Kidney Function Decline in Young Adulthood and Subsequent 24-Hour Ambulatory Blood Pressure in Midlife: The CARDIA Study.

To the Editor:

Hypertension (HTN) and chronic kidney disease (CKD) are comorbid conditions. A decline in kidney function is associated with an increase in blood pressure (BP), and a sustained elevation in BP leads to the progression of kidney function decline. As CKD stage advances, the prevalence of HTN increases, and BP becomes more difficult to control.

Previous studies have focused on the associations between HTN and longitudinal changes in the estimated glomerular filtration rate (eGFR). Patients with HTN or with higher baseline office BP were found to have faster rates of eGFR decline on follow-up in population studies., Outside the office setting, ambulatory BP monitoring (ABPM) is the reference standard to measure the BP. ABPM provides data on the average BP over 24 hours and can identify several BP profiles by comparing the office and out-of-office BP readings. Characterizing BP profiles such as masked HTN (nonhypertensive inside of the office but hypertensive outside of the office) or sustained HTN (hypertensive inside and outside of the office) among patients with and without CKD may help identify the risk of target organ damage and adverse clinical outcomes. Conversely, incident CKD and more rapid declines in eGFR may be a risk factor for masked or sustained HTN. Although cross-sectional studies have shown an association between masked HTN and reduced eGFR, no study has evaluated the association between eGFR decline and incident CKD as a predictor of BP profiles assessed using ABPM., Therefore, the purpose of this cross-sectional study was to examine the association of the development of CKD with subsequent BP on ABPM.

We conducted post hoc analyses of data from CARDIA (Coronary Artery Risk Development in Young Adults), a prospective cohort study conducted in the United States to understand risk factors for the development of cardiovascular disease. Age, race, and sex were self-reported at baseline and were confirmed at the year 2 examination. The remaining clinical data were obtained at the year 30 examination. Serum creatinine, urine albumin, and creatinine samples from year 10, 15, 20, 25, and 30 examinations were collected. The CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation was used to calculate the eGFR (mL/min/1.73m2), and albuminuria was defined as having a urinary albumin-to-creatinine ratio of >30 mg/g. We defined the development of CKD as having an eGFR of <60 mL/min/1.73m2 or albuminuria at any time point between year 10 and 30. The office BP was measured by a certified staff following a standardized protocol. A validated monitor and appropriate cuff size (Item S1) was used for the 24-hour ABPM at the year 30 visit. We calculated the mean 24-hour, mean awake, and mean nocturnal BP, ABPM phenotypes (masked HTN [BP: office, <140/90 mm Hg; 24 h, ≥130/80 mm Hg], sustained normotension [BP: office, <140/90 mm Hg; 24 h, <130/80 mm Hg), and dipping patterns (dippers vs nondippers: reduction in the mean BP from day to night of 10%–20% vs <10%). For the present analysis, we included participants who had a valid ABPM.,

We used linear and logistic regression to evaluate the association between the development of CKD and systolic BP on ABPM, masked HTN (vs sustained normotension), and nondipping systolic BP. We adjusted for demographic and clinical characteristics. All analyses were stratified by whether participants were on antihypertensive medications at year 30, as we hypothesized that the prevalence of ABPM phenotypes would be higher for participants taking antihypertensives. The sensitivity analysis included using the CKD-EPI 2021 equation (without race) and using diastolic BP to define ABPM phenotypes.,

We included 763 participants with complete ABPM (Item S1). The participants who were not on antihypertensive medications (450 [59%]) had a mean age of 54.6 ± 3.7 years; 256 (57%) were women; 226 (50%) were African American; eGFR (year 30) was 90.8 ± 15.4 mL/min/1.73m2; and 62 (14%) had CKD. The participants who were on antihypertensive medication (313 [41%]) had a mean age of 55.0 ± 3.7 years; 201 (64%) were women; 68 (22%) were African American; eGFR (year 30) was 90.4 ± 20.1 mL/min/1.73m2; and 103 (33%) had CKD. Among all adults (receiving and not receiving antihypertensive medications), the development of CKD was not associated with the mean 24-hour systolic and diastolic BP or the mean awake systolic and diastolic BP after adjusting for demographic and clinical characteristics (Table 1; Item S1). However, among those not receiving antihypertensives, CKD was associated with higher nocturnal systolic BP (β [95% CI]: 3.6 [0.40-6.79]) and nocturnal diastolic BP (β [95% CI]: 2.8 [0.84-4.77]) (Table 1; Item S1). We found no association between masked HTN or nondipping BP and the development of CKD in the fully adjusted models (Table 2; Item S1) These findings were consistent even when defining CKD using the CKD-EPI 2021 equation.

