Thiazide-associated hyponatremia: a retrospective cohort study comparing hydrochlorothiazide vs indapamide vs chlorthalidone

doi:10.21203/rs.3.rs-4887478/v1

Download PDF

Research Article

Thiazide-associated hyponatremia: a retrospective cohort study comparing hydrochlorothiazide vs indapamide vs chlorthalidone

https://doi.org/10.21203/rs.3.rs-4887478/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Hyponatremia is a crucial complication of therapy with thiazide diuretics.

Aim

This study compares the epidemiological and biochemical profiles and hospital course of patients using hydrochlorothiazide (HCTZ), indapamide (INDA), and chlorthalidone (CTD) admitted with thiazide-associated hyponatremia.

Method

Retrospectively obtained data from the hospital’s digital registries. The epidemiological and biochemical parameters between the HCTZ, INDA, and CTD groups were compared. The correlation between dose and biochemical parameters in each group was performed. The thiazide groups without diuretic co-medication were compared (HCTZ vs INDA) and the correlation between dose and biochemical parameters in each group was performed.

Results

A comparison of the HCTZ (n = 135), INDA (n = 125), and CTD (n = 27) groups identified difference in serum potassium (s-K; p = 0.03). The hyponatremia correction rate was slower in the CTD group at 96 h after admission (p < 0.001). After the exclusion of diuretic co-medication, the HCTZ group (n = 64/135) showed a higher prevalence of ARBs, s-K (both p < 0.001) and a lower median (IQR) equipotent dose (12.5 (o) mg vs 2.5 (1.2) mg), prevalence of ACE-I (p < 0.001), and eGFR (p = 0.03), when compared to the INDA group (n = 109/125).

Conclusion

Except for s-K, we observed no significant difference in biochemical and epidemiological profiles between HCTZ, INDA, and CTD. After excluding the influence of other diuretics, we observed higher s-K in the HCTZ group compared to the INDA group, potentially explained by the lower equipotent dose of HCTZ. The CTD group showed a statistically significant trend of slower hyponatremia correction.

thiazide diuretics

hyponatremia

thiazide-associated hyponatremia

hydrochlorothiazide

indapamide

chlorthalidone

TAH, regardless of associated thiazide, is a biochemically well-described phenomenon that mostly affects a relatively narrow group of patients.
The different availability of fixed combination products could be a crucial factor in modifying the biochemical profile and daily dose of each thiazide.
The daily dose of any thiazide did not correlate with the severity of hyponatremia on admission.
In a real-world setting, the half-life of thiazide could be a clinically relevant factor modifying the correction rate of hyponatremia.

The term thiazide diuretics refers to a heterogeneous group of sulfur-containing compounds, which could be subdivided into compounds containing the benzothiadiazine ring (thiazide or sometimes called thiazide-type) and those lacking this chemical structure, which are called thiazide-like [1]. Both groups share the inhibition of thiazide-sensitive Na/Cl − cotransporter (NCC) in the proximal segment of the distal tubule of the nephron as their principal mechanism of action. However, it is essential to distinguish between thiazide and thiazide-like diuretics beyond their chemical structure due to differences in their pleiotropic effects, pharmacokinetics parameters, and available evidence-based trials [1, 2].

Both thiazide and thiazide-like diuretics, even 60 years after their introduction to clinical practice, have been a mainstay in the therapy of primary hypertension [3]. The trend of preferring thiazide-like diuretics such as indapamide (INDA) or chlorthalidone (CTD) over thiazide diuretics such as hydrochlorothiazide (HCTZ) is notable at present, mainly due to their greater antihypertensive efficacy and better cardiovascular morbidity and mortality data [2, 4, 5]. However, the superiority of thiazide-like diuretics has not been confirmed in large population trials [6–8]. The situation regarding the adverse effects of these subgroups and their comparison is even less clear. Generally, taking low doses of thiazide or thiazide-like diuretics is considered a rational strategy to prevent adverse outcomes without substantially compromising the benefits of their use [1]. However, even relatively low doses are observed to cause severe adverse events such as hypotonic hyponatremia.

Hyponatremia is a well-known and clinically important complication in the therapy with both, thiazide and thiazide-like diuretics [9]. In connection with hypotonic hyponatremia and thiazides, usually two terms are used: thiazide-induced hyponatremia (TIH) and thiazide-associated hyponatremia (TAH) [10, 11]. TIH refers to hyponatremia caused clearly by thiazide or thiazide-like diuretics. TIH is dose-independent, affecting only susceptible populations, and shows great re-challenge potential [12]. The more extensive term TAH encompasses the hyponatremia that occurs when taking thiazide or thiazide-like diuretics. Since diagnosing TIH is highly challenging in actual clinical practice, the term TAH is more widely used [10, 11].

