Nseir et al. Crit Care (2021) 25:177 https://doi.org/10.1186/s13054-021-03588-4 RESEARCH Open Access Relationship between ventilator‑associated pneumonia and mortality in COVID‑19 patients: a planned ancillary analysis of the coVAPid cohort Saad Nseir1,2* , Ignacio Martin‑Loeches3,4, Pedro Povoa5,6, Matthieu Metzelard7, Damien Du Cheyron8, Fabien Lambiotte9, Fabienne Tamion10, Marie Labruyere11, Demosthenes Makris12, Claire Boulle Geronimi13, Marc Pinetonde  Chambrun14, Martine Nyunga15, Olivier Pouly16, Bruno Mégarbane17, Anastasia Saade18, Gemma Gomà19, Eleni Magira20, Jean‑François Llitjos21, Antoni Torres22, Iliana Ioannidou23, Alexandre Pierre24, Luis Coelho5, Jean Reignier25, Denis Garot26, Louis Kreitmann27, Jean‑Luc Baudel28, Guillaume Voiriot29, Damien Contou30, Alexandra Beurton31, Pierre Asfar32, Alexandre Boyer33, Arnaud W. Thille34, Armand Mekontso‑Dessap35, Vassiliki Tsolaki12, Christophe Vinsonneau36, Pierre‑Edouard Floch37, Loïc Le Guennec38, Adrian Ceccato39, Antonio Artigas40, Mathilde Bouchereau1, Julien Labreuche41, Alain Duhamel41 and Anahita Rouzé1,2the coVAPid study group Abstract Background: Patients with SARS-CoV-2 infection are at higher risk for ventilator-associated pneumonia (VAP). No study has evaluated the relationship between VAP and mortality in this population, or compared this relationship between SARS-CoV-2 patients and other populations. The main objective of our study was to determine the relation‑ ship between VAP and mortality in SARS-CoV-2 patients. Methods: Planned ancillary analysis of a multicenter retrospective European cohort. VAP was diagnosed using clinical, radiological and quantitative microbiological criteria. Univariable and multivariable marginal Cox’s regression models, with cause-specific hazard for duration of mechanical ventilation and ICU stay, were used to compare out‑ comes between study groups. Extubation, and ICU discharge alive were considered as events of interest, and mortal‑ ity as competing event. Findings: Of 1576 included patients, 568 were SARS-CoV-2 pneumonia, 482 influenza pneumonia, and 526 no evidence of viral infection at ICU admission. VAP was associated with significantly higher risk for 28-day mortality in SARS-CoV-2 (adjusted HR 1.70 (95% CI 1.16–2.47), p = 0.006), and influenza groups (1.75 (1.03–3.02), p = 0.045), but not in the no viral infection group (1.07 (0.64–1.78), p = 0.79). VAP was associated with significantly longer duration of mechanical ventilation in the SARS-CoV-2 group, but not in the influenza or no viral infection groups. VAP was associated with significantly longer duration of ICU stay in the 3 study groups. No significant difference was found in *Correspondence: s-nseir@chru-lille.fr 1 Médecine Intensive‑Réanimation, CHU de Lille, F‑59000 Lille, France Full list of author information is available at the end of the article © The Author(s) 2021. 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. Nseir et al. Crit Care (2021) 25:177 Page 2 of 11 heterogeneity of outcomes related to VAP between the 3 groups, suggesting that the impact of VAP on mortality was not different between study groups. Interpretation: VAP was associated with significantly increased 28-day mortality rate in SARS-CoV-2 patients. How‑ ever, SARS-CoV-2 pneumonia, as compared to influenza pneumonia or no viral infection, did not significantly modify the relationship between VAP and 28-day mortality. Clinical trial registration: The study was registered at ClinicalTrials.gov, number NCT04359693. Keywords: Ventilator-associated pneumonia, Mortality, COVID-19 Background SARS-CoV-2 pneumonia, (2) influenza (A or B) pneumo- Coronavirus disease-19 (COVID-19) pandemic is ongo- nia, or (3) no viral infection. ing, and several million patients have been hospital- The Ethics Committee, and Institutional Review Boards ized in intensive care units (ICUs) worldwide. A large approved the study protocol (Comité de Protection des percentage of COVID-19 patients admitted to the ICU Personnes Ouest VI; approved by April 14, 2020; regis- require invasive mechanical ventilation, and are at higher tration number RIPH:20.04.09.60039) as minimal-risk risk for ventilator-associated pneumonia (VAP) [1]. In research using data collected for routine clinical prac- the large multicenter European coVAPid study [2], SARS- tice, and waived the requirement for informed consent. CoV-2 infection was associated with higher risk for VAP, The study was registered at ClinicalTrials.gov, number and ventilator associated tracheobronchitis (VAT), as NCT04359693. compared to patients with influenza, or no viral infec- SARS-CoV-2 infection was confirmed by positive poly- tion at ICU admission. Other recent studies [3–7] con- merase chain reaction (PCR) testing of a nasopharyngeal firmed these results and reported a high incidence of or respiratory secretions samples. Influenza pneumonia VAP ranging from 44 to 86%. This could be explained by was diagnosed based on a positive nasopharyngeal or air- the prolonged duration of mechanical ventilation in these way secretions PCR test. patients, high rate of acute respiratory distress syndrome Definitions (ARDS), common use of corticosteroids, and perhaps Ventilator‑associated lower respiratory tract infection specific factors related to severe acute respiratory syn- The diagnosis of VA-LRTI was based on the presence of drome Coronavirus 2 (SARS-CoV-2) infection. at least two of the following criteria: body temperature of Several studies, performed in general ICU populations more than 38.5 °C or less than 36.5 °C, leucocyte count and using different methods, showed higher mortality greater than 12,000 cells per μL or less than 4000 cells per rates, and longer duration of mechanical ventilation and μL, and purulent tracheal secretions [14, 15]. Addition- ICU stay in VAP patients, as compared with those with ally, all