New Treatment for blood pressure: Baroreflex activation therapy (BAT)

Baroreflex activation therapy (BAT) is a new method to treat patients with advanced high blood pressure. It works by electrically stimulating the carotid sinus, leading to reduced sympathetic tone.

Impressively systolic BP dropped on average by 35 mmHg and diastolic BP by 17 mmHg. About 25% of patients reduced the number of anti-hypertensive medications by half. Full benefits were seen within the first 6 months of procedure, but more importantly results were maintained in 6 years.

I believe more research is needed to approve BAT for mainstream clinical use, given the procedural nature of treatment.




May 2017

Baroreflex activation therapy is a novel technique for treating patients with resistant hypertension. Although short-term studies have demonstrated that it lowers blood pressure, long-term results have not yet been reported.

The aim of the present study is to assess the long-term efficacy and safety of baroreflex activation therapy. Long-term follow-up data were analyzed from all patients who had been included in 1 of the 3 trials that focused on treatment-resistant hypertensive patients.

Altogether, 383 patients were available for analysis: 143 of these had completed 5 years of follow-up and 48 patients had completed 6 years of follow-up. In the entire cohort, office sBP fell from average 179 to 144 mmHg (P<0.0001), whereas office dBP dropped from average 103 to 85 mmHg (P<0.0001). Heart rate fell from 74 to 71 beats per minute (P<0.02).

The effect of baroreflex activation therapy is greater than average in patients with signs of heart failure and less than average in patients with isolated systolic hypertension. In ≈25% of patients, it was possible to reduce the number of medications from a median of 6 to a median of 3. Temporary side effects, related to either the surgical procedure or the cardiovascular instability, do occur, but they do not require specific measures and resolve over time.

After a follow-up of 6 years, baroreflex activation therapy maintains its efficacy for persistent reduction of office blood pressure in patients with resistant hypertension without major safety issues.

More information from the article:

Baroreflex activation therapy (BAT) is a novel way of treating hypertensive patients who respond inadequately to medical therapy. After short-term human studies had shown that electric stimulation of the carotid sinus can lower blood pressure, the DEBuT-HT study (Device-Based Therapy of Hypertension) demonstrated a substantial and sustained reduction in blood pressure over a period of 3 months in treatment-resistant hypertensive patients.

Subsequently, the Rheos Pivotal Trial evaluated the effect of BAT in a double-blind, randomized, prospective, sham-controlled trial in which patients were randomized to receive BAT either immediately or 6 months after implantation of the Rheos device. This study also showed that BAT can safely reduce blood pressure in patients with resistant hypertension in the long run.

At this point, however, the technique has not yet been introduced into clinical practice. One of the drawbacks is that it is an invasive method, and there is still some concern as to whether efficacy is sustained over longer periods of time. In this regard, the 3 trials that have evaluated this device-based therapy (the US Rheos Feasibility Trial, the DEBuT-HT Trial, and the Rheos Pivotal Trial) offered a unique opportunity to address this problem. Indeed, after the formal completion of both trials, patients were followed up at regular intervals to assess the long-term effect of BAT on blood pressure and heart rate. The purpose of the present report, therefore, is to describe the follow-up results of BAT during a 6-year period.


The present analyses show that BAT has a sustained effect on blood pressure after 6 years of follow-up. The largest part of the final drop in blood pressure occurs already within the first year of treatment, notably in the first 6 months, and stabilizes thereafter. Although nonresponsiveness did occur, there were only 26 patients (7% of total) in this cohort in whom systolic pressure fell by <10 mm Hg. Overall, about half of the patients reached the target pressure of 140 mmHg systolic or below and remained relatively stable during the remainder of follow-up.

The effects of BAT were similar across several subgroups and were not modified by sex, smoking status or the presence of diabetes mellitus, coronary artery disease, and stroke. Interestingly, in Afro-Americans, blood pressure fell to the same degree as it did in whites. In contrast, SYMPLICITY HTN-3, involving renal denervation that also lowers sympathetic activity, had a lesser effect in this population. This is probably because of the lack of total renal denervation in that trial. That same trial also showed that renal denervation lowers systolic blood pressure less effectively in patients >65 years of age. BAT, however, showed no effect of age on its potential to lower systolic blood pressure. However, diastolic pressure did not fall as much in those older than 60 years compared with younger patients.

Thus, there may be some apparent differences in the efficacy of renal denervation compared with BAT, even though both techniques are indicated to reduce sympathetic traffic. It is conceivable that these differences explain, at least in part, why BAT can still reduce blood pressure in patients who had previous renal denervation. In a small group of predominantly female patients, the initial fall in blood pressure was followed by unresponsiveness to stimulation. Why this occurred is not entirely clear. 

It should be stressed that the present data relate to office blood pressure measurements only. In drug trials, the effects of medication on blood pressure usually are less impressive when based on ambulatory blood pressure measurements than when based on office pressures. The same may be true for the effect of BAT. Nevertheless, a recent study from Germany showed that BAT significantly reduced ambulatory pressures as well.


Isolated Systolic Hypertension

In the entire group of patients, BAT significantly reduced pulse pressure, and one would expect, therefore, that this treatment would be particularly beneficial in patients with ISH. This, however, appeared not to be the case, as the pressure drop was attenuated in patients with ISH. One could argue that this was because of the fact that they already had a lower diastolic pressure to begin with. However, this cannot be the sole explanation, as we found no convincing evidence of a relationship between the height of blood pressure and the change in pressure. A possible explanation could be that ISH occurs mainly in older patients, and sympathetic activity is less pronounced in this group. Increased arterial stiffness and reduced baroreceptor sensitivity which are more prevalent at higher age may also contribute to a lesser response in patients with ISH. Interestingly, renal denervation is also less effective in these patients.


