Polok et al. Critical Care (2022) 26:224 https://doi.org/10.1186/s13054-022-04082-1 RESEARCH Open Access Noninvasive ventilation in COVID‑19 patients aged ≥ 70 years—a prospective multicentre cohort study Kamil Polok1,2, Jakub Fronczek1, Antonio Artigas3, Hans Flaatten4,5, Bertrand Guidet6,7, Dylan W. De Lange8, Jesper Fjølner9, Susannah Leaver10, Michael Beil11, Sigal Sviri12, Raphael Romano Bruno13, Bernhard Wernly14,15, Bernardo Bollen Pinto16, Joerg C. Schefold17, Dorota Studzińska1, Michael Joannidis18, Sandra Oeyen19, Brian Marsh20, Finn H. Andersen21,22, Rui Moreno23,24, Maurizio Cecconi25,26, Christian Jung13 and Wojciech Szczeklik1* on behalf of COVIP Study Group Abstract Background: Noninvasive ventilation (NIV) is a promising alternative to invasive mechanical ventilation (IMV) with a particular importance amidst the shortage of intensive care unit (ICU) beds during the COVID-19 pandemic. We aimed to evaluate the use of NIV in Europe and factors associated with outcomes of patients treated with NIV. Methods: This is a substudy of COVIP study—an international prospective observational study enrolling patients aged ≥ 70 years with confirmed COVID-19 treated in ICU. We enrolled patients in 156 ICUs across 15 European coun- tries between March 2020 and April 2021.The primary endpoint was 30-day mortality. Results: Cohort included 3074 patients, most of whom were male (2197/3074, 71.4%) at the mean age of 75.7 years (SD 4.6). NIV frequency was 25.7% and varied from 1.1 to 62.0% between participating countries. Primary NIV failure, defined as need for endotracheal intubation or death within 30 days since ICU admission, occurred in 470/629 (74.7%) of patients. Factors associated with increased NIV failure risk were higher Sequential Organ Failure Assessment (SOFA) score (OR 3.73, 95% CI 2.36–5.90) and Clinical Frailty Scale (CFS) on admission (OR 1.46, 95% CI 1.06–2.00). Patients initially treated with NIV (n = 630) lived for 1.36 fewer days (95% CI − 2.27 to − 0.46 days) compared to primary IMV group (n = 1876). Conclusions: Frequency of NIV use varies across European countries. Higher severity of illness and more severe frailty were associated with a risk of NIV failure among critically ill older adults with COVID-19. Primary IMV was associated with better outcomes than primary NIV. Clinical Trial Registration NCT04​321265, registered 19 March 2020, https://​clini​caltr​ials.​gov. Keywords: COVID-19, Noninvasive ventilation, Frailty, Intensive care unit, Elderly Introduction Coronavirus disease 2019 (COVID-19) led to an unprecedented disruption of everyday life and insuffi- *Correspondence: wojciech.szczeklik@uj.edu.pl ciency of healthcare systems around the world [1]. As 1 Centre for Intensive Care and Perioperative Medicine, Jagiellonian University a result, tremendous efforts were made by researchers Medical College, ul. Wrocławska 1‑3, 30 – 901 Kraków, Poland to elucidate the disease pathophysiology, find effective Full list of author information is available at the end of the article treatments, and develop vaccines [2–5]. COVID-19 © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Polok et al. Critical Care (2022) 26:224 Page 2 of 12 typically involves the respiratory system and may lead Methods to an acute respiratory distress syndrome (ARDS) Study design with a poor prognosis and frequent need for invasive This is a substudy of the COVIP study, an interna- mechanical ventilation [6]. It is estimated that approxi- tional prospective cohort study that recruited patients mately 9% of hospitalized patients become critically ill aged ≥ 70 years with confirmed COVID-19 admitted to and require transfer to an intensive care unit (ICU) [7]. the ICU. The COVIP study aims to assess outcomes and Unfortunately, the sheer volume of the most severely factors associated with the outcomes in the population of ill patients repeatedly led to an overload of ICUs. As elderly ICU patients with COVID-19. It is a part of the a consequence, the management of severe hypoxemic Very old Intensive Care Patients (VIP) research network, acute respiratory failure (ARF) required an adjustment which includes critical care physicians and researchers to the reality of the global pandemic. from around the world and is focused on investigating According to the current guidelines, the primary the management and outcomes of VIPs [14, 17]. Patients method of respiratory support in patients with ARDS included in this substudy were recruited in 156 centres is invasive mechanical ventilation (IMV). Conversely, from 15 countries between March 2020 and April 2021. noninvasive ventilation (NIV) is reserved for selected Detailed information about participating countries and patients with mild ARDS. It should be used with cau- number of enrolled patients is summarized in Table 1. tion and in constant preparedness for endotracheal National study coordinators were responsible for gaining intubation [8]. However, due to the shortage of avail- local ethical permission, supervision of patient recruit- able ICU beds, a significant proportion of COVID-19 ment, and recruitment of ICUs. Ethical consent proce- patients with ARDS have been treated with noninva- dures were highly variable across participating countries sive methods, including high-flow oxygen therapy and (Additional file 1: Table S1). The study was performed noninvasive ventilation, often outside the ICU [9–11]. in accordance with the Declaration of Helsinki and its This approach was based on available evidence of amendments. reduced intubation and mortality rates in patients with hypoxemic non-hypercapnic ARF treated with NIV [12]. Study population and data collection Due to multimorbidity and frailty, critically ill elderly The COVIP study included patients admitted to the patients have a particularly poor prognosis [13]. Simi- ICU and aged ≥ 70 years in whom SARS-CoV2 infection lar analyses among patients with COVID-19 confirmed was confirmed using reverse transcription-polymerase that increasing age and degree of frailty are related to chain reaction (RT-PCR). Should a patient be previ- worse outcomes in this population [14]. Compared to ously enrolled in the COVIP study, he or she would not NIV, endotracheal intubation and IMV are associated be recruited again upon transfer, readmission or for any with more discomfort and a higher risk of complica- other circumstance. tions, e.g., ventilator-associated pneumonia. Hence, Information about patients was gathered using elec- NIV may be a particularly appealing therapeutic tronic case report forms. The date of the ICU admis- option in elderly patients with COVID-19, including sion was labeled as day 1, and all dates were numbered those with the “do not intubate” order. The available sequentially from that day on. The study personnel gath- evidence suggests that NIV is superior to high-flow ered detailed clinical data on each patient, including nasal oxygen therapy (HFNOT) and conventional oxy- baseline demographic and clinical characteristics (see gen therapy in terms of decreasing 30-day intubation definitions of comorbidities in Additional file 1: Table S2), rate in these patients, although no effect on mortality Sequential Organ Failure Assessment (SOFA), and Clini- was observed [15, 16]. To date, there are no large-scale cal Frailty Scale (CFS) scores at admission. Based on the studies describing the use of NIV among old patients CFS score, patients were categorized as fit (1–3 points), in