Abstract
Background Whether ambulatory BP monitoring is of value in evaluating risk for outcomes in patients with CKD is not clear.
Methods We followed 1502 participants of the Chronic Renal Insufficiency Cohort (CRIC) Study for a mean of 6.72 years. We evaluated, as exposures, ambulatory BP monitoring profiles (masked uncontrolled hypertension, white-coat effect, sustained hypertension, and controlled BP), mean ambulatory BP monitoring and clinic BPs, and diurnal variation in BP—reverse dipper (higher at nighttime), nondipper, and dipper (lower at nighttime). Outcomes included cardiovascular disease (a composite of myocardial infarction, cerebrovascular accident, heart failure, and peripheral arterial disease), kidney disease (a composite of ESKD or halving of the eGFR), and mortality.
Results Compared with having controlled BP, the presence of masked uncontrolled hypertension independently associated with higher risk of the cardiovascular outcome and the kidney outcome, but not with all-cause mortality. Higher mean 24-hour systolic BP associated with higher risk of cardiovascular outcome, kidney outcome, and mortality, independent of clinic BP. Participants with the reverse-dipper profile of diurnal BP variation were at higher risk of the kidney outcome.
Conclusions In this cohort of participants with CKD, BP metrics derived from ambulatory BP monitoring are associated with cardiovascular outcomes, kidney outcomes, and mortality, independent of clinic BP. Masked uncontrolled hypertension and mean 24-hour BP associated with high risk of cardiovascular disease and progression of kidney disease. Alterations of diurnal variation in BP are associated with high risk of progression of kidney disease, stroke, and peripheral arterial disease. These data support the wider use of ambulatory BP monitoring in the evaluation of hypertension in patients with CKD.
Podcast This article contains a podcast at https://www.asn-online.org/media/podcast/JASN/2020_09_24_JASN2020030236.mp3
Hypertension is common in patients with CKD, and is an important risk factor for morbidity and mortality in this population.1 There has been increasing interest in incorporating BP measured outside the clinical setting into the management of hypertension.2 Twenty-four-hour ambulatory BP monitoring (ABPM) is recommended as the optimal technique to evaluate BP outside the clinic.3 In addition to BP readings obtained during the course of the 24-hour period, the value of ABPM is in the identification of profiles based on comparison of clinic with out-of-clinic BP readings. White-coat effect is the profile where BP in the clinic is higher than BP measured outside the clinic4; masked uncontrolled hypertension is the reverse, where BP is lower in the clinic when compared with out-of-clinic measurements.5 In the general population, there is some uncertainty about the prognostic significance of white-coat hypertension; recent data suggest the prognosis of white-coat hypertension may not be as benign as previously thought, particularly in patients who are untreated.6 There is more consistency in the general population that masked uncontrolled hypertension is associated with higher risk of cardiovascular disease, comparable with, or higher than, that of patients with sustained hypertension.7 In patients with CKD, the prevalence of masked uncontrolled hypertension is higher (30%–59%) than in patients in the general population (10%–25%).8 However, less is known about the long-term prognostic significance of masked uncontrolled hypertension in patients with CKD. Cross-sectional studies have shown associations between masked uncontrolled hypertension and subclinical markers of target organ damage in patients with CKD.9 Masked uncontrolled hypertension has been associated with higher risk of cardiovascular outcomes and mortality,10–⇓12 but association with progression of kidney disease has been inconsistent.13,14 In addition, although nighttime BP is often elevated in patients with CKD,15 the long-term prognostic significance of nighttime BP, independent of clinic BP, is unclear.16
The objective of this paper is to evaluate the association between profiles of BP defined by ABPM (white-coat effect, masked uncontrolled hypertension, and sustained hypertension) and long-term kidney and cardiovascular outcomes, and mortality in patients with CKD, using data from the Chronic Renal Insufficiency Cohort (CRIC) Study. In addition, the association between 24-hour, day, and nighttime BPs; diurnal variation of BP; and clinical outcomes was examined.
