Is insulin resistance an essential component of PCOS?

The influence of confounding factors

D. Cibula

Department of Obstetrics and Gynecology, Charles University, Apolinarska 18, Prague 2, 120 00, Czech Republic e-mail: david.cibula{at}iol.cz


    Abstract
 Top
 Abstract
 Introduction
 Evidence of insulin resistance...
 Non-confirmatory data
 Potential explanations for...
 Conclusions
 References
 
Insulin resistance is often considered a regular component of polycystic ovary syndrome (PCOS). Many interventional studies assume a state of insulin resistance in all patients. However, the evidence is based on small samples that are often insufficient to adjust for significant confounding factors. Moreover, several studies have not confirmed differences in insulin sensitivity between women with PCOS and healthy controls, especially in non-obese patients. This debate article provides an overview of the data published regarding the presence of abnormal or normal insulin sensitivity in PCOS. In conclusion, available data offer evidence that a substantial subgroup of women with PCOS have insulin sensitivity comparable with healthy controls if matched carefully for potential confounding factors.

Key words: insulin/insulin resistance/insulin sensitivity/PCOS


    Introduction
 Top
 Abstract
 Introduction
 Evidence of insulin resistance...
 Non-confirmatory data
 Potential explanations for...
 Conclusions
 References
 
The typical morphology of polycystic ovaries was described >150 years ago (Chereau, 1844Go). Much later, in the 1930s, two professors, Stein and Leventhal, linked polycystic ovaries to the clinical symptoms amenorrhoea, infertility and obesity (Stein and Leventhal, 1935Go). From then on, the syndrome, having the eponym Stein–Leventhal syndrome, did not attract much interest. However, an increased number of relevant articles appeared in the 1980s and 1990s. Among others, the most important reason for the interest of many subspecialties were data documenting the association of polycystic ovary syndrome (PCOS) with hyperinsulinaemia. This new evidence started an ongoing discussion of whether PCOS creates a significant risk for type 2 diabetes and ischaemic heart disease. The question of whether insulin resistance is a regular component of the syndrome has serious consequences both for counselling of patients and for the indication of long-term treatment.


    Evidence of insulin resistance in PCOS
 Top
 Abstract
 Introduction
 Evidence of insulin resistance...
 Non-confirmatory data
 Potential explanations for...
 Conclusions
 References
 
The first to draw attention to the presence of hyperinsulinaemia in women with PCOS was Burghen in 1980 (Burghen et al., 1980Go). He found an elevation of immunoreactive insulin in eight obese women with PCOS compared with six obese controls and a correlation between insulin and androgens. Three years later, Chang et al. (1983Go) showed higher insulin levels during an oral glucose tolerance test (OGTT) in 10 non-obese PCOS patients compared with 10 height- and weight-matched controls. The weight of the patients and controls ranged between 93 and 119% of ideal body weight; the waist to hip ratio (WHR) values were not stated. A higher insulin concentration was described at the same time by others in small groups of patients (Pasquali et al., 1983Go; Shoupe et al., 1983Go).

Later, Dunaif et al. (1989Go) used an euglycaemic clamp, which is considered a gold standard method for evaluation of insulin sensitivity. During clamp, a constant infusion with insulin is administered and euglycaemia is maintained; at steady state, the amount of infused glucose equals the amount taken up by the tissues and is used for measurement of sensitivity to insulin. Dunaif demonstrated a decrease in insulin-mediated glucose disposal rate in women with PCOS. Obesity and PCOS had an additive deleterious effect on insulin sensitivity. Controls in this study were carefully adjusted for age, body mass index (BMI), and body composition determined by hydrostatic weighing; however, the numbers of subjects included in the individual groups were small (11 non-obese and eight obese controls). These results were confirmed by the same author in another study where insulin action was determined in adipocytes and in vivo with a sequential multiple insulin dose euglycaemic clamp (Dunaif et al., 1992Go). PCOS was associated with decreased insulin sensitivity in both obese and non-obese patients.

