Dietary Fat Intake and Risk of Uterine Leiomyomata: A Prospective Ultrasound Study

Abstract

Uterine leiomyomata (UL) are associated with severe reproductive morbidity and are the primary indication for hysterectomy in the United States. A recent prospective cohort study of Black women reported positive associations between intakes of marine-sourced ω-3 fatty acids and UL risk. We examined whether intakes of dietary fat were associated with UL incidence in a 5-year prospective study of premenopausal Black women living in Detroit who underwent serial ultrasound. At baseline (2010–2012) and 20, 40, and 60 months of follow-up, participants underwent transvaginal ultrasound. Among 1,171 UL-free women at baseline, incident UL were detected in 277 women. Cox regression was used to estimate hazard ratios and 95% confidence intervals for the association of dietary fat and UL incidence. Intakes of total fat and saturated, monounsaturated, polyunsaturated, and trans-fat were not appreciably associated with UL incidence. Intake of the marine ω-3 polyunsaturated fatty acid, docosahexaenoic acid, was associated with 49% higher UL incidence (quartile 4 vs. 1: hazard ratio = 1.49, 95% confidence interval: 1.04, 2.14; P for trend = 0.01). Intakes of total marine ω-3 polyunsaturated fatty acids were similarly associated with elevated UL incidence (hazard ratio = 1.35, 95% confidence interval: 0.94, 1.93; P for trend = 0.03). It remains unclear whether the fatty acids or persistent environmental pollutants drive the association.

Abbreviations

BWHS

Black Women's Health Study

CI

confidence interval

HR

hazard ratio

MUFA

monounsaturated fatty acid

PUFA

polyunsaturated fatty acid

SELF

Study of the Environment, Lifestyle, and Fibroids

SFA

saturated fatty acid

UL

uterine leiomyoma

Uterine leiomyomata (UL, or fibroids) are associated with severe reproductive morbidity and are the primary indication for hysterectomy in the United States (1–3). Among reproductive-aged women, the prevalence of clinical diagnosis is approximately 25%–30% (4). However, a large proportion of UL are undiagnosed (5), and data from standardized ultrasound screening of women 35–49 years of age indicated that the estimated cumulative incidence of ultrasound-detectable UL by age 50 years was >70% (4). Black women have a 2- to 3-fold higher incidence of clinically diagnosed UL compared with White women and experience greater symptom severity (3, 4, 6). UL are also responsible for significant medical expenditures (7), with pronounced racial disparities in health-care costs, that disproportionately affect Black women (8). Aside from racial and reproductive factors, few established risk factors for UL have been identified (1, 9), and these factors do not fully explain the racial disparity in incidence (3, 6).

The underlying sources of dietary fat intake differ between Black and White women in the United States, with Black women generally consuming more dietary fat (10), and a greater proportion of fat from meat and seafood but less fat from dairy, compared with White women (11, 12). To address whether dietary fat intake could explain part of the racial disparity in UL incidence, it is worthwhile to examine the relationship between dietary fat and UL. The Black Women’s Health Study (BWHS), a cohort comprised exclusively of Black women, was the first prospective study to examine associations of dietary fat with UL risk. Investigators reported positive associations of individual dietary monounsaturated fatty acids (MUFAs) and marine-sourced ω-3 polyunsaturated fatty acids (PUFAs) and inverse associations for individual saturated fatty acids (SFAs) with UL risk (13). No appreciable associations were observed for intakes of total fat or most fat groups, with the exception of total MUFA intake, for which a positive association was reported. The results from this study, however, were likely affected by outcome misclassification because identification of UL cases was based on self-report.

To expand on the existing literature, we examined the association between dietary fat intake and UL incidence in the Study of Environment, Lifestyle, and Fibroids (SELF), a prospective cohort study of Black women who underwent serial ultrasound screening for incident UL.

METHODS

The Study of Environment, Lifestyle and Fibroids

SELF is a prospective cohort of Black women living in the Detroit, Michigan, metropolitan area (14). Recruitment of SELF participants occurred from 2010 through 2012, with community outreach through radio, television, newspapers, event booths, and informational letters to women in the Henry Ford Health System, the clinical institution collaborating on SELF (14). Enrolled women (n = 1,693) were self-identified as Black/African-American, were 23–35 years of age without self-reported history of hysterectomy (partial or total, verified at baseline ultrasound), and reported no prior diagnoses of UL, cancer, or autoimmune diseases requiring regular medication. Women who were pregnant at the time of recruitment delayed their enrollment into the study until after delivery to assure optimum ultrasound imaging.

