Visual Abstract
Keywords: CKD
Abstract
Key Points
Chlorthalidone reduces the amount of fluid and the BP, but fluid volume reduction is not the cause of lowering of BP.
It is not volume loss but the response to volume loss such as the synthesis of substances that lower BP is important.
Background
Chlorthalidone (CTD) in a chronic kidney disease randomized trial demonstrated a robust reduction in systolic BP in stage 4 CKD. In this study, we explore the mechanisms underlying the antihypertensive effect of CTD.
Methods
In this prespecified analysis, we analyzed the contributions of baseline levels of 24-hour urinary sodium and aldosterone and the changes from baseline to 4 weeks in the multiple mediators reflecting volume status in a causal mediation analysis framework. Baseline levels of these mediators served as covariates. No power calculation for this analysis was performed.
Results
Of the 160 patients randomized, 140 (87.5%) were included in this analysis. Compared with placebo, CTD within 4 weeks reduced weight −1.5% (95% confidence interval [CI], −2.2 to −0.7) and volume −1.4% (95% CI, −2.2 to −0.6), stimulated plasma renin 40.5% (95% CI, 25.4% to 57.4%) and serum aldosterone 40.2% (95% CI, 11.7% to 76%), and reduced plasma N-terminal pro-B-type natriuretic peptide levels −19.4% (95% CI, −33.8% to −1.9%). Mediation analysis revealed the following results: for weight change, the total effect on systolic BP was −10.8 mm Hg (95% CI, −16 to −5.7), of which weight change (indirect effect) accounted for −0.9 mm Hg (95% CI, −4.2 to 2.5) and BP change independent of weight (direct effect) accounted for −10 mm Hg (−15.7 to −4.2). Thus, the percent mediation was 8.1% (95% CI, −22.4 to 38.5). Baseline excretion of 24-hour sodium or aldosterone or any of the changes in the above mediators examined accounted for <2 mm Hg BP drop and were not significant for any of the mediators.
Conclusions
CTD improved BP control among patients with advanced CKD independent of baseline urinary sodium, aldosterone, weight loss, or changes in the renin-angiotensin system or N-terminal pro-B-type natriuretic peptide.
Clinical Trial registry name and registration number:
CTD in chronic kidney disease ClinicalTrials.gov number: NCT02841280.
Introduction
Among patients with advanced CKD, hypertension has multiple etiologies.1 These include sympathetic activation, increased arterial stiffness, and endothelial dysfunction, possibly due to endothelin excess or inactivation of nitric oxide synthase, e.g., via increased levels of asymmetric dimethyl arginine.1 However, the foremost factors thought to be causally related to the pathogenesis of hypertension are believed to be the activation of the renin-angiotensin-aldosterone system (RAAS) and primary volume overload.
In patients with CKD, whether it is primary volume overload or excess RAAS activation that mediates hypertension remains unclear. If volume overload was the primary factor in causing hypertension, then renin and aldosterone levels are expected to be suppressed. If so, diuretic treatment would produce the greatest drop in BP. On the other hand, if volume overload were secondary to RAAS activation, high levels of aldosterone would be expected to associate with the greatest BP lowering. Furthermore, in patients with essential hypertension, it has been suggested that after the initiation of diuretic therapy, the RAAS response blunts the BP-lowering response: the more robust the increase in renin, the lower the decline in BP.2,3 In such patients, the use of blockers of the renin-angiotensin system, such as β-blockers, angiotensin-converting enzyme inhibitors, or angiotensin receptor blockers, might produce the greatest BP-lowering effect.
For CKD, no data exist for the relative contributions of primary volume excess or RAAS activation in producing BP lowering. In this paper, we report a prespecified analysis of the data from a randomized Chlorthalidone in Chronic Kidney Disease (CLICK) trial to explore the pathophysiology of hypertension in CKD. Specifically, we focus on the roles that extracellular volume, RAAS action, and ventricular stretching, as measured by using N-terminal pro-B-type natriuretic peptide (NT-proBNP), play in the mechanisms leading to BP reduction.
Methods
We performed a causal mediation analysis of the data from the CLICK trial. CLICK was a placebo-controlled, double-blind, randomized control trial of chlorthalidone (CTD) versus placebo in patients with advanced CKD and treated but poorly controlled hypertension. This study was approved by the Institutional Review Board of Indiana University and the Research and Development Committee of the Richard L. Roudebush Veterans Administration Medical Center, Indianapolis, IN. All patients provided written informed consent before participation. The trial design4 and the main results have been published.5 In this study, we highlight the key features of the trial that are relevant to this analysis.
