| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Medicine (T.L.S., L.L.S., K.C.P., A.M.D., A.J.B.), Division of Endocrinology (T.L.S., K.C.P., A.J.B.), and Division of Gastroenterology (A.M.D.), Duke University Medical Center, Durham, North Carolina 27710; and Duke University (N.D.H.), Durham, North Carolina 27710
Address all correspondence and requests for reprints to: Tracy L. Setji, M.D., Department of Medicine, Division of Endocrinology, Duke University Medical Center, Box 3611, Durham, North Carolina 27710. E-mail: setji001{at}mc.duke.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of the study was to determine the prevalence and characteristics of NASH and abnormal aminotransferase activity in women with PCOS.
Design: The study is a retrospective chart review.
Setting: The setting is an academic endocrinology clinic.
Patients: Patients were 200 women with PCOS, defined as irregular menses and hyperandrogenism.
Main Outcome Measures: Biopsy-documented NASH and aminotransferase levels were the main outcome measures.
Results: Fifteen percent (29 of 200) had aspartate aminotransferase and/or alanine aminotransferase more than 60 U/liter. Women with aminotransferase elevations had lower high-density lipoprotein (HDL) (41 vs. 50 mg/dl, P = 0.006), higher triglycerides (174 vs. 129 mg/dl, P = 0.024), and higher fasting insulin (21 vs. 12 µIU/ml, P = 0.036) compared with women with normal aminotransferases. Six women (mean age 29 yr) with persistent aminotransferase elevations underwent liver biopsy. All six had NASH with fibrosis. Compared with the 194 of 200 PCOS women who did not undergo biopsy, women with biopsy-documented NASH had lower HDL (median 34 vs. 50 mg/dl, P < 0.001), and higher triglycerides (245 vs. 132 mg/dl, P = 0.025), fasting insulin (26 vs. 13 µIU/ml, P = 0.038), aspartate aminotransferase (144 vs. 22 U/liter, P < 0.001), and alanine aminotransferase (143 vs. 28 U/liter, P < 0.001).
Conclusion: Abnormal aminotransferase activity is common in women with PCOS. Low HDL, high triglycerides, and high fasting insulin were associated with abnormal aminotransferase activity. Some women already had evidence of NASH with fibrosis. Further studies are needed to evaluate whether to screen PCOS women for liver disease at an earlier age than is currently recommended for the general population.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
NAFLD is characterized by the accumulation of fat in the liver occurring in the absence of other causes, such as alcohol intake. An estimated 4.6% of women in the United States have unexplained aminotransferase elevations thought to be secondary to NAFLD (16). NAFLD accounts for up to 70% of cases of cryptogenic chronic hepatitis (17). There is a spectrum of manifestations of NAFLD that includes steatosis alone, steatosis with nonspecific inflammation, and NASH (18, 19). NASH is characterized by steatosis, ballooning degeneration of hepatocytes, and fibrosis. The dying hepatocytes sometimes trigger polymorphonuclear infiltration (20, 21). Patients with NASH are at a markedly increased risk of cirrhosis and liver-related death, with rates of 25 and 10% over 10 yr, respectively (19). Because steatosis (22, 23, 24, 25), steatohepatitis, and possibly fibrosis (23, 25) are reversible, early detection is important. Thus, it is important to identify clinical and biochemical characteristics that are associated with an increased risk of developing NASH.
