الفهرس 
IntroductionOnly 14 pages are availabe for public view
Abstract
Polycystic ovary syndrome (PCOS) is an endocrine and metabolic disorder, with a genetic origin (Ewens et al., 2010; Kosova and Urbanek,2012), influenced by developmental and environmental factors (Franks and Berga, 2012). Biochemical hyperandrogenemia (HA) and/or clinical manifestations of hyperandrogenism are a principal feature of PCOS Hoffman et al., 2012). PCOS accounts for 90% to 95% of women attending infertility clinics with anovulation (Brassard et al., 2008). In those with PCOS and infertility, 90% are overweight (Lim et al., 2013).PCOS women are more likely to have upper body fat distribution compared with weight-matched controls. Greater abdominal or visceral adiposity is associated with greater insulin resistance (IR), which could exacerbate the reproductive and metabolic abnormalities in PCOS (Lord et al., 2006b).Clomiphene citrate (CC) has been the gold standard treatment for induction of ovulation in women with polycystic ovary syndrome for many decades owing to its simplicity of use, low cost, relative safety and efficacy(Radosh, 2009). Although 60% to 85% of patients will ovulate on CC, only about one-half will conceive (Neveu et al., 2007). If ovulation cannot be achieved with clomiphene citrate administration at doses of 150 mg/day, then the patient is said to be clomiphene citrate resistant CCR) (Costello and Ledger, 2012a). If such patients could be identified early, time could be saved by changing CCR anovulatory patients to alternative treatments such as gonadotropin therapy or LOD. Many authors have worked extensively on assessing possible predictors of treatment outcomes based on initial screening characteristics (Imani et al., 1998; Imani et al., 2000; El-Halawaty et al., 2007; Rausch et al., 2009).The aim of the present work was to study the relationship between visceral adiposity, ovarian blood flow and serum Anti-Müllerian hormone Summary 169 169 level and to study the effect of these parameters on the response to ovulation induction in infertile women with polycystic ovary syndrome. The study was carried out on 150 patients with PCOS. The initial screening characteristics assessed included demographic and clinical [age,type and duration of infertility, cycle history and hyperandrogenic manifestations (acne, hirsutism and acanthosis nigricans)], laboratory(FSH, LH, AMH, insulin, glucose, HOMA-IR, DHEAS and testosterone),ultrasonic (ovarian volume, ovarian stromal PI and uterine artery PI) and markers of obesity and fat distribution (BMI, WC and visceral fat area atthe level of L4-L5 by computed tomography).Initial CC doses were 50 mg daily for 5 days starting on cycle day3. In the case of an absent response, doses were increased to 100 and 150 mg daily in subsequent cycles. First ovulation with CC was used as the end
point.After a complete follow-up, 110 patients (73.3%) ovulated. Fortypatients (26.7 %) remaining anovulatory, were considered non-responders.Cycle history, BMI, WC, total testosterone, AMH, fasting insulin,HOMA-IR, mean ovarian volume, and VFA were all significantly higher in non-responders than in responders while ovarian stromal PI was significantly lower in non-responders than in responders. Serum AMH showed a strong positive correlation with ovarian volume and a positive correlation with BMI, WC, VFA, testosterone, LH and HOMA-IR and a strong negative correlation with ovarian stromal PI. Ovarian stomal pulsatility index (OPI) showed a negative correlation with testosterone,insulin, HOMA-IR and VFA and a strong negative correlation with WC,AMH and LH. Visceral fat area (VFA) by CT showed strong positive
correlation with BMI, WC, testosterone, DHEAS, insulin, HOMA-IR,ovarian volume and uterine artery PI and a positive correlation with AMH,
while it showed a negative correlation with ovarian stromal PI.
Summary 170
170
As regards prediction of response to CC, testosterone was the best
single predictor with AUC (0.89). Each of the three main variables (AMH,
OPI and VFA) of the study was significant predictor. The cut-off for AMH
to predict CC resistance was ≥ 3.2 ng/ml with AUC 0.79 with sensitivity
70.6 % and specificity 82.5 % while that for ovarian stromal PI was ≤ 0.92
with sensitivity 57.3 % and specificity 72.5 % and for VFA was ≥ 83.6 cm2
with sensitivity 78.2 % and specificity 72.5 %.
Two prediction models have been developed using logistic
regression; the first (Model 1) included themain 3 screening characteristics
of the study (AMH, OPI and VFA) had an AUC of 0.87, the second (Final
model) included all the screening variables that can significantly predict
response to CC, five variables have been retained in the model (BMI,
insulin, testosterone, AMH and OPI) and the AUC improved to 0.98. The
difference between the predictive capacity of the two models was
significant.
So based on our work, PCOS patients who are less likely to respond
to CC can be predicted based on initial characteristics, such as cycle
history, BMI, WC, total testosterone, AMH, fasting insulin, HOMA-IR,
mean ovarian volume, ovarian stromal PI and VFA.