Patients
CLICK recruited patients who had stage 4 CKD (eGFR <30 ml/min per 1.73 m2 but ≥15 ml/min per 1.73 m2) and uncontrolled hypertension, confirmed by 24-hour ambulatory BP monitoring (ABPM) after a 2-week period during which antihypertensive medications were standardized. Uncontrolled hypertension was defined as having an average 24-hour ABPM ≥130 mm Hg systolic or ≥80 mm Hg diastolic while on antihypertensive treatment. The trial excluded patients whose 24-hour ABPM ≥160 mm Hg systolic or ≥100 mm Hg diastolic, as well as those who had a stroke, myocardial infarction, or hospitalization due to heart failure within 3 months, those on high-dose loop diuretics (furosemide >200 mg/d or torsemide >100 mg/d), or those who were using a thiazide or a thiazide-like diuretic in the previous 12 weeks. At the time of randomization, all patients were put on an angiotensin-converting enzyme inhibitor or an angiotensin receptor blocker, or a β-blocker. Patients were recruited from the Indiana University Hospitals, Eskenazi Hospital, and the Richard L. Roudebush Veterans Administration Medical Center, all in Indianapolis, IN.
This analysis was limited to 140 (87.5%) of 160 randomized patients who had paired measurements of 24-hour ambulatory BP.
Study Visits and BP Measurements
Detailed trial protocol is available at www.nejm.org/doi/suppl/10.1056/NEJMoa2110730/suppl_file/nejmoa2110730_protocol.pdf. In brief, the trial had nine study visits, including four before randomization over 3 weeks, 4 after randomization over 12 weeks, and a final visit 2 weeks after the study drug discontinuation. Methods for clinic BP measurement, ABPM, and medication standardization were detailed in the trial protocol.
Randomization and Blinding
Enrolled patients were randomized in equal numbers to either CTD or placebo, stratified by loop diuretic use. Treatment assignments were masked for the study investigators, treating physicians, patients, and outcome assessors. The study pharmacist, who did not interact with patients, maintained the blinding and dispensed the medication according to the randomization sequence.
Mediator and Outcome Measures
Total body volume was measured using air displacement plethysmography (Life Measurement, Inc., Concord, CA). Serum NT-proBNP was measured using an immunoassay (Elecsys proBNP II STAT Immunoassay run on the Cobas 8100 analyzer, Roche Diagnostics, Indianapolis, IN). Plasma renin was measured using an immunoassay (Human Renin Quantikine ELISA kit, R&D Systems, Inc., Minneapolis, MN). Plasma and urine aldosterone were also measured using an immunoassay (Aldosterone Parameter Assay Kit, R&D Systems, Inc.). The trial's primary outcome was the change in 24-hour systolic ambulatory BP from baseline to 12 weeks and was measured using the Spacelabs 90207 ambulatory monitor (Spacelabs Healthcare, Snoqualmie, WA).
Statistical Methods
Figure 1 shows the directed acyclic graph of the causal mediation model. Causal mediation analyses were performed using standard methods.6,7 The methodological assumptions underlying causal mediation analysis proposed by Lee et al.8 have been summarized in a tabular form in our recent publication.9 First, linear regression models were used to quantify the effects of the treatment on the mediators. Then, linear regression models were fitted to depict the effect of each mediator on the outcome (change in systolic BP between baseline and end of the study). There were eight mediators considered in this analysis. Of these, six mediators were changes from baseline to 4 weeks in total body weight, total body volume, plasma renin, plasma aldosterone, aldosterone-to-renin ratio, and NT-proBNP. Two other mediators only had baseline measurements and were 24-hour urinary sodium and aldosterone excretion rates. All mediators were log-2 transformed to correct the skewness of their distributions before analysis. The same set of covariates were included in the mediator and outcome models, including age (dichotomized at the sample median), sex, race (Black or non-Black), angiotensin-converting enzyme inhibitor or angiotensin receptor blocker use (yes or no), loop diuretic use (yes or no), or K-sparing diuretic use (yes or no). Furthermore, all models included the baseline level of the mediator variable and treatment×mediator interaction. The natural direct effects (independent of the mediator), natural indirect effects (dependent on the mediator), total effects, and percent mediation were calculated using a bootstrap method with 1000 replicates. In a sensitivity analysis, we reduced dimensionality from key change mediators measured from baseline to 4 weeks using principal component analysis and fitted a mediation model to this principal component factor while using the baseline volume, renin, aldosterone, and NT-proBNP as covariates. The model included the treatment×mediator interaction. All statistical analyses were performed using Stata 18.0 (StataCorp, College Station, TX).
