No evidence of hypoglycemia or hypotension in older adults during 6 months of flax lignan supplementation in a randomized controlled trial: A safety evaluation

Abstract

Context: The natural health product, BeneFlax, is a standardized flaxseed [Linum usitatissimum L. (Linaceae)] lignan enriched product with evidence of product quality and known quantity of the bioactive component, lignan. The acceptance of this natural health product for its various health benefits requires greater evidence of its safety in the general population.

Objective: We determined whether flaxseed lignan causes clinical hypoglycemia or hypotension in healthy older adults as an important aspect of safety.

Materials and methods: Participants aged 49–87 years were randomized in a double-blind trial to receive flaxseed lignan (543 mg/day in BeneFlax) or placebo while completing a 6-month walking program. The 94 participants who completed the study were stratified by age (<65 years versus ≥65 years) and treatment category to determine whether older adults were more susceptible to adverse effects.

Results: After 6 months of treatment, average plasma glucose level (5.4 ± 0.6 mmol/L), systolic blood pressure (127 ± 14 mm Hg), and diastolic blood pressure (80 ± 9 mm Hg) were within normal clinical range. Controlling for sex and body mass index covariates resulted in no observed differences between plasma glucose or blood pressure measurements between treatment or age groups (p > 0.05). No incidents of hypoglycemia or hypotension were observed during BeneFlax treatment, suggesting that 543 mg falls at or below the no observable adverse effect level (NOAEL).

Discussion and conclusion: These data suggest the flaxseed lignan product BeneFlax does not pose a risk of hypoglycemia or hypotension in healthy adults aged 49–87 years.

• Blood glucose
• blood pressure
• flaxseed lignan
• supplement

Introduction

Flaxseed (Linum usitatissimum L.) foods and derived products are beneficial for improving blood lipid profiles (Bloedon et al., 2008; Lucas et al., 2002) and to protect against some types of cancer (Demark-Wahnefried et al., 2008; Jenab & Thompson, 1996) in human clinical trials and animal models. While other flaxseed constituents may contribute to these effects, we recently reported that the lignans of flaxseed may provide antioxidant and anti-estrogen protection to cell and tissue processes, thus contributing to overall protection from these chronic conditions (Adolphe et al., 2010).

Following oral consumption, the predominant lignan in flaxseed, secoisolariciresinol diglucoside (SDG), was converted to the mammalian lignans, enterodiol (ED) and enterolactone (EL), by bacteria in the human colon (Hu et al., 2007; Jenab & Thompson, 1996). During absorption, EL and ED undergo further metabolism by phase II biotransformation processes with extensive formation of glucuronide and sulfate conjugates, thereby reducing the oral bioavailability of EL and ED (Adlercreutz et al., 1995; Dean et al., 2004). Despite a lower oral bioavailability, the mammalian lignans seem to mediate beneficial health effects.

EL and ED have structural similarity to estradiol, the active form of estrogen in the body. Binding of these lignans to estrogen receptors can exert weak estrogenic or anti-estrogenic effects (Carreau et al., 2008; Penttinen-Damdimopoulou et al., 2009; Penttinen et al., 2007), and such activity is believed to decrease the risk of hormone-sensitive cancers (Buck et al., 2010). EL and ED also possess antioxidant activity (Kitts et al., 1999; Prasad, 2000) which may contribute to a reduction in risk of diseases and conditions associated with increased inflammation and oxidative damage such as metabolic syndrome (Cornish et al., 2009). Finally, lignans may provide health benefits through the induction of phase II proteins (e.g., antioxidant enzymes), thus promoting the scavenging of oxidants or decreasing the probability of oxidant formation (Juurlink, 2001; Wang et al., 1998).

Flaxseed contains varying amounts of SDG, depending on a variety of factors such as growing locations and conditions (Westcott et al., 2002). In the literature, the uncertainly concerning the health benefits of flaxseed lignans stems from the indeterminate amount of lignan administered in clinical trials when flaxseed products such as whole flaxseed or defatted flaxseed meal are used as the lignan source (Bloedon et al., 2008; Cornish et al., 2009; Demark-Wahnefried et al., 2008; Hallund et al. 2008; Lucas et al., 2002; Pan et al., 2007; Zhang et al., 2008). Recently, standardized flaxseed products have become available for commercial sale and use. Archer Daniels Midland has produced BeneFlax, which contains a standardized amount of SDG and can thus provide a known dose of lignan. Furthermore, with product standardization comes the ability to positively relate study outcomes to a known ingredient in the well-characterized and reproducible product.

