Advertisement

Dietary fat intake, erythrocyte fatty acids, and risk of uterine fibroids

  • Holly R. Harris
    Correspondence
    Reprint requests: Holly Harris, Sc.D., Program in Epidemiology, Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue North, M4-B859, Seattle, Washington 98109-1024.
    Affiliations
    Program in Epidemiology, Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, Washington

    Department of Epidemiology, University of Washington, Seattle, Washington
    Search for articles by this author
  • A. Heather Eliassen
    Affiliations
    Channing Division of Network Medicine, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts

    Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, Massachusetts
    Search for articles by this author
  • David R. Doody
    Affiliations
    Program in Epidemiology, Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, Washington
    Search for articles by this author
  • Kathryn L. Terry
    Affiliations
    Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, Massachusetts

    Obstetrics and Gynecology Epidemiology Center, Brigham and Women’s Hospital, Boston, Massachusetts
    Search for articles by this author
  • Stacey A. Missmer
    Affiliations
    Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, Massachusetts

    Division of Adolescent and Young Adult Medicine, Department of Medicine, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts

    Department of Obstetrics, Gynecology, and Reproductive Biology, College of Human Medicine, Michigan State University, Grand Rapids, Michigan
    Search for articles by this author

      Objective

      To prospectively evaluate the association between dietary fat intake and risk of uterine fibroids; and to evaluate the association between erythrocyte membrane fatty acid (FA) levels and fibroid risk.

      Design

      Prospective cohort study. Cox proportional hazard models were used to calculate hazard ratios and 95% confidence interval (CI). In a subset of participants 34 individual FAs were measured and logistic regression analysis was used to estimate odds ratios (ORs) and 95% CI for the association between FA tertiles and fibroids.

      Setting

      Not applicable.

      Patient(s)

      Premenopausal US women (81,590) in the Nurses’ Health Study II, aged 25–42 years at enrollment in 1989 for whom diet was assessed by a food frequency questionnaire. A total of 553 participants with erythrocyte FA measurements.

      Intervention(s)

      Not applicable.

      Main Outcome Measure(s)

      Cases of fibroids were defined on the basis of self-reported ultrasound or hysterectomy confirmation.

      Result(s)

      A total of 8,142 cases of ultrasound-confirmed or hysterectomy-confirmed were diagnosed during an 18-year period (1991-2009). No associations were observed between intake of any dietary fats and fibroids in the multivariable models. However, when erythrocyte FAs were examined, an inverse association was observed between total n-3 polyunsaturated FAs and likelihood of fibroids (OR for third versus first tertile, 0.41; 95% CI 0.19–0.89). In addition, total trans FAs were associated with more odds of fibroids (OR for third tertile, 3.33; 95% CI 1.50–7.38).

      Conclusion(s)

      Our findings provide preliminary suggestions that n-3 polyunsaturated FAs and trans FAs may play a role in fibroid etiology; however, these results should be confirmed in future studies.
      Ingesta dietética de grasas, ácidos grasos de los eritrocitos y riesgo de miomas uterinos.

      Objetivo

      Evaluar de manera prospectiva la asociación entre la ingesta de grasa en la dieta y el riesgo de miomas uterinos; y evaluar la asociación entre los niveles de ácidos grasos (AF) de la membrana del eritrocito y el riesgo de miomas

      Diseño

      Estudio prospectivo de cohortes. Se utilizaron modelos de riesgo proporcionales de Cox para calcular las razones de riesgo y el intervalo de confianza (IC) del 95%. En un subconjunto de participantes, se midieron 34 AF individuales y se utilizó un análisis de regresión logística para estimar los odds ratios (OR) y el IC del 95% para la asociación entre terciles de FA y fibromas.

      Ubicación

      No aplica.

      Paciente(s)

      Mujeres estadounidenses premenopáusicas (81, 590) en el Nurses ’Health Study II, con edades comprendidas entre los 25 y los 42 años en el momento de la inscripción en 1989 para quienes la dieta se evaluó mediante un cuestionario de frecuencia alimentaria. Un total de 553 participantes con mediciones de FA en eritrocitos.

      Intervención(es)

      No aplica

      Principales medidas de resultado(s)

      Los casos de miomas fueron definidos en base de la ecografía reportadas por ellas mismas o en la confirmación durante una histerectomía.

      Resultados

      Un total de 8.142 casos fueron confirmado por ecografía o confirmados tras una histerectomía fueron diagnosticados durante un período de 18 años (1991-2009). No se observaron asociaciones entre la ingesta de grasas alimentarias y los fibromas en los modelos multivariables. Sin embargo, cuando se examinaron los AG de eritrocitos, se observó una asociación inversa entre los ácidos grasos poliinsaturados n-3 totales y la probabilidad de miomas (OR para el tercer tercil frente al primer tercil, 0.41; IC del 95%: 0.19-0.89). Además, los ácidos grasos trans totales se asociaron con más probabilidades de miomas (OR para el tercer tercil, 3.33; IC del 95%: 1.50 - 7.38).

      Conclusión(es)

      Nuestros hallazgos aportan sugerencias preliminares sobre que los ácidos grasos poliinsaturados n-3 y los ácidos grasos trans puedan desempeñar un papel en la etiología de los miomas; sin embargo, estos resultados deben confirmarse en estudios futuros.

      Key Words

      Discuss: You can discuss this article with its authors and other readers at https://www.fertstertdialog.com/users/16110-fertility-and-sterility/posts/62242-28672
      Uterine fibroids are the most common pelvic tumor in reproductive aged women (
      • Cramer S.
      • Patel A.
      The frequency of uterine leiomyomas.
      ). They are the primary indication for hysterectomies in the United States with ≥200,000 procedures annually and are a leading cause of hospitalizations for gynecologic conditions unrelated to pregnancy (
      • Velebil P.
      • Wingo P.A.
      • Xia Z.
      • Wilcox L.S.
      • Peterson H.B.
      Rate of hospitalization for gynecologic disorders among reproductive-age women in the United States.
      ). Although nonmalignant, fibroids are frequently associated with pelvic pain, abnormal uterine bleeding, infertility, and adverse pregnancy outcomes. Despite the high morbidity and health care costs associated with fibroids, the etiology is not fully understood, and few modifiable risk factors have been identified.
      Dietary factors may play a role in fibroid etiology due to their potential to modify endogenous hormones as well as their inflammatory effects. For example, trans-fat intake influences circulating levels of interleukin (IL) 6 and other inflammatory markers (
      • Baer D.J.
      • Judd J.T.
      • Clevidence B.A.
      • Tracy R.P.
      Dietary fatty acids affect plasma markers of inflammation in healthy men fed controlled diets: a randomized crossover study.
      ,
      • Mozaffarian D.
      • Pischon T.
      • Hankinson S.E.
      • Rifai N.
      • Joshipura K.
      • Willett W.C.
      • et al.
      Dietary intake of trans fatty acids and systemic inflammation in women.
      ,
      • Mozaffarian D.
      • Rimm E.B.
      • King I.B.
      • Lawler R.L.
      • McDonald G.B.
      • Levy W.C.
      Trans fatty acids and systemic inflammation in heart failure.
      ) and a chronic inflammatory milieu has been hypothesized to promote fibroid development (
      • Wegienka G.
      Are uterine leiomyoma a consequence of a chronically inflammatory immune system?.
      ). However, to our knowledge, only one prospective study (
      • Wise L.A.
      • Radin R.G.
      • Kumanyika S.K.
      • Ruiz-Narváez E.A.
      • Palmer J.R.
      • Rosenberg L.
      Prospective study of dietary fat and risk of uterine leiomyomata.
      ), the Black Women’s Health Study (BWHS) has examined the association between dietary fat intake and fibroid risk, observing a small increased risk with intake of several specific omega-3 fatty acids (FAs) and no clear associations with total fat or other fat subtypes. Fish consumption was a large contributor to omega-3 FA consumption in this cohort, thus environmental contaminants through fish intake could explain this increased risk. The BWHS did not measure circulating FAs that capture dietary intake and FA metabolism, more precisely reflecting the internal dose. Furthermore, the FA composition of the erythrocyte membrane reflects dietary intake for months, in contrast to serum and plasma that may only reflect dietary intake for days to weeks (
      • Arab L.
      Biomarkers of fat and fatty acid intake.
      ). Examination of these biomarkers may provide additional insight into fibroid etiology.
      In the present study we used data from the prospective Nurses’ Health Study II (NHSII) to investigate whether intake of dietary fats was associated with ultrasound-confirmed or hysterectomy-confirmed uterine fibroids during an 18-year follow-up period. In a subset of women, we also examined the association between FAs measured in erythrocyte membranes and subsequent fibroid risk.

