High-intensity focused ultrasound for

potential treatment of polycystic

ovary syndrome: toward a
noninvasive surgery
Islam A. Shehata, M.D., M.Sc.,a,b John R. Ballard, Ph.D.,a Andrew J. Casper, Ph.D.,a Leah J. Hennings, D.V.M.,c
Erik Cressman, M.D., Ph.D.,d and Emad S. Ebbini, Ph.D.a
a
College of Science and Engineering, University of Minnesota, Minneapolis, Minnesota; b Department of Diagnostic and
Interventional Radiology, Cairo University, Cairo, Egypt; c Department of Pathology, University of Arkansas for Medical
Sciences, Little Rock, Arkansas; and d Department of Diagnostic Radiology, University of Minnesota, Minneapolis,
Minnesota

Objective: To investigate the feasibility of using high-intensity focused ultrasound (HIFU), under dual-mode ultrasound arrays
(DMUAs) guidance, to induce localized thermal damage inside ovaries without damage to the ovarian surface.
Design: Laboratory feasibility study.
Setting: University-based laboratory.
Animal(s): Ex vivo canine and bovine ovaries.
Intervention(s): DMUA-guided HIFU.
Main Outcome Measure(s): Detection of ovarian damage by ultrasound imaging, gross pathology, and histology.
Result(s): It is feasible to induce localized thermal damage inside ovaries without damage to the ovarian surface. DMUA provided sen-
sitive imaging feedback regarding the anatomy of the treated ovaries and the ablation process. Different ablation protocols were tested,
and thermal damage within the treated ovaries was histologically characterized.
Conclusion(s): The absence of damage to the ovarian surface may eliminate many of the complications linked to current laparoscopic
ovarian drilling (LOD) techniques. HIFU may be used as a less traumatic tool to perform LOD.
(Fertil Steril 2014;101:54551. 2014 by American Society for Reproductive Medicine.) Use your smartphone
Key Words: Polycystic ovary syndrome (PCOS), high-intensity focused ultrasound (HIFU), to scan this QR code
dual-mode ultrasound arrays (DMUA), laparoscopic ovarian drilling (LOD), infertility and connect to the
discussion forum for
this article now.*
Discuss: You can discuss this article with its authors and with other ASRM members at http://
fertstertforum.com/shehataia-ultrasound-polycystic-ovary-syndrome/ * Download a free QR code scanner by searching for QR
scanner in your smartphones app store or app marketplace.

P
olycystic ovarian syndrome hirsutism, anovulation secondary to tus and were associated with a reduc-
(PCOS) is a complex endocrine hyperandrogenism, and increased tion in levels of circulating androgen
disorder that is classied as the incidence of type II diabetes mellitus (1). For treatment of anovulatory infer-
most common gynecological condition owing to increased insulin resistance (1). tility, clomiphene citrate is the rst-line
in women of reproductive age (1) and First-line treatment of PCOS is medical therapy to induce ovulation in
the most common cause of anovulatory conservative, including weight loss, addition to the aforementioned mea-
infertility (2). It is estimated that about an exercise program, and the use of sures (1). About 20% of patients do
8%10% of women worldwide have insulin-sensitizing agents like metfor- not respond to clomiphene citrate (5).
PCOS (3, 4). The main clinical min. These measures were found to Those patients are eligible for the
manifestations of PCOS are obesity, reduce the risk of type II diabetes melli- second-line treatment, which may be
gonadotropins or surgery (1). Gonado-
Received August 4, 2013; revised September 26, 2013; accepted October 14, 2013; published online tropins are expensive, require intensive
November 26, 2013.
I.A.S. has nothing to disclose. J.R.B. has nothing to disclose. A.J.C. has nothing to disclose. L.J.H. has monitoring, and are associated with a
nothing to disclose. E.C. has nothing to disclose. E.S.E. has nothing to disclose. higher incidence of ovarian hyperstim-
Reprint requests: Islam A. Shehata, M.D., M.Sc., 12, Abd El-Aziz Eldreny Street, Apt #18, El-Manyal
11451, Cairo, Egypt (E-mail: islamhifu@gmail.com). ulation syndrome (OHSS) and multiple
ovulation (6). The current surgical op-
Fertility and Sterility Vol. 101, No. 2, February 2014 0015-0282/$36.00 tion is laparoscopic ovarian drilling
Copyright 2014 American Society for Reproductive Medicine, Published by Elsevier Inc.
http://dx.doi.org/10.1016/j.fertnstert.2013.10.023 (LOD), using electrocautery or laser.

