CLINICAL RESEARCH

www.jasn.org

Masked Uncontrolled Hypertension in CKD Rajiv Agarwal, Maria K. Pappas, and Arjun D. Sinha Indiana University School of Medicine; Richard L. Roudebush Veterans Affairs Administration Medical Center, Indianapolis, Indiana

ABSTRACT Masked uncontrolled hypertension (MUCH) is diagnosed in patients treated for hypertension who are normotensive in the clinic but hypertensive outside. In this study of 333 veterans with CKD, we prospectively evaluated the prevalence of MUCH as determined by ambulatory BP monitoring using three definitions of hypertension (daytime hypertension $135/85 mmHg; either nighttime hypertension $120/70 mmHg or daytime hypertension; and 24-hour hypertension $130/80 mmHg) or by home BP monitoring (hypertension $135/ 85 mmHg). The prevalence of MUCH was 26.7% by daytime ambulatory BP, 32.8% by 24-hour ambulatory BP, 56.1% by daytime or night-time ambulatory BP, and 50.8% by home BP. To assess the reproducibility of the diagnosis, we repeated these measurements after 4 weeks. Agreement in MUCH diagnosis by ambulatory BP was 75–78% (k coefficient for agreement, 0.44–0.51), depending on the definition used. In contrast, home BP showed an agreement of only 63% and a k coefficient of 0.25. Prevalence of MUCH increased with increasing clinic systolic BP: 2% in the 90–110 mmHg group, 17% in the 110–119 mmHg group, 34% in the 120–129 mmHg group, and 66% in the 130–139 mmHg group. Clinic BP was a good determinant of MUCH (receiver operating characteristic area under the curve 0.82; 95% confidence interval 0.76–0.87). In diagnosing MUCH, home BP was not different from clinic BP. In conclusion, among people with CKD, MUCH is common and reproducible, and should be suspected when clinic BP is in the prehypertensive range. Confirmation of MUCH diagnosis should rely on ambulatory BP monitoring.

J Am Soc Nephrol 27: ccc–ccc, 2015. doi: 10.1681/ASN.2015030243

CKD affects 11% (19.2 million) of the United States adult population;1,2 among them hypertension occurs in at least 70%.3 Controlling hypertension is an important strategy to reduce cardiovascular morbidity and mortality and end-stage renal disease. Yet some large randomized trials that have lowered BP have failed to show an improvement in cardiovascular or renal outcomes.4–6 One explanation may be that techniques to assess BP in these patients were suboptimal. Whereas non-traditional risk factors may certainly be important, data from the Framingham Heart Study demonstrate that it is the traditional risk factors that mediate the inflammatory state; this in turn impairs vascular function.7 Thus, it appears appropriate to refocus our attention on the traditional risk factors— such as blood pressure—and how to better assess and treat hypertension. Masked hypertension (MHTN), which is defined as normal BP in the clinic but elevated BP at home is seen in 10–20% of the general population that does not J Am Soc Nephrol 27: ccc–ccc, 2015

receive antihypertensive therapy.8,9 A meta-analysis of 36 studies incorporating 25,629 participants reported its prevalence as 17%.10 People receiving antihypertensive therapy who have hypertension out of office but have normal BP in the clinic are said to have masked uncontrolled hypertension (MUCH). In studies of CKD patients, all of which recruited a limited number of patients, a meta-analysis also found a high prevalence of MUCH.11 The lack of recognition of elevated BP outside the clinic may lead to inadequate

Received March 4, 2015. Accepted May 18, 2015. Published online ahead of print. Publication date available at www.jasn.org. Correspondence: Dr. Rajiv Agarwal, Department of Medicine, Indiana University School of Medicine; Richard L. Roudebush Veterans Administration Medical Center, 1481 West 10th Street, Indianapolis, IN 46202. Email: [email protected] Copyright © 2015 by the American Society of Nephrology

ISSN : 1046-6673/2703-ccc

1

CLINICAL RESEARCH

www.jasn.org

treatment of those with normal clinic BP. In the general population, data consistently show that the lack of recognition of high BP outside the clinic—masked hypertension—is associated with increased cardiovascular damage8,12 and cardiovascular morbidity and mortality.9,13–18 Although emerging data in CKD also support this concept,19–28 whether patients with CKD also have a high prevalence of MHTN, and if so what the determinants of this condition are, and how reproducible this diagnosis is, remains to be demonstrated prospectively in studies dedicated to this purpose. Furthermore, the ability of home BP to diagnose MHTN and its reproducibility remains scarcely studied even in the general population. In this prospective study, we report the prevalence of masked hypertension, test various definitions on their effect on the prevalence rates of MHTN, determine the value of baseline clinic BP on the prevalence rates of MHTN, and evaluate its reproducibility. Finally, we tested the ability of home BP to diagnose MHTN confirmed by 24-hour ambulatory BP monitoring (ABPM).

RESULTS

An overview of the study design over six visits over a 5-week period is shown in Figure 1. Baseline characteristics of participants are shown in Table 1. Typical of a veteran population, participants were older, mostly white men, and two-thirds were past smokers. A high prevalence of diabetes mellitus, sleep apnea, and cardiovascular disease was noted. The average eGFR was 44 ml/min per 1.73m2 and the median urinary albumin/creatinine ratio was 30 mg/g (interquartile range 6–228 mg/g). All but six participants were receiving antihypertensive drugs for BP control and the average number of antihypertensive drugs used was 3.1.