Abnormal values of insulin sensitivity were confirmed by several other studies, although mostly on small samples. Dunaif and Finegood (1996Go) examined 28 women with PCOS and 29 age- and weight-matched controls by a modified frequently sampled i.v. glucose tolerance test (IVGTT). This technique is believed to produce data with accuracy comparable with the euglycaemic clamp. Moreover, it enables the evaluation of insulin secretion at the same time. The authors found significantly decreased insulin sensitivity in women with PCOS (P ≤ 0.001) and concurrently an abnormal acute insulin response to glucose (AIRg), significantly increased by obesity but not by PCOS. Only after calculation of the disposition index (which expresses the relationship between insulin sensitivity and insulin secretion) was this significantly influenced by obesity and PCOS. In another study, Lasco et al. (1995Go) found that glucose disposal rate values (M index; defined as the amount of glucose supplied to maintain blood glucose levels during the last period of the clamp) were identical in women with PCOS (n = 10) and in obese controls (n = 6). However, this should not be surprising when both groups had comparable BMI values. In the study by Diamanti-Kandarkis et al. (1995Go), the glucose disposal rate values were decreased in obese (n = 10) and non-obese (n = 8) subgroups of women with PCOS in comparison with controls. Although the controls had comparable mean values for weight, BMI and WHR, from the description of the study it follows that they were not weight matched. Morales et al. (1996Go) found a negative trend of decreasing insulin sensitivity from lean controls, to lean PCOS, to obese controls, and finally to obese PCOS. In this study, insulin sensitivity was determined by a rapid IVGTT. However, each group included only eight subjects. Another study compared insulin sensitivity in nine women with PCOS, six obese type 2 diabetics and five controls (Park et al., 2001Go). The insulin sensitivity index (ISI; defined as the ratio of the glucose disposal rate to the insulin concentration at the end of the clamp) was significantly lower in the PCOS group in comparison with the controls. Even in this study, the controls were not matched by anthropometric parameters, although their mean values of BMI and WHR were comparable with those of the patients. In yet another study, focused on the influence of hormonal therapy on insulin action, the controls were also not matched by BMI (Armstrong et al., 2001Go). The authors documented lower insulin sensitivity in 11 patients with PCOS in comparison with 13 controls. While the mean BMI of women with PCOS was not significantly higher, the WHR values were not stated. In 2001, Toprak et al. focused on non-obese patients with PCOS (n = 12), which they compared with age- and weight-matched controls (n = 10) (Toprak et al., 2001Go). In the PCOS group, significantly lower insulin sensitivity was found (P < 0.001).


    Non-confirmatory data
 Top
 Abstract
 Introduction
 Evidence of insulin resistance...
 Non-confirmatory data
 Potential explanations for...
 Conclusions
 References
 
Some relevant papers did not confirm decreased insulin sensitivity in women with PCOS. Moreover, they pointed out several confounding factors which, if not considered, can lead to false-positive results.

Holte et al. (1994Go) published a study in which they examined 41 women with PCOS and 39 controls using the euglycaemic clamp. They found a significant correlation between BMI and ISI, which was greater in the group with PCOS. Differences in ISI between both groups were significant only in subjects with a higher BMI. Moreover, the differences between the two groups disappeared completely after adjustment for truncal–abdominal skin folds. In this study, insulin resistance was largely determined by body fat distribution. Simultaneously, based on the results from the IVGTT, they found an increased early insulin response present over the entire range of BMI. In 1995, the same authors studied to what extent insulin action can be altered by change in weight (Holte et al., 1995Go). During weight reduction (mean weight loss 12.4 kg), they confirmed significant improvement of insulin sensitivity in PCOS patients. After weight reduction, no differences in ISI were found when compared with BMI-matched controls.

Ehrmann et al. (1995Go) examined insulin sensitivity in relation to the presence of a hereditary risk, expressed as a family history of non-insulin-dependent diabetes mellitus (NIDDM). Using IVGTT, they were not able to find any difference in ISI or first phase insulin secretion between women with PCOS and controls. Only after the combination of insulin sensitivity and insulin secretory rate did significant differences become apparent. However, better results were obtained for a group of women with PCOS with a negative family history in comparison with healthy controls with a positive family history of NIDDM. Abnormalities of insulin secretion in PCOS patients were determined largely by hereditary factors.

Some other studies did not confirm decreased insulin sensitivity specifically in lean women with PCOS. Morin-Papunen et al. (2000Go) divided the women with PCOS and controls into lean and obese subgroup. Significant differences in ISI and early phase insulin secretion were found only between obese PCOS and obese controls. The same results were confirmed by our study, where we performed an euglycaemic clamp in 48 lean and 29 obese women with PCOS (peer-reviewed data: unpublished, Vrbikova J. and Cibula D.). In this large sample, we did not find a difference in ISI between lean PCOS women and healthy controls. Comparable results of insulin action between PCOS and healthy controls were also published by Ovesen et al. (1993Go) in a study on a small number of lean women.