While research continues into the potential for SDG to provide health benefits, a need also exists to examine whether unintended effects result from the administration of high lignan extract supplementation. Older humans, i.e., age ≥65 years may benefit most from antioxidant activity and anti-estrogen effects of lignans to ameliorate chronic inflammation. This age group, though, may be at greater risk for side-effects because adverse drug reactions and drug–drug interactions are more common in older adults than the general population due to polypharmacy, age-related changes in physiology, and/or underlying chronic disease (Mangoni & Jackson, 2004).

In proposing a clinical trial of SDG supplementation in adults over 60 years of age living in long-term care (Whiting, 2010), concerns were raised about reported reductions in blood glucose levels (Pan et al., 2007) and blood pressure (Cornish et al., 2009) with SDG supplementation and whether this could cause hypoglycemia or hypotension in older, frailer participants. As a consequence, we decided to monitor blood glucose and blood pressure in participants who were given 300 mg of SDG (or placebo) in our ongoing clinical trial (NCT01234506).

To address the concern of potential age-related adverse effects, we used data from a completed community-based 6-month efficacy intervention trial of BeneFlax (Cornish et al., 2009), containing 543 mg SDG, which had not been previously evaluated for age or treatment effects on hypoglycemia or hypotension. We analyzed the study data in two ways: (1) we examined all blood glucose and blood pressure measurements for any incidents of hypoglycemia or hypotension, respectively; and (2) after stratifying by age (<65 years versus ≥65 years) and treatment (lignan versus placebo), we determined if SDG affected blood glucose or blood pressure levels more in older than younger participants, after controlling for sex and body mass index (BMI).

Materials and methods

Clinical trial overview

A randomized, double-blind, placebo-controlled study design was used to assess the effects of flaxseed lignan complex supplementation during exercise training on a metabolic syndrome composite score and osteoporosis risk in older adults (Cornish et al., 2009). The study was conducted between March 2005 and March 2006. One hundred participants aged 49–87 years were randomized to receive flaxseed lignan (543 mg/day in a 4050 mg complex) or placebo while completing a 6-month walking program (30–60 min/day, 5–6 days/week). The lignan and placebo were in tablet form, donated by Archer Daniels Midland Co. (Decatur, IL). Participants were excluded from the study if they (1) were taking pharmaceutical agents that lower cholesterol levels; (2) had type I or II diabetes; (3) were smokers; (4) were diagnosed with inflammatory bowel disease; (5) were taking hormone replacement therapy; (6) ingested flaxseed supplements (flax oil, flax lignan, or flax fiber) within the previous 2 months; or (7) were involved in vigorous exercise training (≥3 times/week). The trial was conducted as intent-to-treat and missing data points from individual cases for blood work were accounted for by carrying over the last data point. The tablet containing the active flaxseed lignan complex provided 543 mg flax extract composed of 32.9% SDG (Archer Daniels Midland Co., Decatur, IL). The placebo tablet contained 550 mg of 90% maltodextrin and 10% caramel color. Ethics approval for the study was received from the Biomedical Research Ethics Board of The University of Saskatchewan. Before the start of the study, all participants gave their written informed consent.

Ninety-five participants (55 females and 40 males) had sufficient data for the secondary analysis (see following). Only 92 had complete data as published elsewhere (Cornish et al., 2009). Overnight fasting blood samples for serum glucose (using a LX20 Beckman Coulter analyzer; Beckman Coulter Canada Inc., Mississauga, ON) were measured every 2 months. Resting blood pressure and body composition was assessed at baseline and 6 months. After 5 min of sitting, resting blood pressure was measured with a standard sphygmomanometer by a trained exercise specialist (Canadian Society for Exercise Physiology – Certified Exercise Physiologist) at the same time of day on each occasion.