We have limited and partially conflicting results from head-to-head trials comparing the incidence of TAH among patients using thiazide or thiazide-like diuretics. Contrary to TIH, the incidence of TAH seems dose-dependent [13]. Available data show a lower risk of TAH for HCTZ compared to CTD in equal milligram-to-milligram dose per day [14]. However, there is no significant difference in the risk of TAH with a 2-fold higher daily dose of HCTZ to CTD [14]. According to these results, a population-based cohort study by Edwards et al. did not find a difference in risk of TAH between HCTZ and CTD with matched doses of 2:1, respectively [8]. On the contrary, a previous population-based cohort study by Dhalla et al. determined a higher incidence of TAH in CTD vs HCTZ groups with no relationship to dose ratios [15]. The half-life is assumed to play a key role in the risk of TAH development as HCTZ shows a relatively short half-life (6 h), and CTD stands on the opposite side of the half-life spectrum (40–60 h) [14]. Available data for INDA, with a half-life (14 h) between HCTZ and CTD, do not show a statistically significant difference in the incidence of TAH compared to HCTZ [16].

It is essential to state that the pathophysiology of TAH is a complex process, including the increase in fluid intake, impaired free-water excretion, and osmotic inactivation of sodium [10]. Moreover, genetic factors could be crucial in specific populations. These genetic factors include polymorphism in the renal outer medullary potassium channel (ROMK) [17] or the apical prostaglandin transporter of the distal nephron [18]. Therefore, determining the role the different pleiotropic effects of individual thiazides may have, besides different half-lives, would be of interest.

Aim

Despite the high prevalence of TAH among patients admitted for hypotonic hyponatremia, comparative studies on individual thiazides or thiazide-like diuretics regarding their impact on the course of hospitalization, severity of hyponatremia, and biochemical profile are sparse.

This study aims to compare epidemiological and biochemical profiles, and the course of hospitalization of patients using HCTZ, INDA, and CTD admitted for TAH. Further, this study aims to analyze the effect of dose on admission biochemical parameters and evaluates the eventual dose-dependent relationship.

Ethics approval

Written informed consent was obtained from all the patients as a part of the informed consent to hospital admission. The study design was approved by the Ethics Committee of The Tomas Bata Hospital (2023-22).

The study design was set according to our previous studies with the extension of the enrolling period [19, 20]. The design was purposely set to reflect actual clinical practice, in which thiazide diuretics are commonly combined with other antihypertensive and diuretic therapy.

Inclusion and Exclusion Criteria

Patients admitted to the Internal Medicine Department of the Tomas Bata Hospital in Zlín during the period between 1.1.2016–31.12.2022 with a diagnosis of hypotonic hyponatremia (defined as serum sodium ≤ 135 mmol/L and serum osmolality < 280 mmol/kg) were enrolled. Serum osmolality was calculated by the formula: serum osmolality = (2x serum Na [mmol/L]) + serum urea [mmol/L] + serum glucose [mmol/L] and/or directly measured. Exclusion criteria were: one of the measured or calculated serum osmolality ≥ 280 mmol/kg, the need for dialysis, age under 18 years, and combination therapy of ≥ 2 thiazides. Further, patients with continued use of thiazide were excluded from the analysis of the course of hospitalization.

Patient Enrollment

Patients were identified by retrospective data analysis of the hospital’s digital registries. All patients were hospitalized in the Internal Medicine Department during the inclusion period for the diagnosis of E87.1 (hypo-osmolality and hyponatremia) and/or E22.2 (SIADH) according to the International Statistical Classification of Diseases and Related Health Problems, the tenth revision (ICD-10). The study assembly is shown in Fig. 1.

Data Collection

Blood samples and urine spots were obtained at admission. The biochemical analysis of blood samples included serum Na (s-Na), K (s-K), Cl (s-Cl), osmolality (s-osmolality), urea (s-urea), creatinine (s-creatinine), uric acid (s-UA), and glucose (s-glucose). Urine spots were biochemically analyzed for urine osmolality (u-osmolality), Na (u-Na), K (u-K), Cl (u-Cl), and creatinine (u-creatinine). The estimated glomerular filtration rate (eGFR) was calculated by the CKD-EPI formula based on s-creatinine. Fractional excretions (FE) of Na, K, Cl, and uric acid (UA) were calculated by the formula FE = (u-Ion x s-creatinine)/(s-Ion x u-creatinine)*100.

