Tải bản đầy đủ - 0 (trang)
3 The Crucial Role of Folliculo-Luteal Function in Recurrent Miscarriage

3 The Crucial Role of Folliculo-Luteal Function in Recurrent Miscarriage

Tải bản đầy đủ - 0trang

72



5



Recurrent Miscarriage and Folliculo-Luteal Function



who had had three or more spontaneous clinical abortions before the 24th week of

pregnancy – confirmed in their anamnesis – to have recurrent miscarriage (foetal

weight <500 g and/or foetal length <300 mm, Berry et al. 1995; Christiansen et al.

2005). This was in order to ensure that our results are comparable with those of other

researchers (as then the usual criteria of RM was three or more abortions). As the aim

of our study was to investigate the effect of FLF normalisation generally in recurrent

miscarriage, we did not apply any selection criteria (immunological, anatomical, etc.

causes) apart from the number of previous clinical abortions (with at least one intact

tuboovarian complex, verified ovulation and normospermia). We enrolled 120

patients in our randomised, placebo-controlled study (Siklósi et al. 2012).



5.3.2



Power Analysis



In our preliminary studies, we achieved a successful birth rate of above 90 % with

physiological FLF after three or more abortions, while it failed to reach 60 % in

untreated patients. Therefore, we calculated a minimum 30 % difference regarding

birth rates between the treated and the placebo group when determining the patient

population size. To get a significance level of 0.01 with 99 % power (α) during a

two-sampled t-test with a difference of 30 %, 90 patients are needed. We factored in

a 10–15 % drop out rate and 10–15 % infertility rate per year (especially in the placebo group). Therefore, we started our research with 120 patients.



5.3.3



Treatment Protocol



We randomised patients by numbers generated by a computer. We used clomiphene

citrate (CC) (Clostilbegyt, EGIS) for the normalisation of FLF in the treatment

group, together with the regular control of luteal P and E2. We applied CC treatment

between the 5th and 9th day of the cycle, with an initial dose of 5 × 100 mg if the

progesterone level was below 15 ng/ml, and 5 × 50 mg if the P value was over 15 ng/

ml, and increased the dosage by 5 × 50 mg per cycle until we reached the physiological average (21.0 ng/ml), or the physiological minimum (17 ng/ml). We recommended that patients use traditional contraceptive methods until the achievement of

the physiological value. We advised patients to try to get pregnant only after the

physiological value was achieved, along with the continued application of the dosage required for normalisation. P and E2 values were controlled in each cycle until

conception took place. Patients in the placebo group took identical-looking pills in

similar numbers together with regular luteal P and E2 control until conception took

place. Because of the frequent blood sampling, we designed our study for a duration

of 1 year, although in most patients conception occurred much earlier.

The main goal of our study was to investigate the birth rates with physiological

FLF and the prevalence of various obstetrical complications in the two groups and

to analyse the relations of FLF and pregnancy outcomes in all pregnancies. For the

statistical methods and hormone measurements applied, see Chap. 1.



5.3



The Crucial Role of Folliculo-Luteal Function in Recurrent Miscarriage



5.3.4



73



Results



According to the above, we started our study after the randomisation of 120 patients.

Out of these, three patients did not show up for further treatment (two in the treated,

one in the placebo group), three patients gave up the appointed treatment (two in the

treated, one in the placebo group) and four patients did not get pregnant over the

1-year period in the placebo group. Thus, the further analysis concerning pregnancy

outcomes could be performed on 110 patients, 56 in the treated and 54 in the placebo group. The baseline characteristics of patients did not differ significantly

between the two groups (Table 5.2).

The baseline P value (13.6 ± 3.3 ng/ml) of all 110 patients was significantly

(p < 0.001) lower than the physiological (21.0 ± 2.0 ng/ml). It exceeded the lowest

physiological threshold (17 ng/ml) in 18 % of patients. Based on the physiological

values we established, FLI occurred in 82 % in our unselected patient population,

and ≤10 ng/ml P value was measured in 16 %. To achieve physiological luteal P

values, the administration of 5 × 100 mg CC in 46.6 %, 5 × 150 mg CC in 44 % and

5 × 200 mg in 9.6 % was necessary in the patients of the treated group.