Figure 1.
Directed acyclic graph of the causal mediation model. Direct effect is induced by treatment on systolic BP independent of the mediator and the indirect effects through mediation by a volume marker. The volume markers included change in total body weight, total body volume, renin, aldosterone, aldosterone-to-renin ratio, and NT-proBNP. NT-proBNP, N-terminal pro-B-type natriuretic peptide.
Results
Table 1 shows the baseline characteristics of the 140 of the 160 participants who had paired measurements of baseline and end-of-study 24-hour ambulatory systolic BP. The mean age was 67 years, 78% were men, and 39% were Black. The mean number of antihypertensive medications was 3.4; 63% were on angiotensin converting enzyme inhibitors or angiotensin receptor blockers at baseline, 9% were on either spironolactone or amiloride, and 60% were on loop diuretics. Seated clinic BP was 139.3/67.8 mm Hg, mean eGFR was 23.2 ml/min per 1.73 m2, and median urine albumin to urine creatinine ratio was 597 mg/g creatinine at baseline.
Table 1.
Baseline characteristics of the study sample
| Variable | Placebo | CTD | Total |
|---|---|---|---|
| Participants, n (%) | 73 (52) | 67 (48) | 140 (100) |
| Age (yr), mean (SD) | 67.3 (10.8) | 66.3 (11.3) | 66.9 (11) |
| Men, n (%) | 58 (79.5) | 51 (76.1) | 109 (77.9) |
| Black race, n (%) | 28 (38.4) | 27 (40.3) | 55 (39.3) |
| Antihypertensive medications, n (SD) | 3.4 (1.5) | 3.4 (1.3) | 3.4 (1.4) |
| Using ACEi or ARB, n (%) | 40 (54.8) | 48 (71.6) | 88 (62.9) |
| Using loop diuretics, n (%) | 42 (57.5) | 42 (62.7) | 84 (60) |
| Using K-sparing diuretics, n (%) | 10 (13.7) | 3 (4.5) | 13 (9.3) |
| Systolic BP (mm Hg), mean (SD) | 138.1 (15.4) | 140.7 (15.1) | 139.3 (15.3) |
| Diastolic BP (mm Hg), mean (SD) | 67.2 (13.8) | 68.4 (12.3) | 67.8 (13.1) |
| eGFR, ml/min per 1.73 m2, mean (SD) | 22.9 (4.3) | 23.5 (4.1) | 23.2 (4.2) |
| Spot urine albumin-to-urine creatinine ratio (mg/g), median (IQR) | 889 (96–1736) | 449 (175–1972) | 597 (140–1850) |
| Urinary Na excretion (mEq/24 h), median (IQR) | 109 (75–148) | 135 (94–173) | 122 (86–163) |
| Urinary aldosterone excretion (mcg/24 h), median (IQR) | 4.44 (2.47–6.41) | 4.71 (2.91–6.47) | 4.71 (2.73–6.46) |
ACEi, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; CTD, chlorthalidone; IQR, interquartile range.
Table 2 shows the mean levels of the volume markers at baseline and their changes from baseline to 4 weeks. At 4 weeks, CTD significantly changed several putative volume markers in the expected direction. Within 4 weeks, weight and volume decreased, plasma renin and serum aldosterone increased, and NT-proBNP decreased, but the aldosterone-to-renin ratio did not change in response to CTD treatment. The 24-hour urinary excretion rates of sodium and aldosterone were measured only at baseline but not at 4 weeks. Baseline levels of these two mediators are provided in Table 1.
Table 2.
