Effect of the combination of bumetanide plus chlorthalidone on hypertension and volume overload in patients with chronic kidney disease stage 4–5 KDIGO without renal replacement therapy: a double-blind randomized HEBE-CKD trial

Approximately 85% of patients with advanced chronic kidney disease (glomerular filtration rate < 30 ml/min) suffer from hypertension mostly as a result of volume increase [1]. Given that volume overload is the primary contributing factor to the genesis and maintenance of hypertension in chronic kidney disease patients, loop diuretics have been widely used for treating this subgroup of patients [2]. However, chronic loop diuretic use leads to reduced efficacy over time because cells from the distal convoluted tubule undergo hypertrophy in response to the abnormally high sodium concentrations resulting from blockage of the Na–K-Cl cotransporter (NKCC) in Henle’s loop, facilitating the distal reabsorption of sodium and chlorine [3]. This phenomenon is known as resistance to loop diuretics and is manifested by the presence of volume overload and arterial hypertension despite the chronic use of loop diuretics [4]. One approach to overcome loop diuretic resistance is the addition of a thiazide-type diuretic to produce diuretic synergy via "sequential nephron blockade," first described more than 40 years ago in patients with heart failure [5]; however, some studies that show the potential positive effect of adding a thiazide diuretic in patients with chronic kidney disease reinforce the need for a randomized trial to demonstrate the safety and efficacy of thiazides in advanced chronic kidney disease [6, 7]. Thiazide diuretics decrease blood pressure by blocking the thiazide-sensitive Na + -Cl − symporter, preventing reabsorption of sodium (Na +) and chloride (Cl −) ions from the distal convoluted tubule. When this occurs, NaCl and water are released to the lumen, and the amount of urine produced per day increases; however, their use is not recommended in advanced chronic kidney disease due to their suspected lack of natriuretic and antihypertensive effects when the glomerular filtration rate falls below 30 ml/min [7]. This notion has been recently disputed, as the effect of hydrochlorothiazide has been recognized in patients with advanced chronic kidney disease [8]. In the same way, the CLICK study was recently published, which showed the efficacy of chlorthalidone in controlling blood pressure in patients with CKD stage 4 KDIGO [9]. The aim of our study was to evaluate the additional effect of co-administration of bumetanide with chlorthalidone on blood pressure and volume overload in patients with advanced chronic kidney disease and chronic loop diuretic treatment and without renal replacement therapy.

Fifty-one patients were assessed for eligibility, and 34 were randomized (17 in each group). Finally, sixteen participants from each group were included in the analysis. Two participants (one from each group) were excluded from the analysis because they did not complete the follow-up. The recruitment period lasted 3 months (Fig. 1).

Our results demonstrated the effect of combined bumetanide and chlorthalidone therapy in patients with advanced chronic kidney disease. Among the significant findings, a direct effect of combined therapy with respect to total body water levels and blood pressure was found throughout the study. Trials up to the time of this study evaluating the efficacy and safety of bumetanide plus chlorthalidone treatment with diuretics in patients with chronic kidney disease are limited [6]. Current KDIGO guidelines do not recommend thiazide diuretic use in advanced chronic kidney disease, arguing that the efficacy of this type of diuretic diminishes proportionally with the glomerular filtration rate, suggesting the substitution of thiazide diuretics with a loop diuretic when the glomerular filtration rate dips below 30 ml/min [7]. This recommendation originated from a study conducted in 1961, in which a poor natriuretic effect of thiazides was observed in two patients with glomerular filtration rates below 15 ml/min [14]. Hoshino et al. found that the combination of hydrochlorothiazide and loop diuretics improved blood pressure levels and decreased proteinuria in eleven advanced stage type 2 DKD patients with severe edema [15]. Previously, in the nineties, Fliser et al. showed that the combination of torsemide with a thiazide diuretic significantly increased urinary sodium and chloride elevation in ten patients [16]. Recently, Agarwal et al. showed in an excellent clinical trial the positive impact on blood pressure of adding chlorthalidone in patients with stage 4 CKD. This trial compared the use of chlorthalidone vs placebo in patients with stage 4 chronic kidney disease and poorly controlled hypertension. A reduction of 10 mmHg in 24-h systolic blood pressure in patients who were treated with chlorthalidone was found, even if they were previously treated with loop diuretics, although these patients had a higher risk of significantly increasing serum creatinine levels [9]. In our study, the mean blood pressure reduction was found to be higher than the results observed in other trials. This may be due to the important role of volume overload on hypertension in patients with chronic kidney disease. There is evidence in patients undergoing hemodialysis that less than 5% of patients who are in dry weight, maintain a systolic pressure above 140 mmHg [17]. Agarwal et al. also demonstrated a significant decrease in systolic and diastolic pressure in patients who achieved dry weight [18]. All the patients included in this study suffered from significant volume overload before being randomized to the study, so it is likely that phenomenon that probably influenced our outcomes. The results obtained in our study not only demonstrate the antihypertensive effect of thiazide diuretics in advanced chronic kidney disease but also demonstrate the benefit of combined therapy in treating volume overload.