VOL. 101 NO. 2 / FEBRUARY 2014 545

ORIGINAL ARTICLE: REPRODUCTIVE SCIENCE
The main rationale behind this surgery is to thermally ablate
some of the androgen-producing ovarian tissue to relieve the
hyperandrogenism linked to PCOS. The surgery is also
thought to enhance the response of treated ovaries to intrinsic
gonadotropins through different mechanisms (7). LOD is less
expensive than gonadotropins and provides comparable
success rates with less incidence of OHSS and multiple preg-
nancy (6, 8). However, the surgery is traumatic and can be
complicated by postoperative pelvic adhesions or accidental
bowel injury (9, 10). On this basis, a noninvasive form of
thermal therapy capable of producing thermal damage
within ovaries while minimizing ovarian trauma may be a
new approach for the treatment of PCOS.
High-intensity focused ultrasound (HIFU) is a noninva-
sive technology used for tissue ablation. Intensied ultrasonic
beams can be focused to a point using therapeutic trans-
ducers. This causes damage at the focus without signicant
damage before or after the focal point. HIFU damage is mainly
achieved through a thermal effect, although other effects like
cavitation and mechanical effects also contribute to the focal
damage (11). HIFU is more known in the medical community
for its use in ablating body tumors, particularly uterine
broids and prostate cancer (1113).
Both ultrasound and magnetic resonance imaging (MRI)
are currently used to guide and monitor HIFU therapy. MRI
monitoring makes use of shifts in proton resonance frequency
in response to focal temperature rise to construct thermal
maps. These maps were found to accurately track temperature
changes at the focal point and are accepted clinically as
indicators of tissue ablation (13, 14). On the other hand,
clinical ultrasound monitoring depends on detection of the
echogenic changes that develop as a result of bubble
activity at the focal point. This bubble activity is attributed
to the inertial cavitation and tissue boiling that are
commonly encountered during tissue ablation (13, 15).
Current clinical ultrasound-guided HIFU systems use two
separate transducers, one for imaging and another one for
HIFU therapy. The two transducers are spatially aligned along
the same axes (11); otherwise any misalignment can cause
spatial misregistration between the imaged echogenic
changes and the actual ablated tissue.
Dual-mode ultrasound arrays (DMUAs) are advanced
transducers that use the same elements for simultaneous im-
aging and treatment by HIFU. In other words, this paradigm
provides a single transducer that does imaging and therapy
at the same time. This strategy ensures accurate and complete
spatial registration between the imaged and targeted tissue.
With a high imaging frame rate, reaching up to 1,000 frames
per second, DMUAs can provide sensitive feedback regarding
the ongoing ablation process and evolving echogenic
changes. Inherent DMUA imaging capabilities are provided
through two imaging modes, single transmit focus (STF)
and synthetic aperture (SA). Broadly speaking, STF imaging
provides precise feedback regarding the ablation process at
the focus itself, while SA imaging provides better assessment
