Background
Kidney calculi are a common urinary system disease, and their incidence is increasing annually [
1]. In the clinical guideline of the American College of Physicians, the prevalence of kidney calculi in men and women was 13% and 7%, respectively, and the 5-year recurrence rate of untreated kidney calculi was 35–50% [
2]. Correspondingly, the high recurrence rate yielded a certain economic burden to patients with kidney calculi. Kidney calculi are caused by abnormal accumulation of some crystalline substances, such as calcium, oxalic acid, uric acid, and cystine, and organic matrices, such as matrix A and acid mucopolysaccharide, in the kidney. Most kidney calculi are composed of calcium; calcium oxalate stones are the most common, accounting for 74.8% of all cases of stones [
3]. In response to increasing incidence of kidney calculi, various interventions have been performed to prevent the occurrence of stones, including dietary interventions and medical treatments [
4‐
6]. Among them, thiazide diuretics are the commonly used drugs for preventing recurrent kidney calculi [
7,
8]. However, some clinical trials reported that thiazide diuretics had no significant prophylactic effect compared with the placebo [
9,
10]. We also found that the recommended grade of thiazide diuretics to prevent recurrent kidney calculi was not consistent among guidelines [
2,
7,
8,
11]. Further, no meta-analyses have yet been conducted to prove that thiazide diuretics can prevent recurrent renal calculi. Therefore, we performed a systematic review and meta-analysis of clinical trials that investigated thiazide diuretics for the prevention of recurrent renal calculi to provide evidence-based medical data for use in clinical practice. We aimed to evaluate the effect and safety of thiazide diuretics on recurrent kidney calculi patients, comparing the incidence of recurrent kidney calculi and the 24-h urinary calcium level with that of placebo or no medication group in randomized controlled trials (RCTs).
Methods
Trial search
Two researchers independently searched all studies with keywords of thiazide diuretics and kidney calculi published in the PubMed, EMBASE, and Cochrane Library databases. The following search terms were used: “kidney stones”, “kidney calculi”, “renal calculi”, “nephrolith”, “thiazide diuretics”, “sodium chloride cotransporter inhibitors”. The article search was limited by study design of RCTs. The search was performed until December 31, 2018. In addition, the references of the included studies were also retrieved to supplement the relevant research, including gray literatures (e.g., clinical trials). When the opinions of the two researchers differed, a third researcher was consulted. When there were other languages used, we sought help from linguists.
Inclusion and exclusion criteria
The inclusion and exclusion criteria for the studies were determined prior to data extraction. The inclusion criteria included the following: (1) RCTs with thiazide diuretic administration as the intervention and placebo or no medication as the control condition; (2) patients with renal calculi or hypercalciuria as the study subjects; and (3) test indicators including at least one of the following: number of patients with new stones, 24-h urinary calcium level, 24-h urinary oxalate level, and serum calcium level. Conversely, the exclusion criteria included the following: (1) non-RCTs; (2) RCTs without a treatment group or placebo group; (3) trials with incomplete or no data available; and (4) other study types, e.g., abstract, case report, and review.
Risk of bias assessment
Two investigators independently searched for studies using different search strategies and screened for studies that met the inclusion criteria. The researchers evaluated the quality of the included RCTs using the Cochrane Library risk bias assessment tool. The seven items used to assess bias in each trial included random sequence generation, allocation concealment, double blindness of participants and trial performers, blindness of outcome assessment, incomplete outcome data, selective reporting, and other biases. Each quality item was divided and categorized into high risk, undefined risk, and low risk. The quality of the included trials was rated as low quality, high quality, or medium quality according to the following criteria: (1) if random sequence generation or allocation concealment was assessed to be of a high risk, the trial would be considered to be of low quality regardless of the risk of other projects; (2) if random sequence generation and allocation concealment were assessed to be of a low risk, and all other items were assessed to be of a low or an undefined risk, the test would be of high quality; and (3) if the tests did not meet the high risk or low risk criteria, the quality of the trial was considered moderate [
12].
Two investigators extracted the following data from the included studies: primary author; year of publication; sex, age; sample size; interventions; control group; number of lost visits; and follow-up time. When the trials were greater than two sets or had multi-factor designs, we only extracted content relevant to this study.
Statistical analysis
We performed a meta-analysis and calculated the relative ratio (RR), risk difference (RD), standardized mean difference (SMD), and 95% confidence interval (CI). The pooled RR and RD were used to estimate the efficiency of thiazide diuretics on recurrent kidney calculi by using Mantel–Haenszel method. The pooled SMD were used to evaluate the effect of thiazide diuretics on 24-h urinary calcium level by using inverse variance method. We used
I2 statistics to assess statistical heterogeneity. An
I2 value of 0–25% indicates no significant heterogeneity; 26–50%, low heterogeneity; 51–75%, moderate heterogeneity; and > 75%, high heterogeneity [
13]. Data were pooled using the fixed-effects model when the
I2 value was < 50%; data were pooled using the random-effects model when the
I2 value was > 50%. The test level was set at
α = 0.05;
P values of 0.05 were considered to indicate that the difference was statistically significant. We used the Review Manager 5.2 software (The Cochrane Collaboration, Oxford, UK) to perform the meta-analysis and forest plot analysis and the Stata 13.0 software (Stata Corp, College Station, TX) to conduct the publication bias test (Egger’s test). Evidence quality grading was performed for each outcome measure with reference to the GRADE criteria, and recommendations for the prevention of recurrent kidney calculi using thiazide diuretics were reassessed on the basis of a decision table formed according to the recommendations of the WHO Handbook for Guideline Development [
14]. To assess whether the efficacy of thiazide diuretics in preventing recurrent kidney calculi is related to their clinical features, we performed a subgroup analysis based on the duration of drug action; we also did a subgroup analysis based on quantitative methods of 24-h urinary calcium. For the robustness of the results, we conducted a sensitivity analysis.