Two small studies have recently suggested that women with PCOS may be at an increased risk for NAFLD. A retrospective study of 70 women with PCOS, evaluated at an infertility clinic, found abnormal alanine aminotransferase (ALT) activity in 30% of the population (26). Another study demonstrated ultrasound evidence of hepatic steatosis in 48 of 88 (55%) women with PCOS. Interestingly, only seven of the 48 women (15%) had elevated liver chemistries (27). Whether these women are at an increased risk for progressing to NASH and/or cirrhosis is unknown. Two studies of adults with abnormal liver tests who were referred to hepatologists concluded that advanced NAFLD was rare before the age of 45–50 yr (28, 29). Thus, it might be argued that young women with PCOS are unlikely to have liver damage. In contrast, the young age of these women could have important prognostic implications for developing liver complications later in life. Therefore, the aims of this retrospective study were to identify women with PCOS seen in our university-based medical endocrinology clinic who have biopsy-documented NASH and compare the characteristics of these women to women with PCOS who do not have biopsy-documented NASH. Additionally, we sought to determine the prevalence of abnormal aminotransferase activity, in the absence of other causes, in the cohort of PCOS women seen at our endocrinology clinic and compare the characteristics of these women to PCOS women with normal aminotransferase activity. Our hypothesis was that women with abnormal aminotransferase activity, including those with NASH, would have characteristics associated with metabolic syndrome.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Subjects
The study population was composed of women between the ages of 18 and 50 yr with PCOS seen by two of the investigators (A.J.B., K.C.P.) in the medical endocrinology clinic at Duke University Medical Center. The women were identified using the International Classification of Diseases code for PCOS. A list of all patients with a diagnosis of PCOS who were treated in this endocrine clinic in a 5-yr period (1999–2004) was generated. Women were referred from a variety of sources, including obstetric-gynecology clinics, primary care clinics, subspecialty clinics, and self-referral. Charts for each patient were reviewed to confirm that they were between 18 and 50 yr of age and had a clinical diagnosis of PCOS based on the 1990 National Institutes of Health consensus definition (30). Patients were included in the analysis if they had a history of irregular menstrual cycles and evidence of androgen excess, including hirsutism, acne, male-pattern hair loss, or an elevated testosterone level. Patients were excluded if their charts indicated the presence of any of the following exclusion criteria: 1) other causes of irregular menstrual cycles or androgen excess including hyperprolactinemia, uncontrolled thyroid disease, congenital adrenal hyperplasia, premature ovarian failure, Cushing’s syndrome, androgen-secreting tumor, or pregnancy; 2) history of known liver disease or other medical problem thought to cause an elevation in liver enzymes; 3) history of medication use clinically thought to cause an elevation in liver enzymes; or 4) history of significant alcohol consumption, defined as greater than one alcoholic beverage per day.
We identified 275 women with the diagnosis of PCOS evaluated during the specified time period. Of these, 14 women did not meet the inclusion criteria (four did not meet the age criteria, and 10 did not meet the study diagnostic criteria for PCOS). Aminotransferase levels were not documented in 54 women. Seven women were excluded from the analysis for the following reasons: four had perioperative elevations in their liver enzymes (three had a cholecystectomy, one had gastric bypass surgery), one had concomitant autoimmune hepatitis, and two had liver enzyme elevations thought to be secondary to medications. Therefore, 200 women were included in the analysis.
The following parameters were identified in each patient from the date closest to their initial PCOS evaluation, if available: age, self-reported race, blood pressure, height, weight, waist circumference, total cholesterol, triglycerides, high-density lipoprotein (HDL), and results from a 2-h glucose tolerance test including both insulin and glucose levels. Glucose tolerance tests consisted of a 75-g oral glucose challenge with fasting and 2-h glucose and insulin levels. Fasting lipids were used for this analysis if available, and nonfasting lipids were substituted if they were not. Testosterone levels were recorded. Over the course of the 5 yr in which patients were evaluated, several different measures of testosterone were performed including total testosterone, free testosterone, free androgen index, and bioavailable testosterone. All aspartate aminotransferase (AST) and ALT were reviewed. If the patient had an abnormal AST and/or ALT (AST or ALT > 60 U/liter at Duke Clinical Laboratory), the following additional data were collected to exclude other potential causes, if available: alkaline phosphatase, total bilirubin, viral hepatitis serologies, anti-nuclear antibody, ferritin, total iron binding capacity, iron, 
1 anti-trypsin, anti-smooth muscle antibody, anti-mitochondrial antibody, ceruloplasmin, ultrasound or other liver imaging results, and liver biopsy results. Outside labs were accepted only if the reference range was clearly documented. All liver biopsies were performed and reviewed at Duke University Medical Center.
