Tải bản đầy đủ - 0 (trang)
2 A Quantitative Method for Diagnosing Folliculo-Luteal Function

2 A Quantitative Method for Diagnosing Folliculo-Luteal Function

Tải bản đầy đủ - 0trang

2.2



15



A Quantitative Method for Diagnosing Folliculo-Luteal Function



prog(ng/ml)



prog(nmol/l)



N = 42



30



100



25



80



20



60



15

40

10

20



5



A



B



0



0

+1 +2 +3 +4 +5 +6 +7 +8 +9 +10 +11 +12 +13 +14 +15



*: days following ovulation



+4 … +9



Fig. 2.2 Serum progesterone level during cycles with spontaneous conception, followed by birth,

from 287 P levels measured during 42 cycles. A: serum progesterone level on the days following

ovulation * (average, SD, 2SD). B: distribution of the average of three measured values between

the 4th and 9th days following ovulation



prog(ng/ml)



N = 38



prog(nmol/l)

100



30



80



25

20



60



15

40

10

A



B



5

0



20



0

* -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 M



-9... -4



* : days preceding menstruation



Fig. 2.3 Serum progesterone level during physiological cycles from 234 measured P levels during 38 cycles. A: serum progesterone level on the days before menstruation * (average, SD, 2SD).

B: distribution of the average of three measured values between the 4th and 9th day before

menstruation



2



16



Diagnosis of Folliculo-Luteal Function



Table 2.1 Quantitative characteristics of the episodic secretion of serum hormones

Coefficient of variation, 95 % CI

Min (%)

LH

16.1

FSH

7.1

Prolactin

7.4

Progesterone

16.2

Oestradiol

8.5

Oestrone

8.0

Testosterone

10.2

Androstenedione

8.9

DHEA

7.2

DHEA sulphate

9.7

Cortisol

21.5



Max (%)

48.2

26.3

34.5

26.4

30.2

17.8

23.1

20.3

25.3

16.2

43.2



Average (%)

30.3

15.1

19.2

20.2

23.0

11.6

15.4

14.3

18.5

12.5

30.6



95 % CI (%)

60.6

30.2

38.4

40.3

46.0

23.2

30.8

28.6

37.0

25.0

61.2



From the analysis of 25 values measured every 10 min between 8 and 12 h in 8 patients in the

middle luteal phase



3.18) and E2 = 354 pg/ml (=1299 pmol/l, SI multiplication factor 3.67) (from 287

measured P and E2 levels during 42 cycles) (Fig. 2.2).

When analysing the characteristics of physiologic cycles without conception, we

found a 6-day P plateau between the 4th and 9th day before menstruation, with an

average value similar to the aforementioned plateau: P = 21.0 ng/ml and E2 = 338 pg/

ml (from 234 measured P levels during 38 cycles). In physiological cycles, the P

values measured during the P plateau were higher than the values usually accepted

as physiological (P ≥ 10 ng/ml).

Ad 2. Measuring serum P levels is unsuitable for characterising luteal function,

especially since Filicori et al. (1984) proved the markedly episodic secretion of P. In

their report they found that P levels measured every 10 min for 24 h during the luteal

phase can exhibit a sevenfold variation in the same patient during the course of the

studied day. However, this fluctuation was significantly less between 8 am and midday, when the highest value was no more than twice the lowest value.

We investigated the quantitative features of the episodic secretion of P and other

hormones mostly from the aspect of clinical practice (Siklósi et al. 1984a, b, c). In

accordance with the currently accepted clinical practice, we used the values measured within a few hours after waking to describe the levels of different hormones.

Thus, we can avoid diagnostic errors resulting from diurnal secretion that has been

confirmed in the case of several hormones (cortisol, prolactin, androgens, etc.).

Although in this way we do not measure the daily average, the values measured in

the morning may be proportional to the daily value. Therefore, we examined the

quantitative relations of the episodic secretion of different hormones – particularly

the ones used in reproductive endocrinology – during the morning period (between

8 h and 12 h) (Table 2.1).

