True or False? 'Low Fat' for the Lean, 'Low Carb' for the Obese and Insulin Resistant - Pilot Study Confirms Often Heard Dieting Advise on a Surface Level , However, ...

Low carb, or fat? Left or right?  Which one should you chose and why? Shall you go by your body weight, your insulin sensitivity or your personal food preferences?

"Low Fat for the Lean, Low Carb for the Obese and Insulin Resistant," this quote from the headline sounds like a reasonable advise if you look at the existing evidence on low carbohydrate dieting, which appears to excel whenever the subjects are significantly overweight and insulin resistant. Studies that would do a head-to-head comparison of the two to confirm the accuracy of the hypothesis that "because they are insulin resistant, avoiding carbohydrates will aid people with (pre-)diabetes in losing weight" are non-existent... well, I should probably say they "were" nonexistent; a recent pilot study by Gardner et al does after all just that: compare the weight loss response of insulin sensitive vs. resistant individuals who consume either a low fat or a low carbohydrate diet over 6 months.
The results of Gardner's study have recently been published ahead of print in the venerable scientific journal  Obesity; and they are... interesting, but as it was to be expected for a pilot study with "only" 61 participants in four groups, more research will be needed to make definite conclusions.

Figure 1: Weight loss (kg) after 6 months on the respective in insulin resistant and sensitive subjects (Gardner. 2015).

If you look at the main study outcome, i.e. the weight loss in Figure 1, for example, it would appear as if the previously cited statement from the headline of today's SuppVersity article was confirmed. Due to the large intra-group (=between individuals in one group) differences in all four diet groups (see error bars in Figure 1, they are about as long as the mean difference, we may speak of a trend or a tendency, that appears to confirm the previously stated hypothesis that an insulin resistant individual is better off avoiding carbs when dieting while an insulin sensitive one should stick to the mainstream low-fat recommendations (remember all subjects were overweight, none of the was athletic).

Figure 2: Proportions of carbohydrates, fats, and proteins for each diet at baseline, 3 months, and 6 months (Gardner. 2015).

This wouldn't be a SuppVersity article, though, if a brief glimpse at the main result was everything it had to offer. Let's first take a look at the reported energy- and macronutrient intakes (Figure 2) of the healthy, premenopausal women and men (age 18-50) with stable (>2 months) BMIs and an age between 28 years and 40 years (aside from the increased number of subjects with metabolic syndrome in the insulin resistant group, there were no noteworthy inter-group differences at baseline).
As you can see significant inter-group differences were observed only for the macronutrient composition; and even though these differences were significant, we are nor talking about a ketogenic vs. no-fat diet. Rather than that, both diets had a relatively balanced macronutrient profile, albeit with different main energy sources (carbs vs. fat). What did not differ for the groups, though was the total energy intake (Figure 2). This is interesting, because, in theory, the subjects were allowed ot eat as much as they wanted; and still, the data in Figure 2 tells you that they restricted their energy intake by noteworthy 30% on average (the ~600kcal-deficit also explains why the subjects even lost weight), with no difference between the subjects on the allegedly more satiating low carb vs. low fat diet.

Figure 3: Fiber, added sugar and saturated fat intake in g/1000kcal (Gardner. 2015).

Visible differences existed, obviously, for the intake of fiber, added sugars and saturated fats (Figure 3). These differences, which are characteristic of low fat vs. low carb diets, may also be the reason a recent study by Mansoor et al. (see red box below) found differences in the effects on triglycerides (increases with high CHO and even more so sugar intake), HDL and LDL (both increase w/ high saturated fat intakes) when they compiled the results of the contemporary low fat vs. low carb trials.
It is thus no wonder that the data from  blood analysis of the study at hand (Figure 4) mirrors the results of the studies Mansoor et al. (2015) analyzed for their meta-analysis that is about to be published in the British Journal of Nutrition (see red box):

Figure 4: Changes in LDL-C, HDL-C, Trigs and fasting glucose after 3 & 6 months (Gardner. 2015).

With the most significant differences being observed for triglycerides and LDL, the situation is very similar to the one the Mansoor et al. describe in their review. In this regard, it is also worth mentioning that the differences between the groups were reduced, when the subjects started to increase their fat or carbohydrate intake by 5g on a 1-4 week basis to eventually end up at their individual "that's how I could eat for the rest of my life"-level (compare the 3 months with the 6 months data).

This doesn't solve the dilemma we're in, though: without further data on particle sizes and oxidative status of the LDL molecules, etc. it is virtually impossible to make a reliable prediction which of the two diets is going to have the higher long-term health benefits. What may be even more important, though, is that we must not forget that it is not debatable that both diets triggered significant weight loss and measurable health improvements, especially in the subjects with pre-existing insulin resistance (blue and orange bars in Figure 4).
References:

• Gardner, et al. "Weight Loss on Low-Fat vs. Low-Carbohydrate Diets by Insulin Resistance Status Among Overweight Adults and Adults with Obesity: A Randomized Pilot Trial" Obesity (2015): Ahead of print.
• Mansoor, et al. "Effects of low-carbohydrate diets v. low-fat diets on body weight and cardiovascular risk factors: a meta-analysis of randomised controlled trials." British Journal of Nutrition (2015): First view article.