Clinical symptoms and medication history were recorded as part of the admission protocol or obtained during the hospital stay by the patient or their caretakers if the patient was not conscious at admission. Mortality was obtained directly from the hospital’s digital register. All data were analyzed retrospectively.

Statistical Analysis

Normal distribution was tested using the‎ Shapiro–Wilk test. Since almost all parameters are not normally distributed, we used non-parametric tests for further processing. Group comparisons of categorical variables were done with a chi-square test. For group comparisons of quantitative variables, the Mann-Whitney test for two medians and the Kruskal-Wallis test for three medians were done. Correlations were assessed using Spearman’s rank correlation analysis. The statistical analyses were performed with SOFA Statistics version 1.5.3 (Paton-Simpson & Associates Ltd, Auckland, New Zealand) and R Statistical Software (R Core Team; 2021).

The inclusion criteria for the analysis of admission parameters were met by 284 patients (HCTZ, n = 135; INDA, n = 122; CTD, n = 27). Inclusion criteria for analysis of the course of hospitalization were met by 277 patients (HCTZ, n = 133; INDA, n = 118; CTD, n = 26.) A crude comparison of HCTZ vs INDA vs CTD groups is shown in Table 1. It established differences in the prevalence of potassium-sparing diuretics (p < 0.001), angiotensin II receptor antagonists (ARBs; p < 0.001), angiotensin-converting enzyme inhibitors (ACE-I; p < 0.001), and s-K (p = 0.03). Also, a difference in the correction rate of hyponatremia at 96 h after admission was observed (p < 0.001). The correction rate of hyponatremia is shown in Fig. 2.

No statistically significant difference was observed for the rest of the monitored parameters, including s-Na (p = 0.83) and mortality (p = 0.11). The median daily dose (interquartile range; IQR) was found to be 25 (12.5) mg in the HCTZ group, 2.5 (1.25) mg in the INDA group, and 12.5 (12.5) mg in the CTD group.

Table 2 shows the correlations between dose or s-K and biochemical parameters. They revealed no statistically significant results.

Further, patients using thiazide with another diuretic co-medication were excluded (n = 106). The remaining five CTD patients were excluded to avoid biasing the statistical analysis. A comparison of patients using thiazide without diuretic co-medication (HCTZ, n = 64; INDA, n = 109) is shown in Table 3. It shows a lower prevalence of female gender (p = 0.03), eGFR (p = 0.03), and higher s-K (p < 0.001) in the HCTZ group, when compared to the INDA group. The median daily dose (IQR) was found to be 12.5 (0) mg in the HCTZ group and 2.5 (1.2) mg in the INDA group.

The correlations between dose or s-K with biochemical parameters in patients using thiazide without diuretic co-medication are shown in Table 4. They revealed a negative correlation between HCTZ dose and s-K (rs = − 0.32; p = 0.03), U-K (rs = − 0.43; p = 0.01), and U-osm (rs = − 0.36; p = 0.03). In the INDA group, a positive correlation between s-K and FE-K was observed (rs = 0.29; p = 0.02). For the rest of the observed parameters, we did not find any significant correlation.

Statement of key findings

This study compares the epidemiological and biochemical profiles and hospital course of patients using HCTZ, INDA, and CTD admitted with TAH. Despite different half-lives and pleiotropic effects of individual thiazides, we determined no significant difference in biochemical and epidemiological profiles between HCTZ, INDA, and CTD, except for s-K. After excluding the influence of other diuretics, we identified higher s-K in the HCTZ group compared to the INDA group, potentially explained by the lower equipotent dose of HCTZ compared to INDA. Further, in conditions of real-world practice, differences between individual thiazides did not play a significant role in the mortality. However, we observed a statistically significant trend of slower hyponatremia correction in the CTD group.

Strengths and weaknesses

This is the first study directly comparing individual thiazides regarding their biochemical profile, including their fractional excretions. The study's design reflects the real use of thiazide diuretics in common clinical practice, where thiazides are routinely combined with other antihypertensive and diuretic therapy.

On the other hand, the results of this work must be considered in the light of serious limitations. Firstly, some measurements are partially absent. Secondly, different ranges of individual thiazides in fixed combinations could significantly affect their biochemical profiles and obscure their pharmacological effects. Thirdly, we could not follow proper patient adherence due to the study's retrospective setting. Fourthly, the number of patients in the CTD group was comparatively minor. Fifthly, we are not able to determine the duration of thiazide treatment before hospital admission.