      Materials and methods

       Study Population

      The NHSII is an ongoing prospective cohort study established in 1989. At baseline, 116,429 US female registered nurses aged 25–42 years completed a questionnaire that collected information on demographic and lifestyle factors, anthropometric variables, and disease history. Follow-up questionnaires are sent biennially to update information on exposures and disease status. Additional study details have been provided elsewhere (
      • Solomon C.G.
      • Willett W.C.
      • Carey V.J.
      • Rich-Edwards J.
      • Hunter D.J.
      • Colditz G.A.
      • et al.
      A prospective study of pregravid determinants of gestational diabetes mellitus.
      ). The cumulative follow-up for NHSII is >95%. Ongoing consent was assumed upon return of the completed questionnaire. This study was approved by the institutional review boards of the Harvard T.H. Chan School of Public Health and the Brigham and Women’s Hospital, Boston, Massachusetts.
      Between 1996 and 1999, 29,611 NHSII participants, aged 32–54 years, provided blood samples and answered a short questionnaire at blood collection including information on date, time, and number of hours since last food intake. After overnight shipment all samples were processed into plasma, white blood cell, and red blood cell components and were stored at ≤-130°C in continuously monitored liquid nitrogen freezers. Further details of the blood collection procedure for NHSII have been described previously (
      • Tworoger S.S.
      • Sluss P.
      • Hankinson S.E.
      Association between plasma prolactin concentrations and risk of breast cancer among predominately premenopausal women.
      ). The NHSII participants who provided a blood sample were similar to the total cohort (
      • Tworoger S.S.
      • Sluss P.
      • Hankinson S.E.
      Association between plasma prolactin concentrations and risk of breast cancer among predominately premenopausal women.
      ).

       Analytical Populations

      Follow-up for the analyses that used the diet data began in 1991 when 97,813 NHSII participants first returned the food frequency questionnaire (FFQ) and concluded in 2009, the last year uterine fibroid incidence was assessed on the biennial questionnaire. Criteria for exclusion were as follows: implausible total energy intake (<600 kcal/d or >3,500 kcal/d), blank entries for >70 food items on the 1991 FFQ, or a diagnosis of uterine fibroid or cancer diagnosis (other than non-melanoma skin cancer) before June 1991. The analytical cohort was limited to those who were premenopausal and had intact uteri. After these exclusions, 81,590 premenopausal women with dietary information remained.
      Participants in the erythrocyte membrane FA analysis were selected from 794 control participants in a previous nested case-control study that examined erythrocyte membrane FAs and breast cancer risk (
      • Hirko K.A.
      • Chai B.
      • Spiegelman D.
      • Campos H.
      • Farvid M.S.
      • Hankinson S.E.
      • et al.
      Erythrocyte membrane fatty acids and breast cancer risk: a prospective analysis in the nurses' health study II.
      ). From the original 794 controls, we excluded those with a fibroid diagnosis before blood draw (n = 65), hysterectomy before blood draw (n = 87), and those who were postmenopausal or missing menopausal status (n = 89). After these exclusions 553 women with previously measured erythrocyte membrane FA levels remained.

       Dietary Assessment

      Diet was assessed in 1991, 1995, 1999, 2003, and 2007 using an FFQ that listed >130 food items. Participants were asked how often, on average, they consumed each type of food or beverage during the previous year. For each food item, nine responses were possible, ranging from never or less than once per month to six or more times per day. Nutrient intakes were calculated by first multiplying the portion size of a single serving of each food by its reported frequency of intake for the total amount of food consumed, and then multiplying the total amount consumed by the nutrient content of the food, and then summing across all food items. The questionnaire included information about specific types of margarine and fats used for baking and frying and this was incorporated into the nutrient calculations. Nutrient values in foods were obtained from the US Department of Agriculture (Nutrient Data Laboratory), food manufacturers, independent academic sources, and our own FA analyses of commonly used margarines, cooking oils, and baked foods (
      • Dial S.
      The Antioxidant Vitamins C and E.
      ,
      • Holland G.
      • Unwin I.
      • Buss D.
      • Paul A.
      • Dat S.
      The composition of foods.
      ,
      USDA
      USDA Nutrient Database for Standard Reference.
      ). The food composition database has been updated every 4 years to account for changes in the food supply including updated FA analyses. The reproducibility and validity of the NHSII FFQ has been previously reported (
      • Salvini S.
      • Hunter D.J.
      • Sampson L.
      • Stampfer M.J.
      • Coldtiz G.A.
      • Rosner B.
      • et al.
      Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption.
      ,
      • Willett W.C.
      • Sampson L.
      • Stampfer M.J.
      • Rosner B.
      • Bain C.
      • Witschi J.
      • et al.
      Reproducibility and validity of a semiquantitative food frequency questionnaire.
      ,
      • Yuan C.
      • Spiegelman D.
      • Rimm E.B.
      • Rosner B.A.
      • Stampfer M.J.
      • Barnett J.B.
      • et al.
      Validity of a dietary questionnaire assessed by comparison with multiple weighed dietary records or 24-hour recalls.
      ) with deattenuated correlation coefficients between the FFQ and 7-day diet records of 0.67 for total fat, 0.69 for saturated fat (SFA), 0.57 for polyunsaturated fat (PUFA), 0.56 for monounsaturated fat (MUFA), 0.69 omega-3 FAs, 0.65 for cholesterol (
      • Yuan C.
      • Spiegelman D.
      • Rimm E.B.
      • Rosner B.A.
      • Stampfer M.J.
      • Barnett J.B.
      • et al.
      Validity of a dietary questionnaire assessed by comparison with multiple weighed dietary records or 24-hour recalls.
      ), and 0.66 for fish intake (
      • Salvini S.
      • Hunter D.J.
      • Sampson L.
      • Stampfer M.J.
      • Coldtiz G.A.
      • Rosner B.
      • et al.
      Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption.
      ).

       Laboratory Assays

      Erythrocyte FA concentrations were assayed in Dr. Hannia Campos’ laboratory at the Harvard T.H. Chan School of Public Health using gas-liquid chromatography. A detailed description of the laboratory process has been published elsewhere (
      • Baylin A.
      • Kim M.K.
      • Donovan-Palmer A.
      • Siles X.
      • Dougherty L.
      • Tocco P.
      • et al.
      Fasting whole blood as a biomarker of essential fatty acid intake in epidemiologic studies: comparison with adipose tissue and plasma.
      ). Masked replicates from pooled specimens were included for quality control. Of the 34 FAs included in our analysis, 8 FAs with levels close to the detection limit had coefficients of variation between 20% and 95% (lauric acid, mystristic acid, pentadecanoic acid, mysristoleic acid, docosadienoic acid, palmitelaidic acid, linolelaidic acid, and octadecadienoic acid). Additional details are available elsewhere (
      • Hirko K.A.
      • Chai B.
      • Spiegelman D.
      • Campos H.
      • Farvid M.S.
      • Hankinson S.E.
      • et al.
      Erythrocyte membrane fatty acids and breast cancer risk: a prospective analysis in the nurses' health study II.
      ).

       Fatty Acids

      We examined erythrocyte FAs individually and in the following groups by type: SFAs, MUFAs, n-3 PUFAs, n-6 PUFA, and trans FAs. In addition, we calculated the ratio of total n-6 PUFA to total n-3 PUFA, as this ratio has been hypothesized to predict several chronic inflammatory diseases (
      • Simopoulos A.P.
      The importance of the omega-6/omega-3 fatty acid ratio in cardiovascular disease and other chronic diseases.
      ). The saturation indices, SIn-7 (palmitic/palmitoleic acid) and SIn-9 (stearic/oleic acid), were considered as indicators of the steroyl coenzyme-A desaturase activity (
      • Chajes V.
      • Joulin V.
      • Clavel-Chapelon F.
      The fatty acid desaturation index of blood lipids, as a biomarker of hepatic stearoyl-CoA desaturase expression, is a predictive factor of breast cancer risk.
      ,
      • Patel P.S.
      • Sharp S.J.
      • Jansen E.
      • Luben R.N.
      • Khaw K.T.
      • Wareham N.J.
      • et al.
      Fatty acids measured in plasma and erythrocyte-membrane phospholipids and derived by food-frequency questionnaire and the risk of new-onset type 2 diabetes: a pilot study in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Norfolk cohort.
      ). We also examined SFA and trans FA primarily from milk or meat from cattle or other ruminants (15:0 + 17:0 + 16:1n–7t), termed dairy derived FAs, and trans FA from partially hydrogenated oils (18:1 trans+18:2 trans), termed industrial trans for this analysis. Previous work has shown that erythrocyte content of FAs are closely correlated with plasma content with an average correlation of 0.72 (
      • Sun Q.
      • Ma J.
      • Campos H.
      • Hankinson S.E.
      • Hu F.B.
      Comparison between plasma and erythrocyte fatty acid content as biomarkers of fatty acid intake in US women.
      ).