episodes of infection needed microbiological con- no VAP [8–13]. To our knowledge, no study to date has firmation, with the isolation in the endotracheal aspirate specifically addressed the effect of VAP on mortality in of at least ­105 colony-forming units (CFU) per mL, or in SARS-CoV-2 patients, or the impact of SARS-CoV-2 bronchoalveolar lavage of at least 1 ­ 04 CFU per mL. VAT infection on the relationship between VAP and mortality. was defined with the above-mentioned criteria with no Therefore, we conducted this planned ancillary study of radiographic signs of new pneumonia. VAP was defined the coVAPid European multicenter cohort to determine by the presence of new or progressive infiltrates on chest the impact of VAP on mortality in SARS-CoV-2 patients. X-ray. Only first episodes of VAT and VAP occurring We also aimed to determine the impact of SARS-CoV-2 more than 48 h after starting invasive mechanical ventila- pneumonia, as compared with influenza pneumonia or tion were analyzed. All VA-LRTI episodes were prospec- no viral infection at ICU admission, on the relationship tively identified, and chest X-rays were reviewed by at between VAP and mortality. least two attending physicians. In case of disagreement, a third physician was asked to interpret the radiograph. Methods Late-onset VAP was defined as VAP diagnosed after 4 Study design and population days of invasive mechanical ventilation [15]. Initial anti- This study is a planned ancillary analysis of the coVAPid biotic treatment was considered as appropriate when at cohort study. Briefly, coVAPid was a multicenter retro- least one antibiotic, matching the in vitro susceptibility of spective observational European cohort study. Eligibility the pathogen causing VAP, was given to treat this infec- criteria included age equal or above 18 years, the need tion [16]. for invasive mechanical ventilation for more than 48 h, and one of the following criteria at ICU admission: (1) Nseir et al. Crit Care (2021) 25:177 Page 3 of 11 Outcomes the event to patients who are at risk (without the event of The primary outcome of this ancillary study was 28-day interest and without the competing event for mechanical all-cause mortality. Secondary outcomes included dura- ventilation duration and length of ICU stay) [18]. tion of mechanical ventilation, and ICU length of stay HRs associated with VAT, VAP, and no VA-LRTI were censored at 28 days. derived as effect size using time period without VA-LRTI as reference. To assess whether the association between Statistical analysis VA-LRTI and outcomes differed between the three dis- Quantitative variables were expressed as medians (inter- ease groups, a heterogeneity test was performed [19], quartile range) and categorical variables were expressed comparing the HRs for VA-LRTI (VAP, VAT, VAT + VAP as numbers (percentage). Patient characteristics at ICU combined) between study groups. admission and during ICU stay were described accord- To avoid case-deletion in multivariate analyses due ing to study disease groups (SARS-CoV-2 pneumonia to presence of missing data in covariates, multivariable vs. Influenza pneumonia vs. no viral infection) without Cox’s models were performed after handling missing formal statistical comparisons. Patient’s outcomes (over- data on covariates by using multiple imputation proce- all survival, mechanical ventilation duration, length of dure [20]. This imputation was performed using regres- ICU stay) were described according to study disease sion-switching approach (chained equations with m = 20 groups using survival analysis approach by estimating the obtained) under the missing at random assumption using cumulative incidence of event of interest (death, extuba- all baseline characteristics (see Table 1), disease group tion alive and ICU discharge alive) censored at 28-days. and outcomes (event status and log of event time), with Cumulative incidence of death was estimated using a predictive mean matching method for quantitative Kaplan–Meier method, cumulative incidence of extuba- variables and logistic regression model (binary, ordi- tion alive and ICU discharge alive were estimated using nal or multinomial) for categorical variables. Estimates Kalbfleisch and Prentice method, considering death as obtained in the different imputed data sets were com- competing event [17]. bined using Rubin’s rules [21]. We assessed the effect of SARS-CoV-2 pneumonia Statistical testing was performed at the two-tailed α compared to the two other disease groups on patient’s level of 0.05. Data were analyzed using the SAS software outcomes using a univariable and multivariable marginal package, release 9.4 (SAS Institute, Cary, NC). Cox’s regression models to account clustered (center) data, with cause-specific hazard for mechanical ventila- Results tion duration and length of ICU stay (considering extuba- Patient characteristics tion alive and ICU discharge alive as event of interest, and In total, 1576 patients were included from March 2016 death as competing event). Pre-specified potential pre- through May 2020 (568 in SARS-CoV2, 482 in influ- dictors of patient outcomes (age, gender, SAPS II, Charl- enza, and 526 in no viral infection groups). 399 (25.3%) son score, McCabe classification, shock and ARDS) were VAP, and 167 (10.6%) VAT first episodes were diagnosed included as covariates into multivariable models. Hazard in study patients. 