of the perifocal region owing to an increased eld of view.
DMUAs have been used for in vivo testing of other potential
medical applications (16), and information about the full ca-
pabilities of DMUAs is available in the literature (17, 18).
546
In this pilot study, we hypothesized that HIFU can be used
as a less traumatic alternative tool to achieve the same ther-
apeutic outcome as LOD. The main objective of this study
was to demonstrate the feasibility of using HIFU, under guid-
ance and monitoring of DMUA, to cause localized damage
within the ovarian stroma without damage to the ovarian sur-
face. The secondary objectives were to tune the exposure
levels needed to achieve well-localized damage in ovaries
and to histologically characterize the damage in the ovarian
tissue.
MATERIALS AND METHODS
HIFU Transducer
A 3.5-MHz, 64-element DMUA (Imasonic) with a central
fenestration was used in this study. The transducer elements
were arranged in two groups, 32 elements each, above and
below the central fenestration. Through the central fenestra-
tion, a 7.5-MHz linear diagnostic transducer (HST 15-8/20,
Ultrasonix) was introduced and spatially aligned with the
DMUA. This setup allowed for the collection and comparison
of both DMUA and conventional B-mode ultrasound imaging
data. This integrated transducer design increased the ability to
localize and monitor tissue ablation using multiple imaging
modes.
Ovaries
No institutional review board or Institutional Animal Care
and Use Committee approval were required for these
ex vivo animal experiments. Ex vivo canine and bovine
ovaries were obtained from the American Preclinical Services
and Lindenfelser's Farm Fresh Meat, respectively. After
dissection, ovaries were placed in a holder lled with molten
gelatin. Only part of the ovary, sufcient to hold the ovary in
place, was immersed in the gelatin, while the ovarian surface
through which HIFU therapy took place was kept above the
level of the molten gelatin. Gelatin was then left in the refrig-
erator for 45 minutes to solidify.
Experimental Setup and Ablation Protocols
The tissue holder containing the gelatin-xed ovaries was
attached to a three-dimensional positioning system (Parker
Daedal) and placed in a tank of degassed deionized water in
front of the integrated DMUA transducer (Supplemental
Fig. 1). After identication of the internal anatomical details
of the ovaries using both DMUA and conventional B-mode
imaging, the tissue holder was adjusted to obtain the best
plane to start the treatment (usually the plane showing size-
able ovarian follicles to be used as landmarks during tissue
cutting). HIFU was then applied under imaging guidance.
Two ablation protocols were tested in these experiments.
The rst was a single-shot protocol to produce a single large
lesion. The other protocol, the grid protocol, was used to stack
smaller lesions side by side in a grid design that was intended
to form a conuent large ablation zone. To tune the exposure
parameters for both protocols, different exposure times were
tested (2 seconds, 1 second, 750 ms, and 500 ms) with an esti-
mated focal intensity range of 4,1004,700 W/cm2.
VOL. 101 NO. 2 / FEBRUARY 2014

