2 SUBJECTS AND METHODS

2.1  Patients 

↓16

Forty six infertile women with PCOS, aged between 22 to 39 years, were recruited in the department of reproductive medicine at the University Clinic Charité, Berlin, Germany.

Criteria of PCOS were as follows:

↓17

  1. Oligomenorrhea (less than four cycles in the last 6 months) or amenorrhoea (no menses in the last 4 months),
  2. Hyperandrogenemia defined by the free androgen index (FAI) greater then 5.6
  3. Presence of polycistic ovaries according to the ultrasound criteria of Adams and co-workers,1986.
  4. Hirsutism (Ferriman-Gallwey Score > 6) (Ferriman & Gallwey, 1961)
  5. LH/FSH ratio >2

Women with two or more criteria were considered as having PCOS.

All women had normal prolactin levels. The women were in good general health, euthyroid, and none had been taking hormonal medication or oral contraceptives during the three months before the initial investigation, nor during the study. None of the subjects had renal or liver dysfunction. The latter two criteria were determined to prevent risk of lactic acidosis due to metformin use.

↓18

Impaired glucose tolerance was not the obligate criteria for PCOS, therefore oGTT was not included in the screening programme.

Other exclusionary criteria comprised significant cardiovascular disease, bulimia nervosa, active cancer, intake of drugs affecting a body weight like diuretics, participation in dietary programs for weight reduction, and participation in another investigational study within the past 30 days.

Table 1 reflects the main characteristics of the patients.

↓19

Table 1: Clinical characteristics of PCOS women recruited in the trial

Age

BMI

WHR

Hirsutism score

Acne

Menstrual cycle

PCO in US

Involuntary Childlessness (years)

 

1

24

46.4

0.86

3

yes

Regular

yes

2

 

2

23

33.5

0,72

10

yes

Oligomenorrhea

yes

2

 

3

24

38.6

0.69

2

yes

Oligomenorrhea

yes

1

 

4

32

38.1

0.88

14

yes

Oligomenorrhea

yes

5

 

5

36

31.3

0.82

8

yes

Oligomenorrhea

yes

1

 

6

32

42.4

0.77

11

yes

Regular

yes

2

 

7

30

40.0

0.8

9

yes

Regular

no

2

 

8

39

39.2

0.8

12

yes

Regular

yes

7

 

9

35

45.7

0.78

10

yes

Oligomenorrhea

yes

3

 

10

29

42.0

0.87

9

yes

Oligomenorrhea

yes

7

 

11

31

45.0

0.83

12

yes

Regular

yes

7

 

12

36

46.0

0.82

11

no

Regular

yes

5

 

13

30

43.6

0.9

9

yes

Oligomenorrhea

yes

3

 

14

32

39.5

0.77

4

no

Oligomenorrhea

yes

3

 

15

32

49.9

0.78

2

no

Oligomenorrhea

yes

10

 

16

32

37.0

0.98

1

no

Oligomenorrhea

yes

4

 

17

33

44.0

0.95

8

yes

Oligomenorrhea

yes

4

 

18

30

38.8

0.78

4

yes

Amenorrhoea

yes

6

 

19

28

48.0

0.85

11

no

Oligomenorrhea

yes

7

 

Age

BMI

WHR

Hirsutism score

Acne

Menstrual cycle

PCO in US

Involuntary Childlessness (years)

 

20

23

37.0

0.85

10

no

Oligomenorrhea

yes

2

 

21

37

39.3

0.89

2

no

Oligomenorrhea

yes

15

 

22

29

41.3

0.8

12

yes

Regular

yes

4

 

23

24

46.2

0.88

23

yes

Amenorrhoea

yes

4

 

24

36

36.6

0.84

12

yes

Oligomenorrhea

yes

3

 

25

28

36.9

0.78

13

yes

Amenorrhoea

yes

3

 

26

27

33.5

0.84

10

yes

Amenorrhoea

yes

3

 

27

31

32.7

0.89

8

yes

Regular

yes

4

 

28

31

39.0

0.9

6

yes

Oligomenorrhea

yes

6

 

29

28

41.0

0.84

9

yes

Oligomenorrhea

yes

6

 

30

29

41.0

0.98

1

no

Oligomenorrhea

yes

6

 

31

21

44.1

0.8

3

no

Oligomenorrhea

yes

2

 

32

32

35.9

0.8

1

no

Regular

yes

3

 

33

24

42.7

0.8

12

no

Amenorrhoea

yes

4

 

34

28

42.1

0.9

10

yes

Oligomenorrhea

yes

4

 

35

36

30.2

0.82

10

yes

Regular

yes

7

 