Bivariate Distribution of Clinic and Ambulatory BP at Baseline and Repeat Visits

All 333 participants who were recruited had screening clinic measurement of ,140/90 mmHg. Of these 333 participants, 38 had inadequate or absent ambulatory BP recordings and were not classifiable, therefore the remaining 295 formed the basis of this report. Over three visits, 33 (11%) participants had an average clinic BP $140/90 mmHg at baseline (Figure 2A). Among them, only one (3%) had isolated clinic hypertension defined as well controlled daytime and nighttime hypertension, but 32 (97%) had uncontrolled hypertension (UCH). The distribution of UCH is shown in Figure 2B: one (3%) had isolated daytime hypertension (UCH-D), four (13%) had isolated nighttime hypertension (UCH-N), and 27 (84%) had both daytime and nighttime hypertension (UCH-DN). At baseline, 262 participants had clinic BP ,140/90 mmHg. Of these, 115 (44%) had well controlled ambulatory BP both during the day and during the night (controlled hypertension) whereas 147 (56%) had MUCH (Figure 2A). The distribution of MUCH was as follows: eight (6%) had isolated daytime hypertension (MUCH-D), 77 (52%) had isolated nighttime hypertension (MUCH-N), and 62 (42%) had both daytime and nighttime hypertension (MUCH-DN) (Figure 2B). Repeat measurements 4 weeks later were available in 274 of 295 (93%) participants with classifiable hypertension at baseline. The proportions of participants with isolated clinic hypertension, UCH, controlled hypertension, and MUCH at the repeat visit were similar to those noted at baseline (Figure 2C). Furthermore the categories of UCH and MUCH were also similar (Figure 2D). Prevalence of MUCH Using Various Definitions

Using the conventional definition of MUCH as clinic BP ,140/90 mmHg and daytime ambulatory BP $135/85 mmHg, 70 (27%) participants had MUCH whereas 192 (73%) had well controlled hypertension (Table 2). Repeat visit revealed prevalence of MUCH as 28% (n=69) and that of well controlled hypertension as 72% (n=177). In contrast to the above definition, using the 24-hour ABPM threshold of $130/80 mmHg to define hypertension revealed a higher prevalence of MUCH: 33% at baseline visit and 37% at repeat visit. Furthermore, by using the most liberal definition of daytime or nighttime hypertension increased the prevalence of MUCH to 56% and 57% at baseline and repeat visit, respectively. In comparison to ABPM-based definitions, home BP monitoring-based diagnosis revealed the prevalence of MUCH to be about 46% on each of the two occasions.

Figure 1. Overview of study design. After a brief history and physical examination, participants had BP measured in a seated position. The technique of HBPM was explained and a self-inflating oscillometric device was dispensed. For one week, each participant recorded home BP twice daily. ABPM was performed over 24 hours. After one month hiatus, the study was repeated as in the initial month. CBP, clinic BP; HBPM, home BP monitoring.

2

Journal of the American Society of Nephrology

Reproducibility of Categories and the Diagnosis of MUCH

There were 204 pairs of complete datasets available for clinic and ambulatory BP J Am Soc Nephrol 27: ccc–ccc, 2015

www.jasn.org

CLINICAL RESEARCH

Table 1. Baseline characteristics of the study sample Variable

All (n=333)

With ABPM at week 1 (n=295)

Clinic NT at week 1 (n=262)

With ABPM at week 4 (n=274)

Clinic NT at week 4 (n=246)

Age (years) Male sex, n (%) Race White, n (%) Black, n (%) Other, n (%) Hispanic, n (%) Weight (kg) Body mass index, (kg/m2) Comorbid illnesses Diabetes mellitus, n (%) Diabetic retinopathy, n (%) Diabetic neuropathy, n (%) Myocardial infarction, n (%) Percutaneous coronary revascularization, n (%) Coronary artery bypass graft, n (%) Congestive heart failure, n (%) Stroke, n (%) Cardiac resuscitation, n (%) Pacemaker or defibrillator, n (%) Peripheral vascular disease, n (%) Sleep apnea, n (%) Hepatitis C, n (%) Smoking Never, n (%) Past, n (%) Current, n (%) Laboratory tests Albumin (g/dl) Hemoglobin (g/dl) Estimated glomerular filtration rate (ml/min per 1.73 m2) Urine albumin/creatinine ratio (g/g) Antihypertensive drugs (n) Angiotensin converting enzyme inhibitors, n (%) Angiotension receptor blockers, n (%) b-Blockers, n (%) a-Blockers, n (%) Centrally acting agents, n (%) Dihydropyridine calcium-channel blockers, n (%) Potassium-sparing diuretics, n (%) Loop diuretics, n (%) Non-dihydropyridine calcium-channel blockers, n (%) Thiazide diuretics, n (%) Vasodilators, n (%)

69.469.9 327 (98%)

69.569.9 290 (98%)

69.369.9 257 (98%)

69.5610 269 (98%)

69.9610 241 (98%)

268 (80%) 55 (17%) 10 (3%) 4 (1%) 93.8616.9 30.764.7

236 (80%) 50 (17%) 9 (3%) 4 (1%) 94.1617.1 30.864.7

213 (81%) 40 (15%) 9 (3%) 4 (2%) 94.7617.3 30.964.8

221 (81%) 44 (16%) 9 (3%) 3 (1%) 93.9616.9 30.864.6

201 (82%) 36 (15%) 9 (4%) 3 (1%) 94.3616.8 30.964.5

219 (66%) 73 (33%) 101 (46%) 97 (29%) 81 (24%) 74 (22%) 67 (20%) 42 (13%) 6 (2%) 37 (11%) 74 (22%) 92 (28%) 17 (5%)

195 (66%) 66 (22%) 89 (30%) 87 (30%) 76 (26%) 64 (22%) 58 (20%) 35 (12%) 5 (2%) 31 (11%) 64 (22%) 80 (27%) 13 (4%)