Hyperstimulation did not occur in our patient population during the application of

the controlled treatment protocol presented here. The average P level during the

Table 5.2 Baseline characteristics of patients in the treated and the placebo group

Variables compared

Age

Years: average ± SD*

Range

> 35 years: N (%)

Body mass index kg/m2**

Abortions in the anamnesis

Average ± SD

Range

3 abortions – N (%)

4 abortions – N (%)

5 abortions – N (%)

6–7 abortions – N (%)

≥1 late-term abortion

Birth in the anamnesis

Sec. infertility (>2 years)

Baseline progesterone ng/ml

Progesterone <10 ng/ml

Progesterone >17 ng/ml

Baseline oestradiol (pg/ml) in the luteal phase

*



Treated group

n = 56



Placebo group n = 54



31.1 ± 5.6

22–42

14 (25 %)

22.7 ± 3.5



31.3 ± 5.5

21–42

13 (24 %)

22.9 ± 2.6



3.8 ± 1

(3–7)

30 (53 %)

16 (29 %)

6 (11 %)

4 (7 %)

17 (30 %)

15 (36 %)

8 (11 %)

13.1 ± 3.5

8 (14 %)

10 (18 %)

151 ± 47



3.7 ± 0.7

(3–5)

23 % (42.5 %)

23 % (42.5 %)

8 (15 %)



13 (24 %)

21 (39 %)

7 (13 %)

14.2 ± 3.2

9 (17 %)

10 (19 %)

155 ± 49



Plus-minus values represent average ± SD

No significant difference demonstrated between the groups



**



74



5



Recurrent Miscarriage and Folliculo-Luteal Function



conception cycle was 27.3 ± 5.1 ng/ml (average ± SD) in the treated group and

18.6 ± 4.2 ng/ml in the placebo group.

Pregnancy outcomes are presented in Table 5.3. While the birth rate was significantly higher in the physiological group (93 and 56 %, odds ratio (OR):10.4),

whereas the abortion rate was significantly higher in the placebo group (44 and 7 %,

OR: 10.4). In the physiological group, out of the four abortions, three (5.4 %)

occurred in the first trimester and one (1.7 %) in the second trimester. Preterm birth

and newborns with birth weight under 2500 g were significantly more frequent in

the placebo group. The prevalence of IUGR was also higher in the placebo group

(10 and 23 %, OR:2.8), although this difference was not significant (p = 0.0923).

Table 5.3 Pregnancy outcomes in recurrent miscarriage after three or more abortions



Characteristics

Abortion/pregnancy

95 % CI, patients

Abortion I. trimester//

pregnancy

95 % CI, patients

Abortion II. trimester/

pregnancy

95 % CI, patients

Age > 35 years, abortion/

pregn.

95 % CI, patients

Age ≤ 35 years, abortion/

pregn.

