The 3D vascular status of the follicle after HCG ...

Human Reproduction ppH.1–u5,m20.06Reprod. Advance Access published December 19, 2006 doi:10.1093/humrep/del472

The 3D vascular status of the follicle after HCG
administration is qualitatively rather than quantitatively
associated with its reproductive competence

Daniel H.Mendez Lozano1, Nelly Frydman2, Jean M.Levaillant1, Ste´phanie Fay1,
Rene´ Frydman1 and Renato Fanchin1,3

1Department of Obstetrics and Gynecology and Reproductive Medicine and 2Department of Biology and Genetics of Reproduction,
INSERM Unit 782, Clamart, Universite´ Paris XI, Le Kremlin-Biceˆtre, France
3To whom correspondence should be addressed at: Department of Obstetrics and Gynecology and Reproductive Medicine, Hoˆpital
Antoine Be´cle`re, 157, rue de la Porte de Trivaux, 92141 Clamart, France. E-mail: [email protected]

BACKGROUND: The objective of this study was to determine whether the vascular status of a single pre-ovulatory
follicle is associated quantitatively and/or qualitatively with its reproductive competence. METHODS: We studied 61
monofollicular IVF-embryo transfer cycles. Just before single oocyte retrieval, follicle vascularization was detected by
transvaginal power-Doppler, 3-dimensionally reconstructed, and analysed quantitatively by coloured/gray voxel ratio
[vascularization index (VI)] and qualitatively by blood cell displacement [flow index (FI)] calculation. Cycles were
sorted in two sets of two groups: low VI ( 8%, n 5 44) and high VI (>8%, n 5 17); low FI ( 30, n 5 22) and
high FI (>30, n 5 39). RESULTS: Patients’ characteristics, fertilization rates, and embryo morphology were compar-
able in all groups. In contrast, clinical pregnancy rates/oocyte retrieval (4% versus 33%, P < 0.009) and implantation
rates (11% versus 50%, P < 0.04) were markedly poorer in the low as compared to the high FI groups, respectively,
but remained similar between the low and the high VI groups (22% versus 23% and 38% versus 44%, respectively).
CONCLUSIONS: A qualitative (FI) rather than quantitative (VI) relationship exists between vascular status and
functional quality of the follicle after HCG administration.

Key words: blood flow/embryo implantation/follicular vascularization/power Doppler

Introduction by regulating oxygen supply to the oocytes (Gaulden, 1992;
Van Blerkom et al., 1997).
During the menstrual cycle, LH-driven luteinization is
marked by a profound increase in the vascularization of the During the last 15 years, numerous studies using
pre-ovulatory follicle (Bourne et al., 1991; Collins et al., bi-dimensional Doppler technologies attempted to non-inva-
1991; Campbell et al., 1993; Janson, 1975; Neulen et al., 1995; sively confirm the possible relationship between vasculariza-
Brannstrom et al., 1998; Wulff et al., 2001; Phan et al., 2006). tion and reproductive competence of ovarian follicles. Most
Throughout this process, some granulosa cell-derived products of them concurred to show a positive association between
are likely to promote (vascular endothelial growth factor, fibro- follicle vascularization and ovarian responsiveness to
blast growth factor and angiopoietins) (Gruemmer et al., 2005) controlled ovarian stimulation (Zaidi et al., 1996; Bassil
or inhibit (acid hyaluronic and 2-methoxyestradiol) (Hazzard et al., 1997; Engmann et al., 1999), oocyte retrieval rate
et al., 1999; Koga et al., 2000) blood vessel outgrowth from (Nargund et al., 1996a,b; Oyesanya et al., 1996; Engmann
the thecal vascular plexus towards the inner compartments of et al., 1999) or embryo morphology (Nargund et al., 1996a,b;
the corpus luteum. This suggests that granulosa cells play an Van Blerkom et al., 1997; Huey et al., 1999).
important role in the recruitment of blood vessels into the
early corpus luteum (Redmer et al., 1991; Zheng et al., 1993; Moreover, the issue of whether follicular vascularization is
Redmer and Reynolds, 1996), which implies that luteinized associated (Nargund et al., 1996a,b; Chui et al., 1997;
follicle vascularization, granulosa cell activity and possibly Bhal et al., 1999; Coulam et al., 1999; Engmann et al., 1999)
oocyte competence may be interrelated events (Picton et al., or not (Tekay et al., 1995; Kan et al., 2006; Palomba et al.,
1998). In addition, it is possible that the vascularization status 2006; Ng et al., 2006) with embryo implantation remains
of ovarian follicles influences their reproductive competence controversial. The lack of concordance among published data
may be, at least in part, attributed to methodological limitations

