Classification of normogonadotropic infertility: polycystic ovaries diagnosed by ultrasound versus endocrine characteristics of polycystic ovary syndrome

F E R T I L I T Y AND S T E R I L I T Y 'It

Vol. 67, Nn. 3, March 1997

Copyright ' 1997 American Society fi~r Reproductive Medicine

Printed ,hi m.id-/h!e pclpcr in U. S. A.

Classification of normogonadotropic infertility: polycystic ovaries diagnosed by ultrasound versus endocrine characteristics of polycystic ovary syndrome* Evert J. P. van Santbrink, M.D.¢ Wim C. Hop, M.Sc., Ph.D.S Bart C. J. M. Fauser, M.D., Ph.D.t§ D(ikzigt Academic Hospital and Erasmus University Medical School, Rotterdam. The Netherlands

Objective: To investigate the predictive value of polycystic ovaries for endocrine signs of' polycystic ovary syndrome IPCOS). Design: Controlled descriptive study. Setting: Academic tertiary care fertility clinic. Patient(s): Normogonadotropic (FSH levels between 1 and 10 mIU/mL conversion factor to SI unit, 1.01 oligomenorrheic or amenorrheic women visiting our fertility clinic and a control group of regularly cycling, healthy, normal weight volunteers recruited by advertisement. I n t e r v e n t i o n ( s ) : Single blood samples and transvaginal sonography were performed. Main O u t c o m e Measure(s): Serum levels of FSH, LH, androstenedione IA), and T and ovarian volume, ovarian stroma density, and follicle number. Result(s): In control women, the 95th percentile was calculated for ovarian volume, follicle number, and stroma count as well as endocrine parameters. The use of these upper limits of normal in the study group resulted in 217 ~66~ }patients with polycystic ovaries on ultrasound {defined as increased mean ovarian volume andJor mean follicle number per ovary}, whereas only 120 (36c~) patients exhibited elevated serum androgens (increased A and/or T concentrations) and 155 I47~ I showed elevated LH levels. Sensitivity and specificity of single or combined sonographic parameters for prediction of elevated serum LH or androgen concentrations were limited. Conclusion(s): In the study group of normogonadotropic oligomenorrhea or amenorrheic infertile women, we set strict cutofflevels for various criteria used in the literature for defining PCOS. Groups defined by sonographic or endocrine PCOS criteria did overlap, but sonographic parameters had limited predictive value for abnormal hormone serum levels. Fertil SteriP , 1997;67:452-8

Key Words: Polycystic ovaries, PCOS, transvaginal sonography, infertility, testosterone, androstenedione, anovulation, LH

Polycystic ovary s y n d r o m e IPCOS) r e p r e s e n t s a h e t e r o g e n o u s group of patients with a variety of underlying abnormalities. Moreover, the lack of unifor-

Received July 30, 1996; revised and accepted November 22, 1996. * Supported by the "Stichting Voortplantingsgeneeskunde" Rotterdam, The Netherlands. Division of Reproductive Medicine, Department of Obstetrics and Gynecology. $ Department of Epidemiology and Biostatistics. § Reprint requests: Bart C. J. M. Fauser, M.D., Ph.D., Division of Reproductive Medicine, Department of Obstetrics and Gynecology, Dijkzigt Academic Hospital, Dr. Molewaterplein 40, 3015 GD Rotterdam, The Netherlands (FAX: 31-10-43673061. 452

van Santbrink et al. Diagnostics of PCOS

mity in criteria used for P C O S diagnosis f u r t h e r adds to confusion s u r r o u n d i n g this syndrome. Initially, it was r e p o r t e d by Stein a n d L e v e n t h a l (1) t h a t m o r p h o l o g y diagnosis of polycystic ovaries was associated with a specific medical h i s t o r y and characteristic findings upon physical e x a m i n a t i o n (infertility, o l i g o m e n o r r h e a or a m e n o r r h e a , and masculinization I. More recently, diagnosis of P C O S was b a s e d on clinical s y m p t o m s a n d biochemical p a r a m e t e r s . At present, some i n v e s t i g a t o r s apply j u s t a single biochemical m a r k e r such as elevated s e r u m LH conc e n t r a t i o n s (2} or high a n d r o g e n levels (3) to define PCOS, w h e r e a s others use a c o m b i n a t i o n of these two p a r a m e t e r s (4). Fertility and Sterility"