Methods
Study Population
The CRIC Study is a prospective, multicenter, observational cohort study of participants with CKD. The study design and baseline characteristics of participants have been described previously.17,18 Between 2003 and 2008, 3939 participants, aged 21–74 years, with eGFR between 20 and 70 ml/min per 1.73 m2 were enrolled for participation in the study. Exclusion criteria included a diagnosis of polycystic kidney disease and active immunosuppression for GN as well as cirrhosis, class 3 or 4 heart failure, HIV infection, cancer, and pregnancy. The study protocol was approved by the institutional review board of each participating site, and written informed consent was obtained from all participants. ABPM was obtained in 1502 participants and forms the basis of this report. Participants with any of the following were not eligible for ABPM: arm circumference >50 cm, night-shift workers, breast cancer requiring mastectomy or radiation, and previous ESKD.9 Clinical and demographic comparisons of CRIC participants who had ABPM measurements versus those who did not have ABPM measured has been previously published.9
Main Exposures
Clinic BP was measured three times during all clinic visits by trained study staff, following standardized protocols based on American Heart Association standards for BP measurement using aneroid sphygmomanometers; the average of these three seated measurements in the right arm after a 5-minute rest period was used to define clinic BP. BP measurement was not obtained in the left arm. The clinic BP measured at the visit when the ABPM was applied are used in these analyses.
ABPM
ABPM measures were obtained during the second phase of the CRIC Study. The first phase of CRIC was between 2003 and 2007, in which 3939 participants were recruited to participate in the study. ABPM was conducted between 2008 and 2012 during the second phase of CRIC; therefore, participants who had died, were lost to follow-up, or who did not reconsent for the second phase were not available for measurement of ABPM. Of patients enrolled in phase 2, participants were chosen randomly for measurement of ABPM. Selection of participants was stratified by clinical site. The coordinating center notified the sites if a participant was chosen for ABPM, and the site approached the participant to further evaluate for exclusion criteria, and obtained the ABPM measurement. The average time from the CRIC enrollment visit to ABPM was 5.1 years. For the purposes of this manuscript, baseline was defined as the measures obtained at the time of ABPM measurement, or the closest prior visit. Using the SpaceLabs 90207 or 90217 monitor, BP readings were obtained in the nondominant arm every 30 minutes throughout the day and night. The recording was considered valid if there were at least 14 readings between 06:00 am and midnight, and at least six readings between midnight and 06:00 am.19 Participants with insufficient readings (n=57) were not included in the analyses.9 Nighttime BP was defined by the average of readings between midnight and 06:00 am, and daytime was defined as 06:00 am to midnight. In sensitivity analyses, daytime BP was defined as 09:00 am to 09:00 pm. The ABPM profiles were defined for the primary analyses as follows7:
masked uncontrolled hypertension (clinic systolic BP <140 mm Hg and diastolic BP <90 mm Hg, and 24-hour systolic BP ≥130 mm Hg or diastolic BP ≥80 mm Hg);
white-coat effect (clinic systolic BP ≥140 mm Hg or diastolic BP ≥90 mm Hg, and 24-hour systolic BP <130 mm Hg and diastolic BP <80 mm Hg);
sustained hypertension (clinic systolic BP ≥140 mm Hg or diastolic BP ≥90 mm Hg, and ambulatory 24-hour systolic BP ≥130 mm Hg or diastolic BP ≥80 mm Hg); and
controlled BP (clinic systolic BP <140 mm Hg and diastolic BP <90 mm Hg, and 24-hour systolic BP <130 mm Hg and diastolic BP <80 mm Hg).
In sensitivity analyses, BP thresholds as defined by the 2017 American College of Cardiology/American Heart Association hypertension guidelines were used.2
There is some inconsistency in terminology in the literature in the field: some investigators use the term “masked hypertension” and “white-coat hypertension,” whereas others use “masked uncontrolled hypertension” and “white-coat effect” for patients on antihypertensive drug therapy.20 Most patients in this study were treated with antihypertensive drug therapy (Table 1). Therefore, for consistency, we use the terms masked uncontrolled hypertension and white-coat effect throughout this paper.