By quantitative analysis of the episodic secretion of P during the middle luteal

phase (between 8 and 12 h), we found (Siklósi et al. 1984a, b, c) that the deviation



2.2



17



A Quantitative Method for Diagnosing Folliculo-Luteal Function



%



_

x = 100 %



%



150



150



100



100

250 min



50



50



%



%



150



150



100



100



250 min



250 min

50



50



250 min



Fig. 2.4 Episodic secretion of serum progesterone during the middle secretory phase in eight

healthy women. Each line connects the values measured in one woman between 8 am and midday, expressed as the percentage of the average of 25 individual values (100 %) (Siklósi et al.

1984a, b, c)



of a single P level from the average serum level is ±40.3 % (95 % CI) (Fig. 2.4).

Day-to-day variation of P was ±49.7 % (95 % CI) during the 6-day plateau period.

The barely 10 % difference between episodic secretion and day-to-day variation

implies that under the 6-day plateau, the daily secretion of P can be considered

relatively stable, except for the episodic secretion, and the levels measured in the

morning do not show extreme variation. Similarly to other researchers (Filicori

et al. 1984; Soules et al. 1988; Fujimoto et al. 1990; ASRM 2008), we also found

that a single P measurement is insufficient to describe luteal function quantitatively. Thus, for the more exact characterisation of luteal function, we used the

average of three P values measured every other day during the 6-day plateau in the

luteal phase, which reduced the deviation from average serum levels that result

from the episodic secretion and day-to-day variation to nearly a third of this amount

(±17.8 %, 95 % CI).

Values measured during the plateau showed at most a twofold difference, even

when examining 42 patients (14 and 28 ng/ml, 95 % CI), and only a 1.5-fold difference when using the average of three measured values (17 and 25 ng/ml, 95 % CI).

In the clinical practice, this means that when using average values, the deviation

from the average is lower than ±8.9 % in 66 % of cases and lower than ±17.8 % in



2



18



%



_

x = 100 %



%

150



150



100



100



250 min



250 min

50



50



%



%



150



150



100



100

250 min



50



Diagnosis of Folliculo-Luteal Function



50



250 min



Fig. 2.5 Episodic secretion of serum oestradiol during the middle luteal phase in eight

healthy women. Each line connects the values measured in one woman between 8 am and midday, expressed as the percentage of the average of 25 individual values (100 %) (Siklósi et al.

1984a, b, c)



95 % of cases. This deviation seemed acceptable even in the clinical practice. The

differences caused by episodic secretion and day-to-day variation of serum oestradiol (E2) were ±46.1 % and ±55.0 %, respectively (Fig. 2.5, Siklósi et al. 1984b).

We also characterised average luteal E2 by three measured values during the luteal

phase.

Besides allowing a more precise quantitative diagnosis, using the average of

three P values obtained every other day during the P plateau greatly decreased the

deviation of physiological values as well (from 21.0 ± 3.5 ng/ml to 21.0 ± 2.0 ng/ml).

Consequently, the physiological minimum (average – 2SD) rose from 14.0 ng/ml to

17 ng/ml (see Figs. 2.2 and 2.3, part B). Based on our studies, folliculo-luteal function can be considered physiological if the average of three values of P measured

between the 4th and 9th day before menstruation or after ovulation is typically

21 ng/ml (21.0 ± 2.0 ng/ml; average ±SD; SI conversion factor, 3.18) and at least

17 ng/ml (average – 2SD) (Siklósi 1991). The average luteal E2 in these cycles was

345 pg/ml.

During the luteal phase, P and E2 are mainly produced by the corpus luteum, and

as we also found a strong correlation between the average luteal P and E2 (r = 0.91,

N = 107, Fig. 2.6), we primarily used the average luteal P to describe luteal function.

