Hrev_master [page 56] [Emergency Care Journal 2023; 19:11237] Emergency Care Journal 2023; volume 19:11237 Abstract SARS-CoV-2 related pneumonia is characterized by moderate- to severe hypoxemia often associated with hypocapnia the prog- nostic role of which is poorely documented. Our aim in the present study was to evaluate if hypocapnia can predict the need for Non- Invasive Mechanical Ventilation (NIMV) in this setting. We prospectively studied 52 subjects with moderate-severe SARS- CoV-2 related pneumonia. All the following data were collected at admission to the Emergency Department and processed by unuvariate and multivariate analysis: clinical and laboratory data, blood gas analysis in room air and lung ultrasound. A total of 33 out of 52 subjects (63.4%) underwent NIMV. At univariate analy- sis PaCO2 was inversely associated to the need for NIMV (OR 0.82, CI 95% 0.689-0.976, p 0.025). At multivariate analysis PaCO2 predicted the need for NIMV independently from age, gen- der, number of comorbidities, d-dimer, CRP, PaO2 and LUS SCORE (OR 0.838, CI 95% 0.710-0.988, p .035). Our data suggest that hypocapnia could be an early predictor of clinical worsening in these patients independently from other known predictors of unfavourable outcome, reflecting the occurrence of a deep and fre- quent respiratory pattern possibly related to the generation of excessive transpulmonary pressure swings leading to a Self- Induced Lung Injury (P-SILI). Further studies are needed for vali- dating these data on greater populations. Introduction About 3 years passed but Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) pandemic is still not ended. Although the situation is ameliorated, mainly after vaccine intro- duction, more than 6 millions people died worldwide so far.1 Also, as a consequence of the continuous appearance of novel variants able to elude immune system protection induced by both vaccines and acute infection, the global incidence of new cases and mortal- ity in fragile subjects are still high.1 In these years many papers concerning the risk stratification of covid patients have been pub- lished and some systematic review and meta-analysis are also available.2-4 Unfortunately, all these systematic review and meta- analysis showed that the proposed models are at elevated risk of bias and have low performance, mainly because of high hetero- geneity and lack of prospective design.2-4 Moreover, many of these studies are based on scoring systems already used for other dis- eases or focused their attention on generic parameters common to other diseases, such as epidemiological data (age, gender), number and type of comorbidities, laboratory data including inflammation index, D-Dimer, serum lactic dehydrogenase, creatinine, glomeru- lar filtration rate, haemoglobin levels and platelets count, thus being not specific for SARS-CoV-2 related disease; only few stud- ies focused their attention on respiratory parameters and the most common used was the PaO2/FiO2 ratio.2-4 The clinical spectrum of Coronavirus Disease 19 (COVID-19) can range from asymptomatic infection to severe pneumonia with acute respiratory distress syndrome and death.5 It is well known that one of the major characteristics of SARS-CoV-2 related pneu- monia is the presence of a moderate to severe hypoxemia not asso- ciated to proportional clinical signs of respiratory distress.6 Almost always hypoxemia is associated with hypocapnia. Covid pneumo- nia is characterized by an initial phase of prevalent interstitial and microvascular involvement, which leads to a condition of altered microvascular regulation and diffuse alveolar damage and, at least Correspondence: Paolo Groff, Pronto Soccorso, Azienda Ospedaliera di Perugia. Piazzale Menghini 1, 06129, Perugia. E-mail: paolo.groff@ospedale.perugia.it Key words: hypocapnia; NIMV; SARS-CoV-2. Conflict of interest: the authors declare no conflict of interest. Contribution: PG contributed to the study conception