Well, I've completed the second fasting cycle and now I have some data to share. I weighed myself each morning using a digital scale that uses bioelectrical impedance analysis to measure body fat. The method is not very accurate, but is very consistent - so changes probably mean more than absolute fat percentage.
As a reminder, the diet is a variation on the now popular 5:2 diet, or periodic calorie restriction. The difference is that this one restricts both calories and protein for 5 days in a row, with the remainder of the month eating "ad libitum" - whatever you like. Details and a summary of the original research are given in my previous post. The study showed substantial benefits across a range of different health measures.
The time series starts at the beginning of the previous feed period. I've multiplied total weight by fat percentage to give fat mass (lower series, right hand axis) and (100 - fat percentage) by total mass to give lean mass (upper series, left hand axis). The blue and green data are during the feeding period, pink data are during the fasting (or fasting-mimicking) period.
The first thing to notice is that there is a rapid and sustained loss of both fat and lean tissue during the fasting period - from the data, perhaps 300g per day of each. Assuming that lean tissue loss has about the same calorific value as lean meat, about 100 Cal per 100g, that means I'm deriving perhaps 300 Cal per day by catabolising lean tissue. Fat has around 900 Cal per 100g, so I'm getting about 2700 Cal per day by burning fat stores. If I have the food intake about right, there is another 700 Cal from food each day. This works out 3700 Cal per day, somewhat higher than the average adult male energy requirement (2400 Cal per day), so either I'm missing something, or I have a higher than average energy requirement. This might have something to do with having a relatively high surface area, small volume (I'm tall and lightly built) and it being a bit cold (late winter). ie, I'm using a lot of food calories just to stay warm. Another possible reason is that the scale is over-estimating my body fat percentage and the fat mass reduction is not as large as indicated.
A commenter asked about lean muscle loss last time - and we can see lean mass loss in the graph. The mice in the study linked had a loss of organ weight during fasting, notably kidneys, liver and heart, but not brain, spleen or lungs. It would seem likely that at least part of the mass loss in the graph above is due to this. Part is probably lean muscle tissue.
The next thing to notice is the recovery phase preceding the fast. I least-squares fit a simple model to the fat and the lean mass data for the feeding period only. Expecting that change would be most rapid at the start and tail off towards an equilibrium at the end, I fit an exponential decay curve like
where m is mass (either fat, or lean), a, b and c are constants and t is time (day number).
The values of the parameters probably aren't interesting, but you can see from the graph that the lean mass recovers much, much faster than fat mass. Lean tissue recovery seems to be complete within a week, whereas fat mass is still rising when the next fasting cycle hits. This makes intuitive sense, as lean tissue would be the main priority for our bodies, fat is to a large extent energy storage and perhaps a second priority. It wouldn't make sense to store fat in preference to repairing muscle and internal organs that have been stressed by starvation. This observations could also explain the trend to visceral fat reduction that didn't quite reach significance in the research study after three diet cycles: from the above, it looks like we force a large reduction in fat during the fasting cycle, which the body isn't quite able to make up for during the feed cycle that follows (on the other hand, the same appears not to be true for lean tissue). So we might expect to see an overall reduction in body fat (hopefully, mainly visceral fat) over a number of feed / fast cycles. In addition to the other benefits (lower levels of IGF-1 etc.) which are a bit harder to measure at home!
Anyhow, we shall see. Stay tuned for more data and analysis as they come in.