (extremely severe pain). The severity of pain during
The appropriate IRB approved the study, and written any shoulder activity that occupied the most time per day
informed consent was obtained from all subjects. Sixty was recorded. Each patient reported the VAS score every
patients with AC from the outpatient service of a local day at the same time after the initial treatment until 2 con-
medical center were enrolled in the study between April secutive days when the VAS score had declined by ≥30%,
2011 and May 2013. All diagnoses of AC were confirmed via which was defined as the onset time of significant pain
shoulder ultrasonography. Exclusion criteria included all relief. Secondary clinical outcome measures were the score
patients known to have cervical disk disorder, malignancy, of shoulder pain and disability index (SPADI), a validated,
stroke, mental problems, deformities of the affected shoul- self-reported questionnaire that is used to measure pain (5
der joint, or trauma history in the previous 4 weeks. The items) and disability (8 items) associated with a shoulder
research assistant randomized the 60 enrolled patients by disorder. The pain subscale ranges from 0 to 50, the disabil-
dividing sealed envelopes into 2 groups: the intervention ity subscale ranges from 0 to 80, with the total scale rang-
group containing patients who received 12 weeks of physi- ing from 0 to 130. An increasing score indicates increasing
cal therapy after 1 dose of PRF lesioning of the SSN and the pain or disability.33 Another secondary clinical outcome
control group containing patients who received 12 weeks measure was the PROM, which was evaluated by the same
of physical therapy alone. All patients were instructed to physiotherapist using a universal goniometer according to
avoid any other treatments for their shoulder pain, includ- the method reported by Palmer and Epler.34 For example,
ing analgesic medicines, acupuncture, or injections during medial or lateral rotation was measured using the goniom-
the course of this study, and they were asked to inform us if eter axis in the elbow through the longitudinal axis of the
they had received these treatments. humerus with the stationary arm remaining at 0° and the
movement arm remaining parallel to the forearm.
PRF of the SSN Under UG
The same physician performed musculoskeletal ultraso- Data Analysis
nography for all patients. The linear transducer was posi- Statistical analyses were performed using SPSS 12.0 soft-
tioned just cephalad and parallel to the scapular spine. The ware for Windows. Because of a high dropout rate (18
trapezius muscle, supraspinatus muscle, and bony fossa patients were lost to follow-up), we examined the dis-
were easily identified when the transducer was gradually tribution of both groups with the Kolmogorov–Smirnov
moved in the cephalad direction. The SSN was identified test, because normal distribution did not exist. Therefore,
as a hyperechoic structure, 3 to 4 cm deep and below the nonparametric statistics was used to increase the power.
transverse scapular ligament in the scapular notch.20,29 A Demographic data were analyzed with the Mann–Whitney

September 2014 • Volume 119 • Number 3 www.anesthesia-analgesia.org 687

Ultrasound-Guided Pulsed Radiofrequency for Adhesive Capsulitis

Figure 1. A, Positioning of the ultrasonographic

transducer and needle for radiofrequency. B,
Real-time imaging of needle insertion for the ultra-
sound-guided suprascapular nerve intervention.

Figure 2. Study flow diagram.

Table 1.  Baseline Demographic and Clinical Characteristics of Study Participants

Intervention group (N = 21) Control group (N = 21) P value
Age (y) (SD) 55.0 ± 9.2 57.1 ± 10.9 0.588
Duration (mo) (SD) 6.8 ± 6.3 7.3 ± 6.5 0.668
Female/male (N) 16/5 17/4 0.707
Lesion site
 Right side 7 9 0.525
 Left side 14 12
VAS 6.5 ± 1.3 6.3 ± 1.5 0.990
SPADI pain 62.2 ± 15.7 56.9 ± 14.6 0.284
SPADI disability 51.5 ± 11.6 49.2 ± 13.3 0.497
SPADI total 55.6 ± 11.9 52.1 ± 12.7 0.450
Passive flexion 124.8 ± 19.9 128.5 ± 13.6 0.614
Passive extension 32.4 ± 10.3 36.4 ± 12.8 0.605
Passive abduction 82.2 ± 23.3 87.6 ± 20.9 0.358
Passive medial rotation 43.3 ± 15.4 45.5 ± 5.6 0.696
Passive lateral rotation 34.6 ± 13.2 38.9 ± 16.2 0.413
VAS = visual analog scale; SPADI = shoulder pain and disability index.