28-day mortality was 28.8% (164 of ratios (HRs) for SARS-CoV-2 pneumonia vs. Influenza 568 patients), 22% (125 of 482 patients), and 32.9% (173 pneumonia, and HRs for SARS-CoV-2 pneumonia vs. no of 526 patients) in SARS-CoV-2, influenza, and no viral viral infection were derived from Cox’s models as effect infection groups, respectively. Older age, higher SAPS size. A HR > 1 indicates a decrease in survival, mechani- II and SOFA score, comorbidities, septic shock, cardiac cal ventilation duration and length of ICU stay, whereas a arrest, and acute kidney injury rates at ICU admission HR < 1 indicates an increase in survival, mechanical ven- were more common in non-survivors, as compared to tilation duration and length of ICU stay. survivors in the 3 study groups. During ICU stay, per- In each disease group, we assessed the association of centage of patients who received corticosteroids, prone first episodes of VA-LRTI with patient’s outcomes using positioning, and ECMO was higher in non-survivors univariable and multivariable marginal Cox’s regression as compared to survivors. The characteristics of study models, with cause-specific hazard for mechanical venti- patients are presented in Tables 1, and 2. lation duration and length of ICU stay and by considering The cumulative incidence of 28-day mortality, extuba- the first VA-LRTI occurrence as a time-dependent covar- tion alive, and ICU discharge alive, according to study iate (3-levels categorical variable: No VA-LRTI, vs. VAT, groups, are presented in Additional file 1: Fig. 1. SARS- vs. VAP); as well as binary variable: No VA-LRTI vs. VA- CoV-2 infection was associated with significantly longer LRTI (VAT or VAP). This model accounted for exposure duration of mechanical ventilation, as compared to influ- time of VA-LRTI, by comparing at each follow-up time enza and no viral infection groups (Fig. 1). SARS-CoV-2 point, the current VA-LRTI status of patients who have Nseir et al. Crit Care (2021) 25:177 Page 4 of 11 Table 1 Patient characteristics at ICU admission according to disease group, and 28-day mortality SARS-CoV-2 pneumonia Influenza pneumonia No viral infection Alive (n = 402) Dead (n = 166) Alive (n = 350) Dead (n = 132) Alive (n = 344) Dead (n = 182) Age, ­yearsa 62 (53–70) 70 (62–78) 61 (52–70) 65 (54–72) 63 (52–72) 70 (60–76) Men 281/402 (69.9) 126/166 (75.9) 219/350 (62.6) 79/131 (60.3) 239/342 (69.9) 114/182 (62.6) Body mass index, kg/m2 b 28.7 (25.5–33.6) 29.1 (26.0–33.0) 27.5 (23.1–32.3) 27.6 (23.5–31.7) 26.3 (22.7–29.8) 26.9 (23.2–33.3) Severity scores SAPS ­IIc 38 (31–51) 48 (38–61) 48 (37–60) 58 (45–72) 51 (39–63) 63 (51–73) SOFA ­scored 6 (3–8) 7 (4–10) 8 (5–10) 10 (7–13) 8 (5–11) 9 (7–12) Comorbidity scores MacCabe classification Non-fatal 347/382 (90.8) 128/161 (79.5) 249/332 (75.0) 75/124 (60.5) 216/318 (67.9) 99/171 (57.9) Fatal < 5 years 33/382 (8.6) 29/161 (18.0) 78/332 (23.5) 36/124 (29.0) 90/318 (28.3) 47/171 (27.5) Fatal < 1 year 2/382 (0.5) 4/161 (2.5) 5/332 (1.5) 13/124 (10.5) 12/318 (3.8) 25/171 (14.6) Charlson Comorbidity ­Indexe 2 (1–3) 4 (2–5) 3 (2–5) 4 (2–6) 3 (2–5) 5 (3–6) Chronic diseases Diabetes mellitus 113/399 (28.3) 55/166 (33.1) 72/344 (20.9) 32/130 (24.6) 82/338 (24.3) 50/181 (27.6) Chronic renal failure 11/395 (2.8) 22/164 (13.4) 25/346 (7.2) 14/129 (10.9) 22/339 (6.5) 23/182 (12.6) Cardiovascular disease 59/396 (14.9) 44/164 (26.8) 80/346 (23.1) 37/130 (28.5) 72/337 (21.4) 62/181 (34.3) Chronic heart failure 12/394 (3.0) 9/164 (5.5) 19/345 (5.5) 18/130 (13.8) 24/336 (7.1) 26/182 (14.3) COPD 26/396 (6.6) 11/164 (6.7) 96/344 (27.9) 33/131 (25.2) 63/339 (18.6) 35/182 (19.2) Chronic respiratory failure 12/394 (3.0) 8/164 (4.9) 56/344 (16.3) 11/131 (8.4) 28/336 (8.3) 21/182 (11.5) Cirrhosis 6/395 (1.5) 2/164 (1.2) 10/345 (2.9) 6/130 (4.6) 18/335 (5.4) 18/181 (9.9) Immunosuppression 29/395 (7.3) 23/164 (14.0) 61/348 (17.5) 46/131 (35.1) 71/340 (20.9) 46/180 (25.6) Active smoking 20/396 (5.1) 9/164 (5.5) 122/346 (35.3) 27/130 (20.8) 102/337 (30.3) 35 /182 (19.2) Alcohol abuse 29/394 (7.4) 5/164 (3.0) 65/345 (18.8) 20/130 (15.4) 89/337 (26.4) 43/182 (23.6) Location before ICU admission Home 187/402 (46.5) 84/166 (50.6) 212/349 (60.7) 63/132 (47.7) 180/344 (52.3) 85/182 (46.7) Hospital ward 154/402 (38.3) 61/166 (36.7) 104/349 (29.8) 53/132 (40.2) 147/344 (42.7) 83/182 (45.6) Another ICU 61/402 (15.2) 21/166 (12.7) 33/349 (9.5) 16/132 (12.1) 17/344 (4.9) 14/182 (7.7) Admission category Medical 401/402 (99.8) 166/166 (100.0) 348/350 (99.4) 132/132 (100.0) 302/344 (87.8) 165/182 (90.7) Surgical 0/402 (0.0) 0/166 (0.0) 0/350 (0.0) 0/132 (0.0) 11/344 (3.2) 6/182 (3.3) Trauma 1/402 (0.2) 0/166 (0.0) 2/350 (0.6) 0/132 (0.0) 31/344 (9.0) 11/182 (6.0) Recent hospitalization (< 3 months) 26/401 (6.5) 18/165 (10.9) 41/348 (11.8) 31/131 (23.7) 90/342 (26.3) 58/182 (31.9) Recent antibiotic treatment (< 3 months) 50/402 (12.4) 24/165 (14.5) 56/347 (16.1) 39/130 (30.0) 56/342 (16.4) 47/182 (25.8) Causes for ICU admission Shock 58/394 (14.7) 44/163 (27.0) 144/343 (42.0) 66/127 (52.0) 147/336 (43.8) 97/179 (54.2) Acute respiratory failure 371/401 (92.5) 150/166 (90.4) 316/349 (90.5) 117/131 (89.3) 196/334 (58.7) 83/179 (46.4) ARDS 271/398 (68.1) 115/165 (69.7) 157/342 (45.9) 63/127 (49.6) 44/330 (13.3) 28/179 (15.6) Neurological failure 13/385 (3.4) 13/163 (8.0) 51/339 (15.0) 18/126 (14.3) 128/331 (38.7) 63/178 (35.4) Cardiac arrest 1/384 (0.3) 2/163 (1.2) 14/338 (4.1) 11/127 (8.7) 40/329 (12.2) 44/179 (24.6) Acute kidney injury 48/385 (12.5) 48/163 (29.4) 84/337 (24.9) 49/124 (39.5) 87/327 (26.6) 49/178 (27.5) Values are as no./No. (%) or median (interquartile range) McCabe classification of comorbidities and likelihood of survival, likely to survive > 5 years, 1–5 years, < 1 year; Chronic renal failure, KDOQI CKD classification stage 4 or 5 (creatinine clearance < 30 ml/mn); Chronic heart failure, NYHA class III or IV; Heart disease, ischemic heart disease or atrial fibrillation; Cirrhosis, Child–Pugh score B or C; Immunosuppression if haematological malignancy, allogenic stem cell transplant, organ transplant, HIV or immunosuppressive drugs; More than one cause for ICU admission is possible a 1 missing value in influenza group; b 160 missing values (SARS-CoV-2, n = 32; influenza, n = 68; controls, n = 60); c 87 missing values (SARS-CoV-2, n = 