Our findings could have a positive clinical impact in management of volume overload in the chronic kidney disease population, whose prognosis has already been proven to improve with a reduction in volume overload. Szu-Chun Hung et al. employed a model to predict mortality based on patients’ extracellular water to total body water ratio. This ratio expresses the amount of extracellular water that exists compared to total body water. In healthy people, the index is between 0.36 and 0.39, however, it has been observed that it is increased in patients with advanced chronic kidney disease and has been associated with multiple adverse outcomes. Patients were classified into two groups based on the extracellular water to total body water ratio with a cutoff of ± 48%, and after four years of follow-up, they observed twice the mortality rate in those with levels ≥ 48% [19]. These results could be of particular interest when applied to this trial since, before the intervention, both groups had an ECW/TBW ratio > 50%; however, at the end of follow-up, only the combined treatment group exhibited a reduction to 46%, which could be impactful as a possible benefit in terms of mortality in this population.

Another finding was that treatment with loop diuretics alone was insufficient for blood pressure and volume overload control compared with bumetanide plus chlorthalidone. This was evidenced by the fact that in our trial, all patients had a history of chronic loop diuretic use (most were on furosemide); however, when they were randomized, furosemide was changed to bumetanide, a more potent loop diuretic with a dose equivalent to 160 mg furosemide QD [20]. This protocol, along with our findings, allowed us to demonstrate that high doses of loop diuretics are not sufficient by themselves to achieve adequate blood pressure and volume overload in patients with advanced chronic kidney disease without renal replacement therapy. When employing high doses of bumetanide, the possibilities of poor efficacy of the loop diuretic secondary to pharmacokinetic processes associated with diuretic resistance, such as insufficient dosage, reduced absorption, low bioavailability, and interaction with uremic toxins, are diminished [21]. We consider that this difference could be explained by chronic loop diuretic use and previously described adaptive changes at the distal convoluted tubule, in which patients exhibit higher sodium reabsorption, lower natriuresis, and, as a consequence, loss of loop diuretic efficacy [22, 23].

Thiazide diuretic use, particularly chlorthalidone, seems to provide additional benefits; chlorthalidone has a potency at least threefold that of hydrochlorothiazide, as well as a longer half-life [24]. A recent meta-analysis demonstrated that chlorthalidone has a larger antihypertensive effect than hydrochlorothiazide with a comparable safety profile [25]. Other studies have demonstrated that chlorthalidone may be superior to hydrochlorothiazide with respect to inducing regression of left ventricular hypertrophy and preventing cardiovascular events [26, 27]. As described in other studies, our results suggest that the effect of bumetanide plus chlorthalidone therapy on blood pressure levels could be related to the natriuretic effect with subsequent volume depletion produced by thiazide diuretics [28]. Even studies conducted in anuric patients failed to demonstrate an antihypertensive effect of thiazides independent of natriuresis and volume depletion [29]. Regarding the dose of chlorthalidone used in our study, we recognize that these doses are higher than the 12.5 to 25 mg/day dose that is usually recommended in patients with arterial hypertension [30]; however, doses > 50 mg (optimal doses) have been evaluated and defined in the context of high-dose diuretic therapy, as derived from studies that evaluated the dose–response curve of thiazide diuretics [31]. The pharmacokinetics of chlorthalidone doses between 50 and 200 mg reveal optimal erythrocyte half-life values and urinary excretion in patients with CKD [12]. In addition, we considered pharmacokinetic adjustment according to the deterioration in the glomerular filtration rate present in our patients to achieve the expected natriuretic effect [13]. We must recognize that there are no update pharmacokinetic studies of chlorthalidone in patients with advanced chronic kidney disease, which represents a challenge and a weakness of our study. Based on the above and based on the results of our work, it would be important to design pharmacokinetic studies of chlorthalidone with or without loop diuretics in patients with advanced chronic kidney disease as a future area of research.