Description and changes from baseline to 4 weeks in mediator variables
| Variable | Placebo | CTD | Percent Change from Baseline to 4 wk in Placebo | Percent Change from Baseline to 4 wk in CTD | Percent Difference in CTD Change Minus Placebo Change |
|---|---|---|---|---|---|
| Weight (kg) | 93.3 (88 to 98.6) | 95.6 (90.4 to 100.9) | 0.3 (−0.2 to 0.8) | −1.2 (−1.8 to −0.7) | −1.5 (−2.2 to −0.7) |
| Total body volume (L) | 92.2 (86.8 to 97.7) | 94.7 (89.1 to 100.2) | 0.2 (−0.4 to 0.7) | −1.2 (−1.8 to −0.7) | −1.4 (−2.2 to −0.6) |
| Plasma renin concentration (pg/ml) | 2337 (1846 to 2959) | 2461 (1932 to 3134) | 12.5 (4.6 to 20.9) | 59.4 (46.8 to 73.1) | 40.5 (25.4 to 57.4) |
| Plasma aldosterone concentration (pg/ml) | 309 (251 to 380) | 271 (224 to 329) | 15.1 (−3.5 to 37.4) | 63.7 (37 to 95.5) | 40.2 (11.7 to 76) |
| Ratio of aldosterone to renin | 0.13 (0.09 to 0.19) | 0.11 (0.08 to 0.16) | 2.4 (−13.6 to 21.3) | 2.7 (−13.7 to 22.2) | −1.1 (−21.8 to 25.3) |
| NT-proBNP (pg/ml) | 687 (503 to 939) | 615 (427 to 885) | −11.5 (−22 to 0.3) | −28.4 (−38.4 to −16.8) | −19.4 (−33.8 to −1.9) |
Numbers are represented as means, and 95% confidence interval bounds of the estimates are within parenthesis. All analyses are adjusted for the following variables: age, sex, race, use of angiotensin converting enzyme inhibitor or angiotensin receptor blocker, loop diuretic use, and K-sparing diuretic use at baseline. At the 4-week visit, three patients had missing weights, seven had missing volume, and 20 had missing N-terminal pro-B-type natriuretic peptide. CTD, chlorthalidone; NT-proBNP, N-terminal pro-B-type natriuretic peptide.
Table 3 shows the results of causal mediation analysis for each of the individual biomarkers with a change in 24-hour systolic BP. Despite rapid reduction in weight, total body volume, and NT-proBNP and increases in the plasma levels of renin and aldosterone within 4 weeks, the contributions of these markers on overall reduction in 24-hour systolic BP at 12 weeks were small. Treatment effects mediated through the volume markers accounted for <2 mm Hg drops of the nearly 11 mm Hg total reduction in systolic BP. The 95% confidence intervals of each of these small changes crossed zero and, therefore, were not statistically significant. Plasma renin change accounted for 7%, and NT-proBNP for 8% of the BP-lowering effect; neither was statistically significant. Baseline 24-hour urine Na and urine aldosterone were not significantly associated with systolic BP change at 12 weeks either.
Table 3.
Decomposition of total 24-hour systolic BP change at 12 weeks into direct, indirect, and total effects and percent mediated
| Variable | Indirect Effect | Direct Effect | Total Effect | Percent Mediation |
|---|---|---|---|---|
| Weight (kg) | −0.9 (−4.2 to 2.5) | −10 (−15.7 to −4.2) | −10.8 (−16 to −5.7) | 8.1 (−22.4 to 38.5) |
| Total body volume (L) | −0.2 (−3.1 to 2.7) | −11 (−16.3 to −5.7) | −11.2 (−15.8 to −6.6) | 1.7 (−24 to 27.3) |
| Plasma renin concentration | −0.8 (−5 to 3.5) | −10 (−15.9 to −4) | −10.7 (−15 to −6.5) | 7.1 (−32.3 to 46.5) |
| Serum aldosterone concentration | 0.5 (−1 to 1.9) | −11.2 (−16.1 to −6.4) | −10.8 (−15.3 to −6.2) | −4.5 (−17.9 to 8.9) |
| Ratio of aldosterone to renin | 0 (−0.7 to 0.6) | −10.8 (−15.4 to −6.2) | −10.8 (−15.4 to −6.2) | 0.2 (−6 to 6.4) |
| 24-h urine aldosterone | −1.7 (−4.3 to 0.9) | −9.2 (−13.8 to −4.5) | −10.9 (−15.8 to −6) | 15.8 (−6.7 to 38.3) |
| 24-h urine sodium | −0.3 (−1.3 to 0.6) | −10.3 (−15.1 to −5.6) | −10.6 (−15.2 to −6.1) | 2.9 (−6.1 to 12) |
| NT-proBNP | −0.9 (−2.2 to 0.3) | −10.4 (−15.4 to −5.4) | −11.3 (−16.3 to −6.4) | 8.3 (−3.2 to 19.8) |
Numbers within parenthesis are 95% confidence interval bounds of the estimates. The change in each variable on the log scale adjusted for the baseline level also on the log scale is the mediation variable. The two exceptions are the 24-hour urine collections where only the baseline is collected. Therefore, what is reported is the effect of the baseline levels of sodium and aldosterone excretion rate on the BP response to chlorthalidone. Both the mediator and the outcome variable are adjusted for age, sex, race, use of angiotensin converting enzyme inhibitor or angiotensin receptor blocker, loop diuretic use, and K-sparing diuretic use at baseline. At the 4-week visit, three patients had missing weights, seven had missing volume, and 20 had missing N-terminal pro-B-type natriuretic peptide. At baseline, 24-hour urine sodium was missing in 3 patients and 24-hour urine aldosterone in five patients. NT-proBNP, N-terminal pro-B-type natriuretic peptide.