As pelvic ultrasound (US} was introduced in the early 1980s, this new and noninvasive technique allowed morphological assessment of the ovaries on a wide scale. Sonographic appearance of polycystic ovaries became an important criterion for PCOS diagnosis (5-7). In fact, m a n y investigators now use US as the sole criterion for diagnosis of this syndrome (6-8). Sonogn'aphically diagnosed polycystic ovaries can be enlarged (9, 10), contain an increased number of follicles (5, 11/, or exhibit an increased amount or density ofstroma (5, 12). These characteristics are sometimes used as a single parameter or are combined (5, 6, 12). At present, transvaginal sonography with enhanced resolution is most commonly used for US evaluation of ovaries. However, criteria determined by transabdominal US (5) still are used fl'equently t6, 11, 13). Over 10 years after the introduction of pelvic US, still no agreement has been reached on criteria used for polycystic ovary diagnosis nor on its validity. It is unclear to what extent the presence of polycystic ovaries overlaps with other endocrine criteria used to diagnose PCOS such as elevated LH or androgen levels. It appears essential to evaluate various criteria for PCOS diagnosis to create a stronger basis for further uniform evaluation of this entity. MATERIALS AND METHODS S u b j e c t s and S t u d y D e s i g n

This study was approved by the local Ethics Review Committee and informed consent was obtained from all participants. Forty-eight control subjects were selected by poster advertisement and paid for participation. Inclusion criteria were regular menstrual cycle (26 to 30 days), age between 20 and 35 years, normal weight (body mass index bmI, weight divided by square length: 18 to 26 kg/m2), and no hormone therapy for the last 3 months. On cycle day 3, 4, or 5, transvaginal sonography was performed and a blood sample was taken. For sonographic imaging, we used a 6.5 Mhz. transvaginal transducer (model EUB-415; Hitachi Medical Corporation, Tokyo, Japan). The ovaries were localized in relation to the iliac vessels. Follicles appeared as round or ovoid translucent structures. The follicle number was established by scanning each ovary h'om the inner to the outer margin in longitudinal cross-sections (13). The ovarian volume was estimated according to the following formula: ½CAx B x C), where A is the longitudinal diameter, B the anteroposterior diameter, and C the transverse diameter of the ovary (15). Mean follicle number and mean ovarian volume were calculated as the addition of left and right divided by two. Ovarian stroma echogenicity was Vol. 67, No. 3, March 1997

scored as 1 (normal), 2 (moderately increased), or 3 (markedly increased) as described by Pache et al. (14). Total stroma count was the stroma score addition of the left and the right ovary. Three hundred fifty patients attending our fertility clinic between 1991 and 1994 with 1 infertility, 2 oligomenorrhea (interval between periods 35 days) or amenorrhea (absence of vaginal bleeding for ->6 months), and 3 serum FSH concentrations within normal limits (1 to 10 mIU/mL conversion factor to SI unit, 1.0 (16) were included in this study. Transvaginal sonography and blood withdrawal were performed at random. Prolactin concentrations were not used as an exclusion criterion because hyperprolactinemia has been described both in hypogonadotropic as well as in PCOS patients. Timing of sampling may in some cases affect the magnitude of LH serum levels (because of presence of previous exposure to negative feedback action of high P levels or due to midcycle surge LH levels in oligomenorrheic ovulatory patients). All patients with periovulatory E2 levels (E~ 109 pg/mL conversion factor to SI unit, 3.671) or luteal P levels (P 4.7 ng/mL conversion factor to SI unit, 3.180) were excluded in = 8 and n = 12, respectivelyl. This resulted in a total study population of 330 patients. Hormone Estimations

