Malnutrition and Heart Failure: Comments

To the Editor,

.

We read with interest the “Letter to the Editor” from Sánchez-Muñoz et al.¹ concerning the article recently published by our group in Revista Española de Cardiología.² We appreciate their comments, always a welcome stimulus, and it is a pleasure for us to observe the interest aroused by the issue being addressed. While we are presently impressed by the degree to which heart failure (HF) influences nutritional status (NS), if we change the order of these two terms, the responses are paltry. Modestly, but convincingly, our study constitutes a milestone in this field in that it demonstrates the additional difficulties with which the malnourished patient with HF must cope when facing his or her disease.

We would like to comment on some of the points made in their letter.

With respect to the reference to the Short Form Mini Nutritional Assessment (MNA) as the version recommended for clinical use, it would be necessary to specify the source of this recommendation and the criteria on which it has been based. We feel that it would be more appropriate to say that the short form of the questionnaire has been validated for use under the same terms as the original form and, thus, that both can be employed with similar guarantees in clinical practice.³ The MNA had never been used for the assessment of NS in an extensive, heterogeneous cohort of HF patients, and the influence of this status on the prognosis of these patients had never been evaluated. Thus, the use of the original form was obligatory from the scientific point of view. The results of our work, pending further studies, indicate that, in patients with HF, the NS should be evaluated using the original form of the MNA. We are aware of the bias derived from the fact that this questionnaire cannot be used to assess all patients, but this does not make it less interesting as a tool for identifying the high-risk patient.

With respect to the age-MNA controversy, we know that the questionnaire was designed to evaluate NS in the elderly population.⁴ Nevertheless, we pointed out that this consideration was taken into account when patients in their sixties were included. The high sensitivity and specificity of the questionnaire for the detection of malnutrition have encouraged health professionals to utilize it in a younger population, and its use has now been extended to 50-year-old patients with good results.³ In our study, only 5 patients (2.4%) were under the age of 50 years, and the results we reported support the use of this questionnaire in patients with HF regardless of their age: according to the MNA, malnutrition was an independent prognostic factor, independent of age as well.²

With regard to the variables introduced into the multivariate analysis, we have two comments to make. The use of the Charlson index in the multivariate assessment was not very attractive, first because the index proved to be similar in the 3 groups of patients, and second because this index encompasses several factors which, alone, influence the prognosis in the HF patient. Thus, the inclusion of the Charlson index as a variable would have impeded their inclusion separately, for the sake of avoiding the use of variables twice in the same model, and employed individually they provide more accurate and practical information. That said, we remind readers that, of the variables that comprise the Charlson index, those for which differences between groups were detected were included in the multivariate analysis. With respect to the functional class variable, we wish to point out that it constitutes a relevant prognostic factor in patients with stable HF, and our study was carried out in patients with decompensated HF. Thus, nearly all the patients were in functional class IV (this datum was deliberately omitted from the table that shows the patient characteristics because it was obvious), a fact that justifies its not being included as a variable in the Cox multivariate analysis.

Finally, we are of the opinion that a relationship between nutritional intervention and improved survival in malnourished patients with HF must be considered a working hypothesis, rather than a certainty that can be extracted from the results obtained in nutritional interventions carried out in other populations.³ We should not forget the idiosyncrasy of the disease in question and that, as we have observed, malnourished HF patients have a very high mortality rate. Only a positive outcome obtained in clinical trials designed for the purpose would lead us to include this intervention as part of an integrated approach to patients with HF.

Corresponding author: [e-mail]