Walnuts, Weight Loss, and Blood Lipids

The optimal ratio of fat/carbohydrate/protein for promoting healthy weight loss is still a matter of debate. Some researchers have suggested higher fat, lower carbohydrate diets are better for weight loss in insulin resistant people. It is known that the amount and type of dietary carbohydrates and fats can alter plasma lipids in ways that may increase or decrease coronary artery disease (CAD). Weight loss and exercise can also favorably impact many known and suspected CAD risk factors including dyslipidemia and insulin resistance (IR). To better understand how diet composition affects weight loss and blood lipids, Dr. Cheryl Rock and colleagues at UC San Diego examined more than 200 overweight and obese women (mean weight 90Kg and mean age 50y) who were enrolled in a 1-year behavioral weight loss intervention program.None of the subjects had type 2 diabetes, but about half of them had elevated fasting insulin levels and homeostatic model assessment (HOMA) scores that indicated they were insulin resistant (1).All subjects were randomly assigned to one of three diet groups. All three diet groups were instructed to follow a diet with about 1500kcal (+ or - 300kcal) per day. Participants were given a detailed diet prescription and sample meal plans during individual counseling sessions. The overall goal of the dietary guidance was to reduce energy intake, “aiming for a 500- to 1,000-kcal/day deficit relative to expenditure.” Participants were also encouraged to use a web-based diet tracking system and were given a pedometer and encouraged to walk at least 10,000 steps per day (2).One group was instructed to consume a lower fat (20% energy), higher carbohydrate (65% energy) diet [LF]; a second group was instructed to consume a lower carbohydrate (45% energy), higher fat (35% energy) diet [HF-1]; and the third group was instructed to consume a similar, higher fat (35% energy), lower carbohydrate (45% energy) diet but with 1.5oz of walnuts daily [HF-2] that provided nearly half of their daily fat intake. Blood samples and data available from 213 women at baseline and at six months were the focus of the statistical analysis. Triglycerides, total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C) were checked after 6 months. The results showed serum triglycerides fell significantly in all three diet groups. The women in the walnut-rich diet (HF-2) group had their HDL-C increase a bit more than either the LF or HF-1 diet (P<0.05). The walnut-rich diet also reduced LDL-C a bit more in the insulin-sensitive women. Women in both the LF and walnut-rich HF-2 diet groups demonstrated a significant reduction in LDL-C, which tended to be greatest for the insulin-sensitive women in the walnut-rich diet group. The LF group saw a significant drop in TC but also a significant drop in HDL-C, especially for the more insulin?sensitive women (P<0.05).Insulin sensitivity increased and C-reactive protein levels dropped by a similar amount in all 3 diet groups. However, the HOMA scores of the IR subjects on the HF-2 diet fell about half as much as it did in the other two diet groups.Specific dietary instructions for the LF diet included choosing lean protein sources and reduced-fat dairy foods; for the HF-1 diet, participants were instructed to achieve a high monounsaturated fat intake. The walnut group was instructed to eat an average of 42 g (about 1.5 oz) of walnuts per day. The group weight loss intervention included weekly meetings of participants for the first 4 months, biweekly for the next 2 months, and monthly for the last 6 months.Confounding Variables Complicate Interpretation of ResultsWere the blood lipid and insulin sensitivity changes due mainly to reduced energy intake and weight loss, to increased activity, and/or to alterations in various aspects of dietary components?Another problem with the study is that the authors provide no data about to what degree the subjects complied with the 10,000 daily step goal. Still another limitation of the study was the lack of specific quantitative data regarding dietary intake and how well the subjects adhered to dietary instructions. Also, some data were self-reported, and restriction to women and the large number of exclusion criteria during screening suggests the results may not generalize to most of the population.However, one thing they did validate was compliance with the walnut intake in the HF-2 group. This was confirmed by blood fat measurements that reflected their much higher intake of omega-6 and omega-3 PUFA. Compliance appeared very good, perhaps because the walnuts were provided to only the HF-2 subjects at no cost by the California Walnut Commission, who funded the study. The other two groups were instructed not to eat nuts. Therefore, it is safe to assume that the polyunsaturated fatty acid (PUFA) content of the walnuts resulted in much greater intake of both omega-3 and omega-6 PUFA in the HF-2 group compared with the other two diet groups.Dr. Rock et. al. wrote: “The walnut-rich, higher fat diet resulted in the most favorable changes in lipid levels.” Yes, but were those results simply due to the much greater intake of omega-6 and omega-3 PUFA from walnuts? Whether the HF-2 group also consumed less saturated fat and cholesterol, and/or more fiber and plant sterols than the other 2 groups is not known but differential intakes of those other dietary factors could certainly be confound the results.Mean weight loss was 7.5% of initial weight across all 3 groups. Insulin sensitive women on the low-fat diet lost a bit more weight than their counterparts on the two lower carbohydrate diets (8.3% versus 5.4%). About 37% of insulin resistant women became insulin sensitive at 6 months, but the results were similar on all 3 diets.Women who were insulin resistant — but not those who were insulin sensitive — had a significant reduction in triglyceride levels at 6 months in all 3 groups. “These findings are consistent with current understanding of insulin resistance, in which the impaired ability to manage glucose leads to a higher level of triglycerides in response to carbohydrate consumption,” Rock and colleagues wrote. Actually not true, as her data showed serum triglyceride levels fell by 15mg/dl in both the HF-2 and LF groups.Bottom Line: Assuming the subjects ate the 1.5oz of English Walnuts daily (which is likely because they got them for free) their intake of omega-6 and omega-3 PUFA would have been much higher than that of both the low-fat and higher MUFA groups. At 1500kcal, the 35% fat diet would have about 525kcal from fat. So nearly half (about 250kcal) would come from the 1.5 oz of walnuts. So the walnut group consumed far more omega-3 and omega-6 PUFAs than either the other high-fat diet (where they were instructed to consume mostly MUFA) and also much more omega-3 and omega-6 PUFA than the LF diet group. If we assume that all 3 diets had similar amounts of SFA, cholesterol, fiber, plant sterols, etc. then the higher PUFA intake in the HF-2 (walnut) group alone would have predictably produced the somewhat more favorable changes in blood lipids observed. Odds are Dr. Rock knew that and designed the study to show a more favorable blood lipid effects from consuming walnuts. However, the failure of both higher fat diets to more favorably impact weight loss or improve insulin sensitivity was likely disappointing for California’s walnut growers. And while the improvement in blood lipids was largely the result of a modest increase in HDL-C, whether or not this will translate into a reduced CAD is questionable given growing evidence that higher fat diets may increase HDL-C levels at least in part by increasing dysfunctional HDL particles in the blood.By James J. Kenney, Phd, FACNReferences:

1. Turner RC, Holman RR, Matthews D, Hockaday T. et al. Insulin deficiency and insulin resistance interaction in diabetes: estimation of their relative contribution by feedback analysis from basal plasma insulin and glucose concentrations.”. Metabolism 1979;28 (11): 1086–96. doi:10.1016/0026-0495(79)90146-X. PMID 386029.
2. http://m.jaha.ahajournals.org/content/5/1/e002771.full.pdf.