# The Author 2006. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. Page 1 of 5
All rights reserved. For Permissions, please email: [email protected]

D.H.Mendez Lozano et al.

such as subjectivity and insufficiency of measurements, lack of Pharmaceuticals, Courtaboeuf, France). Oocyte retrieval
direct traceability between follicle vascularization and the fate failure was defined by a negative oocyte recovery after three
of oocyte and embryos, and disregard of blood flow dynamics consecutive follicular flushings. Top quality embryo was
in favour of merely quantitative information. defined on Day 2 as those having no multinucleated blasto-
meres, four or five blastomeres and ,20% anucleated frag-
Hence, in the present study, we used an original model that ments (Van Royen et al., 1999). Embryo transfer was
allows the precise tracking of the oocyte – embryo fate to verify performed 2 days after oocyte retrieval. Luteal phase was
whether a quantitative and/or qualitative relationship exists supported with micronized progesterone (Estima Ge´, Effik
between the vascular status of the luteinized dominant follicle, Pharmaceuticals, Bie`vres, France, 600 mg day21) administered
objectively assessed in the whole follicle wall, and its repro- daily by vaginal route starting on the evening of ET.
ductive competence.
Follicular vascularization assessment
Materials and Methods
Just before oocyte retrieval at approximately 9:30 AM, vascu-
Patients larization status of the pre-ovulatory follicle was assessed by
We prospectively studied 61 infertile women, 24 – 40 years transvaginal power Doppler using a 3.7 – 9.3 MHz ultrasound
of age, undergoing 61 consecutive IVF-embryo transfer probe (RIC5-9H, General Electric Medical Systems, Paris,
(IVF-ET) in mono-dominant follicle cycles from July 2005 France) equipped with a 1468-angle rotating head that
to December 2005. All women met the following inclusion allowed data acquisition for further 3-dimensional (3D) recon-
criteria: (i) both ovaries present and devoid of morphological struction. Patients were first examined with a B-mode ultra-
abnormalities; (ii) regular menstrual cycles between 25 and sound to identify the maximal follicular diameter and to rule
35 days; (iii) no current or past diseases affecting the ovaries out the interference of extra-ovarian organs that could hinder
or gonadotrophin or sex steroid secretion, clearance or adequate image quality. Data acquisition was then performed
excretion; (iv) no clinical signs of hyperandrogenism; (v) according to similar, pre-established 3D power Doppler set-
body mass indexes (BMIs) ranging from 18 to 25 kg m22; tings: (i) high tissue harmonic frequency; (ii) colour gain at
(vi) no current hormone therapy; (vii) adequate visualization 24; (iii) zoom at 1.3 fold; (iv) wall motion filter at 2 and (v)
of both ovaries in transvaginal ultrasound scans; and (viii) no pulse repetition frequency at 0.6 kHz. To reduce acquisition
smoking. Aetiologies of infertility were sperm abnormalities time, the swept volume was reduced to an angle of 608, in
(45.9%), tubal abnormalities (24.6%), unexplained infertility which the total follicle was included.
(24.6%) and endometriosis (4.9%). An informed consent was
obtained from all patients and this investigation received the Afterwards, the pre-ovulatory follicle was entirely
approval of our internal Institutional Review Board. 3D-reconstructed offline using a virtual organ computer aided
analysis (VOCAL) technology (General Electric Medical
IVF-ET protocol Systems, Paris, France). For this, the region of interest (ROI)
On cycle day 3, women underwent blood samplings by in the 3D volume, which corresponded to the outer follicle
venipuncture at approximately 9 AM for serum estradiol borders, was manually and progressively (every 158) set. Sub-
(E2), progesterone and LH measurements. Later in the sequently, follicle vascularization was assessed quantitatively
morning, the number and the sizes of early antral follicles and qualitatively in the ROI. For quantitative assessment, we
were assessed by ultrasound equipped with a tissue harmonic used the vascularization index (VI) that is calculated by the
imaging system (Thomas and Rubin, 1998). From cycle day colour voxels/(total voxels–background voxels) ratio. The VI
8 onward, the selection of the dominant follicle was monitored reflects the number of vessels in the follicle as a percentage of
by ultrasound. When its mean diameter exceeded 12 mm, to the total 3D volume. For the qualitative assessment, we used
prevent the risk of premature LH peak and to control further the flow index (FI), a parameter that assesses blood cell displace-
follicular maturation, women were administered s.c. 0.5 mg ment and reflects the mean intensity of colour voxels in a scale
of a GnRH antagonist (cetrorelix acetate; Cetrotide 0.25 mg, going from 0 (minimum intensity) to 100 (maximum intensity)
Serono Pharmaceuticals, Boulogne, France) and 150 IU of (Pairleitner et al., 1999). All ultrasound scans and vasculariza-
HMG (Menopur, Ferring Pharmaceuticals, Gentilly, France) tion analysis were performed by a single operator (D.H.M.L.),
daily until the day of HCG (Gonadotrophine Chorionique who was unaware of oocyte retrieval outcome and IVF-ET
“Endo”, Organon Pharmaceuticals, Saint-Denis, France) results. Intraobserver reproducibility of VI and FI, measure-
administration. Women received a 5000-IU HCG injection ments, assessed by the intraclass correlation coefficients
i.m. when the dominant follicle diameter exceeded 16 mm. (ICCs), were 0.89 and 0.82 for VI and FI, respectively, thereby
confirming data from other investigators (Jarvela et al., 2003;
The oocyte was retrieved approximately 34 h after HCG Merce et al., 2005). In addition, an accessory parameter of fol-
administration. Under transvaginal ultrasound guidance, the licle vascularization (vascularization–flow index) that expresses
follicular fluid from the single pre-ovulatory follicle was aspi- a combination of quantitative and qualitative information on the
rated using a 10-ml syringe. The aspiration needle was kept follicular vascular status was also investigated.
steady inside the follicle until the oocyte was found and iso-
lated. In case of negative oocyte recovery, sequential follicular Definition of follicle vascularization groups
flushings were performed using 10-ml syringes filled with 3 ml
of a balanced salt solution (Tyrode’s salt solution, Eurobio Cycles were sorted into two sets of two groups according to VI
(quantitative index) and FI (qualitative index) values, based on
Page 2 of 5