Europe and evaluating outcomes in this clinical vulnerable (4 points) or frail (5–9 points). We addition- context. ally gathered information about NIV (day of initiation This substudy of the COVIP study aimed to describe and duration) and invasive ventilation (day of intubation the use of NIV in critically ill older adults with and duration of IMV). Patients were included in the pri- COVID-19 admitted to European ICUs. Moreover, we mary NIV group when NIV was the initial mechanical attempted to assess the outcomes in patients treated ventilation modality. At the same time, patients in whom with NIV, identify risk factors for NIV failure, and IMV was introduced as the first respiratory support were compare the effects of primary NIV and primary IMV included in the primary IMV group. in this population. The patients were followed-up throughout the hospi- talization and up to three months after admission to the Polok et al. Critical Care (2022) 26:224 Page 3 of 12 Table 1 Noninvasive ventilation rate and application across included countries Country Number of patients Number of patients using NIV rate % NIV application NIV (95% CI) Primary NIV Post- extubation NIV Austria 40 10 25.0% 9 (90.0) 1 (10.0) (12.7–41.2%) Belgium 174 2 1.1% 2 (100.0) 0 (0.0) (0.1–4.1%) Denmark* 215 73 34.0% 66 (90.4) 6 (8.2) (27.7–40.7%) England 172 94 54.7% 89 (94.7) 5 (5.3) (46.9–62.2%) France 699 170 24.3 124 (72.9) 46 (27.1) (21.2–27.7%) Germany* 284 141 49.6% 124 (87.9) 15 (10.6) (43.7–55.6%) Greece 130 36 27.7% 19 (52.8) 17 (47.2) (20.2–36.2%) Israel 58 20 34.5% 15 (75.0) 5 (25.0) (22.5–48.1%) Netherlands 338 11 3.3% 7 (63.6) 4 (36.4) (1.6–5.8%) Norway 23 12 52.2% 12 (100.0) 0 (0.0) (30.6–73.2%) Poland 129 4 3.1% 2 (50.0) 2 (50.0) (0.9–7.8%) Portugal 91 48 52.7% 39 (81.3) 9 (18.7) (42.0–63.3%) Spain* 408 46 11.3% 30 (66.0) 15 (34.0) (8.4–14.8%) Switzerland 263 93 35.4% 63 (67.7) 30 (32.3) (29.6–41.5%) Wales 50 31 62.0% 29 (93.5) 2 (6.5) (47.2–75.4%) NIV—noninvasive ventilation *Day of NIV initiation was unknown in 2 patients from Germany and 1 patient from Denmark and Spain ICU. The primary endpoint for this substudy was mortal- performed using logistic regression. It included all-cause ity within 30 days from admission to the ICU. The sec- mortality at 30 days as dependent variable and the inter- ondary endpoint was NIV failure defined as death or val from NIV initiation to intubation in days treated as need for intubation within 30 days of admission to the a continuous variable as well as the set of the following ICU. independent variables selected according to available lit- erature and expert knowledge: number of days hospital- Statistical analysis ized prior to ICU admission, age, sex, body mass index We presented categorical variables as numbers (per- (BMI), SOFA score on admission, CFS score on admis- centages) and we compared them using the ­Chi2 test, sion, hypertension, ischaemic heart disease [IHD], diabe- while continuous variables were presented as medians tes, chronic pulmonary disease, chronic kidney disease, with interquartile ranges (IQR) and compared using the congestive heart failure. The linear association between Mann–Whitney test. Comparisons of crude mortality the pre-intubation NIV duration and 30-day mortality between study groups were performed using the log-rank was later presented graphically. test and were visualized using the Kaplan–Meier curves. Evaluation of factors associated with NIV failure was Multivariable analysis of the association between performed using logistic regression with NIV failure as a pre-intubation NIV duration and 30-day mortality was dependent variable and same set of independent variables Polok et al. Critical Care (2022) 26:224 Page 4 of 12 as in the model mentioned above. We performed a sen- 75.7 (SD 4.6) years, and most of the patients were male sitivity analysis for which we excluded patients in whom (2197/3074, 71.4%). The median SOFA score at admis- decision to withhold or withdraw life-sustaining treat- sion was 5 (IQR 3–8). Based on the CFS score patients ment was made within 2 days since admission to the ICU. were classified as fit (67.0%), vulnerable (15.6%) and frail Finally, we assessed intubation rate and factors associated (17.4%). The baseline characteristics of the study group with it in a subgroup of patients from the primary NIV are presented in Table 2. group in whom LST limitation was not introduced dur- ing the NIV therapy period. Noninvasive ventilation application across included To compare survival in the primary NIV and the pri- countries mary IMV groups, we performed a multivariable survival Noninvasive ventilation was used in 791 (25.7%) patients. analysis using restricted mean survival time adjusted for NIV was used as a primary mechanical ventilation the limitation of life-sustaining therapies (LST) during modality in 630 (79.6%) patients and as post-extubation the hospitalization and the above mentioned set of inde- respiratory support in 157 (19.8%) patients, while the day pendent variables. For the comparison of 30-day mor- of NIV introduction was unknown in 4 patients (0.5%). tality between the primary NIV and the primary IMV The frequency of NIV use varied significantly across groups we performed two separate sensitivity analyses: included countries and ranged from 1.1% (95% CI 0.1– (1) after exclusion of patients in whom LST limitation 4.1%) in Belgium to 62.0% (95% CI 47.2–75.4%) in Wales. was not introduced during the initial respiratory treat- NIV was used more commonly as primary therapy in all ment, and (2) after exclusion of patients in whom LST countries, but the distribution of NIV application differed limitation was introduced within 30 days since the ICU depending on the country. We did not observe any evi- admission. dent temporal trend in the frequency of NIV in the study This was a complete case analysis. Data missingness period (Additional file 1: Figure S2). Detailed information maps for each model are presented in Additional file 1: about the frequency and indications for NIV across par- Figure S1. A two-sided p-value < 0.05 was considered sta- ticipating countries is summarized in Table 1. tistically significant. Statistical analyses were performed In the primary NIV group noninvasive ventilation was using the R 4.1.0 software (R Development Core Team, initiated on day 1.0 (IQR 1.0–1.0) and the median dura- Vienna, Austria). tion of NIV therapy was 33.5 h (12.75–72.0). The histo- grams of the day of NIV initiation and the duration of Results NIV therapy are presented in Additional file 1: Figure S3. Characteristics of the study sample The cohort of this COVIP substudy comprised 3158 Clinical outcomes critically ill patients recruited in 156 centres across 15 Mortality at 30 days was 52.9% (333/630) in the primary countries. All analyses were performed among patients NIV group and 14.0% (22/157) in the post-extubation with complete 30-day follow-up (3074/3158, 97.3%). The NIV group. Among the 630 patients primarily treated study flowchart is presented in Fig. 1. The mean age was with NIV, 330 (52.4%) patients eventually required Fig. 1 Study flowchart Polok et al. Critical Care (2022) 26:224 Page 5 of 12 Table 2 Cohort characteristics Characteristics Entire cohort (n = 3074) NIV (n = 791) Primary NIV (n = 630) Post- extubation NIV (n = 157) Age, mean (SD) [years] 75.7 (4.6) 76.4 (4.9) 76.8 (4.9) 74.9 (4.6) Female gender 877 (28.5) 226 (28.6) 181 (28.7) 42 (26.8) BMI [kg/m2] 27.7 (24.8, 