What we already know about non-invasive respiratory support of immunocompromised patients
Acute respiratory failure (ARF) is one of the main complications in immunocompromised patients since it occurs in up to 50% of hematologic patients and in up to 15% of patients with solid tumors or organ transplants. In these patients, ARF represents one of the main indications for intensive care units (ICUs) admission, and the mortality is high (up to 50%), especially in those who require invasive mechanical ventilation [1, 2].  

2017 ERS clinical practice guidelines about non-invasive ventilation (NIV) suggest early non-invasive strategies (including non-invasive ventilation and continuous positive airways pression) for immunocompromised patients with ARF with a conditional recommendation (moderate quality of evidence), since the use of NIV decreases mortality, the need for intubation (IOT) and the rates of hospital-acquired pneumonia in this population [3]. A network metanalysis published in 2020 also confirmes that the use of non-invasive oxygenation strategies compared with standard oxygen therapy is associated with a lower risk of death [4].

High-flow nasal oxygen (HFNO), also called high flow nasal cannulae (HFNC) is a non-invasive form of respiratory support that provides several physiological effects in ARF: HFNO can generate a more stable FIO2 than conventional oxygen therapy (COT), resulting in a better oxygenation; it provides alveolar recruitment through the generation of a high flow and a low level of positive end expiratory pressure (PEEP); it prevents airway dryness and mucociliary disfunction via ensuring heated humidification; it assures a continuous washout of CO2 from the anatomical dead space [5]. The ease of setting and the increased patient’s tolerance favors the use of HFNO in clinical practice worldwide.

In 2021 ERS published clinical practice guidelines about high-flow nasal cannula in ARF [6]. One of the three subgroups addressed by each PICO question was immunocompromised patients. The ERS task force suggests the use of HFNO over COT in hypoxemic respiratory failure as it reduces intubation and escalation to NIV even if it doesn’t produce any effect on mortality. This recommendation is also true for immunocompromised patients, even if the magnitude of the effect on intubation and escalation to NIV may be lower. Comparing HFNO with NIV in hypoxemic respiratory failure, ERS task force suggests the use of HFNO over NIV since HFNO could reduce 28-days mortality, even with a very low certainty of evidence. HFNO also reduces intubation and increases patient comfort. In immunocompromised patients, ERS task force reports possible greater benefits of HFNC over NIV, but these results remain imprecise, and insufficient for a strong recommendation, especially considering that only one trial compares NIV to HFNO in immunocompromised patients with ARF.

Introduction
FLORALI-IM is a multicenter, open-label, randomized clinical trial conducted in 28 French and one Italian ICUs.
The study investigates the effects of HFNO alone compared with HFNO alternating with NIV on mortality at day 28 in immunocompromised patients with ARF.

Methods
Inclusion criteria were: immunosuppression (defined as hematological malignancy active or remitting within the last 5 years, allogeneic stem cell transplantation £ 5 years, active solid cancer, leucopenia < 1 G/L or neutropenia £ 0.5 G/L induced by chemotherapy, solid organ transplantation, acquired immunodeficiency syndrome, systemic steroids ³ 0.5 mg/kg per day of prednisone or equivalent for at least 3 weeks, or immunosuppressive or immunomodulatory drugs), admission to ICUs for acute hypoxemic respiratory failure (detailed as respiratory rate ³ 25 for minutes and a partial pressure of arterial oxygen to inspired fraction of oxygen [PaO2/FiO2] £ 300 mmHg while spontaneously breathing with COT with a flow ³ 10 l/min, HFNO or NIV) and adulthood (³ 18).

Hypercapnic patients (with a partial pressure of carbon dioxide higher than 50 mmHg), patients who could strongly benefit for NIV (those with underlying chronic lung disease, with cardiogenic pulmonary oedema, or postoperative patients), patients with severe shock, those with an urgent indication of intubation, those with a do-not-intubate order and those with a contraindication to NIV were excluded (for detailed definition of each of exclusion criteria see the original article). Also, intubation criteria were predetermined.

Patients were randomized to receive HFNO alone (HFNO alone group) or NIV alternating with HFNO (NIV group) stratifying for the center and for the type of immunosuppression.