Statistical analysis
The mean (SD) and median (25%, 75%) were calculated for all variables and normality of the distributions assessed. Because the data for many of the variables were markedly skewed, nonparametric tests were performed. We made two comparisons: between those with biopsy-documented NASH and those without, and between those with elevated aminotransferase levels, defined as AST and/or ALT more than 60 U/liter, and those without. For the comparisons of PCOS women with NASH vs. those without NASH, exact Wilcoxon rank sum tests were performed. For the comparisons of PCOS women with abnormal aminotransferase activity vs. those with normal aminotransferase activity, asymptotic Wilcoxon rank sum tests were performed. Statistical significance was set at two-sided P value of less than 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
.
|
The individual characteristics of the six women with biopsy-documented NASH are listed in Table 1. Five women had triglycerides more than or equal to 150 mg/dl, and all six had HDL equal to or less than 50 mg/dl. Additionally, four of the six women had complete oral glucose tolerance tests available, two of whom had impaired glucose tolerance.
|
compares the characteristics of PCOS women with biopsy-documented NASH to the PCOS women with no known NASH. The women with NASH had significantly higher triglycerides (median 245 vs. 132 mg/dl, P = 0.025), fasting insulin (26 vs. 13 µIU/ml, P = 0.038), AST (144 vs. 22 U/liter, P< 0.001), and ALT (143 vs. 28 U/liter, P < 0.001). They had lower HDL (34 vs. 50 mg/dl, P < 0.001). There was a tendency toward higher total cholesterol (233 vs. 201 mg/dl, P = 0.064), but it did not reach statistical significance. There was no significant difference in age, weight, body mass index (BMI), blood pressure, testosterone, oral contraceptive use, fasting glucose, or 2-h glucose or insulin.
|
. The group with elevated aminotransferases included the six patients with NASH. The women with elevated aminotransferases had significantly higher triglycerides (median 174 vs. 129 mg/dl, P = 0.024), higher fasting insulin (21 vs. 12 µIU/ml, P = 0.036), lower HDL (41 vs. 50 mg/dl, P = 0.006), and a tendency toward a higher 2-h glucose level (123 vs. 107 mg/dl, P = 0.058), compared with women with normal aminotransferase activity. There was no significant difference in age, weight, BMI, blood pressure, total cholesterol, fasting glucose, or 2-h insulin, although those with elevated aminotransferases were more likely to be on oral contraceptives than those with normal aminotransferases (48% vs. 28%; P = 0.030). Twenty-one of the 29 women with elevated aminotranferases had complete oral glucose tolerance tests available; seven (33%) had impaired glucose tolerance, and one had diabetes mellitus. Waist circumferences were recorded in very few patients (three women with aminotransferase elevations and 27 women with normal aminotransferase activity). Hence, we were unable to assess the impact of waist circumference on aminotransferase elevations. Furthermore, because testosterone was measured four different ways throughout the duration of the 5-yr period (total, bioavailable and free testosterone, and free androgen index), the number of subjects that had each individual measurement was too small to detect any meaningful differences between the groups.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All six (100%) of the women with persistently elevated aminotransferases who had liver biopsies had evidence of steatohepatitis with varying degrees of fibrosis, two of whom had bridging fibrosis. In a study of 354 asymptomatic adults with unexplained elevated liver enzymes, steatohepatitis was demonstrated in only 34% (32). Furthermore, of those with NASH, only 18% had bridging fibrosis. This suggests that the women who underwent liver biopsies in our study had a very high likelihood of advanced disease and that the true prevalence of NASH in this cohort of women may be higher than what is being reported in this article (six of 200 or 3%). Moreover, these women had evidence of not only inflammation, but also fibrosis, on their liver biopsies. Fibrosis is a sign of potentially progressive liver disease (19). Early detection is important because intervention at an early stage of fibrosis may slow, or even stop, progression of liver disease (23, 25).
Our finding that dyslipidemia is associated with abnormal aminotransferase activity (including NASH) in women with PCOS is consistent with the findings of Clark et al. (17), who found that the risk of having NAFLD was increased by a factor of 3 in National Health and Nutrition Examination Survey (NHANES) III subjects who had triglycerides more than 200 mg/dl and by a factor of almost 2 in subjects with an HDL less than 35 mg/dl. In addition, the high fasting insulin level seen in PCOS women with abnormal aminotransferase activity (including NASH) is not surprising given the association between insulin resistance and NAFLD/NASH (12, 13, 14, 15). However, we were surprised to find no statistically significant difference in weight or BMI between the groups. It is possible that our sample size was not large enough to detect this difference. Alternatively, it is also possible that there is a subgroup of women with PCOS who are more prone to developing complications such as dyslipidemia, insulin resistance, and NAFLD, even though they are not more obese than other women with PCOS.