Blood samples were obtained through venipuncture and centrifuged within 2 hours after withdrawal. Serum was stored at -20°C and assayed for FSH and LH by immunoradiometric assay (IRMA) kits provided by Medgenix (Fleurus, Belgium), and A and T by RIA kits provided by Diagnostic Products Corporation (Los Angeles, CA), as described previously (17). Intra-assay and interassay coefficients of variation were 3% and 8 ~ for FSH, 5c~ and 15% for LH, 8% and 11% for A, and 3% and 5% for T, respectively. Data Analysis

Results are presented as the mean +_ SD if normally distributed, or median and range if distributed otherwise. Normal distribution of sonographic parameters was tested with the Kolmogorov-Smirnov goodness of fit test. Normal range was computed according to a p5 to p95 confidence interval according to mean _+ 1.645 x SD if normally distributed (mean ovarian volume, mean follicle number, and LH concentration), or it was antilogged before calculation if parameters were not normally distributed (A and T levels). We compared groups using the MannWhitney and Wilcoxon rank-sum test and the X2 analysis of contingency table. Criteria used for PCOS diagnosis included elevan Santbrink et al. Diagnostics of PCOS

453

1oo

)

)oom II :!JIL__ o_h

~2ot II

25'

r II

50 I - -

50"

40i

_ 40

~30!

-~30~

~26 0

o _|

0 8 "16 24 32 40 t ~ a n Ioa,~Je i'eJmber~o,

Im__

O 10 20 30 40 Mea~ ovana~~ume (mlt

1 2 3 4 5 6 Total suon-.a ¢ounl

Figure 1 Distribution of mean folliclenumber per ovary, mean ovarian volume,and total stroma count(both ovaries) in regularly cyclingcontrols(n = 48; top) and normogonadotropicoligomenorThea or amenorrheicinfertile patients (n = 330; bottom).

vated LH levels, elevated serum androgen levels (increased A and/or T concentrations), and polycystic ovaries on transvaginal sonography (defined as increased mean ovarian volume and/or mean follicle number). The upper limit of normal for sonographic and endocrine parameters was defined based on the 95th percentile of observations in the control population. RESULTS

Control subjects (n = 48) were used to determine the normal range for sonographic and endocrine parameters. Median age was 27 years (range 21 to 35 years}, cycle length was 28 days (range 27 to 30 days}, and BMI was 22 kg/m 2 (18 to 26 kg/m'~). Follicle number per ovary and ovarian volume were distributed normally. Mean number of follicles per ovary was 7.0 _ 1.7 (upper limit: 9 follicles), mean ovarian volume was 6.8 _+ 2.4 mL (upper limit: 10.7 mL), and total stroma count for both ovaries never exceeded 3 (Fig. 1). Mean LH serum concentration was 4.1 _+ 1.8 mIU/mL (upper limit 6.9 mIU/mL conversion factor to SI unit, 1.0), median A level was 2.48 ng/mL (range 1.14 to 5.29 ng/mL, upper limit 4.41 ng/mL conversion factor to SI unit, 3.671), and median T level was 0.58 ng/dL (range 0.29 to 1.12 ng/dL, upper limit 0.89 ng/dL conversion factor to SI unit 3.467). The study population of 330 normogonadotropic oligomenorrheic or amenorrheic women with infertility had a median age of 28 years (range 17 to 41 years), BMI was 25 kg/m '~ (range 18 to 55 kg/m'-'), and median duration of infertility was 2 years (range 1 to 13 years). Prolactin serum levels were elevated in 28 subjects (8%), in this subgroup correlations 454