Interpretation and further research

Characteristics of TAH, in addition to hypotonic hyponatremia itself, usually include slightly lower s-K, higher FE-K, FE-UA, prevalence of female gender, and advanced age [10, 21–23]. As shown in Table 1, a crude comparison of HCTZ, INDA, and CTD shows almost identical biochemical and epidemiological profiles of each molecule, consistent with the previous description of TAH. This depicts TAH, regardless of associated thiazide, as a biochemically well-described phenomenon mostly affecting a relatively narrow group of patients. Notably, the median daily doses (IQR) of HCTZ, INDA, and CTD shown in crude comparison (Table 1) could be considered as equipotent − 25 (12.5) mg vs 2.5 (1.25) mg vs 12.5 (12.5) mg, respectively. It is unknown whether the doses of thiazides shown in the study are equivalent to those used in the general population.

However, a crude comparison of HCTZ, INDA, and CTD groups established differences in the prevalence of potassium-sparing diuretics, ARBs, ACE-I (all p < 0.001), and s-K (p = 0.03). The different prevalence of potassium-sparing diuretics is plausibly due to the various availability of fixed combination products on the market. In Czechia, CTD is available only as a fixed combination with amiloride or atenolol. HCTZ is available both alone and in fixed combination with amiloride or other antihypertensives. INDA is not available in fixed combination with potassium-sparing diuretics at all, only as a single drug or in combination with other antihypertensives. Interestingly, a biochemical analysis suggests that the difference in s-K is not due to the varying prevalence of potassium-sparing diuretics. The CTD group, which has the highest prevalence of potassium-sparing diuretics (CTD: 20/27; 74.1% vs HCTZ: 67/135; 49.6% vs INDA: 5/122; 4.1%), did not exhibit the highest s-K levels. Further, as shown in Table 2, s-K did not correlate with U-K and FE-K for any of the thiazides. Our analysis also did not find a correlation between thiazide dose or eGFR with s-K in any thiazide group.

The incidence of TAH is considered to be a dose-dependent side effect [13]. This relationship does not apply to the severity of hyponatremia as we did not find a significant correlation between dose and s-Na in any thiazide group (Table 2). Also, considering the absence of difference in s-Na between the HCTZ, INDA, and CTD groups (p = 0.83; Table 2), a plausible explanation could be the necessity of reaching a certain hyponatremia level, which leads to the manifestation of clinical symptoms and is subsequently the reason for hospital admission. In our previous study, this threshold level in patients admitted for TAH was lower compared to other etiologies of hyponatremia but without a difference in in-hospital mortality [19].

Table 1 shows that we observed no differences in mortality (p = 0.11) or in the rate of hyponatremia correction between HCTZ, INDA, and CTD groups the next morning, 24 h, and 48 h after admission (p = 0.63; 0.78; 0.47, respectively). However, at 72 h and 96 h after admission, a statistically significant trend of slower rate of hyponatremia correction is notable in the CTD group (approaching a statistical significance at 72 h, p = 0.052; highly significant at 96 h, p < 0.001), compared to the HCTZ and INDA groups (see Fig. 2). This may be attributed to the longer CTD half-life (40–60 h) compared to HCTZ and INDA (6 h and 14 h, respectively) [1]. A plausible explanation for the statistically same initial increase in s-Na could be the admission of patients to the intensive care unit (ICU), where different half-lives of each thiazide are effectively blunted.

Of the patients, 37% (106/284) were taking another diuretic medication besides thiazide, potentially modifying their biochemical profile or even causing the hyponatremia itself [24]. Hence, we excluded patients with concomitant diuretic medication. Subsequently, we excluded the remaining five patients in the CTD group to avoid biasing the statistical analysis. Comparison of thiazides without diuretic co-medication (Table 3) shows a lower prevalence of female gender (p = 0.03), eGFR (p = 0.03), and higher s-K (p < 0.001) in the HCTZ group, when compared to the INDA group. Also, a median daily dose of the HCTZ group (12.5; IQR: 0 mg) could be considered lower when compared to the INDA group (2.5; IQR: 1.2 mg).