       Ascertainment and Definition of Uterine Fibroids

      Starting in 1993, participants were asked on each biennial questionnaire if they ever received a physician diagnosis of uterine fibroids, and if so, the date of diagnosis and whether the diagnosis was confirmed by pelvic exam, ultrasound, or hysterectomy. Cases were defined on the basis of self-reported ultrasound confirmation or hysterectomy confirmation. Participants who reported fibroids not confirmed by ultrasound or hysterectomy did not contribute person-time to that study period but were allowed to reenter the analysis if confirmed by ultrasound or hysterectomy in the future.
      In a previous validation study, a subset of newly diagnosed cases confirmed by ultrasound or hysterectomy (N = 243, 100 white and 143 African American) were mailed a questionnaire on symptoms and requested a review of their medical records (
      • Marshall L.M.
      • Spiegelman D.
      • Barbieri R.L.
      • Goldman M.B.
      • Manson J.E.
      • Colditz G.A.
      • et al.
      Variation in the incidence of uterine leiomyoma among premenopausal women by age and race.
      ). Of the 216 who responded (89%), 6% denied the diagnosis and 34% confirmed the diagnosis but did not release their medical records. Among the cases in which medical records could be obtained, 93% were confirmed. The proportion diagnosed by hysterectomy, myomectomy, examination under anesthesia or ultrasound did not differ between those who did and did not give permission for medical record release. The proportion confirmed by medical record did not differ comparing white (94%) and African-American (92%) participants.

       Statistical Analysis

       Dietary intake analysis

      In the analysis examining dietary intake assessed from the FFQ, participants contributed follow-up time from the return of the 1991 questionnaire until self-report of a uterine fibroid, diagnosis of any cancer (except non-melanoma skin cancer), death, loss to follow-up, hysterectomy, menopause, or until return of the 2009 questionnaire (the last year uterine fibroid incidence was assessed on the biennial questionnaire—at which time the youngest participant was aged 45 years)—whichever occurred first. Cox proportional hazards regression models were used with age and the questionnaire period as the time scale to estimate hazard ratios (HR) and 95% confidence interval (CI). The lowest category of intake of each nutrient or food was used as the reference group. We examined associations with total fat, vegetable fat, animal fat, SFA, trans-unsaturated fat, MUFA, PUFA, long-chain omega-6 FAs, long-chain omega-3 FAs, eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA), and docosapentaenoic acid (DPA). We also examined dietary cholesterol that comes from intake of animal products also high in fats. In addition, we examined the association with dark meat fish intake as it is the main contributor of long-chain omega-3 FA. Long-chain omega-3 FA intake in the NHSII has also been associated with environmental pollutants (e.g., polychlorinated biphenyls), which have been proposed to increase fibroid risk (
      • Lambertino A.
      • Turyk M.
      • Anderson H.
      • Freels S.
      • Persky V.
      Uterine leiomyomata in a cohort of Great Lakes sport fish consumers.
      ) and was suggestively associated with fibroid risk in the BWHS (
      • Wise L.A.
      • Radin R.G.
      • Kumanyika S.K.
      • Ruiz-Narváez E.A.
      • Palmer J.R.
      • Rosenberg L.
      Prospective study of dietary fat and risk of uterine leiomyomata.
      ).
      Cumulative average consumption is reported, as this method captures long-term dietary intake and minimizes measurement error due to within-person variation over time (
      • Hu F.B.
      • Stampfer M.J.
      • Rimm E.
      • Ascherio A.
      • Rosner B.A.
      • Spiegelman D.
      • et al.
      Dietary fat and coronary heart disease: a comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements.
      ). Covariate adjusted models included the following potential confounders that were chosen a priori due to their association with fibroids or dietary factors: total calories, race/ethnicity, age at menarche, infertility, parity, age at first birth, time since last birth, age of first oral contraceptive use, menstrual cycle length, body mass index, smoking, recent gynecologic/breast exam, and use of antihypertensive medications/diastolic blood pressure. Covariates were updated throughout the analysis as new information became available from the biennial questionnaires. Tests for linear trend of the exposures of interest were performed by assigning the median value of each category to all participants in that group.

       Erythrocyte FA analysis

      Tertiles of FAs were determined by the distribution among the controls. Logistic regression analysis was used to estimate odds ratios (ORs) and 95% CI for the association between FA tertiles and fibroids. The final covariate adjusted models included characteristics of the blood draw as well as the following potential confounders that were associated with FA levels in this dataset: age (continuous), blood draw time (1am-8am, 9am-12pm, 1pm-midnight), fasting status (yes/no), blood draw season (Nov–Apr, May–Oct), race (white, nonwhite), parity (nulliparous, 1, 2, 3, 4+), age at first birth (<25, 25–30, 31+ years), time since last birth (<6, 6–12, 13+ years), age at menarche (<12, 12–13, 14+ years), menstrual cycle length (<26, 26–31, 32+ days), and body mass index (<20, 20–21.9, 22–23.9, 24–24.9, 25–26.9, 27–029.9, 30+). In addition, we examined age at first oral contraceptive use, smoking, recent gynecologic/breast exam, and use of antihypertensive medications/diastolic blood pressure as potential confounders, but none were associated with FAs in this dataset and thus were not included in the final model. Tests for linear trend of the exposures of interest were performed by assigning the median value of each category to all participants in that group and additionally by examining the FAs as continuous variables. All statistical analyses were performed using SAS, version 9.4 (SAS Institute Inc.) and all tests of statistical significance were two-sided.