van Santbrink et al. Diagnostics of PCOS

between US and endocrine p a r a m e t e r s were not different from the overall population (data not shown}. Age, BMI, and duration of infertility were distributed equally in both groups (data not shown). Distribution of all sonographic criteria (increased follicle number, ovarian volume, and stroma count based on "the upper limit of normal defined in control subjects) involved in polycystic ovary diagnosis in the total study group are depicted in Figure 2 (top). Increased ovarian volume (41~ of total group) showed 90% overlap with increased follicle n u m b e r (58% of total group) and/or increased total stroma count (45% of total group). For increased follicle n u m b e r or increased total stroma count, overlap with both remaining US p a r a m e t e r s was 74% and 78%, respectively. In the total study population, 217 (66%) patients showed polycystic ovaries on sonographic examination (arbitrarily defined as mean ovarian volume _> 10.8 mL and/or mean follicle n u m b e r per ovary _> 10). One hundred twenty (36%) patients exhibited elevated serum androgens (A _> 4.44 ng/ mL and/or T - 0.92 ng/dL), and 155 (47%) patients showed increased serum LH levels (LH -> 7.0 mIU/ mL). Sonographic and/or endocrine criteria were not increased in 62 (19%} patients of the study group. Three groups with abnormal endocrine or sonographic param et ers had extensive overlap, as can be seen in a Venn diagram showing the distribution of percentages of patients in the various groups (Fig. 2, bottom). Sixty-eight (21%) patients presented with elevated free androgen index (free androgen index = T × 100/sex-hormone binding globulin ratio 9.1 ). Correlations with US p a r a m e t e r s were not different from T (data not shown). In the total study population, 252 (76%) patients were oligomenorrheic and 78 (24%) patients were amenorrheic. Amenorrheic patients had significant higher serum T levels (P = 0.02), mean follicle number (P = 0.04), and total stroma count (P 0.001) (data not shown}. Luteinizing hormone serum concentrations of oligomenorrheic and amenorrheic patients were not significantly different (data not shown). There was no major difference when Venn diagrams were plotted for oligomenorrheic and amenorrheic patients separately (data not shown). Sensitivity and specificity of various sonographic parameters (mean follicle num ber per whole ovary, mean ovarian volume, and total stroma count) was calculated for elevated T, A, and LH concentrations (Table 1). The best sonographic test for prediction of hyperandrogenicity was increased mean ovarian volume, with a sensitivity of 57% and a specificity of 67%. The addition of increased stroma score to mean ovarian volume and follicle num ber did not increase the predictive value of the tests (data not shown). Receiver operating characteristics (ROC) Fertility and Sterility ~

lncreased

troma (45%)

Increased Follicle

number (58%)

Increased ovarian

Volume (41%)

PCO (66 High Androgens (36%)

High LH (47%)

F i g u r e 2 Distribution of sonographic characteristics (increased m e a n n u m b e r of follicles ->10 per ovary, increased m e a n ovarian volume -> 10.8 m L and increased total s t r o m a count for both ovaries >3) (top), and PCOS diagnostic criteria (elevated a n d r o g e n s A -> 4.44 ng/mL }conversion factor to SI unit, 3.671} and/or T >0.92 ng/dL }conversion factor to SI unit, 3.467}, elevated LH -> 7.0 m I U / m L [conversion factor to SI unit, 1.0}, and polycystic ovaries m e a n ovarian volume ~ 10.8 m L and/or m e a n n u m b e r of follicles per ovary > 10)(bottom) in 330 normogonadotropic oligomenorrhea or amenorrheic infertile patients. N u m b e r s given are percentages of the overall s t u d y group. The hatched area in both figures r e p r e s e n t an identical patient group, characterized by polycystic ovaries.

curves of polycystic ovaries (increased ovarian volume and/or follicle number) as a test parameter and endocrine PCOS criteria (elevated LH or T concentrations) showed limited predictive value of sonography for abnormal endocrine PCOS criteria (Fig. 3). The best correlations were between ovarian volume a n d A ( r = 0.36;P 0.001) o r T ( r = 0.35;P 0.001) serum concentration, and follicle number and T (r = 0.33; P 0.001) or LH (r = 0.30; P 0.001) level (data not shown). DISCUSSION