31.0) 27.7 (24.7, 31.3) 27.7 (24.5, 31.3) 27.8 (25.7, 31.5) Prior hospitalization length [days] 2.0 (1.0, 5.0) 2.0 (1.0, 4.0) 2.0 (1.0, 5.0) 2.0 (1.0, 4.0) Symptoms before hospitalization [days] 7.0 (4.0, 10.0) 7.0 (3.0, 10.0) 6.0 (3.0, 9.0) 7.0 (4.0, 10.0) Diabetes 1033 (33.7) 293 (37.1) 241 (38.4) 51 (32.5) Ischemic heart disease 694 (22.9) 190 (24.4) 158 (25.5) 31 (19.7) Chronic renal failure 495 (16.2) 155 (19.7) 134 (21.4) 20 (12.7) Arterial hypertension 2028 (66.2) 528 (67.0) 418 (66.7) 106 (67.5) Pulmonary disease 689 (22.5) 199 (25.3) 159 (25.4) 40 (25.5) Congestive heart failure 455 (15.0) 129 (16.5) 104 (16.8) 25 (15.9) Bacterial coinfection 651 (21.6) 201 (26.1) 161 (26.4) 40 (25.6) SOFA score on admission 5.0 (3.0, 8.0) 4.0 (3.0, 7.0) 4.0 (3.0, 6.0) 6.0 (4.0, 8.0) Frailty status Fit (CFS 1–3) 1912 (67.0) 459 (61.2) 356 (59.6) 102 (68.0) Vulnerable (CFS 4) 444 (15.6) 127 (16.9) 93 (15.6) 32 (21.3) Frail (CFS 5–9) 498 (17.4) 164 (21.9) 148 (24.8) 16 (10.7) Day of NIV initiation 1.0 (1.0, 3.0) 1.0 (1.0, 3.0) 1.0 (1.0, 1.0) 12.0 (8.0, 19.0) Duration of NIV [hours] 34.0 (10.0, 88.0) 34.0 (10.0, 88.0) 34.0 (10.0, 91.3) 33.5 (12.8, 72.0) IMV 2219 (72.2) 490 (61.9) 330 (52.4) 157 (100.0) Day of IMV initiation 1.0 (1.0, 2.0) 2.0 (1.0, 3.0) 2.0 (2.0, 5.0) 1.0 (1.0, 1.0) Vasopressors 2165 (70.8) 497 (63.4) 352 (56.3) 142 (91.6) Renal replacement therapy 488 (15.9) 120 (15.2) 94 (14.9) 25 (15.9) Antibiotics 2767 (90.1) 713 (90.1) 557 (88.4) 153 (97.5) Steroids 2058 (68.6) 565 (74.3) 467 (77.3) 96 (63.2) LST limitation 1189 (39.1) 306 (39.0) 276 (44.1) 28 (18.1) Withholding 981 (32.3) 251 (32.1) 227 (36.3) 23 (14.8) Withdrawal 627 (20.6) 161 (20.5) 150 (24.0) 9 (5.8) BMI—body mass index, CFS—Clinical Frailty Scale, IMV—invasive mechanical ventilation, LST—life-sustaining therapy, NIV—noninvasive ventilation, SOFA— Sequential Organ Failure Assessment invasive mechanical ventilation at a median 1.0 (0.0–3.0) file 1: Table S3. A multivariable analysis revealed that days from the initiation of NIV. Among patients primar- higher SOFA score at admission (OR 3.73, 95% CI 2.36– ily treated with NIV, we did not find sufficient evidence 5.90, p < 0.001) and higher baseline CFS (OR 1.46, 95% CI for a difference in 30-day mortality between patients 1.06–2.00, p = 0.02) were associated with a higher risk for who eventually required endotracheal intubation and the NIV failure, while hypertension was linked to lower odds remaining patients (58.2 vs. 47.0%, log-rank p = 0.32). The of NIV failure (OR 0.50, 95% CI 0.31–0.81, p = 0.005). For Kaplan–Meier curve for this comparison is presented in a sensitivity analysis, we excluded patients in whom the Additional file 1: Figure S4. The association between the decision to withhold or withdraw life-sustaining treat- duration of NIV and 30-day mortality in patients who ment was made within 2 days since admission to the ICU. eventually required intubation is visualized in Fig. 2. This analysis revealed a NIV failure rate accounting to 74.1% (403/544) and showed its association with a higher Factors associated with NIV failure SOFA score on admission (OR 3.38, 95% CI 2.10–5.46, NIV failure, defined as endotracheal intubation or death p < 0.001) and hypertension (OR 0.58, 95% CI 0.35–0.96, within 30 days since ICU admission, occurred in 470/629 p = 0.034). Finally, among patients in the primary NIV (74.7%) of patients primarily treated with NIV (date of group in whom LST limitation was not introduced dur- intubation was unknown for one patient). Univariate ing the initial NIV treatment, the intubation rate was comparison of these groups is summarized in Additional 67.4% (329/488) and was associated with the baseline Polok et al. Critical Care (2022) 26:224 Page 6 of 12 Fig. 2 Association between pre-intubation NIV duration and 30-day mortality. Black line represents OR point estimate across NIV duration prior to endotracheal intubation, while grey areas depicts 95% confidence interval SOFA score (OR 2.56, 95% CI 1.82–3.59, p < 0.001) and found a significantly lower crude 30-day mortality in age (0.70, 95% CI 0.49–0.99, p = 0.044). Additional file 1: the primary IMV group compared to the primary NIV Table S4 presents the mortality, NIV failure rate, and group (47.7 vs. 52.9%, log-rank p = 0.003) (Fig. 3). A intubation rate stratified by the duration of primary NIV. multivariable restricted mean survival time analysis revealed that in the 30-day follow-up patients in the Primary NIV vs. primary IMV and 30‑day mortality primary NIV group lived for 1.36 fewer days (95% CI − Noninvasive ventilation and invasive ventilation were 2.27 to − 0.46 days, p = 0.003) compared to the patients used as primary ventilation modality in 630 and 1876 in the primary IMV group. patients, respectively. Compared to patients initially We performed two sensitivity analyses to account treated with IMV, patients in the primary NIV group for LST limitation in these groups. The first sensitiv- were older (76.8 vs. 75.1, p < 0.001), more frequently ity analysis revealed that, after exclusion of patients in frail (24.8 vs. 13.7%, p < 0.001), more commonly had whom LST limitation was introduced during the ini- diabetes (38.4 vs. 32.5%, p = 0.009), IHD (25.5 vs. tial respiratory treatment (i.e., during NIV in the pri- 20.5%, 0.01), chronic renal failure (21.4 vs. 13.3%, mary NIV group and during IMV in the primary IMV p < 0.001), chronic pulmonary disease (25.4 vs. 21.4%, group), 30-day mortality in the primary NIV was higher p = 0.044), congestive heart failure (16.8 vs. 13.3%, than in the primary IMV group (43.2 vs. 29.5%, log- p = 0.034) and had a lower SOFA score on admission rank p < 0.001). For the second sensitivity analysis, we (4 vs 7, p < 0.001). Life-sustaining treatment was with- excluded all patients in whom LST limitation was intro- held more commonly in the primary NIV group (36.3 duced within 30 days since the ICU admission. It sug- vs. 29.5%, p = 0.002) and withdrawn similarly often in gested a lack of association between the modality of the both groups (24.0 vs. 22.5%, p = 0.495). The distribu- initial respiratory support and 30-day mortality (28.0 tion of LST limitation timing is presented in Additional vs. 29.5%, log-rank p < 0.55). The Kaplan–Meier curves file 1: Figure S5. Detailed univariate comparison of the for these comparisons are presented in Additional groups is presented in Additional file 1: Table S5. We file 1: Figure S6. Polok et al. Critical Care (2022) 26:224 Page 7 of 12 Fig. 3 Kaplan–Meier curve for comparison of 30-day mortality between primary NIV and primary IMV group Discussion variable across the Europe. This is in line with previous This prospective observational study of more than 3000 reports revealing significant international variation in patients aged ≥ 70 years old hospitalized in the ICU due management of ARF in the course of COVID-19 [20]. to severe COVID-19 showed that the use of NIV in this The presented inter-country differences are multifacto- population is highly variable in Europe and is associated rial and are probably related to local availability of ICU with high failure and mortality rates. A longer time to beds, presence of intermediate care units, and national intubation in patients requiring escalation from NIV to management strategies. The clinical relevance of nonin- IMV was associated with an increased 30-day mortality. vasive respiratory strategies increased significantly dur- Finally, a comparison of initial