Echogenic changes developing at the foci were used as an
indication of tissue ablation.
After tissue ablation, treated ovaries were manually sliced
and photographed to document the resultant tissue damage
before being submitted to histology analysis.
Histology Analysis
Slices of treated ovaries were processed and embedded in
parafn, sectioned at 4 mm, stained with hematoxylin and
eosin, and scanned at 200 total magnication (Aperio CS,
Aperio/Leica). Images were evaluated by a veterinary pathol-
ogist without reference to treatment status of ovaries. The
study pathologist had more than 7 years of experience in
identifying and characterizing thermal damage in tissues.
RESULTS
DMUA imaging provided high-resolution imaging for depict-
ing the internal architecture of the ovaries and monitoring the
ablation process. Both DMUA and conventional B-mode im-
aging correlated well, with clear identication of the ovarian
anatomical detail as shown in Supplemental Figure 2.
In canine ovaries, no signicant echogenic changes were
detected at the 500 ms sonication, and gross tissue analysis
did not reveal any sizeable tissue damage. This was consid-
ered as an underexposure. The 1- and 2-second exposures
showed marked echogenic changes, denoting an overexpo-
sure. Gross histology revealed areas of damage corresponding
to the echogenic changes, measuring approximately 3.2  2.7
mm and 6.7  4.8 mm along the maximum cross sections for
the 1- and 2-second exposures, respectively (Fig. 1). The 750
ms sonication resulted in well-localized echogenic changes
and was used to construct a 3  3 grid in canine ovaries.
DMUA imaging was sensitive in detecting echogenic changes
at the focus, as shown in Figure 2.
For the sake of histology analysis, two sets of HIFU dam-
age were done in a bovine ovary, separated by about 7 mm.
Each set was composed of three HIFU shots using a 1-
second exposure time each, with 1-mm separation between
shots. Histological changes ranged from subtle cellular
changes including cytoplasmic eosinophilia and loss of
nuclear detail to ablation with tissue shrinkage, marked
hyperosinophilia, and focal mineralization. Mild changes
were most clearly evident in blood vessels, in which vacuoli-
zation and hyalinization of smooth muscle were variably pre-
sent. Within broader regions of mild damage, roughly circular
to ellipsoid areas of ablated tissue were present. These areas
appeared to correspond to foci of HIFU application (Fig. 3).
The ovarian surface after treatment was kept intact, with
no evidence of damage to the exterior of treated ovaries as
evaluated by comparative visual inspection before and after
HIFU ablation (Fig. 4).
DISCUSSION
There is currently no way to deliver the electrocautery or laser
energy needed for thermal ablation to the inside of the ovary
except by perforating through the surface. This approach
causes undesirable, but apparently unavoidable, damage to
VOL. 101 NO. 2 / FEBRUARY 2014
Fertility and Sterility
the ovarian surface. This brings many procedure-related com-
plications. In addition to the risk of hemorrhage associated
with surgical procedures, damage to the ovarian surface has
been thought to induce postoperative pelvic adhesions. These
adhesions may introduce a mechanical barrier for conception
that may be even more serious than the original hormonal
disturbance. In study published by Mercorio et al. (9), 60%
of the patients developed postoperative pelvic adhesions as
detected by a second-look laparoscopy.
The technique proposed herein may help to eliminate
many of the complications associated with the current LOD
techniques. Results of this study showed that it is feasible to
induce HIFU damage inside the ovary without damage to
the surface, thus potentially avoiding the aforementioned
complications related to ovarian surface injury. In addition,
this therapeutic technique introduces another potentially
signicant advantage, specically, the ability to image the
treatment area at high resolution using the same therapeutic
array, that is, the DMUA approach. DMUAs, with their ability
to provide simultaneous imaging and treatment with HIFU,
provided high-resolution imaging to guide and monitor the
treatment. As opposed to the current laparoscopic approaches
in which surgeons have clear visualization of the exterior of
the ovaries, DMUAs can provide valuable data regarding the
internal ovarian anatomical details. This may help surgeons
to plan and design the ablation in view of the specic anat-
omy of the ovary under treatment. This may also allow
surgeons to localize the damage to the ovarian cortex to avoid
damaging the blood vessels that ramify from the medulla. In
addition, imaging of the ovary can give a visual perception of
the ovarian volume. This may allow for customizing the
extent of the ablation in relation to the volume of the treated
ovary, which may help to reduce the risk of ovarian failure
associated with ovarian overablation.
HIFU was proposed for treatment of PCOS in 1998 (19).
However, the current study differs from the study published
by Moussatov et al. (19) in important ways. First, Moussatov
and colleagues (19) did their experimentations on liver tissue,
based on the fact that the acoustic attenuation coefcients of
liver and ovaries are almost the same. While this may be true
as regards the acoustic properties, there are signicant
anatomical differences between these tissues. Liver tissue is
more of a homogenous parenchyma that is traversed by rather
sizeable blood vessels and bile ducts. In contrast, ovaries have
distinct corticomedullary differentiation, and ovarian vessels
inside the ovaries are much smaller than the hepatic vessels
and more abundant in the medulla. Thus, identifying the
cortico-medullary differentiation in ovaries through imaging
may improve treatment planning and safety. From a prospec-
tive clinical point of view, the heat sink effect within ovaries
may be less pronounced than in liver owing to the discrep-
ancy in blood ow and local perfusion between these tissues.
This point needs to be considered during translation of this
technique to in vivo studies.
The current study differs from the previous study with
regard to the image guidance method. Moussatov and col-
leagues (19) proposed the use of a coaxial standard vaginal
ultrasound probe and/or MRI for monitoring of the ablation
process. In the context of using ultrasound for treatment
547
ORIGINAL ARTICLE: REPRODUCTIVE SCIENCE