    Potential explanations for controversy
 Top
 Abstract
 Introduction
 Evidence of insulin resistance...
 Non-confirmatory data
 Potential explanations for...
 Conclusions
 References
 
From the overview given, it is evident that a few studies did not confirm decreased insulin sensitivity in women with PCOS, especially in lean subgroups. Moreover, there are arguments showing a significant influence of potential confounding factors on insulin sensitivity.

There are several possible explanations for controversy in the literature regarding a prevalence of insulin resistance in PCOS. The more frequent finding of insulin resistance in papers from the USA might be a consequence of different ethnic backgrounds or different lifestyle factors. It was documented by Wijeyaratne et al. (2002Go) that ethnic influence can be significant. When comparing South Asian and Caucasian patients with PCOS in Leeds, they found a lower insulin sensitivity and higher fasting insulin in the former.

Another very important aspect is insufficient consideration of potential confounding factors. It is evident that the results of insulin sensitivity need to be carefully adjusted for many hereditary and acquired parameters. It follows from data obtained in women with PCOS and in healthy women how significant anthropometric parameters are. Even more important than BMI is body fat distribution, which might be expressed as the WHR or simply as waist circumference. It must be strongly advised to use carefully matched control groups for parameters of visceral obesity and BMI if prevalence or severity of insulin resistance is evaluated. It should be mentioned also that obesity cannot be considered as a consequence or even a symptom of PCOS. On the contrary, the presence of obesity substantially influences the phenotype of the disease.

Other important factors which influence insulin sensitivity are hereditary risks. In the study by Ehrmann et al. (1995Go), it was well documented that specifically women with PCOS and a positive family history of type 2 diabetes have a higher risk of abnormal insulin sensitivity and secretion. Consequently, the comparison with a control group should always include data concerning family history (type 2 diabetes) and personal history (gestational diabetes) in both groups.

Also, differences between groups studied can contribute to discrepancies found in the literature. Due to the heterogeneity of the syndrome and weak diagnostic criteria, it is probable that individual authors studied different subgroups of patients. If patients with acne are referred preferentially to one centre, and patients with an irregular cycle or fertility problems are concentrated in other centres, different groups are created, although all subjects fulfil the currently used diagnostic criteria. This last bias might be particularly significant for small sample studies.


    Conclusions
 Top
 Abstract
 Introduction
 Evidence of insulin resistance...
 Non-confirmatory data
 Potential explanations for...
 Conclusions
 References
 
Decreased insulin sensitivity is considered by some authors to be a regular component of PCOS. However, evidence about the presence of insulin resistance is based on small groups of examined patients, which do not sufficiently consider significant confounding factors. Moreover, several studies were published that did not confirm differences in insulin sensitivity when compared with controls, especially in non-obese women.

From the published data, it is evident that although a certain group of women with PCOS have decreased sensitivity to insulin, another sub group of women fulfilling the diagnostic criteria for PCOS have an insulin sensitivity that does not differ from healthy controls. Thus the finding of women with insulin resistance is very significant for clinical practice. Their identification is important for the choice of optimal follow-up and for deciding upon the modality of long-term treatment.


    Acknowledgement
 
This study was supported by grant No. NH/6558-3 of the Internal Grant Agency of the Ministry of Health of the Czech Republic.


    References
 Top
 Abstract
 Introduction
 Evidence of insulin resistance...
 Non-confirmatory data
 Potential explanations for...
 Conclusions
 References
 
Armstrong VL, Wiggam MI, Ennis CN, Sheridan B, Traub AI, Atkinson AB and Bell PM (2001) Insulin action and insulin secretion in polycystic ovary syndrome treated with ethinyl oestradiol/cyproterone acetate. Q J Med 94,31–37.[ISI]

Burghen GA, Givens JR and Kitabchi AE (1980) Correlation of hyperandrogenism with hyperinsulinism in polycystic ovarian disease. J Clin Endocrinol Metab 50,113–116.[Abstract]

Chang RJ, Nakamura RM, Judd HL and Kaplan SA (1983) Insulin resistance in nonobese patients with polycystic ovarian disease. J Clin Endocrinol Metab 57,356–359.[Abstract]

Chereau A (1844) Mémoires pour Servir a l’étude des Maladies des Ovaries. Masson et Cie, Paris, France.

