ETIFIC was a randomized controlled open label trial carried out in 20 Spanish hospitals with HF units (2015-2018) to compare the safety and effectiveness of HF drug titration by HF-nurses vs HF-cardiologists. Its design and results have been previously published.24,25

Patients with de novo HFrEF and New York Heart Association (NYHA) II-III were included after hospitalization in a cardiology ward. Exclusion criteria were planned surgery, contraindication to BB or already receiving target or maximum tolerated dose, home or terminal care, or inability for self-care.

An active supervision system for recruitment, with centralized randomization, 4-month titration period, and a 6-month follow-up period after inclusion were established. A safety and clinical adjudication committee, blinded to the group assignment, monitored the safety of the research activity and evaluated all adverse events. Written informed consent forms were signed. The study was approved by the Clinical Research Ethics Committee of the Basque Country and complied with the Declaration of Helsinki.

ETIFIC confirmed the noninferiority safety and effectiveness of drug titration by HF-nurses vs drug titration by HF-cardiologists.

The previously published study protocol24 was based on clinical practice guidelines.1,2 The titrating professional was the HF-nurse vs the HF-cardiologist. In both cases, drug prescription was the responsibility of the cardiologist. HF-nurse tasks also included clinical assessment, education on self-care, psychosocial support, and care coordination. All HF-nurses and half of the HF-cardiologists were women.

The main objective of this post hoc substudy was to compare the safety and effectiveness of drug titration in women vs men from the ETIFIC study and to assess the possible factors associated with any differences.

To compare the achieved BB mean relative dose (% relative to target dose) after 4 months of titration in women vs men. The % of target dose was defined according to ESC HF guidelines.24

To compare the following between women and men: a) patient selection process and baseline characteristics; b) mean relative doses of angiotensin-converting enzyme inhibitors (ACEI), angiotensin II receptor blockers (ARB), and MRA at 4 months; c) percentage of patients with 100% and ≥ 50% of the target dose; d) mean relative doses and number of visits according to type and gender of the professional; e) percentage of adverse events associated with titration; f) variables influencing target dose achievement; g) rates of cardiovascular mortality and readmissions at 6 months; and h) changes in left ventricular ejection fraction (LVEF), NYHA class, 6-minute walk distance, N-terminal pro b-type natriuretic peptide (NT-proBNP) levels and quality of life scores throughout the study. Variables are shown in the design article.2 Definitions of sex and gender are provided in the supplementary data.

The analysis was performed on an intention-to-treat basis. Both the Student t-test (or nonparametric Wilcoxon test if continuous data were not normally distributed) and the chi-square test (or Fisher exact test) were used to compare the baseline sociodemographic and clinical characteristics of patients in the 2 groups (women vs men). The effect attributable to the intervention was estimated by comparing the differences in the relative dose of BB, ACEI, ARB and MRA achieved between the groups, assessed at 4 months after the start of titration, and the 95% confidence interval was calculated. We performed a multivariate analysis for the primary and secondary endpoints as predefined in the original study design.24 The model was adjusted by the variables established as relevant related factors with a possible effect on dosing, based on a review of the literature, shown in table 1 of the supplementary data. All analyses were performed considering the 2 target populations (women and men). All variables with a P value<.20 were included as explanatory variables in the multivariate model, with the relative dose as the response variable. The multivariate analysis was conducted using ANCOVA within the framework of a linear mixed regression analysis. To take into account the difference between baseline and end of the titration period at 4 months, mixed linear regression models with fixed effects and random effects (specific effect of each participant and center and the effect of time expressed as visit 1 (baseline) and visit 2 (at 4 months) were used. The models were adjusted for women, men, and the total number of patients. These models took into account the longitudinal structure of the 2 repeated measurements, as well as the hierarchical structure of the data. All the statistical analyses were performed using R (version 4.0.4): R Foundation (Statistical Computing, Vienna, Austria). Statistical significance was set at P<.05.

A total of 824 patients with de novo HFrEF, 221 women and 603 men, were evaluated, and 320 patients were included, 83 women and 237 men. Finally, 289 patients (76 women and 213 men) were analyzed at 4 months, and 274 (74 women, 200 men) were analyzed at 6 months. The selection process and causes for exclusion are shown in figure 1, and in table 2 and 3 of the supplementary data.

Figure 1.

Patient flowchart. HF, heart failure; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association.

(0.34MB).

Patient characteristics were generally well-balanced between the 2 groups, without significant differences in LVEF, ischemic heart disease, or NT-proBNP (table 1). However, women were older (4 years), had a higher proportion of systolic blood pressure (SBP) ≤ 100mmHg, lower hemoglobin level, lower 6-minute walk distance, and worse quality of life. In contrast, men had a higher proportion of smokers, alcohol abuse, atrial fibrillation/flutter, and worse scores on the Lawton Instrumental Activities of Daily Living Scale and age-adjusted Charlson index.

Baseline prescription of HF guideline-recommended drugs did not differ significantly. Women more frequently took psychotropic drugs (table 4 of the supplementary data shows other baseline characteristics).

The possible barriers to women's participation were analyzed.4,12 The proportion of women in ETIFIC (25.93%) could be considered underenrolment, given the established reference (< 32%) in a systematic review of 317 HFrEF trials.4 However, participation was higher than that in trials of BB (23%), ACEI (21%), ARB (26%) and MRA (21%) (see references of the supplementary data: BB,1-20; ACEI 21-23; ARB 24,25 MRA 26-28), but lower than that in observational optimization studies (30%)5–10 (see references 29-42 of the supplementary data). This may have been influenced by recruitment in cardiology wards (lower proportion of elderly patients or with comorbidity). The lower proportion of women in cardiology services vs other services is also reflected in the literature20–26 (see references 43-46 of the supplementary data).

