LAUSR

BACKGROUND & AIMS Nutritional status is an important prognostic factor in patients with heart failure (HF). In a pilot study we previously observed that the Mini Nutritional Assessment Short Form tool (MNA-SF) was the best approach for the screening of nutritional status in HF outpatients over other screening tools. The current study aimed to determine whether the MNA-SF has prognostic value in outpatients with HF and whether the impact of malnutrition differs depending on left ventricular ejection fraction (LVEF). METHODS Prospective study performed in outpatients attending a HF clinic at a university hospital. All subjects completed the MNA-SF at study entry. The primary endpoint was all-cause mortality. Secondary end-points were the number of recurrent HF-related hospitalizations and the composite end-point of all-cause death or HF-related hospitalizations. Patients with malnutrition and at risk of malnutrition were merged and considered as having abnormal nutritional status for statistical analysis. RESULTS From October 2016 to November 2017, 555 patients were included (age 69 ± 11.5 years, 71% male, LVEF 44.6 ± 13.2). Abnormal nutritional status was identified in 103 (18.6%) subjects. HF patients with preserved LVEF had a higher proportion of abnormal nutritional status (23%) than patients with HF and mid-range LVEF (HFmrEF) (16.4%) or those with HF with reduced LVEF (HFrEF) (15.9%.). During a mean follow-up of 23.8 ± 6.6 months, 99 patients died (17.8%), 74 were hospitalized due to HF (13.3%) and the composite end-point was observed in 181 (32.6%). In the univariate analysis, abnormal nutritional status was significantly associated with all-cause mortality (p = 0.02) and the composite end-point (p = 0.02) in the total cohort. However, in the multivariate analysis including age, sex, NYHA functional class, BMI, ischemic aetiology, diabetes, hypertension and HF duration, abnormal nutritional status remained significantly associated with all-cause mortality (HR 3.32 [95%CI 1.47-7.52], p = 0.004), and the composite end-point (HR 2.53 [95%CI 1.30-4.94], p = 0.006) only in HFmrEF patients. Patients with abnormal nutritional status suffered double the crude number of recurrent HF-related hospitalizations (16.4 vs. 8.4 per 100 patients-years, p < 0.001). CONCLUSIONS The implementation of MNA-SF as a routine screening tool allowed the detection of abnormal nutritional status in almost one out of five ambulatory HF patients. Nutritional status assessed by the MNA-SF was an independent predictor of all-cause death and the composite end-point of all-cause death or HF-related hospitalization in outpatients with HFmrEF. Furthermore, abnormal nutritional status was significantly related to recurrent hospitalizations across the HF spectrum.