169 (65%) 56 (21%) 75 (29%) 77 (30%) 68 (26%) 55 (21%) 48 (18%) 31 (12%) 3 (1%) 29 (11%) 52 (20%) 69 (26%) 10 (4%)

175 (64%) 55 (20%) 76 (28%) 78 (28%) 66 (24%) 61 (22%) 52 (19%) 35 (13%) 5 (2%) 26 (9%) 60 (22%) 74 (27%) 13 (5%)

154 (63%) 44 (18%) 61 (25%) 71 (29%) 62 (25%) 55 (22%) 45 (18%) 31 (13%) 5 (2%) 26 (11%) 49 (20%) 63 (26%) 8 (3%)

54 (16%) 222 (67%) 57 (17%)

47 (16%) 201 (68%) 47 (16%)

44 (17%) 177 (68%) 41 (16%)

41 (15%) 185 (68%) 48 (18%)

38 (15%) 168 (68%) 40 (16%)

460.5 13.361.8 44.1615.8

460.5 13.261.8 44615.6

460.5 13.361.7 44.5615.8

460.5 13.461.8 44.1615.3

460.5 13.461.8 44.4615.5

0.3360.88 3.161.4 179 (54%) 66 (20%) 227 (68%) 114 (34%) 25 (8%) 140 (42%) 34 (10%) 129 (39%) 14 (4%)

0.3360.85 3.261.4 155 (53%) 64 (22%) 202 (68%) 107 (36%) 23 (8%) 132 (45%) 29 (10%) 114 (39%) 12 (4%)

0.2560.57 3.161.4 141 (54%) 57 (22%) 177 (68%) 98 (37%) 16 (6%) 112 (43%) 28 (11%) 96 (37%) 11 (4%)

0.360.87 3.161.4 145 (53%) 60 (22%) 186 (68%) 95 (35%) 22 (8%) 110 (40%) 30 (11%) 107 (39%) 12 (4%)

0.2760.86 3.161.4 129 (52%) 56 (23%) 167 (68%) 85 (35%) 16 (7%) 92 (37%) 29 (12%) 94 (38%) 12 (5%)

81 (24%) 20 (6%)

72 (24%) 20 (7%)

66 (25%) 13 (5%)

69 (25%) 17 (6%)

64 (26%) 13 (5%)

NT, normotension.

recordings among participants with controlled clinic BP at both weeks that formed the basis of assessment of reproducibility. Agreement was between 75% and 78% regardless of the definition of MUCH (Table 3). The k coefficient for agreement was between 0.441 and 0.509, which is considered moderate. In contrast to ABPM-based diagnoses, in the case of home BP-based diagnosis of MUCH, the agreement J Am Soc Nephrol 27: ccc–ccc, 2015

was only about 63% and k coefficient 0.249, which is considered fair. Effect of Clinic BP on the Prevalence of MUCH

After excluding one participant who had isolated diastolic hypertension, the prevalence of controlled hypertension and UCH stratified by baseline clinic systolic BP is shown in Table 4. MUCH-CKD

3

CLINICAL RESEARCH

www.jasn.org

Figure 2. Bivariate distribution of hypertension categories by visit. (A) Number of participants who had hypertensive clinic BP (left bar) or were normotensive (right bar). If clinic BP was in the hypertensive range, but ambulatory BP was normal isolated clinic hypertension (ICH) was diagnosed, or if ambulatory BP was high, they were said to have uncontrolled hypertension (UCH). If clinic BP was normal and ambulatory BP was normal they had controlled hypertension (CH), or if ambulatory BP was elevated they were diagnosed with MUCH. (B) The reason why participants were diagnosed with UCH or MUCH. UCH during daytime only (UCH-D), daytime or nightime (UCH-DN), or nighttime only (UCH-N) pie-chart is shown. Similarly MUCH breakdown as a pie chart by ambulatory BP monitoring elevation during daytime (MUCH-D), daytime or nighttime (MUCH-DN), and nighttime alone (MUCH-N) are shown. (C) and (D) Data from repeat measurements at week 4.

Increasing levels of clinic systolic BP was associated with increasing numbers of UCH and reducing numbers of controlled hypertension. Accordingly, the prevalence of MUCH increased with increasing clinic systolic BP. Compared with clinic systolic BP ,110 mmHg, increments of 10 mmHg clinic systolic BP were associated with odds ratios of 13.4, 33.8, and 128.3 at baseline visit and 4.3, 13.9, and 28.3 at 4 weeks. Each

10 mmHg increment in clinic systolic BP increased the odds of MUCH by 3.5 (95% confidence interval (95% CI), 2.4–5.0) at baseline visit and 2.7 (95% CI, 2.0–3.8) at 4 weeks. Multivariable adjustments for the following characteristics: age, race, diabetes, smoking, cardiovascular disease (defined as myocardial infarction, stroke, peripheral vascular disease and congestive heart failure), number of antihypertensive drugs, eGFR,

Table 2. Prevalence of masked hypertension using different definitions and over two time periods Baseline visit Daytime ABPM 24-Hour ABPM Daytime and nighttime ABPM Home BP

Repeat visit at 4 weeks

CH (n)

UCH (n)

MUCH (%)

95% CI

CH (n)

UCH (n)

MUCH (%)

95% CI

192 176 115 140

70 86 147 122

26.7 32.8 56.1 46.6

21.3–32.1% 27.1–38.5% 50.1–62.2% 40.5–52.6%

177 154 106 132

69 92 140 114

28.0 37.4 56.9 46.3

22.4–33.7% 31.3–43.5% 50.7–63.1% 40.1–52.6%

MUCH is calculated as UCH3100%/(CH+UCH).