At baseline and every 20 months during a 5-year follow-up period, participants completed computer-assisted telephone and Web-based questionnaires, self-administered paper questionnaires, and attended in-person clinic visits. Participants were queried on their personal medical history, physical activity, lifestyle and behaviors (e.g., smoking, alcohol intake), reproductive history, and use of contraceptives. Transvaginal ultrasound was performed during in-person clinic visits at enrollment (baseline), and at 20, 40, and 60 months of follow-up. Pregnant women were asked to return to the clinic at 4 months postpartum. For women whose pregnancy was identified at follow-up ultrasound, data were recorded if the participant was ≤12 weeks pregnant based upon fetal measures, otherwise they were asked to return to the clinic at 4 months postpartum. Study sonographers were registered diagnostic medical sonographers with ≥3 years of experience in gynecological sonography (15). Sonographers received additional training for the study to assure consistency in completing research documentation about the ultrasound and received regular refresher trainings during the course of the study (14). At each clinic visit trained study staff measured height and weight, from which body mass index was calculated (as weight (kg)/height (m)²). All participants gave written informed consent, and the study was approved by the institutional review boards of the Henry Ford Health System, the National Institute of Environmental Health Sciences, and Boston University Medical Campus.

Dietary measurement

At baseline, women completed a validated Web-based semiquantitative Block food frequency questionnaire (16–18). Participants reported their usual frequencies and serving sizes (in cups: 0.25, 0.50, 1, or 2) of 110 foods and beverages in the past 12 months. Average daily intakes of dietary fat and individual fatty acids were calculated by multiplying the serving-size and season-adjusted frequency of each food item by its fat content as determined by the US Department of Agriculture Food and Nutrient Database for Dietary Studies (19). Data for total dietary fat and fat groups (SFA, MUFA, PUFA, and trans-fatty acids) are given in grams/day, whereas data for individual fatty acids were given in milligrams/day. Summary measures were calculated for total dietary marine ω-3 PUFA, as the sum of eicosapentaenoic (20:5n3), docosapentaenoic (22:5n3), and docosahexaenoic acids (22:6n3), and total ω-6 PUFA as the sum of linoleic (18:2n6) and arachidonic acids (20:4n6). Total fin-fish and shellfish intake was calculated as the sum of serving-size adjusted grams/day of tuna (including tuna salad and tuna casserole), fried fish or fish sandwich, other fish (not fried), shellfish (including shrimp, scallops, and crabs), and oysters.

For the present analysis, we excluded 384 women with prevalent UL identified by ultrasound at enrollment. We excluded an additional 78 women with total energy intakes of <400 or ≥5,000 kcal/day and 60 women who were missing follow-up data, leaving 1,171 women followed for a median of 5 years for incident UL. Among them, we identified 277 incident UL cases (primary outcome) first detected at their 20- (n = 110), 40- (n = 88), or 60-month follow-up visit (n = 79). The remaining women were right-censored at hysterectomy (n = 8), withdrawal from the study after completing at least 1 follow-up visit (n = 66), or their 60-month follow-up visit (n = 820), whichever came first. There were no apparent differences in baseline characteristics among women with no follow-up data (n = 60; excluded), women who withdrew from the study after completing at least 1 follow-up visit (n = 66; censored), and the remaining analytical cohort (n = 1,105) (Web Table 1, available at https://academic.oup.com/aje).

Statistical analysis

Dietary fats were energy-adjusted using the nutrient residual method (20). We categorized dietary variables into quartiles to avoid the assumption of linearity between dietary fats and UL risk and to make our findings comparable with those of the BWHS (13). We used Cox proportional hazards models, with follow-up time as the time metric, to estimate hazard ratios and 95% confidence intervals for the association between dietary fat and UL incidence. In addition to accounting for follow-up time, all models adjusted for age (stratification variable) and total energy intake. A priori we determined additional baseline factors for inclusion in multivariable models; they included education (up to high school, some college, college or advanced degree), body mass index (continuous), age at menarche (in years: <11, 11, 12, 13, ≥14), parity (nulliparous and tertiles of births: 1, 2, and ≥3 births), age at first birth (nulliparous and tertiles of age in years at first birth: <19, 19–22, ≥23), years since last birth (nulliparous and tertiles of years since last birth: 0–2.6, 2.7–5.7, ≥5.8), ever used oral contraceptives, ever used progestin-only injectable contraceptives, alcohol intake (quartiles of drinks/week: <0.33, 0.33–1.27, 1.28–3.98, ≥3.99), smoking status (none, former, current for <10 years, current for ≥10 years), and history of hypertension (yes/no). We calculated P values from tests for trend (P for trend) by assigning the median value to each category of ordinal dietary variables and including them as continuous variables in regression models.