43; influenza, n = 21; controls, n = 21); d 27 missing values (SARS-CoV-2, n = 21; influenza, n = 4; controls, n = 2); e 50 missing values (SARS-CoV-2, n = 19; influenza, n = 11; controls, n = 20) Nseir et al. Crit Care (2021) 25:177 Page 5 of 11 Table 2 Patient characteristics during ICU stay according to disease groups and 28-day mortality SARS-CoV-2 pneumonia Influenza pneumonia No viral infection Alive (n = 402) Dead (n = 166) Alive (n = 350) Dead (n = 132) Alive (n = 344) Dead (n = 182) Antiviral treatment 226/401 (56.4) 96/165 (58.2) 319/349 (91.4) 118/132 (89.4) 19/343 (5.5) 5/180 (2.8) Oseltamivir 30/399 (7.5) 14/164 (8.5) 314/344 (91.3) 116/131 (88.5) 18/342 (5.3) 4/179 (2.2) Remdesivir 21/399 (5.3) 6/164 (3.7) 0/344 (0.0) 0/131 (0.0) 0/342 (0.0) 0/179 (0.0) Lopinavir-Ritonavir 96/399 (24.1) 51/164 (31.1) 0/344 (0.0) 0/131 (0.0) 0/342 (0.0) 0/179 (0.0) Lopinavir-Ritonavir + interferon 12/399 (3.0) 9/164 (5.5) 0/344 (0.0) 0/131 (0.0) 0/342 (0.0) 0/179 (0.0) Hydroxychloroquine 134/399 (33.6) 39/164 (23.8) 0/344 (0.0) 1/131 (0.8) 0/342 (0.0) 0/179 (0.0) Corticosteroids 131/380 (34.5) 71/162 (43.8) 124/345 (35.9) 58/130 (44.6) 97/343 (28.3) 64/182 (35.2) Hydrocortisone 27/377 (7.2) 32/160 (20.0) 64/343 (18.7) 43/130 (33.1) 39/340 (11.5) 41/180 (22.8) Dexamethasone 32/377 (8.5) 16/160 (10.0) 1/343 (0.3) 0/130 (0.0) 6/340 (1.8) 4/180 (2.2) Methylprednisolone 70/377 (18.6) 21/160 (13.1) 58/343 (16.9) 15/130 (11.5) 51/340 (15.0) 17/180 (9.4) Highest daily dose, ­mga 100 (67–133) 71 (50–133) 75 (50–100) 50 (50–100) 63 (50–100) 50 (50–75) Duration, ­daysb 6 (4–9) 6 (4–8) 5 (3–9) 6 (3–9) 4 (2–7) 4 (3–8) Antibiotic treatment 363/378 (96.0) 143/153 (93.5) 309/330 (93.6) 125/128 (97.7) 268/323 (83.0) 147/173 (85.0) Duration, days 7 (5–9) 7 (4–10) 7 (5–11) 7 (4–9) 7 (4–9) 6 (4–9) Prone positioning 263/401 (65.6) 120/166 (72.3) 96/349 (27.5) 55/132 (41.7) 33/340 (9.7) 30/182 (16.5) ECMO 39/402 (9.7) 22/165 (13.3) 38/349 (10.9) 22/131 (16.8) 2/341 (0.6) 3/182 (1.6) Vales are no./No. (%) or median (interquartile range) ECMO, Extracorporeal Membrane Oxygenation; ICU, Intensive Care Unit; MV, Mechanical Ventilation a 11 missing values (SARS-CoV-2, n = 4; influenza, n = 4; controls, n = 3); b16missing values (SARS-CoV-2, n = 7; influenza, n = 3; controls, n = 6) Fig. 1 Unadjusted and Adjusted hazard ratios for 28-day mortality, extubation alive and ICU discharge alive, associated with SARS-CoV-2 pneumonia, versus influenza pneumonia and no viral infection groups. HRs were calculated using cause-specific proportional hazard models, by considering mortality as competing event for MV duration, and length of ICU stay. Adjusted HRs were calculated, including age, gender, simplified acute physiology score II, Charlson score, MacCabe classification, shock, and acute respiratory distress syndrome as pre-specified covariates in Cox’s models (after handling missing values by multiple imputation). A HR > 1 indicates a decrease in survival duration (i.e. an increased risk for mortality), MV duration (i.e. an increased risk for extubation alive) and ICU length of stay (i.e. an increased risk for discharge alive) and a HR < 1 indicates an increase in survival duration (i.e. a decreased risk for mortality), MV duration (i.e. a decreased risk for extubation alive) and ICU length of stay (i.e. a decreased risk for discharge alive). Note that the event of interest for survival is a pejorative event (death) whereas for MV duration and ICU length of stay, the event of interest is a positive event (extubation or discharge alive). Consequently, the detrimental effect of SARS-CoV-2 pneumonia (vs influenza pneumonia and no viral infection groups) was associated with a HR > 1 for overall survival but was associated with a HR < 1 for MV duration and ICU length of stay. HR, hazard ratio; ICU, intensive care unit; MV, mechanical ventilation Nseir et al. Crit Care (2021) 25:177 Page 6 of 11 infection was also associated with significantly higher with MDR was lower in SARS-CoV-2 patients as com- risk for mortality, as compared to influenza group (Fig. 1). pared to the two other groups, as well as percentage of Characteristics of patients with ventilator‑associated patients with late-onset VAP related to MDR (35 of 167 pneumonia patients (21%), 27 of 78 patients (34.6%), and 20 of 57 Antimicrobial treatment was prescribed to 191 of 205 patients (35.1%), in SARS-CoV-2, influenza, and no viral (93.2%), 98 of 105 (93.3), and 82 of 87 (94.3) patients infection groups; respectively). with VAP in SARS-CoV-2, influenza, and no viral infec- tion groups; respectively. Initial antibiotic treatment was Primary and secondary study outcomes appropriate in 145 of 200 (72.5%), 69 of 102 (67.7%), and VAP was associated with significantly higher risk for 54 of 87 (62.1%) VAP patients in SARS-CoV-2, influenza, 28-day mortality in SARS-CoV-2, and influenza groups, and no viral infection group; respectively. Median (inter- but not in no viral infection group (Fig. 2A). However, quartile range) time from starting invasive mechani- the heterogeneity test showed no significant difference in cal ventilation to VAP occurrence was 9 (6, 13) days in the strength of association between VAP and mortality SARS-CoV-2, 9 (5, 13) days in influenza, and 7 (4, 12) across the 3 study groups. days in no viral infection groups. Percentage of patients VAP was associated with significantly longer duration with late-onset VAP was 82.4% (169 of 200 patients), of mechanical ventilation in the SARS-CoV-2 group, but 73.8% (79 of 102 patients), and 65.5% (57 of 87 patients) not in influenza or no viral infection groups. The hetero- in SARS-CoV-2, influenza, and no viral infection group; geneity test showed no significant difference between respectively. Pseudomonas aeruginosa, Enterobacter spp., the 3 groups regarding the association between VAP and and Klebsiella