688   
www.anesthesia-analgesia.org anesthesia & analgesia
Table 2.  All the Measured Variables Before and After Treatment
Intervention group (N = 21) Control group 1 (N = 21)
Mean ± SD P value Mean ± SD P value P value
Onset time of pain relief (day) (SD) 6.1 ± 3.4 28.1 ± 9.2 <0.001*
VAS baseline 6.5 ± 1.3 6.3 ± 1.5 0.990
 VASWK1-FU 3.9 ± 1.3 <0.001* 6.0 ± 1.7 0.092 <0.001*
 VASWK4-FU 2.5 ± 1.3 <0.001* 4.9 ± 2.5 0.001† 0.001†
 VASWK8-FU 2.2 ± 1.6 <0.001* 3.9 ± 2.4 <0.001* 0.011‡
 VASWK12-FU 1.7 ± 1.5 <0.001* 3.3 ± 2.5 <0.001* 0.026‡
Pain SPADI baseline 62.2 ± 15.7 56.9 ± 14.6 0.284
 Pain SPADIWk1-FU 41.5 ± 17.5 <0.001* 55.5 ± 17.6 0.203 0.032‡
 PainSPADIWk4-FU 29.8 ± 15.4 <0.001* 49.2 ± 21.8 0.011‡ 0.004†
 PainSPADIWk8-FU 21.6 ± 12.7 <0.001* 43.3 ± 22.9 0.002† 0.001†
 PainSPADIWk12-FU 17.4 ± 14.4 <0.001* 38.1 ± 22.2 <0.001* 0.002†
Disability SPADI baseline 51.5 ± 11.6 49.2 ± 13.3 0.497
 Disability SPADIWk1-FU 34.8 ± 14.1 <0.001* 46.8 ± 16.6 0.033‡ 0.021‡
 Disability SPADIWk4-FU 24.9 ± 12.7 <0.001* 44.0 ± 17.9 0.008† 0.001*
 Disability SPADIWk8-FU 19.9 ± 12.5 <0.001* 38.7 ± 19.2 0.008† 0.001†
 Disability SPADIWk12-FU 15.0 ± 12.3 <0.001* 35.2 ± 18.0 <0.001* <0.001*
Total SPADI baseline 55.6 ± 11.9 52.17 ± 12.7 0.450
 TotalSPADIwk1-FU 37.3 ± 14.0 <0.001* 50.1 ± 16.1 0.052 0.016‡
 TotalSPADIwk4-FU 26.8 ± 12.6 <0.001* 45.4 ± 18.6 0.003† 0.001*
 TotalSPADIwk8-FU 20.5 ± 11.5 <0.001* 40.5 ± 19.9 0.005† 0.002†
 TotalSPADIwk12-FU 15.6 ± 12.3 <0.001* 36.3 ± 19.0 <0.001* <0.001*
Passive FL baseline 124.8 ± 19.9 128.5 ± 13.6 0.614
 Passive FLwk1-FU 139.9 ± 19.9 <0.001* 132.5 ± 15.3 0.076 0.385
 Passive FLwk4-FU 149.2 ± 17.4 <0.001* 140.9 ± 12.5 0.002† 0.054
 Passive FLwk8-FU 153.9 ± 16.9 <0.001* 145.4 ± 10.3 <0.001* 0.020‡
 Passive FLwk12-FU 160.2 ± 13.7 <0.001* 150.2 ± 10.5 <0.001* 0.010†
Passive EX baseline 32.4 ± 10.3 36.4 ± 12.8 0.605
 Passive EXwk1-FU 39.4 ± 10.8 0.001† 36.0 ± 13.3 0.965 0.326
 Passive EXwk4-FU 46.4 ± 11.6 <0.001* 39.4 ± 15.8 0.164 0.137
 Passive EXwk8-FU 51.7 ± 10.1 <0.001* 44.7 ± 14.3 0.007† 0.108
 Passive EXwk12-FU 54.1 ± 10.6 <0.001* 45.3 ± 12.3 0.006† 0.019‡
Passive Abd baseline 82.2 ± 23.3 87.6 ± 20.9 0.358
 Passive Abdwk1-FU 92.0 ± 26.9 <0.001* 94.5 ± 24.6 0.044‡ 0.537
 Passive Abdwk4-FU 105.3 ± 24.3 <0.001* 102.5 ± 25.8 0.001† 0.489
 Passive Abdwk8-FU 116.5 ± 25.7 <0.001* 110.8 ± 22.3 <0.001* 0.345
 Passive Abdwk12-FU 131.2 ± 27.1 <0.001* 117.6 ± 16.9 <0.001* 0.091
Passive Med. Rot baseline 43.3 ± 15.4 45.5 ± 5.6 0.696
 Passive Med.Rotwk1-FU 53.8 ± 12.4 0.005† 52.3 ± 8.5 0.002† 0.753
 Passive Med.Rotwk4-FU 65.1 ± 14.7 <0.001* 57.0 ± 8.3 0.001* 0.040‡
 Passive Med.Rotwk8-FU 69.3 ± 11.7 <0.001* 60.9 ± 8.2 <0.001* 0.005†
 Passive Med.Rotwk12-FU 75.2 ± 10.5 <0.001* 66.0 ± 6.3 <0.001* 0.002†
 Passive Lat. Rot baseline 34.6 ± 13.2 38.9 ± 16.2 0.413
 Passive Lat.Rotwk1-FU 40.1 ± 11.7 0.001* 44.8 ± 19.4 0.003† 0.544
 Passive Lat.Rotwk4-FU 51.4 ± 15.7 <0.001* 49.1 ± 18.5 <0.001* 0.753
 Passive Lat.Rotwk8-FU 56.3 ± 14.6 <0.001* 52.7 ± 17.8 <0.001* 0.772
 Passive Lat.Rotwk12-FU 62.7 ± 12.4 <0.001* 59.8 ± 16.0 <0.001* 0.427
VAS = visual analog scale; SPADI = shoulder pain and disability index; WK = week; FU = follow up; FL = flexion; EX = extension; Abd = abduction; Med. Rot = medial
rotation; Lat. Rot = lateral rotation.
*P ≦ 0.001.
†P ≦ 0.01.
‡P ≦ 0.05.