Some baseline differences between women vs men were observed in ETIFIC (table 1). No clinical trial or observational optimization study has evaluated the baseline characteristics of women.5–10 (see references 1-42 of the supplementary data). These were analyzed by 2 meta-analyses17,18 and 6 studies with other objectives.15,19–23 The HFrEF women generally had similar characteristics to those in ETIFIC (table 1): they were 1 to 4 years older,15,17,18,20–22 had a lower proportion of smokers,15,18,21 lower alcohol consumption,15 less frequently had atrial fibrillation,15,18,20,21 a lower 6-minute walk distance,22 worse quality of life,22,23 a higher proportion of NYHA III-IV,17,18,20,21 and minimal differences in LVEF (0.5-2%) vs men.15,17–19,21

Drug prescription in ETIFIC showed no significant differences, except for greater prescription of psychotropic drugs in women. A lower prescription of BB15,18 and ACEI in women vs men has been reported in the literature.15,20

No gender differences were found in the relative doses of ACEI/ARB (table 2). Equally, there were no significant differences between women and men during titration in SBP or eGFR (table 10 of the supplementary data), in symptomatic hypotension events (figure 2), or the mean relative dose of ACEI (table 2). However, there was a lower proportion of women with ≥ 50% the target dose of ACEI (table 2).

The multivariate analysis showed that the relative dose of ACEI at baseline and SBP were associated with the achieved doses of ACEI in both women and men (table 5). The association of achieved ACEI dose with the titrating professional, and the number of visits previously demonstrated in ETIFIC25 was also confirmed for women, as shown in table 3 and table 5 (see also table 5 of the supplementary data).

Women achieved significantly higher relative MRA doses at 4 months, associated with higher use of spironolactone and lower use of eplerenone (table 2). No gender differences were found in the multivariate analysis including all patients (table 5). However, there were no differences in the potassium level at baseline and 4 months (table 10 of the supplementary data) or in adverse events associated with potassium levels (figure 2), but potassium ≥ 5 mEq/L was associated with a lower dose only in men in the multivariate analysis (table 5).

No original ACEI, ARB and MRA trial has reported the prescription and mean relative dose achieved disaggregated by gender. Both women and men achieved lower ACEI/ARB doses in ETIFIC than those achieved by all patients in trials. This difference could be explained by the selected population, a higher SBP and a lower prescription of MRA and BB in these trials compared with ETIFIC (see references 21-28 of the supplementary data).

We found limited and heterogeneous information from some observational studies, 2 of them showing lower ACEI and ARB doses achieved by women6,10 but no other.21 In contrast, higher MRA doses have been reported in women.7

The drug prescription rate in the ETIFIC trial was high for both women and men, without significant differences. In addition, the joint prescription of 3 groups of drugs (BB, renin angiotensin system inhibitors and MRA) was 84% in women vs 78% in men, far higher than in trials and observational studies5–10 (see tables 7-9 and references 1-42,49 of the supplementary data).

Both women and men showed good self-care and adherence, although women showed slightly better results. Previous studies of adherence in HF have reported contradictory results in relation to gender (see table 10 and references 50, 51 of the supplementary data).

There were no differences between women and men in adverse events associated with titration (figure 2). Clinical trials and observational optimization studies did not report adverse events disagregated by sex5–10 (see references 1-42 of the supplementary data).

Differences in pharmacodynamics and pharmacokinetics have been described in women that could lead to higher blood levels at the same drug dose, lower tolerance of higher doses or beneficial effects with lower doses. Moreover, differences in renal function could lead to greater adverse events, suggesting that optimal doses in women should be lower.13,16,21 However, the ETIFIC trial showed no significant differences in women vs men in dosage and adverse events.

No gender differences were found in the mortality rate or cardiovascular admissions. Both were lower than in the literature, probably because the patients in this study had novo HF, received therapeutic optimization, close follow-up by the HF program, and showed good self-care and adherence.1,3

Although it did not reach statistical significance, probably due to the small number of events, it could be clinically relevant that there were no HF admissions in women, considering that there were no significant differences in cardiovascular mortality either (figure 3).

Three meta-analyses have shown the effectiveness of BB, ACEI and MRA in reducing mortality and admissions in women.17,18,27 The BB and MRA meta-analysis17,18 and the European Long-Term registry28 observed fewer serious adverse events in women.

A substudy of BIOSTAT21 observed a 30% reduction of death or hospitalization with 50% of the recommended doses of ACEI, ARB and 60% of BB, without finding a greater reduction with higher doses, although the authors reported that the optimal doses for women are unknown. No titration protocol, adverse events, number of visits or reasons for not reaching the target dose were described. Moreover, the BB and MRA prescription and BB doses achieved were significantly lower than in the ETIFIC trial.

A clear improvement in LVEF (14.86% vs 15.32%), NT-proBNP, NYHA, 6-minute walk test and quality of life, with no significant differences between women and men, was shown, except in NYHA class, where men had a better functional class at 6 months than women (table 6).

LVEF recovery has been associated with being female, nonischemic etiology, atrial fibrillation, adherence to BB and shorter duration of HF19 (see references 52-54 of the supplementary data).

Similar improvements in women vs men have also been described in quality of life and 6-minute walk test, associated with the use of BB, ACEI, ARB, and a close follow-up in clinical studies.22

A paradox previously reported in the literature15,20,29 was also observed in women who, despite being older, more symptomatic and having worse quality of life, had lower mortality and HF hospitalization. According to previous publications, this may be due to possible late diagnoses, less access to health care, socioeconomic and educational factors, and physicians’ misperception of women's symptoms and consequent undertreatment.

The ETIFIC clinical trial was not designed with the aim of evaluating gender differences in drug titration. Therefore, the sample in this post hoc analysis was not calculated for this purpose.