The ‘cold case’ of chlortalidone vs. hydrochlorothiazide in hypertension closed by the diuretic comparison project?

Comment on ‘Chlorthalidone vs. Hydrochlorothiazide for Hypertension–Cardiovascular Events’ which was presented at the American Heart Association (AHA) Scientific Sessions 2022 which was published in the New England Journal of Medicine,https://doi.org/10.1056/NEJMoa2212270 The Diuretic Comparison Project1 was a multicenter, pragmatic, open-label trial that investigated whether chlorthalidone (CHLOR) was superior to hydrochlorothiazide (HCTZ) for preventing major adverse cardiovascular events in patients with hypertension. The study randomly assigned patients in the Department of Veterans Affairs health system aged ≥ 65 years with hypertension who had been receiving HCTZ at a daily dose of 25 or 50 mg to continue the ongoing therapy or to switch to CHLOR at a daily dose of 12.5 or 25 mg. Patients taking antihypertensive medications that contained HCTZ combined with other agents were excluded. The primary outcome was a composite of non-fatal myocardial infarction (MI), stroke, heart failure (HF) resulting in hospitalization, urgent coronary revascularization for unstable angina, or non-cancer-related death. A total of 13 523 subjects were randomized (6756 assigned to the CHLOR group and 6767 to the HCTZ group), with a mean age of 72 years, a mean systolic blood pressure (BP) at baseline of 139 mmHg and a mean number of prescribed medications of 2.6. Ninety seven percent of patients were men, 15% were Black, ∼45% had diabetes, and 11% had a history of stroke or MI. At baseline, 95% of the subjects were receiving a daily dose of 25 mg of HCTZ. The median follow-up was 2.4 years. The primary composite outcome occurred in 10.4% of patients in the CHLOR group and in 10% of those treated with HCTZ (hazard ratio [HR], 1.04; 95% confidence interval [CI], 0.94–1.16; P = 0.45). The observed annual event rate was comparable between the two groups (4.5% vs. 4.3% in the CHLOR and HCTZ groups, respectively) and higher than expected (3.0%). There were no differences in the incidence of MI (HR, 1.01; 95% CI, 0.80–1.28), stroke (HR, 1.00; 95% CI, 0.74–1.36), HF hospitalizations (HR, 1.04; 95% CI, 0.87–1.25), revascularization for unstable angina (HR,1.54; 95% CI, 0.77–3.10), non-cancer-related death (HR, 1.01; 95% CI, 0.88–1.17), and death from any cause (HR, 1.00; 95% CI, 0.87–1.13). Among the patients with history of MI or stroke, the incidence of the primary outcome was lower in the CHLOR group compared to HCTZ (14.3% vs. 19.4%; HR, 0.73; 95% CI, 0.57–0.94). On the other hand, patients in the CHLOR group without history of previous MI or stroke had a slightly higher incidence of the primary outcome. The incidence of hypokalemia was higher in the CHLOR-treated patients than in those receiving HCTZ (6.0% vs., 4.1%, P <0.001). Thiazide and thiazide-like diuretics are recommended as a first-line drug class in the pharmacological treatment of hypertension.2–4 Although the recent US guidelines5 recommend the preferred use of longer acting CHLOR, HCTZ remains the preferred clinical choice. The issue whether CHLOR may be superior to HCTZ in terms of reduction of hypertension-related cardiovascular events has been long advocated though never proven, since the results of previous studies are non-conclusive and often conflicting. In the LEGEND (Large-Scale Evidence Generation and Evaluation in a Network of Databases)6 study no significant difference was found in the risk of MI, hospitalized HF or stroke in patients treated with CHLOR compared with HCTZ. However, CHLOR was associated with a significantly higher risk of hypokalemia, acute renal failure, chronic kidney disease and diabetes. In another large study including 29 873 participants, CHLOR did not produce a significant reduction of the composite outcome of death, hospitalization for HF, stroke, or MI compared to HCTZ with an increased incidence of hypokalemia and hyponatremia.7 In a retrospective observational cohort study from the Multiple Risk Factor Intervention Trial (MRFIT) the incidence of cardiovascular events was significantly lower in subjects treated with CHLOR or with HCTZ compared with those who took neither drug. When comparing the two drugs, CHLOR produced a significant reduction of BP and cardiovascular events.8 The results of the Diuretic Comparison Project1 support a comparable efficacy of CHLOR and HCTZ at the dosages tested on hypertension-related cardiovascular events. Major strengths of this study are represented by the large sample of patients enrolled and by the pragmatic methods of the Department of the Veterans Administration Healthcare System which provides a reliable real-world assessment of the two treatments in the routine clinical care. Several limitations of the study, however, deserve to be highlighted. First, almost the entire study population (94.5%) was treated with the lower dose of HCTZ (25 mg/day) at entrance, and hence, according to the study design received the lower dose of CHLOR (12.5 mg/day). Therefore, the results cannot be extended to higher dosages of the two diuretics. Second, patients who received HCTZ in a single-pill combination were excluded by the study. This appears to be in contrast with the recommendations of the most recent international guidelines which recommend single-pill combinations of two or more drugs.2–4,9 On the other hand, ∼87% of the patients in both groups received other anti-hypertensive drugs in free combinations (2.6 drugs on average). No information is provided about the drugs prescribed in both groups in association to CHLOR or HCTZ, nor about the concomitant use of aspirin, statins and glucose-lowering drugs which might have influenced the results. Third, the study population was mostly composed by White, male, relatively old subjects with a mean age of 72 years and thus the results cannot be extended to other demographic groups. Fourth, both physicians and patients were aware of the assigned treatment group, and this might have influenced the therapeutic management during the study follow-up. Indeed, 15% of the participants who had been randomly assigned to receive CHLOR reverted to HCTZ therapy, but only 3.8% of those who had been assigned to continue receiving HCTZ switched to CHLOR. It is important to note that baseline systolic BP remained substantially unaltered during the study and that BP trends were comparable in the two groups not supporting the suggested more potent BP-lowering effect of CHLOR,10 although this may be the consequence of the concomitant antihypertensive therapy. The study confirms that CHLOR may cause, even when the lowest available dose is used, more frequent hypokalemia and hospitalizations due to hypokalemia than HCTZ, as reported in previous studies.6,7 Finally, the substantially divergent influence of CHLOR in patients with or without previous MI or stroke is interpreted by the same authors as an unexpected finding most likely due to chance, and should not be overinterpreted. Based on the results of the current study and of previous evidence, it appears reasonable to conclude that there are no significant differences in the cardiovascular benefits achieved with these commonly used thiazide diuretics in hypertension. Hence, the choice should be mostly based on the availability of associations with other pharmacological classes at different dosages, possibly in single-pill formulations as recommended in the guidelines. Although the Diuretic Comparison Project should be praised for having addressed a long-debated issue, there is room left to explore whether larger doses of these diuretics may reveal a different impact on BP and on cardiovascular outcomes. There are no new data associated with this article. All authors declare no funding for this contribution.