U test for continuous data and the χ2 test for categorical Table 1 shows the physical characteristics of the 2 groups.
data. The Wilcoxon signed rank test was used to compare There were no significant differences between groups in
the outcome measures within each group. The outcomes any of the measured baseline characteristics, VAS or SPADI
at each follow-up period were compared with the baseline scores, or PROM.
values, and the differences between groups were compared Table 2 shows the time to onset of significant pain relief,
using the Mann–Whitney U test. A P value of <0.05 was con- VAS and SPADI scores, and gain of PROM before and after
sidered statistically significant. treatment for the 2 groups. The intervention group had a
notably shorter time to onset of significant pain relief than
RESULTS the control group (6.1 ± 3.4 vs 28.1 ± 9.2 days; P < 0.001). The
Forty-two patients (21 patients in each group) completed the VAS score immediately after PRF lesioning decreased from
study; the patient enrollment flow is outlined in Figure 2. 6.5 ± 1.3 to 4.2 ± 2.2 (P < 0.001). Moreover, the intervention

September 2014 • Volume 119 • Number 3 www.anesthesia-analgesia.org 689

Ultrasound-Guided Pulsed Radiofrequency for Adhesive Capsulitis

group had a significant reduction in their VAS scores at week improved ROM compared with intraarticular corticosteroid
1 follow-up (40% vs 4.7%) compared with the control group injection combined with physical therapy or physical ther-
(P < 0.001). The intervention group also had a statistically apy alone, and that this effect persisted for at least 12 weeks.
significant reduction in the VAS (P < 0.001) and SPADI (P < Nonetheless, studies concerning the application of PRF
0.001) scores and gain of PROM (P < 0.001) compared with lesioning of the SSN under UG are lacking, because there
baseline at all follow-ups. The control group experienced are only 2 case reports.20,24 Huang et al.20 applied PRF lesion-
a significant reduction in their VAS and SPADI scores, and ing of the SSN under UG in 2 patients with AC and reported
gain of PROM (all P < 0.05) compared with the baseline at an effect that lasted for 5 to 6 months. Kim et al.24 examined
all follow-ups, except for the week 1 VAS and SPADI (pain 1 patient with chronic intractable calcified tendonitis of the
and total) scores, the week-1 passive flexion, and the week-1 rotator cuff, and they reported significant reduction of pain
and week-4 passive extension measurements. Comparison 2 months after the application of PRF lesioning under UG in
of the 2 groups indicated significant improvement in the the axillary and SSNs. Our study supports these previous
intervention group throughout the study with respect to claims of a benefit of this method in patients with AC.
the VAS and SPADI scores, and for the most gain in PROM Fluoroscopy- or CT-guided techniques are frequently
(passive flexion: weeks 8 and 12; passive extension in week used in the application of PRF lesioning of the SSN.
12, medial rotation in weeks 4, 8, and 12; all P < 0.05). There Ultrasound can provide an accurate viewing field of the
were no serious adverse effects or complications in either SSN. A randomized controlled trial by Lee et al.41 showed
group, but 4 patients in the intervention group reported that UG intraarticular injection for treatment of AC was
mild tingling or pain at the puncture site. These symptoms significantly more successful for targeting the SSN than a
disappeared 1 hour later without any treatment. No patient blind technique. Under UG, the operator can easily observe
received any other treatments except regular physical ther- the needle, thereby avoiding the possibility of a pneumo-
apy during the course of this study. thorax. Moreover, ultrasound does not involve radiation
exposure; it is less expensive and more convenient than CT
DISCUSSION or fluoroscopy. However, there is some difficulty in apply-
These results indicate that application of PRF lesioning of ing PRF lesioning of the SSN under UG because the SSN is