respiratory support strat- ing COVID-19 due to a dramatic increase in the number egies revealed that primary IMV was associated with of hypoxemic ARF cases. On the one hand, the main lower mortality compared to primary NIV. idea behind the implementation of NIV in this clini- At the moment, NIV is strongly recommended by the cal scenario was to avoid intubation and admission to combined European Respiratory Society and Ameri- the ICU. On the other hand, the high failure rate raised can Thoracic Society task force in two clinical scenarios, concerns that NIV may only delay endotracheal intuba- i.e., hypercapnic ARF due to COPD exacerbation and tion and potentially worsen patients outcomes [21, 22]. cardiogenic pulmonary oedema [18]. Previous reports Other potentially alarming aspects of NIV application in by the VIP Study Group, based on a cohort of patients patients with hypoxemic ARF due to COVID-19 included aged ≥ 80 years admitted to the ICUs before the pan- the risk of self-inflicted lung injury secondary to large demic, showed that every fourth patient received NIV tidal volumes as well as the generation of aerosol increas- [19]. In the light of a universal shortage of ICU beds dur- ing the risk of nosocomial infections [23]. Our study con- ing the COVID-19 pandemic, NIV became a promising firmed that more than half of patients initially treated alternative to IMV in patients with severe hypoxemic with NIV eventually required endotracheal intubation ARF. Such approach was also justified by encouraging and NIV failure, defined as intubation or death within results of a recent Bayesian network analysis [12]. The 30 days since the NIV initiation, was observed in almost current study showed that NIV was used in approxi- 75% of the cases. In comparison, the primary outcome, mately a quarter of patients included in this study; including intubation or death in the continuous posi- however, the frequency of NIV application was highly tive airway pressure (CPAP) group of the RECOVERY Polok et al. Critical Care (2022) 26:224 Page 8 of 12 RS trial, occurred in 36.3% of the patients, and 28-day Taking into account the high failure rate and the mortality and intubation rate in the helmet NIV group of well-established relation between delays in intubation the HENIVOT trial were 15% and 30%, respectively [15, and a poorer prognosis, we compared the outcomes of 16]. This discrepancy may be largely attributed to older patients depending on an initially introduced modal- age, a higher number of comorbidities, more frequent ity of respiratory support. Survival analysis revealed coexistence of frailty, and greater severity of the disease that mortality was higher in patients treated primarily expressed as a higher SOFA score at admission in the with NIV. Similar observation was made in a sensitivity COVIP study sample. Nevertheless, such a high NIV fail- analysis that excluded patients in whom LST was lim- ure rate was expected since elderly ICU patients have a ited during NIV in the primary NIV group and during particularly grim prognosis [24]. Therefore, in order to IMV in the primary IMV group. However, the differ- avoid unnecessary invasive procedures and futile suffer- ence in mortality between the groups disappeared after ing of the patients, LST limitations are commonly intro- the exclusion of all patients in whom LST limitation duced during hospitalization. In our study, any form of was introduced within 30 days since the ICU admis- LST limitation (withhold or withdrawal) was applied at sion. Another interesting observation is similar mortal- some point of hospitalization in more than 40% of cases. ity in the primary IMV group and in patients who were This proportion is markedly higher compared to a cohort treated with NIV and never required intubation (47.7% of merged VIP1 and VIP2 studies in which therapy limi- and 47.0%, respectively) and markedly higher mortality tation was introduced in 32% of patients [19, 25]. We in patients who required intubation after the initial NIV believe that this difference is yet another consequence trial (58.2%). On the one hand, these results are not of the ICUs being overflown by a staggering number of very surprising because NIV is a suboptimal therapy critically ill patients during the ongoing pandemic [26]. for the majority of ARDS cases according to the current As a result, more advanced treatment modalities such as guidelines and its application in this clinical scenario is IMV and extracorporeal membrane oxygenation became a last resort for elderly patients who would likely not be reserved for younger and healthier patients with a more qualified for intubation under pandemic circumstances. favourable prognosis. On the other hand, the results of the second sensitiv- Despite years of research, the risk factors for NIV fail- ity analysis suggest that both primary NIV and primary ure remain unclear. Most of the evidence on this issue is IMV may be associated with a similar survival rate in based on low-quality studies concerning predominantly elderly critically ill COVID-19 patients when LST limi- hypercapnic ARF [27–29]. Some interesting insights were tation is not considered. offered by a prospective observational study including The main strengths of this paper include an inter- over 1800 patients with ARF due to influenza, which sug- national multicentre character of the study and a gested an increasing SOFA score as a predictor of NIV relatively large sample of a very specific population. failure [30]. In our study increasing baseline SOFA score Moreover, the small amount of missing data on 30-day was related to higher risk of NIV failure as well as higher mortality increases the credibility of our findings. We intubation rate in patients without LST limitation during are aware of several limitations of this article. First, the primary NIV therapy. Additionally, we revealed an some details concerning NIV technique were not gath- association between an increasing degree of frailty and ered i.e., type of interface (face mask vs. helmet) and the risk of NIV failure. This is consistent with previous mode of ventilation (CPAP vs. bi-level). Second, we studies describing the impact of frailty on outcomes in did not gather data on HFNOT, an option increasingly elderly patients admitted to the ICU [14, 19]. used in patients with respiratory failure before tracheal It is crucial for a clinician not to cause harm while try- intubation and IMV. Third, participating countries are ing to avoid intubation by using NIV. In general, the use probably largely heterogenous in terms of intubation of NIV is restricted to mild ARDS with a success rate of criteria, which likely affects our results in a significant 78%, decreasing to 58% and 53% for moderate and severe way. Fourth, we did not collect any physiological or ARDS [21]. Our analysis of the association between the clinical data describing the initial response of patients duration of NIV before intubation and 30-day mortality to NIV therapy, which could potentially provide very revealed a poorer prognosis in a subgroup of patients in valuable information, particularly in terms of risk fac- whom NIV duration exceeded 3 days. It suggests that cli- tors for NIV failure. Finally, potentially important data nicians should reevaluate patients in terms of indications on the pre-ICU disease trajectory was not collected. for intubation early in the course of NIV therapy when Therefore, we were unable to