Author + information
- Larry A. Weinrauch, MD†,‡∗ (, )
- Akshay S. Desai, MD, MPH‡,
- Hicham Skali, MD, MSc‡ and
- John A. D'Elia, MD†
- †Joslin Diabetes Center, Boston, Massachusetts
- ‡Cardiovascular Division, Brigham and Women's Hospital, Boston, Massachusetts
- ↵∗Reprint requests and correspondence:
Dr. Larry A. Weinrauch, Harvard Medical School, 521 Mount Auburn Street, Watertown, Massachusetts 02472.
The contribution of resistant hypertension (RH) to morbidity and cost in chronic kidney disease (CKD) populations should not be understated. Failure to control blood pressure (BP) inevitably heralds renal deterioration as well as accompanying increases in cardiovascular morbidity and mortality. Successful antihypertensive treatment to guideline-recommended targets is challenging, due to intravascular volume expansion (sodium/fluid retention) and treatment-related adverse effects limiting patient adherence. The NHANES (National Health and Nutrition Examination Survey) and ALLHAT (Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial) trials both demonstrated that CKD itself is a predictor of cardiac events as a result of failure to achieve adequate BP control (1,2). But to date the management of RH remains empirical, due to a paucity of scientific clinical outcomes data with suboptimal results.
In this issue of the Journal, data are presented with regard to prevalence and prognosis of RH in 436 patients with CKD (3). Consecutive patients with treated hypertension (HTN) and CKD Stage II to V prospectively recruited from 4 nephrology clinics in Italy underwent 48-h ambulatory blood pressure monitoring (ABPM) as well as 6 standard BP measurements during the same time period. Patients were classified into 4 phenotypes on the basis of achievement of guideline-directed targets (4,5) at the time of enrollment (Table 1): true resistance (RH); pseudoresistance; sustained; or treatment responsive HTN.
With this granular categorization of HTN control applying standard office BP measurements and a single session of ABPM, the authors defined a gradient of risk for RH in CKD patients. As anticipated, the risk for cardiorenal events was highest in patients with “true” RH (concordant elevation in ABPM and office measurements). Surprisingly, those with “sustained” and “pseudoresistant” HTN were not at increased cardiovascular risk compared with control subjects. Those with “sustained” HTN were, however, at relatively higher risk for renal events.
These data from a CKD population mirror experience in unselected populations highlighting the unfavorable prognosis of “masked” HTN in comparison with the favorable prognosis of “white coat” HTN (6). All of these patients, however, were hypertensive. These results underscore the importance of ABPM for “reclassification” of the adequacy of BP control in CKD. Use of ABPM identified a large proportion (43% in this study) of subjects for whom BP control seemed adequate by office measurement but was actually suboptimal. Such patients, overlooked during routine surveillance, are under-treated (only 30% on regimen of ≥3 antihypertensive drugs). Phenotyping with ABPM offers the opportunity to improve clinical cardiorenal outcomes by targeted antihypertensive therapy in the sustained HTN group. By contrast, finding patients with pseudoresistant HTN (of whom 100% are on regimen of ≥3 medications) might identify a group exposed to adverse effects of complex antihypertensive treatment without cardiovascular or renal benefit. Similar observations have been previously made in a dialysis population (7).
Although discordance between cardiorenal endpoints in the sustained HTN group is challenging to explain, these patients were younger than those with “true” RH (62 vs. 68 years) and less likely to have diabetes mellitus (25% vs. 64%). The lower incidence of left ventricular hypertrophy and diabetes in those without true RH raises questions with regard to links between insulin resistance and (lack of) efficacy of antihypertensive drugs in CKD populations. The absence of an observed relationship between body mass index, RH, and pseudoresistance in this study (3) contrasts with prior studies linking obesity with increased sympathetic outflow (8) and should spur future investigations.
