In this cohort of patients with chronic HFpEF, we showed that subjects with lower FT3/FT4 ratio were more likely to have a higher clinical severity of HF and higher prevalence of obesity, diabetes and atrial fibrillation. Furthermore, lower levels FT3/FT4 ratio were associated with higher body fat, lower LVEF and higher PASP. Lastly, a lower FT3/FT4 ratio was associated with a higher risk of urgent HF visit, HF hospitalization or cardiovascular mortality.
As previously stated, the ratio of FT3/FT4 has been suggested as an indirect index of peripheral conversion of T4 into T3 [8]. Deiodinases’ activity appears to be disturbed in several acute and chronic illnesses, including HF [4]. Although this decrease in T3 levels is seen as an adaptative response to reduce energy expenditure, questions have been raised regarding the role of this mechanism in HF [6, 11]. Persistently low T3 levels negatively influence myocardial function and structure, impairing cardiac remodeling in a similar way as in HF progression [6, 13]. In HFpEF, lower T3 levels were previously associated with increased symptom burden, higher BNP levels and worse diastolic function [14].
Patients with lower FT3/FT4 ratio had a higher prevalence of metabolic and cardiac comorbidities. In fact, these comorbidities may further impair deiodinase activity. On the other hand, lower levels of active TH can also worsen or predispose to these comorbidities. Hypothyroidism contributes to weight gain, due to a decrease in resting energy expenditure [15], and may increase the vulnerability to arrhythmias such as atrial fibrillation [16].
Lower FT3/FT4 ratio and FT3 levels were associated with lower LVEF and higher deceleration time, which suggests worsening of both systolic and diastolic functions with lower levels of active TH. On the other hand, PASP was found to be higher with higher FT4 and lower FT3/FT4 ratio levels. Even though hyperthyroidism is associated with pulmonary hypertension [1], we did not find association between FT3 and PASP. Pulmonary hypertension is common and a marker of dismal prognosis in patients with HFpEF [17]. It is then plausible that patients with higher levels of PASP have further impairment in deiodinase activity, hence, lower levels of active TH.
Several studies have shown that both lower T3 and higher T4 levels were independent predictors of poor outcomes in HF [18, 19]. In a recent retrospective study, FT3/FT4 ratio was shown to predict cardiovascular and all-cause mortality in HF patients, with a greater predictive value in HFpEF [20]. These studies increase the robustness of our analysis and underline the potential use of FT3/FT4 ratio as a prognostic metric in this subtype of HF.
Some limitations of our analysis are noteworthy. This is an observational study, which limits our ability to comment on causality and generalize our results. Our study had limited statistical power to detect associations of TH and FT3/FT4 ratio with clinical parameters and outcomes; a larger sample size might have shown more consistent associations. Despite our effort to adjust for relevant confounding factors, we cannot exclude that residual confounding may have influenced our results. We only evaluated thyroid hormones in a single moment, not taking into account potential variations of these parameters.
In conclusion, in patients with HFpEF, a lower FT3/FT4 ratio was associated with a worse clinical status, cardiac function and long-term prognosis. A decreased conversion of FT4 to FT3 might be a mechanism associated with HFpEF progression and decompensation. More studies addressing the thyroid-heart interaction in HFpEF are needed.