3D follicle vascularization and IVF-ET outcome

the best likelihood ratio provided by receiver operating charac- the ICC. To characterize the predictability of follicle vascular-
teristic (ROC) analysis. Therefore, the low VI group included ization, we employed relative risk (RR) and its 95% confidence
cycles in which VI values remained 8% (n ¼ 44) and the intervals (CIs) calculation as well as logistic regression analy-
high VI group included those in which they were .8% sis (to exclude explicative variables) and ROC analysis. The
(n ¼ 17), whereas the low FI group comprised cycles in present study was powered to detect anticipated differences
which FI values were 30 (n ¼ 22) and the high FI group of 25% in embryo implantation rates at .80% power at 0.05
included those in which FI remained .30 (n ¼ 39). significance level. A P-value of ,0.05 indicated statistically
significant differences.
Hormonal measurements
Results
Serum E2 and progesterone levels were determined by an auto-
mated multi-analysis system using a chemiluminescence tech- Patients’ characteristics and cycle monitoring data
nique (Advia-Centaur, Bayer Diagnostics, Puteaux, France). Patients’ characteristics and cycle monitoring data in the low
For E2, lower detection limit was 15 pg ml21, linearity up to and high VI and FI groups are detailed in Table I. As shown,
1000 pg ml21, and intra- and inter-assay coefficients of variation ages of patients, BMI values, menstrual cycle lengths, antral
(CVs) were 8% and 9%, respectively. For progesterone, lower follicle counts on Day 3 were comparable within each set of
detection limit was 0.1 ng ml21, linearity up to 60 ng ml21, follicle vascularization groups. Aetiologies of infertility were
and intra- and inter-assay CVs were 8% and 9%, respectively. also similar in both sets of follicle vascularization groups.
Serum LH levels were measured by an immunometric technique Similarly, the day of HCG administration and pre-ovulatory
using an Amerlite kit (Ortho Clinical Diagnostics, Strasbourg, follicle size, endometrial thickness and serum E2, progesterone
France). Lower limit of detection was 0.1 mIU ml21 and intra- and LH levels of the day of HCG were not statistically different
and inter-assay CVs were 5% and 7%, respectively, for LH. between low and high VI cycles and low and high FI cycles.