A lack of consensus on criteria used for PCOS diagnosis is recognized widely (18). Often accurate inforVol. 67, No. 3, March 1997

mation regarding the used definition of PCOS is insufficiently provided in the literature. Moreover, the population from which PCOS patients are obtained is poorly characterized. This may cause discrepant observations in different groups of p a t i e n t s - - a l l termed P C O S - - regarding prevalence of this syndrome, abnormalities present, and t r e a t m e n t outcome. Correlations between sonographic (ovarian volume, follicle number, and stroma count) and endocrine (LH, A, and T serum levels) characteristics in PCOS patients have been described by various authors. For instance, increased serum LH and androgen levels were demonstrated to be related to ovarian enlargement (6, 13). In a group with bilateral polycystic ovaries, on US it was shown that, in patients with a more polycystic ovary-like sonographic image (more follicles per ovary), biochemical disturbances were more apparent (8). Correlation between serum LH levels, androgen levels, and sonographic parameters also was observed in normogonadotropic anovulatory infertility (10), which is confirmed in the present study (data not shown). The predictive value of sonographic characteristics for endocrine abnormalities frequently associated with PCOS (elevated LH, A, or T levels) is hard to find in the literature. Fox and colleagues (7) studied a group of 65 oligomenorrheic or amenorrheic women and suggested t h a t LH and free androgen index were good predictors for polycystic ovaries, although LH and T levels as single markers had a specificity of both 60% and negative predictive values of 46% and 49%, respectively. We found comparable sensitivity and specificity of sonographic parameters (polycystic ovaries) for the prediction of endocrine abnormalities (see Table 1). In the present study, normal values for sonographic and endocrine parameters used for PCOS diagnosis were defined in a well-characterized control population of volunteers (n = 48). These women had strict regular cycles, normal weight, and were aged between 20 and 35 years. Polycystic ovary syndrome diagnostic criteria (increased follicle number and ovarian volume, elevated serum T and A, and elevated serum LH) were defined as the 95th percentile of the control population. The prevalence ofpolycystic ovaries in this population was 8%. Several other studies (19, 20) showed t h a t polycystic ovaries can be observed in approximately 22% of "normal" randomly selected women. Differences in inclusion criteria for the control population ("women not seeking medical treatment" 18 versus rigid criteria used in the present study) and different criteria to define polycystic ovaries easily can explain these seemingly discrepant observations. We chose increased mean ovarian volume and/or increased mean number of follicles per whole ovary as criteria for polycystic v a n S a n t b r i n k e t al.

Diagnostics of PCOS

455

Table 1 Sensitivity, Specificity, Predictive Value, and Significance of Single and Multiple Sonographic Criteria for Polycystic Ovary Diagnosis Versus Single and Multiple Endocrine Characteristics of PCOS

Test parameter Increased ovarian volume (->10.8 mL) Increased follicle number ( ->10 follicles) Increased stroma score (->4) Increased ovarian volume and/or follicle number Increased ovarian volume and follicle number

Reference test

Sensitivity

Specificity

Positive predictive value

Negative predictive value

Total*

Pt

High androgen levels$ High LH levels§ High LH and androgens

57 50 61

67 67 65

49 58 35

74 59 84

64 59 64

<0.001 0.002 <0.001

High androgen levels High LH levels High LH and androgens /

70 70 76

47 50 46

42 56 30

73 65 87

55 60 53

0.03 <0.001 0.001

High androgen levels High LH levels High LH and androgens

52 54 53

57 60 56

40 56 27

68 58 79

55 57 55

NS[[ 0.015 NS

High androgen levels High LH levels High LH and androgens

77 77 84

41 45 40

43 56 30

76 68 89

54 60 50

0.001 <0.001 <0.001

High androgen levels High LH levels High LH and androgens

49 42 53

73 73 71

50 59 35

72 58 83

65 58 67

<0.001 0.005 <0.001

* Percentage correct identified (positive and negative) of total group. X~ test for the association between test parameter and reference test. High androgen levels = A -> 4.44 ng/mL (conversion factor to SI unit, 3.671) and/or T -> 0.92 ng/dL Iconversion factor to SI unit, 3,467).

§ High LH levels = LH -> 7.0 mIU/mL (conversion factor to SI unit, 1.0). IINS, not significant.

o v a r i e s . T h e first c r i t e r i o n w a s c h o s e n b e c a u s e : t r a n s v a g i n a l s o n o g r a p h y h a s b e e n d e s c r i b e d to be a n a c c u r a t e a n d objective w a y to m e a s u r e o v a r i a n v o l u m e (21) a n d m e a n o v a r i a n v o l u m e e x h i b i t s 90% o v e r l a p w i t h t h e r e m a i n i n g two s o n o g r a p h i c c r i t e r i a .