spp. were the most frequently identified duration of mechanical ventilation (Fig. 2B). microorganisms (Table 3). Percentage of VAP patients Table 3 Microorganisms responsible for ventilator-associated pneumonia SARS-CoV-2 pneumonia (n = 205) Influenza pneumonia (n = 107) No viral infection (n = 87) Gram-positive cocci MSSA 20 (9.8) 5 (4.7) 8 (9.2) MRSA 6 (2.9) 4 (3.7) 2 (2.3) Enterococcus spp. 7 (3.4) 2 (1.9) 2 (2.3) Streptococcus pneumoniae 7 (3.4) 1 (1) 2 (2.3) Streptococcus spp. 1 (0.5) 0 (0) 0 (0) Other 0 (0) 2 (1.9) 3 (3.5) Gram-negative bacilli Pseudomonas aeruginosa 51 (24.9) 26 (24.3) 15 (17.2) Enterobacter spp. 37 (18) 15 (14) 12 (13.8) Klebsiella spp. 26 (12.7) 17 (15.9) 12 (13.8) Escherichia coli 19 (9.2) 8 (7.5) 5 (5.7) Acinetobacter baumannii 9 (4.4) 16 (15) 10 (11.5) Stenotrophomonas maltophilia 4 (2) 2 (1.9) 4 (4.6) Serratia marcescens 9 (4.4) 4 (3.7) 1 (1.1) Citrobacter freundii 6 (2.9) 1 (1) 1 (1.1) Citrobacter spp. 5 (2.4) 3 (2.8) 3 (3.5) Proteus mirabilis 5 (2.4) 1 (1) 2 (2.3) Haemophilus influenza 3 (1.5) 5 (4.7) 5 (5.7) Morganella morganii 2 (1) 3 (2.8) 0 (0) Other 26 (12.7) 8 (7.5) 5 (5.7) Polymicrobial 24 (11.7) 8 (7.5) 6 (6.9) Multidrug-resistant isolates* 42 (20.7) 42 (40) 27 (31) Data are presented as N (%) * Missing data: 2, 2 in SARS-CoV-2 and Influenza groups; respectively MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-sensitive Staphylococcus aureus Nseir et al. Crit Care (2021) 25:177 Page 7 of 11 (See figure on next page.) Fig. 2 Association between ventilator-associated lower respiratory tract infections and outcomes. a 28-Day mortality. b Duration of mechanical ventilation. c Length of ICU stay. HRs were calculated using cause-specific proportional hazard models, considering the first VA-LRTI as a time dependent 3-levels categorical variable (No VA-LRTI vs. VAT vs. VAP). Adjusted HRs were calculated including age, gender, simplified acute physiology score II, Charlson score, MacCabe classification, shock, and acute respiratory distress syndrome as pre-specified covariables in Cox’s model. Since the event of interest for 28-Day mortality is a pejorative event (death), whereas for MV duration and ICU length of stay, the event of interest is a positive event (extubation or discharge alive), the detrimental effect of SARS-CoV-2 pneumonia (vs influenza pneumonia and no viral infection groups) was associated with a HR > 1 for 28-Day mortality, with a HR < 1 for MV duration and ICU length of stay. HR, hazard ratio; ICU, intensive care unit; MV, mechanical ventilation; VA-LRTI, ventilator-associated respiratory tract infection; VAP, ventilator-associated pneumonia; VAT, ventilator-associated tracheobronchitis VAP was associated with significantly longer duration on individual data from 6284 patients included in rand- of ICU stay in all study groups. The heterogeneity test omized controlled trials of VAP prevention [8]. The over- showed no significant difference between the 3 groups all attributable mortality of VAP was 13%, with higher regarding the association between VAP and ICU length rates for surgical patients and patients with a mid-range of stay (Fig. 2C). severity score at admission. Attributable mortality was VAT was not significantly associated with increased mainly caused by prolonged exposure to the risk of dying risk for mortality, or longer duration of mechanical ven- due to increased length of ICU stay. However, other tilation in the 3 study groups. VAT was associated with studies suggested that mortality attributable to VAP was significantly longer ICU stay in no viral infection group, small [26, 27]. but not in the two other groups. The heterogeneity test was significant between the 3 groups regarding duration Impact of SARS‑CoV‑2 infection on the relationship of ICU stay (Fig. 2a–c). between VAP and mortality Descriptive characteristics and outcomes of patients Multidrug resistant bacteria (MDR) bacteria and inap- with, or without, VA-LRTI are presented in Additional propriate initial antibiotic treatment are well-known risk file 1: Tables 1, and 2. The association between VA-LRTI factors for mortality in VAP patients [28, 29]. Although and outcomes, according to study groups, is presented in the incidence of MDR, and inappropriate initial antibi- Additional file 1: Fig. 2. otic treatment was lower in COVID-19 patients, as com- pared to the two other groups, the relationship between Discussion VAP and mortality was only significant in COVID-19 Main results patients. This suggests that SARS-CoV-2 infection and The main results of our study are that VAP is associated specific pulmonary lesions might play a role in the sever- with increased 28-day mortality rate and longer dura- ity and outcome of VAP in this population. Neverthe- tion of mechanical ventilation and ICU length of stay in less, the absence of significant heterogeneity between the SARS-CoV-2 patients. SARS-CoV-2 infection, compared three groups suggests that SARS-CoV-2 infection has no with influenza or no viral infection, has no significant significant impact on the relationship between VAP and impact on the relationship between VAP and 28-day mortality. At least three explanations could be provided mortality, neither on the relationship between VAP and for this result. First, potential confounders might have duration of mechanical ventilation, or ICU length of stay. influenced our results. However, careful adjustment was Relationship between VAP and mortality in SARS‑CoV‑2 performed on SAPS II, age, gender, comorbidities, sep- patients tic shock, and ARDS, based on the results of prior stud- Few data are available on mortality rate in SARS-CoV-2 ies [13, 30]. Second, the number of patients with VAP patients with VAP. Although several recent studies evalu- was relatively small in the no viral infection group, and ated the incidence of VAP in these patients, only one the