Predictors of blood pressure response after renal denervation beyond pretreatment blood pressure

Background The principle of initial value (Wilder's law of initial value) proposes that the “direction of response of body function to any agent depends to a large degree on the initial value of that function”. Indeed, in several trials on renal denervation (RDN), pre-treatment blood pressure (BP) has been consistently and repeatedly found to predict the decrease in BP after RDN. Efforts to discover further statistically significant and clinically meaningful predictors of BP response to RDN failed. Objective By use of a new methodological approach, we aimed to determine predictors of BP response after RDN in patients with resistant hypertension. Methods The study population comprised 266 patients with resistant hypertension (mean age 62 years, 34% females, mean BMI 30.5 kg/m2, 27% had coronary heart disease, 42% had diabetes mellitus and 61% had hypercholesterolaemia) who underwent radiofrequency RDN with the Symplicity catheter at the Universities of Homburg and Erlangen. Clinical data including 24h ambulatory BP (ABP) were obtained prior to, and 3, 6 and 12 months after RDN. The primary parameter of response was defined as change in 24-hour systolic ABP after 6 months from pre-treatment values. As expected, change in 24h systolic ABP correlated with pre-treatment 24h systolic ABP (r2 linear = 0.225, p<0.001), with change in 24h systolic ABP = 73.82 + 0.55 x pre-treatment 24h systolic ABP. To overcome the predominant role of the pre-treatment BP that may mask other factors, we calculated for each individual patient the “expected systolic ABP decrease” by applying this regression equation and the “excessive systolic ABP decrease” by subtracting the measured from expected change in 24h systolic ABP. We divided the study population into 2 groups (above [responders] and below [non-responders] of the median change in excessive 24h systolic ABP. Results Neither pre-treatment 24h systolic or diastolic ABP, nor office systolic or diastolic BP differed between the two groups (all p>0.20). Following RDN, 24h systolic ABP decreased in the responders by −23.3±16 vs non-responders +1.4±11 mmHg at 6 month, and 24h systolic ABP values were also significantly lower in responders at 3 and 12 months (all p<0.001), without difference in number of antihypertensive drugs between the groups. Of all clinical variables at baseline, office heart rate (65.6 vs 68.7±12 bpm, p=0.024) and HbA1c (6.07±0.88 vs. 6.37±1.23%, p=0.035) were lower in responders compared with non-responders. Finally, a multiple regression analysis confirmed that pre-treatment 24h systolic ABP (beta +0.565, p<0001), HbA1c (beta −0.167, p=0.004) and office HR (beta +0.106, p=0.057) were independent predictors of decrease in 24h systolic ABP. Conclusion In patients with resistant hypertension, lower HbA1c and office HR were identified as predictors of BP response in addition to pre-treatment BP. This finding may help to identify hypertensive patients who benefit most from RDN. Funding Acknowledgement Type of funding sources: Other. Main funding source(s): Extramural grant provided vy Medtronic INc