4

Journal of the American Society of Nephrology

J Am Soc Nephrol 27: ccc–ccc, 2015

www.jasn.org

Table 3. Reproducibility of masked hypertension using different definitions Concordance statistics Agreement Daytime ABPM 24-Hour ABPM Daytime and nighttime ABPM Home BP

k

95% CI k

P level

78.4% 76.5% 75.5%

0.441 0.305–0.577 ,0.001 0.474 0.339–0.609 ,0.001 0.509 0.372–0.645 ,0.001

62.8%

0.249 0.112–0.386 ,0.001

and log urinary albumin/creatinine ratio did not remove the statistical significance of increasing clinic systolic BP on the odds of MUCH (Table 4). Ability of Home BP to Diagnose Ambulatory BP Diagnosed MUCH

Using the conventional definition of MUCH (elevated day time ambulatory BP ($135/85 mmHg but normal clinic BP ,140/ 90 mmHg), we asked the question whether home BP can detect MUCH and if so what are the diagnostic characteristics of this test. We used receiver operating characteristic (ROC) curves to evaluate the performance of home BP in making a diagnosis of MUCH. The ROC curve to diagnose MUCH using home systolic BP is shown in Figure 3A and B for baseline visit and week 4 visit, respectively. The area under the curve (AUC) of the ROC curve was 0.78 (95% CI, 0.71–0.85) at baseline and 0.72 (95% CI, 0.65– 0.79) at week 4. However, given the above observations that baseline clinic systolic BP is a strong determinant of MUCH, we calculated the diagnostic performance of clinic systolic BP in diagnosing MUCH. The AUC of the ROC curve was 0.82 (95% CI, 0.76–0.87) at baseline and 0.78 (95% CI, 0.72–0.84) at week 4. Comparison of home BP-based ROC AUC to clinic BP-based AUC at baseline (P=0.39) and week 4 (P=0.14) revealed no significant differences in their abilities to diagnose MUCH.

DISCUSSION

The definition of MUCH has an important bearing on the prevalence of this condition. The original definition proposed by Pickering and colleagues29 and one used by the International Database on Ambulatory Blood Pressure in Relation to Cardiovascular Outcomes (IDACO) investigators30 do not take into

CLINICAL RESEARCH

account BP recordings at night. The conventional definition of MUCH used in these studies relies on clinic BP ,140/90 mmHg and daytime ambulatory BP $135/85 mmHg. Using this definition, among participants with CKD with normal clinic BP, the prevalence of MUCH was 27%. However, some investigators use elevation of either daytime or nighttime BP to diagnose MUCH.23 Using this definition more than doubles the prevalence of MUCH to 56%. This is similar to the prevalence reported in those studies. For example, from a Spanish registry of ABPM, 14,840 subjects with treated and controlled conventional BP, MUCH diagnosed by 24-hour ABPM was prevalent in 31%.31 Nighttime hypertension was the exclusive abnormality in 24%. In comparison, in our study, nearly 95% had nighttime BP elevation, and it was the exclusive abnormality in half the participants. This is not surprising given the observation that nondipping among patients with CKD is common;23,32 in fact, it antedates the occurrence of albuminuria in type 1 diabetes mellitus33 and much before GFR declines.34 A second important observation made in this study is that the prevalence of MUCH is strongly dependent on the level of clinic BP. Thus, those who repeatedly have a low clinic systolic BP (,110 mmHg) are unlikely to have MUCH. However, among those with usual clinic BP of 130–139 mmHg, MUCH is prevalent in two of three and those with usual clinic BP of 120–129 mmHg, MUCH is prevalent in one of three. Even among individuals with clinic BP of 110–119 mmHg, the prevalence of MUCH was approximately one in six. To allow direct comparison among cohorts, future studies on the prevalence of MUCH should report prevalence stratified by levels of clinic BP. The reproducibility of the diagnosis of MUCH has previously been tested among 50 untreated participants who had borderline hypertension.35 Using daytime ambulatory BP of $135/85 mmHg to define hypertension, the prevalence of masked hypertension at first visit was 54% and a week later was 53%. The high rates likely reflect the higher baseline BP as most participants already had borderline hypertension. Daytime ambulatory BP was concordant in classifying participants 73% of the time but the k statistic was only 0.47. In a retrospective cohort in which 80% of the participants were being treated and MUCH was present in 25 of 196 (13%) participants, after a mean period of 1.5 years, the k statistic was only 0.26.36 However, treatment for hypertension could be changed and time elapsed between BP recordings was variable; the standard deviation of elapsed time between recordings was 1.5 years. Among 503 untreated Japanese participants, using morning home BP as the standard for out-of-office

Table 4. Prevalence of MUCH defined by daytime ABPM stratified by clinic BP Baseline visit Clinic systolic BP (mmHg) CH (n) UCH (n) MUCH (%) ,110 110–119 120–129 130–139

66 69 39 18

1 14 20 35

1.5 16.9 33.9 66.0

Return visit at 4 weeks OR (95% CI)

aOR

1 1 13.4 (1.7–104.7) 9.1 (1.1–74.9) 33.8 (4.4–262.1) 21.2 (2.6–172.1) 128.3 (16.4–1001.8) 90.3 (11.1–734.3)

CH (n) UCH (n) MUCH (%) 62 53 43 19

3 11 29 26

4.6 17.2 40.3 57.8

OR

aOR

1 1 4.3 (1.1–16.2) 4.9 (1.1–21.2) 13.9 (4.0–48.7) 17.1 (4.1–70.9) 28.3 (7.7–103.9) 43.9 (9.6–201.7)

aOR, adjusted odds ratio; OR, odds ratio. Adjustments were for age, diabetes, smoking, cardiovascular disease, number of antihypertensive drugs, eGFR, and log urinary albumin/creatinine ratio.

