Predictors of success of high-flow nasal cannula in the treatment of acute hypoxemic respiratory failure

R Artacho Ruiz; B Artacho Jurado; F Caballero Güeto; A Cano Yuste; I Durbán García; F García Delgado; J A Guzmán Pérez; M López Obispo; I Quero Del Río; F Rivera Espinar; E Del Campo Molina

doi:10.1016/j.medin.2019.07.012

Predictors of success of high-flow nasal cannula in the treatment of acute hypoxemic respiratory failure

Med Intensiva (Engl Ed). 2021 Mar;45(2):80-87. doi: 10.1016/j.medin.2019.07.012. Epub 2019 Aug 24.

[Article in English, Spanish]

Authors

Affiliations

¹ Servicio de Cuidados Críticos y Urgencias, Hospital de Montilla, Montilla, Córdoba, España; Servicio de Medicina Intensiva, Hospital Cruz Roja, Córdoba, España. Electronic address: rafael_artacho@hotmail.com.
² Emergency Assessment Unit, John Radcliffe, Oxford University Hospital, Oxford, Reino Unido.
³ Servicio de Cuidados Críticos y Urgencias, Hospital de Montilla, Montilla, Córdoba, España; Servicio de Medicina Intensiva, Hospital Cruz Roja, Córdoba, España.
⁴ Servicio de Urgencias, Hospital Quirón-Salud, Córdoba, España.
⁵ Servicio de Cuidados Críticos y Urgencias, Hospital de Montilla, Montilla, Córdoba, España.
⁶ Servicio de Cuidados Críticos y Urgencias, Hospital de Montilla, Montilla, Córdoba, España; Dirección Médica, Hospital Cruz Roja, Córdoba, España.
⁷ Servicio de Medicina Intensiva, Hospital Quirón-Salud, Córdoba, España.

PMID: 31455561
DOI: 10.1016/j.medin.2019.07.012

Abstract

Background: High-flow nasal cannula (HFNC) therapy is used in the treatment of acute respiratory failure (ARF) and is both safe and effective in reversing hypoxemia. In order to minimize mortality and clinical complications associated to this practice, a series of tools must be developed to allow early detection of failure. The present study was carried out to: (i)examine the impact of respiratory rate (RR), peripheral oxygen saturation (SpO₂), ROX index (ROXI=[SpO₂/FiO₂]/RR) and oxygen inspired fraction (FiO₂) on the success of HFNC in patients with hypoxemic ARF; and (ii)analyze the length of stay and mortality in the ICU, and the need for mechanical ventilation (MV).

Methods: A retrospective study was carried out in the medical-surgical ICU of Hospital de Montilla (Córdoba, Spain). Patients diagnosed with hypoxemic ARF and treated with HFNC from January 2016 to January 2018 were included.

Results: Out of 27 patients diagnosed with ARF, 19 (70.37%) had hypoxemic ARF. Fifteen of them (78.95%) responded satisfactorily to HFNC, while four (21.05%) failed. After two hours of treatment, RR proved to be the best predictor of success (area under the ROC curve [AUROC] 0.858; 95%CI: 0.63-1.05; P=.035). For this parameter, the optimal cutoff point was 29rpm (sensitivity 75%, specificity 87%). After 8hours of treatment, FiO₂ and ROXI were reliable predictors of success (FiO₂: AUROC 0.95; 95%CI: 0.85-1.04; P=.007 and ROXI: AUROC 0.967; 95%CI: 0.886-1.047; P=.005). In the case of FiO₂ the optimal cutoff point was 0.59 (sensitivity 75%, specificity 93%), while the best cutoff point for ROXI was 5.98 (sensitivity 100%, specificity 75%). Using a Cox regression model, we found RR<29rpm after two hours of treatment, and FiO₂<0.59 and ROXI>5.98 after 8hours of treatment, to be associated with a lesser risk of MV (RR: HR 0.103; 95%CI: 0.11-0.99; P=.05; FiO₂: HR 0.053; 95%CI: 0.005-0.52; P=.012; and ROXI: HR 0.077; 95%CI: 0.008-0.755; P=.028, respectively).

Conclusions: RR after two hours of treatment, and FiO₂ and ROXI after 8hours of treatment, were the best predictors of success of HFNC. RR<29rpm, FiO₂<0.59 and ROXI>5.98 were associated with a lesser risk of MV.

Keywords: Acute respiratory failure; Cánula nasal de alto flujo; Fallo respiratorio agudo; High-flow nasal cannula; Hipoxemia; Hypoxemia; Predictores de éxito; Predictors of success; ROX index; Índice ROX.