Weight, insulin resistance, blood lipids, and diet quality changes associated with ketogenic and ultra low-fat dietary patterns: a secondary analysis of the DIETFITS randomized clinical trial

Abstract

Background: The DIETFITS trial reported no significant difference in 12-month weight loss between a healthy low-fat and healthy low-carbohydrate diet. Participants were instructed to restrict fat or carbohydrates to levels consistent with a ketogenic or ultra low-fat diet for 2 months and to subsequently increase intakes until they achieved a comfortable maintenance level. Objective: To compare 3- and 12-month changes in body weight and cardiometabolic risk factors between a subsample of participants who reported 3-month fat or carbohydrates intakes consistent with either a ketogenic-like diet (KLD) or ultra low-fat diet (ULF). Design: 3-month and 12-month weight and risk factor outcomes were compared between KLD (n = 18) and ULF (n = 21) sub-groups of DIETFITS participants (selected from n = 609, healthy overweight/obese, aged 18–50 years). Results: Less than 10% of DIETFITS participants met KLD or ULF criteria at 3-months. Both groups achieved similar weight loss and insulin resistance improvements at 3-months and maintained them at 12- months. Significant differences at 3-months included a transient ~12% increase in LDL cholesterol (LDL-C) for KLD with a concomitant greater reduction in log(TG/HDL), a measure of LDL-C’s atherogenic potential. The latter was maintained at 12-months, despite substantial diet recidivism for both groups, whereas LDL-C levels were similar for ULF at baseline and 12-months. KLD participants achieved and maintained the greatest reductions in added sugars and refined grains at 3- months and 12-months, whereas ULF participants reported a 50% increase in refined grains intake from baseline to 12-months. Conclusion: Among the ~10% of study participants that achieved the most extreme restriction of dietary fat vs. carbohydrate after 3 months, weight loss and improvement in insulin sensitivity were substantial and similar between groups. At 12 months, after considerable dietary recidivism, the few significant differences in diet quality and blood lipid parameters tended to favor KLD over ULF.