HFNO was set with a flow of 60 l/min or the highest tolerated, while NIV was set in a pressure support mode, with a PEEP of at least 8 cmH20 and an inspired pressure adjusted to obtain a tidal volume (TV) below 8 ml/Kg of predicted bodyweight (PBW). FiO2 was set to reach an oxygen saturation of 92% or more.

Concerning the timing, the respiratory support assigned to the patients was started within 6 hours from the inclusion and had to be applied for at least 48 h from randomization. The patients of the HFNO alone group received HFNO continuously, while the patients of NIV group received a first session of NIV of 4 hours, and then NIV was applied for at least 12 hours per day, alternating session with HFNO. Both intubation criteria and weaning from HFNO or NIV was standardized (see the original article).

Outcomes
The primary outcome was mortality at day 28 after randomization. Secondary outcomes were: mortality in the ICU, in hospital, at day 90, and at day 180; intubation within 28 days after randomization; length of stay in the ICU and in hospital; number of ventilator-free days between randomization and day 28; number of oxygenation technique-free days (without HFNO or NIV) between randomization and day 28; and tolerance of oxygenation techniques. PaO2/FiO2, respiratory rate, dyspnea (using Likert scale) and tolerance (using visual analogue scale) was assessed 1 h after randomization.

Subgroup analyses were performed according to the etiology of immunosuppression (as previous data suggest that patients with leukopenia or hematological malignancy could benefit from NIV compared to standard oxygen therapy in terms of reduction of intubation rate and death in ICU [7]) and according to PaO2/FiO2 ratio at enrolment, because HFNO could reduce the intubation rate in patients with PaO2/FiO2 ratio equal to or below 200 mm Hg [8].

Study population
497 patients were screened from January 2017 to March 2019, 299 of which were included in the final analysis, matching the calculated sample-size. 154 patients were randomized to the HFNC alone group, 145 to the NIV group.

Median age was 62 (SD 12) in the HFNC alone group versus 65 (12) in the NIV group, and the median SOFA II score was 46 (18) and 45 (15) in the HFNC alone and the NIV group respectively. According to the stratification for the type of immunosuppression, patients with hematological malignancy, solid cancer, AIDS, and solid organ transplant recipient are evenly distributed between the two groups. The mean values of respiratory rate and PaO2/FiO2 were 32 (6) breaths per min and 147 (57) mmHg respectively. A mean flow of 58 l/min (5) and a mean FiO2 of 0.71 (0.22) was set in HFNO group, while initial NIV settings were: a mean pressure support of 7 (3) cmH20, a mean PEEP of 7 (2) cmH20, resulting in a tidal volume of 9.6 ml/kg of PBW. Almost all patients received NIV via face mask.

Results
No statistical difference in mortality rate at days 28 was found in the two groups (mortality rates were 36% and 35% in the HFNC alone group and in the NIV group respectively, p = 0.83). The type of immunosuppression did not influence the primary outcome (p = 0.15).

Mortality in ICU, in hospital, at day 90, at day 180 and the intubation rate at day 28, length of ICU and hospital stay, number of ventilator-free days and of non-invasive oxygenation technique free days at day 28 did not differ between the two groups. No statistical difference was found after performing subgroup analyses.

Focusing on respiratory parameters, PaO2/FiO2 was significantly higher in NIV group 1 h after randomization (mean PaO2/FiO2 HFNC alone group 143 vs. PaO2/FiO2 NIV group 199, p < 0.001), but the difference progressively reduces at 6 to 12 hours and at 24 hours (mean PaO2/FiO2 HFNC alone group 172 vs. PaO2/FiO2 NIV group 196 p < 0.001). No statistical difference was found in time-to-intubation between the two groups.

Discomfort significantly decreased in HFNO alone group 1 hour after randomization compared to NIV group (p = 0.04), while respiratory rate did not reach the statistical significance between the two groups.

