Metabolic approaches to the subclassification of polycystic ovary syndrome

Robert J. Norman, Stacey C. Masters, William Hague, Clive Beng, Peter Pannall, Jim X. Wang

Research output: Contribution to journalArticlepeer-review

81 Citations (Scopus)

Abstract

OBJECTIVES:

To examine the relationship between various hormonal and metabolic variables in a large group of women with unequivocal evidence of polycystic ovarian syndrome (PCOS) to dissect out the metabolic heterogeneity of this condition.
DESIGN:

Cross-sectional observational study of PCOS (n = 122) and non-PCOS (n = 26) subjects.
SETTING:

Reproductive medicine unit in a tertiary teaching hospital.
PATIENTS:

Subjects with presumed PCOS were recruited from the Reproductive Medicine and Gynaecological Clinics and later confirmed as PCOS with recognized criteria. Several other subjects were identified through recruiting reference subjects. The PCOS population consisted of 122 patients. Reference subjects were recruited from partners of male factor infertility patients in the clinics and from the general population (n = 27).
INTERVENTIONS:

A 75 g 2-hour oral glucose tolerance test was performed on all subjects in their midluteal phase. Blood was taken at fasting and at 30, 60, 90, and 120 minutes.
MAIN OUTCOME MEASURES:

Age, body mass index (BMI), waist to hip ratio, levels of integrated glucose and insulin, concentrations of maximum insulin, sex hormone-binding globulin, T, triglyceride, apolipoproteins (Apo A1, B), high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol (LDLC).
RESULTS:

Five clusters could be identified. They are characterized as a nonobese group, a moderately obese group, and three very obese groups. The nonobese group (n = 41, BMI = 24.1) exhibited the lowest level of integrated insulin (236.4 mIU/L or microU/mL) and concentration of serum T (5.5 nmol/L). The moderately obese group had the second lowest level of integrated insulin (497.1 mIU/L) whereas the three very obese groups (n = 15, 13, and 5, respectively) had significantly higher but different levels of integrated insulin (group 3: 850.8 mIU/L; group 4: 1,131.5 mIU/L; and group 5: 1,531.9 mIU/L), triglyceride (group 3: 1.39 mmol/L; group 4: 1.76 mmol/L; and group 5: 2.78 mmol/L [1 mmol/L = 88mg/mL]), Apo B (group 3: 1.18 g/L; group 4: 1.08 g/L; and group 5: 1.55 g/L) and LDLC (group 3: 3.81 mmol/L; group 4: 3.05 mmol/L; and group 5: 5.06 mmol/L [1 mmol/L = 38.6 mg/100 mL]).
CONCLUSIONS:

The metabolic heterogeneity of the PCOS population is reflected at least partly in patients' levels of insulin, lipids, and lipoproteins, dependent and independent of BMI.
Original languageEnglish
Pages (from-to)329-335
Number of pages7
JournalFertility and Sterility
Volume63
Issue number2
DOIs
Publication statusPublished - Feb 1995
Externally publishedYes

Keywords

  • PCOS
  • subclassification of PCOS
  • glucose metabolism
  • insulin resistance

Fingerprint

Dive into the research topics of 'Metabolic approaches to the subclassification of polycystic ovary syndrome'. Together they form a unique fingerprint.

Cite this