Table 1

Association of CKD Development (Compared With Normal Kidney Function Between Years 10 and 30) With Systolic Ambulatory BP of Participants (at year 30) On and Not On Antihypertensive Medications

β (95% CI)	Model 1	Model 2	Model 3
Participants Not On Antihypertensive Medications (n=450)
24-Hour systolic BP	4.43 (0.72 to 8.14)a	3.26 (−0.50 to 7.01)	1.65 (−1.05 to 4.36)
Awake systolic BP	3.39 (−0.52 to 7.29)	2.41 (−1.60 to 6.41)	0.71 (−2.17 to 3.59)
Nocturnal systolic BP	6.47 (2.67 to 10.26)a	5.05 (1.23 to 8.87)a	3.59 (0.40 to 6.79)a
Participants On Antihypertensive Medications (n=313)
24-Hour systolic BP	2.08 (−1.34 to 5.49)	0.80 (−2.69 to 4.29)	0.27 (−2.60 to 3.13)
Awake systolic BP	1.41 (−2.17 to 4.98)	0.24 (−3.39 to 3.87)	−0.35 (−3.31 to 2.61)
Nocturnal systolic BP	3.50 (−0.17 to 7.17)	1.81 (−1.97 to 5.59)	1.18 (−2.22 to 4.59)

Note: Model 1: unadjusted; model 2: adjusted for study center, age, sex, race, and educational attainment; model 3: model 2 + smoking status, history of diabetes, history of hypertension, family history of hypertension, statin use at year 30, body mass index at year 30, total cholesterol at year 30, office systolic at year 30, office diastolic at year 30, and estimated glomerular filtration rate at year 30.

Abbreviations: BP, blood pressure; CKD, chronic kidney disease.

Statistically significant.

Table 2

Association of CKD Development (Compared With Normal Kidney Function Between Years 10 and 30) With Ambulatory BP Phenotypes (at Year 30) Based on The 2005 ACC/AHA Guidelines of Participants On and Not On Antihypertensive Medications

OR (95% CI)	Model 1	Model 2	Model 3
Masked Hypertension vs Sustained Normotension
Participants Not On Antihypertensive Medications (n=450)
Using office BP and 24 hour ambulatory BP	1.70 (0.94-3.04)	1.62 (0.84-3.08)	1.72 (0.60-4.84)
Using office BP and awake ambulatory BP	1.48 (0.82-2.63)	1.34 (0.70-2.53)	1.26 (0.46-3.37)
Using office BP and nocturnal ambulatory BP	1.89 (1.04-3.36)a	1.61 (0.84-3.08)	2.03 (0.78-5.21)
Participants On Antihypertensive Medications (n=313)
Using office BP and 24 hour ambulatory BP	1.29 (0.76-2.21)	0.92 (0.49-1.66)	0.57 (0.24-1.33)
Using office BP and awake ambulatory BP	1.09 (0.63-1.86)	0.86 (0.47-1.55)	0.65 (0.28-1.49)
Using office BP and nocturnal ambulatory BP	1.54 (0.91-2.62)	1.16 (0.65-2.06)	0.94 (0.43-2.02)

Note: Model 1: unadjusted; model 2: adjusted for study center, age, sex, race, and educational attainment; model 3: model 2 + smoking status, history of diabetes, history of hypertension, family history of hypertension, statin use at year 30, body mass index at year 30, total cholesterol at year 30, office systolic at year 30, office diastolic at year 30, and estimated glomerular filtration rate at year 30. The white coat effect and sustained hypertension (Item S1) were prevalent among 8 (2%) and 33 (7%) participants not on antihypertensive medications and among 7 (2%) and 46 (15%) participants on antihypertensive medications, respectively. Because of the limited sample size, we were unable to study the association between these BP phenotypes and incident CKD.

Abbreviations: BP, blood pressure; CKD, chronic kidney disease; OR, odds ratio.

Statistically significant.

We found that overall worsening kidney function was associated with higher nocturnal systolic and diastolic BP among adults not receiving antihypertensive medications, but this association was attenuated by demographic factors and comorbid conditions. These results are on par with the findings of the study by Mojón et al, who showed that the development of CKD was associated with higher nocturnal systolic BP. Our data suggest that the incidence of CKD during young adulthood was not associated with masked HTN and nondipping or higher awake or 24-hour BP in midlife among individuals with mostly normal eGFR. The main limitation of this study include the modest sample size of patients enrolled in the CARDIA study with complete ABPM at year 30 and the cross-sectional nature of this analysis. These findings show that compared with patients without CKD, those with CKD have higher nocturnal BP, particularly if they are not using any antihypertensive medication. Whether noctural BP might be a precursor for CKD should be explored further.