J Am Soc Nephrol 27: ccc–ccc, 2015

MUCH-CKD

5

CLINICAL RESEARCH

www.jasn.org

home BP measurement may be stressful, and readings during the first few hours after waking may be higher than during the rest of the day.39 This might explain why the prevalence of MUCH detected by home BP recordings was higher. Prior studies which suggested the usefulness of using home BP for diagnosing MUCH likely ignored the independent contribution of clinic BP for diagnosing this condition. Nonetheless, our study does not discount the importance of home BP monitoring. MUCH diagnosed by home BP monitoring is associated with an increased risk of cardiovascular events in elderly treated patients with essential hypertension.9 Among those with CKD, MUCH diagnosed by home BP monitoring is associated with an increased risk of ESRD.20,40 An individual level meta-analysis reported by investigators of the International Database of Home Blood Pressure in relation to Cardiovascular OutFigure 3. Receiver operating characteristic (ROC) curves for the diagnosis of MUCH). ROC curves for the diagnostic test performance of systolic clinic and home BP in predicting comes, from five populations in 5008 particMUCH diagnosed by daytime ambulatory hypertension (135/85 mmHg) at baseline visit (A) ipants, suggests that home BP monitoring and after 1 month (B). The area under the curve (AUC) for clinic and home BP monitoring can refine cardiovascular risk assessment above and beyond clinic BP.41 This is particwere similar (P=0.38 and 0.14, respectively) at each of the two visits. ularly true when clinic BP is in the normal range (i.e., among those with MHTN or MUCH). Accordingly, measurement, k coefficient for agreement was 0.58 at 6 months.37 a more logical explanation of our findings is that home BP Our study is the largest specifically to test the reproducibility of monitoring-diagnosed MUCH may detect a different phenotype. MUCH after 4 weeks. Our study had a high completion rate with Our study has limitations. Participants in our study were 295 participants completing the first set of visits and 274 participants restricted to veterans who are older and are predominantly men. completing the second set 4 weeks later. We found agreement of between 75% and 78% and k of 0.44–0.51 similar to that reported Whether our findings apply to younger people and women will need future studies. There are several strengths of our study: our in those with MHTN in a cohort of people with untreated borderline hypertension. The observed concordance in the diagnosis of MUCH study used 24-hour ABPM,the gold-standard method to diagnose suggests that MUCH is more than a statistical phenomenon and may out-of-office hypertension. Our study was prospective, it carefully phenotyped the BP pattern in each individual using several BP have a biologic basis. monitoring methods, used multiple clinic visits to define clinic In this cohort, home BP monitoring detected a prevalence of MUCH that was nearly 70% more than that of the standard hypertension, had a high completion rate, and it assessed reproducibility of MUCH using an adequate sample size for definition of MUCH using ABPM (46% versus 27%). In contrast to ambulatory BP-based definitions, home BP-based diagnosis this purpose. The three clinical implications of our findings are as follows: had lower agreement and poor k suggesting only fair reproduc(1) Among patients with CKD, nearly all of whom were taking ibility. Furthermore, home BP measurements were no better than antihypertensive medications, MUCH is likely to be seen in as clinic BP in making a diagnosis of MUCH. Numerically, the AUC many as one of six if clinic BP is 110–119 mmHg, twice as many under ROC curves was greater for clinic BP than home BP in (one in three) if clinic BP is 120–129 mmHg, and four times as diagnosing MUCH. Thus, there is little reason to suspect lack of many (two of three) if the usual clinic BP is 130–139 mmHg. statistical power as a reason for failure of home BP over clinic BP in diagnosing MUCH. It is possible that a larger sample size Only if clinic BP is ,110 mmHg, is the likelihood of MUCH low (,5%). (2) The use of home BP monitoring to diagnose MUCH might have demonstrated that clinic BP significantly outperforms in CKD is not supported by our study, however we recognize that home BP. Taken together, these findings extend the observations home BP monitoring may detect a different phenotype and that of Viera et al. who also pointed out the inadequacy of home BP MUCH/MHTN diagnosed by home BP monitoring is indepenmonitoring in detecting MHTN.35 In contrast to daytime ambulatory BP recordings, many of which may be taken during activity, dently associated with cardiovascular outcomes. (3) The shortterm reproducibility of MUCH diagnosed by ambulatory BP home BP recordings are always made during rest. Given that measurement regardless of the definition is between 75% and activity generally increases systolic BP it is surprising that home 78%; therefore MUCH may be more than a statistical BP found a higher prevalence of MUCH.38 However, the act of 6

Journal of the American Society of Nephrology

J Am Soc Nephrol 27: ccc–ccc, 2015

www.jasn.org

phenomenon and have a biologic basis.42 Future studies will define the biologic basis of MUCH and the long-term outcomes associated with this common disorder.

CONCISE METHODS This is a prospective study of CKD patients stages 2 to 4 (eGFR defined using the Modification of Diet in Renal Disease (MDRD) equation ,90 ml/min per 1.73 m2 but .15 ml/min per 1.73 m2). For those with stage 2 CKD, albuminuria (A2 or .300 mg/g creatinine) was required for inclusion in the cohort. Those with an initial clinic BP of 140/90 mmHg or less were considered eligible and studied further. Detailed inclusion and exclusion criteria are shown in the supplemental Appendix. After obtaining a clinical history, performing a physical examination, and obtaining basic laboratory tests, measurements of BP in the clinic (average of three visits), home (one week average) and by 24-hour ambulatory monitoring (24 hour average) were performed (for details on measurements, please see supplemental Appendix). BP measurements were repeated after one month to assess their reproducibility. Only those diagnoses confirmed by physician review of the electronic chart of the patient are reported here.

