Anti-hypertensive medication adherence in the REQUIRE trial: post-hoc exploratory evaluation

Of the 136 RH patients enrolled in the REQUIRE trial, 58 were assessed in this analysis. Baseline characteristics of the cohort with available urine data were similar to those of the entire REQUIRE cohort (Supplementary Table 1).

At baseline, 55.2% showed good medication adherence (full: 26 patients, greater partial: 6 patients) and 44.8% had poor adherence (lesser partial: 19 patients, none: 7 patients) (Fig. 1). To the best of our knowledge, this is the first study to conduct a direct evaluation of medication adherence in Japanese patients with RH, though the number of patients studied was small. The proportion of patients with good adherence in our study was consistent with the results from DENERHTN [12] and the Peregrine studies [13], demonstrating low adherence rate of 40–50% for RH patients. Thus, the REQUIRE trial seems to have included many patients with pseudo-RH, or poor BP control due to poor medication adherence.

Fig. 1figure 1

Graphical abstract: Medication adherence at baseline. 55.2% of patients with resistant hypertension showed good adherence, whereas 44.8% showed poor adherence

We evaluated differences in baseline characteristics between patients with good adherence and poor adherence. In the poor adherence group, age (51.0 ± 9.4 vs 58.5 ± 11.4, p = 0.010 by the unpaired t-test) at baseline was lower, and the number of prescribed antihypertensive drugs (3.9 ± 1.6 vs 3.1 ± 1.0, counted on the basis of formulation, p = 0.0214 by the unpaired t-test) and baseline ASBP (170.8 ± 10.7 mmHg vs 160.8 ± 14.9 mmHg, p = 0.006 by the unpaired t-test) were higher.

Next, we examined the effects of baseline medication adherence on the changes in adherence and ASBP at 3 months post-procedure. Figure 2A demonstrates that adherence rates did not change in patients with good adherence undergoing either uRDN or sham and poorly adherent patients undergoing uRDN. However, in poorly adherent patients undergoing sham procedure, medication adherence showed a trend towards improvement (p = 0.059). Fig. 2C shows that in patients with good adherence, uRDN decreased ASBP significantly (p = 0.018), whereas sham did not. In poorly adherent patients, ASBP did not change in the uRDN group, but decreased in the sham group (Fig. 2C, p = 0.035).

Fig. 2figure 2

Change in medication adherence rate and 24-h ambulatory systolic blood pressure after ultrasound renal sympathetic nerve denervation or sham procedure. Actual value at baseline and 3 months for adherence (A) and its change from baseline (B). Actual value at baseline and 3 months for 24-h ambulatory systolic blood pressure (C) and its change from baseline (D). Circles indicate individual values. Data are expressed as the mean + or ± standard deviation. The Wilcoxon signed-rank test and the paired t-test were performed for comparisons of adherence rate and ASBP between baseline and 3 months post-procedure, respectively. The Mann–Whitney U test and the unpaired t-test were performed for comparisons of change in adherence rate and ASBP between two groups, respectively. Values are described in Supplementary Table 3

In addition, the reduction in ASBP following uRDN compared to sham was −10.1 ± 13.3 mmHg versus −1.9 ± 15.3 mmHg, respectively, in patients with good adherence at baseline (Fig. 2D and Supplementary Table 3), which was comparable to that reported in RADIANCE-HTN SOLO, RADIANCE-HTN TRIO and RADIANCE II trials demonstrating the BP-lowering effect of uRDN superior to sham-operated patients [7,8,9].

It was noteworthy that three (21%) of 14 sham-operated patients with poor adherence at baseline showed remarkable ASBP reductions greater than −30 mmHg (Fig. 2D). Such large BP reductions were likely due to improved medication adherence that was poor before the study, rather than to the effect of the therapeutic intervention. Overall, adherence was improved in five (36%) of 14 sham-operated patients with poor adherence at baseline (Fig. 2B).

Given the adherence change over time in some patients, an additional analysis was conducted to look at patients that had good adherence at both baseline and 3 months. A significant BP-lowering effect of uRDN was evident in patients who had good baseline medication adherence and maintained good adherence during the observation period (24-h ASBP, 165.2 ± 14.3 mmHg at baseline, 155.3 ± 18.1 mmHg at 3 months, n = 11, p = 0.042 by the paired t-test), whereas the sham did not show a significant decrease (159.3 ± 1.7 mmHg at baseline, 157.4 ± 18.2 mmHg at 3 months, n = 19, p = 0.595 by the paired t-test).

Although the number of patients in this analysis was a bit less than half of the entire REQUIRE cohort, the baseline patient characteristics were similar. These data suggest that the gradual and large BP reduction in the sham group is in part explained by improvement of medication adherence during the observation period. Taken together, it is possible that REQUIRE included many patients who originally had poor BP control due to poor medication adherence and the BP-lowering effect of uRDN was offset by improved medication adherence during the observation period in such patients, particularly in the sham group.

This study had limitations. First, the long-term stability of drugs in urine samples has not been confirmed. Second, a qualitative measurement was employed; therefore, information about the exact drug doses could not be obtained. Third, the pharmacokinetic characteristics of the drugs and renal and liver function of patients could have influenced the detection of the compound; and several drugs were not measured because they were not listed in the drug panels able to be measured (Supplementary Table 2). A sensitivity analysis including the 48 patients in whom all prescribed drugs or their metabolites were able to be measured confirmed the results described above (Supplementary Tables 46).

In conclusion, this REQUIRE post-hoc analysis showed that about 45% of Japanese RH patients had poor medication adherence. A significant BP-lowering effect of uRDN could be evident in patients having good baseline medication adherence that was maintained during the observation period. Monitoring and maintaining medication adherence is important for future intervention studies in RH.

留言 (0)

沒有登入
gif