FIGURE 1

HIFU-induced tissue damage in canine ovaries. (A) The 2 seconds of sonication resulted in a zone of tissue damage (arrows) measuring about 6.7 
4.8 mm. (B) The 1-second exposure resulted in a smaller zone of damage (arrows) measuring about 3.2  2.7 mm in maximum cross sections.
Shehata. HIFU for potential treatment of PCOS. Fertil Steril 2014.

monitoring, DMUAs may be the best t as single tool for
imaging and therapy, owing to the inherent registration of
imaging and treatment coordinates as well as the high sensi-
tivity for detection of the echogenic changes. MRI can be
helpful for treatment monitoring in animal experiments, but
it may not be as practical in a future clinical practice for
two reasons. If this treatment is applied through a laparo-
scopic approach, then an MRI-compatible surgical suite will
be required, which is expensive and not widely available. If
HIFU is applied through a transvaginal approach, as with
standard transvaginal ultrasound (TVUS), then there may be
some concerns regarding the space limitations of the MRI
gantry. This is especially true with the patient most probably
in a lithotomy position. An integrated transvaginal trans-
ducer, composed of a DMUA and standard diagnostic probe,
may be the best design for a transvaginal approach. Such
an integrated transducer can allow the treating physicians
to rst use the standard diagnostic transducer for a wide gen-
eral pelvic scan and then switch to the DMUA to conduct and
monitor HIFU therapy.
This study was not without limitations. It was challenging
to detect the HIFU damage against the pale background of
ex vivo ovaries. This was even more difcult in bovine
ovaries, where histology was used to detect the damage.
Although ex vivo tissue will exhibit ablation in response to
thermal damage, cellular and tissue changes are often more
subtle than in corresponding in vivo, perfused tissues. In
addition, treatments were planned to ablate tissue between
follicles, to allow for more reliable estimation of damage.
Perifollicular tissue is expected to exhibit different thermal
behavior than follicles, especially cystic follicles, which
have a high uid content. Untreated control ovaries were
not evaluated histologically.
We have used the hyperechogenic changes developing at
the focal point as an indication of ovarian tissue ablation.
Theoretically speaking, ultrasound beams are strongly scat-
tered and attenuated by the bubbles developing at the focal
point (20). Based on this, the echogenic changes may over-
estimate the extent of actual ablated tissue. Since these echo-
genic changes are transient in nature and frequently have
irregular contours, they may not be sufciently accurate to
estimate the exact extent of coagulative necrosis (21). Never-
theless, from a clinical point of view, ultrasound-guided
HIFU was successfully used for the treatment of several

FIGURE 2

DMUA monitoring of HIFU therapy. (A) DMUA imaging showing well-localized echogenic changes (arrows) that developed after a 750-ms
sonication time. (B) Corresponding tissue slice showing localized damage (arrows) that correlated well with the echogenic changes detected by
DMUA imaging. Note the high DMUA imaging resolution that showed the hypoechoic follicles (labeled F) with good correlation with the actual
gross anatomy in panel B.
Shehata. HIFU for potential treatment of PCOS. Fertil Steril 2014.