and design. Material preparation, data collection and analysis were performed by SDV. The first draft of the manuscript was written by SDV and GML and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Availability of data and materials: all data generated or analyzed during this study are available upon request. Ethics approval and consent to participate: the present study was approved by The local Ethical Committee. Written informed consent was obtained from all patients. Informed consent: written informed consent was obtained from a legally authorized representative(s) for anonymized patient information to be published in this article. Received for publication: 3 February 2023. Revision received: 2 March 2023. Accepted for publication: 3 March 2023. This work is licensed under a Creative Commons Attribution 4.0 License (by-nc 4.0). ©Copyright: the Author(s), 2023 Licensee PAGEPress, Italy Emergency Care Journal 2023; 19:11237 doi:10.4081/ecj.2023.11237 Publisher's note: all claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organiza- tions, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its man- ufacturer is not guaranteed or endorsed by the publisher. Hypocapnia as a predictor of the need for non-invasive mechanical ventilation in subjects with SARS-CoV-2 related pneumonia Stefano De Vuono, Sokol Berisha, Laura Settimi, Pasquale Cianci, Alessandra Lignani, Gianmarco Lanci, Maria Rita Taliani, Paolo Groff Emergency Department, Santa Maria della Misericordia Hospital, Perugia, Italy No n- co mm er cia l u se on ly in the first phase, lung compliance is not significantly compro- mised; thus, the patient develops a respiratory pattern character- ized by greater depth of breaths and greater respiratory frequency without feeling the respiratory effort and with consequent hypocapnia development.7 To our knowledge only few previous studies directly or indi- rectly evaluated the potential role of hypocapnia as a predictor of respiratory failure worsening in subjects with SARS-CoV-2 related pneumonia. We aimed our study for evaluating if hypocapnia detected at the time of admission to the Emergency Department could predict the need for Non-Invasive Mechanical Ventilation (NIMV) in sub- jects with SARS-CoV-2 related pneumonia. Materials and Methods We prospectively conducted our study at the field hospital set up near the emergency department of our hospital to accommodate SARS-CoV-2 infected patients in the time period between March and June 2021. Data collection has been interrupted because of changes in SARS-CoV-2 pandemic epidemiology (the end of the pandemic wave) which caused a change in the function of our sub- intensive unit thus not allowing the enrollment of new cases. The end-point of the study was the need for Non-Invasive Mechanical Ventilation (NIMV). Predefined criteria for considering the escalation of respiratory support to NIMV were the presence of SpO2 ≤92% while on VMK or HFNC or PaO2/FiO2 ratio ≤180 mm Hg with FiO2 ≥50%, and at least one between respiratory rate ≥28 breaths/min, severe dys- pnea, signs of increased work of breathing (e.g., use of accessory muscles). All the data (clinical medical history, vital signs, labora- tory parameters, blood gas analysis and lung ultrasound) were obtained at the time of admission to the emergency room. In order to eliminate the possible confounding effect of the supplemental oxygen administered, all blood gas analysis were performed in room air. After blood gas analysis oxygen was administered using Venturi-Mask (VMK) or High Flow Nasal Canula (HFNC) according to clinical judgement. Eligibility criteria were: age ≥18 years old; positive PCR test confirming SARS-CoV-2 infection; clinical signs of acute respira- tory infection and radiological or ultrasound evidence of pneumo- nia; peripheral oxygen saturation (SpO2) ≤92% or arterial partial pressure of oxygen to fraction of inspired oxygen (arterial oxygen tension (PaO2)/FiO2) ratio <300 in