      Results

      During 1,536,355 person-years of follow-up contributed by 81,590 women, 8,142 incident cases of ultrasound-confirmed or hysterectomy-confirmed uterine fibroids were reported. Women with a higher intake of total fat were more likely to be a current smoker or parous, had a higher mean body mass index, and were less likely to have had a recent gynecologic exam (Table 1).
      Table 1Distribution of potential risk factors for uterine fibroids according to total fat intake among women in the Nurses' Health Study II at baseline in 1991.
      CharacteristicTotal fat intake quintile
      12345
      Women, n15,22314,78014,94916,16219,856
      Age, y36.1 (4.6)36.0 (4.6)36.0 (4.6)36.1 (4.6)36.4 (4.6)
      White90.092.692.893.393.6
      BMI, kg/m223.4 (4.5)24.0 (4.7)24.3 (5.0)24.8 (5.3)25.6 (6.0)
      Cigarette smoking
       Never67.267.266.965.964.4
       Past23.322.522.122.120.3
       Current9.510.211.012.115.3
      Age at menarche, y
       <1223.823.422.724.124.5
       1230.029.830.730.330.8
       1327.328.228.627.927.1
       ≥1418.818.618.117.717.6
      Menstrual cycle length, d
       <2616.615.816.716.717.3
       26–3168.269.168.068.367.8
       32–5011.211.011.111.110.5
       ≥51 or irregular cycles3.94.14.13.94.3
      Ever use of oral contraceptives82.083.883.684.685.1
      Age at first use of oral contraceptives, y
       Never18.516.616.715.815.2
       13–165.55.05.15.15.4
       17–2038.640.140.141.742.0
       21–2426.928.628.628.628.8
       ≥2510.69.89.58.88.6
      Nulliparous35.627.724.623.624.7
      Infertility diagnosis6.96.55.96.06.0
      Time since last birth, y
       <16.86.76.45.95.4
       1–329.630.330.529.127.5
       4–515.414.715.115.215.0
       6–710.911.811.911.812.3
       8–99.39.09.710.510.0
       ≥1028.027.526.427.529.8
      Recent gynecologic exam
       No exam12.612.713.414.616.6
      Antihypertensive medication use2.62.42.72.73.2
      Diastolic blood pressure
       <6525.923.022.421.420.0
       65–7448.349.549.047.046.4
       75–8421.022.122.925.225.8
       85–892.93.43.74.04.8
       ±902.02.02.12.42.9
      Total calories, kcal1,778 (547)1,794 (534)1,814 (541)1,814 (550)1,780 (564)
      Total fat intake, g47.0 (5.5)56.6 (1.7)62.0 (1.5)67.5 (1.7)77.4 (5.9)
      Note: Data presented as mean (standard deviation) or percent, unless stated otherwise. All data shown are standardized to the age distributions of the 1991 cohort. BMI = body mass index.
      Total fat intake was not associated with risk of fibroids (Table 2). When specific types of fat were examined there was the suggestion of a higher risk of fibroids with trans-unsaturated fat intake and cholesterol intake in the age-adjusted analyses (HR [95% CI] for fifth quintile vs. first quintile of 1.07 [1.00–1.15]; Ptrend = .02 and 1.10 [1.03–1.18]; Ptrend = .003, respectively). However, the associations were attenuated after adjustment for covariates (HR [95% CI] for fifth quintiles = 1.00 [0.93–1.07]; Ptrend = .99 and 1.04 [0.96–1.11]; Ptrend = .34, respectively). No other dietary fats were associated with fibroids risk (Table 2) nor was dark meat fish intake (results not shown).
      Table 2Hazard ratios and 95% confidence intervals for uterine fibroids according to quintiles of dietary fat intake (energy-adjusted g/d) in the Nurses' Health Study II, 1991-2009.
      CharacteristicCases, nAge-adjusted HR (95% CI)MV HR (95% CI)
      Adjusted for age (continuous), total calories (continuous), race/ethnicity (white, black, Hispanic, Asian, other), age at menarche (y; <11, 11, 12, 13, 14–15, >15), infertility (yes, no), parity (nulliparous, 1, 2, 3, 4+), age at first birth (y; <25, 25–30, >30), time since last birth (y; <1, 1–3, 4–5, 6–7, 8–9, 10–12, 13–15, 16+), age first oral contraceptive use (y; 13–16, 17–20, 21–24, 25+), menstrual cycle length (d; <26, 26–31, 32–50, >50), body mass index (<20, 20–21.9, 22–23.9, 24–24.9, 25–26.9, 27–29.9, 30+), smoking (never, past, current), recent gynecologic/breast exam (no recent exam, recent exam), and use of antihypertensive medications/diastolic blood pressure (no meds <65, no meds 65–74, no meds 75–84, no meds 85–89, no meds 90+, meds <65, meds 65–74, meds 75–84, meds 85–89, meds 90+).
      Total fat
       11,5581.00Referent1.00Referent
       21,6631.040.97, 1.121.040.97, 1.11
       31,6971.070.99, 1.141.060.98, 1.13
       41,7071.091.02, 1.171.081.00, 1.15
       51,5171.030.96, 1.100.980.91, 1.06
      Ptrend
      Determined using category medians.
      .21.92
      Vegetable fat
       11,4701.00Referent1.00Referent
       21,6671.030.96, 1.111.030.96, 1.11
       31,7231.050.97, 1.121.030.96, 1.11
       41,6671.000.93, 1.080.990.92, 1.06
       51,6151.000.93, 1.070.980.91, 1.05
      Ptrend
      Determined using category medians.
      .63.22
      Animal fat
       11,6581.00Referent1.00Referent
       21,6290.990.92, 1.060.970.91, 1.04
       31,6941.050.98, 1.121.030.97, 1.11
       41,6471.050.98, 1.131.030.96, 1.10
       51,5141.020.95, 1.100.990.92, 1.06
      Ptrend
      Determined using category medians.
      .19.83
      Saturated fat
       11,6801.00Referent1.00Referent
       21,7161.020.95, 1.091.010.95, 1.08
       31,6831.040.97, 1.111.040.97, 1.11
       41,6101.030.96, 1.101.020.95, 1.09
       51,4531.000.93, 1.070.980.91, 1.05
      Ptrend
      Determined using category medians.
      .95.64
      Trans-unsaturated fat
       11,5731.00Referent1.00Referent
       21,6741.030.96, 1.101.020.95, 1.09
       31,6341.000.93, 1.070.970.91, 1.04
       41,7241.081.01, 1.161.030.96, 1.11
       51,5371.071.00, 1.151.000.93, 1.07
      Ptrend
      Determined using category medians.
      .02.99
      Monounsaturated fat
       11,4921.00Referent1.00Referent
       21,6431.060.98, 1.131.050.98, 1.13
       31,6821.060.99, 1.141.050.98, 1.13
       41,7451.121.04, 1.201.101.02, 1.17
       51,5801.050.98, 1.131.000.93, 1.08
      Ptrend
      Determined using category medians.
      .07.67
      Polyunsaturated fat
       11,5051.00Referent1.00Referent
       21,6741.020.95, 1.101.010.94, 1.09
       31,7211.040.97, 1.121.020.95, 1.09
       41,6451.000.93, 1.070.960.89, 1.03
       51,5971.040.96, 1.110.980.92, 1.06
      Ptrend
      Determined using category medians.
      .58.31
      Long-chain omega-6 fatty acids
       11,4131.00Referent1.00Referent
       21,6451.020.95, 1.091.010.94, 1.08
       31,7271.020.95, 1.090.990.93, 1.07
       41,7130.990.93, 1.070.960.90, 1.04
       51,6441.000.93, 1.080.960.89, 1.03
      Ptrend
      Determined using category medians.
      .84.13
      Long-chain omega-3 fatty acids
       11,4241.00Referent1.00Referent
       21,6731.050.98, 1.131.040.97, 1.11
       31,6831.010.94, 1.080.980.91, 1.05
       41,6891.020.95, 1.100.990.92, 1.06
       51,6731.050.98, 1.131.010.94, 1.08
      Ptrend
      Determined using category medians.
      .37.81
      Eicosapentaenoic acid (EPA)
       11,2451.00Referent1.00Referent
       21,8351.020.95, 1.101.030.96, 1.10
       31,7231.091.02, 1.181.081.00, 1.16
       41,6521.070.99, 1.151.060.98, 1.14
       51,6871.050.98, 1.131.040.96, 1.12
      Ptrend
      Determined using category medians.
      .27.51
      Docosahexaenoic acid (DHA)
       11,4871.00Referent1.00Referent
       21,6311.020.95, 1.101.020.95, 1.09
       31,6691.081.01, 1.161.081.00, 1.15
       41,7391.050.89, 1.121.030.96, 1.11
       51,6161.101.02, 1.181.081.00, 1.16
      Ptrend
      Determined using category medians.
      .01.06
      Docasapentaenoic acid (DPA)
       11,4831.00Referent1.00Referent
       21,8820.930.87, 0.991.020.95, 1.10
       31,4530.910.85, 0.971.060.99, 1.14
       41,9700.940.88, 1.001.030.96, 1.11
       51,3540.850.80, 0.921.060.98, 1.14
      Ptrend
      Determined using category medians.
      .006.20
      Marine fatty acids (EPA, DHA, and DPA)
       11,5111.00Referent1.00Referent
       21,5821.010.94, 1.081.000.93, 1.08
       31,6411.060.99, 1.141.040.97, 1.12
       41,7561.091.02, 1.171.081.01, 1.16
       51,6521.060.99, 1.141.030.96, 1.11
      Ptrend
      Determined using category medians.
      .04.20
      Cholesterol
       11,5561.00Referent1.00Referent