M e a n follicle n u m b e r a n d t o t a l s t r o m a c o u n t s h o w o n l y 74% a n d 78% o v e r l a p (Fig. 2, top). M e a n follicle n u m b e r w a s c h o s e n as c r i t e r i o n n e x t to m e a n o v a r i a n v o l u m e b e c a u s e it i n c r e a s e s t h e s e n s i t i v i t y as c o m p a r e d w i t h v o l u m e a l o n e ( T a b l e 1). A d d i t i o n a l l y , follicle n u m b e r is t h e m o s t f r e q u e n t l y u s e d a n d i n vestigated sonographic p a r a m e t e r since its introduct i o n as d i a g n o s t i c c r i t e r i o n for p o l y c y s t i c o v a r i e s b y A d a m s et al. (5). T h e v a l u e of s t r o m a d e n s i t y i n PCOS diagnosis has been questioned in recent studies, b e c a u s e it a p p e a r s to be a s u b j e c t i v e p a r a m e t e r (4, 12). T h e v a l u e of s o n o g r a p h i c p a r a m e t e r s a s a s c r e e n i n g t e s t to p r e d i c t e n d o c r i n e a b n o r m a l i t i e s c h a r a c t e r i s t i c for P C O S a p p e a r s to be l i m i t e d i n t h e p r e s e n t s t u d y . B e c a u s e a good s c r e e n i n g t e s t m u s t h a v e h i g h s e n s i t i v i t y (i.e., p e r c e n t a g e t h e t e s t i n d i c a t e s p r e s e n c e of t h e d i s e a s e of t h e t o t a l d i s e a s e gn'oup) r a t h e r t h a n specificity, m e a n follicle n u m b e r as a s i n g l e t e s t a n d m e a n follicle n u m b e r a n d / o r o v a r i a n v o l u m e as a c o m b i n e d t e s t a p p e a r t h e m o s t f a v o r a b l e ( T a b l e 1). A l t h o u g h t h e c o m b i n e d s o n o graphic test has the highest sensitivity, the ROC c u r v e s (Fig. 3) s h o w t h a t t h e a r e a u n d e r t h e c u r v e for t h e p r e d i c t i o n of b o t h e l e v a t e d T or L H l e v e l s did n o t exceed 0.65 a n d t h e r e f o r e m a y be c a t e g o r i z e d a s a t e s t w i t h l i m i t e d d i s c r i m i n a t i v e power.

1.00

0.75

0.50 -

0.25 -

0.00

0.60

0.25

~ 0.50 1 - Specificity

= 0.75

i.O0

Figure 3 Receiver operating charactelqstics curves of mean ovarian volume for (A) elevated LH levels (->7.0 mIU/mL conversion factor to SI unit, 1.0) and (B) elevated T levels (->0.92 ng/dL conversion factor to SI unit, 3.467) in a group of 330 normogonadotropic oligomenorrheic or amenorrheic infertile patients. The diagonal line represents an imaginary test that has no discriminative power. 456

van Santbrink

e t al.