We also determined the average luteal E2 for most patients, similarly to the average

luteal P, although initially this was without any therapeutic consequences. We



A Quantitative Method for Diagnosing Folliculo-Luteal Function



oestradiol (pg/ml)



2.2



19



Progesterone and oestradiol

R = 0,9172

700

600



y = 16.858x - 70.708

R² = 0.84



500

400

300

200

100

0

5



15



25



35

45

progesterone (ng/ml)



Fig. 2.6 The relation between average progesterone and oestradiol levels during the conception

cycle (N = 170)



primarily used the luteal E2 level for the indirect investigation of the preovulatory

E2 peak. The preovulatory level of E2 has a defining role in the development of the

physiologic secretory endometrium, as P can only induce the formation of an adequate secretory endometrium after a sufficient E2 effect (this is also the basis for the

gestagen test). Preovulatory levels of E2 show a strong correlation with the luteal P

and E2 levels during CC treatment as well (Radwanska et al. 1980): they are twice

the average luteal E2 level. In clinical practice it is easier to examine P in relation to

the next menstruation (between the 4th and 9th days), as this way we can avoid

measuring the LH peak. At the same time, direct measurement of the E2 peak would

be hindered by the fact that the exact timing of it is unknown during treatment,

another blood sample would be required and because of the marked fluctuation of

E2 (Siklósi et al. 1984a, b, c), a single measurement would give a false approximation of the value.

The strong correlation between average luteal P and E2 does not mean that the

levels of the two hormones are parallel in each actual case, if this were so, the correlation coefficient (r) would be 1.0. This cannot be expected based on the facts

mentioned above: the average of three P values describe the actual level only with a

maximal deviation of ±17.8 % (95 % CI), similarly to E2. Therefore, the measured

E2 value is not always equivalent of the average P, although because of the strong

correlation this cannot occur frequently. In the majority of abortions that occurred

in pregnancies conceived with physiologic P levels, we obtained low E2 levels

(E2 < 220 pg/ml), which are characteristic of abortion. Because of this, we started to

regularly measure E2 levels along with P levels. We adjusted the treatment so that

both parameters entered the physiological range: the P should reach or exceed the

29.2 ng/ml, while for E2, the 420 pg/ml physiological threshold average value, or at



2



20



%



_

x = 100 %



%

150



150



100



250

min



100



50



50



%



%



150



150



100



100



50



Diagnosis of Folliculo-Luteal Function



250

min



50



250

min



250 min



Fig. 2.7 Episodic secretion of serum cortisol. Each line connects the values measured in one

woman between 8 am and midday, expressed as the percentage of the average of 25 individual

values (100 %) (Siklósi et al. 1984a, b, c)



the very least the physiological minimum of 23 ng/ml and 350 pg/ml, and E2 should

not exceed 800 pg/ml to avoid multiple pregnancies (see later). In the first group of

65 patients who were treated, abortion occurred in one case (1.5 %); all other pregnancies lead to mature birth. In light of this, we can expect therapeutic results to

further improve by the combined use of P and E2.

When investigating the quantitative characteristics of the episodic secretion of 11

different hormones, we demonstrated profound fluctuations in many cases, depending on their half-life in the serum (Table 2.1). The most explicit is the fluctuation of

cortisol (Fig. 2.7), and the fluctuation of prolactin is nearly the same as that of progesterone. Taking the diurnal fluctuation of both hormones into consideration, a

single value measured in the morning is even less appropriate for describing their

daily average than the level of P. Therefore, we subsequently followed the practice

of determining hormonal levels – with exception of P and E2 – from a mixture containing equal amounts of three serum samples obtained during the luteal phase.

Thus, we obtain the average value of the three samples without increasing



2.2



A Quantitative Method for Diagnosing Folliculo-Luteal Function



21



laboratory costs. These hormones do not undergo cycle-dependent changes, only a

slight increase for several days around the LH peak. It would be worth considering

this practice in other specific areas of the general clinical practice, instead of the

usually applied single-value method. Based on the above, a prolactin value measured from a single sample can significantly differ from the average that can occasionally lead to misdiagnosis and inappropriate treatment.

A question often emerges about the extent of episodic secretion of hormones and

what proportion of these results can be attributed to methodological errors. The

extent of fluctuation is a sum of two independent factors: the actual fluctuation and

the methodological error. Determining the values of a given patient was performed

using the same radioimmune series, which means that the methodological error

equals the intra-assay coefficient of variation. The relation of two independent factors can be described by the Gauss error formula: S2 = S12 + S22 (the squared resultant standard deviation equals the sum of squared deviations of the independent

factors). For example, in the case of P, the standard deviation of fluctuation was

20.2 %, the intra-assay coefficient of variation of the method was 5.5 % and so the

actual value of fluctuation is only 0.8 % lower than as measured.