Figure 2 illustrates graphically that changes in markers of extracellular fluid volume at 4 weeks contributed little to the BP change at 12 weeks. For example, weight loss or weight gain had little association with subsequent drops in BP whether participants were on placebo or CTD. However, CTD treatment per se lowered systolic BP independent of weight. Similar relations were seen for changes in total body volume, plasma renin, plasma aldosterone, plasma aldosterone-to-plasma renin ratio, plasma NT-proBNP, and subsequent changes in 24-hour ambulatory systolic BP.
Figure 2.
Changes in markers of extracellular fluid volume at 4 weeks and their association with the change in 24-hour systolic BP at 12 weeks. Volume parameters are depicted as a change from baseline to 4 weeks and for ease of interpretation are either shown as a fold change or percent change from baseline. ARR, aldosterone-to-renin ratio.
Sensitivity Analysis
We reduced dimensionality of the volume parameters using principal component analysis, as noted above. Because weight and volume were highly correlated, we dropped weight from the model. Only one factor emerged with an eigenvalue >1 that explained 48% of the total variance. We fitted a mediation model to this principal component factor while using the baseline volume, renin, aldosterone, and NT-proBNP as covariates. The indirect effect of the principal component was −0.3 mm Hg (95% CI, −3.4 to 2.8 mm Hg) and accounted for 2.5% (95% CI, −24.2% to 29.1%) of the total effect.
Discussion
We explored the mechanisms that underlie BP reduction with CTD in patients with advanced CKD. Our data showed that within 4 weeks, CTD reduced weight and volume, stimulated plasma renin and plasma aldosterone production, and lowered plasma NT-proBNP levels. We have earlier reported a −10.5 mm Hg (95% CI, −14.6 to−6.4) (P < 0.001) change in 24-hour ambulatory systolic BP at 12 weeks. However, none of the changes in the volume markers from baseline to 4 weeks have led to appreciable changes in systolic BP between baseline and 12 weeks. Furthermore, baseline 24-hour excretion of sodium or aldosterone was not associated with a drop in systolic BP at 12 weeks.
The mechanisms of thiazide or thiazide-like diuretics-induced BP reduction have been studied since the 1960s when these drugs were first discovered and marketed. It has been demonstrated that although acute effects of these drugs provoke a reduction in extracellular fluid volume10,11 and cardiac output,12 in the long term, the fall in systemic vascular resistance is responsible for the BP reduction.13–15 The relation of weight loss to BP change induced by these drugs is less clear. For example, Winer, in a carefully done 1961 study, showed that when plasma volume was acutely expanded to pretreatment level, systolic BP rose but did not reach baseline level.16 Furthermore, moderate sodium chloride supplementation of 8 g/d did not abolish the BP-lowering effect of chlorothiazide; however, 20 g/d sodium chloride supplementation for 2 weeks did.16 Winer suggested that “salt loss must involve another mechanism by which peripheral vascular resistance is reduced.”16 By contrast, Freis et al. in 1988 reported that among patients with essential hypertension treated with hydrochlorothiazide, there was a positive correlation between loss in body weight and the decline in diastolic BP.17 Although they found that nonresponders had less weight loss than responders, they acknowledged that BP reductions may vary with the same volume loss among those with hypertension.17 In a 1983 study of 23 patients with essential hypertension, CTD 100 mg given for 6 weeks caused a marked increase in renin, but blunted responsiveness to infused NE, potentially uncovering mechanisms of reduced systemic vascular resistance.18 Although hyper-responsiveness to renin release has been reported to abrogate an antihypertensive response to diuretics,2,3 in our study, all patients were treated with drugs, such as β-blockers, angiotensin converting enzyme inhibitors, or angiotensin receptor blockers, to block hyperactivation of RAAS. Accordingly, it is not surprising that we did not observe a relation between renin response and a blunted antihypertensive effect.