Classification of Hypertension Masked hypertension is a term reserved for people not receiving antihypertensive medications. Given that there were only six participants in our study who were not receiving antihypertensive therapy, we used the term masked uncontrolled hypertension (MUCH) to define participants who had controlled clinic BP (,140/90 mmHg on average of three clinic visits by oscillometric BP measurement) but elevated ambulatory BP.42 Elevated ambulatory BP was defined three different ways: (1) elevated daytime ($135/85 mmHg); (2) elevated 24-hour ($130/80 mmHg); and (3) either elevated daytime or elevated nighttime ($120/70 mmHg) ambulatory BP. To define daytime and nighttime we used patient diaries as noted in the supplemental Appendix. The average weekly home BP was used to define home hypertension if BP was $135/85 mmHg. The prevalence rates of MUCH for each of these definitions and their 95% CIs are reported.

CLINICAL RESEARCH

with MHTN and excellent if 80–100% were diagnosed with MHTN. We calculated that if 78 patients had MHTN it would allow us to estimate reproducibility with a 10.2% margin of error when the true reproducibility rate was 70%. This sample size of 78 patients would also give us a 95% CI to estimate an excellent reproducibility with a 6.7% margin of error when the true reproducibility rate was 90%.

Statistical Analysis Software for data entry at front end was designed for manual and electronic import. For example, home BP and ambulatory BP data were electronically imported. Stored in a relational database, datawere queried for accuracyof data entry and quality. Means, standard deviations, counts and proportions were calculated for baseline data. Prevalence rates of MUCH and their 95% CIs were calculated. Reproducibility was evaluated by using percentage agreement and k statistic. Ordinal logistic regression was used to evaluate the association of declines of systolic clinic BP with the diagnosis of MUCH. Logistic regression was used to evaluate the association of each 10 mmHg increase in clinic systolic BP with the odds of MUCH. Multivariate adjustments were made based on a careful review of the literature accounting for variables associated with MUCH. The ability of clinic and home BP to detect MUCH was evaluated using receiver operating-characteristic curves. All statistical analyses were done with Stata 13.1 (StataCorp., College Station, TX). Nominal level of statistical significance was taken as a two-sided P value of 0.05.

ACKNOWLEDGMENTS This study was supported by a grant from VA Merit Review.

DISCLOSURES R.A. has consulted for several pharmaceutical companies that make antihypertensive drugs including Merck, Takeda, Novartis, Daiichi Sankyo, Abbvie, Bayer, and Johnson and Johnson. The other authors have nothing to declare.

REFERENCES Determination of Sample Size

Estimation of Sample Size for Prevalence of Hypertension In our preliminary study we found the prevalence of masked hypertension by standardized measurements of clinic BP to be 26% (95% CI, 13–39%). We estimated that a sample size of 300 subjects with controlled hypertension in the clinic will allow us to estimate the prevalence of MHTN with a 3.8% margin of error and 95% CI when the true prevalence is as low as 13%, and with a 5.5% margin of error if the true prevalence is as high as 39%.

Estimation of Sample Size for Reproducibility We estimated that approximately 78 patients (26% of 300) will have masked hypertension in a sample of 300 patients who complete both visits. At one month we sought a diagnosis of MHTN by repeating the study. We defined reproducibility to be modest if 60–79% were diagnosed J Am Soc Nephrol 27: ccc–ccc, 2015

1. Coresh J, Astor BC, Greene T, Eknoyan G, Levey AS: Prevalence of chronic kidney disease and decreased kidney function in the adult US population: Third National Health and Nutrition Examination Survey. Am J Kidney Dis 41: 1–12, 2003 2. Coresh J, Selvin E, Stevens LA, Manzi J, Kusek JW, Eggers P, Van Lente F, Levey AS: Prevalence of chronic kidney disease in the United States. JAMA 298: 2038–2047, 2007 3. Coresh J, Wei GL, McQuillan G, Brancati FL, Levey AS, Jones C, Klag MJ: Prevalence of high blood pressure and elevated serum creatinine level in the United States: findings from the third National Health and Nutrition Examination Survey (1988–1994). Arch Intern Med 161: 1207– 1216, 2001 4. Klahr S, Levey AS, Beck GJ, Caggiula AW, Hunsicker L, Kusek JW, Striker G; Modification of Diet in Renal Disease Study Group: The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease. N Engl J Med 330: 877–884, 1994