room air and need for oxygen therapy according to clinical judgement, at screening. Exclusion criteria included: respiratory rate ≥28 breaths/min; PaO2/FiO2 ratio ≤200; need for immediate intubation or for NIMV according to clinical judgement; patients already on NIMV at study screening; septic shock; evidence of multiorgan failure; Glasgow Coma Scale <13; neuromuscular disease; presence of partial pressure of arterial carbon dioxide (PaCO2) >45 mm Hg or history of chronic hyper- capnia. Patients already on long-term oxygen therapy and/ or home NIV/CPAP or with limitation of care based on patients’ or physi- cians’ decision or with the inability to comprehend the study con- tent and give consent were also excluded. At admission we took note of the following comorbidities: Hypertension (HTN), dia- betes, dyslipidaemia, obesity, Chronic Ischemic Cardiomyopathy (CIC), Cerebrovascular Disease (CVD), Peripheric Obliterative Arteriopathy (PAOD), Atrial Fibrillation (AF), Chronic Obstructive Pulmonary Disease (COPD); we reported the sum of them as Number Of Comorbidities (NOC). The following laboratory parameters were included: White Blood Count (WBC) with its relative formula, Haemoglobin (Hb), Platelets (PTL), glycemia, urea, creatinine, Glomerular Filtration Rate (GFR; calculated with MDRD formula), INR, RATIO, d- dimer, Aspartate Aminotransferase (AST), Alanine Aminotransferase (ALT), albumin, serum Lactic Dehydrogenase (LDH), C-Reactive Protein (CRP). All patients underwent standard Lung Ultrasound (LUS) for calculating the LUS SCORE. LUS was performed using the stan- dardized acquisition protocol which provides the acquisition of 12 standard areas (2 posterior, 2 lateral, and 2 anterior).8 Images were acquired by 5 well experienced emergency physicians using a con- vex probe with a frequency of 1-6 Hz (W-Cube i7, Alpinion). For each area the severity of lung findings was described numerically as follows: score 0: the pleural line is continuous and regular and presence of A-lines; score 1: the pleural line is indented and pres- ence of at least 3 well-spaced B-lines; score 2: presence of coales- cent B-lines or presence of the so-called “white lung” pattern; score 3: presence of consolidation. LUS SCORE was calculated by the sum of the highest scores obtained in each area. The local Ethical Committee approved the study (approval number 3948/21). Written informed consent was obtained from all patients. Statistical analysis Given the small sample size we reported the data as median and range. We used Student’s t test and the Mann-Whitney U test to compare parametric and non-parametric variables. For testing the univariate relations among variables we used binary logistic regression analysis and Spearman’s rank correlation coefficients. Binary logistic regression was used for testing the effect of possi- ble confounding factors on the considered end-point. Analyses were performed with SPSS software (version 19.0; SPSS, Inc., Chicago, IL), with statistical significance set at p<0.05. Results In the study period 52 patients, out of 282 admitted to the field hospital, fully met the inclusion criteria (30 males and 22 females); mean age was 61 ± 12 years. Mean number of comorbidities was 1,04 ± 1,03. The most frequent comorbidity was arterial hyperten- sion, followed by dyslipidemia and obesity (see Figure 1). None of the subjects included was affected by CIC, CVD or PAOD. Only Article Figure 1. Percentage of the most frequent comorbidities in the population study [Emergency Care Journal 2023; 19:11237] [page 57] No n- co mm er cia l u se on ly 2/52 (3.8%) of the subjects included were affected by COPD; con- sidering that we excluded from the study hypercapnic subjects it is not likely that the presence of COPD patients in our population altered the results. The other general characteristics of the popula- tion are described in Table 1. In 21/52 subjects (40.3%) oxygen was administered using HFNC, in 31/52 (59.7%) oxygen was administered using Venturi- Mask (VMK). 