Because progestin-only injectable contraceptives strongly suppress ovarian hormones (estradiol and progesterone), we performed a sensitivity analysis in which we additionally excluded 76 women (including 10 with incident UL cases) who were using these contraceptives at baseline. Statistical analyses were performed using SAS, version 9.4 (SAS Institute, Inc., Cary, North Carolina). All reported P values are 2-sided, and P < 0.05 was considered statistically significant.

RESULTS

Baseline characteristics of SELF participants, stratified by total dietary fat intake, are shown in Table 1. With the exception of fish intake, participants’ baseline characteristics were not appreciably associated with total dietary fat intake. Women who consumed more dietary fat were more likely to consume fish. Household income, body mass index, and reproductive factors including age at menarche, parity, age at first birth, years since last birth, and use of injectable progestin-only contraceptives were similar across categories of dietary fat intake.

Table 1.

Characteristics of Participants According to Categories of Total Dietary Fat Intake (n = 1,171), Study of Environment, Lifestyle and Fibroids, Michigan, 2010–2018

	Energy-Adjusted Quartiles of Total Fat, g/day
	<76.96		76.96–84.07		84.08–91.15		≥91.16
Characteristic	No.	%	No.	%	No.	%	No.	%
Age, yearsa	28.10 (3.53)		28.63 (3.45)		28.62 (2.52)		28.36 (3.34)
Education
Up to high school	66	22.60	53	18.09	61	20.82	68	23.21
Some college	151	51.71	158	53.92	147	50.17	155	52.90
College degree or beyond	75	25.68	82	27.99	85	29.01	70	23.89
Household income, $
<20,000	133	46.34	116	39.86	123	42.12	146	50.17
20,001–50,000	110	38.33	124	42.61	121	41.44	101	34.71
>50,000	44	15.33	51	17.53	48	16.44	44	15.12
Body mass indexa,b	33.90 (9.68)		33.51 (9.49)		33.09 (9.89)		34.44 (10.10)
Age at menarche, years
<11	59	20.21	46	15.70	48	16.38	47	16.04
11	56	19.18	59	20.14	62	21.16	65	22.18
12	82	28.08	88	30.03	79	26.96	79	26.96
13	44	15.07	50	17.06	55	18.77	44	15.02
≥14	51	17.47	50	17.06	49	16.72	58	19.80
No. of live birthsa	1.24 (1.37)		1.30 (1.32)		1.33 (1.44)		1.34 (1.55)
Age at first birth, yearsa,c	20.48 (3.74)		20.85 (4.08)		20.39 (3.32)		20.79 (3.84)
Years since last birtha,c	4.91 (3.67)		4.80 (3.65)		5.33 (4.08)		4.40 (3.32)
Current hormonal contraceptive use
None	194	66.44	191	65.19	202	68.94	204	69.62
Oral contraceptive	33	11.30	34	11.60	41	13.99	27	9.22
Intrauterine device (hormonal)	29	9.93	32	10.92	12	4.10	18	6.14
Injectable progestin-only	18	6.16	17	5.80	19	6.48	22	7.51
Other	18	6.16	19	6.48	19	6.48	22	7.51
Oral contraceptive use
Never	79	27.05	77	26.28	96	32.76	87	29.69
Ever	213	72.95	216	73.72	197	67.24	206	70.31
Age at first oral contraceptive use, years
Never	79	27.05	77	26.28	96	32.76	87	29.79
12–17	91	31.16	105	35.84	87	29.69	108	36.99
18–33	122	41.78	111	37.88	110	37.54	97	33.22
Injectable progestin-only contraceptive use
Never	170	58.22	164	55.97	168	57.34	148	50.51
Ever	122	41.78	129	44.03	125	42.66	145	49.49
Alcohol intake, drinks/week
<0.33	72	24.66	69	23.55	77	26.28	72	24.57
0.33–1.27	62	21.23	79	26.96	80	27.30	71	24.23
1.28–3.98	60	20.55	75	25.60	73	24.91	88	30.03
≥3.99	98	33.56	70	23.89	63	21.50	62	21.16
Smoking status
Never smoker	206	70.55	219	74.74	229	78.16	218	74.40
Former smoker	24	8.22	19	6.48	21	7.17	19	6.48
Current, <10 years	47	16.10	42	14.33	32	10.92	47	16.04
Current, ≥10 years	15	5.14	13	4.44	11	3.75	9	3.07
History of hypertension
No	256	87.97	264	91.35	267	91.75	256	88.58
Yes	35	12.03	25	8.65	24	8.25	33	11.42
History of diabetes
No	278	96.19	286	97.95	284	97.59	278	94.88
Yes	11	3.81	6	2.05	7	2.41	15	5.12
Fish intake, g/day
<9.59	84	28.77	83	28.33	78	26.62	47	16.04
9.59–21.39	64	21.92	82	27.99	79	26.96	68	23.21
21.40–39.99	68	23.29	73	24.91	71	24.23	81	27.65
≥40.00	76	26.03	55	18.77	65	22.18	97	33.11
Red meat intake, g/day
<16.53	68	23.29	92	31.40	83	28.33	49	16.72
16.53–28.35	75	25.68	87	29.69	69	23.55	61	20.82
28.36–48.48	71	24.32	66	22.53	80	27.30	74	25.26
≥48.49	78	26.71	48	16.38	61	20.82	109	37.20
Poultry intake, g/day
<11.82	64	21.92	90	30.72	83	28.33	54	18.43
11.82–22.08	75	25.68	80	27.30	81	27.65	58	19.80
22.09–42.81	81	27.74	69	23.55	65	22.18	77	26.28
≥42.82	72	24.66	54	18.43	64	21.84	104	35.49