Baseline characteristics of study population overall, and stratified by ABPM profile (SD)
Diurnal variation in BP was defined as reverse-dipper (night/day systolic BP ratio >1), nondipper (night/day systolic BP ratio 0.9–1), and dipper (night/day systolic BP ratio <0.9) profiles.21
The clinical and demographic characteristics of the participants with ABPM compared with all patients enrolled in the CRIC Study have been previously reported.9 Because ABPM was obtained in phase 2 of CRIC, the characteristics of phase-2 participants who did and did not obtain ABPM was also examined (Supplemental Table 2)22.
Covariates
At study visits, demographic and physical measures, medical history, medication use, and serum and urine for laboratory assessments were collected. Participants were followed annually with in-person clinic visits, and were also contacted by telephone calls approximately 6 months apart. Diabetes was determined as at least one of the following: self-reported insulin or oral hypoglycemic medication, fasting blood glucose ≥126 mg/dl or a nonfasting level ≥200 mg/dl, or a hemoglobin A1c ≥6.5%. GFR was estimated using the CRIC Study equation.23 Self-reported history of any cardiovascular disease at baseline included previous myocardial infarction, coronary revascularization, heart failure, stroke, or peripheral arterial disease.
Outcomes
Outcomes were ascertained from the time the ABPM was performed through June 2018. Cardiovascular events (myocardial infarction, cerebrovascular accident, heart failure, and peripheral arterial disease) were ascertained during the course of the study by asking participants about hospitalizations during the clinic or phone visit. Hospital records were then obtained and were adjudicated using predefined, event-specific guidelines by two clinicians. Heart failure events were determined on the basis of clinical symptoms, radiographic evidence of pulmonary edema, physical examination of the heart and lungs, central venous hemodynamic monitoring data, and echocardiographic imaging in patients who were hospitalized, using criteria consistent with previous studies.24 Diagnosis of probable or definite myocardial infarction was made on the basis of symptoms consistent with acute ischemia, cardiac biomarker levels, and electrocardiograms, as recommended by a consensus statement on the universal definition of myocardial infarction.25 Two neurologists reviewed all hospitalizations and emergency-department visits suggestive of stroke, and adjudicated the stroke outcome based on predefined criteria.26 The kidney outcome was defined as a composite of ESKD or halving of the eGFR. In addition to participant report, ESKD was also ascertained through the United States Renal Data System database. Mortality was ascertained through report from next of kin, retrieval of death certificates or obituaries, review of hospital records, and linkage with the Social Security Mortality Master File. In sensitivity analysis, we also considered cardiovascular mortality. For some of the death events, we adjudicated whether the death event was cardiovascular related (concordant reviews by two physicians determined if a hospitalization was cardiovascular related). Using the adjudicated death events as gold standard, we built a super-learning algorithm,27 based on the cause of death codes from the National Death Index data, to predict the probability of a death event that is cardiovascular related.
Statistical Analyses
Baseline characteristics were compared among the four ABPM hypertension groups (white-coat effect, masked uncontrolled hypertension, sustained hypertension, and controlled BP) as mean and SD or median and interquartile range for continuous variables, and frequency and percentage for categoric variables. P values were calculated using ANOVA, Kruskal–Wallis rank sum test, or Pearson chi-squared test as appropriate.
For associations between the three main exposures such as clinic BP, ABPM profile, and diurnal variation in BP and various cardiovascular, kidney, and mortality outcomes, Cox proportional hazard models were fitted. For each outcome considered, three models were conducted: the first model was unadjusted; model A adjusted for demographic factors (age, sex, center, race/ethnicity) and traditional cardiovascular risk factors such as diabetes, GFR, urinary protein-creatinine ratio, and prior cardiovascular disease history; and model B added clinic or ABPM BP measures. All P values are two sided and statistical significance is defined as P<0.05. All statistical analyses were conducted with SAS, version 9.4 (Cary, NC).
