Human Reproduction vol.14 no.3 pp.
618–621, 1999
Ovarian stromal echogenicity in women with normal
and polycystic ovaries
W.M.Buckett1,4, R.Bouzayen2, K.L.Watkin3,
T.Tulandi1 and S.L.Tan1
1Department
of Obstetrics and Gynecology, McGill University,
Royal Victoria Hospital, 687 Avenue des Pins Ouest, 2Département
d’obstetrique et gynécologie, Université de Montréal, Hôpital Ste
Justine and 3Department of Radiology, McGill University,
Royal Victoria Hospital, Montréal, Québec, Canada H3A 1A1
4To whom correspondence should be addressed
Since the widespread use of transvaginal ultrasound to
diagnose polycystic ovary syndrome (PCOS), a cardinal
feature has been shown to be the presence of a bright,
highly echogenic stroma. This is usually assessed subjec-
tively. The objective of this study was to determine whether
ovarian stromal echogenicity when measured objectively
actually differed between women with polycystic ovaries
and those with normal ovaries. A total of 67 women
underwent a detailed ultrasound assessment before con-
sidering assisted conception treatment. Ovarian morpho-
logy was assessed and total ovarian volume, stromal volume,
peak stromal blood flow velocity and mean stromal echo-
genicity were measured. The stromal index (ratio of mean
stromal echogenicity to mean echogenicity of the entire
ovary) and total stromal echogenicity were also calculated.
Ovarian volume, stromal volume, and stromal peak blood
flow velocity were all significantly higher in ovaries from
women with PCOS. There was no difference in the mean
stromal echogenicity, although the stromal index was signi-
ficantly greater in women with polycystic ovaries. The
apparent subjective increase in stromal echogenicity in
women with polycystic ovaries, as exemplified by the greater
stromal index, is due to a combination of the increased
volume of ovarian stroma and the significantly lower mean
echogenicity of the entire ovary in these women.
Key words: echogenicity/ovary/polycystic ovary syndrome/
stroma/ultrasound
Introduction
Polycystic ovary syndrome (PCOS) is the most frequent
disorder of ovarian function in women of reproductive age
(Franks, 1989). Diagnosis of the syndrome is generally based
clinically on the presence of amenorrhoea/oligomenorrhoea or
symptoms of hyperandrogenism, and biochemically on the
presence of elevated serum luteinizing hormone (LH) and
androgen concentrations. More recently, transvaginal ultra-
sound examination of ovarian morphology has been used to
help make the diagnosis.
618
Swanson et al. (1981) were the first to describe the ultrasound
findings associated with PCOS. The classical image is that of
enlarged ovaries containing an increased number of small
follicles (2–10 mm) encircling the ovarian cortex like a string
Downloaded from https://academic.oup.com/humrep/article/14/3/618/632879 by guest on 03 April 2024
of pearls, although this is sometimes incomplete, and the
presence of an increased bright echogenic stroma (Adams
et al., 1985). The presence of hypertrophic ovarian stroma has
been used by many groups to distinguish between PCOS and
normal ovaries (Conway et al., 1989; Dewailly et al., 1993).
The degree of echogenicity of the ovarian stroma is usually
assessed subjectively but this is open to observer bias where
other features of PCOS are already seen. Methods of quantita-
tively measuring ultrasound image data, often termed textural
feature analysis, have been useful in identifying diffuse diseases
of the liver (Raeth et al., 1985) and brain (Barr et al.,
1995). Using these recent advances in ultrasound software, the
brightness, or echogenicity, of the ovarian stroma can be
determined objectively by measuring the intensity level of the
ultrasound pixels within the stroma displayed on an ultrasonic
image. The mean echogenicity of a given area can then be
calculated.
The objective of this study was to determine whether the
ovarian stromal echogenicity measured objectively differed
significantly between women with PCOS and those with
normal ovaries.
Materials and methods
Transvaginal pelvic ultrasound and Doppler examination were
performed during the early follicular phase of the menstrual cycle
(days 2–4) on 67 consecutive patients who were not taking any
medication as part of routine assessment prior to undergoing in-vitro
fertilization for a variety of indications at the McGill Reproductive
Center.
All ultrasound examinations were performed firstly using a 5 MHz
endovaginal probe with colour and pulsed Doppler facilities (Acuson
XP 128/10®; Acuson Corp., Mountain View, CA, USA), and then
using a 7.5 MHz endovaginal probe with histogram measurement
of echogenicity (Aloka SSD 2000®; Aloka Co., Tokyo, Japan).
All ultrasound examinations were performed by one of the authors
(W.B.).
A systematic examination of the morphology of the uterus and
ovaries was performed using previously described criteria (Zaidi
et al., 1995a; Tan et al., 1996). An ultrasound diagnosis of PCOS was
made when there were greater than 10 small cysts/follicles (2–8 mm
diameter) around a dense core of stroma. These ultrasound findings
were invariably accompanied with an increased ovarian volume. Any
ovaries which contained ovarian cysts (over 12 mm mean diameter)
or ultrasound evidence of endometriomas were excluded from the
analysis.
Intra-ovarian blood flow was assessed by pulsed Doppler examina-
© European Society of Human Reproduction and Embryology

