*3.1. Outcomes*

Overall, starting from the 24-months evaluation, the outcomes of women were more favourable compared to men (Figure 2). During a median follow-up of 149 (IQR 90–232) months, 189 patients (31%) experienced the main outcome measure (44 males with LVRR, 35%; 105 males without LVRR, 34%; 10 females with LVRR, 18%; and 30 females without LVRR, 27%; global log-rank *p* = 0.03) and 128 patients (21%) the secondary outcome measure (36 males with LVRR, 29%; 76 males without LVRR, 24%; 6 females with LVRR, 11%; 24 females without LVRR, 22%; global *p* = 0.06). The cumulative incidence at 10 years of follow-up of specific components of the outcome measure is reported in Table 2. Women experiencing LVRR had the lowest incidence of all-cause mortality/HTx/VAD at 10 years of follow-up compared to the other groups, with an absolute risk reduction of 12% and a relative risk reduction of 71% of the main outcome measure compared to men with LVRR (*p* = 0.04). Interestingly, women without LVRR at 24 months showed a similar incidence of adverse outcomes as males (Figure 2). Noteworthy, the cumulative incidence of arrhythmic events followed the same trend, being lower in women with LVRR than in the other subgroups (*p* = 0.06) with an absolute risk reduction of 6% and a relative risk reduction of 60% of the arrhythmic outcome measure at 10 years of follow-up compared to men with LVRR (*p* = 0.02) (Figure 2).


**Table 1.** Characteristics of the Population at 24-month evaluation.

\* Gender specific volumes (LVEDV/BSA): Normal volumes Females: < 61 mL/m2. Males: < 74 mL/m2; Mild dilation Females: 62–70 mL/m2. Males: 75–89 mL/m2; Moderate Dilation Females: 71–80 mL/m2. Males: 90–100 mL/m2; Severe Dilation Females: > 80 mL/m2. Males: > 100 mL/m2. [10] Legend: ACE-I: Angiotensin Converting Enzyme-Inhibitors; ARBs: Angiotensin Receptor Blockers; BSA: Body Surface Area; CRT: Cardiac Resynchronization Therapy; ICD: Implantable Cardioverter Defibrillator; LBBB: Left Bundle Branch Block; LVEDDI: Left Ventricular End Diastolic Diameter Indexed; LVEDVI: Left Ventricular End Diastolic Volume Indexed; LVEF: Left Ventricular Ejection Fraction; MRA: Mineralocorticoid Receptor Antagonists; NYHA: New York Heart Association; RFP: Restrictive filling pattern; RV: Right ventricular; SBP: Systolic Blood Pressure.

**Figure 2.** Kaplan-Meier curves for the incidence of All-cause mortality/HTx/VAD (Left Panel) and cumulative incidence function for SCD/MVA (Right Panel) according to LVRR and sex. Legend. HTx: Heart Transplantation; LVRR: Left Ventricular Reverse Remodelling; VAD: Ventricular Assist Device. MVA: Major Ventricular Arrhythmias; SCD Sudden Cardiac Death.

**Table 2.** Cumulative incidence of events at 10 years of follow-up (starting from the 24 months evaluation) according to sex and LVRR.


Legend: CV: cardiovascular; HTx: Heart Transplantation; MVA: major ventricular arrhythmias; SCD: Sudden Cardiac Death; VAD: Ventricular assist device.

#### *3.2. Multi-State Model Analysis*

After adjustment for the different variables at the 24 months evaluation (i.e., Age, NYHA class, Sinus Rhythm, Severe LV Dilation, LVEF, Restrictive Filling Pattern, Right Ventricular Dysfunction, and medical therapy) male sex emerged as an independent risk factor of adverse outcomes (HR 1.86, 95% CI 1.07–3.82, *p* = 0.02). To further investigate the relationship between sex and the prognostic role of LVRR over time, a multistate model was built considering LVRR as an intermediate state, with the follow up starting from the baseline. The multi-state model highlights how the occurrence of LVRR over time was strongly associated with better outcomes (HR 0.01, 95% CI 0.001–0.04, *p* < 0.001) and male sex emerged as a strong prognostic factor in patients who experienced LVRR (HR 2.81, 95% CI 1.03–7.64, *p* = 0.04), whereas the impact of sex was diluted in patients without LVRR. Indeed, men with LVRR had a significantly higher probability of experiencing adverse outcomes over time (*p* = 0.04), whereas sex differences were blunted in those without LVRR over time (*p* = 0.52) (Figure 3).

**Figure 3.** Adjusted Cumulative Incidence estimated from the multi-state model of All-cause mortality/ HTx/VAD according to sex in patients with LVRR and without LVRR: Left Ventricular Re. Legend. HTx: Heart Transplantation; VAD: Ventricular Assist Device.

#### **4. Discussion**

Female sex has emerged as an important outcome modifier in different cardiovascular scenarios. In patients with DCM, previous reports highlighted the protective role of female sex towards adverse outcomes over the long-term follow-up [6–8,16,17]. However, besides speculative hypotheses and observational analyses, there is no evidence so far investigating the possible mechanisms underlying this prognostic difference between sexes. Although one possible explanation might dwell in a different sex-specific response to medical treatment and, therefore, a different rate of LVRR with subsequent prognostic implications [5], evidence of that is still unavailable. To date, this is the first study addressing the interaction between sex and LVRR as potential outcome modifier in a large population of well-characterized DCM patients with available follow-up data.

The LVRR is a complex process that usually starts with the introduction of medical therapy and takes up to 24 months to complete [1]. Although several factors have been associated with the occurrence of LVRR over time [5], so far, little is known about the influence of sex on the rate of LVRR. Similarly to previous reports [5], in our population approximately 30% of patients experienced LVRR at 24 months of follow-up and the occurrence of LVRR was strongly associated with better prognosis (HR 0.01, 95% CI 0.001–0.04, *p* < 0.001). Interestingly and unexpectedly, the rate of LVRR was comparable between man and women (Figure 1). Noteworthy, among patients experiencing LVRR, females had an overall better prognosis compared to males during a very long-term follow-up; conversely a comparable prognosis between males and females without LVRR was found. To evaluate the prognostic impact of sex over time, we used a multi-state model considering the occurrence of LVRR as an intermediate state. The LVRR was confirmed as a long-term prognostic predictor and the female sex was strongly associated with better outcomes predominantly in patients experiencing LVRR whereas its prognostic implications were diluted in those not experiencing LVRR (Figure 3).

Despite the optimization of medical and device therapy, at 24-months revaluation women still showed a more advanced phenotype of the disease, characterized by larger LV diameters and a higher incidence of moderate to severe sex-specific dilation, which might partially justify the comparable outcomes in patients without LVRR (Table 1). Despite the more advanced phenotype of DCM observed at 24-month revaluation, women had overall better long-term outcomes than men. This was probably driven by the excellent long-term outcome showed by women experiencing LVRR, compared to

either men with LVRR or patients without LVRR regardless of their sex (Figure 2). Indeed, women experiencing LVRR showed a 71% relative risk reduction of experiencing a composite adverse outcome of all-cause mortality/HTx/VAD compared to men with LVRR. Similar trends were found for arrhythmic events (Figure 2).

In the era of precision medicine, these findings might have important clinical implications, opening new possible scenarios in patients' management. In fact, different treatment strategies might be employed between sexes experiencing LVRR or not over time. Our results highlight an independent prognostic role of female sex, especially after the LVRR is achieved, and opens up novel scenarios to investigate the mechanism underlying this prognostic advantage of women besides response to treatment.