95 % CI, patients

Births/pregnancy

95 % CI, patients

Mature births/birth

95 % CI, patients

Preterm birth/birth

95 % CI, patients

IUGR/birth

95 % CI, patients

Birth weight <2500G/birth

95 % CI, patients

Preeclampsia/birth

95 % CI, patients

Weeks of pregnancy, weight

Twin birth/birth

95 % CI, patients

Week of pregnancy, mean

weight

*



Treated group

Average

P > 17 ng/ml

N = 56



Placebo group

N = 54



7.1 %

3–17 %, 4/56

5.4 %

2–15 %, 3/56



44.4 %

32–58 %, 24/54

33.3 %

22–47 %, 18/54



p < 0.001



1.7 %

0.3–9 %, 1/56



11.1 %

5–22 %, 6/54



p < 0.05



6.8

1–59



14.3 %

4–40 %, 2/14



61.5 %

35–82 %, 8/13



p < 0.05



9.6

1–62



4.8 %

1–16 %, 2/42



39.0 %

26–54 %, 16/41



p < 0.001



12.8

3–60



92.9 %

83–97 %, 52/56

90.4 %

79–96 %, 47/52

9.6 %

4–21 %, 5/52

9.6 %

4–21 %, 5/52

13.5 %

7–25 %, 7/52





55.6 %

42–68 %, 30/54

43.3 %

27–61 %, 13/30

56.7 %

39–73 %, 17/30

23.3 %

12–41 %, 7/30

56.7 %

39–73, 17/30

3.3 %

0.6–17 %, 1/30

Week 32, 1170 g

1.9 %

0.3–10 %, 1/52

Week 28, 1000 g



p < 0.001



10.4

3–32

12.3

4–39

12.3

4–39

2.8

1–10

8.4

3–24





3.3 %

0.6–17 %, 1/30

Week 37, 3000 g



Odds ratio ± 95 % CI compared to the placebo group



Significance



p < 0.001



p < 0.001

p < 0.001

p = 0.09

p < 0.001





NS



Odds

ratio*

95 % CI

10.4

3–32

8.8

2–32



1.7

0–28



5.3



75



The Crucial Role of Folliculo-Luteal Function in Recurrent Miscarriage



Investigating the FLF and pregnancy outcomes in all patients (N = 110), luteal P

and E2 values measured in the conception cycle showed a significant (p < 0.001)

difference in the event of miscarriage, preterm birth and mature birth in the consecutive groups (Fig. 5.1). We also found significant (p < 0.001) differences between the

Pa (ng/ml)

p < 0.001



45

40



p < 0.001



p < 0.001



35

30

25

20

15

10

5

0

Miscarriage

N = 28

a



Preterm birth

N = 22



Term birth

N = 60



mean luteal progesterone in cycle of conception



IUGR newborn

N = 12

b



Eutroph nbb

N = 70



preterm and term births together



E2 a (pg/ml)

600



p < 0.001



p < 0.001



550

500

450



p < 0.001



400

350

300

250

200

150

100

50

0

Miscarriage

N = 28

a



Preterm birth

N = 22



Term birth

N = 60



mean luteal oestradiol in cycle of conception



IUGR newborn.

N = 12

b



Eutroph nbb

N = 70



preterm and term births together



Fig. 5.1 The relationship between serum progesterone and oestradiol levels and pregnancy

outcome



76



5



Recurrent Miscarriage and Folliculo-Luteal Function



P and E2 values measured in the luteal phase of pregnancies ending with the birth

of newborns with retardation and of eutrophic newborns.

By further investigating the relationship of FLF and pregnancy outcome, we

found a strong and significant correlation (r = 0.84) between the average luteal P and

the length of pregnancy (weeks) in the P value interval of 12–23.5 ng/ml (Fig. 5.2).

This correlation was even stronger (r = 0.93) when we excluded the cases of four

abortions with physiological values. Conception did not occur under 12 ng/ml P and

all pregnancies with an average P level under 17 ng/ml aborted. All pregnancies

over the value of 23.5 ng/ml P (except for one abortion) ended with mature birth.

The typical P level of preterm birth cases was 20.2 ± 1.7 ng/ml. We detected a strong

correlation (r = 0.89, p < 0.001) between the average luteal P and the weight of newborns, as in every case with a P value over 23.5 ng/ml, the weight of the newborn

exceeded 2500 g (Fig. 5.3). A similar correlation (r = 0.86) was found between the

luteal P value and the weight percentile of newborns: the typical P value in IUGR

was 19.6 ± 1.5 ng/ml. Other neonatal parameters also showed a strong correlation

with luteal P values: length (r = 0.73) and biparietal diameter (r = 0.76).

When comparing the P values and the neonatal parameters of mature births

between the treated and the control group (N = 47 and 13, respectively), we found a

significant (p < 0.001) difference between P values (28.3 and 23.3 ng/ml) and neonatal parameters also: the average weight of newborns was 3429 g and 2843 g,

respectively, and the weight percentile values were 63 and 30, respectively. As the

luteal levels of P and E2 showed strong mutual correlation (r = 0.91, p < 0.001,

Figure 2.6) and the biometrical relations were similar for both hormones, we preferred the relations of P values in our description.