MUCH-CKD

7

CLINICAL RESEARCH

www.jasn.org

5. Agodoa LY, Appel L, Bakris GL, Beck G, Bourgoignie J, Briggs JP, Charleston J, Cheek D, Cleveland W, Douglas JG, Douglas M, Dowie D, Faulkner M, Gabriel A, Gassman J, Greene T, Hall Y, Hebert L, Hiremath L, Jamerson K, Johnson CJ, Kopple J, Kusek J, Lash J, Lea J, Lewis JB, Lipkowitz M, Massry S, Middleton J, Miller ER III, Norris K, O’Connor D, Ojo A, Phillips RA, Pogue V, Rahman M, Randall OS, Rostand S, Schulman G, Smith W, Thornley-Brown D, Tisher CC, Toto RD, Wright JT Jr, Xu S; African American Study of Kidney Disease and Hypertension (AASK) Study Group: Effect of ramipril vs amlodipine on renal outcomes in hypertensive nephrosclerosis: a randomized controlled trial. JAMA 285: 2719–2728, 2001 6. Ruggenenti P, Perna A, Gherardi G, Garini G, Zoccali C, Salvadori M, Scolari F, Schena FP, Remuzzi G: Renoprotective properties of ACEinhibition in non-diabetic nephropathies with non-nephrotic proteinuria. [see comments] Lancet 354: 359–364, 1999 7. Vita JA, Keaney JF Jr, Larson MG, Keyes MJ, Massaro JM, Lipinska I, Lehman BT, Fan S, Osypiuk E, Wilson PW, Vasan RS, Mitchell GF, Benjamin EJ: Brachial artery vasodilator function and systemic inflammation in the Framingham Offspring Study. Circulation 110: 3604– 3609, 2004 8. Liu JE, Roman MJ, Pini R, Schwartz JE, Pickering TG, Devereux RB: Cardiac and arterial target organ damage in adults with elevated ambulatory and normal office blood pressure. Ann Intern Med 131: 564–572, 1999 9. Bobrie G, Chatellier G, Genes N, Clerson P, Vaur L, Vaisse B, Menard J, Mallion JM: Cardiovascular prognosis of “masked hypertension” detected by blood pressure self-measurement in elderly treated hypertensive patients. JAMA 291: 1342–1349, 2004 10. Verberk WJ, Kessels AG, de Leeuw PW: Prevalence, causes, and consequences of masked hypertension: a meta-analysis. Am J Hypertens 21: 969–975, 2008 11. Bangash F, Agarwal R: Masked hypertension and white-coat hypertension in chronic kidney disease: a meta-analysis. Clin J Am Soc Nephrol 4: 656–664, 2009 12. Cuspidi C, Sala C, Tadic M, Rescaldani M, De Giorgi GA, Grassi G, Mancia G: Untreated masked hypertension and carotid atherosclerosis: a meta-analysis. Blood Press 24: 65–71, 2015 13. Mancia G, Facchetti R, Bombelli M, Grassi G, Sega R: Long-term risk of mortality associated with selective and combined elevation in office, home, and ambulatory blood pressure. Hypertension 47: 846–853, 2006 14. Bobrie G, Clerson P, Ménard J, Postel-Vinay N, Chatellier G, Plouin PF: Masked hypertension: a systematic review. J Hypertens 26: 1715– 1725, 2008 15. Pierdomenico SD, Cuccurullo F: Prognostic value of white-coat and masked hypertension diagnosed by ambulatory monitoring in initially untreated subjects: an updated meta analysis. Am J Hypertens 24: 52– 58, 2011 16. Björklund K, Lind L, Zethelius B, Andrén B, Lithell H: Isolated ambulatory hypertension predicts cardiovascular morbidity in elderly men. Circulation 107: 1297–1302, 2003 17. Fagard RH, Cornelissen VA: Incidence of cardiovascular events in white-coat, masked and sustained hypertension versus true normotension: a meta-analysis. J Hypertens 25: 2193–2198, 2007 18. Stergiou GS, Asayama K, Thijs L, Kollias A, Niiranen TJ, Hozawa A, Boggia J, Johansson JK, Ohkubo T, Tsuji I, Jula AM, Imai Y, Staessen JA; International Database on HOme blood pressure in relation to Cardiovascular Outcome (IDHOCO) Investigators: Prognosis of whitecoat and masked hypertension: International Database of HOme blood pressure in relation to Cardiovascular Outcome. Hypertension 63: 675– 682, 2014 19. Minutolo R, Gabbai FB, Agarwal R, Chiodini P, Borrelli S, Bellizzi V, Nappi F, Stanzione G, Conte G, De Nicola L: Assessment of achieved clinic and ambulatory blood pressure recordings and outcomes during treatment in hypertensive patients with CKD: a multicenter prospective cohort study. Am J Kidney Dis 64: 744–752, 2014