Data cleaning

On secondary analysis of the data, the dependent variables of interest were plasma glucose concentration, systolic blood pressure (SBP), and diastolic blood pressure (DBP). Due to differences in missing data, the SBP and DBP data were cleaned together, while plasma glucose data were cleaned separately. The reason for this was because missing data were replaced by carrying the last data point forward. Patient records with data missing at baseline were deleted from the analysis. There were two identical cases deleted from the SBP and DBP data on this basis and one case in the plasma glucose data. The record deleted from the plasma glucose data was different from the records deleted in the SBP and DBP data. The independent variables assessed in the analysis were treatment group (placebo or lignan) and age category (<65 years or ≥65 years). Covariates considered were BMI and sex; the measured BMI value did not change between baseline and 6-month measurements, and hence the BMI covariate was an average of the two measurements. Sex was included as a dichotomous covariate.

Statistical analysis

Plasma glucose, SBP, and DBP were analyzed separately using a Generalized Linear Model, with repeated measures. Repeated measurements are correlated within subjects and were accounted for using a compound symmetry correlation structure. Covariance parameters were estimated with the restricted maximum likelihood method. Age (≥65 years versus <65 years) and treatment (lignan versus placebo) were the predictors of interest in this analysis. The age range of our current clinical trial is 60–80 years; however, the cut-off age of 65 was chosen because those in the ≥65 years are considered members of the senior population. Sex (female versus male), time (6 months versus baseline), and BMI (continuous) were included as covariates. Hypoglycemia was defined as plasma glucose <3.9 mmol/L, and clinical hypotension, defined as SBP <80 mm Hg.

For each outcome variable three statistical models were fit to the data. Model 1 included the predictors for treatment, age, time, sex, and BMI. Model 2 included a treatment × age category interaction term, in addition to the predictors in Model 1. Model 3 included a treatment × time point interaction term, in addition to the predictors in Model 2. Conclusions did not differ between Models 1, 2, and 3. The results of Model 3 are presented because it tested for differences in treatment effects over time and treatment effects between different age groups. Residuals were normally distributed and the variance was homogeneous. Leverage and Cook’s D statistics did not identify any outliers. The intracluster correlation was 79% for plasma glucose, 70% for SBP, and 67% for DBP, indicating that individuals had similar measurements between baseline and 6 months, and that the majority of the variability in the data is due to differences between individuals. Analysis was performed using PROC MIXED (SAS version 9.2, SAS Institute Inc, Cary, NC). All values reported are the average ±SD.

Results

The average age of the study population was 61.3 ± 6.7 years, and the majority of participants were less than 65 years of age (Table 1). The results in Table 1 show that no significant differences in age, weight, height, or BMI were detected between participants classified by age (<65 years versus ≥65 years) and treatment (placebo versus lignan).

Table 1. Baseline characteristics of participants.

Treatment	Age	n (M, F)	Age (y)	Weight (kg)	Height (cm)	BMI (kg/m^2)
Placebo	<65	33 (12, 21)	58.1 (4.0)	82.8 (14.5)	168 (9)	29.6 (5.5)
Lignan	<65	36 (14, 22)	58.2 (3.6)	78.5 (16.5)	168 (9)	27.7 (4.6)
	p value	0.83^a	0.91^b	0.26^b	0.88^b	0.13^b
Placebo	≥65	13 (6, 7)	71.6 (6.9)	80.1 (16.5)	169 (7)	27.9 (4.3)
Lignan	≥65	13 (7, 6)	67.6 (2.3)	79.1 (15.4)	169 (9)	27.5 (3.6)
	p value	0.69^a	0.067^b	0.87^b	0.96^b	0.78^b

M, males; F, Females. All values are means (SD).

^aChi-square test comparing counts in Treatment × Sex categories within an age category.

^bT-test comparisons between Placebo and Lignan groups within an age category.

At baseline, the study participants had an average plasma glucose level of 5.4 ± 0.6 mmol/L, SBP of 127 ± 12 mm Hg, and DBP of 81 ± 8 mm Hg. Follow-up at 6 months showed average plasma glucose to be identical to baseline, while SBP was 127 ± 14 mm Hg, and DBP was 80 ± 9 mm Hg. No significant differences in unadjusted plasma glucose, SBP, or DBP existed in the study population at baseline or 6 months between treatment groups for those <65 or ≥65 years of age (Table 2). Glucose, SBP, and DBP were within the normal clinical ranges, except for a single participant who had elevated blood glucose (Table 2). This participant was retained in all analyses and did not overly influence the outcome of the analysis.