5.3.5



Discussion



In light of our therapeutic results, our data appears to confirm that the actual physiological luteal P levels are significantly higher (21.0 ± 3.5 ng/ml, average ± SD,

physiological minimum value 17 ng/ml) than we had previously assumed (≥10 ng/

ml). Similar P values (18 ng/ml) were also considered as the physiological luteal

minimum by Dickey et al. (2002) in their study, and other authors also found an

average P level over 20 ng/ml in their physiologic control group (Serle et al. 1994;

Kupesic and Kurjak 1997). A likewise significantly elevated P value was demonstrated in several studies in women with recurrent miscarriage who successfully

gave birth compared to those whose pregnancies ended with abortion: in the luteal

phase preceding pregnancy (Serle et al. 1994) as well as in the first trimester of

pregnancy (Kato et al. 2002; Mattukrishna et al. 2002; Arck et al. 2008).

The prevalence of FLI in RM was found to be 15–17 % when most authors estimated it based on a 10 ng/ml physiological minimum P value (Li et al. 2002/b;

Jaslow et al. 2010). We obtained ≤10 ng/ml average baseline P values at a similar

rate (16 %). When using the physiological minimum value (17 ng/ml) that we

defined, 82 % of patient baseline values suggested FLI. The average baseline P

value in the placebo group (14.2 ng/ml) was significantly lower than the values



5.3



The Crucial Role of Folliculo-Luteal Function in Recurrent Miscarriage



77



Duration of pregnancy (week)



Progesterone and duration of pregnancy

R = 0.8459



45



y = 3,2219x – 32,545

R² = 0.7157



40

35

30

25

20

15

10

5



0

5



15



25



35



45



Progesterone (ng/ml)



Fig. 5.2 The relationship between average progesterone and the length of pregnancy (N = 108)



Progesterone and weight of newborn

R = 0.8996

5000

y = 1,3565x – 4,7238

R2 = 0.8093



Weight of newborn (g)



4500

4000

3500

3000

2500

2000

1500

1000

10



15



20

25

30

Progesterone (ng/ml)



35



40



Fig. 5.3 The relationship between average progesterone in the conception cycle and newborn

weight (N = 80)



measured in the conception cycles (18.6 ng/ml). The baseline P levels exceeded the

physiological minimum in 18 % of the cases, while average P was measured in the

conception cycle in 56 % of the patients. This phenomenon might be explained by

the already known great intra-individual variability of FLF (Davis et al. 1989;

Jones 1991), which probably arises from the extreme sensitivity of cycle function.

As physiological FLF begets optimal fertility, in our investigations of the variable

cycles in the placebo group, conception most often took place in the cycle with the

most adequate FLF, with a P value of >12 ng/ml in each case with baseline P lower



78



5



Recurrent Miscarriage and Folliculo-Luteal Function



than 10 ng/ml. The possible positive psychological impact of the increased frequency of patient-doctor encounters also emerges, similarly to the case of “tender

loving care” (Li et al. 2002; Branch et al. 2010); however, the pregnancy outcomes

we obtained in the placebo group were not better than the data available in the literature. Our results correspond well with the observations of other authors. Bromer

et al., (2009) found endometrial insufficiency in the proliferative phase of

RM. Other researchers noted the decreased receptivity of the secretory endometrium in RM (Li et al. 2002; Taklenburg et al. 2010) and showed that this can be

improved with the effect of P through PIBF (progesterone-induced blocking factor) (Raghupathy et al. 2009).

Still, only therapeutic success can confirm that the diagnostic parameters as we

defined them are correct, and FLI has a causal role in RM. As the corpus luteum

is generated from the dominant follicle, it is essentially determined by folliculogenesis (Li et al. 2002), although an LH peak and luteal LH level in proper time

and extent is also essential for the physiological function of the corpus luteum

(Jones 1991; ASRM 2006a, b). Thus we chose CC treatment with controlled doses

to normalise folliculo-luteal function (ASRM 2006a, b). During CC treatment

with an appropriate dosage, the known feed-back mechanisms generate an adequate oestradiol peak in the optimal maturation stage of the granulosa cells and

the oocyte, which in turn induces an adequate LH peak, prompting ovulation in

the optimal karyotic and cytoplasmic maturation stage of the oocyte. Luteal function emerges as a result of these processes, thus physiological luteal function can

verify that the preceding events were also physiological. In empirical CC treatment, where the therapy is applied without control, many authors have described

the negative effect of CC on the cervical mucus and the endometrium (ASRM

2006a, b) and used this to explain why pregnancy occurs in only some cases and

why monthly pregnancy rates remain behind the physiological. It has long been

known that a significant correlation can be demonstrated between preovulatory

oestradiol levels and luteal P levels during CC treatment as well (Radwanska et al.