study might have been underpowered to detect a study performed in ARDS patients requiring ECMO significant effect. Third, SARS-CoV-2 infection could [6], reported on mortality rate (30%) in this population. be associated with similar impact on the relationship However, this study was performed in a single center, the between VAP and mortality. Previous studies suggested number of patients with SARS-CoV-2 infection was small that COVID-19 patients have similar outcomes as other (n = 50), and no comparison was performed with mortal- patients with similar type of acute illness. For example, ity rate in SARS-CoV-2 patients with no VAP. a recent study reported a 90-day mortality rate of 37% Previous studies, performed in general ICU popu- in COVID-19 ARDS [1], which is in line with previous lations, showed an increased mortality rate in VAP results in non-COVID ARDS [31]. patients [22–25]. A large meta-analysis was performed Nseir et al. Crit Care (2021) 25:177 Page 8 of 11 Nseir et al. Crit Care (2021) 25:177 Page 9 of 11 Strengths and limitations Abbreviations COVID: Coronavirus disease; HR: Hazard ratio; ICU: Intensive care unit; MV: To our knowledge, our study is the first to evaluate the Mechanical ventilation; SARS-CoV-2: Severe acute respiratory syndrome relationship between VAP and mortality in SARS-CoV-2 Coronavirus 2; VA-LRTI: Ventilator-associated respiratory tract infection; VAP: patients. Strengths of this study are the large number of Ventilator-associated pneumonia; VAT: Ventilator-associated tracheobronchitis. included patients, the multicenter design, and the two con- trol groups including patients with influenza pneumonia Supplementary Information or no viral infection. In addition, we carefully adjusted for The online version contains supplementary material available at https://​doi.​ org/​10.​1186/​s13054-​021-​03588-4. potential confounders, using competing risk analyses, and Kalbfleisch and Prentice methods. Cox’s model took into Additional file 1. Online supplementay data. account the immortal time bias, by considering VAP as time-dependent covariable [32]. The discrepancy between Acknowledgements the lower mortality rate in SARS-CoV-2 patients with VAP, In addition to the authors, the coVAPid study group includes the fol- as compared with those with no VA-LRTI (Additional lowing collaborators: Raphaël Favory: CHU de Lille, Centre de Réanimation, file 1: Table 2), and the adjusted hazard ratio showing an F-59000 Lille, France. Sébastien Préau: CHU de Lille, Centre de Réanimation, F-59000 Lille, France. Mercé Jourdain: CHU de Lille, Centre de Réanimation, increased risk for mortality in the former than in the latter F-59000 Lille, France. Julien Poissy: CHU de Lille, Centre de Réanimation, group (Fig. 2 A) is explained by the immortal time bias in F-59000 Lille, France. Piehr Saint Leger: Service de Réanimation polyvalente, the calculation of actual mortality rate. Centre Hospitalier de Valenciennes, Valenciennes, France. Thierry Van der Lin‑ den: Médecine intensive réanimation, Hôpital Saint Philibert GHICL, Université Some limitations of this study should be acknowledged. catholique, Lille, France. Anne Veinstein: CHU de Poitiers, Médecine Intensive First, coVAPid cohort was retrospective. However, VAP Réanimation, CIC 1402 ALIVE, Université de Poitiers, Poitiers, France. Elie was prospectively identified in all centers. Further, the pres- Azoulay: Service de médecine intensive et réanimation, hôpital Saint-Louis, 1 Avenue Claude Vellefaux, 75010, Paris, France. Frédéric Pene: Medical Intensive ence of new infiltrate on chest X-ray was evaluated by at Care Unit, Cochin Hospital, AP-HP. Centre, & Université de Paris, Paris, France. least two physicians. Second, the study was performed in Maelle Martin: Service de Médecine Intensive Réanimation, CHU de Nantes, Europe, and the results may not be generalized to other Nantes, France. Keyvan Razazi: Assistance Publique-Hôpitaux de Paris, CHU Henri Mondor, Service de réanimation médicale, Créteil, France. Gaëtan Plant‑ world regions. Third, number of patients was relatively efeve: Réanimation Polyvalente, CH Victor Dupouy, Argenteuil, France. Muriel small in some subgroups with VAP or VAT, and our study Fartoukh: Sorbonne Université, Assistance Publique-Hôpitaux de Paris, Service might have been underpowered to detect differences de Médecine Intensive Réanimation, Hôpital Tenon, Paris, France. Didier Thevenin: Réanimation polyvalente CH Lens, France. Bertrand Guidet: Service regarding some secondary outcomes. Fourth, in spite of de Médecine Intensive Réanimation, AP-HP, Hôpital Saint-Antoine, Assistance careful adjustment, our analysis might have missed some Publique-Hôpitaux de Paris, 184 rue du Faubourg Saint-Antoine, Paris, 75571 residual confounders. Fifth, the interpretation of chest-x Cedex 12, France. Nicolas Weiss: Sorbonne Université, AP-HP, Hôpital de la Pitié-Salpêtrière, Département de Neurologie, Unité de Médecine Intensive ray is a difficult task in patients with pneumonia or ARDS. Réanimation Neurologique, Paris, France. Achille Kouatchet: Département de In spite of evaluation of chest-X ray by at least two phy- Médecine Intensive-Réanimation, CHU d’Angers, Université d’Angers, 4 rue sicians to confirm the presence of new infiltrate, some Larrey, 49933, Angers Cedex 9, France. Charlotte Salmon: Service de Médecine Intensive Réanimation, CHU de Tours, Hôpital Bretonneau, 2 Bd Tonnellé, patients with VAP could have been misclassified in the 37000 Tours, France. Guillaume Brunin: Service de réanimation, hôpital Duch‑ VAT group. Further, some patients with VAT (12.6%) devel- enne, rue Monod, 62200 Boulogne-sur-Mer, France. Safaa Nemlaghi: Service oped subsequent VAP. However, we repeated the analysis de Pneumologie, Médecine intensive - Réanimation (Département "R3S"), AP-HP, Sorbonne Université, Groupe Hospitalier