Wednesday, January 23, 2013

90 Min Sleep Restriction - How Bad is It Really? Changes in Insulin Resistance Last For One Week - Until the System Adapts. Plus: Subjects Even Reduce Weight & Waistline

In general it's certainly a good idea to rise, when the cock crows on the dung heap, but what if that means you miss 90 min of your precious sleep?
"Tell me how and how long you slept last night and I'll tell you something about your chances of getting / staying lean and healthy!" It sounds more straight forward than the stupid AM vs. PM system you, my American friends, are still sticking to: If you consistently miss just one hour of sleep, you are messing with your insulin sensitivity, body weight and your plasma concentrations of leptin. But is this actually accurate and "how much messing" does it actually take? A close analysis of a recent study by a group of researchers from University of Surrey in Guildford, UK (Robertson. 2013), does at least raise some questions in how far smaller deviations from your usual 24/7 x 365 consistency will mess with your physique and health.

Can you "wake your way" towards obesity?

"Mild sleep restriction" is what scientists call it, when you miss 'only' one hour of your regular 7-8h of sleep every night and "mild sleep restriction" is also what the 19 healthy young normalweight men in the Robertson study have been exposed to over a three-week period.
Just in case you have forgotten about it, you can do the Munich Chronotype test for free (including personal evaluation) right at the website of the University of Munich (click here to be redirected)
Healthy male students, aged 20–30years, BMI 19–26kg/m² were recruited for this randomized–controlled sleep intervention study [...] those with a self-reported sleep length of 7.0–7.5h were invited for more detailed screening.[...] According to self-report, volunteers were not taking prescription or over the counter medication and had a stable weight for >3months. Specific exclusion criteria also included (i) shift-work and travel beyond 2 time zones in the preceding 2months, (ii) high intake of caffeine and alcohol, (iii) extreme morning or evening preference assessed with the Munich Chronotype and Horne Ă–stberg Questionnaire, (iv) a self-reported sleep problem (Pittsburgh Sleep Quality Index global score≥6) and (v) daytime naps in the preceding 4weeks.
Before the actual study began, the habitual sleep patterns were assessed by actigraphy and sleep diary information collected during a two week baseline period. The subjects were then  randomized to either the
  • restricted group spending their regular time-in-bed minus 1.5h per night, or the
  • habitual group who simply followed their habitual sleeping patterns
The participants had to fill out sleep logs in order to assure compliance. Moreover, participants were instructed not to take naps during the day . The actual acquisition of the weight, blood pressure and adiposity data, as well as the leptin and adiponectin levels, and information about  arterial stiffness took place after a 12h overnight fast, at the Centre for Endocrinology Diabetes and Research (the oral glucose tolerance used to evaluate the effects of sleep deprivation on insulin sensitivity took place at the CEDAR, as well). The procedure was conducted four times, initially following the 2-week baseline period and in weekly intervals for the 3 weeks of the intervention, thereafter.
Figure 1: Relative changes of insulin sensitivity (HOMA-IR), fasting blood glucose and adiponectin levels in the course of the three-week study period (Robertson. 2013)
If you simply went by the absolute data and the short but concise summary of the main results the researchers provide in the abstract it would in fact seem as if the provocative statement I at the very beginning of this article would be spot on and you would add a pound of fat to your frame and lose a full week of your precious life-time with each minute of sleep you are missing. If you put things into perspective by calculating the relative changes, you cannot but realize that sleeping 90min less may not be ideal, but it's "medium term"  effects are probably overrated.
Figure 2: Illustrated and annotated version of the graph depicting the change in insulin sensitivity in the habitual (open circles) and sleep restricted (closed circles) group (Robertson. 2013)
In fact, the plot of the changes in insulin sensitivity in figure 2 clearly suggests that there is something like an adaptation process that compensates for the initial reduction in insulin sensitivity once a new 'sleep steady state' is achieved. In the absence of data from the following weeks, we cannot however exclude that the in an even longer-term scenario the reduction of sleep duration from below the 5h mark would not have had a negative impact on the overall and metabolic health of the subjects. In this context, Robertson et al. point out that
"[...] some of the effects of sleep restriction appear transient, i.e. we have beenunable to confirm a change in insulin sensitivity beyond one week of sleep restriction [and that t]he data clearly indicate that the effects of a reduction in sleep duration may change in the course of the exposure to sleep reduction, [which renders any] extrapolation from single visit laboratory studies to epidemiological data problematic." (Roberston. 2013)
If we would yet discard the epidemiological findings and simply ignore the constant increase in fasting blood glucose and rely on the fact that the insulin tolerance returned to baseline after an initial drop, we could even (ab-)use the data from the study at hand to argue in favor of sleep "deprivation" (I am deliberately using quotation marks, because the previously mentioned adapation process would suggest that the subjects were not actually "deprived") as a means to reduce the obesity epidemic. After all the subjects in the sleep restricted group did lose some body weight and reduced their adiposity index, the waist equivalent to the BMI (higher values = higher ratio of waist to body height), by meager, but at least measurable 0.8%.
Believe me: Fridge raiding is your least problem if you don't sleep at night. If you want to know more about the profound metabolic, psychological and even carcinogenic effects of having or not having a regular sleeping rhythm, why a dark room is important and more, read up on the Circadian Rhythm Series
Sleep does not necessary equal sleep: It should also be mentioned that "sleep" data we are talking about is actually data spent in bed and if you take a look at the sleep latency, you will realize that the subjects in the restricted group made up for 4-8 minutes of their 90min sleep debt by simply falling asleep more readily. Another factor to keep in mind is that, at least in my humble opinion, the actual sleep duration of the young men who participated in the study was to low to begin with. This is just my personal N=1 experience, but whenever the demands of life (or my own stupidity) forced me onto a <7h sleeping schedule, I ended up gaining body fat - not necessarily weight, but fat. So if there is something like a "break even" and the subjects were below that already, the effects could be much more pronounced in someone like you, who is (hopefully) getting is 7h of sleep in a pitch black, quiet room every night.
And what's more, in the correlation analysis the scientists conducted, the only parameter with a statistically singnificant correlation with sleep duration was - you won't guess that - body weight! With a positive correlation of +0.271 indicating that shorter sleep durations exhibited - at least in the study at hand - statistically significant correlations with decreases in body weight.