Statistical analysis Overall, in the 61 IVF-ET cycles studied, mean VI values
were at 7.8 + 0.6% and varied from 0.7% to 32.3% whereas
The measures of central tendency and variability were, respect- mean FI values were 33.4 + 0.8 and varied from 17.4 to
ively, the mean and SEM when data distribution was normal, 48.7. We observed a positive correlation between VI and FI
and the median and the ranges when normality could not be values (r ¼ 0.48; P ,0.0001). In line with data described in
ascertained. Normality distribution of the data was assessed Table I, continuous statistical analysis failed to show any sig-
with the Kolmogorov – Smirnov test. Unpaired data were com- nificant relationship between VI or FI values and patients’
pared with the unpaired Student’s t-test or the Mann – Whitney characteristics or cycle monitoring data.
test, when appropriate. Relationship between two continuous
variables was assessed by correlation when they were indepen- Embryology and IVF-ET outcome data
dent from each other and by simple regression when there was Embryology data and IVF-ET outcome data are summarized in
a dependency relationship. The Spearman’s test was used to Table II. Both sets of groups remained similar with regard to
determine if coefficients of correlation (r) were significantly positive oocyte retrieval rate, fertilization rates and prevalence
different from zero. The chi-square and Fisher’s exact test of top quality embryos. Yet, whereas clinical pregnancy (gesta-
were used to compare categorical variables. As mentioned tional sac observed at ultrasound scans at around 7 weeks of
before, the reproducibility of measurements was evaluated by

Table I. Patients’ characteristics and cycle monitoring data in the low and high vascularization index (VI) and flow index (FI) groups

Parameter Low VI (n ¼ 44) High VI (n ¼ 17) P Low FI (n ¼ 22) High FI (n ¼ 39) P

Age (years)a 33.5 + 0.4 (32.6 –34.5)b 33.5 + 0.9 (31.5–35.5) NS 33.7 + 0.7 (35.0 –35.3) 33.4 + 0.5 (32.4–34.4) NS
Body mass index (kg m22)a 21.7 + 0.5 (21.0 –22.6) 21.8 + 0.3 (20.5–22.9) NS 22.2 + 0.5 (21.0–23.4) 21.5 + 0.3 (20.7–22.3) NS
Cycles (days)a 28.4 + 0.3 (27.6 –29.1) 28.4 + 0.5 (27.2–29.5) NS 28.1 + 0.4 (27.1–29.0) 28.5 + 0.3 (27.7–29.3) NS

Indications for IVF-embryo transfer:

Sperm abnormalities (%) 45.4 47.0 NS 45.4 46.1 NS

Tubal abnormalities (%) 25.0 23.5 NS 18.1 28.2 NS

Unexplained infertility (%) 27.2 17.6 NS 31.8 20.5 NS

Endometriosis (%) 2.2 11.7 NS 4.5 5.1 NS
Antral follicle count on day 3a 10.1 + 0.8 (8.3– 11.8)
Day of HCG (days)a 11.8 + 0.4 (11.0 –12.6) 11.2 + 0.9 (9.3–13.1) NS 9.9 + 1.2 (7.3– 12.6) 10.6 + 0.7 (9.1–12.2) NS
Follicle size on the day of HCG (mm)a 17.2 + 0.1 (16.9 –17.6) 13.0 + 0.5 (11.9–14.2) NS 12.2 + 0.6 (10.9–13.5) 12.5 + 0.3 (11.4–12.9) NS
17.1 + 0.2 (16.7–17.6) NS 17.2 + 0.2 (16.6–17.8) 17.2 + 0.1 (16.8–17.5) NS
8.2 + 0.1 (7.8– 8.6) 8.9 + 0.3 (8.1–9.6) NS 8.6 + 0.2 (8.0– 9.2) 8.3 + 0.2 (7.8–8.8)
Endometrial thickness on the 253 + 24 (203– 303) 234 + 20 (190–278) NS 256 + 30 (193– 319) 243 + 23 (194–291) NS
day of HCG (mm)a NS

Estradiol on the day of HCG (pg ml21)a

Progesterone on the day 0.1 + 0.02 (0.1–0.1) 0.2 + 0.07 (0.0–0.4) NS 0.1+ 0.02 (0.0–0.1) 0.1+ 0.03 (0.1– 0.2) NS
of HCG (ng ml21)a 4.1 + 0.4 (3.2– 5.1) 4.2 + 0.7 (2.6–5.8) NS 4.6 + 0.8 (2.9– 6.3) 3.9 + 0.4 (3.0–4.8) NS

LH on the day of HCG (mIU ml21)a

NS, Not significant.
aValues are expressed in mean + SE.
bValues in parentheses are 95% confidence intervals (CIs).