       21,6571.030.96, 1.111.020.96, 1.10
       31,6771.050.98, 1.121.030.96, 1.10
       41,6681.081.00, 1.151.040.97, 1.11
       51,5841.101.03, 1.181.040.96, 1.11
      Ptrend
      Determined using category medians.
      .003.34
      Note: CI = confidence interval; HR = hazard ratio; MV = Multivariable.
      a Adjusted for age (continuous), total calories (continuous), race/ethnicity (white, black, Hispanic, Asian, other), age at menarche (y; <11, 11, 12, 13, 14–15, >15), infertility (yes, no), parity (nulliparous, 1, 2, 3, 4+), age at first birth (y; <25, 25–30, >30), time since last birth (y; <1, 1–3, 4–5, 6–7, 8–9, 10–12, 13–15, 16+), age first oral contraceptive use (y; 13–16, 17–20, 21–24, 25+), menstrual cycle length (d; <26, 26–31, 32–50, >50), body mass index (<20, 20–21.9, 22–23.9, 24–24.9, 25–26.9, 27–29.9, 30+), smoking (never, past, current), recent gynecologic/breast exam (no recent exam, recent exam), and use of antihypertensive medications/diastolic blood pressure (no meds <65, no meds 65–74, no meds 75–84, no meds 85–89, no meds 90+, meds <65, meds 65–74, meds 75–84, meds 85–89, meds 90+).
      b Determined using category medians.
      Characteristics of the subset of women who had erythrocyte FA levels measured are presented in Table 3 by fibroids status. The median time between blood draw and fibroids diagnosis was 5.5 years, with an interquartile range of 3.3–9.3 years. Results were similar between age-adjusted (data not shown) and covariate-adjusted models with results for covariate-adjusted models presented in Table 4. We observed an inverse association between erythrocyte levels of total n-3 PUFA and odds of fibroids (OR [95% CI] for third vs. first tertile = 0.41 [0.19–0.89]; Ptrend = .02). Of the individual n-3 PUFAs examined, the inverse association was strongest for EPA (OR [95% CI] for third tertile = 0.42 [0.19–0.90]; Ptrend = .03).
      Table 3Characteristics at blood collection (1996-1999) of uterine fibroids in 56 cases and 497 controls in the Nurses’ Health Study II, 1996-2009.
      CharacteristicControlsCases
      (n = 497)(n = 56)
      Age at blood draw, y (range), mean (SD)43 (33–52)(4.0)43 (33–50)(3.5)
      White471(94.8)54(96.4)
      Parity
       Nulliparous97(19.5)8(14.3)
       165(13.1)5(8.9)
       2177(35.5)20(35.7)
       3113(22.7)18(32.1)
       ≥445(9.0)5(8.9)
      Age at first birth, y
       <25141(35.3)13(27.1)
       25–30187(46.8)31(64.6)
       ≥3172(18.0)4(8.3)
      Time since last birth, y
       <6102(25.5)14(29.2)
       6–12181(45.3)18(37.5)
       ≥13117(29.3)16(33.3)
      Age at menarche, y
       <1296(19.3)12(21.4)
       12–13315(63.3)35(62.5)
       ≥1486(17.3)9(16.1)
      Menstrual cycle length, d
       <2693(19.2)9(17.0)
       26–31348(71.9)35(66.0)
       ≥3243(8.9)9(17.0)
      BMI (kg/m2), mean (SD)25.2(6.0)26.2(6.4)
      Note: Data presented as n (%), unless stated otherwise. BMI = body mass index; SD = standard deviation.
      Table 4Multivariable-adjusted odds ratio (95% confidence interval) of fibroids according to tertiles of erythrocyte fatty acid concentration among 553 participants (56 fibroids cases, 497 noncases), Nurses' Health Study II.
      CharacteristicTertile 1Tertile 2Tertile 3Ptrend
      Determined using category medians.
      Pcontinuous
      Determined using a continuous variable.
      Total saturated fatty acids1.00 (Referent)0.94 (0.45–1.95)1.12 (0.55–2.27).70.51
       Lauric acid (12:0)1.00 (Referent)0.98 (0.48–2.00)0.92 (0.45–1.87).81.66
       Mystristic acid (14:0)1.00 (Referent)0.83 (0.40–1.74)1.05 (0.52–2.10).82.88
       Pentadecanoic acid (15:0)1.00 (Referent)0.62 (0.29–1.32)1.07 (0.54–2.15).78.39
       Palmitic acid (16:0)1.00 (Referent)1.15 (0.58–2.30)0.92 (0.44–1.93).83.68
       Margric acid (17:0)1.00 (Referent)1.60 (0.75–3.43)1.85 (0.85–4.03).15.07
       Stearic acid (18:0)1.00 (Referent)1.02 (0.49–2.12)1.22 (0.59–2.49).56.08
       Nonadecanoic acid (19:0)1.00 (Referent)1.27 (0.57–2.81)2.08 (0.97–4.46).05.19
       Arachidic acid (20:0)1.00 (Referent)1.03 (0.50–2.15)1.17 (0.57–2.40).66.46
       Behenic acid (22:0)1.00 (Referent)0.94 (0.46–1.92)1.01 (0.49–2.07).96.55
       Tricosanoic acid (23:0)1.00 (Referent)0.96 (0.48–1.90)0.63 (0.29–1.35).22.50
       Lignoceric acid (24:0)1.00 (Referent)0.59 (0.29–1.19)0.68 (0.34–1.36).26.21
      Total monounsaturated fatty acids1.00 (Referent)0.78 (0.39–1.57)0.64 (0.30–1.36).24.12
       Mysristoleic acid (14:1n-5c)1.00 (Referent)0.96 (0.46–2.02)1.22 (0.60–2.47).53.54
       Pentadecenoic acid (15:1n-5c)1.00 (Referent)0.83 (0.42–1.64)0.55 (0.25–1.23).15.06
       Palmitoleic acid (16:1n-7c)1.00 (Referent)0.92 (0.47–1.80)0.48 (0.21–1.10).08.05
       Oleic acid (18:1n-9c)1.00 (Referent)0.39 (0.18–0.84)0.60 (0.30–1.23).11.46
       Octadecenoic acid (18:1n-7c)1.00 (Referent)2.24 (1.09–4.63)1.19 (0.53–2.71).77.94
       Gondoic acid (20:1n-9c)1.00 (Referent)2.03 (0.99–4.18)0.96 (0.43–2.15).88.58
       Nervonic acid (24:1n-9c)1.00 (Referent)1.01 (0.52–1.99)0.65 (0.30–1.38).28.18
      n-3 polyunsaturated fatty acids1.00 (Referent)0.49 (0.24–0.97)0.41 (0.19–0.89).02.02
       Alpha-linolenic acid (18:3n-3c)1.00 (Referent)1.69 (0.84–3.40)0.75 (0.34–1.69).40.74
       Eicosapentaenoic acid (20:5n-3c)1.00 (Referent)0.64 (0.32–1.25)0.42 (0.19–0.90).03.01
       Docosapentaenoic acid (22:5n-3c)1.00 (Referent)0.60 (0.29–1.22)0.73 (0.36–1.48).34.11
       Docosahexaenoic acid (22:6n-3c)1.00 (Referent)0.40 (0.19–0.85)0.57 (0.28–1.15).09.03
      n-6 polyunsaturated fatty acids1.00 (Referent)0.93 (0.44–1.92)1.14 (0.57–2.29).74.81
       Linoleic acid (18:2n-6cc)1.00 (Referent)0.94 (0.45–1.93)1.09 (0.53–2.23).81.58
       Gamma-linolenic acid (18:3n-6c)1.00 (Referent)0.85 (0.42–1.73)0.83 (0.40–1.75).65.29
       Eicosadienoic acid (20:2n-6c)1.00 (Referent)1.05 (0.51–2.13)1.04 (0.51–2.15).92.95
       Dihomogammalinolenic acid (20:3n-6c)1.00 (Referent)0.60 (0.29–1.23)0.70 (0.34–1.42).33.11
       Arachidonic acid (20:4n-6c)1.00 (Referent)0.62 (0.30–1.29)0.90 (0.45–1.80).76.79
       Docosadienoic acid (22:2n-6c)1.00 (Referent)0.85 (0.38–1.89)1.86 (0.90–3.84).05.05
       Aolrenic acid (22:4n-6c)1.00 (Referent)1.12 (0.55–2.29)1.03 (0.50–2.12).93.42
      Total trans fatty acids1.00 (Referent)1.23 (0.52–2.92)3.33 (1.50–7.38).002.004
       Palmitelaidic acid (16:1n-7t)1.00 (Referent)0.90 (0.42–1.92)1.42 (0.68–2.99).30.10
       Linolelaidic acid (18:2n-6t)1.00 (Referent)1.87 (0.81–4.29)2.87 (1.30–6.34).008.04
       Octadecadienoic acid (18:2n-7c)1.00 (Referent)0.86 (0.41–1.81)1.26 (0.63–2.52).48.59
       18:1 trans1.00 (Referent)1.14 (0.48–2.71)3.10 (1.42–6.77).003.002
       18:2 trans1.00 (Referent)1.31 (0.61–2.81)1.64 (0.78–3.46).19.34
      Dairy-derived fatty acids1.00 (Referent)1.26 (0.58–2.72)1.68 (0.79–3.59).17.10
      Industrial trans1.00 (Referent)1.08 (0.46–2.56)3.06 (1.41–6.66).002.003
      Total n-6/n-3 ratio1.00 (Referent)1.23 (0.55–2.74)1.94 (0.91–4.16).08.03
      SI ration-71.00 (Referent)2.03 (0.90–4.56)2.10 (0.90–4.86).10.05
      SI ration-91.00 (Referent)1.19 (0.57–2.46)1.21 (0.58–2.52).62.09
      Note: Adjusted for age (continuous), blood draw time (1am-8am, 9am-12pm, 1pm-midnight), fasting status (yes/no), blood draw season (Nov–Apr, May–Oct), race (white, nonwhite), parity (nulliparous, 1, 2, 3, 4+), age at first birth (y; <25, 25-30, 31+), time since last birth (y; <6, 6–12, 13+), age at menarche (y; <12, 12–13, 14+), menstrual cycle length (d; <26, 26–31, 32+), and body mass index (<20, 20–21.9, 22–23.9, 24–24.9, 25–26.9, 27–29.9, 30+). SI = saturation index.
      a Determined using category medians.
      b Determined using a continuous variable.
      Total trans FA was associated with more odds of fibroids (OR [95% CI] for third tertile = 3.33 [1.50–7.38]; Ptrend = .002) with the largest magnitude of individual trans FA associations with linolelaidic acid and 18:1 trans (OR [95% CI] = 2.87 [1.30–6.34]; Ptrend = .008 and OR [95% CI] = 3.10 [1.42–6.77]; Ptrend = .003, respectively). A significant positive association was also observed with industrial trans FAs (Table 4). No associations were observed for total or individual SFA, MUFA, or n-6 PUFA.