So what? You cannot 'wake yourself lean', can you?

No, probably not. The evidence from epidemiological studies is too overwhelming to throw the insufficient sleep = diabesity hypothesis over board. On the other hand, the results of the study at hand are interesting, as they clearly suggest that (a) our bodies are able to adapt to a news steady state within a certain window of suboptimal to optimal sleeping time and (b) that the short term effects of what I would like to identify as an actue stressor, namely 1-2 weeks of reduced sleep will lead to similarly transient weight loss.

Previous study showed: When you are dieting 40min extra sleep can help you 1lbs extra fat (read more).
increases in cortisol levels do the exact same thing. In the short run, they will spike you up (Lance Armstrong has admitted using corticosteroids to outperform the competition only days ago). Used intermediately, they will begin to gnaw on your body fat and muscles and in the long run they will totally mess up your metabolism (this assumes chronic low level exposure). To cut a long story short: The study at hand clearly suggests that someone without pre-existing health problems, a young man or woman in the prime of your life, won't risk becoming diabetic during the exam preparations or during any other period in his / her life, where he / she has to reduce the habitual and natural sleep durations by 90min for a <1 month in order to cope with external circumstances he or she has no bearing on.

Bottom line: Now, does this mean that we have been overrating the importance of sleep hygiene before? Certainly not. What we may have underestimated, however, are short-term compensatory effects which allow for a brief, but non-negligible time window, in which neither the acute very short-term effects - nor the chronic (very) long-term effects of sleep deprivation will harm you. And though we don't know anything about the size of this window, one thing appears to be certain: It's large enough for exam preparations, but not for years of drinking, partying and doing all sorts of things but sleeping ;-)

  • Robertson MD, Russell-Jones D, Umpleby AM, Dijk DJ. Effects of three weeks of mild sleep restriction implemented in the home environment on multiple metabolic and endocrine markers in healthy young men. Metabolism. 2012 Sep 15.