Page 3 of 5

D.H.Mendez Lozano et al.

Table II. Embryology and IVF-embryo transfer outcome data

Parameter Low VI (n ¼ 44) High VI (n ¼ 17) P RR [95% CI] Low FI (n ¼ 22) High FI (n ¼ 39) P RR [95% CI]

Positive oocyte retrieval ratea (%) 86.4 94.2 NS 1.0 [0.9–1.2] 77.3 94.9 NS 1.2 [0.9– 1.5]

Oocyte fertilization rate (%) 93.7 80.0 NS 0.8 [0.6–1.1] 86.6 90.6 NS 1.0 [0.8– 1.3]

Top quality embryos (%) 48.1 50.0 NS 1.0 [0.4–2.1] 51.6 50.6 NS 0.7 [0.3– 1.4]

Clinical pregnancy rate/OR (%) 22.7 23.5 NS 1.0 [0.3–2.8] 4.5 33.3 0.009 7.3 [1.0– 52.3]

Embryo implantation rate (%) 38.4 44.4 NS 1.1 [0.4–2.7] 11.1 50.0 0.04 4.5 [0.6– 29.7]

RR, relative risk; NS, not significant; OR, oocyte retrieval.
aPercentage of cycles in which the oocyte could be recovered.

amenorrhoea) rates/oocyte retrieval and the implantation rates The mechanisms underlying this relationship remain
(total number of gestational sacs  100/total number of unclear. Despite the fact that VI and FI values were signifi-
embryos transferred) were similar in the low and the high VI cantly correlated, it is conceivable that HCG-driven angiogen-
groups, they were higher in the high FI as compared to the esis is not systematically associated with an increase of the
low FI groups (P ¼ 0.009). blood flow through the follicular wall, as far as it may
be reflected by FI values. According to this hypothesis, it is
In line with this, logistic regression analysis indicated that the blood flow dynamics rather than the vascularized extension
only FI values (P ,0.01; 95% CIs, 1.02 – 1.31) were predictive of the follicle wall that plays a determining role in the repro-
of clinical pregnancy, in contrast with patient’s ages (P ¼ 0.47; ductive competence of the follicle. Indeed, functional defects
0.75 – 1.14), BMIs (P ¼0.55; 0.66– 1.24), the day of HCG of follicle perfusion have been shown to alter the organization
administration (P ¼ 0.54; 0.80 – 1.50), serum E2, progesterone and stability of the metaphase spindle through the reduction of
and LH levels on the day of HCG administration (P ¼ 0.29; the oxygen supply to the oocytes (Gaulden, 1992; Van Blerkom
0.98 – 1.00; P ¼ 0.25; 0.18 – 605.90 and P ¼ 0.33; 0.64 – 1.15, et al., 1997), a phenomenon that possibly exerts a detrimental
respectively), follicular size on the day of HCG administration role in the embryo aptitude to implantation. Furthermore,
(P ¼ 0.56; 0.36 – 1.72), and VI values (P ¼ 0.84; 0.68 – 1.02). growing evidence indicates that, in pre-ovulatory follicles,
the oocyte directly activates several physiological processes
Also, VI could not adequately discriminate the likelihood of that occur in its surrounding granulosa cells, including plasmi-
a clinical pregnancy as the area under the ROC curve was 0.56 nogen activator production (Canipari et al., 1995), and LH
(P ¼ 0.45; 95% CIs, 0.39– 0.74). Conversely, FI values receptor (Joyce et al., 1999), kit ligand (Joyce et al., 1999)
showed good predictability as the area under the ROC curve and anti-Mullerian hormone (Salmon et al., 2004) gene
was 0.71 (P ,0.01; 95% CIs, 0.57 – 0.85). Indeed, the best like- expression. Therefore, by extrapolation, we hypothesize that
lihood ratio for FI was 6.8 with a rounded cutoff value of 30 poor quality oocytes are unable to drive some functional vascu-
which corresponds to a sensitivity of 48.9 (95% CIs, 34.1 – lar modifications that the pre-ovulatory follicle has to undergo
63.9%) and a specificity of 92.9 (95% CIs, 66.1– 99.8%). before ovulation. Finally, it is noteworthy that oocyte retrieval
failure rate, a parameter shown to be increased in women with
Discussion follicle quality defects (Zreik et al., 2000), as well as oocyte
fertilization rate and embryo morphology, remained unchanged
The present study aimed at evaluating the possible relationship irrespective of VI or FI values. This suggests that follicle vas-