      Discussion

      In this large, prospective cohort study we observed no associations between intake of dietary fats (total or specific types) and risk of uterine fibroids. However, in a subset of women with erythrocyte FA measurements, we observed a lower odd of fibroids among women with higher n-3 PUFA erythrocyte levels and a greater odds among those with higher trans FA erythrocyte levels. These findings suggest n-3 PUFAs and trans FAs may be associated with fibroid risk.
      Dietary fat intake may influence the etiology of fibroids through estrogenic or inflammatory effects. A meta-analysis (
      • Wu A.H.
      • Pike M.C.
      • Stram D.O.
      Meta-analysis: dietary fat intake, serum estrogen levels, and the risk of breast cancer.
      ) of 13 intervention studies reported that reducing fat consumption among both premenopausal and postmenopausal women resulted in lower serum estradiol levels. More recently (
      • Wegienka G.
      Are uterine leiomyoma a consequence of a chronically inflammatory immune system?.
      ) it has been hypothesized that systemic chronic inflammation, marked by increased T-helper cytokines and decreased functional regulatory T cells, may lead to the development of fibroids through the formation of fibrous tissue and smooth muscle proliferation. In addition, women with fibroids may be at increased risk of atherosclerosis and hypertension (
      • Boynton-Jarrett R.
      • Rich-Edwards J.
      • Malspeis S.
      • Missmer S.A.
      • Wright R.
      A prospective study of hypertension and risk of uterine leiomyomata.
      ,
      • Faerstein E.
      • Szklo M.
      • Rosenshein N.B.
      Risk factors for uterine leiomyoma: a practice-based case-control study. II. atherogenic risk factors and potential sources of uterine irritation.
      ,
      • He Y.
      • Zeng Q.
      • Dong S.
      • Qin L.
      • Li G.
      • Wang P.
      Associations between uterine fibroids and lifestyles including diet, physical activity and stress: a case-control study in China.
      ), and inflammation has been identified as playing an important role in atherosclerosis development (
      • Ross R.
      Atherosclerosis—an inflammatory disease.
      ). Trans fat intake influences circulating levels of IL-6, IL-1β, tumor necrosis factor-α (TNF-α) and other inflammatory markers (
      • Baer D.J.
      • Judd J.T.
      • Clevidence B.A.
      • Tracy R.P.
      Dietary fatty acids affect plasma markers of inflammation in healthy men fed controlled diets: a randomized crossover study.
      ,
      • Mozaffarian D.
      • Pischon T.
      • Hankinson S.E.
      • Rifai N.
      • Joshipura K.
      • Willett W.C.
      • et al.
      Dietary intake of trans fatty acids and systemic inflammation in women.
      ,
      • Mozaffarian D.
      • Rimm E.B.
      • King I.B.
      • Lawler R.L.
      • McDonald G.B.
      • Levy W.C.
      Trans fatty acids and systemic inflammation in heart failure.
      ), and markers including IL-6, IL-1, and TNF-α have been reported to influence the secretion of enzymes that digest endometrial extracellular matrices (
      • Bischof P.
      Endocrine, paracrine, and autocrine regulation of trophoblastic metalloproteinases.
      ). In contrast, dietary intake and plasma levels of omega-3 FAs has been inversely associated with inflammatory cytokines, including IL-6, TNF-α, and TNF-α receptors (
      • Ferrucci L.
      • Cherubini A.
      • Bandinelli S.
      • Bartali B.
      • Corsi A.
      • Lauretani F.
      • et al.
      Relationship of plasma polyunsaturated fatty acids to circulating inflammatory markers.
      ,
      • Pischon T.
      • Hankinson S.E.
      • Hotamisligil G.S.
      • Rifai N.
      • Willett W.C.
      • Rimm E.B.
      Habitual dietary intake of n-3 and n-6 fatty acids in relation to inflammatory markers among us men and women.
      ). Our results among women with erythrocyte FA levels that indicated a decreased risk with higher n-3 PUFAs concentrations and an increased risk with higher trans FAs concentrations are consistent with these observations.
      To our knowledge only three studies, two case-control, (
      • Chiaffarino F.
      • Parazzini F.
      • la Vecchia C.
      • Chatenoud L.
      • di Cintio E.
      • Marsico S.
      Diet and uterine myomas.
      ,
      • Nagata C.
      • Nakamura K.
      • Oba S.
      • Hayashi M.
      • Takeda N.
      • Yasuda K.
      Association of intakes of fat, dietary fibre, soya isoflavones and alcohol with uterine fibroids in Japanese women.
      ) and one prospective cohort study, (
      • Wise L.A.
      • Radin R.G.
      • Kumanyika S.K.
      • Ruiz-Narváez E.A.
      • Palmer J.R.
      • Rosenberg L.
      Prospective study of dietary fat and risk of uterine leiomyomata.
      ) have examined the association between dietary fat intake and fibroid risk. An Italian hospital-based case-control study (
      • Chiaffarino F.
      • Parazzini F.
      • la Vecchia C.
      • Chatenoud L.
      • di Cintio E.
      • Marsico S.
      Diet and uterine myomas.
      ) of 843 histologically confirmed fibroid cases and 1,557 controls with acute non-gynecologic, non-hormonal, and non-neoplastic conditions (e.g., traumatic injury, nontraumatic orthopedic disorders, surgical conditions, eye disorders) reported no associations between butter, margarine, or oil intake in the year before the study and fibroids risk. That study was not able to adjust for total caloric intake and categorization based on tertiles (i.e., low, intermediate, and high) were used for the three sources of fat intake. Consistent with these null results, a cross-sectional study (
      • Nagata C.
      • Nakamura K.
      • Oba S.
      • Hayashi M.
      • Takeda N.
      • Yasuda K.
      Association of intakes of fat, dietary fibre, soya isoflavones and alcohol with uterine fibroids in Japanese women.
      ) of Japanese women enrolled through a health check-up program (54 fibroids cases and 234 without fibroids) reported no association between total fat or specific subtypes (i.e., SFA, MUFA, PUFA) assessed with a 169-item FFQ and fibroids identified through transvaginal ultrasound.
      Most recently (
      • Wise L.A.
      • Radin R.G.
      • Kumanyika S.K.
      • Ruiz-Narváez E.A.
      • Palmer J.R.
      • Rosenberg L.
      Prospective study of dietary fat and risk of uterine leiomyomata.
      ), the prospective BWHS followed >12,000 African American women for 8 years identifying 2,695 fibroid cases diagnosed by ultrasound, hysterectomy, or surgery. In this analysis, Wise (
      • Wise L.A.
      • Radin R.G.
      • Kumanyika S.K.
      • Ruiz-Narváez E.A.
      • Palmer J.R.
      • Rosenberg L.
      Prospective study of dietary fat and risk of uterine leiomyomata.
      ) et al. reported increased risks of fibroids with intake of specific n-3 PUFAs but no consistent associations with total fat or other fat subtypes. These investigators further reported a greater risk with dark meat fish, which was the main source of n-3 PUFAs in this population. These results are intriguing given that long-chain omega-3 FAs exhibit anti-inflammatory properties (
      • Ferrucci L.
      • Cherubini A.
      • Bandinelli S.
      • Bartali B.
      • Corsi A.
      • Lauretani F.
      • et al.
      Relationship of plasma polyunsaturated fatty acids to circulating inflammatory markers.
      ,
      • Pischon T.
      • Hankinson S.E.
      • Hotamisligil G.S.
      • Rifai N.
      • Willett W.C.
      • Rimm E.B.
      Habitual dietary intake of n-3 and n-6 fatty acids in relation to inflammatory markers among us men and women.
      ) and have previously been associated with positive health benefits such as reduced risk of coronary heart disease (
      • Mozaffarian D.
      • Rimm E.B.
      Fish intake, contaminants, and human health: evaluating the risks and the benefits.
      ) and endometriosis (
      • Missmer S.A.
      • Chavarro J.E.
      • Malspeis S.
      • Bertone-Johnson E.R.
      • Hornstein M.D.
      • Spiegelman D.
      • et al.