Although the exact mechanism of the antihypertensive effect of CTD remains elusive, we can synthesize the available evidence to reach some tentative conclusions. Conditions that are conducive to volume overload, such as the presence of advanced CKD, high sodium intake, and a high aldosterone secretion rate when treated with an effective long-acting thiazide-like diuretic, CTD, results in a reduction of extracellular fluid volume. Although stimulation of the renin-angiotensin system or decline in NT-proBNP are homeostatic responses, overcorrection could cause an increase in BP. However, a reduction in BP was observed in this study. Therefore, we conjecture that either a reduced response to vasoconstrictors such as NE18 or an increase in vasodilators such as nitric oxide,19 kinins,20 or eicosanoids21 ultimately results in a reduction in BP. In a crossover trial of 26 patients with moderate to advanced CKD, Bovee et al. demonstrated that prostaglandin E2 (PGE2) was stimulated with distal diuretics but not sodium restriction.22 PGE2 has numerous effects on the kidney including increase in renal blood flow,23 natriuresis,24 and activation of renal sympathetic afferent nerves.25 These effects of PGE2 can collectively lower BP. Among patients with Bartter and Gitelman syndrome, an altered angiotensin II short- and long-term cell signaling is observed, which suggests altered vascular tone regulation.26 The vascular response to volume depletion appears to be therefore important.27
Our study has limitations. These data were generated in patients with advanced CKD; there were only a few Asian and Hispanic patients and only 22% were women. Whether similar results would be obtained in earlier stages of CKD or more diverse populations is unclear. All biomarker measurements were made at baseline and at 4 weeks except for two. The exceptions were those measurements that required 24-hour urine collection—sodium and aldosterone—and were measured at baseline but not at 4 weeks. A small number of patients did not complete testing, which might have biased the results. A larger study might have yielded statistical significance. Even then, <20% of the effect would be attributable to volume markers, which would still question the clinical significance of the observations. All models were accounted for the baseline level of the biomarkers. For example, prior studies in patients without CKD have suggested a greater BP reduction in those with suppressed baseline renin.28–30 However, our study is consistent with one in CKD, which also could not demonstrate the association of baseline renin and aldosterone and BP response.22 Finally, as with all causal mediation analyses, there may exist unobserved confounders that causally affect both the mediator and the outcome even after conditioning on the observed treatment and pretreatment covariates.
In conclusion, factors other than volume per se or stimulation of the renin-angiotensin system are therefore important in the BP-lowering effect seen among patients with advanced CKD. It is the response to volume depletion that is important. Therefore, we postulate that diuretic treatment in patients with advanced CKD either results in an increase in the production of vasodilators or mitigates the vasoconstrictor effects of circulating mediators, culminating in BP lowering.
Footnotes
See related editorial, “Understanding the BP-Lowering Mechanism of Chlorthalidone in Advanced Kidney Disease,” on pages 949–951.
Disclosures
Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/CJN/B937.
Funding
Supported by National Heart Lung and Blood Institute: R01 HL126903.
Author Contributions
Conceptualization: Rajiv Agarwal.
Data curation: Rajiv Agarwal.
Formal analysis: Rajiv Agarwal.
Funding acquisition: Rajiv Agarwal.
Investigation: Rajiv Agarwal, Arjun D. Sinha.
Methodology: Rajiv Agarwal, Wanzhu Tu.
Project administration: Rajiv Agarwal.
Resources: Rajiv Agarwal.
Software: Rajiv Agarwal.
Supervision: Rajiv Agarwal.
Validation: Rajiv Agarwal.
Visualization: Rajiv Agarwal.
Writing – original draft: Rajiv Agarwal.
Writing – review & editing: Rajiv Agarwal, Arjun D. Sinha, Wanzhu Tu.
Data Sharing Statement
Due to ongoing study and analyses, data are not available to share at present.