between vascularization of the pre-ovulatory follicle and its cularization plays a determining role in the outcome of late
reproductive competence. For this, its design required the fol- rather than early functions of the oocyte, such as adequate
lowing methodological characteristics. First, we included embryo genome activation, an event that takes place after the
IVF-ET candidates undergoing single pre-ovulatory follicle third cycle of blastomere division (Braude et al., 1988).
cycles to properly track the link between the originating follicle
and oocyte/embryo outcome. Second, we elected to assess, by a In conclusion, the enhanced aptitude for implantation of
single operator, vascularization on the entire follicle using auto- embryos originated from follicles presenting high FI values
mated 3D reconstruction and sensitive power Doppler technol- suggests that a close relationship exists between perifollicular
ogy, in an improved methodology as compared to previous blood flow dynamics and reproductive competence of
studies, which were limited to the assessment of random luteinized pre-ovulatory follicles. Further, this phenomenon
vessels around the follicle (Tekay et al., 1995; Nargund et al., apparently is independent of oocyte fertilizability and
1996a; Chui et al., 1997; Palomba et al., 2006). Third, we embryo morphology. The mere quantitative identification of
employed an expert computerized system (VOCAL) that perifollicular blood vessels (VI) probably is a much less sensi-
allows the distinct analysis of quantitative and qualitative tive predictor of follicle reproductive competence than the
aspects (Pairleitner et al., 1999) of follicle vascularization. assessment of blood flow dynamics (FI) in the whole follicle
Even when the RR for FI are high, 7.3 for clinical pregnancy wall. These results encourage us to conduct further studies on
(P , 0.001) and 4.5 for implantation rate (P , 0.04), a larger the administration of vasodilators and/or cardio-tonics in an
study must confirm these data. The results of our investigation effort to improve follicle vascularization and, possibly,
indicated that the vascular status of the pre-ovulatory follicle is oocyte quality. Yet, the issue of whether or not the present
qualitatively (FI) but not quantitatively (VI) related to the data could be extrapolated to controlled ovarian
implantation outcome of the corresponding embryo.

Page 4 of 5

3D follicle vascularization and IVF-ET outcome

hyperstimulated cycles deserves further investigation. If extra- Merce LT, Gomez B, Engels V, Bau S and Bajo JM (2005) Intraobserver
and interobserver reproducibility of ovarian volume, antral follicle count,
polation was valid, the possibility of preferentially selecting to and vascularity indices obtained with transvaginal 3-dimensional
ultrasonography, power Doppler angiography, and the virtual organ
transfer embryos ensuing from follicles displaying high FI computer-aided analysis imaging program. J Ultrasound Med 24, 1279–1287.

values may be a logical measure to improve their implantation Nargund G, Bourne T, Doyle P, Parsons J, Cheng W, Campbell S and
Collins W (1996a) Associations between ultrasound indices of follicular
rates and should also be the matter of additional investigation. blood flow, oocyte recovery and preimplantation embryo quality. Hum
Reprod 11, 109– 113.
References
Nargund G, Doyle PE, Bourne TH, Parsons JH, Cheng WC, Campbell S
Bassil S, Wyns C, Toussaint-Demylle D, Nisolle M, Gordts S and Donnez J and Collins WP (1996b) Ultrasound derived indices of follicular blood
(1997) The relationship between ovarian vascularity and the duration of flow before HCG administration and the prediction of oocyte recovery and
stimulation in in-vitro fertilization. Hum Reprod 12, 1240– 1245. preimplantation embryo quality. Hum Reprod 11, 2512– 2517.

