Until recently I had not come across the term “chronotherapy”, though I can work out that it has something to do with the time you take your medicines, or as wikipedia puts it “treatment scheduling”. So the question I am considering in this post is whether, as patients with chronic kidney disease (CKD) we should be taking some of our antihypertensives (blood pressure medication) at bedtime rather than all at once in the morning.
Blood pressure varies during the 24 hour period, termed “circadian variability”. Surprisingly we still don’t know the exact physiology behind this but one of the factors is probably that while we sleep the body is less alert or ready for flight or flight – reduced activity of the sympathetic nervous system. So during the night a healthy person’s BP will “dip” below the daytime average by around 10-20%. In the papers these people are referred to as “dippers” and being a dipper is generally a good thing – it is linked to lower risk for heart disease and blood vessel problems, lower risks of stroke and death.
However, people with chronic kidney disease and hypertension, of any sort, not just PKD, tend not to have this drop in night-time BP and may be referred to as “non-dippers”. One of the theories to explain this is to do with salt excretion by the kidneys. A struggling kidney cannot excrete sodium so well and so needs a higher pressure in the glomerulus (the filter system) during the night to force the excess sodium out into the urine, thus blood pressure stays higher for longer during the night before it dips, or, eventually, doesn’t dip at all. I will talk about salt another time, but it is important to know that BP in CKD tends to be salt-sensitive: lowering salt intake can lower BP and enhance the effects of the medication, while conversely a high salt diet can offset the efficiency of diuretics and other antihypertensives.
Why does it matter if you have high BP in your sleep? Several studies have shown that a non-dipping BP during the night relates to an increased cardiovascular risk – of heart attacks, strokes and ischaemic events. One showed that a 10mmHg increase in mean night-time systolic BP was associated with a 21% increase in cardiovascular mortality.
I first read the paper by Hermida et al in 2011, shortly after it was published. In a nutshell, this study seems to say that bedtime dosing is associated with reduced cardiovascular risk and better control of 24-hour blood pressure. A study from the University of Vigo, Spain, they took over 600 patients with high blood pressure and chronic kidney disease, excluding diabetics, shift workers and some other co-morbidities. These were randomly allocated to one group where the usual morning dosing schedule was maintained (conventional morning time therapy, CMTT) or the alternative where one or more of the BP medications was taken at bedtime (bedtime hypertension chronotherapy, BTCT) and the follow-up was for an average of 5.4 years (range of 6 months to 8 years). Measurements of ambulatory BP (24hrBP) were taken at the start of the trial and every year, and whenever changes in doses or tablets were made. An important feature of the study was that the investigators were blinded to the time of day medication was taken. The outcome was that patients in the BTCT group had an adjusted risk of cardiovascular disease that was one third that of those in the CMTT group.
This figure bares closer scrutiny. Because the numbers of any one type of event (death/angina/heart failure/heart attack/stroke/arterial occlusion) were small they were grouped together to obtain a composite figure representing risk. They report this as an adjusted risk of 0.31 (95% confidence interval,range 0.21-0.46, P<0.001). Now I never understood statistics at medical school, not until I had to teach it to students did I grasp even the basics, so my explanation will be lacking for any statistician, but that result can be called “highly significant”. The confidence interval is the upper and lower boundaries of the likely true effect – here the range is narrow so the estimate of the true effect is precise. The p value indicates that, if there was actually no difference then the chance of getting the observed results would have been less than 1 in 1000. So the bottom line, statistically, is that the study has reasonable power to detect the effect it did.
BUT statistical significance is not the same as clinical significance. To ascertain that you also need to know how well these findings relate to other patients, real life scenarios and whether other risks might outweigh the apparent benefits. Actually the same group of researchers have previously demonstrated similar statistical results and this paper was intended to demonstrate clinical significance. They wanted to show that the effect on nocturnal BP translated into outcomes, longer survival, fewer events.
One pictorial way they present their results is called the Kaplan-Meier curve. This tries to answer the patient’s question “If I make this change, how long will I survive before I experience an event?”. The graph has two lines – one depicts event-free survival times if no intervention is made (the lower curve) and the other shows the probability of event-free survival if you do make the change. That these lines are far apart is suggestive that there is a significant difference on the two groups of patients – if you take one or more of your BP tablets at bedtime, the chances are you will have more years before you suffer from an event such as a heart attack.
The paper more or less convinced me that I should change the timing of my medication, but further reading persuaded me that there was insufficient information on the potential adverse effects of doing this to make the change.
What other things could happen if you take antihypertensives at night? The potential problems of nighttime dosing include:
- not all medications are suited to night-time dosing, diuretics taken at bedtime could make for a most uncomfortable and disturbed sleep pattern
- a fall in BP at night has been associated with glaucoma
- the optic nerve is at risk from ischaemia from low BP at night
- some strokes may be related to falls in BP at night
So, 4 years ago, after considering those aspects, I decided not to change the timing of my antihypertensives. But recently a review of hypertension management in CKD was published that encourages physicians to consider changing the timing of BP treatments. Note the word “consider”.
What has changed? Additional studies with more patients, meta-analyses of related trials and the increased focus on chronobiology have all added information to feed the decision.
In the end, however, the decision is never going to be a clearcut “this is a good thing”, individual patients need to consider the risks for themselves, potential adverse effects and any other medications they may take. There is increasing evidence that using ambulatory (24-hour) BP measurements is helpful in managing hypertension in CKD and that would be a starting point for the discussion with a renal specialist about when to take medications.
Ben-Dov IZ, Kark JD, Ben-Ishay D, Mekler J, Ben-Arie L, Bursztyn M. Predictors of all-cause mortality in clinical ambulatory monitoring: unique aspects of blood pressure during sleep. Hypertension. 2007;49(6):1235–1241.
Carter BL, Chrischilles EA, Rosenthal G, Gryzlak BM, Eisenstein EL, Vander Weg MW. Efficacy and safety of nighttime dosing of antihypertensives: review of the literature and design of a pragmatic clinical trial. J Clin Hypertens (Greenwich). 2014;16:115-121.
Crespo JJ, Pineiro L, Otero A, et al. Administration-time-dependent effects of hypertension treatment on ambulatory blood pressure in patients with chronic kidney disease. Chronobiol Int. 2013;30(1–2):159–175
Hermida RC, Ayala DE, Mojon A, Fernandez JR. Bedtime dosing of antihypertensive medications reduces cardiovascular risk in CKD. J Am Soc Nephrol. 2011;22(12):2313–2321.
Mallick SR, Rahman M. Nocturnal medications dosing: does it really make a difference in blood pressure control among patients with chronic kidney disease? Curr Hypertens Rep. 2012;14:449-454.
Ohkubo T, Hozawa A, Nagai K, Kikuya M, Tsuji I, Ito S, Satoh H, Hisamichi S, Imai Y. Prediction of stroke by ambulatory blood pressure monitoring versus screening blood pressure measurements in a general population: the Ohasama study. J Hypertens. 2000 Jul; 18(7):847-54.
(My blog has the footer stating it is NOT medical advice and I want to highlight that now – I am NOT making any medical recommendations to anybody. These decisions need to be made alongside a personal physician. I am no longer working clinically and do not advise individuals. )