Diagnostics of PCOS

Only normogonadotropic (World H e a l t h Organiza-

Fertility and Sterility ~

tion class II 21) oligomenorrhea or amenorrheic infertile patients were included in this study. We chose this particular classification based on serum FSH levels within normal limits, because PCOS has been described as part of this group. Subdivision of these patients according to criteria used to identify PCOS (Fig 2., bottom) showed much overlap in the three groups of polycystic ovaries, elevated LH levels, and hyperandrogenemia, whereas 19% of subjects did not fit any criterion. In addition, only 20% of patients fulfilled all criteria. Recent studies claim t h a t US has a tendency to overestimate PCOS (23). This is in line with the present results indicating t h a t 21% of patients present with polycystic ovaries without endocrine signs of PCOS. Elevated serum LH concentration and slightly decreased FSH concentration are well-established criteria for PCOS diagnosis. This also can be expressed as the LH:FSH ratio. Although this ratio has been used as an important diagnostic criterion in some studies (2, 7), recent literature shows reduced emphasis on measurements of serum gonadotropin concentrations for PCOS diagnosis (11, 24). Pulsatile release of LH as well as timing of blood withdrawal and assay variability (RIA versus IRMA) makes it questionable to use this parameter as the sole diagnostic criterion (13, 25). In our study, 85% of patients presenting with elevated serum LH levels suffered from polycystic ovaries and/or elevated serum androgens. Hyperandrogenicity is perceived as the most important biochemical parameter for PCOS diagnosis (18). Steroid hormones of adrenal or ovarian origin (T, A, and dehydroepiandrosterone sulphate) frequently are elevated in PCOS patients. Dehydroepiandrosterone sulphate is a m a r k e r for adrenal hyperandrogenism (26), but generally is excluded as a PCOS diagnostic criterion. Most authors agree t h a t T and A are the best parameters for detection of hyperandrogenism (7, 10, 13). In practice, hyperandrogenicity is defined most frequently as elevated serum A and/or T concentrations. Androstenedione concentration and ovarian enlargement have been reported to be correlated in PCOS (6, 9), which is confirmed in the present study (r = 0.36; P 0.001). However, the most sensitive single androgen for discrimination of PCOS is reported to be T (10, 13), which equals present results. Serum androgen level may be the most valuable marker for PCOS, because this is correlated best to both sonographic observations and elevated LH levels. We conclude t h a t - - a f t e r defining in a control population strict cutoff levels for sonographic and endocrine PCOS diagnostic criteria--considerable overlap between these parameters was found in normogonadotropic oligomenorrhea or amenorrheic infertile women. However, the predictive value of Vol. 67, No. 3, March 1997

polycystic ovaries (on US) for endocrine abnormalities was limited. This cross-sectional study should be followed by a longitudinal follow-up study to investigate which of these initial screening parameters are best predictors for success or complication rates during induction of ovulation. REFERENCES 1. Stein IF, Leventhal ML. Amenorrhea associated with bilateral polycystic ovaries. Am J O b Gyn 1935;29:181-91. 2. Scheele F, Hompes PGA, van der Meer M, Schoute E, Schoemaker J. The effects of a gonadotrophin-releasing hormone agonist on treatment with low-dose follicle-stimulating hormone in polycystic ovary syndrome. Hum Reprod 1993;5: 699-704. 3. Lobe R. Disturbances of androgen secretion and metabolism in polycystic ovary syndrome. Clin Obstet Gynaecol 1985; 12: 605-2O. 4..4adaens Y, Robert Y, Lemaitre L, Fessati P, Dewailly D. Polycystic ovary disease: contribution of vaginal endosonography and reassessment of ultrasonic diagnosis. Fertil Steril 1991;55:1062-8. 5. Adams J, Poison DW, Abdulwahid N, Morris DV, Franks S, Mason HD, et al. Multifollicular ovaries: clinical and endocrine features and response to pulsatile GnRH. Lancet 1985;2:1375-8. 6. Balen All, Conway GS, Kaltsas G, Techatraisak K, Manning P J, West C, et a]. Polycystic ovary syndrome: the spectrum of the disorder in 1741 patients. Hum Reprod 1995;8:210711. 7. Fox R, Corrigan E, Thomas PA, Hull MGR. The diagnosis of polycystic ovaries in women with oligo-amenorroea: predictive power of endocrine tests. Clin Endocrinol (Oxf) 1991;34:127-31. 8. Takahashi K, Eda Y, Okada S, Abu-Musa A, Yoshino K, Kitao M. Morphological assessment of polycystic ovaries using transvaginal ultrasound. Hum Reprod 1993;6:844-9. 9. Puzigaca Z, Prelevic GM, Stretenovic Z, Balint-Peric L. Ovarian enlargement as a possible marker of androgen activity in polycystic ovary syndrome. GynecoI Endocrinol 1991;5:16774. 10. Pache TD, de Jong FH, Hop WC, Fauser BCJM. Association between ovarian changes assessed by transvaginal sonography and clinical and endocrine signs of the polycystic ovary syndrome. Fertil Steril 1993;59:544-9. 11. Obhrai M, Lynch S, Holder G, Jackson R, Tang L, Butt WR. Hormonal studies on women with polycystic ovaries diagnosed by ultrasound. Clin Endocrinol (Oxf) 1990;32:467-74. 12. Dewailly D, Robert Y, Helin I, Ardaens Y, Thomas-Desrousseaux P, Lemaitre L, et al. Ovarian stromal hypertrophy in hyperandrogenic women. Clin Endocrinol (OxD I994;41:55762. 13. Robinson S, Rodin DA, Deacon A, Wheeler MJ, Clayton RN. Which hormone tests for the diagnosis of polycystic ovary disease? Br J Obstet GynaeceI 1992;99:232-8. 14. Pache TD, Hop WC, W]adimiroff JW, Schipper J, Fauser BCJM. Transvaginal sonography and abnormal ovarian appearance in menstrual cycle disturbances. Ultrasound Med Biol 1991; 17:589-93. 15. Sample WF, Lippe BM, Gyepes MT. Grayscale ultrasonog~'aphy of the normal female pelvis. Radiology 1977; 125:477-83. 16. van Santbrink EJP, Hop WC, van Dessel TJHM, de Jong FH, Fauser BCJM. Decremental follicle-stimulating hormone and dominant follicle development during the normal menstrual cycle. Fertil Steril 1995;64:37-43. van Santbrink et