Bhal PS, Pugh ND, Chui DK, Gregory L, Walker SM and Shaw RW (1999) Neulen J, Yan Z, Raczek S, Weindel K, Keck C, Weich HA, Marme D
The use of transvaginal power Doppler ultrasonography to evaluate the and Breckwoldt M (1995) Human chorionic gonadotropin-dependent
relationship between perifollicular vascularity and outcome in in-vitro expression of vascular endothelial growth factor/vascular permeability
fertilization treatment cycles. Hum Reprod 14, 939– 945. factor in human granulosa cells: importance in ovarian hyperstimulation
syndrome. J Clin Endocrinol Metab 80, 1967–1971.
Bourne TH, Jurkovic D, Waterstone J, Campbell S and Collins WP (1991)
Intrafollicular blood flow during human ovulation. Ultrasound Obstet Ng EH, Tang OS, Chan CC and Ho PC (2006) Ovarian stromal vascularity is
Gynecol 1, 53– 59. not predictive of ovarian response and pregnancy. Reprod Biomed Online
12, 43–49.
Brannstrom M, Zackrisson U, Hagstrom HG, Josefsson B, Hellberg P, Granberg
S, Collins WP and Bourne T (1998) Preovulatory changes of blood flow in Oyesanya OA, Parsons JH, Collins WP and Campbell S (1996) Prediction of
different regions of the human follicle. Fertil Steril 69, 435–442. oocyte recovery rate by transvaginal ultrasonography and color Doppler
imaging before human chorionic gonadotropin administration in in vitro
Braude P, Bolton V and Moore S (1988) Human gene expression first occurs fertilization cycles. Fertil Steril 65, 806– 809.
between the four- and eight-cell stages of preimplantation development.
Nature 31, 459–461. Pairleitner H, Steiner H, Hasenoehrl G and Staudach A (1999)
Three-dimensional power Doppler sonography: imaging and quantifying
Campbell S, Bourne TH, Waterstone J, Reynolds KM, Crayford TJ, Jurkovic D, blood flow and vascularization. Ultrasound Obstet Gynecol 14, 139– 143.
Okokon EV and Collins WP (1993) Transvaginal color blood flow imaging
of the periovulatory follicle. Fertil Steril 60, 433– 438. Palomba S, Russo T, Falbo A, Orio F, Jr, Manguso F, Nelaj E, Tolino A,
Colao A, Dale B and Zullo F (2006) Clinical use of the perifollicular
Canipari R, Epifano O, Siracusa G and Salustri A (1995) Mouse oocytes inhibit vascularity assessment in IVF cycles: a pilot study. Hum Reprod 21, 1055–61.
plasminogen activator production by ovarian cumulus and granulosa cells.
Dev Biol 167, 371– 378. Phan B, Rakenius A, Pietrowski D, Bettendorf H, Keck C and Herr D (2006)
hCG-dependent regulation of angiogenic factors in human granulosa lutein
Chui DK, Pugh ND, Walker SM, Gregory L and Shaw RW (1997) Follicular cells. Mol Reprod Dev 73, 878–884.
vascularity– the predictive value of transvaginal power Doppler
ultrasonography in an in-vitro fertilization programme: a preliminary Picton H, Briggs D and Gosden R (1998) The molecular basis of oocyte growth
study. Hum Reprod 12, 191–196. and development. Mol Cell Endocrinol 145, 27–37.

Collins W, Jurkovic D, Bourne T, Kurjak A and Campbell S (1991) Ovarian Redmer DA and Reynolds LP (1996) Angiogenesis in the ovary. Rev Reprod 1,
morphology, endocrine function and intra-follicular blood flow during the 182– 92.
peri-ovulatory period. Hum Reprod 6, 319–324.
Redmer DA, Kirsch JD and Reynolds LP (1991) Production of mitogenic
Coulam CB, Goodman C and Rinehart JS (1999) Colour Doppler indices of factors by cell types of bovine large estrogen-active and estrogen-inactive
follicular blood flow as predictors of pregnancy after in-vitro fertilization follicles. J Anim Sci 69, 237–45.
and embryo transfer. Hum Reprod 14, 1979– 1982.
Salmon NA, Handyside AH and Joyce IM (2004) Oocyte regulation of
Engmann L, Sladkevicius P, Agrawal R, Bekir JS, Campbell S and Tan SL anti-Mullerian hormone expression in granulosa cells during ovarian
(1999) Value of ovarian stromal blood flow velocity measurement after follicle development in mice. Dev Biol 266, 201–208.
pituitary suppression in the prediction of ovarian responsiveness and
outcome of in vitro fertilization treatment. Fertil Steril 71, 22–29. Tekay A, Martikainen H and Jouppila P (1995) Blood flow changes in uterine
and ovarian vasculature, and predictive value of transvaginal pulsed colour
Gaulden ME (1992) Maternal age effect: the enigma of Down syndrome and Doppler ultrasonography in an in-vitro fertilization programme. Hum
other trisomic conditions. Mutat Res 296, 69– 88. Reprod 10, 688– 693.