Figure 1. The outline of, and histogram displaying the echogenicity within, the entire ovary (A) and the ovarian stroma (B) in a woman
with normal ovaries.
Figure 2. The outline of, and histogram displaying the echogenicity within, the entire ovary (A) and the ovarian stroma (B) in a woman
with polycystic ovaries.
tion of blood vessels in the ovarian stroma, as previously described
(Zaidi et al., 1995b, 1996).
The mean echogenicity was defined as the sum of the product of
each intensity level (varying from 0–63) and the number of pixels
for that intensity concentration divided by the total number of pixels
in the measured area, as follows:
mean 5 (Σ.xi.fi)/n
where n 5 total number of pixels in the measured area
x 5 intensity level (from 0 to 63)
f 5 number of pixels corresponding to that level.
The mean echogenicity of the entire ovary and of the ovarian
stroma were separately calculated. The spread of intensities was also
displayed by histogram, where the horizontal axis indicated the
different intensity concentrations and the vertical axis the number of
pixels at each intensity level (Figures 1 and 2).
The total stromal echogenicity was calculated as the product of
the mean stromal intensity and the total number of pixels measured
in the same two-dimensional ultrasound image of the stroma.
The stromal index was then calculated by dividing the mean
stromal echogenicity by the mean echogenicity of the entire ovary in
order to correct for cases where the gain was adjusted to allow
optimal image definition. The stromal index was therefore more than
one if the mean stromal echogenicity was greater than the mean
echogenicity of the entire ovary.
Ovarian stromal echogenicity
Downloaded from https://academic.oup.com/humrep/article/14/3/618/632879 by guest on 03 April 2024
Serum follicle stimulating hormone (FSH), luteinizing hormone
(LH), testosterone, and dihydroepiandrosterone sulphate (DHEAS)
were also measured in the early follicular phase on the day of the
baseline ultrasound scan.
Normally distributed data were expressed as means with 95%
confidence limits and compared using Student’s unpaired two-tailed
t-test. Non-parametric data were expressed as medians with inter-
quartile range and compared using the Mann–Whitney rank sum test.
Results
Of a total of 67 patients who underwent ultrasound assessment
(one of whom had had a previous oophorectomy), five patients
were excluded. Reasons were the presence of endometrioma
(3) or of a simple ovarian cyst .15 mm mean diameter (2).
Of the 123 ovaries included in the study, 46 (37%) had an
ultrasound diagnosis of PCOS.
The total ovarian volume, stromal volume, and peak stromal
blood flow (as assessed by colour Doppler) were all signi-
ficantly higher (P , 0.0001; P , 0.05; and P , 0.001
respectively) in the women with PCOS compared with those
with normal ovaries (Table I).
There was no significant difference in the mean stromal
echogenicity between the two groups, and although the total
619
W.M.Buckett et al.

ovaries which suggests that there is no intrinsic difference in

Table I. Ovarian volume, stromal volume, and peak stromal blood flow the nature of the stroma itself. This is an important finding as
velocity in normal and polycystic ovaries. Values are means with 95%
confidence intervals (CI) or medians with interquartile ranges (IQR) a highly echogenic stroma is a cardinal feature in the ultrasound
diagnosis of PCOS (Conway et al., 1989; MacDougall et al.,
Normal ovaries Polycystic ovaries P value 1992; Dewailly et al., 1993). The significant difference in the
Total ovarian volume (ml) 6.31 (5.60–7.02) 9.94 (8.29–11.59) ,0.0001 stromal index would suggest that the impression of a highly
Stromal volume (ml) 1.90 (1.18–2.78) 2.41 (2.15–3.22) ,0.02 echogenic stroma in PCOS is primarily due to a visual
Peak stromal blood 5.98 (4.84–7.12) 10.45 (8.73–12.17) ,0.0001
flow velocity (cm/s)
perception of the difference in echogenicities between the
stroma and the ovary as a whole with its multiple small cysts.
Clearly the greater the number and the larger the size of the
cysts the greater the stromal index would be since the mean
Table II. Mean stromal echogenicity, mean echogenicity of the entire ovary,
stromal index, and total stromal echogenicity in women with normal ovaries total ovarian echogenicity would be lower.
and PCOS. Values expressed as medians with interquartile ranges (IQR) The greater total stromal echogenicity would also suggest

