To the Editor:
I would like to thank Dr. Hefler-Frischmuth et al. for their fine article (1). In this study, they found that women with polycystic ovary syndrome (PCOS) had significantly elevated serum levels of antihistone and anti-double-stranded DNA (anti-dsDNA) antibodies, whereas serum levels of anti-nuclear antibodies (ANAs) and antinucleosome antibodies were similar between PCOS and healthy controls. When autoimmune markers were correlated with clinical and biochemical parameters, a significant correlation between serum levels of ANAs and serum thyroid stimulating hormone (TSH) were established.
It has been reported that there is a close link between metabolic syndrome characterized by central obesity, insulin resistance, hypertension, and dyslipidemia and immunological diseases (2). Between 30% and 47% of patients with PCOS have the metabolic syndrome and obesity, especially central obesity, insulin resistance and dyslipidemia are also associated with chronic inflammation and rheumatic diseases (2, 3). Autoantibodies may be directed to steroid hormone-producing cells in adrenal and ovary in autoimmune diseases (4).
In this study the authors reported that when serologic markers of autoimmunity were correlated with clinical and biochemical parameters, a significant correlation between serum levels of ANAs and serum TSH was established. Waist-to-hip ratio and lipid levels were important parameters in inflammation and autoimmunity and were not defined in the text. It would be better if the demographic and biochemical parameters of the PCOS and control groups were shown in detail to see the statistical different parameters in each group. Insulin resistance is another important parameter in autoimmunity and considered as abnormal oral glucose tolerance test (OGTT), but abnormal OGTT was not defined. The percentage of women with hirsutism was high, the mean body mass index (BMI) of the study group was 28.1 that can be defined as overweight, and none of women with PCOS had regular menses in the study, however insulin resistance were low and it would be expected to be higher in this group. The definition of PCOS was not demonstrated. There is a great diversity of symptoms, signs, hormonal profiles and diagnostic criteria among women with PCOS that may influence the statistics and results of the study. The data about personal and family history, menstrual irregularities, parity, and desire for a child, smoking and drinking habits, and current medication were sampled, but not detailed in the text. The exclusion criteria of the study should be mentioned.
The results of Pearson’s correlation analysis were not given in detail. I think p and r values of each parameter must be given in the text. Did these markers have close associations between PCOS status? Were all of these correlations made in overall group or in PCOS group? I think all of these must be answered and discussed in the text.
Dr. Fatma Ferda Verit
Department of Obstetrics and Gynecology
Harran University, Faculty of Medicine
1. Hefler-Frischmuth K, Walch K, Huebl W, Baumuehlner K, Tempfer C, Hefler L. Serologic markers of autoimmunity in women with the polycystic ovary syndrome. Fertil Steril 2009; Mar 16 [Epub ahead of print].
2. Escárcega RO, García-Carrasco M, Fuentes-Alexandro S, Jara LJ, Rojas-Rodriguez J, Escobar-Linares LE , et al. Insulin resistance, chronic inflammatory state and the link with systemic lupus erythematosus-related coronary disease. Autoimmun Rev 2006;6(1):48-53.
3. Dessein PH, Joffe BI. Insulin resistance and impaired beta cell function in rheumatoid arthritis. Arthritis Rheum 2006;54(9):2765-75.
4. Forges T, Monnier-Barbarino P, Faure GC, Béné MC. Autoimmunity and antigenic targets in ovarian pathology. Hum Reprod Update 2004;10(2):163-75.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2009.08.009
The Authors Respond:
We thank Dr. Fatma Ferda Verit for her comments. Many of these are interesting. After consulting with all our co-authors and our statistician and having in mind all comments the reviewers have previously made, we decided not to present the demographic and biochemical parameters of the women with polycystic ovary syndrome (PCOS) and controls in detail. We thank Dr. Verit for pointing us to the “abnormal oral glucose tolerance test (OGTT).” An impaired glucose tolerance, i.e. an abnormal OGTT, was ascertained when at least one value was above the cut-off values of 110 and 140 mg/dL for the 0 minutes=fasting level and the 120 minutes value, respectively.
Regarding the definition of PCOS, we point the readers to the inclusion criteria, which are given in full detail within the manuscript. We agree with Dr. Verit that there is a great diversity of symptoms, signs, hormonal profiles and diagnostic criteria among women with PCOS that may influence the statistics and results of the study. After consulting with our statistician, we decided not to present all p and r values within the text in order not to confuse our readers.
We thank Dr. Verit that we have the opportunity to clarify that serum levels of anti-nuclear antibodies (ANAs), anti-histone, anti-nucleosome, and anti-double-strand DNA (anti-dsDNA) antibodies were correlated with clinical (body mass index, age) and biochemical parameters (thyroid-stimulating hormone, serum luteinizing hormone, serum follicle-stimulating hormone, serum testosterone, serum androstendion, serum dehydroepiandrosterone sulfate (DHEA-S), serum creatinine, fasting blood glucose level, presence/absence of an oral glucose tolerance test only in the subgroup of women with PCOS.
Katrin Hefler-Frischmuth, MD
Department of Laboratory Medicine
Lukas Hefler, MD
Department of Obstetrics and Gynecology
Medical University Vienna
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2009.08.010