Statistics from Altmetric.com
We have read with great interest the editorial by Beattie and Johnson,1 discussing the role of subcutaneous furosemide (SCF) as an off-label palliative symptomatic treatment feasible for patients with decompensated advanced heart failure (DAHF), whose preferences led to being attended to in their homes, thus avoiding inappropriate hospitalisations, as demonstrated by Zacharias and Johnson in their recently published study.2
In this sense, we have performed an observational retrospective study with the aims to determine the effects of SCF dose amount on hospital length of stay, and time to symptom improvement, among patients with DAHF. We also compared the results with a cohort of patients treated with intravenous furosemide on wards.
Between January 2008 and January 2012, all episodes of DAHF consecutively attended to at home were recruited. Inclusion criteria were: older than 18 years, heart failure, New York Heart Association (NYHA) class III or IV symptoms, treated at home with SCF by a Southern Spanish university hospital at-home unit. Analysed variables were: number of hospital admissions before and after inclusion, initial dose, total dose, mean daily dose, time to symptom recovery (dyspnoea remission or oedema-significant improvement) and length of hospital stay. Qualitative variables were described as numbers or percentages, quantitative variables were described as medians (p25–p75), except for age (mean±SD). Qualitative variables were divided into quartiles. Then, χ2 analyses and non-parametric tests (Mann–Whitney U test for median comparison) were performed, as although several variables had a normal distribution, the small size made these analyses more appropriate.
We compared 34 consecutive episodes in 17 patients suffering from DAHF and treated with SCF, with 63 episodes in 27 patients treated with intravenous furosemide (IVF), which were included for analysis. Among all the patients, 64.7% vs 58.7% were women, their median age was 84 years (76–88.5) vs 83 (77–88), and their median Barthel index 40 (20–55) vs 20 (0–50) in the SCF/IVF subgroups (p value>0.05 for variable comparison), respectively. The median number of hospitalisations and emergency department consults in the previous year, treatment episodes, complex chronic conditions, Charlson index and drugs/patient differed slightly, but not significantly between both subgroups. All patients were polypathological and highly polymedicated, receiving 12 (5.5–15.75) drugs/patient.
A total SCF/IVF dose of 720 (480–1125) vs 720 (320–1200) mg/episode (p value=0.666), and daily dose of 160 (99–250) vs 103.75 (80–171.43) mg/day (p value=0.005) was administered. The initial dose of furosemide was 160 (120–250) mg vs 80 (60–160) (p value=0.000). The median length of stay per episode was 12.5 (8–16) vs 7 (5–12) days (p value=0.001). After discharge, the number of hospital admissions was 1 (2) vs 0 (1), and number of emergency department consults was 1 (2) vs 0 (1) (both comparisons p value>0.05). There were five adverse events leading to stoppage of SCF: three haematomas, one severe local skin infection, with favourable outcome under oral antibiotics, and one drip removal.
When the initial SCF/IVF furosemide dose was divided into quartiles, 0/27 patients received a dose lower than 60 mg/day, 14/16 patients received a dose between 60 and 120 mg/day, 8/15 patients 121–240 mg/day, and 12/5 a dose higher than 240 mg/day. By contrast with the SCF subgroup, a majority of patients treated with IVF (42.86%) received a dose <60 mg/day. Higher initial doses showed a relation with shorter days of stay (figure 1). This dose-effect was less evident among patients treated with IVF. Instead, higher total dose/episode was associated with longer lengths of stay.
When assessing different outcome variables, only median number of emergency consults (1 year after hospital discharge) differed significantly between subgroups (p=0.003). There were 90% of patients (78.6% vs 96.2%, p value=0.115) who died during the period of study, accounting for a median survival time (SCF vs IVF) of 26.51 (10.8–253.31) vs 17.67 (8.84–292–58) (p value=0.937) days. Also, survival analyses showed an association between higher survival times and Barthel index >40 (p=0.01), or receiving β-blocker treatment (p=0.03).
Most of the families of patients treated with SCF expressed being satisfied, or very satisfied, with the procedure (15/17), and none of them had any knowledge of the possibility of being treated at home.
In conclusion, SCF is a feasible off-label palliative treatment for patients with DAHF who want to be attended to in their homes, thus avoiding unnecessary hospitalisations which would worsen comfort and could be disappointing to their families. Clinical trials are needed to establish efficacy and safety of SCF, and allow the subcutaneous route of administration approval for this drug.
The patients with DAHF treated with SCF in our area were mostly older women, highly disabled, polymedicated polypathological patients. Higher initial doses of SCF could be beneficial in terms of shorter lengths of stay.
Collaborators José Molina, Verónica Alfaro-Lara, Ricardo Parra Alcaraz, Salvador Sobra Calderón, Julián Garrido Nieto.
Contributors All authors have contributed to the manuscript.
Competing interests None.
Ethics approval Ethical committee of the University Hospital Virgen del Rocío.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.