548 VOL. 101 NO. 2 / FEBRUARY 2014

 Fertility and Sterility

FIGURE 3

Histology analysis of HIFU damage in bovine ovaries. (A) Ultrasound image showing the design of HIFU therapy in bovine ovary. Two sets of HIFU
lesions (cigar shaped) were placed close to an ovarian follicle (labeled F) as a landmark. The arrow shows the direction of HIFU beam. (B) Echogenic
changes (outlined) were seen by the end of sonication for set no. 2. (C) Hematoxylin and eosin staining of the treated ovary (40 magnication)
showing a rather ellipsoid area of tissue damage corresponding to one of the HIFU shots. (D) 200 magnication of the damaged area (above the
line) showing evidence of tissue ablation in the form of hypereosinophilia (coagulation), nuclear pyknosis, and tissue shrinkage.
Shehata. HIFU for potential treatment of PCOS. Fertil Steril 2014.

malignant tumors using the echogenic changes as an indica-
tion of tissue ablation. Satisfactory clinical outcomes have
been reported for the treatment of liver and prostate cancer,
for examples (11). In a study published by Leslie et al. (22), a
clear correlation was established between the extent of the
echogenic changes during ablation of liver tumors and post-
operative MRI estimates of ablated tissue. Thus it was
concluded that the echogenic changes provided safe predica-
tion of the extent of the ablated zone (22). Putting this
debate in consideration, we believe that hyperechogenic
changes may be adequately used for treatment monitoring
for this particular application. Ovaries in PCOS are not ma-
lignant. Thus, even if the hyperechogenic changes slightly
overestimate the extent of actual ablation, this will not be
as risky as with treatment of cancerous tissues. In addition,
although the aim of the treatment is to ablate some of the
ovarian tissue, clinical improvement will be essentially
achieved through reduction in the levels of circulating
androgen. This point needs further evaluation through
in vivo experimentation to explore whether a clear correla-
tion can be established between the extent of tissue ablation
and the decreases in the level of circulating androgen.
Finally, alternative MRI monitoring may not be as practical
from a clinical point of view, as discussed before.
In vivo studies are also required to further investigate the
treatment protocols and probe design. There is high exibility
as regards the way by which HIFU can be used in potential
future clinical application. A laparoscopic approach using
an interstitial dual-mode transducer can be used for this pur-
pose. The other approach, which can eliminate the surgical
risk of laparoscopy, is a transvaginal approach. It is techni-
cally feasible to miniaturize the DMUA to t on a transvaginal
probe so that the treatment can be conducted as in a standard
TVUS setting. Moussatov et al. (19) estimated the distance
from the tip of a TVUS probe to midovary in patients with
PCOS to be approximately 2 cm. Taking into consideration
the constraints of designing a transvaginal probe, a miniatur-
ized DMUA can be designed with a focal length of about 3 cm.
The phased-array nature of DMUAs can allow for steering the
focus for additional millimeters in the axial and lateral