^a Values are expressed as mean (standard deviation).

^b Body mass index calculated as weight (kg)/height (m)².

^c Among parous women.

Associations of energy-adjusted quartiles of total fat and individual fatty acid intakes with UL incidence are presented in Table 2 (reduced table) and Web Table 2 (full table). Hazard ratios comparing the highest versus lowest quartiles of fat consumption were: 1.08 (95% confidence interval (CI): 0.76, 1.52) for total dietary fat, 0.97 (95% CI: 0.67, 1.38) for SFA, 1.17 (95% CI: 0.83, 1.64) for MUFA, 0.91 (95% CI: 0.64, 1.31) for PUFA, and 0.83 (95% CI: 0.59, 1.18) for trans-fatty acids. Greater intake of each of the 3 marine ω-3 PUFAs was positively associated with UL incidence (quartiles 4 vs. 1: docosahexaenoic acid (PUFA 22:6) hazard ratio (HR) = 1.49, 95% CI: 1.04, 2.14 (P for trend = 0.01); eicosapentaenoic acid (PUFA 20:5) HR = 1.25, 95% CI: 0.87, 1.81 (P for trend = 0.06); docosapentaenoic acid (PUFA 22:5) HR = 1.22, 95% CI: 0.85, 1.75 (P for trend = 0.21)). The hazard ratios for quartiles 2, 3, and 4 versus 1 of total dietary marine ω-3 PUFA were 0.87 (95% CI: 0.59, 1.29), 1.14 (95% CI: 0.79, 1.66), and 1.35 (95% CI: 0.94, 1.93), respectively, and a linear trend was observed (P for trend = 0.03). Additionally, there was a suggestive association with intake of palmitoleic acid (MUFA 16:1) (HR = 1.42, 95% CI: 0.99, 2.04; P for trend = 0.10; Web Table 2). Hazard ratios for intakes of individual SFAs and trans-fatty acids, as well as the remaining MUFAs and PUFAs did not suggest appreciable associations.

Table 2.