1980). If the uncontrolled CC dose only slightly increases preovulatory oestradiol

levels, it will fail to compensate the adverse cervical and endometrial effects of

CC. Whereas by using a CC treatment in a controlled dose that causes the normalisation of luteal P values, based on the above-mentioned correlation, we can

assume that the preovulatory oestradiol level is high enough to ensure the development of an adequate cervical mucus and proliferatory endometrium via competitive antagonism, while the secretory transformation is ensured by the

physiological P effect. The effect of CC treatment has a great individual variance,

while it is always dose dependent in a given patient (Dickey and Holtkamp 1996)

as we also observed during FLF normalisation. This further emphasises the necessity of treatment control. Our findings seem to confirm the above described mechanism of effect. The improved pregnancy rates with physiological FLF in the

treated group compared to the placebo group (average of 30 % and 20 %, respectively) suggest that the cervical mucus was adequate, and as for the endometrium,

we observed that pregnancy outcomes are essentially defined by luteal P and E2

levels and are independent of CC treatment (Table 5.3).



5.3



The Crucial Role of Folliculo-Luteal Function in Recurrent Miscarriage



79



In our randomised, placebo-controlled study involving 110 unselected patients

with RM, we confirmed the crucial role of FLF in RM. With physiological FLF,

93 % of pregnancies ended in a birth. This also seems to indicate that the previously

suspected risk factors of RM altogether only play a primary causal role in 7 % of

patients (95 % CI:3–17 %).

By the quantitative analysis of the relations between FLF and pregnancy outcomes, we found a significant difference between the average P and E2 levels of

pregnancies ending in abortion, preterm birth and mature birth in the consecutive

groups and between the levels in the luteal phases of pregnancies resulting in IUGR

and eutrophic newborns. Moreover, a strong and highly significant correlation was

found between the average luteal P and pregnancy outcome (length of pregnancy)

and between the average luteal P and newborn parameters (weight, weight percentile, length, BPD) in the placebo group and in the total pregnancies as well. These

interrelations seem to prove that the circumstances of implantation and placentation, determined by FLF, essentially define the characteristics of the developing

placenta and thus the future outcome of pregnancy and the characteristics of the

newborn.

Although in conceptions that occurred with P levels that were above the physiological minimum, we applied (17 ng/ml) every pregnancy ended in birth both in

the placebo and the treated group except for four abortions. In pregnancies conceived with a P level between 17 and 23 ng/ml preterm birth, IUGR and neonatal

weight under 2500 g still occurred frequently. On the basis of the above described

relations, we needed to revise the P values regarding physiological FLF. We defined

the typical P values of physiological cycles through the examination of conception

cycles that ended in birth. At that time we could not have predicted the above

detailed relations of FLF and pregnancy outcomes, and this was the reason why

cycles in which pregnancy ended in preterm birth or the birth of a newborn with

retardation were also included in the study. According to our recent studies – recognising the connection between FLF and pregnancy outcome – from the aspect of

reproduction, we regard the average P values measured in the luteal phase of pregnancies resulting in the birth of a singular, eutrophic, mature newborn weighing

over 2500 g to be physiological (29.4 ± 2.9 ng/ml [average ± SD]). In our research

material, pregnancies (N = 46) conceived with P values over the 23.5 ng/ml physiological minimum of average P level (average-2SD) – except for one spontaneous

abortion – all ended with the birth of a mature, eutrophic newborn, even in the single

case of twins. Dickey et al. (1995) obtained P values in the luteal phase of successful pregnancies concordant with our findings: they found an average P level of

27 ng/ml in spontaneous cycles and 32 ng/ml in CC-treated cycles.