the SSN under UG combined with physical therapy pro- relatively deep, making accurate tracking with sonography
vided better and faster relief from pain, reduced disability, somewhat difficult.
and improved PROM (except abduction and lateral rota- In the early stages of AC, the aim of management is to
tion) than physical therapy alone in patients with AC, an alleviate pain so that rehabilitation can be used to restore
effect that persisted for at least 12 weeks. To the best of our normal shoulder ROM and activity. Previous studies indi-
knowledge, this is the first prospective, randomized, con- cated that some patients could not finish rehabilitation
trolled study to investigate the effect of PRF lesioning of the because of the presence of intractable pain, leading to sub-
SSN under UG in patients with AC. sequent studies for investigating the effect of pain relief
Sluijter et al.35 first reported the use of PRF lesioning before rehabilitation. One of the most frequently used meth-
for the relief of chronic pain in 1998. Subsequent studies ods is intraarticular injection with a steroid. However, one
concluded that this method was safe, nondestructive, had review concluded that an intraarticular steroid injection
reproducible results, and provided long-lasting relief from before rehabilitation provided only a short-term benefit.42
pain of different etiologies.12–14 Rohof16 first reported the Our study is the first to investigate the effect of PRF lesion-
application of PRF lesioning to the SSN in 2002. Other stud- ing of the SSN before physical therapy. Similar to the results
ies also reported that PRF lesioning of the SSN relieved acute of Abdelshafi et al.,40 we found that this method provided
and chronic shoulder pain from multiple etiologies.2,12,15 long-term relief from pain as well as improved the SPADI
Shah and Racz15 and Gurbet et al.18 reported improved scores and ROM. The rapid onset of pain relief after PRF
shoulder function and pain relief for at least 12 weeks after lesioning of the SSN apparently makes it easier for patients
each PRF procedure. Luleci et al.27 performed a prospective to tolerate rehabilitation therapy. However, more research is
open-label study of 57 patients, and they showed statisti- needed before we can recommend widespread adoption of
cal reduction in pain that lasted for up to 6 months after this method as a treatment for AC.
use of PRF lesioning of the SSN. Several other studies also The results of this study must be viewed in light of its
reported that this procedure provided relief from pain and limitations. First, the number of cases was relatively small,
improved the ROM for 12 weeks to 6 months.17,19,26 Although although we had more subjects than most other published
the mechanism of PRF lesioning is not yet clear, one study studies in this field. Second, we did not have measurements
indicated that it may modify neuronal membranes.36 Other of active ROM, limiting the generalizability of our results.
studies of the rat cervical dorsal root ganglia suggested that Third, it would be better to follow up patients for a longer
PRF lesioning had a selective effect on small-diameter C- time. The significant improvement of the VAS and SPADI
and A-δ fibers37 and increased c-Fos immunoreactivity in scores as well as some PROM by week 12 in the interven-
the dorsal horn up to 1 week after treatment.38 tion group compared with the control group continued to
Multiple studies about the use of an SSNB under UG also be observed. Therefore, we believe the effect could continue
reported a significant effect.29–32,39,40 Abdelshafi et al.40 stud- for >12 weeks if the follow-up period is extended. The high
ied patients with chronic shoulder pain and reported that an dropout rate is the fourth limitation. Nine patients dropped
SSNB under UG combined with physical therapy provided out of the study in each group for personal reasons such as
greater relief from pain, an improved SPADI score, and health problems, family issues, or busy schedules rather than