However, the propriety of our diagnostic parameters and the causal role of

folliculo-luteal insufficiency (FLI) in UI and HA can only be proved by the success

of therapy. As the corpus luteum originates from the dominant follicle, its characteristics are primarily determined by folliculogenesis, although the adequate degree

of the LH peak during the appropriate maturity stage of granulosa cells and the

adequate luteal LH level also play an important role. Therefore, we used clomiphene citrate (CC) that acts through the stimulation of folliculogenesis, as a treatment for FLI. The individual effect of CC is extremely variable (Dickey and

Holtkamp 1996). For this reason, we increased the CC dosage under regular supervision measurement of average luteal P until the physiological FLF was reached

(Siklósi 1991; Siklósi et al. 2012). The preovulatory level of E2 correlates strongly

with the luteal P and E2 levels (Radwanska et al. 1980); thus physiological luteal P

levels indicate physiological folliculogenesis and preovulatory E2 levels at the same

time. Moreover, an adequate preovulatory E2 peak induces an adequate LH peak in

the optimal karyotic and cytoplasmic maturity stage of the oocyte.

The observation that stimulating ovulatory cycles in UI multiplies the pregnancy

rate implies that the menstrual cycle can be inadequate for conception even if ovulation has been confirmed (Mervil et al. 2010). Our hypothesis was that unrecognized

hormonal insufficiency of the menstrual cycle, which the formerly applied methods

failed to demonstrate, can be the main cause of UI. Even when ovulation occurs,

low preovulatory oestradiol (E2) and then insufficient luteal P and E2 levels can

lead to the formation of a deficient secretory endometrium that is inappropriate for

implantation and placentation (folliculo-luteal insufficiency, FLI). We supposed

there is only mild FLF insufficiency in habitual abortion (HA), implantation takes

place but the pregnancy is aborted because of inadequate placentation.

Our diagnostic method described here lived up to our expectations, and after

the excellent initial results, we have regularly applied it in our clinical practice.

Average luteal P was significantly (p < 0.001) lower in UI than the physiological



22



2



Diagnosis of Folliculo-Luteal Function



range (11.3 ± 3.3 and 21.0 ± 2.0 ng/ml), and normalisation of FLF with controlled

treatment made the fertility of patients physiological (average monthly pregnancy rate for 12 months was 26.6 %, while the yearly cumulative pregnancy rate

was 98.5 %). In HA, the luteal P value was also significantly (p < 0.001) lower

than the physiological range (13.6 ± 3.1 and 21.0 ± 2.0 ng/ml); the controlled normalisation of FLF before conception decreased not only the occurrence of miscarriages but also that of preterm birth and IUGR. These results suggest that the

average luteal P is suitable for the quantitative diagnosis of the complete menstrual cycle (Chaps. 5 and 6).



2.3



Discussion



The majority of authors describe luteal function with a single P value measured

between the 4th and 10th day before menstruation (Balasch and Vanrell 1987; Li

and Cooke 1991), although several authors have demonstrated a markedly episodic

secretion of serum P. Based on our studies, using the average of three P values

decreases the maximum deviation (95 % CI) from the average serum level nearly to

a third of the single value (Siklósi et al. 1984a, b, c). Comparing the individual values obtained during the P plateau period and the distribution of individual averages,

they markedly differ from each other (Figs. 2.2 and 2.3). These results further support our formerly stated opinion that the examination of several P values at the same

time is required to quantitatively describe FLF.

We found a 6-day P plateau in both physiological study groups during the

luteal phase, and most authors likewise found the P level relatively constant

between the 4th and 9th day after ovulation in physiologic luteal phase (Abraham

et al. 1974; Hull et al. 1982). Jones (1949) observed a similar 6-day plateau in the

pregnanediol excretion during physiological cycles. Other researchers report the

occurrence of a P and E2 peaks along with lower hormone values in the luteal

phase (Li and Cooke 1991), which is a characteristic of insufficient luteal phase

in our experience.