Gruemmer R, Klein-Hitpass L and Neulen J. (2005) Regulation of gene Thomas JD and Rubin DN (1998) Tissue harmonic imaging: why does it work?
expression in endothelial cells: the role of human follicular fluid. J Mol J Am Soc Echocardiogr 11, 803– 808.
Endocrinol 34, 37–46.
Van Blerkom J, Antczak M and Schrader R (1997) The developmental potential
Hazzard TM, Molskness TA, Chaffin CL and Stouffer RL (1999) Vascular of the human oocyte is related to the dissolved oxygen content of follicular
endothelial growth factor (VEGF) and angiopoietin regulation by fluid: association with vascular endothelial growth factor levels and
gonadotrophin and steroids in macaque granulosa cells during the perifollicular blood flow characteristics. Hum Reprod 12, 1047–1055.
peri-ovulatory interval. Mol Hum Reprod 5, 1115–1121.
Van Royen E, Mangelschots K, De Neubourg D, Valkenburg M, Van
Huey S, Abuhamad A, Barroso G, Hsu MI, Kolm P, Mayer J and Oehninger S de Meerssche M, Ryckaert G, Eestermans W and Gerris J (1999)
(1999) Perifollicular blood flow Doppler indices, but not follicular pO2, Characterization of a top quality embryo, a step towards single-embryo
pCO2, or pH, predict oocyte developmental competence in in vitro transfer. Hum Reprod 14, 2345–2349.
fertilization. Fertil Steril 72, 707– 712.
Wulff C, Dickson SE, Duncan WC and Fraser HM (2001) Angiogenesis in the
Janson PO (1975) Effects of the luteinizing hormone on blood flow in the human corpus luteum: simulated early pregnancy by HCG treatment is
follicular rabbit ovary, as measurated by radioactive microspheres. Acta associated with both angiogenesis and vessel stabilization. Hum Reprod
Endocrinol 79, 122– 133 16, 2515–2524.

Jarvela IY, Sladkevicius P, Tekay AH, Campbell S and Nargund G (2003) Zaidi J, Barber J, Kyei-Mensah A, Bekir J, Campbell S and Tan SL (1996)
Intraobserver and interobserver variability of ovarian volume, gray-scale Relationship of ovarian stromal blood flow at the baseline ultrasound scan
and color flow indices obtained using transvaginal three-dimensional to subsequent follicular response in an in vitro fertilization program.
power Doppler ultrasonography. Ultrasound Obstet Gynecol 21, 277–282. Obstet Gynecol 88, 779– 784.

Joyce IM, Pendola FL, Wigglesworth K and Eppig JJ (1999) Oocyte regulation Zheng J, Redmer DA and Reynolds LP (1993) Vascular development and
of kit ligand expression in mouse ovarian follicles. Dev Biol 214, 342–353. heparin-binding growth factors in the bovine corpus luteum at several
stages of the estrous cycle. Biol Reprod 49, 1177–1189.
Kan A, Ng EH, Yeung WS and Ho PC (2006) Perifollicular vascularity in poor
ovarian responders during IVF. Hum Reprod 21, 1539–1544. Zreik TG, Garcia-Velasco JA, Vergara TM, Arici A, Olive D and Jones EE (2000)
Empty follicle syndrome: evidence for recurrence. Hum Reprod 15, 999–1002.
Koga K, Osuga Y, Tsutsumi O, Momoeda M, Suenaga A, Kugu K, Fujiwara T,
Takai Y, Yano T and Taketani Y (2000) Evidence for the presence of Submitted on July 27, 2006; resubmitted on November 08, 2006; accepted on
angiogenin in human follicular fluid and the up-regulation of its November 16, 2006
production by human chorionic gonadotropin and hypoxia. J Clin
Endocrinol Metab 85, 3352– 3355. Page 5 of 5