Downloaded from https://academic.oup.com/humrep/article/14/3/618/632879 by guest on 03 April 2024

that the subjective impression of highly echogenic stroma in
Normal ovaries Polycystic ovaries P value
women with PCOS may be partly due to the increased stromal
Mean stromal echogenicity 20.0 (17.9–24.3) 19.6 (17.8–23.8) NS volume. The reason that statistical significance was not reached
Mean total ovarian 17.6 (15.8–22.4) 16.7 (14.6–18.6) ,0.05 is probably because of the small sample size.
echogenicity
Total stromal echogenicity 5844 (4574–9559) 6717 (4718–11004) NS Histological examination of ovaries from women with PCOS
Stromal index 1.10 (1.06–1.21) 1.20 (1.12–1.31) ,0.0001 has shown a five-fold increase in the sub-cortical (medulla)
stroma (Hugheson, 1982). The findings of the present study
NS 5 not significant. that the mean stromal echogenicity is comparable in the two
PCOS 5 polycystic ovary syndrome. groups of women are consistent with the histological evidence
that although the stromal volume is increased, there is no other
change which could account for an increased echogenicity
Table III. Concentrations of serum follicle stimulating hormone (FSH),
serum luteinizing hormone (LH), serum testosterone and serum per se.
dihyroepiandrosterone sulphate (DHEAS) and LH:FSH ratio in women with It has been suggested that vascular endothelial growth factor
normal and polycystic ovaries. Values are expressed as means with 95% (VEGF) has a role in the maintenance of perifollicular blood
confidence intervals (CI).
flow (Van Blerkom et al., 1997) and recent evidence shows a
Normal ovaries Polycystic ovaries P value positive correlation between VEGF and ovarian stromal blood
Serum FSH (IU/l) 8.89 (7.46–10.32) 6.85 (6.10–7.60) ,0.05
flow velocities in women with ultrasound diagnosed polycystic
Serum LH (IU/l) 5.19 (4.29–6.09) 6.78 (5.51–8.05) ,0.05 ovaries and PCOS (Agrawal et al., 1998). This increased
LH:FSH ratio 0.59 (0.51–0.67) 1.01 (0.80–1.22) ,0.01 vascularity, possibly mediated by VEGF, is therefore probably
Serum testosterone (nmol/l) 1.09 (0.95–1.23) 1.37 (1.20–1.54) ,0.05
Serum DHEAS (mmol/l) 3.70 (3.23–4.17) 5.27 (3.77–6.77) ,0.05
responsible for the formation of increased stroma and the
ultimate phenotype associated with PCOS.
In conclusion, the ultrasonic measurement of mean ovarian
stromal echogenicity was higher in the PCOS group this did stromal echogenicity adds little to the ultrasound diagnosis of
not reach statistical significance (Table II). PCOS and cannot be recommended in the routine ultrasound
The stromal index (ratio of mean stromal echogenicity to assessment of the pelvis. Although the stromal echogenicity
mean echogenicity of the entire ovary) was higher in the PCOS appears subjectively brighter, this is primarily a reflection of
group (P , 0.0001), partly as a result of the reduced mean the difference between the echogenicity of the stroma compared
echogenicity of the entire ovary in the PCOS group (P , 0.05). with that of the entire ovary. There was an increased stromal
The endocrine parameters of PCOS were all significantly index as well as an increased stromal volume in women with
raised in the PCOS group (Table III), although the serum FSH polycystic ovaries. Although the measurement of ovarian and
was lower than in the control group.
stromal volume and intra-ovarian blood flow were not primary
outcomes in this study, we recommend their routine measure-
Discussion ment in women undergoing induction of ovulation because of
This study confirms the results of previous studies that there their predictive value in the responsiveness of the ovary
is a high incidence of ultrasound diagnosed PCOS in women to hormonal stimulation (MacDougall et al., 1992; Zaidi
undergoing in-vitro fertilization (IVF) for a variety of indica- et al., 1996b).
tions (Polson et al., 1988; MacDougall et al., 1992). We found
that the total ovarian volume and the stromal volume are
increased in women with PCOS. We also confirmed that the References
peak ovarian stromal blood flow velocity is significantly higher Adams, J., Polson, D.W., Abdulwahid, N. et al. (1985) Multifollicular ovaries:
in polycystic ovaries than in normal ovaries (Zaidi et al., clinical and endocrine features and response to pulsatile gonadotrophin
releasing hormone. Lancet, ii, 1375–1378.
1995b; Battaglia et al., 1997).
Agrawal, R., Sladkevicius, P., Engmann, L. et al. (1998) Serum vascular
Interestingly, the mean stromal echogenicity was no higher endothelial growth factor concentrations and ovarian stromal blood flow
in women with PCOS compared with women with normal are increased in women with polycystic ovaries. Hum. Reprod., 13, 651–655.