Actual estimates indicate that nearly 40% of patients diagnosed with advanced chronic kidney disease decide not to start renal replacement therapy, choosing conservative treatment instead [32]. Our results support the notion that bumetanide plus chlorthalidone therapy could be an alternative in patients with advanced chronic kidney disease who have not yet accepted the initiation of renal replacement therapy to reduce volume overload and optimize blood pressure control, with the aim of achieving a better quality of life. However, this pharmacological therapy with diuretics will always be limited by the prerequisite of residual uresis in patients subject to this treatment modality.

On the other hand, BNP has been correlated with volume overload in patients with chronic kidney disease measured by bioimpedance [33, 34]; however, in our trial, BNP levels did not behave in a manner that reflected a reduction in volume overload, as evidenced by bioimpedance vector analysis. This result can be explained by results obtained in other studies in patients subject to renal replacement therapy in which it was demonstrated that a significant reduction in BNP levels (< 100 pg/dL, only attainable after multiple hemodialysis sessions) is associated with ECW/TBW ratios < 40% [35, 36]. For this reason, we consider that even though our patients exhibited decreased ECW/TBW ratio, at the end of follow-up, none achieved the ideal dry weight. Therefore, the reduced volume overload observed in the bumetanide plus chlorthalidone group was not sufficient to diminish cardiac muscle stress or achieve a significant reduction in BNP levels.

We observed the same phenomenon in plasma urea levels and glomerular filtration rate (GFR), both associated with a nonsignificant deterioration in patients on bumetanide plus chlorthalidone treatment compared with bumetanide plus placebo. The proportion of patients in the bumetanide plus chlorthalidone group who experienced an increase in plasma creatinine levels could be conceptually defined as having acute kidney injury [37]. Nevertheless, grading by AKIN or RIFLE classifications (utilized for acute kidney failure diagnosis) has not been applied in advanced chronic kidney disease populations [38]. Among the reasons for not applying these criteria to diagnose acute kidney failure in patients with advanced chronic kidney disease and chronic diuretic use is the fact that urinary volume measurements, considered part of the diagnostic criteria for acute kidney failure, cannot be applied because urinary volume in this population is variable and dependent on residual renal tubular function [39]. Employing serum creatinine elevation as a diagnostic criterion for acute kidney failure, considering that basal creatinine levels in the patients included in our study were near 4 mg/dL, would not be adequate, as serum creatinine in these stages of kidney disease is more a marker of nutritional status than of kidney function. However, we decided to use two cut-off points for serum creatinine levels to assess renal function impairment. The first, an elevation greater than 0.3 mg/dl, a cut-off point with high sensitivity, to detect differences in creatinine elevation between both groups, as was the case in the study. The second, a double elevation of baseline creatinine levels, a cut-off point with high specificity that could be more related to a clinically significant elevation, which did not occur in any of the groups.

One of the main adverse effects reported in the literature associated with thiazide use is hydroelectrolytic imbalance, hyperuricemia or an increased diabetes risk [40‐42]. Our results did not show any significant differences in uric acid risk (there was no gout attack in any group), blood glucose, potassium or sodium levels at the end of follow-up. There was one case of acute myocardial infarction in the intervention group, which required endoprosthesis insertion via catheterization. Cardiovascular function improved significantly following intervention; however, kidney function remained unchanged, with final GFR readings similar to baseline.

This study has some others limitations. First, the follow-up in our study was short; therefore, further prospective trials are needed to assess the long-term safety and efficacy of this regimen. Second, there was patient noncompliance with dietary recommendations, even though all patients were evaluated and counseled in terms of renal nutrition, with an emphasis on sodium dietary restriction, as it is known that sodium excretion is directly proportional to kidney function and sodium dietary intake [43]. Third, this trial was done at a single center which could limit generalizability. Fourth, this study does not include hard clinical outcomes, which may be interesting to investigate in future studies. Finally, most of our patients did not have a basal echocardiogram at the time of enrollment. Having an estimate of ventricular function beforehand may have allowed us to identify the presence of cardiac failure or changes in left ventricle ejection fraction or telediastolic volume, which may have explained the observed BNP behavior.

In conclusion, our study demonstrated the efficacy of chlorthalidone and bumetanide on volume overload and blood pressure reduction in advanced chronic kidney disease patients. Our results suggest that at the one-month follow-up, bumetanide plus chlorthalidone therapy is effective in patients with advanced chronic kidney disease. Volume overload reduction was adequately tolerated in this population; however, further studies with larger samples and longer follow-up periods are warranted. Based on this trial and those conducted previously on thiazide use for advanced chronic kidney disease, we propose that current guidelines should be reviewed to perhaps change the notion that thiazides should not be used in advanced CKD patients, as it has been demonstrated that this class of diuretics can be helpful in the treatment of this population.