Table 2. Blood glucose, SBP, and DBP by treatment group, age category, and time point.

				Unadjusted means (SD)
Outcome	Treatment	Age	n (M, F)	Baseline	6 months	Range
Glucose (mmol/L)	Placebo	<65	33 (12, 21)	5.3 (0.4)	5.3 (0.4)	4.7–6.3
	Lignan	<65	35 (14, 21)	5.2 (0.6)	5.3 (0.5)	4.3–6.9
			p value	0.67	0.63
	Placebo	≥65	13 (6, 7)	5.6 (0.6)	5.5 (0.6)	4.4–6.5
	Lignan	≥65	13 (7, 6)	5.6 (0.9)	5.6 (1.1)	4.8–9.1^a
			p value	0.80	0.71
SBP (mm Hg)	Placebo	<65	33 (12, 21)	127 (13)	126 (17)	92–179
	Lignan	<65	36 (14, 22)	128 (11)	126 (11)	96–146
			p value	0.76	0.87
	Placebo	≥65	12 (6, 6)	126 (18)	125 (16)	96–154
	Lignan	≥65	12 (7, 5)	128 (9)	132 (13)	112–154
			p value	0.73	0.17
DBP (mm Hg)	Placebo	<65	33 (12, 21)	82 (7)	82 (9)	62–110
	Lignan	<65	36 (14, 22)	83 (7)	81 (7)	66–98
			p value	0.64	0.83
	Placebo	≥65	12 (6, 6)	74 (9)	74 (10)	60–88
	Lignan	≥65	12 (7, 5)	81 (8)	81 (8)	65–92
			p value	0.050	0.077

M, males; F, Females; p values are generated from T-test comparisons between Placebo and Lignan groups within an age category (<65 or ≥65 years) and time point (Baseline or 6 months).

^aOne participant was observed to have an elevated glucose concentration of 9.1 mmol/L. The range of glucose concentration was 4.8–6.9 mmol/L without the outlying participant. All statistics presented were generated with the outlying participant included in the analysis.

In the analysis adjusted for sex and BMI, the interaction of treatment (lignan versus placebo) and time was not significant for plasma glucose, SBP, or DBP (Table 3). Similarly, the interaction of treatment and age category revealed no significant differences in plasma glucose, SBP, or DBP (Table 3). These results indicate that plasma glucose, SBP, and DBP responded in a similar way to lignan and placebo treatments over time and between age categories.

Table 3. Adjusted glucose, SBP, and DBP by treatment group, age category, and time point.

				Adjusted means (SD)^a
Outcome	Treatment	Age	n (M, F)	Baseline	6 Months	Fixed effects	p Value
Glucose	Placebo	<65	33 (12, 21)	5.3 (0.2)	5.3 (0.2)	Treatment	0.73
	Lignan	<65	35 (14, 21)	5.3 (0.2)	5.3 (0.2)	Time	0.62
	Placebo	≥65	13 (6, 7)	5.5 (0.2)	5.5 (0.2)	Age	0.24
	Lignan	≥65	13 (7, 6)	5.6 (0.2)	5.7 (0.2)	Treatment × Age	0.76
						Treatment × Time	0.18
SBP	Placebo	<65	33 (12, 21)	127 (4)	126 (4)	Treatment	0.62
	Lignan	<65	36 (14, 22)	127 (5)	127 (4)	Time	0.46
	Placebo	≥65	12 (6, 6)	126 (5)	125 (5)	Age	1.00
	Lignan	≥65	12 (7, 5)	130 (4)	130 (4)	Treatment × Age	0.70
						Treatment × Time	0.52
DBP^b	Placebo	<65	33 (12, 21)	82 (2)	82 (2)	Treatment	0.43
	Lignan	<65	36 (14, 22)	83 (2)	81 (2)	Time	0.73
	Placebo	≥65	12 (6, 6)	74 (3)	74 (3)	Age	0.0023
	Lignan	≥65	12 (7, 5)	82 (2)	80 (2)	Treatment × Age	0.095
						Treatment × Time	0.41

^aAdjusted for sex and BMI.