620
 Ovarian stromal echogenicity

Barr, L.L., McCullough, P.J., Ball, W.S. Jr et al. (1995) Quantitative

sonographic feature analysis of clinical infant hypoxia: a pilot study. Am.
J. Neuroradiol., 17, 1025–1031.
Battaglia, C., Artini, P.G., Ganazzi, A.D. et al. (1997) Color Doppler analysis
in oligo- and amennorrheic women with polycystic ovary syndrome.
Gynecol. Endocrinol., 11, 105–110.
Conway, G.S., Honour, J.W. and Jacobs, H.S. (1989) Heterogenicity of the
polycystic ovary syndrome: clinical, endocrine, and ultrasound features in
556 patients. Clin. Endocrinol., 30, 459–470.
Dewailly, D., Duhamel, A., Rober, Y. et al. (1993) Interrelationship between
ultrasonography and biology in the diagnosis of polycystic ovarian
syndrome. Ann. N. Y. Acad. Sci., 687, 206–216.
Franks, S. (1989) Polycystic ovary syndrome: a changing perspective. Clin.
Endocrinol., 31, 87–120.
Hugheson, P.E. (1982) Morphology and morphogenesis of the Stein–Leventhal
ovary and of so-called ‘hyperthecosis’. Obstet. Gynecol. Surv., 37, 59–77.
MacDougall, M.J., Tan, S.L., Balen, A.H. and Jacobs, H.S. (1992) A controlled

Downloaded from https://academic.oup.com/humrep/article/14/3/618/632879 by guest on 03 April 2024

study comparing patients with and without polycystic ovaries undergoing
in-vitro fertilization. Hum. Reprod., 8, 233–237.
Polson, D.W., Wadsworth, J., Adams, J. and Franks, S. (1988) Polycystic
ovaries: a common finding in normal women. Lancet, ii, 70.
Raeth, U., Schlaps, D., Limburg, B. et al. (1985) Diagnostic accuracy of
computerized B-scan texture analysis and conventional ultrasonography in
diffuse parenchymal and malignant liver disease. J. Clin. Ultrasound, 13,
87–99.
Swanson, M., Sauerbrie, E.E. and Cooperberg, P.L. (1981) Medical
implications of ultrasonically detected polycystic ovaries. J. Clin.
Ultrasound, 9, 219–222.
Tan, S.L., Zaidi, J., Campbell, S. et al. (1996) Blood flow changes in the
ovarian and uterine arteries during the normal menstrual cycle. Am. J.
Obstet. Gynecol., 175, 625–631.
Van Blerkom, J., Antczak, M. and Schrader, R. (1997) The development
potential of the human oocyte is related to the dissolved oxygen content
of follicular fluid: association with vascular endothelial growth factor
concentrations and perifollicular blood flow characteristics. Hum. Reprod.,
12, 1047–1055.
Zaidi, J., Jurovic, D., Campbell, S. et al. (1995a) Description of circadian
rhythm in uterine artery blood flow during the peri-ovulatory period. Hum.
Reprod., 10, 1642–1646.
Zaidi, J., Campbell, S., Pitroff, R. et al. (1995b) Ovarian stromal blood flow
in women with polycystic ovaries – a possible new marker for diagnosis?
Hum. Reprod., 10, 1992–1996.
Zaidi, J., Tan, S.L., Pitroff, R. et al. (1996a) Blood flow changes in the intra-
ovarian arteries during the peri-ovulatory period – relationship to the time
of day. Ultrasound Obstet. Gynecol., 7, 135–140.
Zaidi, J., Barber, J., Kyei-Mensah, A. et al. (1996b) Relationship of ovarian
stromal blood flow at the baseline ultrasound scan to subsequent follicular
response in an in vitro fertilization program. Obstet. Gynecol., 88, 779–784.

Received on June 18, 1998; accepted on December 2, 1998

621