VOL. 101 NO. 2 / FEBRUARY 2014 549

ORIGINAL ARTICLE: REPRODUCTIVE SCIENCE
FIGURE 4
Ovarian surface before and after HIFU therapy. (A) Canine ovary xed in gelatin before therapy. (B) The same ovary after HIFU therapy showing
smooth surface with no evidence of damage, despite the damage induced within the ovarian stroma, shown in Figure 2.
Shehata. HIFU for potential treatment of PCOS. Fertil Steril 2014.
directions, which can give more exibility as regards localiza-
tion of the HIFU focus. A transvaginal approach allows easy
and direct access to the ovaries. Even in rare cases where
bowel loops may interpose between the HIFU probe and the
ovary, angulation and tilting of the probe may help to get
these loops away from the path of the HIFU beam. It is impor-
tant not to have any bowel loops between the transducer and
the treated ovary because gas inside bowel loops may result in
serious reections of the HIFU beam, causing accidental
collateral damage.
Several animal models are available for studying PCOS
(23). To investigate the transvaginal approach in particular,
a sheep animal model may be needed. Corriedale and Suffolk
female sheep were used by Pimentel et al. (24) to investigate
transvaginal ovarian drilling and may be considered for
investigating this application as well.
As for the ablation protocols, two protocols were tested
in this preliminary study. The single-lesion high-exposure
protocol produced larger lesions in a shorter time. However,
it is known that such overexposures are less controlled in
terms of the lesion size and localization. They can carry
the risk of inducing surface damage. The fact that there
was no surface damage in this study's experiments should
be accepted with caution since these ex vivo experiments
were done in degassed water at room temperature. This pro-
vided continuous cooling of the ovarian surface, which will
not be the case for in vivo studies. The other protocol is the
grid protocol, which requires more time to stack the individ-
ual small lesions side by side along a specic grid design.
Although time consuming, this protocol is more controlled
and can allow for the design of different sizes and shapes
of ablation. In our experiments, the 750-ms exposure was
the best to produce such controlled damage in canine
ovaries. Of note, the lengthy nature of the grid protocol
can be shortened by using a multiple focusing technique,
in which the transducer can produce multiple simultaneous
foci to ablate larger zones (25).
In conclusion, the results of this pilot study showed that it
is feasible to use HIFU to induce localized thermal damage
inside ovaries under DMUA guidance without signicant
550
damage to the ovarian surface in this model system. In clinical
practice, treatment preferences are inuenced by many fac-
tors, including clinical assessment and regional guidelines.
The aim of this therapeutic technique is to give physicians a
potentially easier and safer tool to perform therapy. The ther-
apeutic approach proposed herein can be also investigated for
treatment of other localized ovarian diseases, like
endometriosis.
REFERENCES
1. American College of Obstetricians and Gynecologists. ACOG practice
bulletin. Polycystic ovary syndrome. Obstet Gynecol 2009;114:93649.
2. Adams J, Polson DW, Franks S. Prevalence of polycystic ovaries in women with
anovulation and idiopathic hirsutism. Br Med J (Clin Res Ed) 1986;293:3559.
3. Azziz R, Carmina E, Dewailly D. The Androgen Excess and PCOS Society
criteria for the polycystic ovary syndrome: the complete task force report.
Fertil Steril 2009;91:45688.
4. Shannon M, Wang Y. Polycystic ovary syndrome: a common but often un-
recognized condition. J Midwifery Womens Health 2012;57:22130.
5. Imani B, Eijkemans MJC, teVelde ET, Habbema JDF, Fauser BCJM. Predictors
of patients remaining anovulatory during clomiphene citrate induction of
ovulation in normogonadotropic oligoamenorrheic infertility. J Clin Endocri-
nol Metab 1998;83:23615.
6. Farquhar C, Brown J, Marjoribanks J. Laparoscopic drilling by diathermy or
laser for ovulation induction in anovulatory polycystic ovary syndrome. Co-
chrane Database Syst Rev 2012:CD001122.
7. Hendriks ML, Ket JC, Hompes PG, Homburg R, Lambalk CB. Why does
ovarian surgery in PCOS help? Insight into the endocrine implications of
ovarian surgery for ovulation induction in polycystic ovary syndrome. Hum
Reprod Update 2007;13:24964.
8. Fernandez H, Morin-Surruca M, Torre A, Faivre E, Defeux X, Gervaise A.
Ovarian drilling for surgical treatment of polycystic ovarian syndrome: a
comprehensive review. Reprod Biomed Online 2011;22:55668.
9. Mercorio F, Mercorio A, Di Spiezio Sardo A, Barba GV, Pellicano M, Nappi C.
Evaluation of ovarian adhesion formation after laparoscopic ovarian drilling
by second-look minilaparoscopy. Fertil Steril 2008;89:122933.
10. Api M, Gorgen H, Cetin A. Laparascopic ovarian drilling in polycystic ovary
syndrome. Eur J Obstet Gynecol Reprod Biol 2005;119:7681.
11. Shehata IA. Treatment with high intensity focused ultrasound: secrets re-
vealed. Eur J Radiol 2012;81:53441.
12. Jenne JW, Preusser T, G unther M. High-intensity focused ultrasound: princi-
ples, therapy guidance, simulations and applications. Z Med Phys 2012;22:
31122.
VOL. 101 NO. 2 / FEBRUARY 2014