Results
The mean age of the study population (n=1502) was 63.1±10.3 years; 49.5% of participants had diabetes, 38.7% were non-Hispanic Black, and 38.9% had self-reported cardiovascular disease at baseline (Table 1). The mean GFR was 46 ml/min per 1.73 m2. Masked uncontrolled hypertension was seen in 416 (27.6%) participants, white-coat hypertension in 60 (3.9%), sustained hypertension in 283 (18.8%), and controlled BP in 743 (49.4%) participants. There were several differences in clinical and demographic characteristics at baseline across these groups (Table 1). The reverse-dipper profile was seen in 15.64%, nondipper profile in 47.14%, and dipper profile in 37.22% of study participants; baseline characteristics stratified by these categories are presented in Supplemental Table 1. There were minor differences in clinical characteristics between participants who had ABPM data available compared with those not selected for collection of ABPM data (Supplemental Table 2).
After mean follow-up of 6.72 years, the composite cardiovascular outcome was seen in 333 (22.1%) participants, and 363 (24.1%) participants reached ESKD or had a 50% decline in GFR. After adjustment of traditional risk factors and clinic BP, presence of masked uncontrolled hypertension was associated with higher risk of the composite cardiovascular outcome (hazard ratio [HR], 1.5; 95% CI, 1.12 to 2.01) and the kidney outcome (HR, 1.77; 95% CI, 1.31 to 2.39), but not all-cause mortality (HR, 1.28; 95% CI, 0.94 to 1.76) compared with participants with controlled BP (Figure 1, Table 2). There was no statistically significant association between the presence of white-coat hypertension and risk of cardiovascular outcomes, kidney outcomes, or mortality (Table 2). Participants with sustained hypertension were at higher risk of developing the composite cardiovascular outcome (HR, 1.9; 95% CI, 1.2 to 3), but not kidney outcomes and mortality compared with participants with controlled BP. These associations were consistent in analyses of each component of the composite cardiovascular outcome and for cardiovascular mortality (Supplemental Table 3). The associations between ABPM profiles and the composite cardiovascular outcomes were consistent across subgroups of age, sex, race, and diabetes status (Supplemental Figure 1). Masked uncontrolled hypertension and sustained hypertension were associated with higher risk of the kidney outcome in participants over the age of 65 years (P value for interaction=0.005; Supplemental Figure 2).
Masked hypertension is independently associated with higher risk of the composite cardiovascular outcome and renal outcome compared to patients with controlled blood pressure. Association between ABPM profiles and clinical outcomes. *P<0.05, compared with patients with controlled BP.
Association of ABPM profile with composite cardiovascular outcome, kidney outcome, and mortality
In sensitivity analyses, using the alternate threshold for the definition of hypertension as recommended by the 2017 American College of Cardiology/American Heart Association (<130/80 mm Hg), results were consistent with the primary analyses, although white-coat hypertension using this definition was associated with higher risk of all-cause mortality (Supplemental Table 4). We also conducted sensitivity analyses using definitions of the ABPM profiles based on daytime BP (Supplemental Table 5). The associations between ABPM profiles and clinical outcomes are qualitatively similar in this analysis to those seen with definitions based on 24-hour BP. To further evaluate the effects of GFR on the association between ABPM profiles and clinical outcomes, we conducted analyses including GFR as an interaction term. The interaction term for GFR was not statistically significant for the composite cardiovascular outcome (P=0.27) and all-cause mortality (P=0.68), suggesting the association is consistent across the continuum of GFR. However, the interaction term for GFR was statistically significant (P=0.003) for the kidney outcome. In analyses stratified by level of GFR (Supplemental Table 6), the association between ABPM profiles and kidney outcomes is most prominent in patients with GFR <30 ml/min per 1.73 m2.