al.

Diagnostics of PCOS

457

17. Fauser BCJM, Pache TD, Lamberts S, de Jong FH, Hop WC, Dahl KD. Serum immunoreactive and bioactive luteinizing hormone and follicle-stimulating hormone levels in women with cycle abnormalities with and without polycystic ovary disease. J Clin Endocrinol Metab 1991;73:811-7. 18. DunaifA, Givens JR, Haseltine F, Meln-iam GR, editors. The polycystic ovary syndrome. Cambridge: Blackwell Scientific, 1992. 19. Poison DW, Wadswol~h J, Adams J, Franks S. Polycystic ovaries--a common finding in normal women. Lancet 1988; 1: 870-2. 20. Farquhar CM, Birdsall M, Manning P, Mitchell JM, France JT. The prevalence of polycystic ovaries on ultrasou~nd scare ning in a population of randomly selected women. Aust NZ J Obstet Gynaecol 1994;34:67-72. 21. el Tabbakh GH, Lotfy I, Azab I, Rahman HA, Southren AL, Aleem FA. Correlation of the ultrasonic appearance of the ovaries in PCOD and the clinical, hormonal and laparoscopic findings. Am J Obstet Gynecol 1986;154:892-5.

458

v a n S a n t b r i n k et al.

Diagnostics of PCOS

22. World Health Organization. WHO manual for the standardized investigation and diagnosis of the infertile couple. Cambridge: Cambridge University Press, 1993. 23. Abdel Gadir A, Khatim MS, Mowafi RS, Alnaser HMI, Muharib NS, Shaw RW. Implications of ultrasonically diagnosed polycystic ovaries. I. Correlations with basal hormonal profiles. Hum Reprod 1992;4:453-7. • 24. Lobo RA, Kletzky OA, Campeau JD, diZerega GS. Elevated bioactive luteinizing hormone in women with the polycystic ovary syndrome. Fertil Steril 1983;39:674-8. 25. Fauser BCJM, Pache TD, Hop WC, de Jong FH, Dahl KD. The significance of a single serum luteinizing hormone measurement in women with cycle disturbances: discrepancies between immunoreactive and bioactive hormone estimates. Clin Endocrinol (Oxf} 1992;37:445-52. 26. Rittmaster RS. Androgen conjugates as a measure of hyperandrogenism. Semin Reprod Endocrinol 1994; 12:45-50.

Note. Additional references are available upon the author by request.

Fertility and Sterility ®