determine whether and there are no signs of improvement. This could poten- for how long NIV or HFNOT were used before admis- tially prevent excessive delays in intubation and therefore sion to the ICU. improve patients’ outcomes. Polok et al. Critical Care (2022) 26:224 Page 9 of 12 Conclusions Intensiv, Regionshospitalet Horsens, Horsens, Denmark; Helene Korvenius Nedergaard: Intensive Care Unit, Odense University Hospital, Odense, In conclusion, NIV is used in approximately one in Denmark; Anders Thais Bjerregaard: Intensiv , Sygehus Lillebælt, Kolding, four elderly patients with COVID-19 treated in the Denmark; Ida Riise Balleby: Intensiv, Regionshospitalet Viborg, Viborg, European ICU, and its use varies significantly across Denmark; Kasper Andersen: Department of Anaesthesia and Intensive Care, Sygehus Sønderjylland, Aabenraa, Denmark; Maria Aagaard Hansen: Intensiv the European countries. Initial application of NIV was Afdeling, Regionshospitalet Herning, Herning, Denmark; Stine Uhrenholt: associated with a high risk of failure and increased Department of Anaesthesia and Intensive Care, Nordsjællands Hospital, mortality compared to patients in whom IMV was Hillerød, Denmark; Helle Bundgaard: Intensiv, Regionshospitalet Randers, Randers, Denmark; Jesper Fjølner: Department of Intensive Care, Aarhus the first mechanical ventilation modality. In addition, University Hospital, Aarhus, Denmark; Aliae AR Mohamed Hussein: Medical we found evidence of the association between higher ICU and Isolation Centers, Assiut University Hospital, Assiut, Egypt; Rehab SOFA and CFS scores and an increased risk of NIV fail- Salah: Cardiology ICU, One day surgery hospital, Nasr city, Egypt; Yasmin Khairy NasrEldin Mohamed Ali: MICU, Minia University Hospitals, Minia, Egypt; Kyrillos ure. Finally, among patients primarily treated with NIV Wassim: MICU, Quweisna central hospital, Quweisna, Egypt; Yumna A. Elgazzar: who eventually required IMV, delay in endotracheal MICU, Mayo Isolation Hospital, Cairo Governorate, Egypt; Samar Tharwat: intubation was associated with increased 30-day mor- Mansoura university Hospital, Temi El amdid, Mansoura, Egypt; Ahmed Y. Azzam: Alazhar University Hospitals, Cairo, Egypt; Ayman Abdelmawgoad tality. Careful monitoring during the first days of NIV Habib: intermediate ccu, one day surgery, nasr city, Egypt; Hazem Maarouf is essential to ensure that patients showing no sign of Abosheaishaa: MICU, Mostafa Mahmoud Specialized Hospital, Giza, Egypt; improvement from NIV, who are expected to benefit Mohammed A Azab: Sherif Mokhtar Cairo University ICU, Kar Al-Ainy Cairo University Hospital, Cairo, Egypt; Susannah Leaver: General Intensive care, St from an escalation to IMV, are offered endotracheal George´s University Hospitals NHS Foundation trust, London, England; Arnaud intubation promptly. Galbois: Medico-surgical ICU, Hôpital Privé Claude Galien, Quincy sous Sénart, France; Bertrand Guidet: Medical intensive care unit, Saint Antoine, Paris, France; Cyril Charron: Medical intensive care unit, Hôpital Ambroise Paré, Abbreviations Boulogne Billancourt, France; Emmanuel Guerot: Medical intensive care unit, ARDS: Acute respiratory distress syndrome; ARF: Acute respiratory failure; CFS: Hopital Européen Georges Pompidou, Paris, France; Guillaume Besch: Clinical Frailty Scale; COVID-19: Coronavirus disease 2019; HFNOT: High-flow Medico-surgical ICU, CHU de Besançon, Besançon, France; Jean-Philippe nasal oxygen therapy; ICU: Intensive care unit; IHD: Ischaemic heart disease; Rigaud: Medical intensive care unit, Dieppe General Hospital, Dieppe, France; IMV: Invasive mechanical ventilation; NIV: Noninvasive ventilation; IQR: Inter- Julien Maizel: Medical intensive care unit, CHU Amiens, Amiens, France; Michel quartile ranges; LST: Life-sustaining therapies; RT-PCR: Reverse transcription Djibré: Medico-surgical ICU, Tenon, Paris, France; Philippe Burtin: Surgical ICU, polymerase chain reaction; SOFA: Sequential organ failure assessment; VIP: Clinique Du Millenaire, Montpellier, France; Pierre Garcon: Medico-surgical ICU, Very old intensive care patients. Marne La Vallee, Jossigny, France; Saad Nseir: Medical intensive care unit, CHU Lille, Lille, France; Xavier Valette: Medical intensive care unit, CHU de Caen, Caen, France; Nica Alexandru: Medico-surgical ICU, Compiegne Noyon Supplementary Information Hospital, Compiegne, France; Nathalie Marin: Medical intensive care unit, The online version contains supplementary material available at https://​doi.​ Cochin, Paris, France; Marie Vaissiere: Medico-surgical ICU, CH Pau, Pau, France; org/​10.​1186/​s13054-​022-​04082-1. Gaëtan Plantefeve: Medico-surgical ICU, Victor Dupouy, Argenteuil, France; Thierry Vanderlinden: Medical intensive care unit, CH Saint Philibert, Lomme lez Lille, France; Igor Jurcisin: Medico-surgical ICU, Beaujon, Clichy, France; Additional file 1. Supplementary Figure 1. Missingness maps for each Buno Megarbane: Medical intensive care unit, Lariboisière, Paris, France; Anais model included in the paper. Supplementary Figure 2. Histograms Caillard: Surgical ICU, Lariboisière, Paris, France; Arnaud Valent: Surgical ICU, showing (A) number of recruited patients (B) proportion of primary NIV Saint-Louis, Paris, France; Marc Garnier: Surgical ICU, Saint Antoine, Paris, (C) proportion of primary IMV and (D) 30-day mortality stratified by study France; Sebastien Besset: Medico-surgical ICU, Louis Mourier, Colombes, month. Supplementary Figure 3. Histograms showing distribution of NIV France; Johanna Oziel: Medico-surgical ICU, Avicenne, Bobigny, France; initiation day and NIV duration stratified by group. Suuplementary Fig- Jean-herlé Raphalen: Medico-surgical ICU, Centre hospitalier de Versailles, Le ure 4. Kaplan-Meier curves comparing survival in patients with primary Chesnay, France; Stéphane Dauger: Pediatric Intensive and Intermediate Care NIV stratified by intubation status. Supplementary Figure 5. Distribu- Unit, Robert Debré, Paris, France; Guillaume Dumas: Medical intensive care tion of LST withhold and withdrawal timing. Supplementary Figure 6. unit, Saint-Louis, Paris, France; Bruno Goncalves: Medico-surgical ICU, Kaplan-Meier curves for the sensitivity analyses. Supplementary Table 1. Sainte-Anne, Paris, France; Gaël Piton: Medical ICU, CHU de Besancon, Details concerning ethical approval and patient`s consent requirements Besançon, France; Eberhard Barth: Anesthesiologic Intensive Care Department, in participating countries. Supplementary Table 2. Definitions of University Hospital Ulm, Ulm, Germany; Ulrich Goebel: Klinik für Anästhesie comorbidities. Supplementary Table 3. Comparison of patients primarily und operative Intensivmedizin, St. Franziskus-Hospital Münster, Münster, treated with NIV stratified by NIV failure. Supplementary Table 4. 