The association between BP as a continuous variable and clinical outcomes is presented in Table 3. Higher mean 24-hour systolic BP was associated with high risk of the composite cardiovascular outcome (HR per SD, 1.43; 95% CI, 1.24 to 1.66), kidney outcome (HR per SD, 1.42; 95% CI, 1.22 to 1.64), and mortality (HR per SD, 1.29; 95% CI, 1.11 to 1.51). Nighttime systolic BP was associated with kidney outcomes (HR per SD, 1.52; 95% CI, 1.23 to 1.89), but was not associated with cardiovascular outcomes or mortality. Higher daytime systolic BP was associated with higher risk of the composite cardiovascular outcome and mortality. Higher clinic BP was associated with lower risk of the composite cardiovascular outcome, but higher risk of the kidney outcome (Table 3). The results were consistent for cardiovascular mortality, myocardial infarction, stroke, heart failure, and peripheral arterial disease (Supplemental Table 7). To further evaluate the effects of GFR on the association between ABPM measures (as continuous variables) and clinical outcomes, we conducted analyses including GFR as an interaction term. The interaction term for GFR was not statistically significant for most BP variables and outcomes. However, the interaction term with GFR was statistically significant for the association of 24-hour systolic, daytime systolic, and nighttime systolic BP with the cardiovascular outcome. In stratified analyses for these variables, associations were generally stronger at higher levels of GFR (Supplemental Table 8). In sensitivity using alternate definitions for daytime BP (09:00 am to 09:00 pm), association between daytime BP and outcomes was qualitatively similar to the association seen with the primary definition (Supplemental Table 9).
Association between BP (as a continuous variable, mm Hg) and composite cardiovascular outcome, kidney outcome, and mortality
Participants with the reverse-dipper profile of diurnal variation of BP were at higher risk of kidney outcomes (HR per SD, 2.03; 95% CI, 1.44 to 2.85) and peripheral arterial disease compared with participants with the dipper profile (Supplemental Table 10, Table 4). Participants with the nondipper profile were at higher risk of stroke and peripheral arterial disease compared with participants with the dipper profile.
Association of diurnal profile of BP with composite cardiovascular outcome, kidney outcome, and mortality
We conducted exploratory analyses to include 24-hour systolic BP as a covariate in models evaluating the association between ABPM profiles and outcomes. In these models, ABPM profiles (white-coat effect, masked uncontrolled hypertension, and sustained hypertension; Supplemental Table 11) were no longer statistically significantly associated with the composite cardiovascular outcomes, kidney outcomes, and mortality. After adjustment for 24-hour systolic BP, diurnal profile of BP was not statistically significantly associated with the composite cardiovascular outcome or mortality, but was significantly associated with kidney outcomes (Supplemental Table 12). We also evaluated BP and antihypertensive medication use over time (Supplemental Table 13, A–D) The pattern of office BP and antihypertensive medication use was generally stable over the course of follow-up (although this was not statistically tested).
Discussion
In this cohort of participants with CKD, the presence of masked uncontrolled hypertension was associated with high risk of cardiovascular disease and progression of kidney disease. BP measures obtained using ABPM (24-hour mean, daytime, and nighttime BP) were more strongly associated with higher risk of cardiovascular disease, kidney disease, and mortality than clinic BP. Alterations of diurnal variation in BP (reverse-dipper and nondipper profiles) were associated with high risk of progression of kidney disease, stroke, and peripheral arterial disease.