13. Zhou YF. High intensity focused ultrasound in clinical tumor ablation. World
J Clin Oncol 2011;2:827.
14. Hynynen K. MRIgHIFU: a tool for image-guided therapeutics. J Magn Reson
Imaging 2011;34:48293.
15. Rabkin BA, Zderic V, Crum LA, Vaezy S. Biological and physical mechanisms
of HIFU-induced hyperecho in ultrasound images. Ultrasound Med Biol
2006;32:17219.
16. Shehata IA, Ballard JR, Casper AJ, Liu D, Troutman M, Ebbini ES. Feasibility of
targeting atherosclerotic plaques by high-intensity focused ultrasound: an
in vivo study. J Vasc Interv Radiol 2013;24:18807.e2.
17. Ebbini ES, Yao H, Shrestha A. Dual-mode ultrasound phased arrays for
image-guided surgery. Ultrason Imaging 2006;28:20120.
18. Casper AJ, Liu D, Ballard JR, Ebbini ES. Real-time implementation of a dual-
mode ultrasound array system: in vivo results. IEEE Trans Biomed Eng 2013;
60:27519.
19. Moussatov AG, Baker AC, Duck FA. A possible approach to the treatment of
polycystic ovarian syndrome using focused ultrasound. Ultrasonics 1998;36:
893900.
VOL. 101 NO. 2 / FEBRUARY 2014
Fertility and Sterility
20. Makin IR, Mast TD, Faidi W, Runk MM, Barthe PG, Slayton MH. Miniaturized
ultrasound arrays for interstitial ablation and imaging. Ultrasound Med Biol
2005;31:153950.
21. Goldberg SN, Gazelle GS, Mueller PR. Thermal ablation therapy for focal ma-
lignancy: a unied approach to underlying principles, techniques, and diag-
nostic imaging guidance. Am J Roentgenol 2000;174:32331.
22. Leslie T, Ritchie R, Illing R, Ter Haar G, Phillips R, Middleton M, et al. High-
intensity focused ultrasound treatment of liver tumours: post-treatment
MRI correlates well with intra-operative estimates of treatment volume. Br
J Radiol 2012;85:136370.
23. Walters KA, Allan CM, Handelsman DJ. Rodent models for human polycystic
ovary syndrome. Biol Reprod 2012;86:149.
24. Pimentel AM, Kobayashi D, Kliemann LM, Franjdlich R, Capp E, Corleta HV.
Transvaginal ultrasound ovarian diathermy: sheep as an experimental
model. J Ovarian Res 2012;5:1.
25. Ebbini ES, Cain CA. Multiple-focus ultrasound phased-array pattern synthe-
sis: optimal driving-signal distributions for hyperthermia. IEEE Trans Ultrason
Ferroelectr Freq Control 1989;36:5408.
551
ORIGINAL ARTICLE: REPRODUCTIVE SCIENCE

SUPPLEMENTAL FIGURE 1

The experimental setup. (A) Gelatin-xed ovaries (labeled O) are placed within a tissue holder (labeled H) above a tank containing degassed water
(labeled W) that serves as a coupling medium for ultrasound propagation. (B) The tissue holder is immersed in the degassed water in front of the
integrated DMUA transducer (labeled T). The arrow shows the direction of the HIFU beam.
Shehata. HIFU for potential treatment of PCOS. Fertil Steril 2014.

551.e1 VOL. 101 NO. 2 / FEBRUARY 2014

 Fertility and Sterility

SUPPLEMENTAL FIGURE 2

Correlation between conventional B-mode and DMUA imaging. (A) Conventional B-mode imaging obtained from the diagnostic transducer within
the central fenestration of the integrated transducer showing a peripheral slightly hypoechoic ovarian cortex (labeled C) and a central relatively
more echogenic medulla (labeled M). (B) Corresponding DMUA imaging showing the same corticomedullary differentiation as in panel A.
Shehata. HIFU for potential treatment of PCOS. Fertil Steril 2014.

VOL. 101 NO. 2 / FEBRUARY 2014 551.e2