33/52 of the subjects included (63.4%) reached the end point requiring NIMV. Mean time from admission to NIMV start was 20,3 ± 13 hours. In Table 2 comparison between patients treated with VMK ver- sus HFNC are shown. The differences of the general characteris- tics between patients who needed NIV and patients who did not are shown in Table 3. At univariate analysis we found an inverse statistically signif- icant association between PaCO2 and need for NIMV. On the con- trary we did not find any significant association between need for NIMV and age, gender, number of comorbidities, systolic or dias- tolic blood pressure, heart rate, respiratory rate, white blood cells count, neutrophils, lymphocyte, platelets, blood urea, creatinine, glomerular filtration rate, AST, ALT, D-dimer, LDH, CRP, PaO2 and LUS SCORE (Table 4). We did not find any significant association between PaCO2 and respiratory rate (rho -0,257, p 0.075) and between PaCO2 and LUS SCORE (rho -0.060, p 0.679). The oxygen delivery system used (VMK or HFNC) was not significantly associated with the need for NIMV (OR 0.571, IC 95% 0.178-1.832, p .346). At multivariate analysis PaCO2 levels were able to predict the need for NIMV independently from age, gender, number of comor- bidities, D-dimer, CRP, PaO2 and LUS SCORE (OR 0.838, IC 95% 0.710-0.988, p .035; Table 5). Discussion Our results suggest that hypocapnia could be a good predictor of rapid respiratory failure worsening in subjects affected by SARS-CoV-2 related pneumonia independently from other already Article Table 2. Comparison between patients treated with VMK versus HFNC VMK HFNC p Age, yrs 65 ± 12 62 ± 14 ns Days from symptoms onset 7.5 ± 3 7.2 ± 3 ns Days from positive swab 5.8 ± 3 6.3 ± 3 ns Number of comorbidities 1.2 ± 1.18 0.76 ± 0.7 ns SBP, mmHg 139 ± 22 140 ± 14 ns DBP, mmHg 80 ± 14 81 ± 10 ns Heart rate, bpm 93 ± 16 86 ± 19 ns Respiratory rate, breaths/min 25 ± 5 25 ± 4 ns Body temperature, °C 36.8 ± 0.7 36.9 ± 08 ns WBC, 103/mmc 7094 ± 2764 7829 ± 3512 ns HB, mg/dL 14.3 ± 1.4 13.4 ± 1.8 ns PTL, 103/mmc 193 ± 66 197 ± 59 ns Glycemia, mg/dL 161 ± 60 137 ± 33 ns Urea, mg/dL 42 ± 23 39 ± 12 ns Creatinin, mg/dL 0.87 ± 0.26 0.81 ± 0.19 ns GFR, mL/min/1,73 m2 87 ± 21 88 ± 16 ns D-dimer, µg/mL 710 ± 337 952 ± 1086 ns AST, IU/L 75 ± 60 86 ± 182 ns ALT, IU/L 67 ± 68 62 ± 99 ns Albumin, mg/dL 3.6 ± 0.3 3.6 ± 0.3 ns LDH, IU/L 412 ± 140 474 ± 260 ns CRP, mg/mL 7.6 ± 4.4 26.7 ± 69 ns pH 7.47 ± 0.04 7.47 ± 0.05 ns PaO2, mmHg 59 ± 6 57 ± 5 ns PaCO2, mmHg 31 ± 5 30 ± 4 ns HCO3-, mEq/L 23 ± 4 22 ± 3 ns P/F 281 ± 28 272 ± 21 ns Table 1. General characteristics of the population Median Range Age, years 61.5 36-89 Days from symptoms onset 7 0-15 Days from positive swab 6 0-15 SBP, mmHg 141.5 80-174 DBP, mmHg 82 45-103 HR, bpm 93 50-120 RR, breaths/min 26 14-36 Body Temperature, °C 36.7 35-38 WBC, 103/mm3 7020 1140-14690 N, % 82.8 36.7-2.7 L, % 13.2 2.7-53.7 M, % 4.9 1-11.4 E, % 0.000 0.0-0.6 B, % 0.000 0.0-0.4 Hb, g/dl 13.7 10-18.6 PLT, 103/mm3 185 91-361 Glycemia, mg/dl 133.5 92-315 Urea, mg/dl 38 13-110 Creatininemia, mg/dl 0.81 0.42-1.6 GFR, ml/min/1,73 m2 93.1 41-116 D-dimer, µg/ml 676 224-5363 AST, UI/L 50 23-881 ALT, UI/L 43 13-480 Albumin mg/dl 3.6 2.9-4.5 LDH, UI/L 393 206-1476 CRP, mg/dl 7 1.6-326 LUS SCORE 12 0.0-36 Blood gas analysis pH 7.47 7.25-7.54 PaO2 57.4 45.3-77.5 PaCO2 30.8 17.7-42.7 HCO3-, mEq/L 22.7 14-31 P/F 274.6 217-371 SBP: systolic blood pressure; DBP: diastolic blood pressure; HR: heart rate; RR: respiratory rate; WBC: white blood cells; N: neutrophils; L: lymphocyte; M: monocyte; E: eosinophils; B: basophils; Hb: haemo- globin; PTL: platelets; GFR: glomerular filtration rate; AST: aspartate aminotransferase; ALT: alanine aminotransferase; LDH: serum lactic dehydrogenase; CPK: creatine phosphokinase; CRP: C-reactive pro- tein; LUS SCORE: Lung Ultrasound SCORE [page 58] [Emergency Care Journal 2023; 19:11237] No n- co mm er cia l u se