Measurement of BP outside the traditional clinic setting is an important aspect of management of hypertension. This allows identification of profiles of BP comparing clinic and out-of-clinic readings. Masked uncontrolled hypertension, defined as the profile in which out-of-clinic BP is high in the setting of normal clinic BP, is associated with increased cardiovascular risk in the general population.7 However, this has been less well studied in patients with CKD. Although the prevalence of masked uncontrolled hypertension is high in patients with CKD,20,28,29 the long-term prognostic significance is less well defined, particularly in diverse populations. In studies done in China, Japan, and Italy, masked uncontrolled hypertension was associated with a higher risk of cardiovascular events and mortality.10–⇓12 However, studies evaluating the association between masked uncontrolled hypertension and progression of kidney disease have demonstrated conflicting results; some,10,11 but not all, studies have shown a higher risk of progression of kidney disease in patients with masked uncontrolled hypertension.13,14 In a large and diverse cohort of participants with CKD, we demonstrate that the presence of masked uncontrolled hypertension is associated with high risk of cardiovascular disease and progression of kidney disease. This risk is independent of other risk factors that may influence these outcomes, including office BP, and consistent across subgroups of age, sex, race, and diabetes. The association between masked uncontrolled hypertension and the kidney outcome is most prominent in patients with lower levels of GFR. Our findings support the value of ABPM in identifying this group of patients at high risk for cardiovascular and kidney outcomes. Clinical use of ABPM has been limited in the past by variable insurance coverage of the costs of performing the procedure. However, recent changes by the Centers for Medicare and Medicaid services allow the use of ABPM for detecting masked hypertension.30 In this context, the robust association of masked hypertension with important clinical outcomes seen in our study supports the wider use of ABPM in patients with CKD. However, it also important to note that participants with elevated BP both on clinic and ambulatory BP (sustained hypertension) were at high risk of clinical outcomes, reinforcing the importance of BP control in preventing clinical outcomes in patients with CKD. This is consistent with recommendations regarding BP control in CKD in national guidelines.2
The mechanisms mediating the higher BP outside the clinic seen in masked hypertension remain poorly understood; adherence with medication31 and increased sympathetic nervous system activity,32 particularly with exercise,33 are some mechanisms that have been studied. Similarly, the optimal approach to management of patients with masked hypertension remains uncertain; current guidelines recommend intensification of antihypertensive drug therapy in patients with masked hypertension.2 Large clinical trials are underway in patients with essential hypertension to evaluate the best management options in masked uncontrolled hypertension.34 Our findings underscore the need for future research to better understand the mechanisms and optimal treatment of masked hypertension in patients with CKD.
The prevalence of white-coat effect in our study is consistent with other cohorts of patients with CKD, although it may vary depending on the population studied and the method of clinic BP measurement.8 The long-term prognostic significance of white-coat hypertension is unclear; a recent meta-analyses in patients with hypertension in the general population suggests that white-coat hypertension is a marker of high risk of cardiovascular disease in patients who are untreated, and not in patients treated with antihypertensive drug therapy.6 In patients with CKD, studies have shown conflicting results: some studies show no association of white-coat effect and cardiovascular disease,10 whereas others show that patients with a white-coat effect are at higher risk for mortality.13 In our cohort of patients who were mostly treated for hypertension, the presence of white-coat effect was not associated with a higher risk for cardiovascular outcomes, kidney outcomes, and mortality compared with participants with controlled BP. However, white-coat hypertension when defined using the lower threshold of 130/80 mm Hg was associated with higher risk of mortality, and white-coat hypertension was associated with higher risk of the kidney outcome in participants with lower levels of GFR. The relatively small number of patients with white-coat hypertension in our study may limit the power to detect associations; further research is needed to evaluate the long-term outcomes of patients with white-coat hypertension and CKD.
Our study demonstrates that ambulatory BP is strongly associated with of cardiovascular outcomes, kidney outcomes, and mortality, independent of clinic BP. The associations of ambulatory BP with cardiovascular outcomes were stronger in participants with higher levels of GFR. In addition, after adjustment for 24-hour systolic BP, profiles of ABPM (white-coat and masked hypertension) were no longer statistically significantly associated with clinical outcomes. This reinforces the importance of ambulatory systolic BP in patients with CKD. Of note, clinic BP, as a continuous variable, was not associated with risk of cardiovascular outcomes in our study after adjustment for traditional risk factors. This may reflect, in part, the analytic approach used in this paper which was limited to the subset with ABPM; previous papers from CRIC have shown that participants in the highest quartile of BP were at higher risk of developing cardiovascular events compared with those in the lowest quartile.35 Interestingly, in the Italian cohort of patients with CKD, office BP was also not associated with risk of clinical outcomes.16 This may reflect the fact that office BP measurement does not capture nighttime BP, which may be an important predictor of outcomes. However, it is important to note most hypertension clinical trials focused on clinic BP, and current guidelines recommend hypertension management based on clinic BP.2
Elevated nighttime BP and lack of nocturnal decline in BP are often seen in patients with CKD.21 Most,16,36 but not all,37 studies suggest that elevated nighttime BP and lack of nocturnal decline in BP are associated with high risk for clinical outcomes in patients with CKD. In our study, nighttime BP was strongly associated with kidney outcomes; in addition, participants with the reverse-dipper profile were about twice as likely to develop the kidney outcome compared with those whose BP declined at night, even after adjustment for 24-hour ambulatory systolic BP. Participants with blunted diurnal variation of BP were at higher risk for stroke and peripheral arterial disease. This underscores the value of ABPM in evaluating risk of progression of kidney disease, and the need for better understanding of the pathophysiology of elevated BP at night in these patients. Simple changes in timing of antihypertensive drug therapy have not been shown to significantly lower nighttime BP in patients with CKD.38 Newer intervention strategies, based on better understanding of the factors that influence BP at night, are needed to optimize nighttime BP; these need to be tested in well-designed studies to assess effect on clinical outcomes.