Dickey et al. (1992) also found similar, 22.0 ng/ml average P values in the luteal

phase of cycles resulting in birth, and Leach et al. (1997) in physiologic cycles and

UI (24.0 ng/ml and 13.7 ng/ml). In our studies, we measured P average values that

were 16 % higher on average between the 6th and 10th weeks of pregnancies that

resulted in birth, than during the luteal phase. Several authors measured an average

P level of 23.6 ng/ml and 24.4 ng/ml (N = 121 and 54) between the 5th and 7th week

of successful pregnancies where conception happened without intervention (Long

et al. 1994; Cowan et al. 1994). These results correspond to the physiological values

that we obtained during the luteal phase of the cycle with spontaneous conception

and between the 6th and 10th week of pregnancy (luteal phase and the 6–10th week

of pregnancy, 21.0 ng/ml and 24.2 ng/ml).

Almost every author agrees that the sum of P levels measured every day during the luteal phase (the so-called luteal index) or the integral of P values would

be the most appropriate for quantitatively describing luteal function, although



2.3



Discussion



23



PROG (ng/ml)



PROG (nmol/l)



25



80



20



60



15

40

10

20



5



0

OVULA- 1

TION



0

2



3



4



5



6



7



8



9



10 11 12 13



M



the mean value of plateau is

proportionlal to the area under curve



Fig. 2.8 Simplified luteal index



this method is not applicable routinely in the clinical practice because it is complicated and costly (Jones et al. 1974; Balasch and Vanrell 1987; McNeely and

Soules 1988; Cook et al. 1991; Batista et al. 1993). If we consider that the average value of the P plateau is proportional to the sum of daily measured P values,

then the determination of P plateau value by obtaining several (in our case, three

samples measured every other day) serum samples can be considered as a simplified luteal index and is a procedure that patients tolerate well (Fig. 2.8). In

our second study group, the average luteal index in case of 21 ng/ml plateau

value is 189 ng/ml/luteal phase and in case of the physiological minimum of

17 ng/ml, is 153 ng/ml/luteal phase (ninefold the average value of the plateau).

A similar average physiologic luteal index has been reported by other authors as

well (Wu and Minassian 2008; Batista et al. 1993). Batista et al. (1993) demonstrated that the average of three P values obtained on the 4th, 7th and 10th day

of the luteal phase correlates strongly with the value of the luteal index (r = 0.93,

f = 43).

Most authors consider the examination of several cycles with concordant results

as a requirement for diagnosing FLI, while others find it satisfactory to evaluate a

single cycle (Li and Cooke 1991; Olive 1991). Due to the large individual variability

of FLF, the chance of repeated occurrence is different for each patient, but the conception cycle is decisive in the aspect of pregnancy outcome. The maintained normalisation of FLF is thus absolutely justified in infertility and habitual abortion



24



2



Diagnosis of Folliculo-Luteal Function



(Chaps. 5 and 6), while, at the same time, there are no risks or disadvantages of

controlled FLI treatment.

Since the first case report and naming – “pseudocorpus luteum insufficiency” – by Keller et al. (1979), a number of authors have presumed the role of

decreased endometrial sensitivity to P in the development of unexplained infertility (Li and Cooke 1991). This opinion emerges from the fact that in the case

of physiologic P values, FLI has repeatedly been demonstrated by histologic

examination; however, in these studies the authors considered a low, that is,

5–10 ng/ml, P value to be the minimal physiological threshold (Balasch et al.

1982; Spirtos et al. 1985; Li and Cooke 1991). Apart from the diagnostic inaccuracy of endometrial biopsy, this seems to prove that the P values considered

as physiological were too low, rather than the reduced sensitivity of the endometrium. This appears to support the propriety of the higher physiological range

established in our studies.

Numerous authors view the so-called luteinized unruptured follicle (LUF)

syndrome as a separate disorder (Katz 1988). In the case of LUF, the follicle

diameter is significantly smaller than the physiological and is usually associated with low P levels. As this is a rarely repeated condition and it can be successfully treated by applying various stimulation therapies (Katz 1988; Check

et al. 1992), we and some other authors (Petsos et al. 1987) consider this to be

a form of FLI arising from abnormalities in folliculogenesis and not as an individual disorder.