690   
www.anesthesia-analgesia.org anesthesia & analgesia
because of a lack of effect. In fact, most of the patients dropped 5. Vorster W, Lange CP, Briët RJ, Labuschagne BC, du Toit DF, Muller
out in the early period (before the fourth week), and there- CJ, de Beer JF. The sensory branch distribution of the suprascapu-
lar nerve: an anatomic study. J Shoulder Elbow Surg 2008;17:500–2
fore the missing data would not influence our final results. In
6. Karataş GK, Meray J. Suprascapular nerve block for pain relief
addition, the missing data were omitted from the statistical in adhesive capsulitis: comparison of 2 different techniques.
analysis. Finally, it would be better to use a sham procedure or Arch Phys Med Rehabil 2002;83:593–7
an active non-PRF intervention in the control group. 7. Brown DE, James DC, Roy S. Pain relief by suprascapular
nerve block in gleno-humeral arthritis. Scand J Rheumatol
1988;17:411–5
CONCLUSIONS 8. Emery P, Bowman S, Wedderburn L, Grahame R. Suprascapular
This study shows that PRF lesioning of the SSN under UG nerve block for chronic shoulder pain in rheumatoid arthritis.
provides faster onset and longer term pain relief as well as BMJ 1989;299:1079–80
decreases disability in patients with AC. Furthermore, PRF 9. Vecchio PC, Adebajo AO, Hazleman BL. Suprascapular
nerve block for persistent rotator cuff lesions. J Rheumatol
lesioning can be easily repeated if the pain recurs because
1993;20:453–5
there is no destruction of tissue. This method is also con- 10. Jones DS, Chattopadhyay C. Suprascapular nerve block for the
venient and safe under UG. The application of PRF lesion- treatment of frozen shoulder in primary care: a randomized
ing of the SSN under UG appears to have clinical value, trial. Br J Gen Pract 1999;49:39–41
although additional large long-term studies are needed to 11. Cosman ER. A comment on the history of the pulsed radio-
frequency technique for pain therapy. Anesthesiology
confirm our results. E
2005;103:1312
12. Martin DC, Willis ML, Mullinax LA, Clarke NL, Homburger
DISCLOSURES JA, Berger IH. Pulsed radiofrequency application in the treat-
Name: Yung-Tsan Wu, MD. ment of chronic pain. Pain Pract 2007;7:31–5
Contribution: This author helped design the study, con- 13. Van Zundert J, Patijn J, Kessels A, Lamé I, van Suijlekom H, van
Kleef M. Pulsed radiofrequency adjacent to the cervical dorsal
duct the study, data collection, data analysis, and manuscript root ganglion in chronic cervical radicular pain: a double blind
preparation. sham controlled randomized clinical trial. Pain 2007;127:173–82
Attestation: Yung-Tsan Wu attests to the integrity of the origi- 14. Cohen SP, Sireci A, Wu CL, Larkin TM, Williams KA, Hurley
nal data and the analysis reported in this manuscript. RW. Pulsed radiofrequency of the dorsal root ganglia is supe-
Name: Cheng-Wen Ho, MD, PhD. rior to pharmacotherapy or pulsed radiofrequency of the inter-
Contribution: This author helped data analysis and manu- costal nerves in the treatment of chronic postsurgical thoracic
pain. Pain Physician 2006;9:227–35
script preparation.
15. Shah RV, Racz GB. Pulsed mode radiofrequency lesioning of
Attestation: Cheng-Wen Ho attests to the integrity of the origi- the suprascapular nerve for the treatment of chronic shoulder