      A prospective study of dietary fat consumption and endometriosis risk.
      ). However, it is consistent with a modest increased risk of prostate cancer observed with higher EPA and DPA blood levels (
      • Crowe F.L.
      • Appleby P.N.
      • Travis R.C.
      • Barnett M.
      • Brasky T.M.
      • Bueno-de-Mesquita H.B.
      • et al.
      Circulating fatty acids and prostate cancer risk: individual participant meta-analysis of prospective studies.
      ). In addition, consumption of fish that contain persistent organic pollutants has been previously associated with fibroids incidence (
      • Lambertino A.
      • Turyk M.
      • Anderson H.
      • Freels S.
      • Persky V.
      Uterine leiomyomata in a cohort of Great Lakes sport fish consumers.
      ). In our FFQ-based analysis among a predominantly white population we observed no significant associations between any dietary fats, including long-chain omega-3 FAs, or dark meat fish, and risk of uterine fibroids. Although in the NHSII and BWHS dark meat fish was the largest contributor of n-3 PUFAs. Differences in types of dark meat fish consumed, other sources of n-3 PUFAs, and/or differing exposures to environmental contaminants through these sources could have contributed to the disparate results.
      To our knowledge this is the only study to have examined the association between erythrocyte FAs and fibroid risk. The differing results observed between dietary FAs assessed with FFQs compared with erythrocyte FA levels in this analysis deserves further discussion. Circulating FAs originate from many sources, including dietary intake as well as being synthesized and/or transformed in vivo. The latter of these sources cannot be assessed by dietary intake. Thus, the FA composition of the erythrocyte membrane likely represents an integrated measure of the interactions between dietary FA intake, other dietary factors, and internal transformation of FAs, which may explain the association observed with the erythrocyte FAs and not with the dietary fat intake assessed with FFQs. In a similar population in the Nurses’ Health Study, moderate-to-strong correlations were observed between erythrocyte FAs and FA intake assessed by the cumulative average method from three FFQs with adjusted Spearman correlation coefficients of 0.54 (P<.01) for the n-3 PUFA DHA and 0.48 (P<.01) for trans FAs (
      • Sun Q.
      • Ma J.
      • Campos H.
      • Hankinson S.E.
      • Hu F.B.
      Comparison between plasma and erythrocyte fatty acid content as biomarkers of fatty acid intake in US women.
      ), indicating that erythrocyte concentrations of these specific FAs, which are largely of exogenous origin, are suitable biomarkers for long-term FA dietary intake.
      At present this is the largest study to examine the association between dietary fat intake and fibroid risk with 18 years of follow-up and multiple dietary assessments. Although the FFQ has been previously validated (
      • Salvini S.
      • Hunter D.J.
      • Sampson L.
      • Stampfer M.J.
      • Coldtiz G.A.
      • Rosner B.
      • et al.
      Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption.
      ,
      • Willett W.C.
      • Sampson L.
      • Stampfer M.J.
      • Rosner B.
      • Bain C.
      • Witschi J.
      • et al.
      Reproducibility and validity of a semiquantitative food frequency questionnaire.
      ,
      • Yuan C.
      • Spiegelman D.
      • Rimm E.B.
      • Rosner B.A.
      • Stampfer M.J.
      • Barnett J.B.
      • et al.
      Validity of a dietary questionnaire assessed by comparison with multiple weighed dietary records or 24-hour recalls.
      ), some error in self-report is expected. Thus our assessment of erythrocyte membrane FA levels on a subset of women complemented the FFQ data, allowing us to consider dietary intake and endogenous synthesis and transformation of FAs providing new insight into the potential association between dietary fats and fibroids risk. The use of erythrocyte measures of FAs instead of plasma was a further strength as erythrocytes are likely to represent long-term intake better than plasma due to their longer half-life (
      • Katan M.B.
      • Deslypere J.P.
      • van Birgelen A.P.
      • Penders M.
      • Zegwaard M.
      Kinetics of the incorporation of dietary fatty acids into serum cholesteryl esters, erythrocyte membranes, and adipose tissue: an 18-month controlled study.
      ).
      Fibroid diagnosis was collected through self-report, which includes the potential for outcome misclassification. To address this, we restricted our case definition to those reporting a diagnosis that was confirmed with ultrasound or hysterectomy. Based on the results from the previous validation study conducted in this cohort, we are confident that women reporting a fibroid diagnosis have been diagnosed with fibroids (
      • Marshall L.M.
      • Spiegelman D.
      • Barbieri R.L.
      • Goldman M.B.
      • Manson J.E.
      • Colditz G.A.
      • et al.
      Variation in the incidence of uterine leiomyoma among premenopausal women by age and race.
      ). However, we cannot quantify how many women with undiagnosed fibroids are present in this cohort. Baird et al. (
      • Baird D.D.
      • Dunson D.B.
      • Hill M.C.
      • Cousins D.
      • Schectman J.M.
      High cumulative incidence of uterine leiomyoma in black and white women: ultrasound evidence.
      ) reported that 43% of white women of reproductive age showed ultrasound evidence of having an undiagnosed fibroid. The presence of undiagnosed disease usually leads to a bias to the null (i.e., it reduces the probability of observing true associations rather than generating false-positive associations). However, it is difficult to design studies of fibroids to prevent this phenomenon. Even among studies in which controls or women without fibroids have an ultrasound to confirm that they do not have fibroids, the time of fibroid onset among the cases is unknown and exposure status before fibroid onset is difficult to determine. In addition, these cross-sectional studies do not have ability to prospectively collect dietary data for many years or blood samples before fibroid diagnosis as is possible with our study design. We observed statistically significant results despite this potential misclassification, suggesting that the associations observed with erythrocyte FA levels may be stronger than we observed. In regard to generalizability, although our study participants were limited to US female registered nurses at study enrollment, there is no strong rationale as to why the association between erythrocyte FA levels and fibroid/fibroids incidence would differ in this population compared with women in the general population. However, as the NHSII is a predominantly white cohort, we had insufficient numbers to examine these associations by race. This is an important limitation as black women are disproportionately impacted by fibroids and some risk factors for fibroids could differ between racial and ethnic groups.
      Finally, we must acknowledge limitations of the erythrocyte analyses. With the limited sample size and multiple FAs examined in our biomarker analyses we may have observed significant associations due to chance. If we account for multiple comparison using a Bonferroni correction none of the observed associations would be statistically significant, thus our results should be interpreted with caution. Furthermore, it is currently not clear whether our erythrocyte FA measures capture the relevant etiologic window for fibroid development. However, erythrocytes likely reflect longer term dietary intake given their half-life of 120 days compared with serum measures that reflect intake during only a few days.
      In conclusion, our prospective analysis suggests that n-3 PUFAs and trans FAs may play a role in the incidence of clinically relevant uterine fibroids. As these factors are influenced by dietary intake, increasing long-chain omega-3 FA intake and decreasing trans FA intake should be further examined as potentially modifiable risk factors in the etiology of uterine fibroids.

