We read with great interest the article by Segrelles Calvo G et al., which provided an important assessment of the effectiveness of non-invasive mechanical ventilation (NIMV) in elderly patients.1 The study demonstrates that some of the limiting factors of such effectiveness are the comorbidity and greater tendency to readmit elderly patients over time, influencing decision-making and prognosis of NIMV. The success of applying NIMV in the Respiratory Monitoring Unit (RMU) in elderly patients highlights the need to promote the creation and consolidation of these specific healthcare areas for elderly patients who require NIMV outside the Intensive Care Unit (ICU). We believe that there are relevant issues in this paper:

• 1)

  With respect to the age factor, the evidence available in the literature on the benefits of NIMV shows that age does not imply a poorer response. In the study by Scarpazza et al.,2 which mainly included patients with Chronic Obstructive Pulmonary Disease (COPD) in acute respiratory failure (ARF), 54 patients of a cohort of 62 patients aged over 75 years responded to NIMV; thus the NIMV failure rate (need for orotracheal intubation (OTI) or mortality) was 12.9% of patients and the survival one year after discharge was 70%. In the study by Nava et al.,3 41 patients aged over 75 years were randomised to receive NIMV; 38 patients responded successfully and three (i.e. 7.3%) required OTI, one of whom died. In the series by Segrelles Calvo et al.1 fewer than 3.5% of patients required transfer to ICU and OTI. In this series, the mortality in the group over 75 years on admission was 21.4%, slightly higher than that found in previous studies.2,3 The do-not-intubate order could have influenced their results. The article states that 65% of patients were considered as do-not-intubate. It would be interesting to know more precisely how the order was established, and if patients were informed in earlier stages or after the first hospital admission. A higher frequency of heart failure as a cause of admission may also have influenced their results, in contrast to other recent studies which have mostly included patients with COPD or restrictive disorders. Heart failure was the major cause of mortality in their study. It would be interesting to know the separate mortality rate in patients with exacerbation of COPD in their series, and to analyse whether there are other factors not considered that may account for this higher mortality.

• 2)

  Another aspect that we consider important is the health care organisation of NIMV in readmissions. The study by Segrelles Calvo et al. is one of the first to be carried out in a RMU in elderly patients, and it would be interesting to know whether the same protocol of care is also followed in readmissions in these same specific health care areas of the hospital. Aspects such as stabilisation of these patients in specific areas of the hospital and the low rate of transfer to ICU are relevant. In this respect, the authors do not explain whether patients who are readmitted are treated equally effectively in the RMU, or whether its use would be limited in subsequent admissions. It would be useful to know how the readmissions arose and where we draw the line with this approach.

• 3)

  With respect to the higher incidence of some complications in the cohort, such as heart or renal failure, the study does not suggest whether there were any limitations in the treatment of these complications.

• 4)

  It would be interesting to learn the rate of readmissions of patients discharged with home mechanical ventilation (HMV), comparing it with those who were discharged without NIMV. In the case of patients with chronic respiratory failure secondary to a restrictive disease who require HMV, some studies have been published showing that age does not influence the results in terms of 5-year survival, and that HMV reduces the number of readmissions and increases survival.4 The implementation of long-term HMV after the first admission in patients in the series could reduce the tendency to readmission at six months.

• 5)

  Finally, it would be desirable to know the cost variables of NIMV in this population for the hospital system.5,6 We believe that it would be interesting to have further evidence on the effectiveness of NIMV in elderly patients, especially in stages after the first hospital admission.