on ly known predictors. In our study PaCO2 levels are the only parame- ter significantly associated to the need for NIMV also after correc- tion for the main variables currently considered as severity disease predictors, whether they are patient pre-existing characteristics (age, gender, number of comorbidities), laboratory parameters (inflammation index and D-dimer) or ultrasound data (LUS SCORE), able to quantify the entity of lung involvement. It is likely that hypocapnia is expression of one of the patho- physiological mechanisms involved in respiratory failure worsen- ing of covid patients. SARS-CoV-2 related pneumonia follows always the same evolution phases that we can conceptually divide into phases based on the intensity of treatment required.9 There is a first inflammatory phase without atelectasis which correspond to patient arrival to the emergency department. Subsequently, the dis- ease can evolve to a phase characterized by edema and atelectasis which correspond to patient’s admission to semi-intensive or inten- sive care unit and finally the disease can resolve or can evolve towards a fibrotic phase.9 The first phase is characterized by a pre- served lung compliance which allows for a respiratory pattern characterized by deep and frequent breathing.9 The occurrence of this respiratory pattern that causes hypocapnia on one hand, can generate an excessive swing of transpulmonary pressures on the other hand, inducing a real risk of a patient self-induced lung injury (P-SILI),10,11 thus maintaining a vicious circle that can pro- mote the evolution to a more severe phase of the disease. Thus, we believe that the degree of hypocapnia detected at emergency department arrival help us in identifying those patients at higher risk of self-induced lung injury and thus at higher risk of rapid res- piratory failure worsening. Surprisingly our data analysis shows that PaCO2 levels are not associated with respiratory rate. Maybe in the first phase of SARS-CoV-2 related pneumonia the increase in respiratory depth is more pronounced than the increase in respi- ratory rate. In our data also LUS SCORE is not significantly associated with PaCO2 levels, suggesting that the degree of lung involvement does not influence the presence of hypocapnia. In line with the results of COVID-HIGH trial,12 the first multi- center randomized controlled trial to report results on the use of high-flow nasal oxygen in hospitalized patients with COVID-19 and mild hypoxemia, in our study the type of oxygen delivery sys- tem (VMK or HFNC) is not associated with the need for NIMV. Article Table 3. Comparison between NIMV and non-NIMV patients. NO-NIMV NIMV p Age, yrs 62 ± 11 60 ± 12 ns Days from symptoms onset 7.2 ± 4 7.3 ± 2 ns Days from positive swab 6.7 ± 3 5.5 ± 3 ns Number of comorbidities 1 ± 0.9 1 ± 1 ns SBP, mmHg 140 ± 21 140 ± 18 ns DBP, mmHg 81 ± 13 80 ± 12 ns Heart rate, bpm 86 ± 17 92 ± 17 ns Respiratory rate, breaths/min 24 ± 3 25 ± 5 ns Body temperature, °C 36.8 ± 0.8 36.8 ± 0.8 ns WBC, 103/mmc 7475 ± 3571 7274 ± 2802 ns HB, mg/dl 13.8 ± 1.2 14.1 ± 1.8 ns PTL, 103/mmc 222 ± 75 184 ± 55 .045 Glycemia, mg/dl 135 ± 34 161 ± 57 .047 Urea, mg/dl 40 ± 12 42 ± 22 ns Creatinin, mg/dl 0.84 ± 0.19 0.85 ± 0.25 ns GFR, ml/min/1,73 m2 87 ± 18 88 ± 20 ns D-dimer, µg/ml 805 ± 413 806 ± 869 ns AST, IU/L 50 ± 21 95 ± 152 ns ALT, IU/L 52 ± 32 74 ± 98 ns Albumin, mg/dl 3.6 ± 0.2 3.6 ± 0.3 ns LDH, IU/L 369 ± 95 471 ± 232 ns CRP, mg/ml 7 ± 5 20 ± 55 ns pH 7.47 ± 0.04 7.47 ± 0.05 ns PaO2, mmHg 60 ± 7 58 ± 4 ns PaCO2, mmHg 32 ± 4 29 ± 5 .055 HCO3-, mEq/L 24 ± 4 22 ± 3 .017 P/F 284 ± 34 274 ± 20 ns Table 4. Univariate analysis among variables included in the study and need for NIMV OR 95% CI p Age 1.001 0.955 – 1.048 0.981 Gender 1.1196 0.387 – 3.697 0.756 Number of comorbidities 0.982 0.568 – 1.698 0.948 SBP 1.003 0.974 – 1.033 0.851 DBP 0.992 0.948 – 1.038 0.725 HR 1.029 0.993 – 1.065 0.112 RR 1.123 0.981 – 1.286 0.093 WBC 1.000 1.000 – 1.000 0.416 