References
- 1.Campese VM, Mitra N, Sandee D. Hypertension in renal parenchymal disease: why is it so resistant to treatment? Kidney Int. 2006;69(6):967–973. doi: 10.1038/sj.ki.5000177 [DOI] [PubMed] [Google Scholar]
- 2.Vaughan ED, Jr., Carey RM, Peach MJ, Ackerly JA, Ayers CR. The renin response to diuretic therapy: a limitation of antihypertensive potential. Circ Res. 1978;42(3):376–381. doi: 10.1161/01.res.42.3.376 [DOI] [PubMed] [Google Scholar]
- 3.Gavras H Waeber B Kershaw GR, et al. Role of reactive hyperreninemia in blood pressure changes induced by sodium depletion in patients with refractory hypertension. Hypertension. 1981;3(4):441–447. doi: 10.1161/01.hyp.3.4.441 [DOI] [PubMed] [Google Scholar]
- 4.Agarwal R, Cramer AE, Balmes-Fenwick M, Sinha AD, Ouyang F, Tu W. Design and baseline characteristics of the chlorthalidone in chronic kidney disease (CLICK) trial. Am J Nephrol. 2020;51(7):542–552. doi: 10.1159/000508700 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Agarwal R Sinha AD Cramer AE, et al. Chlorthalidone for hypertension in advanced chronic kidney disease. N Engl J Med. 2021;385(27):2507–2519. doi: 10.1056/NEJMoa2110730 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Imai K, Keele L, Tingley D. A general approach to causal mediation analysis. Psychol Methods. 2010;15(4):309–334. doi: 10.1037/a0020761 [DOI] [PubMed] [Google Scholar]
- 7.VanderWeele T. ProQuest. Explanation in Causal Inference: Methods for Mediation and Interaction. Oxford University Press; 2015. [Google Scholar]
- 8.Lee H Cashin AG Lamb SE, et al.; AGReMA group. A guideline for reporting mediation analyses of randomized trials and observational studies: the AGReMA statement. JAMA. 2021;326(11):1045–1056. doi: 10.1001/jama.2021.14075 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Agarwal R Tu W Farjat AE, et al.; FIDELIO-DKD and FIGARO-DKD Investigators. Impact of finerenone-induced albuminuria reduction on chronic kidney disease outcomes in type 2 diabetes: a mediation analysis. Ann Intern Med. 2023;176(12):1606–1616. doi: 10.7326/M23-1023 [DOI] [PubMed] [Google Scholar]
- 10.Leth A. Changes in plasma and extracellular fluid volumes in patients with essential hypertension during long-term treatment with hydrochlorothiazide. Circulation. 1970;42(3):479–485. doi: 10.1161/01.cir.42.3.479 [DOI] [PubMed] [Google Scholar]
- 11.van Brummelen P, Schalekamp MA. Body fluid volumes and the response of renin and aldosterone to short- and long-term thiazide therapy of essential hypertension. Acta Med Scand. 1980;207(4):259–264. doi: 10.1111/j.0954-6820.1980.tb09718.x [DOI] [PubMed] [Google Scholar]
- 12.Frohlich ED, Schnaper HW, Wilson IM, Freis ED. Hemodynamic alterations in hypertensive patients due to chlorothiazide. N Engl J Med. 1960;262:1261–1263. doi: 10.1056/NEJM196006232622502 [DOI] [PubMed] [Google Scholar]
- 13.Conway J, Lauwers P. Hemodynamic and hypotensive effects of long-term therapy with chlorothiazide. Circulation. 1960;21(1):21–27. doi: 10.1161/01.cir.21.1.21 [DOI] [PubMed] [Google Scholar]
- 14.Lauwers P, Conway J. Effect of long-term treatment with chlorothiazide on body fluids, serum eletrolytes, and exchangeable sodium in hypertesive patients. J Lab Clin Med. 1960;56:403–408. PMID: 14414615 [PubMed] [Google Scholar]
- 15.van Brummelen P, Man in 't Veld AJ, Schalekamp MA. Hemodynamic changes during long-term thiazide treatment of essential hypertension in responders and nonresponders. Clin Pharmacol Ther. 1980;27(3):328–336. doi: 10.1038/clpt.1980.44 [DOI] [PubMed] [Google Scholar]
- 16.Winer BM. The antihypertensive mechanisms of salt depletion induced by hydrochlorothiazide. Circulation. 1961;24(4):788–796. doi: 10.1161/01.cir.24.4.788 [DOI] [PubMed] [Google Scholar]