According to the literature, about 50 % of pregnancy losses in RM (three or more

abortions) are caused by randomly occurring, mainly numerical chromosomal

abnormalities (CA) that emerge during oogenesis or sometimes spermiogenesis (Li

et al. 2002; Christiansen et al. 2005). Pursuant to the investigations of Ogasawara

et al. (2000), the occurrence of abortion increases with the number of previous

aborted pregnancies, and although the ratio of CAs decreases, their prevalence

remains almost invariably 20 % pregnancies, irrespective of the number of abortions



80



5



Recurrent Miscarriage and Folliculo-Luteal Function



in the anamnesis. In the placebo group of our study, 24 pregnancies are aborted, and

based on the above, CAs only could play a primary causal role in half of the abortions, or in 20 % of total pregnancies (i.e. in 12 cases). As every aborted pregnancy

in the placebo group was conceived with FLI, this suggests a possible connection

between FLI and the emergence of random CAs. It is known that the FSH and E2

levels are measured during the proliferatory phase, and also the preovulatory E2 and

LH peak are significantly lower in FLI. Moreover, the main inducing factor of LH

receptors is FSH; thus decreased FSH levels may lead to the diminution of the biological effect of LH on the level of granulosa and endometrial cells through the

decreased number of LH receptors. As appropriate E2 levels play an essential role

in the cytoplasmic maturation process of the oocyte, and the size and biological

effectivity of LH peak have a decisive role in the division processes of the oocyte

(Jones 1991; Li et al. 2002; Mitwaly et al. 2005), it is possible that FLI has a role in

the emergence of numerical CAs. Only 7 % of pregnancies are aborted in the physiological group instead of the expected 20 % with CAs. Even if every lost pregnancy

in this group was caused by CAs, CAs could still only occur in one-third of the

expected cases. Therefore, we can assume that the normalisation of FLF – through

the normalisation of preovulatory FSH, E2 and LH receptor levels and the adequate

LH peak – might play a role in preventing a large proportion of random CAs.

However, this assumption needs further investigation.

In our earlier studies (Siklósi 1991, 1992), pregnancy loss after three or more

previous abortions occurred in six cases. We were able to repeat the treatment in

three out of these, and to our surprise, after the second procedure, pregnancy with

physiological FLF ended with birth in each case. This may suggest that in these

patients, the first abortion with physiological FLF was not primarily caused by some

consequent causal factor (e.g. immunological) but rather an accidental cause. We

can thus draw the conclusion that it is worth repeating the treatment after pregnancy

losses with physiological FLF.

To our knowledge, this is the first randomised, placebo-controlled study that

seems to confirm the crucial role of FLF in RM and in preterm birth and IUGR that

occurs two to four times more often in RM patients. Moreover, our method applied

for the more exact quantitative characterisation of FLF made it possible to recognise

the relationship between pregnancy outcomes and FLF that determines placental

features and between FLF and foetal characteristics by the application of parametric

biometry methods.



5.3.6



Summary



Our results allow the following major conclusions: the typical hormonal values

(P and E2) of physiological FLF appear to be significantly higher than it was

previously thought. Using the average of three P values measured every other

day significantly decreases the deviation of singular P values and increases the

diagnostic reliability of P. From the aspect of reproduction, the average P values measured in the luteal phase of pregnancies resulting in the birth of a



5.4



Successful Treatment of Recurrent Miscarriage



81



singular, eutrophic, mature newborn can be regarded as physiological

(29.3 ± 2.9 ng/ml, average ± SD), and the physiological minimum is 23.5 (average-2SD). The normalisation of FLF appears to be the most effective treatment

of RM. Applying CC treatment in a controlled, individually determined manner

seems to be a simple, reliable method for treating FLI. FLI plays a crucial role

in RM and 93 % of pregnancies with physiological FLF results in birth. In the

development of RM, every primary causal role other than FLI altogether

accounts for a mere 7 % (95 % CI: 3–17 %) of cases. FLF, via determining the

characteristics of the developing placenta, has a crucial role not only in the outcome of pregnancy but in the further development of the foetoplacental unit and

thus in defining foetal parameters, which altogether provides an opportunity to

prevent preterm birth and IUGR – two conditions that occur two to four times

more frequently in RM.



5.4



Successful Treatment of Recurrent Miscarriage

by the Normalisation of Folliculo-Luteal Function



Our randomised, placebo-controlled study seems to confirm our presumption.