36

28

36.3

0.89

8

yes

Oligomenorrhea

yes

1

 

37

30

37.6

0.81

3

no

Oligomenorrhea

yes

1

 

38

31

43.0

0.92

16

no

Amenorrhoea

yes

10

 

39

32

44.9

0.82

9

yes

Amenorrhoea

yes

5

 

Age

BMI

WHR

Hirsutism score

Acne

Menstrual cycle

PCO in US

Involuntary Childlessness (years)

 

40

27

31.0

0.88

2

yes

Oligomenorrhea

yes

1

 

41

28

48.6

0.84

2

no

Regular

yes

5

 

42

28

32.6

0.81

2

no

Oligomenorrhea

yes

1

 

43

29

37.7

0.89

3

no

Amenorrhoea

yes

1

 

44

37

36.8

0.86

9

no

Oligomenorrhea

yes

1

 

45

26

37.4

0.9

11

yes

Amenorrhoea

yes

5

 

46

33

45.0

0.8

9

no

Amenorrhoea

yes

10

 

The mean age of patients was 29.9 ± 4.2 years. All subjects were overweight (mean BMI 38.1; range 28.1 – 49.0 kg/m²) and the large majority of women (87.5%) had abdominal fat distribution defined by waist-to-hip ratio (WHR) greater than 0.8.

↓20

11 (23.9%) patients revealed regular menstrual cyclicity, 25 (54.3%) patients had less than four cycles in the last 6 months and 10 (21.7%) women had chronic amenorrhoea.

Polycistic ovaries were demonstrated by transvaginal ultrasound in all volunteers although its presence is not required for diagnosis of the polycystic ovary syndrome (Franks, 1995).

The study protocol was approved by the ethic commission of the University Clinic Charité, Berlin, Germany and written informed consent was obtained from each woman before study.

↓21

Metformin was helpfully granted by Berlin-Chemie, Germany.

2.2 Study Design

Therapy with metformin was initiated in a randomised double-blind trial of 1500 mg daily. Each woman took one 500 mg metformin tablet or an identical placebo tablet orally each morning, afternoon and evening for sixteen weeks. Patients were advised that they may have minor gastrointestinal side effects. These could be diarrhea, abdominal discomfort, anorexia, nausea, and rarely, a metallic taste in the mouth (Dandona et al., 1983).

The trial included five monthly visits. Participants came to the Division of Reproductive Medicine between 7.00 and 10.00 a.m. after a 12-hour overnight fast. During each visit ( at week 0, 4, 8, 12 and 16) all the women underwent the following investigations:

↓22

Group therapy with aspects in nutrition and physical activity was conducted monthly. Each woman received individual counselling by a dietician.

At the first, third and fifth visits (at week 0, 8 ,16) blood for endocrine and metabolic measurements was obtained. Venous blood for the endocrine investigations was obtained before the start of the metabolic studies.A 75-g oral glucose was administered, and venous blood was obtained for glucose and insulin determination, basally, at 60 and 120 min through the catheter. The response of glucose and insulin to the oGTT was analysed by calculating the AUC (area under the curve).

↓23

Pill bottles with 90 metformin or placebo tablets for the next month were given at each visit.

As a monitor of general drug safety, a complete blood count with differential hepatic and renal characteristics was performed at baseline and after trial.

The study was conducted from September 2002 to July 2003 in the Department for Reproductive Endocrinology, University Hospital Charité, Berlin.

↓24

Limitation of the Study

All blood samples were obtained without regarding the day of menstrual cycle.

2.3 Methods 

Anthropometric Measurements

BMI was calculated as weight (kg) divided by height (m) squared. Waist measurements were made with a soft tape midway between the lowest rib margin and the iliac crest in the standing position. The hip circumstance was measured over the widest part of the gluteal region.

Blood Pressure Measurements

↓25

Blood pressure was measured by a mercury sphygmomanometer in the sitting position, after a rest of at least 10 minutes.

Ultrasound Examination

Ultrasound assessments were conducted before the treatment, and at the first, fourth, eighth, twelfth and sixteenth week. Transvaginal ultrasonography (Logic TM 700,Probe 7.5 MHz, GE Medical Systems, USA) was performed to evaluate the ovaries, number of follicles and endometrium.

The ovaries were defined as “polycistic” when they were enlarged (diameter over 30 mm) with more than 8 small cysts, which are typically arranged peripherally around an increased echogenic stroma (Adams et al, 1986).

Endocrine Investigations

↓26

A determination of FSH, LH, DHEAS, SHBG, testosterone, estradiol, progesterone and metabolic parameters was done in the central laboratory of Charité Hospital.