Universitaire Pitié-Salpêtrière on the relationship between VAP and mortality, includ- Charles Foix, 75013, Paris, France. David Meguerditchian: Intensive Care Unit, ing VAT patients and the results did not differ (Additional Pellegrin-Tripode Hospital, University Hospital of Bordeaux, Bordeaux, France. file 1: Fig. 2). Sixth, no information was collected on timing Laurent Argaud: Service de Médecine Intensive - Réanimation, Hospices Civils de Lyon, Hôpital Edouard Herriot, 5, place d’Arsonval, 69437, Lyon of appropriate initial antibiotic treatment. Finally, patients Cedex 03, France. Sebastian Voicu: Réanimation Médicale et Toxicologique, were only followed-up until day 28. Therefore, outcomes at Hôpital Lariboisière, Université de Paris, INSERM UMRS-1144, Paris, France. day 60, or day 90 could not be evaluated. Charles-Edouard Luyt: Service de Médecine Intensive Réanimation, Institut de Cardiologie, Hôpital Pitié-Salpêtrière, Assistance Publique-Hôpitaux de Paris (APHP), Sorbonne Université, 47-83, Boulevard de l’Hôpital, 75651, Paris Cedex Conclusions 13, France. Benjamin Kowalski: Service de réanimation et de soins intensifs, To conclude, our results suggest that VAP is associated centre hospitalier de Douai, route de Cambrai, Douai, France. Edgar Moglia: Critical Care Department, Hospital Universitari Parc Taulí, Sabadell, Spain. Luis with increased 28-day mortality rate and longer dura- Morales: Intensive Care Unit, Hospital Universitari Sagrat Cor, and Universidad tion of mechanical ventilation and ICU length of stay in Autonoma de Barcelona, Barcelona, Spain Intensive Care Unit, Hospital Unive SARS-CoV-2 patients. However, SARS-CoV-2 infection, rsitari Sagrat Cor, Barcelona, Spain; Institut d’Investigacio i Innovacio Parc Tauli, Sabadell, Spain. Antonia Koutsoukou: 1st Department of Pulmonary Medicine compared with influenza or no viral infection, has no and Intensive Care Unit, National and Kapodistrian University of Athens, significant impact on the relationship between VAP and "Sotiria" Chest Hospital, Athens, Greece. Spyros D. Mentzelopoulos: 1st Depart‑ 28-day mortality, or on the relationship between VAP and ment of Intensive Care Medicine, National and Kapodistrian University of Ath‑ ens Medical School, Evaggelismos Hospital, Athens, Greece. David Nora: Poly‑ duration of mechanical ventilation, and ICU length of valent Intensive Care Unit, São Francisco Xavier Hospital, Centro Hospitalar de stay. Further studies are needed to confirm our findings. Lisboa Ocidental, Lisbon, Portugal. Sean Boyd: 3Department of Intensive Care Medicine, Multidisciplinary Intensive Care Research Organization (MICRO), Nseir et al. Crit Care (2021) 25:177 Page 10 of 11 St. James’s Hospital, St. James Street, Dublin 8, Dublin, Eire, Ireland. Julien Pitié‑Salpêtrière, Assistance Publique-Hôpitaux de Paris (APHP), Sorbonne Maizel: Medical ICU, Amiens university hospital, Amiens, France. Pierre Cuchet: Université, 47‑83, Boulevard de L’Hôpital, 75651 Paris Cedex 13, France. 15 ICU, Department of Medical Intensive Care, Caen University Hospital, F-14000 Roubaix Hospital, Roubaix, France. 16 Médecine Intensive Réanimation, Hôpital Caen, France. Quentin Delforge: Service de Médecine Intensive Réanimation, Saint Philibert GHICL, Université Catholique, Lille, France. 17 Réanimation Médi‑ Centre hospitalier de Béthune. Jean-Pierre Quenot: Department of Intensive cale Et Toxicologique, Hôpital Lariboisière, Université de Paris, INSERM UMRS- Care, François Mitterrand University Hospital, Dijon, France. Déborah Boyer: 1144, Paris, France. 18 Service de Médecine Intensive Et Réanimation, Hôpital 42Medical Intensive Care Unit, Rouen University Hospital, Rouen, France. Catia Saint-Louis, 1 Avenue Claude Vellefaux, 75010 Paris, France. 19 Critical Care Cilloniz: Pulmonology Department, Hospital Clinic of Barcelona, Ciber de Department, Hospital Universitari Parc Taulí, Sabadell, Spain. 20 1St Department Enfermedades Respiratorias (Ciberes, CB06/06/0028)-Institut d’Investigacions of Intensive Care Medicine, National and Kapodistrian University of Athens Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain. Medical School, Evaggelismos Hospital, Athens, Greece. 21 Medical Intensive Care Unit, Cochin Hospital, AP‑HP. Centre, Université de Paris, Paris, France. 22 Authors’ contributions Department of Pulmonology, Hospital Clinic Barcelona, University of Barce‑ AR, IML, PP, JL, AD, and SN designed the study. JL, and AD performed the lona, IDIBAPS, CIBERES, ICREA, Barcelona, Spain. 23 1St Department of Pulmo‑ statistical analyses. All authors collected study data. The first draft was written nary Medicine and Intensive Care Unit, National and Kapodistrian University by SN, and AR. All authors participated in writing, and revising the manuscript. of Athens, “Sotiria” Chest Hospital, Athens, Greece. 24 Réanimation Polyvalente, CH Lens, Lens, France. 25 Service de Médecine Intensive Réanimation, CHU de Funding Nantes, Nantes, France. 26 Service de Médecine Intensive Réanimation, CHU This study was supported in part by a grant from the French government de Tours, Hôpital Bretonneau, 2 Bd Tonnellé, 37000 Tours, France. 27 Service through the «Programme Investissement d’Avenir» (I-SITE ULNE) managed by de Médecine Intensive ‑ Réanimation, Hospices Civils de Lyon, Hôpital the Agence Nationale de la Recherche (coVAPid project). The funders of the Edouard Herriot, 5, place d’Arsonval, 69437 Lyon Cedex 03, France. 28 Service study had no role in the study design, data collection, analysis, or interpreta‑ de Médecine Intensive Réanimation, AP‑HP, Hôpital Saint‑Antoine, Assistance tion, writing of the report, or decision to submit for publication. Publique-Hôpitaux de Paris, 184 rue du Faubourg Saint‑Antoine, 75571 Paris Cedex 12, France. 