Our study has several strengths. To our knowledge, this the largest cohort of participants with CKD with ABPM measures available. Participants are well characterized, allowing for adjustment of other factors that may influence outcomes. Measurement of clinic BP by trained staff using a standardized protocol, long duration of prospective follow-up, and rigorous adjudication of clinical outcomes are other strengths of this study. However, there are important limitations: our findings are based on a single measure of ABPM, and whether or not repeated measures over time that evaluate longitudinal change improve ability to predict outcomes remains to be seen. In addition, home BP monitoring, commonly used to evaluate out-of-clinic BP, was not available in CRIC participants. Therefore, we are unable to compare home BP with ABPM as a predictor of outcome; this is an important area for future research. Recent studies have made a distinction between white-coat effect in patients treated and untreated with antihypertensive drug therapy.6 Given the relatively small number of patients with white-coat effect, and that most were treated, this could not be further evaluated in our study. Although the results of the ABPM study were shared with patients and providers, CRIC is an observational study, and management of hypertension was performed by primary clinicians; therefore, this study cannot evaluate the effect of hypertension treatment based on ABPM. Finally, the CRIC Study population has a high proportion of patients who are treated and controlled; whether or not similar results will be seen in less well-treated populations remains to be seen.
In summary, in this cohort of patients with CKD, masked uncontrolled hypertension is associated with high risk of cardiovascular disease and progression of kidney disease. BP measures obtained from ABPM are more strongly associated with cardiovascular disease, kidney disease, and mortality compared with clinic BP. Alterations of diurnal variation in BP are associated with high risk of progression of kidney disease, stroke, and peripheral arterial disease. These data support the wider use of ABPM in the evaluation of hypertension in patients with CKD. Future research is needed to better understand the pathophysiology, and to develop optimal strategies to reduce adverse outcomes among individuals with high-risk ABPM profiles.
Disclosures
All authors have nothing to disclose.
Funding
Funding for the CRIC Study was obtained under a cooperative agreement from National Institute of Diabetes and Digestive and Kidney Diseases, grants U01DK060990, U01DK060984, U01DK061022, U01DK061021, U01DK061028, U01DK060980, U01DK060963, and U01DK060902. In addition, this work was supported in part byPerelman School of Medicine at the University of Pennsylvania Clinical and Translational Science Award, National Institutes of Health (NIH)/National Center for Advancing Translational Sciences (NCATS) grant UL1TR000003, Johns Hopkins University grant UL1 TR-000424, the Clinical and Translational Science Collaborative of Cleveland, from NCATS component of NIH and NIH Roadmap for Medical Research grant UL1TR000439, the Michigan Institute for Clinical and Health Research grant UL1TR000433, the University of Illinois at Chicago Clinical and Translational Science Award grant UL1RR029879, the Tulane COBRE for Clinical and Translational Research in Cardiometabolic Diseases grant P20 GM109036-COBRA; University of Maryland grant GCRC M01 RR-16500; National Center for Research Resources grant UCSF-CTSI UL1 412 RR-02413; and the Leonard C. Rosenberg Foundation. J. Lash reports receiving grants from NIH (U01DK060980), during the conduct of the study. M. Rahman reports receiving grants from NIH (U01DK061021), during the conduct of the study. R. Townsend reports receiving grants from NIH (U01DK060984), during the conduct of the study.