A determination of insulin, free testosterone, leptin and IGF-I levels has been accomplished in the laboratory of Institute for Experimental Endocrinology, Charité Hospital.

Follicle–stimulating hormone (FSH), luteinizing hormone (LSH) and prolactin were measured by means of immunoassay sandwich method using direct chemiluminescence–technology ( Firm Chiron Diagnostics ACS, USA). Results were expressed in IU/ml and μg/l (prolactin). Intraassay and interassay variations varied from 0.3% to 2.7% and 2.2% to 2.9% for FSH; 1.5% to 2.9% and 2.3% to 3.0% for LH; 2,7% to 3,3% and 1,4% to 4,7% for prolactin.

↓27

Testosterone (T) and dehydroepiandrosterone-sulfate (DHEAS) were determined by the competitive immunoassay method. ADVIA Centauer Testosterone-Assay (USA) was used for testosterone determination. IMMULITE Analyzer (USA) was used for DHEAS determination. Intraassay coefficients of variation were between 2.3% and 6.2%, for T, 6.8% and 9.5% for DHEAS. Interassay variations were between 1.4% and 4.7% for T , 10.8% and 16.6% for DHEAS.

Levels of sex hormone-binding globulin (SHBG) in serum were determined by the Immunometric Assay method (IMMULITE Analyzer, USA). Results were expressed in nmol/L. Intraassay and interassay variations varied from 4.1% to 7.7% and 5.5% to 8.9%, respectively.

Estradiol levels were determined by the competitive immunoassay method using direct chemiluminescence–technology (Chiron Diagnostics ACS, USA). Results were expressed in p/ml. Intraassay and interassay variations varied from 4.5% to 8.1% and from 6.3% to 12.1%, respectively.

↓28

Progesterone was determined in serum by the competitive immunoassay method using direct chemiluminescence–technology (Firm: ADVIA Centauer, USA). Intraassay and interassay variations varied from 1.9% to 5.7% and 3.7% to 12.7%, respectively.

Thyroid hormones were determined by the competitive immunoassay method using direct chemiluminescence–technology (Firm Chiron Diagnostics ACS, USA). Intraassay and interassay coefficients of variation were between 2.41%-2.48% and 2.05%-5.31% for TSH; 1.81%-2.57 and 1.48%-3.48% for fT3.

 IGF-I was determined by the radioimmunassay with double antibodies technique (Adaltis, Italy) with intraassay coefficients of variation 2.9%-5.9% and interassay coefficients from 4.3% to 7.1%, respectively.

↓29

Serum leptin concentrations were determined by the human-Leptin–radioimmunassay (Mediagnost, Tübingen) with intraassay variation less than 5% and interassay variation less than 7.6%.

Free testosterone was determined by the RIA kits by firm Diagnostic Systems Laboratories GmbH (Sinsheim). Intraassay variations were between 3.7% to 6.2%, interassay variations were between 7.3% to 9.7%. The detection limit of the assay came to 0.18 pg/ml.

Metabolic Parameters

Oral Glucose Tolerance Test

↓30

Patients were asked to follow 300 g carbohydrate preparatory diet for 3 days before the test. An oral glucose tolerance test (oGGT) was performed in the morning after a 12-hours fast.

For determination of glucose and insulin levels blood samples were collected basally and after 60 and 120 min. A glycemic response to the OGTT was defined according to the criteria of the American Diabetes Association, 1997.

Table 2: Criteria for the diagnosis of diabetes mellitus

Normal

Impaired glucose tolerance

Diabetes mellitus

Fasting glucose (mg/dl)

< 110

110-125

>125

120 min post 75g glucose load (mg/dl)

<140

140-200

>200

↓31

Levels of glucose were determined by the enzymatic test by Roche Company,

Germany. Intraassay and interassay variations were 0.8% to 1.0% and 1.7% to 1.9%, respectively.

Insulin levels were determined by the radioimmunassay by Adaltis Italia S.p.A (Bologna, Italy). Intraassay variations were between 4.5% and 7.4%, interassay variations were between 4.2% and 8.0%.

↓32

Insulin resistance

The evaluation of insulin resistance in PCOS patients was assessed with Avignon insulin sensitivity index (SiM) (Avignon et al., 1999). Values of sensitivity index less than one characterize insulin resistance. Values of sensitivity index greater than one are supposed to be normal.