Discussion
Satisfactory treatment outcomes have not yet been obtained for kidney calculi, and their recurrence rate remains a concern [
21‐
24]. Previously, multiple RCTs and reviews have reported a decreased incidence of recurrent kidney calculi with diet control (e.g., high fluid intake and calcium intake limitation) and thiazide diuretic, alkali citrate, and allopurinol administration. Herein, the overall efficacy of thiazide diuretics in reducing the incidence of recurrent renal calculi was limited when compared to the placebo and untreated groups and the pooled RD with 95% CI was − 0.23 (− 0.30 to − 0.16). The quality of evidence was low owing to the insufficient samples and the inconsistent results among the studies when the GRADE criteria were used to assess the incidence of stone events. This result considerably decreased our expectation of the efficacy of thiazide diuretics in reducing the incidence of recurrent kidney calculi.
We reviewed the clinical guidelines and found that the 2016 updated edition of the Canadian Urological Association guidelines considered thiazide diuretics as a highly recommended drug for preventing recurrent kidney calculi (level of evidence: 1–3 and grade A–B recommendation, based on the Oxford levels of evidence and grades of recommendation). However, the grade according to the American College of Physicians guidelines was weak (grade: weak recommendation, moderate-quality evidence).
In addition to concerns regarding stone events, we also observed a decrease in the serum potassium level and an increase in the uric acid level in some thiazide diuretic groups among the eight studies. Further, 3.7–20% of the patients in the thiazide diuretic groups withdrew from the trials owing to the development of adverse reactions (e.g., hypokalemia, elevated uric acid levels, and abnormal blood glucose and cholesterol levels). Clinically, the main adverse reactions of thiazide diuretics are as follows: (1) water and electrolyte disturbance, such as hypokalemia and hyponatremia; (2) cardiovascular problems, such as blood volume insufficiency and orthostatic hypotension; (3) gastrointestinal reactions, such as anorexia, nausea, gastric irritation, and constipation; (4) central nervous system problems, such as dizziness, paresthesia, and headache; and (5) abnormalities in related metabolic indicators, such as hyperglycemia and elevated total cholesterol levels [
25‐
27]. Makam et al. [
28] showed that 14.3% of thiazide users and 6.0% of non-users had adverse reactions (serum sodium level of < 135 mmol/L; serum potassium level of < 3.5 mmol/L; and estimated glomerular filtration rate reduction by > 25% compared with that at baseline) (
P < 0.05). In the ALLHAT trial, the incidence of newly diagnosed diabetes was 17.1% in patients with metabolic syndrome after using chlorthalidone for 4 years; that in patients without metabolic syndrome was 7.7% (
P < 0.05) [
29].
However, the patients’ compliance was poor. Dauw et al. [
30] found that when patients with kidney calculi were treated with a single prophylactic drug, the proportion of patients in the thiazide diuretic groups who followed the doctor’s advice was only 42.5%. Even in the presence of cardiovascular diseases (e.g., hypertension), patients in the thiazide diuretic groups were equally less docile (42%) [
31].
Taken together, thiazide diuretics yielded several adverse reactions, and the patients’ compliance was low. Based on the decision table formed by the recommendation in the WHO Handbook for Guideline Development and combined with the degree of benefits for patients with kidney calculi, patients’ compliance, adverse reactions caused by long-term medication, and economic burden, thiazides are not recommended for use in preventing recurrent kidney calculi.
Our study also has some limitations: (1) although we searched each database using keywords, not all relevant documents were included, such as unpublished literature; (2) most of the studies included have been conducted several years ago; (3) the kidney calculi investigated in this study were all calcium stones. Whether thiazide diuretics have any effect on other types of kidney calculi remains unclear; thus, the results of this meta-analysis should be interpreted with caution; (4) some of the included studies assessed experimental groups and control groups with other measures as diet, fluid therapy or potassium chloride, which may be a source of heterogeneity; and (5) we only described the adverse reactions qualitatively and didn’t conduct meta-analysis of the safety of thiazide diuretics for recurrent kidney calculi.
Conclusions
Long-term use of thiazide diuretics can reduce the incidence of recurrent renal calculi and 24-h urinary calcium level. However, the benefits are insufficient, and the quality of evidence is low. Considering the adverse effects, patients’ preferences, and economic burden of long-term medication, we do not recommend the use of thiazide diuretics to prevent recurrent kidney calculi.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.