N 1.054 0.999 – 1.113 0.054 L 0.952 0.899 – 1.009 0.098 PLT 0.995 0.986 – 1.004 0.277 Urea 1.017 0.984 – 1.050 0.315 GFR 0.997 0.968 – 1.028 0.861 Creatininemia 1.664 0.142 -19.567 0.685 AST 1.015 0.993 – 1.037 0.197 ALT 1.006 0.995 – 1.017 0.297 D-dimer 1.000 0.999 – 1.001 0.843 LDH 1.003 0.998 – 1.007 0.227 CRP 1.106 0.988 – 1.239 0.079 PaO2 0.899 0.799 – 1.011 0.076 PaCO2 0.859 0.749 – 0.984 0.028 LUS SCORE 1.075 0.994-1.163 0.072 Table 5. Multivariate analysis among predictors of the need for NIMV OR 95% CI p Age 0.981 0.923-1.043 0.526 Gender 1.217 0.283-5.243 0.792 N° comorbidities 0.743 0.340-1.588 0.432 D-dimer 0.999 0.998 – 1.001 0.273 CRP 1.114 0.946 – 1.312 0.196 PaO2 0.928 0.808-1.066 0.291 PaCO2 0.838 0.710-0.988 0.035 LUS SCORE 1.062 0.963-1.171 0.228 [Emergency Care Journal 2023; 19:11237] [page 59] No n- co mm er cia l u se on ly To our knowledge only few previous studies explored the role of hypocapnia as a possible predictor of disease severity. Some studies indirectly explored the possible role of hypocapnia using the standard PaO2, which corrects PaO2 for PaCO2 levels, for cal- culating the P/F ratio.1,13 In the study by Prediletto et al, the authors showed that using the standard PaO2 is more accurate than the tra- ditional P/F in predicting in-hospital mortality,13 while, on the con- trary, Maraziti et al. showed that calculating the P/F ratio using the standard PaO2 is not able to predict this outcome.14 More directly a retrospective study on 165 covid subjects requiring NIMV showed that PaCO2 levels are significantly lower in non- survivors.15 Similarly, in a previous study we showed that the ratio between PaCO2 levels and the fraction of inspired oxygen was able to predict the need for invasive mechanical ventilation indepen- dently from several already known predictors of severe disease.16 In line with these studies, that indirectly demonstrated the pre- dictive role of clinical worsening of hypocapnia, our data prospec- tively highlight its potential role in predicting the need for NIMV in a patient population entering the emergency department for acute respiratory failure related to SARS-CoV-2 infection, inde- pendently of a number of confounding factors. Another interesting aspect is that the mean time from admis- sion to NIMV start was only about 20 hours. We suppose that hypocapnia allows the recognition of those subjects in which the pathophysiological mechanism causing the P-SILI is ongoing and thus allows the early recognition of those subjects at higher risk of early deterioration. This could also explain why we did not find significant association among the already known predictors of severe disease and the need for NIMV. It is likely that parameters such age, gender, comorbidities, inflammation and prothrombotic state are more efficient in predicting long-term outcomes, such as mortality or intubation, than a short-term outcome. In conclusion our data suggest that hypocapnia seems to be a good predictor of the need for NIMV in subjects affected by SARS-CoV-2 related pneumonia, independently from other already known predictors of unfavorable outcome. Moreover, we observed that the majority of subjects needed NIMV support after few hours from admission suggesting that hypocapnia allows the early recognition of the subjects at higher risk of rapid deteriora- tion. Certainly, a better risk stratification of covid patients, helping physicians in choosing the better allocation of covid patients is needed and certainly further multicentric studies are needed for validating these data. Limitations The main limitation of our study is the small simple size. Unfortunately changes in SARS-CoV-2 pandemic epidemiology at the time of data collection (the end of the pandemic wave) caused a change in the function of our sub-intensive unit thus not allowing the enrollment of new cases. The study is monocentric and therefore the results may not be extensible to different contexts. This mean