Acknowledgments
Dr. Mahboob Rahman received research funding from Bayer and Duke Clinical Research Institute, and honoraria from Relypsa and Reata, unrelated to this work. Dr. Matthew R. Weir reports personal fees from Janssen, personal fees from AstraZeneca, personal fees from Merck, personal fees from Boehringer-Ingelheim, personal fees from Vifor/Relypsa, personal fees from Otsuka, and personal fees from Boston Scientific, outside the submitted work.
Dr. Mahboob Rahman, Dr. James Lash, and Dr. Raymond R. Townsend designed the study; Dr. Mahboob Rahman, Ms. Jeanne Charleston, Dr. Debbie L. Cohen, Dr. Edward J. Horowitz, Dr. James Lash, Dr. Matthew R. Weir, Dr. Paul E. Drawz, Dr. Sarah Schrauben, and Dr. Raymond R. Townsend participated in data collection; Ms. Xue Wang, Dr. Joshua D. Bundy, Ms. Jeanne Charleston, Dr. Lama Ghazi, and Dr. Dawei Xie participated in data analyses; Dr. Mahboob Rahman, Ms. Jeanne Charleston, and Dr. Paul E. Drawz drafted and revised the paper; and all authors contributed to review and edits of the manuscript, and approved the final version of the manuscript.
Supplemental Material
This article contains the following supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2020030236/-/DCSupplemental.
Supplemental Table 1. Clinical and demographic characteristics stratified by diurnal BP profile.
Supplemental Table 2. Clinical and demographic characteristics of CRIC participants enrolled in phase 2 stratified by ABPM data.
Supplemental Table 3. Association of ABPM profiles with cardiovascular mortality and components of the composite cardiovascular outcome.
Supplemental Table 4. Association between ABPM profiles (defined by 2017 ACCAHA guidelines) and clinical outcomes.
Supplemental Table 5. Association between ABPM profiles (defined by daytime BP) and clinical outcomes.
Supplemental Table 6. Association between ABPM profiles and clinical outcomes with addition of interaction terms for GFR.
Supplemental Table 7. Association of BP as a continuous variable and clinical outcomes.
Supplemental Table 8. Association of BP as a continuous variable and cardiovascular outcomes including interaction term for GFR.
Supplemental Table 9. Association between day time BP (defined between 0900 and 2100) and clinical outcomes.
Supplemental Table 10. Association of diurnal profile of BP with cardiovascular outcomes and CV mortality.
Supplemental Table 11. Association between ABPM profiles and outcomes adding 24 hours systolic BP as a covariate.
Supplemental Table 12. Association between diurnal profiles and outcomes adding 24 hours systolic BP as a covariate.
Supplemental Table 13. BP and antihypertensive drug therapy stratified by ABPM profile over the course of the study.
Supplemental Figure 1. Association between ABPM phenotypes and the composite cardiovascular outcomes in sub groups of age, gender, race and diabetes.
Supplemental Figure 2. Association between ABPM profiles and kidney outcome (ESKD or 50% decline in GFR) sub groups of age, gender, race and diabetes.
Footnotes
↵* The CRIC Study Investigators includes: Lawrence J. Appel, Jiang He, Jing Chen, Harold I. Feldman, Alan S. Go, James P. Lash, Robert G. Nelson, Mahboob Rahman, Panduranga S. Rao, Vallabh O. Shah, Raymond R. Townsend, and Mark L. Unruh.
Published online ahead of print. Publication date available at www.jasn.org.
See related editorial, “Much to Fear about MUCH,” on pages 2496–2499.
- Copyright © 2020 by the American Society of Nephrology
References
In this issue
Podcast