30-day Germany; Eberhard Barth: IOI-Interdisziplinäre Operative Intensivmedizin, mortality, NIV failure rate and intubation rate stratified by the duration of University Hospital Ulm, Ulm, Germany; Anselm Kunstein: MX01, Uniklinik primary NIV. Supplementary Table 5. Comparison of patients primarily Düsseldorf, Düsseldorf, Germany; Michael Schuster: Anästhesie-Intensivsta- treated with NIV and IMV. tion, Universitätsmedizin der Johannes Gutenberg-Universität Mainz, Mainz, Germany; Martin Welte: Interdiszipinaere Operative Intensivstation, Klinik fuer Anaesthesiologie und operative Intensivmedizin, Klinikum Darmstadt GmbH, Acknowledgments Darmstadt, Germany; Matthias Lutz: Internistische Intensivstation, Uniklinik COVIP Study Group Philipp Eller: Allgemeine Medizin Intensivstation, Medical Schleswig Holstein Campus Kiel, Kiel, Germany; Patrick Meybohm: Klinik für University Graz, Graz, Austria; Michael Joannidis: Division of Intensive Care and Anästhesie und operative Intensivmedizin, University Hospital Würzburg, Emergency Medicine, Department of Internal Medicine, Medical University Würzburg, Germany; Stephan Steiner: ICU, St Vincenz, Limburg, Germany; Innsbruck, Innsbruck, Austria; Dieter Mesotten: Department of Intensive Care, Tudor Poerner: ITS, Marienhospital Aachen, Aachen, Germany; Hendrik Haake: Ziekenhuis Oost-Limburg, Genk, Belgium; Pascal Reper: Department of Internistische Intensivstation I und II, Kliniken Maria Hilf, Mönchengladbach, Intensive Care, CHR Haute Senne, Soignies, Belgium; Sandra Oeyen: Germany; Stefan Schaller: 43i, Charité - Universitätsmedizin Berlin, Berlin, Department of Intensive Care, Ghent University Hospital, Ghent, Belgium; Germany; Stefan Schaller: 44i, Charité - Universitätsmedizin Berlin, Berlin, Walter Swinnen: Department of Intensive Care, AZ Sint-Blasius, Dendermonde, Germany; Stefan Schaller: 8i, Charité - Universitätsmedizin Berlin, Berlin, Belgium; Helene Brix: Intensiv Behandling, Herlev og Gentofte Hospital, Herlev, Germany; Detlef Kindgen-Milles: CIA1, University Hospital Duesseldorf, Denmark; Jens Brushoej: Intensiv, Slagelse, Slagelse, Denmark; Maja Villefrance: Duesseldorf, Germany; Christian Meyer: Intensivstation, Evangelisches Polok et al. Critical Care (2022) 26:224 Page 10 of 12 Krankenhaus Düsseldorf, Düsseldorf, Germany; Muhammed Kurt: 32, Hospital de Tortosa Verge de la Cinta, Tortosa, Spain; Cesar Aldecoa: Florence-Nightingale Krankenhaus, Duesseldorf, Germany; Karl Friedrich Kuhn: Reanimación Quirurgica, Hospital Universitario Rio Hortega, Valladolid, Spain; 144i, Charité - Universitätsmedizin Berlin, Berlin, Germany; Winfried Randerath: Aida Fernández Ferreira: Servicio de Medicina Intensiva CHUVI, Hospital Alvaro Intensivpflege Bethanien, Krankenhaus Bethanien GmbH, Solingen, Solingen, Cunqueiro, Vigo, Spain; Begoña Zalba-Etayo: Unidad de Cuidados Intensivos, Germany; Jakob Wollborn: Anaesthesiologiesche Intensivtherapiestation, Clínico Universitario Lozano-Blesa, Zaragoza, Spain; Isabel Canas-Perez: Medical Center - University of Freiburg, Freiburg, Germany; Zouhir Dindane: Servicio de Medicina Intensiva - Unidad 2, Hospital Universitario Río Hortega, Interdisziplinäre Intensivstation, Städtische Kliniken Mönchengladbach, Valladolid, Spain; Luis Tamayo-Lomas: Servicio de Medicina Intensiva - Unidad Mönchengladbach, Germany; Hans-Joachim Kabitz: I01, Klinikum Konstanz, 3, Hospital Universitario Río Hortega, Valladolid, Spain; Cristina Diaz-Rodriguez: Konstanz, Germany; Ingo Voigt: Kardiologisch-internistische Intensivstation, Servicio de Medicina Intensiva - Unidad 4, Hospital Universitario Río Hortega, Elisabeth-Krankenhaus Essen, Essen, Germany; Gonxhe Shala: Station 2, Valladolid, Spain; Susana Sancho: General, H. Universitari i Politècnic La Fe, Johanna Etienne Krankenhaus, Neuss, Germany; Andreas Faltlhauser: Valencia, Spain; Jesús Priego: UCI CHUO, Complexo Hospitalario Universitario Interdisziplinäre Intensivmedizin, Kliniken Nordoberpfalz AG, Klinikum Weiden, Ourense, Ourense, Spain; Enas M.Y. Abualqumboz: Khartoum Bahri Hospital, Weiden, Germany; Greece; Nikoletta Rovina: ICU 1st Department of Pulmonary Khartoum, Sudan; Momin Majed Yousuf Hilles: Medical Military Hospital, Medicine Athens Medical School, National and Kapodistrian University of Khartoum, Sudan; Mahmoud Saleh: Wad Medani teaching Hospital, Wad Athens, Sotiria Hospital, Athens, Greece; Zoi Aidoni: ICU, University General Medani, Sudan; Nawfel Ben-HAmouda: Service de Médecine Intensive Adulte Hospital Ahepa, Thessaloniki, Greece; Evangelia Chrisanthopoulou: 2nd (SMIA), Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland; Andrea Department of Critical Care, University Hospital (Attikon), Haidari, Greece; Roberti: , Clinica Luganese Moncucco, Lugano, Switzerland; Alexander Antonios Papadogoulas: ICU, General Hospital of Larissa, Larissa, Greece; Dullenkopf: IPS KSF, Spital Thurgau Frauenfeld, Frauenfeld, Switzerland; Yvan Mohan Gurjar: Critical Care Medicine, Sanjay Gandhi Postgraduate Institute of Fleury: Intensive Care Unit, Fribourg Hospital, Fribourg, Switzerland; Bernardo Medical Sciences (SGPGIMS), Lucknow, India; Ata Mahmoodpoor: General, Bollen Pinto: Intensive Care Unit (Service des Soins Intensifs), Geneva Imam Reza, Tabriz, Iran; Abdullah Khudhur Ahmed: Baghdad Teaching University Hospitals, Geneva, Switzerland; Joerg C. Schefold: Dept of Intensive Hospital, Baghdad, Iraq; Ireland; Brian Marsh: Department of Critical Care Care Medicine, Inselspital Bern, Bern, Switzerland; Mohammed Al-Sadawi: ICU, Medicine, Mater Misericordiae University Hospital, Dublin, Ireland; Ahmed SUNY Downstate, Brooklyn, USA Elsaka: Covid ICU, Cork University Hospital, Cork, Ireland; Israel; Sigal Sviri: Corona ICU, Hadassah University Medical Center, Jerusalem, Israel; Vittoria Author contributions Comellini: Terapia Intensiva Respiratoria, Policlinico S. Orsola-Malpighi, Conception and design of the study were contributed by KP, JF, AA, HF, BG, Bologna, Italy; Libya; Ahmed Rabha: MICU, Askar, Suq Elkamis, Libya; Hazem DL, JF, SL, MB, SS, RB, BW, BP, JC, DS, MJ, SO, BM, FA, RM, MC, CJ, WS. Acquisition Ahmed: MICU, Tripoli University Hospital, Tripoli, Libya; Mexico; Silvio A of the data was contributed by KP, JF, AA, HF, BG, DL, JF, SL, MB, SS, RB, BW, BP, Namendys-Silva: Department of Critical Care Medicine, Instituto Nacional de JC, DS, MJ, SO, BM, FA, RM, MC, CJ, WS. Analysis was contributed by KP, JF, WS. Ciencias Medicas y Nutricion Salvador Zubiran, Mexico City, Mexico; Morocco; Interpretation of the data was contributed by KP, JF, AA, HF, BG, DL, JF, SL, MB, Abdelilah Ghannam: Service de Réanimation - Institut National d´Oncologie, SS, RB, BW, BP, JC, DS, MJ, SO, BM, FA, RM, MC, CJ, WS. Drafting the manuscript CHU Ibn Sina de Rabat, Rabat, Morocco; Martijn Groenendijk, ICU Department, was contributed by KP, AA, JF, WS. Critical revision of the manuscript was Alrijne Zorggroep, Leiderdorp, The Netherland; Marieke Zegers, Intensive Care contributed by KP, JF, AA, HF, BG, DL, JF, SL, MB, SS, RB, BW, BP, JC, DS, MJ, SO, department Radboudumc, Radboudumc, Nijmegen, The Netherland; Dylan de BM, FA, RM, MC, CJ, WS. KP, JF, AA, HF, BG, DL, JF, SL, MB, SS, RB, BW, BP, JC, DS, Lange, ICU departement, UMC Utrecht, Utrecht, The Netherland; Alex Cornet, MJ, SO, BM, FA, RM, MC, CJ, WS contributed to final approval of the manu- Intensive Care Center, Medisch Spectrum Twente, Enschede, The Netherland; script. Agreement to be accountable for all aspects of the work in ensuring Mirjam Evers, ICU Department, Canisius Wilhelmlina Ziekenhuis, Nijmegen, that questions related to the accuracy or integrity of any part of the work are The Netherland; Lenneke Haas, Intensive care, Diakonessenhuis Utrecht, appropriately investigated and resolved by KP, JF, AA, HF, BG, DL, JF, SL, MB, Utrecht, The Netherland; Tom Dormans, Zuyderland Heerlen, Zuyderland SS, RB, BW, BP, JC, DS, MJ, SO, BM, FA, RM, MC, CJ, WS. All authors read and Medical Center, Heerlen, The Netherland; Willem Dieperink, Department of approved the final manuscript. Critical Care, University Medical Center Groningen, Groningen, The Netherland; Luis