Part of the explanation for the lower equipotent dose in the HCTZ group compared to the INDA group after the exclusion of patients taking another diuretic co-medication can be again demonstrated in the different range of products on the market. The fixed HCTZ/amiloride combination is only available in strengths of 25/2.5 mg or 50/5 mg, respectively. HCTZ is also available in other fixed preparations, including strengths of 12.5 mg even in combination with maximum daily doses of telmisartan (80 mg), candesartan (32 mg), irbesartan (300 mg), losartan (100 mg), and valsartan (320 mg), as well as in combination HCTZ/bisoprolol 6,25/10 mg, respectively. We revealed that 54.1% (20/37) of patients in the HCTZ group without diuretic co-medication using ARBs were taking 12.5 mg of HCTZ in combination with maximum daily doses of ARBs. INDA is available in fixed combination predominantly as INDA/perindopril, always in ratio 2.5/0.625 mg, 5/1.25 mg, or 10/2.5 mg, respectively, and is registered as a combination INDA/telmisartan 2.5/80 mg, respectively. Thus, prescribers may aim to add substantially reduced doses of thiazide diuretic to the current antihypertensive medication, preferably in a single, more compound-containing pill. With HCTZ, this goal is easier to achieve.

Also, the very choice/availability of another antihypertensive in fixed combination with thiazide could be relevant for the biochemical profile of TAH. When the prescriber prefers to choose a fixed combination of thiazide with RAAS axis inhibitor, two primary combinations exist: HCTZ/ARBs and INDA/perindopril. ARBs undergo predominantly hepatic metabolism to form inactive metabolites, while the active metabolite of perindopril, perindoprilat, is eliminated primarily by renal excretion. Therefore, we may attribute lower eGFR in the HCTZ group compared to the INDA group (p = 0.03) to preference for ARBs in patients with chronic kidney disease. On the other hand, obtained eGFR value of 68 (IQR: 31) ml/min in the HCTZ group is not consistent with significant decrease of renal function. HCTZ is not metabolized and is excreted renally, unlike the INDA, which is metabolized extensively hepatically. Thus, regardless of origin, lower eGFR in the HCTZ group prolongs the half-life of HCTZ. Additionally, since half-life is a crucial factor affecting TAH incidence, further investigation is warranted to fully understand its impact.

The lower eGFR in the HCTZ, compared to the INDA group, may also explain the higher s-K in the HCTZ group when compared to the INDA group (p < 0.001) as a negative correlation between eGFR and s-K is well-known [25]. Nevertheless, we identified no correlation between s-K and eGFR in either the HCTZ or INDA groups (Table 4). Since we observed a negative correlation between HCTZ dose and s-K (rs = − 0.32; p = 0.03), one plausible explanation for the higher s-K in the HCTZ group compared to the INDA group is the lower equipotent dose of HCTZ. Remarkably, we identified a negative correlation between HCTZ dose and U-K (rs = − 0.43; p = 0.01); U-osm (rs = − 0.36; p = 0.03).

Except for s-K, we discovered no significant difference in biochemical and epidemiological profiles between HCTZ, INDA, and CTD. After excluding the influence of other diuretics, we identified higher s-K in the HCTZ group compared to the INDA group potentially explained by the lower equipotent dose of HCTZ. Further, in conditions of real-world practice, differences between individual thiazides did not play a significant role in the mortality. However, the CTD group showed a statistically significant trend of slower hyponatremia correction.

Funding

We had no funding.

Disclosure statement

We declare no conflict of interest.

