Advertisement

Risk factors for uterine fibroids: time to build on what we have learned

Published:September 06, 2020DOI:https://doi.org/10.1016/j.fertnstert.2020.07.059
      Uterine leiomyomata, or fibroids, are common, and symptomatic fibroids can severely reduce quality of life (
      • Wise L.A.
      • Laughlin-Tommaso S.K.
      Epidemiology of uterine fibroids: from menarche to menopause.
      ). Current medical and surgical treatment options are expanding but are frequently transient and may conflict with fertility plans. On this background of high prevalence, high symptom burden, and treatment options with potential adverse effects, the search for modifiable exposures to reduce fibroid prevalence is important. The recent investigation by Harris et al. (
      • Harris H.R.
      • Eliassen A.H.
      • Doody D.R.
      • Terry K.L.
      • Misser S.A.
      Dietary fat intake, erythrocyte fatty acids, and risk of uterine fibroids.
      ), adds to the growing literature on the association of diet and fibroid risk.
      Inflammation may play a role in the cause of fibroids (
      • El Andaloussi A.
      • Chaudhry Z.
      • Al-Hendy A.
      • Ismail N.
      Uterine fibroids: bridging genomic defects and chronic inflammation.
      ), and dietary components may enhance or mitigate chronic inflammation. Dietary fats include both pro- and anti-inflammatory fatty acids, creating a complex exposure. Leveraging up to 18 years of follow-up of premenopausal participants in the Nurses’ Health Study II, Harris et al. (
      • Harris H.R.
      • Eliassen A.H.
      • Doody D.R.
      • Terry K.L.
      • Misser S.A.
      Dietary fat intake, erythrocyte fatty acids, and risk of uterine fibroids.
      ) investigate the associations of self-reported current dietary fat intake and blood biomarkers of fatty acid intake measured at baseline, with the risk of fibroids. They find little evidence of associations with dietary intake of fat. However, they report suggestive inverse associations with erythrocyte fatty acid (EFA) measures of some n-3 polyunsaturated fatty acids, which are anti-inflammatory, and increased risk with total trans fatty acids, which are pro-inflammatory (
      • Harris H.R.
      • Eliassen A.H.
      • Doody D.R.
      • Terry K.L.
      • Misser S.A.
      Dietary fat intake, erythrocyte fatty acids, and risk of uterine fibroids.
      ). The inclusion of blood biomarkers is innovative and provides some advantages over dietary measures: they integrate diet, supplements, and metabolic sources of fatty acids, and they are a more objective and less error-prone measure than dietary self-report.
      Although the analysis uses data from a well-designed prospective study, there are limitations for fibroid research. To address some of the uncertainty with self-reported diagnosis of fibroids, Harris et al. (
      • Harris H.R.
      • Eliassen A.H.
      • Doody D.R.
      • Terry K.L.
      • Misser S.A.
      Dietary fat intake, erythrocyte fatty acids, and risk of uterine fibroids.
      ) limit the study’s outcome to self-reported fibroids detected surgically or by ultrasonography. However, to the extent that the dietary factors of interest could affect the presence of symptoms or be correlated with health care access, the analysis could suffer from bias. In addition, reliance on clinically detected fibroids misses the many undiagnosed fibroids, and it does not adequately address the issue of temporality because clinical detection of fibroids often comes years after tumor development (
      • Wise L.A.
      • Laughlin-Tommaso S.K.
      Epidemiology of uterine fibroids: from menarche to menopause.
      ). Although larger symptomatic fibroids are important in terms of treatment, smaller presymptomatic fibroids detected through ultrasonographic screening may be the relevant outcome for primary and secondary prevention.
      The inclusion of smaller and possibly asymptomatic fibroids will also help identify important windows of susceptibility. The current study benefited from updated dietary data every 2 years, which was averaged to provide a more stable and timely measure of exposure. Interestingly, although there was no strong evidence of an association with prospectively updated dietary data, Harris et al. (
      • Harris H.R.
      • Eliassen A.H.
      • Doody D.R.
      • Terry K.L.
      • Misser S.A.
      Dietary fat intake, erythrocyte fatty acids, and risk of uterine fibroids.
      ) did observe an association with EFA, a biomarker measured only at baseline. Identifying windows of susceptibility and possible points of intervention would benefit from analysis of time-varying exposures to fibroid initiation as opposed to clinical detection based on symptoms. Large prospective studies of fibroids with ultrasonographic screening are needed to identify risk factors for fibroid incidence and growth.