- 17.Freis ED, Reda DJ, Materson BJ. Volume (weight) loss and blood pressure response following thiazide diuretics. Hypertension. 1988;12(3):244–250. doi: 10.1161/01.hyp.12.3.244 [DOI] [PubMed] [Google Scholar]
- 18.Weidmann P, Beretta-Piccoli C, Meier A, Keusch G, Gluck Z, Ziegler WH. Antihypertensive mechanism of diuretic treatment with chlorthalidone. Complementary roles of sympathetic axis and sodium. Kidney Int. 1983;23(2):320–326. doi: 10.1038/ki.1983.22 [DOI] [PubMed] [Google Scholar]
- 19.Shultz PJ, Tolins JP. Adaptation to increased dietary salt intake in the rat. Role of endogenous nitric oxide. J Clin Invest. 1993;91(2):642–650. doi: 10.1172/JCI116244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ferri C, Bellini C, Carlomagno A, Perrone A, Santucci A. Urinary kallikrein and salt sensitivity in essential hypertensive males. Kidney Int. 1994;46(3):780–788. doi: 10.1038/ki.1994.333 [DOI] [PubMed] [Google Scholar]
- 21.Nasjletti A. Arthur C. Corcoran Memorial Lecture. The role of eicosanoids in angiotensin-dependent hypertension. Hypertension. 1998;31(1 Pt 2):194–200. doi: 10.1161/01.hyp.31.1.194 [DOI] [PubMed] [Google Scholar]
- 22.Bovee DM Visser WJ Middel I, et al. A randomized trial of distal diuretics versus dietary sodium restriction for hypertension in chronic kidney disease. J Am Soc Nephrol. 2020;31(3):650–662. doi: 10.1681/ASN.2019090905 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Elsner D, Muntze A, Kromer EP, Riegger GA. Prostaglandin E2 in dogs with heart failure: hemodynamic, hormonal, and renal effects. J Cardiovasc Pharmacol. 1992;20(4):547–553. doi: 10.1097/00005344-199210000-00006 [DOI] [PubMed] [Google Scholar]
- 24.Stokes JB, Kokko JP. Inhibition of sodium transport by prostaglandin E2 across the isolated, perfused rabbit collecting tubule. J Clin Invest. 1977;59(6):1099–1104. doi: 10.1172/JCI108733 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kopp UC Cicha MZ Smith LA, et al. Dietary sodium modulates the interaction between efferent and afferent renal nerve activity by altering activation of α2-adrenoceptors on renal sensory nerves. Am J Physiol Regul Integr Comp Physiol. 2011;300(2):R298–R310. doi: 10.1152/ajpregu.00469.2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Calo LA. Vascular tone control in humans: insights from studies in Bartter's/Gitelman's syndromes. Kidney Int. 2006;69(6):963–966. doi: 10.1038/sj.ki.5000253 [DOI] [PubMed] [Google Scholar]
- 27.Kurtz TW, DiCarlo SE, Pravenec M, Morris RC. Changing views on the common physiologic abnormality that mediates salt sensitivity and initiation of salt-induced hypertension: Japanese research underpinning the vasodysfunction theory of salt sensitivity. Hypertens Res. 2019;42(1):6–18. doi: 10.1038/s41440-018-0122-5 [DOI] [PubMed] [Google Scholar]
- 28.Brooks CS, Johnson CA, Kotchen JM, Kotchen TA. Diuretic therapies in low renin and normal renin essential hypertension. Clin Pharmacol Ther. 1977;22(1):14–20. doi: 10.1002/cpt197722114 [DOI] [PubMed] [Google Scholar]
- 29.Adlin EV, Marks AD, Channick BJ. Spironolactone and hydrochlorothiazide in essential hypertension. Blood pressure response and plasma renin activity. Arch Intern Med. 1972;130(6):855–858. doi: 10.1001/archinte.130.6.855 [DOI] [PubMed] [Google Scholar]
- 30.Blaufox MD, Lee HB, Davis B, Oberman A, Wassertheil-Smoller S, Langford H. Renin predicts diastolic blood pressure response to nonpharmacologic and pharmacologic therapy. JAMA. 1992;267(9):1221–1225. doi: 10.1001/jama.1992.03480090069029 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Due to ongoing study and analyses, data are not available to share at present.