Food for thoughts 

• FLORALI-IM is the first trial comparing NIV alternating with HFNO and HFNO alone in immunocompromised patients. The trial is precisely designed, procedures are detailed, and almost all criteria are standardized among centers (e.g., intubation criteria and weaning of HFNO and NIV criteria), leaving no arbitrary choice to the attending physician. Sample size is reached, while mortality observed in the two groups is higher than the one reported in other previous studies, as FLORALI-IM includes severe patients that are often excluded by other studies (e.g., severe neutropenia), perhaps making the study population more comparable to the “real” one that occupies ICUs.
• The study finds that HFNO therapy could be an alternative to NIV for immunocompromised patients, supporting 2021 ERS clinical practice guidelines about the use of HFNO in ARF, which suggests the use of HFNO over NIV in immunocompromised patient. Moreover, HFNO appears to be easier to set and to manage than NIV.
• Even if NIV improves gas exchange and requires a lower inspired fraction of oxygen from one hour to 24 hours after randomization, these respiratory parameters are not associated with a long-term influence on the primary outcome.
• After randomization, mean TV value is 9.6 ml PBW, suggesting that, in this population, protective ventilation is difficult to reach (TV < 8 ml PBW). TV after 1 hour of NIV do not differ between intubated and non-intubated patients (p = 0,63). Similar comparison in the later phases of hospitalization is not available.
• mortality seems to be higher in NIV group intubated patients than in HFNO alone group intubated patients (64% vs. 51%, respectively), even if these data are not comparable since samples are no longer randomized. There are not differences in time-to-intubation between the two groups, suggesting that neither HFNC nor NIV alternating with HFNC delays intubation compared to the other.

Conclusion
In immunocompromised patients admitted to ICU for acute respiratory failure, mortality rates at day 28 did not differ between patients treated with HFNO alone or NIV alternating with HFNO.

References

1. Azoulay E, Mokart D, Kouatchet A, Demoule A, Lemiale V. Acute respiratory failure in immunocompromised adults. Lancet Respir Med. 2019 Feb;7(2):173-186.
   
   
2. Azevedo LCP, Caruso P, Silva UVA, Torelly AP, Silva E, Rezende E, et al. Outcomes for patients with cancer admitted to the ICU requiring ventilatory support: results from a prospective multicenter Chest 2014;146(2):257-266.
   
   
3. Rochwerg B, Brochard L, Elliott MW, Hess D, Hill NS, Nava S, Navalesi P Members Of The Steering Committee, Antonelli M, Brozek J, Conti G, Ferrer M, Guntupalli K, Jaber S, Keenan S, Mancebo J, Mehta S, Raoof S Members Of The Task Force. Official ERS/ATS clinical practice guidelines: noninvasive ventilation for acute respiratory failure. Eur Respir J. 2017 Aug 31;50(2):1602426.
   
   
4. Ferreyro BL, Angriman F, Munshi L, Del Sorbo L, Ferguson ND, Rochwerg B, Ryu MJ, Saskin R, Wunsch H, da Costa BR, Scales DC. Association of Noninvasive Oxygenation Strategies With All-Cause Mortality in Adults With Acute Hypoxemic Respiratory Failure: A Systematic Review and Meta-analysis. JAMA. 2020 Jul 7;324(1):57-67.
   
   
5. Sklar MC, Mohammed A, Orchanian-Cheff A, Del Sorbo L, Mehta S, Munshi L. The impact of high-flow nasal oxygen in the immunocompromised critically ill: a systematic review and metaanalysis. Respir Care 2018; 63: 1555–66.
   
   
6. Oczkowski S, Ergan B, Bos L, Chatwin M, Ferrer M, Gregoretti C, Heunks L, Frat JP, Longhini F, Nava S, Navalesi P, Ozsancak Uğurlu A, Pisani L, Renda T, Thille AW, Winck JC, Windisch W, Tonia T, Boyd J, Sotgiu G, Scala R. ERS clinical practice guidelines: high-flow nasal cannula in acute respiratory failure. Eur Respir J. 2022 Apr 14;59(4):2101574
   
   
7. Hilbert G, Gruson D, Vargas F, et al. Noninvasive ventilation in immunosuppressed patients with pulmonary infiltrates, fever, and acute respiratory failure. N Engl J Med 2001; 344: 481–87.
   
   
8. Frat J-P, Thille AW, Mercat A, et al. High-flow oxygen through nasal cannula in acute hypoxemic respiratory failure. N Engl J Med 2015;372: 2185–96