      Given the current scarcity of data on strong risk factors for fibroids beyond age, parity, and race/ethnicity, control of confounding in fibroid studies remains challenging (
      • Wise L.A.
      • Laughlin-Tommaso S.K.
      Epidemiology of uterine fibroids: from menarche to menopause.
      ). The current study had data on a wide range of confounders updated over the course of the study. The dietary analysis was able to adjust for time-varying confounders, and this adjustment was important for many of the individual dietary fats. The attenuation in the point estimates, particularly for trans fats, suggests that there may have been residual confounding, potentially by factors that have yet to be identified. The analysis of EFA, adjusted solely for baseline covariates, raises the possibility that uncontrolled confounding (particularly by changes in parity) may have biased the observed associations. This study benefited from frequent participant follow-up and high retention. Future studies of fibroids should aim for frequent comprehensive follow-up examinations so that confounding from important time-varying covariates can be well controlled.
      The burden of fibroids is particularly high among Black women in the United States, in whom incidence is earlier, which may account for their more symptomatic fibroids and greater risk of requiring surgery for symptom control. Associations between dietary fat and fibroid risk have been explored by Black Women’s Health (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.
      ) and the Study of Environment Lifestyle and Fibroids (SELF) (
      • Brasky T.M.
      • Bethea T.N.
      • Wesselink A.K.
      • Wegienka G.R.
      • Baird D.D.
      • Wise L.A.
      Dietary fat intake and risk of uterine leiomyomata: a prospective ultrasound study.
      ), both of which include exclusively Black women and only had measures of self-reported diet. In SELF, fibroids were prospectively detected by serial ultrasonographic examinations. These two studies found similarly null associations for dietary intake of most types of fat (total, saturated, unsaturated, polyunsaturated, and trans fats). Investigators from both SELF and BWHS, however, reported positive associations with dietary intake of marine-sourced long-chain ω-3 fatty acids, in particular docosahexaenoic acid. There was also a weakly elevated association for dietary docosahexaenoic acid in the current study (
      • Harris H.R.
      • Eliassen A.H.
      • Doody D.R.
      • Terry K.L.
      • Misser S.A.
      Dietary fat intake, erythrocyte fatty acids, and risk of uterine fibroids.
      ). All three studies highlight possible differences in overall diet composition by race/ethnicity and the presence of environmental contaminants in sources of marine fatty acids. Although large overall, the current study was underpowered to assess associations among Black participants, and SELF and BWHS were unable to explore associations among non-Black participants. Future studies must be powered to examine associations among Black and nonwhite participants to fully understand dietary patterns and fibroid risk.
      The search for modifiable risk factors for fibroids is just beginning. This study exemplifies many of the study design considerations needed for optimal research: repeated measures of the exposure and covariates, biomarkers of exposures, large sample size with high retention, long-term follow-up, and appropriate analytic methods. To advance fibroid research, however, we need to apply these strengths to large prospective studies with ultrasonographic screening and sufficient numbers of young Black and other non-White women. Given the high burden of disease, improved and larger studies are needed to identify the risk factors and relevant time windows that will ultimately provide opportunities to prevent fibroids.

      Acknowledgments

      The authors thank Dr. Donna Baird for helpful comments on early drafts of this commentary.

      References

        • Wise L.A.
        • Laughlin-Tommaso S.K.
        Epidemiology of uterine fibroids: from menarche to menopause.
        Clin Obstet Gynecol. 2016; 59: 2-24
        • Harris H.R.
        • Eliassen A.H.
        • Doody D.R.
        • Terry K.L.
        • Misser S.A.
        Dietary fat intake, erythrocyte fatty acids, and risk of uterine fibroids.
        Fertil Steril. 2020; 114: 837-847
        • El Andaloussi A.
        • Chaudhry Z.
        • Al-Hendy A.
        • Ismail N.
        Uterine fibroids: bridging genomic defects and chronic inflammation.
        Semin Reprod Med. 2017; 35: 494-498
        • 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
        • Brasky T.M.
        • Bethea T.N.
        • Wesselink A.K.
        • Wegienka G.R.
        • Baird D.D.
        • Wise L.A.
        Dietary fat intake and risk of uterine leiomyomata: a prospective ultrasound study.
        Am J Epidemiol. 2020 June 17; https://doi.org/10.1093/aje/kwaa097

      Linked Article