Recurrent miscarriages and the two to four times more frequent preterm birth and

IUGR in RM are mainly caused by FLI of varying degree. The strong correlations

between FLF and pregnancy outcome and between FLF and foetal parameters further support the determining role of FLF in RM. The normalisation of FLF provides

the most effective treatment of RM so far. We applied our method regularly thereafter, in cases with two preceding abortions as well. Since the recommendation of the

ASRM in 2008, most authors consider two abortions as the diagnostic criteria for

the disorder, while in Hungary we have used this definition from the beginning

(Zoltán 1975; Papp 1999). In this chapter, we present our results achieved with the

application of this method on a representative patient population and the conclusions that can be drawn based on this. Because of the applied altered definition of

RM, we separately evaluate also our results achieved after two abortions and after

at least three abortions.



5.4.1



Patients and Methods



Studies included in this book were performed on patients of the Endocrine Unit of

the 2nd Department of Obstetrics and Gynaecology, Semmelweis University. We

diagnosed recurrent miscarriage in patients who had two or more verified spontaneous clinical abortions in the anamnesis that occurred before the 24th week of pregnancy (foetal weight <500 g and/or foetal length <300 mm, Berry et al. 1995;

Christiansen et al. 2005). When enrolling patients in the study, we did not apply any

selection criteria (e.g. thrombophilia, immunological, anatomical, etc. causes) other

than the numbers of abortions (in the case of at least one intact tuboovarian unit,

normospermia and confirmed ovulation).



82



5



Recurrent Miscarriage and Folliculo-Luteal Function



Altogether, 510 patients participated in our study, and in the therapy, out of

which 102 did so repeatedly. In the analysis of FLF and pregnancy outcome

(N = 684), we also included the data of a further 72 pregnancies of our patients,

where the average luteal P was determined but did not reach the physiological value.

As control group, we used the data of pregnant women – with the same group

size as the treatment group – who had also had two or more preceding abortions but

did not have any treatment before conception and were admitted to our institute with

their pregnancies.



5.4.2



Treatment Protocol



We used clomiphene citrate (CC) (Clostilbegyt, EGIS) and/or low-dosage corticoid

treatment (LDCT) in the treatment group to normalise FLF, together with regular

luteal P control. We applied CC between the 5th and 9th day of the cycle, with an

initial dose of 5 × 100 mg if the P level was below 15 ng/ml and 5 × 50 mg if the P

value was over 15 ng/ml, and increased the dose by 5 × 50 mg per cycle until the

physiological average was achieved (21.0 ng/ml) but at least until we reached the

physiological minimum (17 ng/ml). We used LDCT continuously administered in

the form of 0.5 mg dexamethasone (DEX) every evening (see Chap. 4). We suggested that the patient use traditional contraceptive methods until the physiological

values were reached. We advised patients to get pregnant only after the physiological value was achieved, along with the continued application of the dosage required

for normalisation.

The main goal of our studies was to investigate pregnancy outcomes during our

treatment method in comparison with the untreated control group, to analyse FLF

and pregnancy outcomes and to draw theoretical conclusions based on our representative patient population.



5.4.3



Results



Table 5.4 summarises the characteristics of the treatment and the control group. We

failed to demonstrate significant differences between the patient characteristics of

the two groups.

The baseline average luteal serum P values in all 510 patients (13.6 ± 3.3 ng/ml,

average ± SD) was significantly (p < 0.001) lower than the physiological

(21.0 ± 2.0 ng/ml) (Table 5.4). The average P value exceeded the 17 ng/ml physiological threshold in 18 % of patients and was under 10 ng/ml (usually considered as

the physiological minimum) in 16 % of patients. Pregnancy took place under the

course of 1–11 cycles in case of physiological FLF (3.1 cycle on average). The average monthly pregnancy rate was 35 %, and the cumulative pregnancy rates over 3,

6, 9 and 11 months were 72 %, 91 %, 98 % and 100 %, respectively. In patients

(N = 85) who did not conceive for over 2 years after their last abortion (over 24–240

cycles, 45 cycles on average), we observed similar monthly and cumulative



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

3 The Crucial Role of Folliculo-Luteal Function in Recurrent Miscarriage

Tải bản đầy đủ ngay(0 tr)

×