Romundstad: Department of Critical Care and Emergencies, Oslo Funding University Hospital, Rikshospitalet Medical, Oslo, Norway; Britt Sjøbø: General COVIP study did not have any funding. Publication of this article was funded ICU, Haukeland University Hospital, Bergen, Norway; Finn H. Andersen: Dept. by the Priority Research Area qLife under the program “Excellence Initia- Anesthesia and Intensive Care, Surgical ICU, Ålesund Hospital, Ålesund, tive – Research University” at the Jagiellonian University in Krakow (06/ Norway; Hans Frank Strietzel: ICU, Kristiansund Hospital Helse Møre og IDUB/2019/94). Romsdal HF, Kristiansund N, Norway; Theresa Olasveengen: Surgical ICU, Oslo University Hospital, Oslo, Norway; Michael Hahn: ICU, Haugesund Hospital, Availability of data and materials Haugesund, Norway; Miroslaw Czuczwar: II Department of Anesthesiology The datasets used and/or analysed during the current study are available from and Intensive Care, First Independent Teaching Hospital No. 1, Lublin, Poland; the corresponding author on reasonable request. Ryszard Gawda: Department of Anesthesiology and Intensive Care, Opole University Hospital, Opole, Poland; Jakub Klimkiewicz: COVID-19 ICU, Military Institute of Medicine, Warsaw, Poland; Maria de Lurdes Campos Santos: Declarations Infectious Diseases ICU, Centro Hospitalar e Universitário São João, Porto, Portugal; André Gordinho: Serviço de Medicina Intensiva, Hospital de Beatriz Ethics approval and consent to participate Ângelo, Loures, Portugal; Henrique Santos: D, Centro Hospitalar Tráz os Montes Ethical consent procedures were highly variable across participating countries e Alto Dour, Vila Real, Portugal; Rui Assis: Serviço de Medicina Intensiva, Centro and while it was possible to recruit patients without informed consent in Hospitalar do Médio Tejo, Abrantes, Portugal; Ana Isabel Pinho Oliveira: some countries, in the remaining countries obtaining informed consent Unidade de Cuidados Intensivos Polivalente, Centro Hospitalar de Tondela- was mandatory. The study complied with the Declaration of Helsinki and Viseu, EPE, Viseu, Portugal; Saudi Arabia; Mohamed Raafat Badawy: MICU, its amendments. Details concerning ethical approval and patient`s consent Fundeni Clinical Institute, Jeddah, Saudi Arabia; Spain; David Perez-Torres: UVI requirements in participating countries are presented in Additional file 1: Polivalente y Coronaria, Hospital Universitario Río Hortega, Valladolid, Spain; Table S4. Gemma Gomà: Intensive Care Department, Corporacion Sanitaria Universitaria Parc Tauli, Sabadell, Spain; Mercedes Ibarz Villamayor: Intensive Care Unit, Consent for publication Hospital Universitario Sagrado Corazon, Barcelona, Spain; Angela Prado Mira: Not applicable. Intensive Care Unit, Hospital General Universitario de Albacete, Albacete, Spain; Patricia Jimeno Cubero: ICU Segovia, Complejo Hospitalario de Segovia, Competing interests Segovia, Spain; Susana Arias Rivera: Intensive Care and Burn Unit, Universitario Joerg C. Schefold declares that the Dept. of Intensive Care Medicine Bern has/ de Getafe, Getafe, Spain; Teresa Tomasa: General ICU, Germans Trias i Pujol had research and/or development/consulting contracts with (full disclosure): Hospital, Badalona, Spain; David Iglesias: UCI Burgos, Hospital Universitario de Orion Corporation, Abbott Nutrition International, B. Braun Medical AG, CSEM Burgos, Burgos, Spain; Eric Mayor Vázquez: Unidad de Cuidados Intensivos, SA, Edwards Lifesciences Services GmbH/SA, Kenta Biotech Ltd, Maquet Polok et al. Critical Care (2022) 26:224 Page 11 of 12 Critical Care AB, Omnicare Clinical Research AG, and Nestlé. Educational grants 5. Osuchowski MF, Winkler MS, Skirecki T, Cajander S, Shankar-Hari M, Lach- were received from Fresenius Kabi; GSK; MSD; Lilly; Baxter; Astellas; AstraZen- mann G, et al. The COVID-19 puzzle: deciphering pathophysiology and eca; B. Braun Medical AG, CSL Behring, Maquet, Novartis, Covidien, Nycomed, phenotypes of a new disease entity. Lancet Respir Med. 2021;9(6):622–42. Pierre Fabre Pharma (Roba Pharma); Pfizer, Orion Pharma. The money went https://​doi.​org/​10.​1016/​S2213-​2600(21)​00218-6. into departmental funds. No personal financial gain applies. All other authors 6. McGonagle D, Bridgewood C, Meaney JFM. A tricompartmental model of do not have any conflict of interest to declare related to this manuscript. lung oxygenation disruption to explain pulmonary and systemic pathol- ogy in severe COVID-19. Lancet Respir Med. 2021;9(6):665–72. https://​doi.​ Author details org/​10.​1016/​S2213-​2600(21)​00213-7. 1 Centre for Intensive Care and Perioperative Medicine, Jagiellonian University 7. Docherty AB, Mulholland RH, Lone NI, Cheyne CP, De Angelis D, Diaz- Medical College, ul. Wrocławska 1‑3, 30 – 901 Kraków, Poland. 2 Department Ordaz K, et al. Changes in in-hospital mortality in the first wave of COVID- of Pulmonology, Jagiellonian University Medical College, Kraków, Poland. 19: a multicentre prospective observational cohort study using the WHO 3 Critical Care Department, Corporacion Sanitaria Universitaria Parc Tauli, CIBER Clinical Characterisation Protocol UK. Lancet Respir Med. 2021;9(7):773– Enfermedades Respiratorias, Autonomous University of Barcelona, Sabadell, 85. https://​doi.​org/​10.​1016/​S2213-​2600(21)​00175-2. Spain. 4 Department of Anaesthesia and Intensive Care, Haukeland Univer- 8. Fernando SM, Ferreyro BL, Urner M, Munshi L, Fan E. Diagnosis and sity Hospital, Bergen, Norway. 5 Department of Clinical Medicine, University management of acute respiratory distress syndrome. Can Med Assoc J. of Bergen, Bergen, Norway. 6 INSERM, UMR_S 1136, Institut Pierre Louis 2021;193(21):E761–8. d’Epidémiologie et de Santé Publique, Equipe: Epidémiologie Hospitalière 9. Xu W, Sun N-N, Gao H-N, Chen Z-Y, Yang Y, Ju B, et al. Risk factors Qualité et Organisation des Soins, Sorbonne Universités, UPMC Univ Paris 06, analysis of COVID-19 patients with ARDS and prediction based on 75012 Paris, France. 7 Assistance Publique - Hôpitaux de Paris, Paris, France. machine learning. Sci Rep. 2021;11(1):2933. https://​doi.​org/​10.​1038/​ 8 Department of Intensive Care Medicine, University Medical Center, University s41598-​021-​82492-x. Utrecht, Utrecht, The Netherlands. 9 Department of Anaesthesia and Intensive 10. Reyes LF, Murthy S, Garcia-Gallo E, Irvine M, Merson L, Martin-Loeches I, Care, Viborg Regional Hospital, Viborg, Denmark. 10 Department of Critical Care et al. Clinical characteristics, risk factors and outcomes in patients with Medicine, St George’s Hospital, London, UK. 11 Medical Intensive Care Unit, severe COVID-19 registered in the International Severe Acute Respira- Hadassah Medical Center, Jerusalem, Israel. 12 Department of Medical Intensive tory and Emerging Infection Consortium WHO clinical characterisation Care, Hadassah Medical Center and Faculty of Medicine, Hebrew University protocol: a prospective, multinational, multicentre, observational study. of Jerusalem, Jerusalem, Israel. 13 Department of Cardiology, Pulmonology ERJ Open Res. 2022;8(1):552–2021. and Vascular Medicine, Medical Faculty, Heinrich-Heine-University Duessel- 11. Cammarota G, Esposito T, Azzolina D, Cosentini R, Menzella F, Aliberti S, dorf, Moorenstraße 5, 40225 Duesseldorf, Germany. 14 Department of Internal et al. Noninvasive respiratory support outside the intensive care unit for Medicine, General Hospital Oberndorf, Teaching Hospital of the Paracelsus acute respiratory failure related to coronavirus-19 disease: a systematic Medical University, Salzburg, Austria. 