NephJC Shorts: BedMed and Chronotherapy

Why was this study done? 

Chronotherapy - i.e. using the circadian rhythms to time the drug administration, has been tested in hypertension, specifically to help with nocturnal hypertension which is associated with cardiovascular outcomes. There are some notable (or infamous) Spanish trials, MAPEC and HYGIA, which have reported implausibly high benefits with bedtime dosing of BP lowering medications. This does not make much sense, as the magnitude of the benefit is larger than any BP medication trial (e.g. HOPE with ramipril) and also because the drugs we use now are quite long acting and time of day dosing should not matter hugely. We have litigated this issue in the past (see NephJC coverage; Turgeon et al, Hypertension 2021) based on critical appraisal and pharmacology. But the real test is to reproduce these interventions in an RCT. BedMed was designed in this environment. 

How was the trial done, and what did it show? 

BedMed was a publicly funded trial with a similar design as the Spanish trials. It was a pragmatic trial, recruiting people with hypertension on one or more BP lowering medication, from primary practices across 5 Canadian provinces. Only those with sleep-disrupting shift work, those with glaucoma, and those living in continuing care facilities were excluded. The last group were instead enrolled in the sister BedMed Frail trial (Garrison et al, JAMA NO 2025).  The intervention was bedtime dosing of all BP meds, compared to taking meds when arising. The research team recommended one timing change per week until all once-daily antihypertensives were per allocation. The bedtime group took BP medication when readying for bed; the morning group took BP medication on arising. The primary outcome was a composite of all-cause mortality and MACE. The trial was powered for a very conservative (compared to the outrageous odds reported from Spain) of a 255 relative risk reduction, needing 406 events. 

Over 5 years, 5073 individuals were screened, 3357 were randomized to bedtime (n = 1677) vs morning (n = 1680) antihypertensive use. Notably, 56% of the participants were female, median age was 67 years old; 18% had diabetes, 11% had known coronary artery disease, 24% had sleep apnea, and 7% had CKD. Just over half the participants were only on 1 medication, and ~ 2% were on more than 4 medications. About 3% were lost to all follow up, and though ~ 15% dropped out, they could be followed in the administrative databases for outcomes. RASi (~ 66%) followed by CCBs (~ 28%), followed by diuretics (~ 27%) were the commonly used agents. As expected, over a median 4.6 years, the primary outcome event rate was 2.3 per 100 patient-years in the bedtime group and 2.4 per 100 patient-years in the morning group (adjusted HR, 0.96; 95% CI, 0.77-1.19; p = 0.70). 

Figure 2 from Garrison et al, JAMA 2025.  Effect of Medication Timing on Cardiovascular Events and Death

There was no difference in any of the components of the composite outcome, or the secondary outcomes, or the safety outcomes. Notably, glaucoma was not worse with bedtime dosing (nocturnal hypotension has been associated with glaucoma progression, e.g. Charlson et al, Ophthalmology 2014). The subgroups showed no signals. Adherence was pretty high with one or more once-daily antihypertensive drugs taken at the allocated time by 88% of bedtime and 97% of morning participants. A 24 hour ABPM was performed in a subset of participants (302 of the 3372) and did demonstrate a lower nighttime BP with bedtime dowing  (SBP 116.5 mm Hg vs 123.9 mm Hg; difference, −7.4 mm Hg; p < .001 and DBP 62.9 mm Hg vs 65.5 mm Hg; difference, −2.7 mm Hg; p = .02).

What are the implications? 

This is, hopefully, the final nail in the chronotherapy coffin. This trial follows on the TIME trial (Mackenzie et al, Lancet 2022), with over 21,000 patients and demonstrated no benefit with nighttime dosing as well (unadjusted HR 0.95, 95% CI 0.83–1·10; p=0.53). TIME did report 3.6% higher excessive visits to the toilet during the day or night for those on nighttime dosing. These two trials otherwise clearly demonstrate that the nighttime/bedtime taking of BP medications has no special benefit. The accompanying BedMed Frail trial, in 776 participants living in continuing care facilities, reported very similar results (adjusted HR, 0.88; 95% CI, 0.71-1.11; P = .28). This is also the classic setting when a software driven systematic review will be classic GIGO (see Kuang et al, Am J CV Drugs 2025 as an example). If you find the Spanish studies dragging your pooled outcome into heterogeneous territory, doing silly subgroup analyses is not the answer. Read the robust RCT rather than the silly SR. 

What should we make of the lower nighttime BP seen in BEDMED and the lack of benefit despite a 7.4/2.7 mm Hg lower nocturnal BP? Similar to this, in TIME, mean morning-assessed BP (home, not ABPM) was lower in the evening dosing group than in the morning dosing group (SBP difference of 1.8 mm Hg lower [p<0.0001]; DBP difference of 0·4 mm Hg lower [p=0.023]). On the other hand, conversely, evening-assessed BP was lower in the morning dosing group than in the evening dosing group (SBP difference of 1.1 mm Hg [p<0.0001]; DBP difference of 0.9 mm Hg [p<0.0001]). This might account for some of the wash in terms of clinical benefits. It also might be true that nocturnal hypertension is an epiphenomenon, and the association of the non-dipping pattern with bad CV outcomes may be from the causes of non-dipping (e.g. diabetes, CKD, volume overload, sleep apnea, sympathetic activity) and that changing the BP number without changing the underlying cause doesn't help. Lastly, it is also possible that there might be a benefit due to lower nocturnal BP, masked by a counterbalanced harm from a lower BP and hypoperfusion, which thus provides a neutral effect. 

In conclusion, medications should be taken when they can be best remembered to be taken (adherence is a huge problem in hypertension: eg ~ 35% of those with resistant hypertension may be non-adherent, Bourque et al, Am J Hypertension 2023). For more, watch a fun discussion with the TIME and BedMed authors here.

Summary by Swapnil Hiremath
Reviewed by Cristina Popa