Based on our studies, luteal function can be deemed physiological if the average of three P values obtained every other day (between 8 am and midday

between the 4th and 9th day before menstruation) is typically 21 ng/ml and

minimally 17 ng/ml and the average value of luteal E2 is 345 pg/ml. The propriety of physiological diagnostic parameters that we defined and the causal role of

FLI in reproduction can only be proven if the normalisation of FLF results in

pregnancy and the pregnancy rates and outcomes correspond to the physiological range. In the following chapters, we review our results achieved in different

disorders.

In our studies, we analysed the role of FLI in unexplained infertility and habitual

abortion according to the criteria listed above. In the initial phase of our work, we

did not suspect that FLF has a crucial role in the entire outcome of pregnancy, in the

occurrence of preterm birth and in the development of IUGR. We investigated hormonal characteristics of the physiological cycle through the examination of conception cycles that resulted in birth. However, preterm birth and IUGR occurred in

several of these births. Starting out from the outcomes of pregnancy, the physiological P and E2 values that we defined required retrospective correction. P values

measured during the luteal phase of pregnancies resulting in a singular, mature and

eutrophic newborn can be considered physiological in the aspect of reproduction: in

such cycles, the luteal average of P is 29.2 ± 3.1 ng/ml (physiological minimum

23.0 ng/ml) and the luteal average of E2 is 420 pg/ml (physiological minimum

350 pg/ml), as we will demonstrate later on (Fig. 2.9).



25



References



FSH LH(IU/l)



E2 (pg/ml) PROG. (ng/ml)



50



30



800



PROG



E2



25



40



LH



600

20



FSH

15



400



30



LH



10



20



5



10



200

FSH



0



0

M



2



4



6



8



10 12 14 16 18 20 22 24 26 M



Fig. 2.9 Hormonal characteristics of the physiological menstrual cycle from the aspect of

reproduction



References

Abraham GE. Solid-phase radioimmunoassay of oestradiol-17beta. J Clin Endocrinol Metab.

1969;29:866–70.

Abraham GE, Maroulis GB, Marshall JR. Evaluation of ovulation and corpus luteum function

using measurements of plasma progesterone. Obstet Gynecol. 1974;44:522–5.

Alexander SE, Aksel S, Yeoman RR, Hazelton JM. Gonadotropin and ovarian hormone dynamics

in luteal phase defects. Am J Obstet Gynecol. 1992;166:652–7.

ASRM. Progesterone supplementation during the luteal phase and in early pregnancy in the treatment of infertility: an educational bulletin. Fertil Steril. 2008;2008(90):789–92.

ASRM. Diagnostic evaluation of the infertile female: a committee opinion. Practice Committee of

American Society for Reproductive Medicine. Fertil Steril. 2012a;98:302–7.

ASRM. The clinical relevance of luteal phase deficiency: a committee opinion. The Practice

Committee of the American Society for Reproductive Medicine. Fertil Steril. 2012b;98:1112–7.

Ayabe T, Tetsu T, Tsutsami O, Mitsuhashi N, Momoeda M, Taketani Y. Impaired follicular growth

and abnormal luteinizing hormone surge in luteal phase defect. Fertil Steril. 1994;81:652–6.

Bakos O, Lundkvist O, Bergh T. Transvaginal sonographic evaluation of endometrial growth and

texture in spontaneous ovulatory cycles – a descriptive study. Hum Reprod. 1993;8:99–806.

Balasch J, Vanrell JA, Marquez M, Rivera F, Gonzalez Merlo J. Luteal phase in infertility: problems of evaluation. Int J Fertil 1982;27:60–2.

Balasch J, Vanrell JA. Corpus luteum insufficiency and fertility: a matter of controversy. Hum

Reprod. 1987;2:557–67.

Batista MC, Cartledge TP, Merino MJ, Axiotis C, Platia MP, Merriam GR, Loriaux DL, Nieman

LK. Midluteal phase endometrial biopsy does not accurately predict luteal function. Fertil

Steril. 1993;59:294–300.



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

2 A Quantitative Method for Diagnosing Folliculo-Luteal Function

Tải bản đầy đủ ngay(0 tr)

×