Heart Failure

A striking finding is the potentiated response in patients with signs of heart failure, a condition that is associated with a markedly enhanced sympathetic drive. In this regard, the results support those of other, recent studies of BAT in patients with heart failure and reduced ejection fraction. Although we did not routinely assess ejection fraction in our patients, their higher systolic pressure at baseline at least suggests that their ejection fraction was not severely impaired.

Medication Need

A heterogeneous pattern occurred in the number of medications taken over time. In roughly one-third of the patients, medication use remained stable at a median of 5 medications per day. This means that hypertension became more treatable with BAT. Whether this is related to an independent effect of BAT on blood pressure, in addition to the pharmacological treatment, cannot be determined from these data. It is possible that BAT makes the cardiovascular system more sensitive to the action of antihypertensive drugs.

In 39% of the patients, however, we found that not only were more medications taken but also a greater amount of electric stimulation was programmed. In 27% of the patients, the opposite was true, that is, less medications, as well as less stimulation power. Although these divergent responses cannot be readily explained, the data are at least compatible with the theory that ≈60% of patients (those with less or stable medication) exhibit some degree of true baroreceptor resetting after BAT so that hypertension becomes less severe or more receptive to pharmacological treatment. In the remainder, resetting is either absent or less pronounced. Nevertheless, even in the group with more drugs and more electric stimulation, blood pressure was reduced to almost normal levels.


Adverse Events

As is common with first-generation devices, the Rheos system was not free from adverse events. The events, which typically occurred in the initial phases after implantation, have already been described with respect to the first-generation technology in earlier publications and which greatly improved with the introduction of the second-generation Barostim neo technology. During prolonged follow-up, other complications were reported, but not to the extent that Rheos became unsafe, and they resolved with time.



There are several limitations of the present study. The most important one is the fact that 2 of the 3 studies were nonrandomized, and all lacked a control group during prolonged follow-up. Only in the pivotal trial, there was a randomized comparison of true BAT versus sham BAT but only for a period of 6 months. The ideal control group for long-term evaluation would consist of patients in whom the device had been implanted but not activated. However, to follow patients with an implanted and deactivated device for so many years would be unethical. The second best would be to compare this group with another group of patients with optimal medical treatment. Logistically, this would not be easy to do and one would not expect such large differences in blood pressure as we observed in the present analysis. Still, the results of the present study should be interpreted with caution.

Second limitation is the variable follow-up period. Theoretically, this could have biased the results, as patients who do not do as well with the device are likely to drop out at an earlier stage so that the better patients remain in follow-up. We have tried to overcome this objection by presenting the data for the various cohorts with different periods of follow-up. In so doing, we found no evidence that shorter follow-up periods were related to less efficacy of the device. Third, efficacy data are based on office pressures only, and it is possible that the data are less impressive on 24-hour ambulatory monitoring.

Finally, adherence to drug treatment should be considered. Indeed, lack of adherence could have led to a false diagnosis of treatment resistance and, hence, the inclusion of the wrong type of patients. Indeed, a recent study of patients who had been referred for renal denervation showed that a significant proportion was, in fact, not adherent rather than treatment-resistant. This is a problem with all hypertension trials and cannot be solved until reliable, easy-to-use, and cheap methods are available to measure adherence. Recently, it was proposed to measure antihypertensive drug levels in the patient by means of liquid chromatography–tandem mass spectrometry. However, this method is expensive and can detect only those individuals who have not taken medication for a prolonged period of time, not the ones who take their drugs off and on.

To get at least some insight in patients’ behavior, we used, before inclusion, Medication Events Monitoring System track caps in a small subset of patients, allowing us to include only those with adequate adherence. However, we still cannot rule out the possibility that these patients became temporarily more adherent because they knew they were being monitored, the so-called Hawthorne effect. Although the existence of this effect has been challenged, one could argue that it also played a role during the follow-up phase, in that patients may have taken their antihypertensive drugs more punctually. Although we cannot entirely exclude such an effect, the results of the pivotal trial speak against this option. In that trial, the effect of the device was larger in the group in which BAT was started immediately than in the group with deferred BAT. In the latter group, blood pressure began to drop more substantially when the device had been switched on. Because patients did not know to which group they belonged, it seems unlikely that these differences in response were because of a Hawthorne effect.



We have shown that chronic BAT durably lowers blood pressure in treatment-resistant hypertensive patients. It is markedly effective when hypertension is complicated by heart failure with preserved ejection fraction. Altogether, these results justify further development and implementation of device-based therapy for resistant hypertension and heart failure, including those with reduced ejection fraction.

Moreover, there is a need to assess whether unilateral stimulation with new and smaller devices will be as effective as the Rheos system. With the further development of renal denervation, it becomes important also to know whether baroreceptor activation therapy is efficacious in patients with previous renal denervation and vice versa.

Because BAT requires an invasive (surgical) procedure, it is worthwhile to explore whether alternative methods such as those related to nanotechnology can eventually make the method better suitable. Finally, our results prompt for further research into how baroreceptors precisely work in hypertension, in particular on the long run, and how we can select a priori those individuals who are going to benefit most from baroreceptor activation.


Editorial on the subject (click here)