^bAge main effect (the DBP of participants ≥65 years was, on average, 7.5 ± 2.4 mm Hg lower than participants <65 years; p = 0.0023; reported as mean ± SD).

Discussion

Our findings are applicable to middle-aged and older adults; however, we are focusing on adults 65 years and older as this group represents our current demographic in an ongoing clinical trial. We wanted assurance that our subjects were not at risk of hypoglycemia or hypotension associated with BeneFlax use.

Observed main findings were that SDG supplementation did not cause incidents of clinical hypoglycemia, defined as laboratory confirmed hypoglycemia with plasma glucose <3.9 mmol/L, or of clinical hypotension, defined as SBP <80 mm Hg. Furthermore, no differences between treatment and placebo groups were observed over time across different age categories with respect to SBP, DBP, or plasma glucose. The differences that were found represented generally weak effects. After controlling for BMI and sex, the largest effect was found between age categories for DBP where older participants had lower DBP. Younger participants are expected to have higher DBP as this physiological parameter is known to decrease with age (Franklin et al., 1997).

The literature currently lacks information regarding the safety of flax lignan complex supplementation in adults between 60 and 80 years of age. Specifically, the effects on blood glucose and blood pressure have not been examined in an older population. Concern about flax use in older people may come from vascular effects reported in rats consuming a diet consisting of 20% flaxseed, which showed a vasorelaxant effect in hypertensive rats. In that animal study, however, the response in hypertensive rats to flaxseed was greater than that of normotensive rats and the mechanism was blood pressure-independent (Skoff et al., 2011). Hallund et al. (2008) previously investigated the effects of giving 543 mg of SDG to healthy post-menopausal women aged 61 ± 7 years and saw no significant change in body weight, SBP, or DBP. While their study did not separate effects between older and young participants, we report similar findings as our data suggest no age-related differences in these parameters.

There may be concern that flaxseed lignan consumption may induce hypoglycemia as it can lower blood glucose. However, not all hypoglycemic agents lower blood glucose to the point of hypoglycemia, e.g., contrast glipizide (which maintains glycemic control) with glyburide which does not (Aspinall et al., 2011; Fravel et al., 2011). In a population of Type II diabetics (63 ± 7 years), a dose of 360 mg SDG produced no effect on fasting glucose (Pan et al., 2007). A 600 mg dose of SDG given to hypercholesterolemic patients (53 ± 12 years) produced a statistically significant decrease in fasting glucose after 6 and 8 weeks of treatment; however, no statistically significant decrease was observed when a 300 mg dose was given (Zhang et al., 2008). The reduction of blood glucose was more pronounced in those individuals with baseline fasting blood glucose levels of 5.83 mmol/L, suggesting a low risk for SDG to induce hypoglycemia in people without hyperglycemia. In those with hyperglycemia, SDG may ameliorate the high glucose levels rather than pose a risk for clinical hypoglycemia.

A limitation of our analysis is that it represents a secondary analysis of age effects. Initial participant selection only specified ≥50 years. Consequently, a preponderance of participants fell in the age category of <65 years with much fewer participants in the ≥65 years group, the group for which we wished to focus our safety analysis. Unequal group sizes can reduce the power to detect effects, but by using a Generalized Linear Model, groups with balanced data are not required for valid inferences. Furthermore, our analysis is for the average response in our study population, which can be misleading as it may fail to detect adverse effects among a small number of people. Inspection of individual level data, though, did not reveal any extreme observations masked by the average. Finally, the selection criteria for this study biased the study towards a fairly healthy population of individuals.

Conclusion

As our research question was related to the safety of flaxseed lignans in an older population, we conclude that BeneFlax at a daily dose of 543 mg does not appear to pose a risk of causing hypoglycemia or hypotension in healthy older adults aged 65 years and older. This suggests a dose of 543 mg BeneFlax falls at or below the no observable adverse effect level (NOAEL) for hypoglycemia or hypotension.

Declaration of interest

The authors report no declarations of interest. Original study (Cornish et al., 2009) was funded by the Heart & Stroke Foundation of Saskatchewan. The funding for this publication has been supported in part through a Health Research Team Grant from the Saskatchewan Health Research Foundation (SHRF).