Amen RI, Sirirat R, Oda K, et al. Effect of walnut supplementation on dietary polyphenol intake and urinary polyphenol excretion in the Walnuts and Healthy Aging Study. Nutrients. March 2, 2023;15(5):1253. doi: 10.3390/nu15051253.

Walnuts (Juglans spp., Juglandaceae) has been shown to lower blood lipids and modify inflammation and endothelial dysfunction, reducing risks of cardiovascular disease. Walnuts' reported benefits are greater than can be accounted for by their fat content. Evidence suggests that plant polyphenols in walnuts and other nuts boost antioxidant defenses and reduce inflammation. Polyphenol-rich diets increase high-density lipoprotein cholesterol and reduce low-density lipoprotein-cholesterol, blood lipids, and inflammatory markers.

Among tree nuts, walnuts have the highest levels of polyphenols. The walnut kernel and its seed coat, or pellicle, are richest in polyphenols. Walnuts' hundreds of compounds include flavonoids, flavanols, phenolic acids, hydrolyzable and condensed tannins, and lignans. In a study of dietary composition changes, patients with type 2 diabetes who ate walnuts had higher energy (Kcal), protein, total fat, and magnesium intake than those who did not, and lower intake of salt, empty calories, and dairy products. Healthy participants had similar results, with increases in ω-3 and -6 polyunsaturated fatty acids (FAs), dietary fiber, calcium, phosphorus, and zinc, and reductions in animal protein, carbohydrates, and saturated FAs. In this secondary analysis of data from the Walnuts and Healthy Aging (WAHA) study, the authors report for the first time effects of walnut intake on total polyphenols and subclass intake, and total urinary excretion of polyphenol metabolites.

The WAHA study, a two-year, dual-centered, randomized, parallel group, observer-blinded, controlled trial (RCT) of effects of supplementation of usual diets with English walnuts (J. regia) at 15% of Kcal intake (30-60 g/d) on aging outcomes in elderly participants, was conducted in Barcelona, Spain, and at Loma Linda University (LLU; Loma Luz, California) in 2014-2016. It included ambulatory men and women 63-79 years old. Exclusion criteria were inability to take neuropsychological tests; history or incidence of stroke, head trauma, or brain surgery; psychiatric illness, e.g., major depression; uncontrolled diabetes or hypertension; morbid obesity; habitual intake of tree nuts (> two servings/wk), fish oil, flax (Linum usitatissimum, Linaceae) seed oil, and/or soy (Glycine max, Fabaceae) lecithin; or allergy to walnuts.

This analysis used data from the LLU site only. Of 656 potential participants screened there, 300 did not meet criteria, and 356 were randomized to control or walnut groups. Of these, 22 were excluded at the end of the RCT due to missing 24-h dietary recall (24h DR) data, leaving 334. For this secondary analysis, 34 more were excluded for other missing data, leaving 146 in the control group and 154 in the walnut group (t = 300).

Demographic, anthropometric, dietary and lifestyle data were collected at baseline. Assays were conducted at one laboratory at study's end. Five unscheduled 24h DRs per participant (t = 1242), each including > one weekend day, were taken via telephone and face-to-face visits. DRs yielded details of spices, fats, and oils consumed, among other food groups. DRs were at diverse intervals to capture seasonal variation. Using the Nutritional Data System for Research, the authors determined nutrient totals for foods, drinks, and dietary supplements reported, corrected for portion size. Polyphenols were estimated via the Phenol-Explorer database v. 3.6. as total flavonoids, phenolic acids, flavanols, and lignans. By matching DRs, nutrient composition, and phenolic profiles, contributions of specific foods to total polyphenols and polyphenol subclasses could be estimated.

There were more women than men in the cohort, but groups were similar in gender, other demographic and anthropometric parameters, and smoking status. During the RCT, the walnut group had significantly higher Kcal, dietary fiber, total fat (P < 0.001 each), monosaturated FA (P = 0.012), total polyphenol, flavonoids, phenolic acid (P < 0.001 each), and flavanol (P = 0.036) but not lignan intakes. Among nuts reported, walnuts contributed significantly to intake of total polyphenols and all subclasses but lignans. Among food groups defined, fruits provided significantly higher total polyphenols in the walnut vs. placebo arm, as did vegetables for flavonoid (P < 0.05) and lignan (P not shown) intake. A difference in flavonoid intake from spices favored walnuts over placebo (0.1 ± 0.7 vs. 0.02 ± 0.1 mg/g; P not shown).

Values for beverages (excluding dairy milk), legumes, grains, fats and oils, chocolate (Theobroma cacao, Malvaceae), and miscellaneous foods (mixed foods, dairy products), separately calculated, showed no significantly different between-group polyphenol contributions. Fish are not named in these food groups, and it is unclear where eggs would be counted; both have non-plant polyphenols. In fasting (but not first void) urine collected at baseline and after years 1 and 2, total polyphenols were assayed. At year 1, urinary polyphenol excretion had risen to near significance (P = 0.066) in the walnut group, but not in placebo and not at year 2. At year 2, a significant negative association in the walnut group between total urinary polyphenols and the log of total dietary flavonoids (P = 0.0316) lost significance after adjustment for creatinine (P = 0.0785). There were no other associations between dietary and urinary polyphenols. Blood was collected at the same data points, but no analyses are reported.

Participants in this RCT had higher polyphenol intake than in other United States studies. Urine samples may have been insufficient in rapidly-metabolized polyphenols and did not account for polyphenols transformed by gut microbiota and later excreted in feces or urine. Future RCTs should use first morning void or 24h urine collection, according to the authors. Individual variation, poor bioavailability, and effects of food matrices on polyphenol metabolization complicate nutritional studies. Amounts and ratios of polyphenols in walnuts differ by species cultivar, and conditions of growth, storage, preparation, and extraction, e.g., use/degree of heat or defatting. Of nine authors, three received research funds from the CA Walnut Council (Folsom, CA), funder of the RCT; one received other compensation from that organization. Other authors had no conflicts of interest.

—Mariann Garner-Wizard