that the potential role of hypocapnia need an external validation. Finally, we did not include in our analysis the Rox index, which was validated as a predictor of hospitalization and mortality of covid patients in the setting of the emergency department.17 However ROX index is defined as the ratio of pulse oximetry/frac- tion of inspired oxygen (SpO2/FiO2) to Respiratory Rate (RR). We wanted to focus our attention on the role of hypocapnia for the pos- sible pathophysiological role in identifying the subjects at higher risk of P-SILI and for eliminating the possible confounding effect of the supplemental oxygen we included in our study only subjects with blood gas analysis performed in room air. For this reason, we believe that calculating the ROX index in this contest could be at high risk of bias, without adding new information respect to the already existing data on ROX index in covid patient. References 1. WHO. WHO Coronavirus (COVID-19) Dashboard. Accessed: 27th January 2023. Available from: https://covid19.who.int 2. Wynants L, Van Calster B, Collins GS, et al. Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal. BMJ 2020;369:m1328 3. Booth A, Reed AB, Ponzo S, et al. Population risk factors for severe disease and mortality in COVID-19: A global systematic review and meta-analysis. PLoS ONE 2021;16:e0247461. 4. Chu K, Alharahsheh B, Garg N, et al. Evaluating risk stratification scoring systems to predict mortality in patients with COVID-19. BMJ Health Care Inform 2021;28:e100389. 5. Wu Z, McGoogan JM. Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China: sum- mary of a report of 72,314 cases from the Chinese Center for Disease Control and Prevention. JAMA 2020;323:1239-42. 6. Tobin MJ, Laghi F, Jubran A. Why COVID-19 silent hypoxemia is baffling to physicians. Am J Respir Crit Care Med 2020;202:356-360. 7. Dhont S, Derom E, Van Braeckel E, Depuydt P, Lambrecht BN. The pathophysiology of 'happy' hypoxemia in COVID-19. Respir Res 2020;21:198. 8. Mojoli F, Bouhemad B, Mongodi S, Lichtenstein D. Lung ultrasound for critically ill patients. Am J Respir Crit Care Med 2019;199:701– 14. 9. Gattinoni L, Gattarello S, Steinberg I, et al. COVID-19 pneumonia: pathophysiology and management. Eur Respir Rev 2021;30:210138. 10. Brochard L, Slutsky A, Pesenti A. Mechanical ventilation to minimize progression of lung injury in acute respiratory failure. Am J Respir Crit Care Med 2017;195:438-42. 11. Grieco DL, Menga LS, Eleuteri D, Antonelli M. Patient self-inflicted lung injury: implications for acute hypoxemic respiratory failure and ARDS patients on non-invasive support. Minerva Anesthesiol 2019;85:1014-23. 12. Crimi C, Noto A, Madotto F, et al. High-flow nasal oxygen versus conventional oxygen therapy in patients with COVID-19 pneumonia and mild hypoxaemia: a randomised controlled trial. Thorax 2022;0:1–8. 13. Prediletto I, D’Antoni L, Carbonara P et al. Standardizing PaO2 for PaCO2 in P/F ratio predicts in-hospital mortality in acute respiratory failure due to COVID-19: a pilot prospective study. Eur J Intern Med 2021;92:48-54. 14. Maraziti G, Becattibi C. Early variation of respiratory indexes to pre- dict death or ICU admission in severe acute respiratory syndrome coronavirus-2-related respiratory failure. Respiration 2022;101:632- 637. 15. Gupta B, Jain G, Chandrakar S, Gupta N, Agarwal A. Arterial blood gas as a predictor of mortality in COVID pneumonia patients initiated on noninvasive mechanical ventilation: a retrospective analysis. Indian J Crit Care Med 2021;25:866–71. 16. De Vuono S, Cianci P, Berisha S, et al. The PaCO2/FiO2 ratio as out- come predictor in SARS-COV-2 related pneumonia: a retrospective study. Acta Biomed 2022;93:e2022256. 17. Gianstefani A, Farina G, Salvatore V, et al. Role of ROX index in the first assessment of COVID-19 patients in the emergency department. Intern Emerg Med 2021;16:1959–65. Article [page 60] [Emergency Care Journal 2023; 19:11237] No n- co mm er cia l u se on ly