Diamanti-Kandarakis E, Mitrakou A, Hennes MJI, Platanissiotis D, Kaklas N, Spina J, Georgiadou E, Hoffmann RG, Kissebah AH and Raptis S (1995) Insulin sensitivity and antiandrogenic therapy in women with polycystic ovary syndrome. Metabolism 44,525–531.[ISI][Medline]

Dunaif A and Finegood DT (1996) {beta}-Cell dysfunction independent of obesity and glucose intolerance in the polycystic ovary syndrome. J Clin Endocrinol Metab 81,942–947.[Abstract]

Dunaif A, Segal KR, Futterweit W and Dobrjansky A (1989) Profound peripheral insulin resistance, independent of obesity, in polycystic ovary syndrome. Diabetes 38,1165–1174.[Abstract]

Dunaif A, Segal KR, Shelly DR, Green G, Dobrjansky A and Licholai T (1992) Evidence for distinctive and intrinsic defects in insulin action in the polycystic ovary syndrome. Diabetes 41,1257–1266.[Abstract]

Ehrmann DA, Sturis J, Byrne MM, Karrison T and Rosenfield RL (1995) Insulin secretory defects in polycystic ovary syndrome. J Clin Invest 96,520–527.[ISI][Medline]

Holte J, Bergh T, Berne Ch, Berglund L and Lithell H (1994) Enhanced early insulin response to glucose in relation to insulin resistance in women with polycystic ovary syndrome and normal glucose tolerance. J Clin Endocrinol Metab 78,1052–1058.[Abstract]

Holte J, Bergh T, Berne Ch, Wide L and Lithell H (1995) Restored insulin sensitivity but persistantly increased early insulin secretion after weight loss in obese women with polycystic ovary syndrome. J Clin Endocrinol Metab 80,2586–2593.[Abstract]

Lasco A, Cucinotta D, Gigante A, Denuzzo G, Pedullo M, Trifiletti A and Frisina N (1995) No changes of peripheral insulin resistance in polycystic ovary syndrome after long-term reduction of endogenous androgens with leuprolide. Eur J Endocrinol 133,718–722.[ISI][Medline]

Morales AJ, Laughlin GA, Butzow T, Maheshwari H, Baumann G and Yen SSC (1996) Insulin, somatotropic, and luteinizing hormone axes in lean and obese women with polycystic ovary syndrome: common and distinct features. J Clin Endocrinol Metab 81,2854–2864.[Abstract]

Morin-Papunen LC, Vahkonen I, Koivunen RM, Ruokonen A and Tapanainen JS (2000) Insulin sensitivity, insulin secretion, and metabolic and hormonal parameters in healthy women and women with polycystic ovary syndrome. Hum Reprod 15,1266–1274.[Abstract/Free Full Text]

Ovesen P, Moller J, Ingerslev HJ, Jorgensen JO, Mengel A, Schmitz O, Alberti KG and Moller N (1993) Normal basal and insulin-stimulated fuel metabolism in lean women with the polycystic ovary syndrome. J Clin Endocrinol Metab 77,1636–1640.[Abstract]

Park KH, Kim JY, Ahn CW, Song YD, Lim SK and Lee HC (2001) Polycystic ovarian syndrome (PCOS) and insulin resistance. Int J Gynaecol Obstet 74,261–267.[CrossRef][ISI][Medline]

Pasquali R, Casimirri F, Venturoli S, Paradisi R, Mattioli L, Capelli M, Melchionda N and Labo G (1983) Insulin resistance in patients with polycystic ovaries: its relationship to body weight and androgen levels. Acta Endocrinol 104,110–116.[ISI][Medline]

Shoupe D, Kumar DD and Lobo RA (1983) Insulin resistance in polycystic ovary syndrome. Am J Obstet Gynecol 147,588–592.[ISI][Medline]

Stein IF and Leventhal ML (1935) Amenorrhea associated with bilateral polycystic ovaries. Am J Obstet Gynecol 29,181–186.

Toprak S, Yonem A, Cakir B, Guler S, Azal O, Ozata M and Corakci A (2001) Insulin resistance in nonobese patients with polycystic ovary syndrome. Horm Res 55,65–70.[CrossRef]

Wijeyaratne CN, Balen AH, Barth JH and Belchetz PE (2002) Clinical manifestation and insulin resistance (IR) in polycystic ovary syndrome (PCOS) among South Asians and Caucasians: is there a difference? Clin Endocrinol 57,343–350.[CrossRef][ISI][Medline]