Impact of a strong fiber, low-fat diet intervention... Part #3

Impact of a strong fiber, low-fat diet intervention... Part #3

Table 2 summarizes body weight and intakes of energy, fiber, and percent energy intake from carbohydrate, fat, and protein at baseline and 12 months in the two study groups. At 12 months, mean BMI was 27.6 kg/m2 (standard deviation [SD], 5.6 kg/m2) in the comparison group and 27.7 kg/m2 (SD, 6.4 kg/m2) in the intervention group, which did not differ from baseline. Similar to the characteristics of a previously described subsample of WHEL study subjects, weight loss was not observed in the majority of these women: 218 women (75%) were within 5% of their baseline weight at 12 months, weight loss of ≥ 5% baseline weight occurred in 35 women (12%), and 38 women (13%) gained > 5% of their baseline weight at 12 months. Level of physical activity did not change significantly from baseline to 12 months in either study group (data not shown).

Table 2. Dietary Intakes and Body Weight of Participants at Baseline and 12 Months, by Study Group Assignment

	Comparison Group (n = 138)		Intervention Group (n = 153)
	Baseline	12 Months	Baseline	12 Months
Energy, kcal/d
Mean	1,736	1,590	1,719	1,640
SD	383	345	365	337
Percent energy from fat
Mean	28	27	28	21
SD	7	7	7	7*
Percent energy from carbohydrate
Mean	56	57	57	66
SD	9	10	7	9
Percent energy from protein
Mean	16	17	16	16
SD	3	3	3	3
Dietary fiber, g/d
Mean	22	22	22	29
SD	8	9	9	10
Body weight, kg
Mean	74	75	74	74
SD	17	16	19	18

As shown in Table 2, there were no significant differences in the intakes of the study groups at baseline. At 1-year follow-up, the intervention group had a significantly lower intake of energy from fat and a significantly higher intake of fiber and carbohydrate than the comparison group (P < .001). As expected, an inverse correlation between energy intake from fat and fiber intake was observed (Pearson r = –0.39; P < .05), likely due to the nutritional guidance provided and the distributions of these dietary constituents in foods. Serum concentrations of reproductive steroid hormones and SHBG did not differ significantly between the two study groups at baseline (Table 3). A significant difference in the baseline to 1-year change in serum bioavailable estradiol concentration in the intervention versus comparison group was observed (P < .05). The change in serum total estradiol in the intervention versus comparison group differed in the expected direction, but this difference did not reach statistical significance (P < .10). Within the intervention group, bioavailable estradiol concentration decreased significantly (P < .05). Changes in total estradiol and SHBG concentrations observed in the women in the intervention group were not statistically significant (P < .10). No significant changes in serum hormone and SHBG concentrations were noted in the comparison group participants. Table 3. Serum Hormone and Sex Hormone-Binding Globulin Concentrations of the Participants at Baseline and 12 Months, by Study Group Assignment

	Comparison Group (n = 138)		Intervention Group (n = 153)
	Baseline	12 Months	Baseline	12 Months
Estradiol, pmol/L
Mean	72	77	91	64
SD	107	132	172	42
Bioavailable estradiol, pmol/L*
Mean	33	36	41	28
SD	46	68	78	25

Weight loss was associated with an increase in SHBG concentration in both study groups. Women who lost ≥ 5% of initial body weight from baseline to one year (n = 35) exhibited a significant increase in serum SHBG concentration (from an average of 72.0 to 84.8 nmol/L; P < .05). Women on tamoxifen (n = 172), who constituted 59% of the study subjects, exhibited a greater decline in estradiol and bioavailable estradiol concentrations than did women not on tamoxifen (n = 119; Table 4). However, an important difference between these two subgroups is that the women on tamoxifen versus those not on tamoxifen had a significantly lower BMI at baseline (averaging 26.8 v 28.8 kg/m2, respectively; P < .01). Also, analysis across subgroups categorized by study arm and tamoxifen usage is constrained by significantly different baseline bioavailable estradiol concentrations (P < .001), and the subgroup cell sizes and variances in these subgroup data are shown in Table 4. Table 4. Serum Estradiol and Bioavailable Estradiol Concentrations in Participants Categorized by Study Group Assignment and Tamoxifen Use

	Estradiol (baseline)	Estradiol (12 Months)	Change in Estradiol	Bioavailable Estradiol* (baseline)	Bioavailable Estradiol* (12 Months)	Change in Bioavailable Estradiol*
Intervention Group
On tamoxifen (n = 94)
Mean	100	63	–36	39	23	–16
SD	204	47	183	88	27	76
Not on tamoxifen (n = 59)
Mean	78	65	–13	44	35	–9
SD	100	32	92	57	19	48
Significance (P value)	.83	.53	.30	.001	.0001	.52
Comparison Group
On tamoxifen (n = 78)
Mean	73	54	–18	25	20	–6
SD	121	15	22	31	11	30
Not on tamoxifen (n = 60)
Mean	72	106	33	42	56	14
SD	83	196	189	60	98	87
Significance (P value)	.59	.006	.07	.0001	.0001	.10
All Participants
On tamoxifen (n = 172)
Mean	87	59	–28	33	22	–11
SD	172	36	158	69	22	60
Not on tamoxifen (n = 119)
Mean	75	86	10	43	46	14
SD	92	142	150	58	72	87
Significance (P value)	.92	.01	.03	.0001	.0001	.02

Results of the linear regression analysis, in which changes in bioavailable estradiol and total estradiol concentrations were modeled as a function of changes in dietary intakes of fiber and percent energy from fat, are shown in Table 5. Change in fiber intake, but not change in fat intake, was significantly independently related to change in bioavailable estradiol and total estradiol concentration (P < .05), when controlled for baseline serum hormone concentration, tamoxifen usage, and change in BMI. Table 5. Regression Models for Changes in Serum Estradiol and Bioavailable Estradiol Concentrations

Dependent Variable	Dietary Factor	ß Coefficient	Significance (P)
Change in bioavailable estradiol concentration (R2 = 0.41)	Change in fiber intake	–0.0096	.002
	Change in percent energy intake from fat	–0.0032	.38
Change in total estradiol concentration (R2 = 0.51)	Change in fiber intake	–0.0062	.02
	Change in percent energy intake from fat	–0.0046	.15