Pattern of hormonal and Doppler parameter changes in women with PCOS undergoing ovarian diathermy. * = Signi®cant differences of each variable (comparison of before and 6±10 weeks after operation) ( P < 0.05). 

Context in source publication

Context 1

... total of 79 women was recruited but 27 cases were excluded from the analysis because they had not completed the measurements. Therefore, 52 patients were used for the ®nal analysis. Data analysis showed a fall in the serum concentrations of LH, testosterone and LH/FSH ratio starting on day 2 after LOD. Hormonal pro®le and Doppler parameters, before, 2 days and 6±10 weeks after the operation are shown in Figure 1. LH decreased from 16.86 T 4.53 pre-operatively to 11.7 T 4.82 mIU/ml (6±10 weeks after operation) ( P = 0.001). Mean T SD serum concentrations of testosterone before and 6±10 weeks after the operation were 1.18 T 0.32 and 0.72 T 0.28 ng/ ml respectively ( P = 0.001). When compared with pre- operation levels, serum concentration of FSH increased from 6.24 T 1.85 before operation to 7.55 T 1.98 mIU/ml 6±10 weeks after operation ( P = 0.03). The mean T SD of PI and RI (6±10 weeks after operation) were signi®cantly higher than those in pre-operation values ( P = 0.001), and that of PSV was signi®cantly lower ( P = 0.001). LH/FSH ratio decreased from 2.67 T 0.55 before LOD to 1.59 T 0.65, 6±10 weeks thereafter. Changes in serum hormonal concentrations and Doppler blood ̄ow velocity and 95% con®dence interval of the differences, before and 6±10 weeks after LOD are shown in Table I. We found signi®cant negative correlations between LH and PI ( r = 1434 0.43, P = 0.001), testosterone and PI ( r = 0.40, P = 0.003), testosterone and RI ( r = 0.30, P = 0.043), LH/FSH ratio and PI ( r = 0.53, P < 0.001) and RI ( r = 0.43, P = 0.001). Correlations between hormonal and Doppler parameter changes are shown in Table II. Of all the women, 73.1% ovulated as indicated by mid-luteal serum progesterone levels ( b 5 ng/ml) and leading follicular diameter ( b 15 mm). After adjustment, PI increased signi®cantly in women who ovulated after LOD ( P = 0.001). Although statistically insigni®cant, in the adjusted analyses, an increase in RI was observed in women that ovulated after operation (Figure 2). In post-operation analyses, the changes in Doppler indices in women who did not ovulate were not signi®cant when compared with their pre-operation values (Figure 3). All variables (PSV, PI, RI, LH, FSH, testosterone, age, progesterone, follicular size) had normal distribution. The results of this study demonstrate that ovarian stromal blood ̄ow velocity declined after LOD in women with PCOS. Hormonal alteration that occurred after LOD in our patients was consistent with previous reports (Naether, 1993; Liguri et al ., 1996). The ovulation and pregnancy rate after LOD varied from 20±80% (Greenblatt and Casper, 1993; Tiitinen et al ., 1993; Donesky and Adashi, 1995). Our study had been designed to evaluate the Doppler parameters of ovarian stroma and hormonal pro®le including ovulation before and after LOD. Thus long-term follow-up and pregnancy rate were not considered. Colour Doppler ultrasound permits accurate non- invasive assessment of blood ̄ow to the reproductive organs. It has been used for the assessment of uterine and ovarian blood ̄ow in normal cycles (Sladkevicius et al ., 1993; Zaidi, 2000), PCOS (Ajossa et al ., 2001; Zaidi et al ., 1995; 1998; Zaidi, 2000), and after IVF attempts (Faver et al ., 1993; Balakier and Stronell et al ., 1994). The vascular changes observed during the entire folliculogenesis process seem to play an important role in ovulation (Campbell et al ., 1993; Sladkevicius et al ., 1993; Balakier et al ., 1994; Dolz et al ., 1999). Regarding this point, some reports now exist in the literature arguing that ovulation in humans depends on changes in blood ̄ow to the follicle and that the main blood supply to the Graa®an follicle is directed predominantly to a wreath of blood vessels that lie along the inner border of theca interna (Macchiarelli et al ., 1995). Our study is the ®rst study in the literature to report a signi®cant correlation between some Doppler indices and hormonal changes including ovulation after ovarian diathermy (Table II). In this study, we have demonstrated a dramatic fall in ovarian blood ̄ow in parallel with LH and testosterone level, and LH/FSH ratio 6±10 weeks after surgery (Table I). We were also able to show that PI signi®cantly decreased in women who ovulated after ovarian diathermy. Although statistically not signi®cant, in the adjusted analyses, RI increased ( P > 0.05) when measured 6±10 weeks after operation in subjects that