Factors responsible for inadequate HTN control in populations with true resistant or sustained HTN are not clearly illuminated. Relevant factors in broader populations with RH include poor adherence to prescribed therapies, use of substances interfering with therapy (nonsteroidal anti-inflammatory drugs, licorice, ephedrine, thyroid hormone, contraceptives, and licit or illicit sympathomimetic drugs), lifestyle factors including obesity and dietary sodium intake, inadequate treatment (particularly inadequate use of diuretic agents), untreated secondary causes (including renovascular disease, obesity/sleep apnea, hyperaldosteronism, and Cushing's syndrome), insulin resistance/diabetes, and genetic factors. Optimal management of patients with inadequate ambulatory BP control in CKD requires additional investigation. Many treatments effective in non-CKD populations (e.g., sodium restriction, combination renin-angiotensin-aldosterone system blockade, aldosterone antagonists) are ineffective or associated with higher risk in populations with CKD (9,10).
Nonpharmacologic therapies such as renal sympathetic denervation are particularly promising in this regard (11,12). Ongoing trials of this therapy in populations with RH defined by office BP data should take account of these outcome data, which highlight the potential for inadvertent inclusion of lower-risk patients with “pseudoresistance” who would not be expected to benefit and exclusion of those with “sustained” HTN and CKD who might be better candidates. We must not repeat (uncontrolled, underpowered) trial errors of unsuccessful renovascular manipulations of past decades, chasing surrogate successes (13,14), and avoidance of “falling behind,” while opening a deluge of expensive procedures that expose patients to thus far unreported adverse outcomes. We hope that the 48 open trials currently listed at ClinicalTrials.gov(15) will modify flawed designs to answer appropriate safety and outcome questions. These same reservations apply to pharmacologic agents under investigation (aldosterone synthase inhibitor, nitric oxide donor, and the like) (16) with none demonstrated to be safe or useful in CKD populations.
There are study limitations, many acknowledged by the authors. This study is observational, and important differences between groups and subsequent therapies might account for variations in long-term outcomes. A largely European population, from a single region, cannot be generalized to other cohorts with HTN and CKD, and cross-validation is needed. Nonetheless, this work highlights a linkage between readily available technology and important clinical endpoints (not surrogates). This report (3) adds to the growing published data highlighting limitations of office BP measurements for management of HTN, exposes additional gaps in application of evidence-based antihypertensive treatment, and provides a basis for future research.
In clinical practice, there is increasing reliance on automated BP measurement taken in substandard conditions by auxiliary personnel with limited clinical experience. Even when properly done, random office measurement of BP might be an inadequate surrogate for the risk for long-term outcomes. A single session of ABPM might provide a reliable estimate of the efficacy of antihypertensive treatment and more efficiently direct selection of appropriate therapy in patients with CKD while costing less than a single session of dialysis. Home BP self-monitoring has not been demonstrated to mirror these clinical outcome results (17–19). These ABPM-driven results suggest a need to re-evaluate our appropriate device usage criteria to improve long-term clinical outcomes and financial impact among hypertensive CKD populations. We should not accept false positive and negative rates of this magnitude in defining adequate BP control in populations on the precipice of heart failure and end-stage renal disease or the actual prevalence of RH that represents ≤25% of the population identified as such.
An important observation worth emphasizing is that a marker is used (APBM result) to reclassify patients and to determine whether reclassification has a measurable impact on healthcare outcome in the population. Future studies will determine whether treatment decisions on the basis of knowledge of this marker improved outcomes among those at high risk. It is time to determine whether measurement of ABPM improves healthcare outcome and to determine the impact of ABPM management reclassifications among heart failure populations. Absent such studies, a potentially useful, noninvasive technology will remain underused, and the financial impact of poorly controlled HTN on prevalence of end-stage renal disease and heart failure will continue to increase.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Desai has served as a consultant to Novartis, Boston Scientific, Reata, and Intel; received research grant from AtCor Medical, Inc.; and received travel honoraria from Amgen, Inc. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2013 American College of Cardiology Foundation
- De Nicola L.,
- Gabbai F.B.,
- Agarwal R.,
- et al.
- Esler M.D.,
- Krum H.,
- Schlaich M.,
- Schmieder R.E.,
- Böhm M.,
- Sobotka P.A.,
- Symplicity HTN-2 Investigators
- Weinrauch L.A.,
- D'Elia J.A.
- ↵ClinicalTrials.gov [Internet]. Bethesda, MD: National Library of Medicine (US). Renal Denervation and Hypertension: Open Studies. Available at: http://clinicaltrials.gov/ct2/results?term=Renal+Denervation+and+Hypertension%3A+Open+Studies. Accessed May 8, 2013.