8

Journal of the American Society of Nephrology

20. Minutolo R, Agarwal R, Borrelli S, Chiodini P, Bellizzi V, Nappi F, Cianciaruso B, Zamboli P, Conte G, Gabbai FB, De Nicola L: Prognostic role of ambulatory blood pressure measurement in patients with nondialysis chronic kidney disease. Arch Intern Med 171: 1090– 1098, 2011 21. Mitsnefes M, Flynn J, Cohn S, Samuels J, Blydt-Hansen T, Saland J, Kimball T, Furth S, Warady B; CKiD Study Group: Masked hypertension associates with left ventricular hypertrophy in children with CKD. J Am Soc Nephrol 21: 137–144, 2010 22. Okada T, Nakao T, Matsumoto H, Nagaoka Y, Tomaru R, Iwasawa H, Wada T: Prognostic significance of home blood pressure control on renal and cardiovascular outcomes in elderly patients with chronic kidney disease. Hypertens Res 32: 1123–1129, 2009 23. Pogue V, Rahman M, Lipkowitz M, Toto R, Miller E, Faulkner M, Rostand S, Hiremath L, Sika M, Kendrick C, Hu B, Greene T, Appel L, Phillips RA; African American Study of Kidney Disease and Hypertension Collaborative Research Group: Disparate estimates of hypertension control from ambulatory and clinic blood pressure measurements in hypertensive kidney disease. Hypertension 53: 20–27, 2009 24. Kayrak M, Gul EE, Kaya C, Solak Y, Turkmen K, Yazici R, Guney I, Altintepe L, Turk S, Ozdemir K: Masked hypertension in renal transplant recipients. Blood Press 23: 47–53, 2014 25. Cha RH, Kim S, Ae Yoon S, Ryu DR, Eun Oh J, Han SY, Young Lee E, Ki Kim D, Kim YS: Association between blood pressure and target organ damage in patients with chronic kidney disease and hypertension: results of the APrODiTe study. Hypertens Res 37: 172–178, 2014 26. Czy_zewski Ł, Wyzgał J, Kołek A: Evaluation of selected risk factors of cardiovascular diseases among patients after kidney transplantation, with particular focus on the role of 24-hour automatic blood pressure measurement in the diagnosis of hypertension: an introductory report. Ann Transplant 19: 188–198, 2014 27. Iimuro S, Imai E, Watanabe T, Nitta K, Akizawa T, Matsuo S, Makino H, Ohashi Y, Hishida A; Chronic Kidney Disease Japan Cohort Study Group: Clinical correlates of ambulatory BP monitoring among patients with CKD. Clin J Am Soc Nephrol 8: 721–730, 2013 28. Gabbai FB, Rahman M, Hu B, Appel LJ, Charleston J, Contreras G, Faulkner ML, Hiremath L, Jamerson KA, Lea JP, Lipkowitz MS, Pogue VA, Rostand SG, Smogorzewski MJ, Wright JT, Greene T, Gassman J, Wang X, Phillips RA; African American Study of Kidney Disease and Hypertension (AASK) Study Group: Relationship between ambulatory BP and clinical outcomes in patients with hypertensive CKD. Clin J Am Soc Nephrol 7: 1770–1776, 2012 29. Pickering TG, Davidson K, Gerin W, Schwartz JE: Masked hypertension. Hypertension 40: 795–796, 2002 30. Franklin SS, Thijs L, Li Y, Hansen TW, Boggia J, Liu Y, Asayama K, Björklund-Bodegård K, Ohkubo T, Jeppesen J, Torp-Pedersen C, Dolan E, Kuznetsova T, Stolarz-Skrzypek K, Tikhonoff V, Malyutina S, Casiglia E, Nikitin Y, Lind L, Sandoya E, Kawecka-Jaszcz K, Filipovsky J, Imai Y, Wang J, Ibsen H, O’Brien E, Staessen JA; International Database on Ambulatory blood pressure in Relation to Cardiovascular Outcomes Investigators: Masked hypertension in diabetes mellitus: treatment implications for clinical practice. Hypertension 61: 964–971, 2013 31. Banegas JR, Ruilope LM, de la Sierra A, de la Cruz JJ, Gorostidi M, Segura J, Martell N, García-Puig J, Deanfield J, Williams B: High prevalence of masked uncontrolled hypertension in people with treated hypertension. Eur Heart J 35: 3304–3312, 2014 32. Agarwal R: Regulation of circadian blood pressure: from mice to astronauts. Curr Opin Nephrol Hypertens 19: 51–58, 2010 33. Lurbe E, Redon J, Kesani A, Pascual JM, Tacons J, Alvarez V, Batlle D: Increase in nocturnal blood pressure and progression to microalbuminuria in type 1 diabetes. N Engl J Med 347: 797–805, 2002 34. Agarwal R, Light RP: GFR, proteinuria and circadian blood pressure. Nephrol Dial Transplant 24: 2400–2406, 2009

J Am Soc Nephrol 27: ccc–ccc, 2015

www.jasn.org

35. Viera AJ, Hinderliter AL, Kshirsagar AV, Fine J, Dominik R: Reproducibility of masked hypertension in adults with untreated borderline office blood pressure: comparison of ambulatory and home monitoring. Am J Hypertens 23: 1190–1197, 2010 36. Ben-Dov IZ, Ben-Arie L, Mekler J, Bursztyn M: Reproducibility of whitecoat and masked hypertension in ambulatory BP monitoring. Int J Cardiol 117: 355–359, 2007 37. Kawabe H, Saito I: Reproducibility of masked hypertension determined from morning and evening home blood pressure measurements over a 6-month period. Hypertens Res 30: 845–851, 2007 38. Agarwal R, Light RP: Physical activity and hemodynamic reactivity in chronic kidney disease. Clin J Am Soc Nephrol 3: 1660–1668, 2008 39. Kario K, Pickering TG, Umeda Y, Hoshide S, Hoshide Y, Morinari M, Murata M, Kuroda T, Schwartz JE, Shimada K: Morning surge in blood pressure as a predictor of silent and clinical cerebrovascular disease in elderly hypertensives: a prospective study. Circulation 107: 1401– 1406, 2003

J Am Soc Nephrol 27: ccc–ccc, 2015

CLINICAL RESEARCH

40. Agarwal R, Andersen MJ: Prognostic importance of clinic and home blood pressure recordings in patients with chronic kidney disease. Kidney Int 69: 406–411, 2006 41. Asayama K, Thijs L, Brguljan-Hitij J, Niiranen TJ, Hozawa A, Boggia J, Aparicio LS, Hara A, Johansson JK, Ohkubo T, Tzourio C, Stergiou GS, Sandoya E, Tsuji I, Jula AM, Imai Y, Staessen JA; International Database of Home Blood Pressure in Relation to Cardiovascular Outcome (IDHOCO) investigators: Risk stratification by self-measured home blood pressure across categories of conventional blood pressure: a participant-level meta-analysis. PLoS Med 11: e1001591, 2014 42. Franklin SS, O’Brien E, Thijs L, Asayama K, Staessen JA: Masked hypertension: a phenomenon of measurement. Hypertension 65: 16–20, 2015

This article contains supplemental material online at http://jasn.asnjournals.org/ lookup/suppl/doi:10.1681/ASN.2015030243/-/DCSupplemental.

MUCH-CKD

9