ovulated after the operation but PSV showed no change. The pathophysiology of abnormal ovarian blood ̄ow in PCOS is not clearly understood. One possible explanation is that serum estradiol (E 2 ) might have a role as the moderator of uterine and ovarian vascularity (Steer et al ., 1990; de Ziegler et al ., 1991; Zaidi, 2000). Greenblatt and Casper (1987) showed a fall in E 2 level starting the ®rst day after LOD, reaching the minimum level by day 4 after operation and beginning to rise thereafter. Thus the hypothesis of any correlation between serum E 2 levels and ovarian blood ̄ow changes remains elusive. On the other hand, a signi®cant decrease in vascular impedance to blood ̄ow in the ovarian artery (Deutinger et al ., 1989), and in vessels around the follicles, in correlation with an increase in the number of follicles and serum E 2 concentration (Weiner et al ., 1993), was observed after ovarian stimulation with gonadotrophins. As we demonstrated, ovarian blood ̄ow decreased starting on day 2 following the operation and remained low for at least 2 months. Considering these observations and the data reported by Schurz et al . (1993), it seems that some factors other than E 2 could be the cause of increased ovarian stromal vascularity in PCOS. Dolz et al . (1999) suggested that different mechanisms may be responsible for the haemodynamic anomalies that are uni- formly observed in patients who do not undergo the type of luteal conversion occurring in normally cycling women. They suggested that the abnormal haemodynamic patterns may be due to an abnormal timing of LH-dependent prostaglandin release. Bourne and co-workers (1991) described a direct correlation between LH levels, prostaglandin activity and blood ̄ow changes in the ovary. An alteration in the ®nely tuned timing for release of speci®c prostaglandins is likely to interfere with ovulation in humans. Engmann et al . (1999b) showed that ovarian stromal artery blood ̄ow velocity declines after short term (2±3 weeks) treatment with GnRH agonist and increases signi®cantly on the day of hCG administration. The decline in ovarian artery blood ̄ow velocity after GnRH agonist therapy is unlikely to be due to a hypoestrogenic effect. There is evidence that GnRH agonist therapy has a direct inhibitory effect on granulosa and luteal cell function and may play an important role in processes such as follicular atresia and luteal regression (Sharpe et al ., 1982); therefore the ovaries are quiescent after GnRH agonist therapy. Primordial or smaller preantral follicles do not have any special vascular supply of their own and derive their blood supply from stromal blood vessels (Findaly, 1986). Subsequent growth of primary follicles leads to development of a vascular network with increased 1435 follicular blood ̄ow. Thus the stromal blood ̄ow velocity in an inactive or quiescent ovary may re ̄ect the baseline blood ̄ow perfusion. Laparoscopic ovarian diathermy may result in the reduction in the number of small and intermediate follicles that usually seen in PCOS, it has the same effect on ovarian stromal tissue (Naether, 1993; Liguri et al ., 1996). Regarding these effects and the above-mentioned mechanism by which ovarian stromal blood ̄ow declined after GnRH agonist therapy (Findaly, 1986), we can hypothesize that the decline in ovarian stromal blood ̄ow velocity could be the result of the direct electrical and/or thermal effects of LOD. Considering the increased ovarian stromal blood ̄ow velocity in PCOS (Battaglia et al ., 1995; Zaidi et al ., 1995) and its possible effects on ovarian steroidogenesis, there might be a possible bene®cial effect of diminished ovarian stromal blood ̄ow velocity on ovarian steroidogenesis in PCOS. Our data shed no light on these possibilities since we did not measure E 2 or prostaglandins and no data regarding the direct effect of diminished ovarian stromal blood ̄ow on ovarian steroidogenesis is available. In this study, we reported our preliminary ®ndings regarding the effects of LOD on ovarian stromal blood ̄ow. The results show that Doppler indices of ovarian stromal blood ̄ow signi®cantly changed after LOD and these changes are signi®cantly correl- ated with hormonal changes and subsequent ovulation. Our results provide a potential new avenue for evaluation of ovarian stromal blood ̄ow changes after LOD. These data also suggest that the measurement of ovarian stromal blood ̄ow by colour Doppler may be of value in predicting the prognosis of PCOS related problems after LOD. However, we believe that further research on a larger sample size is needed to determine whether an interaction occurs between LOD, ovarian stromal blood ̄ow and ovarian ...