Associations of Dietary Fat Intake With Uterine Leiomyomata Risk (n = 1,171), Study of Environment, Lifestyle and Fibroids, Michigan, 2010–2018

	Energy-Adjusted Quartiles of Fat Intakea
	1			2			3			4
Dietary Fat	No. of Cases	HR b	95% CI	No. of Cases	HR b	95% CI	No. of Cases	HR b	95% CI	No. of Cases	HR b	95% CI	P for Trend
Total fat	67	1.00	Referent	68	0.89	0.62, 1.27	67	0.80	0.56, 1.14	75	1.08	0.76, 1.52	0.78
Total SFA	66	1.00	Referent	71	0.92	0.65, 1.32	74	1.05	0.74, 1.48	66	0.97	0.67, 1.38	0.98
Total MUFA	67	1.00	Referent	70	0.90	0.63, 1.29	60	0.72	0.50, 1.04	80	1.17	0.83, 1.64	0.53
Total PUFA	61	1.00	Referent	71	1.10	0.76, 1.58	79	1.28	0.89, 1.83	66	0.91	0.64, 1.31	0.67
ω-3 fatty acids
α-linolenic acid	66	1.00	Referent	81	1.10	0.78, 1.56	67	0.84	0.59, 1.21	63	0.82	0.58, 1.18	0.14
Eicosapentaenoic acid	61	1.00	Referent	59	0.83	0.56, 1.23	78	1.14	0.78, 1.66	79	1.25	0.87, 1.81	0.06
Docosapentaenoic acid	61	1.00	Referent	68	1.01	0.69, 1.47	70	1.00	0.68, 1.47	78	1.22	0.85, 1.75	0.21
Docosahexaenoic acid	56	1.00	Referent	67	1.04	0.71, 1.52	68	1.12	0.74, 1.59	86	1.49	1.04, 2.14	0.01
Total marine ω-3	59	1.00	Referent	61	0.87	0.59, 1.29	76	1.14	0.79, 1.66	81	1.35	0.94, 1.93	0.03
Total trans-fat	73	1.00	Referent	64	0.75	0.52, 1.07	75	0.91	0.64, 1.30	65	0.83	0.59, 1.18	0.53

Abbreviations: CI, confidence interval; HR, hazard ratio; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; SFA, saturated fatty acid.

^a Category cutpoints and minimally adjusting regression models are presented in Web Table 2.

^b Adjusted for age, education, income, body mass index, age at menarche, parity, age at first birth, years since last birth, oral contraceptive use, progestin-only injectable contraceptive use, alcohol intake, smoking, history of hypertension, and total energy intake.

Associations of fish intake and UL risk are given in Table 3. Similar to the marine ω-3 PUFAs, hazard ratios for total fish intake were also elevated: hazard ratio = 1.19 (95% CI: 0.93, 1.71), hazard ratio = 1.33 (95% CI: 0.93, 1.91), and hazard ratio = 1.37 (95% CI: 0.94, 2.00) for quartiles 2, 3, and 4 versus 1 (P for trend = 0.14), respectively. Associations were generally similar for fin fish and shellfish.

Table 3.

Associations of Fish Intake With Uterine Leiomyomata Risk (n = 1,171), Study of Environment, Lifestyle and Fibroids, Michigan, 2010–2018

	Quartiles of Fish Intake
	1			2			3			4
Fish	No. of Cases	HR a	CI	No. of Cases	HR a	CI	No. of Cases	HR a	CI	No. of Cases	HR a	CI	P for Trend
Total fishb	61	1.00	Referent	69	1.19	0.93, 1.71	76	1.33	0.93, 1.91	71	1.37	0.94, 2.00	0.14
Fin fishc	57	1.00	Referent	79	1.50	1.05, 2.14	73	1.41	0.98, 2.03	68	1.40	0.95, 2.06	0.27
Shellfishd	69	1.00	Referent	60	1.17	0.81, 1.67	68	1.38	0.97, 1.97	80	1.26	0.89, 1.79	0.31

Abbreviations: CI, confidence interval; HR, hazard ratio.

^a Adjusted for age, education, income, body mass index, age at menarche, parity, age at first birth, years since last birth, oral contraceptive use, progestin-only injectable contraceptive use, alcohol intake, smoking, history of hypertension, and total energy intake.

^b The cutpoints are as follows, quartile 1, <9.59 g/day; quartile 2, 9.59–21.39 g/day; quartile 3, 21.40–39.99 g/day; and quartile 4, ≥40.00 g/day.

^c The cutpoints are as follows, quartile 1, <7.27 g/day; quartile 2, 7.27–16.03 g/day; quartile 3, 16.04–29.77 g/day; and quartile 4, ≥29.78 g/day.

^d The cutpoints are as follows, quartile 1, <1.00 g/day; quartile 2, 1.00–3.07 g/day; quartile 3, 3.08–9.10 g/day; and quartile 4, >9.10 g/day.

