Use of HFNC in COVID-19 patients in non-ICU setting: Experience from a tertiary referral centre of north India and a systematic review of literature (preprint)

ABSTRACT

Introduction The rapid surge of cases and insufficient numbers of intensive care unit (ICU) beds have forced hospitals to utilise their general wards for administration of non-invasive respiratory support including HFNC(High Flow Nasal Cannula) in severe COVID-19. However, there is a dearth of data on the success of such advanced levels of care outside the ICU setting. Therefore, we conducted an observational study at our centre, and systematically reviewed the literature, to assess the success of HFNC in managing severe COVID-19 cases outside the ICU. Methods A retrospective cohort study was conducted in a tertiary referral centre where records of all adult COVID-19 patients (over 18 years) requiring HFNC support were between September and December 2020 were analysed. HFNC support was adjusted to target SpO2 over 90% and respiratory rate less than 30 per min. The clinical, demographic, laboratory, and treatment details of these patients were retrieved from the medical records and entered in predesigned proforma. Outcome parameters included duration of oxygen during hospital stay, duration of HFNC therapy, length of hospital stay and death or discharge. HFNC success was denoted when a patient did not require escalation of therapy to NIV or invasive mechanical ventilation, or shifting to the ICU, and was eventually discharged from the hospital without oxygen therapy; otherwise, the outcome was denoted as HFNC failure. Systematic review was also performed on the available literature on the experience with HFNC in COVID-19 patients outside of ICU settings using the MEDLINE, Web of Science and Embase databases. Statistical analyses were performed with the use of STATA software, version 12, OpenMeta[Analyst], and visualization of the risk of bias plot using robvis. Results Thirty-one patients receiving HFNC in the ward setting, had a median age of 62 (50 - 69) years including 24 (77%) males. Twenty-one (68%) patients successfully tolerated HFNC and were subsequently discharged from the wards, while 10 (32%) patients had to be shifted to ICU for non-invasive or invasive ventilation, implying HFNC failure. Patients with HFNC failure had higher median D-dimer values at baseline (2.2 mcg/ml vs 0.6 mcg/ml, p=0.001) and lower initial SpO2 on room air at admission (70% vs 80%, p=0.026) as compared to those in whom HFNC was successful .A cut-off value of 1.7 mg/L carried a high specificity (90.5%) and moderate sensitivity (80%) for the occurrence of HFNC failure. Radiographic severity scoring as per the BRIXIA score was comparable in both the groups(11 vs 10.5 out of 18, p=0.78 ). After screening 98 articles, total of seven studies were included for synthesis in the systematic review with a total of 820 patients, with mean age of the studies ranging from 44 to 83 years and including 62% males. After excluding 2 studies from the analysis, the pooled rates of HFNC failure were 36.3% (95% CI 31.1% - 41.5%) with no significant heterogeneity (I2 =0%, p=0.55). Conclusions Our study demonstrated successful outcomes with use of HFNC in an outside of ICU setting among two-thirds of patients with severe COVID-19 pneumonia. Lower room air SpO2 and higher D-dimer levels at presentation were associated with failure of HFNC therapy leading to ICU transfer for endotracheal intubation or death. Also, the results from the systematic review demonstrated similar rates of successful outcomes concluding that HFNC is a viable option with failure rates similar to those of ICU settings in such patients.