Cholesterol, triglycerids, creatinine, alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyltransferase (GGT) were determined by the enzymatic colorimetric test by Firm Roche, Germany. Results were expressed in mg/dl for cholesterol, triglycerids, creatinine and in U/l for ALT, AST and GGT. Intraassay variations laid between 0.7% to 1% for cholesterol, 0.9% to 1.5% for triglycerids, 0.7%-0.9% for creatinine, 0.7%-2.9% for ALT, 1.1%-2.1% for AST and 0.3%-1.5% for GGT.

↓33

 

HbA1c was determined by the VARIANT analysis system using high performance liquid technology. Intraassay variations laid between 0.94% to 1.53%, interassay variations laid between 1.3% to 1.8%.

2.4 Free Androgen Index

The androgen status was assessed by calculating the ratio of the total testosterone (TT) concentration to the concentration (or binding capacity) of SHBG. It is typically calculated on a molar/molar basis and rescaled by a factor ten, or one hundred or one thousand (Wheeler MJ., 1995).

↓34

In this study the free androgen index was calculated as follows: (TT in nmol/L / SHBG in nmol/L) X 100.

2.5 Questionnaires

Patients answered questions relating to personal and family history, gynaecologic history like age of menarche, menstrual activity, contraception, miscarriages, extrauterine pregnancies, operations; weight reduction, smoking and sport activities. They were additionally asked to keep a nutritional protocol for seven days to estimate the nutritional status. Both questionnaires are attached to this paper.

2.6 Dietary Composition

Before treatment, all patients kept a nutrition diary for four days. The nutrition status of each patient was analysed by Nutrition software Diät 2000. Group therapy focussing on nutrition and physical activity started at the first visit and was conducted monthly. After treatment, patients were asked to keep a second nutrition diary and nutrition status was further analysed.

↓35

Lifestyle recommendations for PCOS women included a low fat, high carbohydrates intake for breakfast and high protein intake in the evening. Increased consumption of fibre, wholegrain bread, fruit and vegetables was also recommended. It is suggested that such diets aid in increased weight loss owing to the increased satiating power of protein (Mikkelsen, 2002), and improve insulin sensitivity through maintenance of lean body mass.

2.7 Metformin Treatment

Metformin (Siofor®, Berlin-Chemie, Berlin, Germany) is a white crystalline compound with a molecular formula of C4H11N5▪HCl and a molecular weight of 165.63. Siofor tablets contain 500 mg or 850 mg of metformin hydrochloride. Each tablet contains the inactive ingredients povidone and magnesium stearate. The absolute bioavailability is approximately 50-60%. Following oral administration, 90% of the absorbed drug is eliminated via the renal route within the first 24 hours, with a plasma elimination half-life of approximately 6.2 hours (Physicians’ Desk Reference, 2000).

Metformin inhibits hepatic glucose production, decreases intestinal absorption and promotes glucose uptake and utilization by peripheral tissues at the post-receptor level. It increases the number of insulin receptors but not insulin concentrations and therefore does not cause hypoglycaemia in normoglycemic patients.

↓36

Common side effects include gastrointestinal symptoms such as nausea, vomiting, flatulence and diarrhea. Metformin cannot be used in patients with renal impairment (serum creatinine level > 1.4 mg/dL), congestive heart failure, or liver dysfunction.

The use of metformin increases the concentrations of lactic acid through induction of conversion of glucose to lactate by the intestinal wall. There is no evidence to date that metformin therapy is associated with an increased risk of lactic acidosis or with increased levels of lactate compared with other antihyperglycemic treatments if the drugs are prescribed under study conditions, taking into account contraindications (Salpeter et al., 2003). The doses of metformin used for the treatment of type 2 diabetes mellitus are 1-2 x 500-850 mg/d (Herold et al., 2004).

2.8 Randomization 

Randomization was conducted in a double blind fashion; patients received either placebo or metformin according to the code provided by the pharmacy department of the University Clinic Charité, Berlin, Germany. The randomization code was not broken until the last patient completed all observations.

2.9 Statistical Analysis 

↓37

The integrated glucose (area under the curve glucose [AUCg]) and insulin (AUCi) responses during the OGTT were determined using the trapezoidal rule. Normal distribution of continuous variables were tested by the Kolmogorov-Smirnov test. Both groups were compared by the paired Wilcoxon nonparametric sign rank test. Correlation analysis was performed between variables using Spearman correlation coefficient. All statistical evaluations were performed by the SPSS (Statistical Package for the Social Sciences), Version 11.0. Differences were considered to be significant with P values less than 0.05, and data are reported as the mean ± SD.

Graphics are presented as box plots. The heavy bars represent the median values; the lower and upper limits of the boxes represent the interquartile range (25th and 75th percentiles); the I bars indicate the extreme values.

Figure II: Box plot


© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 4.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated:
10.04.2008