After excluding women who used progestin-only injectable contraceptives at baseline, findings for marine ω-3 PUFA were stronger (Table 4). For docosahexaenoic acid specifically, the hazard ratio comparing quartile 4 versus 1 was 1.59 (95% CI: 1.10, 2.30; P for trend < 0.01). Findings for palmitoleic acid were unchanged (data not shown).

Table 4.

Associations of Marine ω-3 Polyunsaturated Fatty Acid Intake With Uterine Leiomyomata Risk Among Women Not Using Injectable Progestin-Only Contraceptives at Baseline (n = 1,095), Study of Environment, Lifestyle and Fibroids, Michigan, 2010–2018

	Energy-Adjusted Quartiles of Fat Intake a
	1			2			3			4
ω -3 Fatty Acids	No. of Cases	HR b	CI	No. of Cases	HR b	CI	No. of Cases	HR b	CI	No. of Cases	HR b	CI	P for Trend
Eicosapentaenoic acid	57	1.00	Referent	55	0.85	0.56, 1.28	77	1.18	0.80, 1.74	78	1.27	0.88, 1.85	0.06
Docosapentaenoic acid	57	1.00	Referent	65	1.11	0.74, 1.64	68	1.10	0.74, 1.64	77	1.28	0.89, 1.85	0.18
Docosahexaenoic acid	51	1.00	Referent	64	1.12	0.75, 1.66	67	1.26	0.85, 1.87	85	1.59	1.10, 2.30	<0.01
Total marine ω-3	55	1.00	Referent	57	0.92	0.62, 1.38	75	1.25	0.86, 1.84	80	1.41	0.97, 2.03	0.02

Abbreviations: CI, confidence interval; HR, hazard ratio.

^a Category cutpoints are given in Web Table 2.

^b Adjusted for age, education, income, body mass index, age at menarche, parity, age at first birth, years since last birth, oral contraceptive use, progestin-only injectable contraceptive use, alcohol intake, smoking, history of hypertension, and total energy intake.

DISCUSSION

In this prospective cohort study of reproductive-aged Black women who underwent serial ultrasound during a 5-year period, dietary intakes of total fat and most individual fatty acids were not appreciably associated with UL incidence. However, higher dietary intakes of total marine ω-3 PUFAs, particularly docosahexaenoic acid, were associated with increases in UL incidence. To our knowledge, this study represents the first prospective ultrasound study of dietary fat and UL incidence.

In SELF, findings regarding intake of the marine ω-3 PUFA docosahexaenoic acid (quartile 4 vs. 1: HR = 1.49) were consistent with a prior report from the BWHS (2,695 cases), in which higher intakes of marine ω-3 PUFA (along with their fish sources) were associated with 13%–21% increased risks of UL (13). Although hazard ratios were imprecise, our findings for the marine ω-3 PUFAs eicosapentaenoic acid (HR = 1.25) and docosapentaenoic acid (HR = 1.22) are similar in magnitude to those of the BWHS. Two other studies have reported on fish intake and UL risk in White women (21, 22). Similar to our findings on fish intake, a cohort study examining Great Lakes sport-caught fish intake reported 20% increase in risk of self-reported UL associated with fish consumption (22). In contrast, an Italian clinic-based case-control study reported 30% reductions in the odds of UL among women in the second (odds ratio = 0.7, 95% CI: 0.5, 0.8) and third tertiles (odds ratio = 0.7, 95% CI: 0.6, 0.9) of fish consumption relative to the first (21); however, this study might have been prone to recall bias.

Our observation of 43% higher UL risk for greater intakes of palmitoleic acid (which derives from vegetable oils and animal fats) is not clearly supported by data from the BWHS (quintile 5 vs. 1: HR = 1.09, 95% CI: 0.97, 1.23) (13). Additionally, in the BWHS, total MUFA (HR = 1.15, 95% CI: 1.02, 1.30), as well as individual MUFAs oleic acid (HR = 1.13, 95% CI: 1.00, 1.27) and erucic acid (HR = 1.17, 95% CI: 1.04, 1.32), were positively associated with UL risk (13). Although imprecise, our results for intake of total MUFA (HR = 1.17) and oleic acid (HR = 1.23) were also similar in direction and magnitude to those found in the BWHS (Web Table 2).