      References

        • Cramer S.
        • Patel A.
        The frequency of uterine leiomyomas.
        Am J Clin Pathol. 1990; 94: 435-438
        • Velebil P.
        • Wingo P.A.
        • Xia Z.
        • Wilcox L.S.
        • Peterson H.B.
        Rate of hospitalization for gynecologic disorders among reproductive-age women in the United States.
        Obstet Gynecol. 1995; 86: 764-769
        • Baer D.J.
        • Judd J.T.
        • Clevidence B.A.
        • Tracy R.P.
        Dietary fatty acids affect plasma markers of inflammation in healthy men fed controlled diets: a randomized crossover study.
        Am J Clin Nutr. 2004; 79: 969-973
        • Mozaffarian D.
        • Pischon T.
        • Hankinson S.E.
        • Rifai N.
        • Joshipura K.
        • Willett W.C.
        • et al.
        Dietary intake of trans fatty acids and systemic inflammation in women.
        Am J Clin Nutr. 2004; 79: 606-612
        • Mozaffarian D.
        • Rimm E.B.
        • King I.B.
        • Lawler R.L.
        • McDonald G.B.
        • Levy W.C.
        Trans fatty acids and systemic inflammation in heart failure.
        Am J Clin Nutr. 2004; 80: 1521-1525
        • Wegienka G.
        Are uterine leiomyoma a consequence of a chronically inflammatory immune system?.
        Med Hypotheses. 2012; 79: 226-231
        • Wise L.A.
        • Radin R.G.
        • Kumanyika S.K.
        • Ruiz-Narváez E.A.
        • Palmer J.R.
        • Rosenberg L.
        Prospective study of dietary fat and risk of uterine leiomyomata.
        Am J Clin Nutr. 2014; 99: 1105-1116
        • Arab L.
        Biomarkers of fat and fatty acid intake.
        J Nutr. 2003; 133: 925s-932s
        • Solomon C.G.
        • Willett W.C.
        • Carey V.J.
        • Rich-Edwards J.
        • Hunter D.J.
        • Colditz G.A.
        • et al.
        A prospective study of pregravid determinants of gestational diabetes mellitus.
        JAMA. 1997; 278: 1078-1083
        • Tworoger S.S.
        • Sluss P.
        • Hankinson S.E.
        Association between plasma prolactin concentrations and risk of breast cancer among predominately premenopausal women.
        Cancer Res. 2006; 66: 2476-2482
        • Hirko K.A.
        • Chai B.
        • Spiegelman D.
        • Campos H.
        • Farvid M.S.
        • Hankinson S.E.
        • et al.
        Erythrocyte membrane fatty acids and breast cancer risk: a prospective analysis in the nurses' health study II.
        Intern J Cancer. 2018; 142: 1116-1129
        • Dial S.
        The Antioxidant Vitamins C and E.
        in: Tocopherols and tocotrienols in key foods in the US diet. The American Oil Chemists' Society Press, Champaign, IL1995: 327-342
        • Holland G.
        • Unwin I.
        • Buss D.
        • Paul A.
        • Dat S.
        The composition of foods.
        in: The Royal Society of Chemistry and Ministry of Agriculture, Fish and Foods. 1991 (Cambridge, UK)
        • USDA
        USDA Nutrient Database for Standard Reference.
        1993
        • Salvini S.
        • Hunter D.J.
        • Sampson L.
        • Stampfer M.J.
        • Coldtiz G.A.
        • Rosner B.
        • et al.
        Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption.
        Int J Epidemiol. 1989; 18: 858-867
        • Willett W.C.
        • Sampson L.
        • Stampfer M.J.
        • Rosner B.
        • Bain C.
        • Witschi J.
        • et al.
        Reproducibility and validity of a semiquantitative food frequency questionnaire.
        Am J Epidemiol. 1985; 122: 51-65
        • Yuan C.
        • Spiegelman D.
        • Rimm E.B.
        • Rosner B.A.
        • Stampfer M.J.
        • Barnett J.B.
        • et al.
        Validity of a dietary questionnaire assessed by comparison with multiple weighed dietary records or 24-hour recalls.
        Am J Epidemiol. 2017; 185: 570-584
        • Baylin A.
        • Kim M.K.
        • Donovan-Palmer A.
        • Siles X.
        • Dougherty L.
        • Tocco P.
        • et al.
        Fasting whole blood as a biomarker of essential fatty acid intake in epidemiologic studies: comparison with adipose tissue and plasma.
        Am J Epidemiol. 2005; 162: 373-381
        • Simopoulos A.P.
        The importance of the omega-6/omega-3 fatty acid ratio in cardiovascular disease and other chronic diseases.
        Exp Biol Med. 2008; 233: 674-688
        • Chajes V.
        • Joulin V.
        • Clavel-Chapelon F.
        The fatty acid desaturation index of blood lipids, as a biomarker of hepatic stearoyl-CoA desaturase expression, is a predictive factor of breast cancer risk.
        Curr Opin Lipidol. 2011; 22: 6-10
        • Patel P.S.
        • Sharp S.J.
        • Jansen E.
        • Luben R.N.
        • Khaw K.T.
        • Wareham N.J.
        • et al.
        Fatty acids measured in plasma and erythrocyte-membrane phospholipids and derived by food-frequency questionnaire and the risk of new-onset type 2 diabetes: a pilot study in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Norfolk cohort.
        Am J Clin Nutr. 2010; 92: 1214-1222
        • Sun Q.
        • Ma J.
        • Campos H.
        • Hankinson S.E.
        • Hu F.B.
        Comparison between plasma and erythrocyte fatty acid content as biomarkers of fatty acid intake in US women.
        Am J Clin Nutr. 2007; 86: 74-81
        • Marshall L.M.
        • Spiegelman D.
        • Barbieri R.L.
        • Goldman M.B.
        • Manson J.E.
        • Colditz G.A.
        • et al.
        Variation in the incidence of uterine leiomyoma among premenopausal women by age and race.
        Obstet Gynecol. 1997; 90: 967-973
        • Lambertino A.
        • Turyk M.
        • Anderson H.
        • Freels S.
        • Persky V.
        Uterine leiomyomata in a cohort of Great Lakes sport fish consumers.
        Env Res. 2011; 111: 565-572
        • Hu F.B.
        • Stampfer M.J.
        • Rimm E.
        • Ascherio A.
        • Rosner B.A.
        • Spiegelman D.
        • et al.
        Dietary fat and coronary heart disease: a comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements.
        Am J Epidemiol. 1999; 149: 531-540
        • Wu A.H.
        • Pike M.C.
        • Stram D.O.
        Meta-analysis: dietary fat intake, serum estrogen levels, and the risk of breast cancer.
        J Natl Cancer Inst. 1999; 91: 529-534
        • Boynton-Jarrett R.
        • Rich-Edwards J.
        • Malspeis S.
        • Missmer S.A.
        • Wright R.
        A prospective study of hypertension and risk of uterine leiomyomata.
        Am J Epidemiol. 2005; 161: 628-638
        • Faerstein E.
        • Szklo M.
        • Rosenshein N.B.
        Risk factors for uterine leiomyoma: a practice-based case-control study. II. atherogenic risk factors and potential sources of uterine irritation.
        Am J Epidemiol. 2001; 153: 11-19
        • He Y.
        • Zeng Q.
        • Dong S.
        • Qin L.
        • Li G.
        • Wang P.
        Associations between uterine fibroids and lifestyles including diet, physical activity and stress: a case-control study in China.
        Asia Pac J Clin Nutr. 2013; 22: 109-117
        • Ross R.
        Atherosclerosis—an inflammatory disease.
        N Engl J Med. 1999; 340: 115-126
        • Bischof P.
        Endocrine, paracrine, and autocrine regulation of trophoblastic metalloproteinases.
        Early Pregnancy. 2001; 5: 30-31
        • Ferrucci L.
        • Cherubini A.
        • Bandinelli S.
        • Bartali B.
        • Corsi A.
        • Lauretani F.
        • et al.
        Relationship of plasma polyunsaturated fatty acids to circulating inflammatory markers.
        J Clin Endocrinol Metab. 2006; 91: 439-446
        • Pischon T.
        • Hankinson S.E.
        • Hotamisligil G.S.
        • Rifai N.
        • Willett W.C.
        • Rimm E.B.
        Habitual dietary intake of n-3 and n-6 fatty acids in relation to inflammatory markers among us men and women.
        Circulation. 2003; 108: 155-160
        • Chiaffarino F.
        • Parazzini F.
        • la Vecchia C.
        • Chatenoud L.
        • di Cintio E.
        • Marsico S.
        Diet and uterine myomas.
        Obstet Gynecol. 1999; 94: 395-398
        • Nagata C.
        • Nakamura K.
        • Oba S.
        • Hayashi M.
        • Takeda N.
        • Yasuda K.
        Association of intakes of fat, dietary fibre, soya isoflavones and alcohol with uterine fibroids in Japanese women.
        Br J Nutr. 2009; 101: 1427-1431
        • Mozaffarian D.
        • Rimm E.B.
        Fish intake, contaminants, and human health: evaluating the risks and the benefits.
        JAMA. 2006; 296: 1885-1899
        • Missmer S.A.
        • Chavarro J.E.
        • Malspeis S.
        • Bertone-Johnson E.R.
        • Hornstein M.D.
        • Spiegelman D.
        • et al.
        A prospective study of dietary fat consumption and endometriosis risk.
        Hum Reprod. 2010; 25: 1528-1535
        • Crowe F.L.
        • Appleby P.N.
        • Travis R.C.
        • Barnett M.
        • Brasky T.M.
        • Bueno-de-Mesquita H.B.
        • et al.
        Circulating fatty acids and prostate cancer risk: individual participant meta-analysis of prospective studies.
        J Natl Cancer Inst. 2014; 106: 1-10
        • Katan M.B.
        • Deslypere J.P.
        • van Birgelen A.P.
        • Penders M.
        • Zegwaard M.
        Kinetics of the incorporation of dietary fatty acids into serum cholesteryl esters, erythrocyte membranes, and adipose tissue: an 18-month controlled study.
        J Lipid Res. 1997; 38: 2012-2022
        • Baird D.D.
        • Dunson D.B.
        • Hill M.C.
        • Cousins D.
        • Schectman J.M.
        High cumulative incidence of uterine leiomyoma in black and white women: ultrasound evidence.
        Am J Obstet Gynecol. 2003; 188: 100-107