15 Institute of General Practice, Family review and meta-analysis. Crit Care. 2021;25(1):268. https://​doi.​org/​10.​ Medicine and Preventive Medicine, Paracelsus Medical University of Salz- 1186/​s13054-​021-​03697-0. burg, Salzburg, Austria. 16 Department of Acute Medicine, Geneva University 12. Ferreyro BL, Angriman F, Munshi L, Del Sorbo L, Ferguson ND, Rochwerg Hospitals, Geneva, Switzerland. 17 Department of Intensive Care Medicine, B, et al. Association of noninvasive oxygenation strategies with all-cause Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland. mortality in adults with acute hypoxemic respiratory failure: a systematic 18 Division of Intensive Care and Emergency Medicine, Department of Internal review and meta-analysis. JAMA. 2020;324(1):57–67. https://​doi.​org/​10.​ Medicine, Medical University Innsbruck, Innsbruck, Austria. 19 Department 1001/​jama.​2020.​9524. of Intensive Care 1K12IC, Ghent University Hospital, Ghent, Belgium. 20 Depart- 13. Fronczek J, Polok K, de Lange DW, Jung C, Beil M, Rhodes A, et al. ment of Critical Care Medicine, Mater Misericordiae University Hospital, Dublin, Relationship between the Clinical Frailty Scale and short-term mortality Ireland. 21 Department of Anaesthesia and Intensive Care, Ålesund Hospital, in patients ≥ 80 years old acutely admitted to the ICU: a prospec- Ålesund, Norway. 22 Department of Circulation and Medical Imaging, Norwe- tive cohort study. Crit Care. 2021;25(1):231. https://​doi.​org/​10.​1186/​ gian University of Science and Technology, Trondheim, Norway. 23 Faculdade s13054-​021-​03632-3. de Ciências Médicas de Lisboa ‑ Nova Médical School, Hospital de São José, 14. Jung C, Flaatten H, Fjølner J, Bruno RR, Wernly B, Artigas A, et al. The Centro Hospitalar Universitário de Lisboa Central, Lisbon, Portugal. 24 Facul- impact of frailty on survival in elderly intensive care patients with COVID- dade de Ciências da Saúde, Universidade da Beira Interior, Covilhã, Portugal. 19: the COVIP study. Crit Care. 2021;25(1):149. https://​doi.​org/​10.​1186/​ 25 Department of Anesthesia and Intensive Care Medicine, Humanitas Clinical s13054-​021-​03551-3. and Research Center – IRCCS, Via Alessandro Manzoni 56, 20089 Rozzano, 15. Perkins GD, Ji C, Connolly BA, Couper K, Lall R, Baillie JK, et al. An MI, Italy. 26 Department of Biomedical Sciences, Humanitas University, Pieve adaptive randomized controlled trial of non-invasive respiratory Emanuele, Rozzano, MI, Italy. strategies in acute respiratory failure patients with COVID-19. medRxiv. 2021;2021.08.02.21261379 Received: 18 March 2022 Accepted: 27 June 2022 16. Grieco DL, Menga LS, Cesarano M, Rosà T, Spadaro S, Bitondo MM, et al. Effect of helmet noninvasive ventilation vs high-flow nasal oxygen on days free of respiratory support in patients with COVID-19 and moderate to severe hypoxemic respiratory failure: the HENIVOT randomized clinical trial. JAMA. 2021;325(17):1731–43. https://​doi.​org/​10.​1001/​jama.​2021.​ References 4682. 1. Blumenthal D, Fowler EJ, Abrams M, Collins SR. Covid-19—implications 17. Van HPV, Beil M, Guidet B, Sviri S, Jung C, De LD, et al. A new multi- for the health care system. N Engl J Med. 2020;383(15):1483–8. https://​ national network studying very old intensive care patients. Anaesiol doi.​org/​10.​1056/​NEJMs​b2021​088. Intensive Ther. 2021;53(4):1–6. 2. Sterne JAC, Murthy S, Diaz JV, Slutsky AS, Villar J, Angus DC, et al. Associa- 18. Rochwerg B, Brochard L, Elliott MW, Hess D, Hill NS, Nava S, et al. Official tion between administration of systemic corticosteroids and mortality ERS/ATS clinical practice guidelines: noninvasive ventilation for acute among critically Ill patients with COVID-19: a meta-analysis. J Am Med respiratory failure. Eur Respir J. 2017;50(2):1602426. Assoc. 2020;324(13):1330–41. 19. Guidet B, de Lange DW, Boumendil A, Leaver S, Watson X, Boulanger 3. Group TWHOREA for C-19 T (REACT) W. Association between administra- C, et al. The contribution of frailty, cognition, activity of daily life and tion of IL-6 antagonists and mortality among patients hospitalized for comorbidities on outcome in acutely admitted patients over 80 years COVID-19: a meta-analysis. JAMA. 2021;326(6):499–518. https://​doi.​org/​ in European ICUs: the VIP2 study. Intensive Care Med. 2020;46(1):57–69. 10.​1001/​jama.​2021.​11330. https://​doi.​org/​10.​1007/​s00134-​019-​05853-1. 4. Barda N, Dagan N, Ben-Shlomo Y, Kepten E, Waxman J, Ohana R, et al. 20. Azoulay E, de Waele J, Ferrer R, Staudinger T, Borkowska M, Povoa P, et al. Safety of the BNT162b2 mRNA Covid-19 vaccine in a nationwide setting. International variation in the management of severe COVID-19 patients. N Engl J Med. 2021;385(12):1078–90. https://​doi.​org/​10.​1056/​NEJMo​ Crit Care. 2020;24(1):486. https://​doi.​org/​10.​1186/​s13054-​020-​03194-w. a2110​475. 21. Bellani G, Laffey JG, Pham T, Madotto F, Fan E, Brochard L, et al. Nonin- vasive ventilation of patients with acute respiratory distress syndrome. Polok et al. Critical Care (2022) 26:224 Page 12 of 12 Insights from the LUNG SAFE Study. Am J Respir Crit Care Med. 2016;195(1):67–77. https://​doi.​org/​10.​1164/​rccm.​201606-​1306OC. 22. Carteaux G, Millán-Guilarte T, De Prost N, Razazi K, Abid S, Thille AW, et al. Failure of noninvasive ventilation for de novo acute hypoxemic respira- tory failure: role of tidal volume*. Crit Care Med. 2016;44(2):282–90. 23. Ute Muti-Schüenemann GE, Szczeklik W, Solo K, Khabsa J, Thomas R, Borowiack E, et al. Update Alert 3: ventilation techniques and risk for transmission of coronavirus disease, including COVID-19. Ann Intern Med. 2021. https://​doi.​org/​10.​7326/​L21-​0424. 24. Mayerhöfer T, Klein SJ, Peer A, Perschinka F, Lehner GF, Hasslacher J, et al. Changes in characteristics and outcomes of critically ill COVID-19 patients in Tyrol (Austria) over 1 year. Wien Klin Wochenschr. 2021;133(23):1237– 47. https://​doi.​org/​10.​1007/​s00508-​021-​01945-5. 25. Fronczek J, Polok K, Nowak-Kózka I, Wludarczyk A, Górka J, Czuczwar M, et al. Frailty increases mortality among patients = 80 years old treated in Polish ICUs. Anaesthesiol Intensive Ther. 2018;50(4). 26. Jung C, Flaatten H, de Lange D, Beil M, Guidet B. The relationship between treatment limitations and pressure on intensive care units in elderly patients. Intensive Care Med. 2022;48(1):124–5. https://​doi.​org/​10.​ 1007/​s00134-​021-​06553-5. 27. Duan J, Wang S, Liu P, Han X, Tian Y, Gao F, et al. Early prediction of noninvasive ventilation failure in COPD patients: derivation, internal vali- dation, and external validation of a simple risk score. Ann Intensive Care. 2019;9(1):108. https://​doi.​org/​10.​1186/​s13613-​019-​0585-9. 28. Ferreira C, Dias C, Mendes M, Moita J. Predictors of failure of noninvasive ventilation (NIV) in acute respiratory failure due to chronic obstructive pulmonary disease. Eur Respir J. 2013;42(Suppl 57):P2492. 29. Ozyilmaz E, Ugurlu AO, Nava S. Timing of noninvasive ventilation failure: causes, risk factors, and potential remedies. BMC Pulm Med. 2014;14(1):19. https://​doi.​org/​10.​1186/​1471-​2466-​14-​19. 30. Rodríguez A, Ferri C, Martin-Loeches I, Díaz E, Masclans JR, Gordo F, et al. Risk factors for noninvasive ventilation failure in critically ill subjects with confirmed influenza infection. Respir Care. 2017;62(10):1307–15. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions