Forty-two patients were included from September 2 to December 9, 2020. The median (interquartile) age of the patients was 72 (65–78) years. A total of 69% were male. Fifty-two percent of the patients were treated for hypertension. One patient had to be intubated by the medical team in charge of the transfer before the transfer and thus was excluded from the analysis. This was the only patient excluded from the analysis. Two patients had to be intubated during the transfer by Angers SMUR. For these two patients, the respiratory rate (RR) and oxygen saturation (SaO2) values before intubation were entered as “final” respiratory requirements. The overall analysis of patients showed no significant variation in the respiratory requirements of the patients during the transfer (Fig. 1). At the beginning of the transfer, the RR and SaO2 medians were 30/min (26–32) and 92% (90–95), respectively. At intensive care unit admission, the RR and SaO2 were measured at 28/min (22–32) and 93% (90–95), respectively. Ventilation settings were not changed significantly during the transfer period: At the beginning of the transfer, the flow and FiO2 medians were 50 L/min (50–60) and 75% (53–100), respectively. At intensive care unit admission, debit and FiO2 were 50 L/min (50–60) and 80% (50–100), respectively. Overall, 52% of patients were intubated during their stay in the ICU, including three patients intubated before or during transfer. The median Rox score at the start of the transfer was 3.3 in patients intubated and 6.8 in non-intubated patients, after exclusion of one patient with a treatment limitation decision. No patient had to be urgently intubated upon arrival in the ICU. Survival at 1 month after admission to the ICU was 74%.
Discussion
There are very few descriptions in the literature of medical transfer of adult COVID-19 patients needing HFNC, so our study population cannot be compared to most studies. The respiratory requirements of the patients corresponded to the indications for HFNC. However, the respiratory status of some patients was particularly severe. This may explain the high proportion of intubated patients during their ICU stay. For these patients, other causes of decompensation could have compounded the patients’ respiratory efforts, such as pulmonary embolism, heart failure, and superinfection. But in our cohort, ICU follow-up ruled out these causes of initial decompensation. When HFNC is started, the objectives are a decrease in the respiratory rate and an increase in oxygen saturation. Then, during mobilizations and ambulance transport, the objective is to maintain the stability of these respiratory requirements. Between the patient’s room and the intensive care ambulance, during which the power supply to the heater humidifier is interrupted, humidification and heating of the air-oxygen mixture are no longer guaranteed. Did this lack of warming and humidification contribute to the sudden deterioration of the respiratory state of the two patients who were intubated? This led us to recommend that the need for a prolonged interruption of power supply would contraindicate a transfer with HFNC, unless the transfer is after a prior weaning trial. Interhospital transfer with HFNC is very high-risk, and intubation remains indicated in the most unstable patients.
However, 48% of patients benefited from HFNC and were thus able to avoid intubation during their transfer and ICU stay; for these patients, intubation would probably have been indicated in the absence of available HFNC techniques. Ideally, we would have liked to compare the survival of two cohorts of patients. Are these patients (who are transferred on HFNC) having overall better survival rates or approximately the same as their intubated cohorts? The main obstacle to using HFNC in COVID-19 patients was the fear of an increased risk of contamination for personnel due to the aerosolization produced when using HFNC. However, a few trials did not appear to show that there is an increased risk of contamination with HFNC compared with other modes of oxygenation, including high concentration masks [7,8,9]. Our small trial needs to be expanded to a larger patient population for more concrete evidence of HFNC benefits in interhospital transfers. Finally, alternatives to HFNC are possible, such as non-invasive ventilation or continuous positive airway pressure.