The overall prevalence of abnormal aminotransferase activity (15%) in our cohort of women with PCOS was lower than the percentage found by Schwimmer et al. (30%) (26). However, we used different definitions of abnormal aminotransferase activity. Schwimmer et al. defined abnormal aminotransferase activity as an ALT more than 35 U/liter or AST more than 40 U/liter, which was the upper limit of normal in their clinical laboratory. Our definition of abnormal aminotransferase activity was higher: AST or ALT more than 60 U/liter, the upper limit of normal in our laboratory. If we use the same cutoffs as Schwimmer et al., the prevalence rate of abnormal aminotransferase activity in our population of women with PCOS is similar (28%) to the San Diego cohort. Furthermore, the prevalence of unexplained abnormal aminotransferase activity is much higher in both PCOS populations than was found in women in the NHANES database (4.6%) (16). The aminotransfersase cutoff used for the NHANES study was even lower: ALT and/or AST more than 31 U/liter. If we apply the same criteria to our sample, approximately 38% of the women would have elevated aminotransferase activity. Finally, it has been suggested that an ALT more than 19 U/liter should be considered abnormal in women (33). Eighty-four percent of the women in our study had an ALT more than 19 U/liter. Thus, the cutoff used in our study was very conservative relative to those used in other studies. However, even the use of this conservative assessment indicates that the prevalence of abnormal aminotransferase activity is higher in women with PCOS than in the general population of women in the United States.
Although this is not a controlled trial, another observation that might be significant is the improvement in aminotransferase activity through lifestyle modifications, as well as metformin therapy, even in the women with biopsy-documented NASH. Lifestyle modifications including diet, exercise, and weight loss are generally recommended as treatment modalities for overweight individuals with evidence of NAFLD (34). Small studies have demonstrated beneficial effects on liver enzymes (35) as well as liver histology (22, 24) through weight loss by diet alone or in combination with exercise. There has been little published on the benefits of weight loss in patients with more advanced liver disease (25). We previously reported a woman with PCOS and NASH who had marked improvement in her histological findings (documented by liver biopsy) with diet, exercise, and moderate weight loss (36). Additionally, four of the six women with biopsy-documented NASH in this retrospective study demonstrated normalization of aminotransferases with either diet and exercise alone or in combination with metformin. These findings support the role for further study of lifestyle modifications and insulin sensitizers in patients with advanced liver disease.
There are a number of limitations present in this study. First, our patient population is from our medical endocrinology clinic, and therefore, our population is a convenience sample rather than a sample recruited from the community. Moreover, because Duke University Medical Center is a single, tertiary referral center, there is a referral filter bias. However, many of our patients are also self-referred, which may decrease this bias. Second, alcohol consumption, one of the exclusion criteria, was determined by self-report and may not be accurate in all subjects. Third, although liver enzymes are commonly used to screen for NAFLD, normal liver enzymes do not exclude the presence of NAFLD. Thus, it is likely that NAFLD, including NASH, is underestimated in our study.
Finally, this study was retrospective and complete data were not available on all of the patients. Imaging studies were available in only 16 of the 29 subjects with abnormal aminotransferase activity. In addition, although a complete evaluation to exclude other liver pathology was negative in 11 of the 29 subjects that had an ALT and/or AST more than 60 U/liter, 18 subjects did not undergo a complete evaluation. Therefore, some of the remaining 18 subjects may have had another underlying cause for their aminotransferase elevations. In the NHANES study, aminotransferase elevations were unexplained in 69% of subjects and, thus, assumed to be secondary to NAFLD, but 31% of the subjects had another identifiable cause for their liver enzyme elevations (16). If we assume that 31% of the remaining 18 women (i.e. six individuals) who did not undergo a complete liver evaluation in our study actually had some cause of liver disease other than NAFLD, the prevalence of NAFLD in our PCOS population would be approximately 12% (23 of 200 individuals). Although prospective studies are needed to eliminate problems associated with incomplete data, published information supports our conclusion that the prevalence of NAFLD is higher in young women with PCOS than in the general adult population.
