NIV in Immunocompromised Patients

The first study that assessed NIV in immunocompromised patients was published by Hilbert et al. in 2001 [25]. In this single-center randomized controlled trial, 52 immunosuppressed patients (not only patients with hematologic malignancies but several modalities of immunosuppression) were enrolled if they had pulmonary infiltrates, fever, and hypoxemic acute respiratory failure, defined by the presence of dyspnea at rest, respiratory rate greater than 30 breaths per minute, and a PaO₂/FiO₂ ratio of less than 200 mmHg while breathing oxygen. These patients were randomly allocated to receive either standard oxygen treatment via face mask or intermittent NIV. Compared to standard oxygen therapy, the group treated with NIV had lower rates of endotracheal intubation (12/26 patients vs. 20/26, p = 0.03), and in-hospital mortality (50% vs. 81%, p = 0.02). The main limits of this study were that the patients were recruited at a single center and that the mortality rate among patients managed with invasive mechanical ventilation was considerably higher than current rates (90%). Antonelli et al. reported the efficacy of NIV in reducing the need of endotracheal intubation and invasive mechanical ventilation in immunocompromised patients after solid organ transplantation with hypoxemic respiratory failure [28]. Forty patients with a PaO2/FiO2 ratio of less than 200 mmHg while breathing oxygen and active use of accessory respiratory muscles were randomized to receive either NIV or treatment with supplemental oxygen via Venturi mask. The group treated with NIV demonstrated significantly lower rates of endotracheal intubation (20% vs. 70%, p = 0.002) and ICU mortality (20% vs. 50%, p = 0.05). However, no significant difference was found in the in-hospital mortality rate. A more recent randomized controlled trial investigated the potential role of early use of continuous positive airway pressure (CPAP) in patients with hematologic malignancies delivered before ICU admission [59]. Forty patients of the ward with bilateral infiltrates, respiratory rate greater than 25 breaths/minute, and an oxygen saturation of less than 90% while breathing on room air were randomized to receive oxygen (FiO2 = 50%) either by face mask or helmet CPAP at 10 cmH₂O. Overall, significantly fewer patients treated with CPAP required ICU admission and NIV or invasive mechanical ventilation (4 vs. 16 patients; p = 0.0002). This was associated with an improved survival. However, these very encouraging results have not been confirmed in subsequent observational [60] and randomized clinical studies [61]. To assess the potential benefit of early NIV in reducing the mortality rate among immunocompromised patients who developed acute hypoxemic respiratory failure, Lemiale underwent a randomized controlled trial [6]. In 28 intensive care units from France and Belgium, 374 immunocompromised patients with PaO2 less than 60 mmHg on room air, or respiratory rate greater than 30/min, or signs of respiratory distress, were randomized to receive NIV or conventional oxygen therapy. Of note, patients were stratified according to the cause of immune deficiency in two groups, one with hematologic malignancy or solid cancer and one with solid organ transplant or long- term/high-dose immunosuppressive treatment. No difference was found between groups with regard to the primary endpoint, the mortality rate at 28 days after randomization (NIV 24.1% vs. oxygen 27.3%; 95% CI: -12.1 to 5.6; p = 0.47). Also the analysis of the two prespecified subgroups did not result in any significant difference. Nevertheless, it must be outlined that this trial was underpowered since the authors anticipated a mortality of 35% in the oxygen-treated group, whereas the observed mortality rate was 27.3%. As a result, the possibility of drawing definitive conclusions and a clinically meaningful effect based on the study findings is limited. A recent post hoc analysis, restricted to the subgroup of immunosuppressed patients of the FLORALI trial, in a population of patients with acute hypoxemic respiratory failure suggested no benefit of NIV neither for intubation nor for survival [62]. In contrast to what it was expected, NIV may not be the panacea for the main reason than discussed in non-immunosuppressed patients. The possible role of injurious breathing pattern, potentially aggravated by NIV, may hence provide another explanation for the lack of efficacy of NIV in immunocompromised patients.