Intermittent Fasting

It has been known for a long time that caloric restriction (a significant reduction in food intake – to 60-70% of normal levels) prolongs life,and improves health in many species (even including yeast!). Many people are interested in caloric restriction (for example, a caloric restriction support group on Yahoo! has 1,687 members) but relatively few people are willing to actually follow such an extreme diet. However, evidence that the benefits of caloric restriction result from the activation of a genetic program (involving insulin signaling and the gene SIRT1) rather than reduced metabolic activity per se has opened up the possibility that benefits can be achieved by other means. The current issue of Cell (February 25, 2005) is devoted to reviews on aging and includes an article by Lenny Guarente (PubMed) that makes this point in particular. One hope is finding a drug or dietary supplement that will do the job (perhaps resveratrol). I have paid more attention to the possibility that the same genetic program may be induced by intermittent fasting.

At the beginning of May last year (2004) I started a diet that involves not eating on Mondays, Wednesdays and Fridays. My diet was inspired in particular by a study (Anson et al., 2003) that reported beneficial effects on glucose metabolism and neuronal resistance to injury without an overall reduction in calorie intake. In this study, mice were assigned to four groups (ad libitum, intermittent fasting, caloric restriction and pair fed; this last group was given as much food as the intermittent fasting group ate, but on a daily basis). The intermittent fasting group did as well as the caloric restriction group on a variety of tests but enjoyed almost as much food as the ad libitum group (by making up the difference on days when food was available). A very recent paper by Hsieh et al. 2005 (Effects of Caloric Restriction on Cell Proliferation in Several Tissues in Mice: Role of Intermittent Feeding: PubMed; Am J Physiol Endocrinol Metab) reaches similar conclusions studying cell proliferation.

I fast between dinner of one day and dinner the next day, but eat dinner every day. The fast consists of no calories at all (I allow myself water, tea and coffee), and the precise duration depends on when we have dinner, according to the details of each day's schedule. While I normally fast three times each week, I am flexible. For example, I might allow myself to schedule lunch on a Friday after having fasted on Monday and Wednesday. In fact, I've done this only a few times since starting the diet. One week I postponed Friday's fast so that I fasted on Monday, Wednesday and Saturday that week.

The health benefits of intermittent fasting in humans are certainly not established, and I'm not recommending this diet for anyone else. I find it likely that caloric restriction works in humans because it works in so many other species. However, it is possible that the genetic program is already constitutive in humans. It is also possible that the particular diet I have selected fails to induce the pathway. Perhaps longer but less frequent fasts would be more effective, or many fewer fasts would be sufficient. I don't know, but I am encouraged by the growing evidence that the induction of extreme longevity can be uncoupled from caloric restriction per se.

I don't find this diet to be terribly difficult. In fact, skipping meals can be convenient (I'm busy, after all). For about three weeks after starting the diet I would experience mild hypoglycemic symptoms on the fast days (I was nervous but tired, just a little irritable and not terribly productive) and became sleepy when I ate again. These symptoms have passed. I now feel pretty much the same on fast days and other days. This diet is not about losing weight. I lost about 10 pounds over the first six weeks and then stabilized (and I am starting to gain some of this back). There are side effects. In particular, I don't think that it's a good thing to routinely drink strong coffee on an empty stomach (which is what I'm doing three days each week). I am aware that this diet is the opposite of the standard "healthy eating" advice that advocates a big breakfast and regular meals, and I must confess that I have passed up opportunities to discuss it with my doctor (although I have mentioned it to friends who are doctors). On the other hand, fasts of various kinds are a part of almost all religious traditions. I suspect that a study with direct tests of the relevant biomarkers (reduced serum glucose and insulin levels, or even elevated expression of SIRT) is possible, and will eventually be carried out, but I've decided not to wait for it. In the meantime, I can only quote the endorsement of my friend Peter Roemer, who was responding to an article in "The Economist", (March 31, 2005: "Even a slight decrease in calories may lead to longer lifespans") about the Hsieh et al. article cited above:

"OK, I'm giving it a try. Of course, this is exactly what Steve has been doing for a while. And look, he's still living!!!"

This also appears on SteveMount.com as Posting 2.