Ernst ME, Fravel MA. Thiazide and the Thiazide-Like Diuretics: Review of Hydrochlorothiazide, Chlorthalidone, and Indapamide. Am J Hypertens. 2022;35:573–86.
DiNicolantonio JJ, Bhutani J, Lavie CJ, et al. Evidence-based diuretics: focus on chlorthalidone and indapamide. Future Cardiol. 2015;11:203–17.
McNally RJ, Morselli F, Farukh B, et al. A review of the prescribing trend of thiazide-type and thiazide-like diuretics in hypertension: A UK perspective. Br J Clin Pharmacol. 2019;85:2707–13.
Liang W, Ma H, Cao L, et al. Comparison of thiazide-like diuretics versus thiazide-type diuretics: a meta-analysis. J Cell Mol Med. 2017;21:2634–42.
Olde Engberink RHG, Frenkel WJ, van den Bogaard B, et al. Effects of Thiazide-Type and Thiazide-Like Diuretics on Cardiovascular Events and Mortality. Hypertension. 2015;65:1033–40.
Ishani A, Cushman WC, Leatherman SM, et al. Chlorthalidone vs. Hydrochlorothiazide for Hypertension–Cardiovascular Events. N Engl J Med. 2022;387:2401–10.
Hripcsak G, Suchard MA, Shea S, et al. Comparison of Cardiovascular and Safety Outcomes of Chlorthalidone vs Hydrochlorothiazide to Treat Hypertension. JAMA Intern Med. 2020;180:542–51.
Edwards C, Hundemer GL, Petrcich W, et al. Comparison of Clinical Outcomes and Safety Associated With Chlorthalidone vs Hydrochlorothiazide in Older Adults With Varying Levels of Kidney Function. JAMA Netw Open. 2021;4:e2123365.
Mannheimer B, Bergh CF, Falhammar H, et al. Association between newly initiated thiazide diuretics and hospitalization due to hyponatremia. Eur J Clin Pharmacol. 2021;77:1049–55.
Filippone EJ, Ruzieh M, Foy A. Thiazide-Associated Hyponatremia: Clinical Manifestations and Pathophysiology. Am J Kidney Dis. 2020;75:256–64.
Burst V, Grundmann F, Kubacki T, et al. Thiazide-Associated Hyponatremia, Report of the Hyponatremia Registry: An Observational Multicenter International Study. Am J Nephrol. 2017;45:420–30.
Friedman E, Shadel M, Halkin H, et al. Thiazide-induced hyponatremia. Reproducibility by single dose rechallenge and an analysis of pathogenesis. Ann Intern Med. 1989;110:24–30.
Ravioli S, Bahmad S, Funk G-C, et al. Risk of Electrolyte Disorders, Syncope, and Falls in Patients Taking Thiazide Diuretics: Results of a Cross-Sectional Study. Am J Med. 2021;134:1148–54.
van Blijderveen JC, Straus SM, Rodenburg EM, et al. Risk of Hyponatremia with Diuretics: Chlorthalidone versus Hydrochlorothiazide. Am J Med. 2014;127:763–71.
Dhalla IA, Gomes T, Yao Z, et al. Chlorthalidone Versus Hydrochlorothiazide for the Treatment of Hypertension in Older Adults. Ann Intern Med. 2013;158:447–55.
Roush GC, Ernst ME, Kostis JB, et al. Head-to-Head Comparisons of Hydrochlorothiazide With Indapamide and Chlorthalidone. Hypertension. 2015;65:1041–6.
Huang C-C, Chung C-M, Hung S-I, et al. Clinical and Genetic Factors Associated With Thiazide-Induced Hyponatremia. Med (Baltim). 2015;94:e1422.
Ware JS, Wain LV, Channavajjhala SK, et al. Phenotypic and pharmacogenetic evaluation of patients with thiazide-induced hyponatremia. J Clin Invest. 2017;127:3367–74.
Klhůfek J, Šálek T. Thiazide-associated hyponatremia in internal medicine patients: analysis of epidemiological and biochemical profiles. Postgrad Med. 2022;134:487–93.
Klhůfek J, Vodička M. Furosemide-associated hyponatremia in internal medicine patients: Analysis of epidemiological and biochemical profiles. Eur J Intern Med. 2023;114:138–40.
Liamis G, Filippatos TD, Elisaf MS. Thiazide–associated hyponatremia in the elderly: what the clinician needs to know. J Geriatr Cardiol JGC. 2016;13:175–82.
Nadal J, Channavajjhala SK, Jia W, et al. Clinical and Molecular Features of Thiazide-Induced Hyponatremia. Curr Hypertens Rep. 2018;20:31.
Burst V, Grundmann F, Kubacki T, et al. Thiazide-Associated Hyponatremia, Report of the Hyponatremia Registry: An Observational Multicenter International Study. Am J Nephrol. 2017;45:420–30.
Mannheimer B, Falhammar H, Calissendorff J, et al. Non-thiazide diuretics and hospitalization due to hyponatraemia: A population-based case-control study. Clin Endocrinol (Oxf). 2021;95:520–6.
Ueda Y, Ookawara S, Ito K, et al. Changes in urinary potassium excretion in patients with chronic kidney disease. Kidney Res Clin Pract. 2016;35:78–83.

Tables 1 to 4 are available in the Supplementary Files section.

Tables.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Thiazide-associated hyponatremia: a retrospective cohort study comparing hydrochlorothiazide vs indapamide vs chlorthalidone

Status:

Version 1

Abstract

Background

Aim

Method

Results

Conclusion

Figures

Key Points

Introduction

Methods

Statistical Analysis

Results

Discussion

Conclusion

Declarations

References

Tables

Supplementary Files

Status:

Version 1