nal data and the analysis reported in this manuscript. pain. Pain Physician 2003;6:503–6
Name: Yi-Ling Chen, MD. 16. Rohof OJ. Radiofrequency treatment of peripheral nerves. Pain
Contribution: This author helped conduct the study and data Pract 2002;2:257–60
collection. 17. Keskinbora K, Aydinli I. [Long-term results of suprascapu-
lar pulsed radiofrequency in chronic shoulder pain]. Agri
Attestation: Yi-Ling Chen approved the final manuscript.
2009;21:16–21
Name: Tsung-Ying Li, MD. 18. Gurbet A, Türker G, Bozkurt M, Keskin E, Uçkunkaya N, Sahin
Contribution: This author helped conduct the study and data S. [Efficacy of pulsed mode radiofrequency lesioning of the
collection. suprascapular nerve in chronic shoulder pain secondary to
Attestation: Tsung-Ying Li approved the final manuscript. rotator cuff rupture]. Agri 2005;17:48–52
Name: Kuei-Chen Lee, PT, MS. 19. Liliang PC, Lu K, Liang CL, Tsai YD, Hsieh CH, Chen HJ.
Contribution: This author helped conduct the study and data Pulsed radiofrequency lesioning of the suprascapular nerve
for chronic shoulder pain: a preliminary report. Pain Med
collection.
2009;10:70–5
Attestation: Kuei-Chen Lee approved the final manuscript. 20. Huang CC, Tsao SL, Cheng CY, Hsin MT, Chen CM. Treating
Name: Liang-Cheng Chen, MD, MS. frozen shoulder with ultrasound-guided pulsed mode radiofre-
Contribution: This author helped design the study, con- quency lesioning of the suprascapular nerve: two cases. Pain
duct the study, data collection, data analysis, and manuscript Med 2010;11:1837–40
preparation. 21. Eyigor C, Eyigor S, Korkmaz OK, Uyar M. Intra-articular cor-
ticosteroid injections versus pulsed radiofrequency in painful
Attestation: Liang-Cheng Chen attests to the integrity of the
shoulder: a prospective, randomized, single-blinded study.
original data and the analysis reported in this manuscript and Clin J Pain 2010;26:386–92
is the archival author. 22. Korkmaz OK, Capaci K, Eyigor C, Eyigor S. Pulsed radiofre-
This manuscript was handled by: Spencer S. Liu, MD. quency versus conventional transcutaneous electrical nerve
stimulation in painful shoulder: a prospective, randomized
REFERENCES study. Clin Rehabil 2010;24:1000–8
1. Cleland J, Durall CJ. Physical therapy for adhesive capsulitis: 23. Gofeld M, Restrepo-Garces CE, Theodore BR, Faclier G. Pulsed
systematic review. Physiotherapy 2002;88:450–7 radiofrequency of suprascapular nerve for chronic shoulder
2. Binder AI, Bulgen DY, Hazleman BL, Roberts S. Frozen pain: a randomized double-blind active placebo-controlled
shoulder: a long-term prospective study. Ann Rheum Dis study. Pain Pract 2013;13:96–103
1984;43:361–4 24. Kim JS, Nahm FS, Choi EJ, Lee PB, Lee GY. Pulsed radiofre-
3. Shaffer B, Tibone JE, Kerlan RK. Frozen shoulder. A long-term quency lesioning of the axillary and suprascapular nerve in cal-
follow-up. J Bone Joint Surg Am 1992;74:738–46 cific tendinitis. Korean J Pain 2012;25:60–4
4. Ritchie ED, Tong D, Chung F, Norris AM, Miniaci A, 25. Gabrhelik T, Michalek P, Adamus M, Mikova M, Dolecek L.
Vairavanathan SD. Suprascapular nerve block for postopera- Effect of pulsed radiofrequency therapy on the suprascapu-
tive pain relief in arthroscopic shoulder surgery: a new modal- lar nerve in shoulder pain of various aetiology. Ir J Med Sci
ity? Anesth Analg 1997;84:1306–12 2010;179:369–73