We did not observe associations for the remaining fat groups or individual fatty acids. In the BWHS, there were relatively small inverse associations for intakes of SFAs caprylic acid (HR = 0.89, 95% CI: 0.78, 0.99) and caproic acid (HR = 0.86, 95% CI: 0.76, 0.97) (13). In the present study, hazard ratios for caprylic acid (HR = 0.92) were consistent, but those for caproic acid (HR = 1.04) were contrary to those reported previously (13) (Web Table 2).

The present study has several limitations. Foremost, with 1,171 women under study and 277 incident cases, the study had low power to identify small associations between dietary fats and UL risk. The use of dietary self-report is also subject to error (23). Given the study’s prospective design, we expect this error to be nondifferential, but it could have attenuated our results. Further, the FFQ did not collect data on the specific types of fish consumed, which would be expected to add nondifferential error in the estimation of marine ω-3 PUFAs. Unfortunately, we did not measure fatty acids in circulation (e.g., plasma phospholipids), which could provide a more reliable estimate of association with UL risk. We additionally could not account for supplemental intakes of ω-3 PUFAs. As is the case with any observational study, residual confounding could have influenced our findings. For instance, participants were not queried at baseline on childhood abuse (24–26) or past uterine infections, which may confound the fat-UL association. In addition, analyses were not adjusted for prevalent diabetes due to small numbers, nor did they account for multiple comparisons. Finally, it is important to note that our results among a study population comprised exclusively of Black women cannot fully address whether dietary fats explain the racial disparity in UL incidence.

These limitations notwithstanding, the present study has several strengths. It focuses on a population of women at high risk of UL. To our knowledge, ours is only the second prospective study and the first ultrasound-based study to examine associations of dietary fat intake with UL risk. The utilization of serial ultrasound to classify UL and minimize detection bias is a major strength, given the high proportion of UL that are undiagnosed (4). We know of no prior study of diet and UL risk has been able to discern incident from prevalent disease using ultrasound, a detection method with high sensitivity and specificity relative to histologic evidence (27). Finally, we were able to adjust for a wide range of potential confounders, including known and suspected risk factors for UL.

There is no clear explanation for the increased UL risk observed among women with high intakes of marine ω-3 PUFA. Indeed, a wealth of evidence suggests that marine ω-3 PUFA would be expected to have an inhibitory effect on uterine neoplasms given that they are thought to reduce inflammation through the inhibition of nuclear factor κ-B (28, 29), which acts as a transcription factor for targets associated with inflammation, including interleukin-6 and cyclooxygenase-2. Because eicosapentaenoic acid and docosahexaenoic acid are incorporated into cell phospholipids partly at the expense of arachidonic acid (ω-6 PUFA), they attenuate the ability of cyclooxygenase enzymes to synthesize proinflammatory eicosanoids (29, 30). Among observational studies, marine ω-3 PUFAs have been associated with reductions in circulating C-reactive protein, interlukin-6, and tumor necrosis factor-α (31). In human trials, fish oil supplementation reduced several blood biomarkers of inflammation, including C-reactive protein (32, 33) and tumor necrosis factor-α (32, 34), when contrasted against a placebo.

An alternative hypothesis is that persistent pollutants such as polychlorinated biphenyls or mercury—both of which bioaccumulate in fish—might explain the positive associations that we observed; however, evidence remains limited. In a cross-sectional analysis of the National Health and Nutrition Examination Survey (35), unadjusted geometric mean levels of whole blood mercury were significantly higher in women with UL versus without UL, and positive associations were found with self-reported UL diagnosis. Data from a recent prospective analysis of blood and urinary mercury concentrations and UL prevalence, as determined by laparoscopic surgery (likely missing intramural and submucosal UL, the most common types of UL), did not support these findings (36). In the same study, authors reported that high visceral fat concentrations of several polychlorinated biphenyls were associated with 52%–88% increased prevalence of UL (37).

The present prospective study of Black women who underwent serial ultrasound provides additional evidence that intakes of marine ω-3 PUFAs, particularly docosahexaenoic acid, are associated with increased UL risk. If the association is real, it remains unclear whether the fatty acids themselves or persistent environmental pollutants are an underlying causal agent. Additional prospective studies with careful measurement of dietary marine ω-3 fatty acids or biomarkers of their intake, as well as ultrasound-detected UL, are needed. Biomarker measurements of exposure to persistent organic and inorganic pollutants would further help clarify the extent to which the association is explained by environmental pollutants, the fatty acids themselves, or other factors.