In conclusion, abnormal aminotransferase activity is relatively common in women with PCOS and is associated with high triglycerides, low HDL, and elevated fasting insulin levels. Thus, these characteristics may help identify PCOS women who are at increased risk of fatty liver disease. In addition, the PCOS women in our study were diagnosed with NASH in their 20s and 30s. The young age of many women with PCOS, and the relatively advanced stage of NASH seen on the biopsies of our patients, suggests the possibility of significant risk for long-term complications from liver disease. Therefore, evaluation for liver disease in women with PCOS may be warranted at an earlier age than has been previously recommended for the general population. Further investigation regarding screening, prevalence, and treatment modalities for fatty liver in women with PCOS is needed. Additionally, a longitudinal study following women with PCOS and NAFLD may provide important insight into this disease.
|
Footnotes |
|---|
Disclosure Statements: T.L.S., N.D.H., L.L.S., K.C.P., and A.J.B. have nothing to declare. A.M.D. consults for Icoria and Axscan Scandipharm and has received lecture fees from Axcan Scandipharm and GlaxoSmithKline.
First Published Online February 21, 2006
Abbreviations: ALT, Alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; HDL, high-density lipoprotein; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; NHANES, National Health and Nutrition Examination Survey; PCOS, polycystic ovary syndrome.
Received December 20, 2005.
Accepted February 14, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. K. Dowman, J.W. Tomlinson, and P.N. Newsome Pathogenesis of non-alcoholic fatty liver disease QJM, November 13, 2009; (2009) hcp158v1. [Full Text] [PDF]
|
|
|
|
 |
|
 E. Vassilatou, S. Lafoyianni, A. Vryonidou, D. Ioannidis, L. Kosma, K. Katsoulis, E. Papavassiliou, and I. Tzavara Increased androgen bioavailability is associated with non-alcoholic fatty liver disease in women with polycystic ovary syndrome Hum. Reprod., November 3, 2009; (2009) dep380v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Cussons, G. F. Watts, T. A. Mori, and B. G. A. Stuckey Omega-3 Fatty Acid Supplementation Decreases Liver Fat Content in Polycystic Ovary Syndrome: A Randomized Controlled Trial Employing Proton Magnetic Resonance Spectroscopy J. Clin. Endocrinol. Metab., October 1, 2009; 94(10): 3842 - 3848. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Denzer, D. Thiere, R. Muche, W. Koenig, H. Mayer, W. Kratzer, and M. Wabitsch Gender-Specific Prevalences of Fatty Liver in Obese Children and Adolescents: Roles of Body Fat Distribution, Sex Steroids, and Insulin Resistance J. Clin. Endocrinol. Metab., October 1, 2009; 94(10): 3872 - 3881. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Palomba, A. Falbo, F. Zullo, and F. Orio Jr. Evidence-Based and Potential Benefits of Metformin in the Polycystic Ovary Syndrome: A Comprehensive Review Endocr. Rev., February 1, 2009; 30(1): 1 - 50. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Barbieri Update in Female Reproduction: A Life-Cycle Approach J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2439 - 2446. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. A. Raskauskas, R. A. Fisken, G. Targher, G. Lippi, P. Moghetti, L. Mascitelli, F. Pezzetta, M. R. Goldstein, and J. E. Nestler Metformin in Polycystic Ovary Syndrome N. Engl. J. Med., April 24, 2008; 358(17): 1866 - 1868. [Full Text] [PDF]
|
|
|
|
|
|
T. M. Barber, S. J. Golding, C. Alvey, J. A. H. Wass, F. Karpe, S. Franks, and M. I. McCarthy Global Adiposity Rather Than Abnormal Regional Fat Distribution Characterizes Women with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., March 1, 2008; 93(3): 999 - 1004. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Ibanez, A. Lopez-Bermejo, L. del Rio, G. Enriquez, C. Valls, and F. de Zegher Combined Low-Dose Pioglitazone, Flutamide, and Metformin for Women with Androgen Excess J. Clin. Endocrinol. Metab., May 1, 2007; 92(5): 1710 - 1714. [Abstract] [Full Text] [PDF]
|
|
Read all eLetters
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