September 2014 • Volume 119 • Number 3 www.anesthesia-analgesia.org 691

Ultrasound-Guided Pulsed Radiofrequency for Adhesive Capsulitis

26. Kane TP, Rogers P, Hazelgrove J, Wimsey S, Harper GD. Pulsed 35. Sluijter ME, van Kleef M, Barendse GA, Weber W. Thermal
radiofrequency applied to the suprascapular nerve in painful transport during radiofrequency current therapy of the inter-
cuff tear arthropathy. J Shoulder Elbow Surg 2008;17:436–40 vertebral disc. Spine (Phila Pa 1976) 1998;23:745
27. Luleci N, Ozdemir U, Dere K, Toman H, Luleci E, Irban A. 36. Cosman ER Jr, Cosman ER Sr. Electric and thermal field effects in
Evaluation of patients’ response to pulsed radiofrequency tissue around radiofrequency electrodes. Pain Med 2005;6:405–24
treatment applied to the suprascapular nerve in patients 37. Hamann W, Abou-Sherif S, Thompson S, Hall S. Pulsed
with chronic shoulder pain. J Back Musculoskelet Rehabil radiofrequency applied to dorsal root ganglia causes a
2011;24:189–94 selective increase in ATF3 in small neurons. Eur J Pain 2006;
28. Sites BD, Brull R. Ultrasound guidance in peripheral regional 10:171–6
anesthesia: philosophy, evidence-based medicine, and tech- 38. Van Zundert J, de Louw AJ, Joosten EA, Kessels AG, Honig W,
niques. Curr Opin Anaesthesiol 2006;19:630–9 Dederen PJ, Veening JG, Vles JS, van Kleef M. Pulsed and con-
29. Harmon D, Hearty C. Ultrasound-guided suprascapular nerve tinuous radiofrequency current adjacent to the cervical dorsal
root ganglion of the rat induces late cellular activity in the dor-
block technique. Pain Physician 2007;10:743–6
sal horn. Anesthesiology 2005;102:125–31
30. Peng PW, Wiley MJ, Liang J, Bellingham GA. Ultrasound-
39. Herring AA, Stone MB, Nagdev A. Ultrasound-guided supra-
guided suprascapular nerve block: a correlation with fluoro-
scapular nerve block for shoulder reduction and adhesive cap-
scopic and cadaveric findings. Can J Anaesth 2010;57:143–8
sulitis in the ED. Am J Emerg Med 2011;29:963.e1–3
31. Børglum J, Bartholdy A, Hautopp H, Krogsgaard MR, Jensen K. 40. Abdelshafi ME, Yosry M, Elmulla AF, Al-Shahawy EA, Adou
Ultrasound-guided continuous suprascapular nerve block for Aly M, Eliewa EA. Relief of chronic shoulder pain: a com-
adhesive capsulitis: one case and a short topical review. Acta parative study of three approaches. Middle East J Anesthesiol
Anaesthesiol Scand 2011;55:242–7 2011;21:83–92
32. Taskaynatan MA, Ozgul A, Aydemir K, Koroglu OO, Tan AK. 41. Lee HJ, Lim KB, Kim DY, Lee KT. Randomized controlled trial
Accuracy of ultrasound-guided suprascapular nerve block mea- for efficacy of intra-articular injection for adhesive capsulitis:
sured with neurostimulation. Rheumatol Int 2012;32:2125–8 ultrasonography-guided versus blind technique. Arch Phys
33. MacDermid JC, Solomon P, Prkachin K. The Shoulder Pain and Med Rehabil 2009;90:1997–2002
Disability Index demonstrates factor, construct and longitudi- 42. Blanchard V, Barr S, Cerisola FL. The effectiveness of corticoste-
nal validity. BMC Musculoskelet Disord 2006;7:12 roid injections compared with physiotherapeutic interventions
34. Palmer ML, Epler EM. Fundamentals of Musculoskeletal for adhesive capsulitis: a systematic review. Physiotherapy
Assessment Techniques. 2nd ed. New York: Lippincott, 1998 2010;96:95–107

692   
www.anesthesia-analgesia.org anesthesia & analgesia