29 Sorbonne Université, Assistance Publique‑Hôpitaux de Availability of data and materials Paris, Service de Médecine Intensive Réanimation, Hôpital Tenon, Paris, France. 30 All data needed to evaluate the conclusions in this Article are present and Réanimation Polyvalente, CH Victor Dupouy, Argenteuil, France. 31 Service de tabulated in the main text or the appendix. This article is the result of an origi‑ Pneumologie, Médecine Intensive ‑ Réanimation (Département “R3S”), AP‑HP, nal retrospective cohort. For individual de-identified raw data that underlie Sorbonne Université, Groupe Hospitalier Universitaire Pitié‑Salpêtrière Charles the results reported in this article, please contact the corresponding author. Foix, INSERM, UMRS1158 Neurophysiologie Respiratoire Expérimentale Et Clinique, Paris, France. 32 Département de Médecine Intensive‑Réanimation, CHU D’Angers, Université D’Angers, 4 rue Larrey, 49933 Angers Cedex 9, Declarations France. 33 Intensive Care Unit, Pellegrin‑Tripode Hospital, University Hospi‑ tal of Bordeaux, Bordeaux, France. 34 CHU de Poitiers, Médecine Intensive Ethics approval and consent to participate Réanimation, CIC 1402 ALIVE, Université de Poitiers, Poitiers, France. 35 APHP, The Ethics Committee, and Institutional Review Boards approved the study CHU Henri Mondor, Service de Médecine Intensive RéanimationUniversité protocol (Comité de Protection des Personnes Ouest VI; approved by April 14, Paris Est‑Créteil, Faculté de Santé, Groupe de Recherche Clinique CARMASIN‑ 2020; registration number RIPH:20.04.09.60039) as minimal-risk research using SERM U955, Institut Mondor de Recherche Biomédicale, 94010 Créteil, France. data collected for routine clinical practice, and waived the requirement for 36 Service de Médecine Intensive Réanimation, Centre Hospitalier de Béthune, informed consent. Réseau de Recherche Boréal, 62408 Béthune, France. 37 Service de Réani‑ mation, Hôpital Duchenne, Rue Monod, 62200 Boulogne‑sur‑Mer, France. Consent for publication 38 Sorbonne Université, AP‑HP, Hôpital de La Pitié‑Salpêtrière, Département Not applicable. de Neurologie, Unitéde Médecine Intensive Réanimation Neurologique, Paris, France. 39 Intensive Care Unit, Hospital Universitari Sagrat Cor, and Ciber de Competing interests Enfermedades Respiratorias (Ciberes, CB06/06/0028)-Institut D’Investigacions AR received personal fees from MaatPharma, IML received personal fees from Biomèdiques August Pi I Sunyer (IDIBAPS), Barcelona, Spain. 40 Critical Care MSD, and Gilead. AA received personal fees from Lilly Foundation, and grants Center, Corporacion Sanitaria Universitaria Parc Tauli, CIBER Enfermedades Res‑ from Grifols and Fischer & Paykel. CEL received personal fees from Bayer, piratorias, Autonomous University of Barcelona, Parc Tauli 1, 08028 Sabadell, Merck, Aerogen, Biomérieux, ThermoFischer Brahms, and Carmat. SN received Spain. 41 Univ. Lille, CHU Lille, ULR 2694-METRICS: Évaluation Des Technologies personal fees from MSD, Bio Rad, BioMérieux, Gilead, and Pfizer. All other de Santé Et Des Pratiques Médicales, 59000 Lille, France. authors declare no competing interests. Received: 22 February 2021 Accepted: 27 April 2021 Author details 1 Médecine Intensive‑Réanimation, CHU de Lille, F‑59000 Lille, France. 2 Inserm U1285, CNRS, UMR 8576‑UGSF‑Unité de Glycobiologie Structurale et Fonc‑ tionnelle, Univ. Lille, Lille, France. 3 Department of Intensive Care Medicine, Multidisciplinary Intensive Care Research Organization (MICRO), St. James’s References Hospital, St. James Street, Dublin 8, Dublin, Eire, Ireland. 4 Hospital Clinic, 1. COVID-ICU Group on behalf of the REVA Network and the COVID-ICU IDIBAPS, Universided de Barcelona, CIBERes, Barcelona, Spain. 5 Polyvalent Investigators. Clinical characteristics and day-90 outcomes of 4244 criti‑ Intensive Care Unit, São Francisco Xavier Hospital, Centro Hospitalar de Lisboa cally ill adults with COVID-19: a prospective cohort study. Intensive Care Ocidental, and NOVA Medical School, CHRC, New University of Lisbon, Lisbon, Med. 2021;47:60–73. Portugal. 6 Center for Clinical Epidemiology and Research Unit of Clinical 2. Rouzé A, Martin-Loeches I, Povoa P, Makris D, Artigas A, Bouchereau M, Epidemiology, OUH Odense University Hospital, Odense, Denmark. 7 Medical Lambiotte F, Metzelard M, Cuchet P, Boulle Geronimi C, Labruyere M, ICU, Amiens University Hospital, Amiens, France. 8 Department of Medical Tamion F, Nyunga M, Luyt C-E, Labreuche J, Pouly O, Bardin J, Saade A, Intensive Care, Caen University Hospital, 14000 Caen, France. 9 Service de Asfar P, Baudel J-L, Beurton A, Garot D, Ioannidou I, Kreitmann L, Llitjos J-F, Réanimation Polyvalente, Centre Hospitalier de Valenciennes, Valenciennes, Magira E, Mégarbane B, Meguerditchian D, Moglia E, Mekontso-Dessap France. 10 Medical Intensive Care Unit, Rouen University Hospital, Normandie A, et al. Relationship between SARS-CoV-2 infection and the incidence Université, UNIROUEN, Inserm U1096, FHU-REMOD-VHF, 76000 Rouen, France. 11 of ventilator-associated lower respiratory tract infections: a European Department of Intensive Care, François Mitterrand University Hospital, multicenter cohort study. Intensive Care Med. 2021;47:188–98. Dijon, France. 12 Intensive Care Unit, University Hospital of Larissa, University 3. Maes M, Higginson E, Pereira-Dias J, Curran MD, Parmar S, Khokhar of Thessaly, 41110 Biopolis Larissa, Greece. 13 Service de Réanimation Et de F, Cuchet-Lourenço D, Lux J, Sharma-Hajela S, Ravenhill B, Hamed I, Soins Intensifs, Centre Hospitalier de Douai, Route de Cambrai, Douai, France. 14 Heales L, Mahroof R, Solderholm A, Forrest S, Sridhar S, Brown NM, Service de Médecine Intensive Réanimation, Institut de Cardiologie, Hôpital Baker S, Navapurkar V, Dougan G, Bartholdson Scott J, Conway Morris A. Nseir et al. Crit Care (2021) 25:177 Page 11 of 11 Ventilator-associated pneumonia in critically ill patients with COVID-19. 17. Austin PC, Lee DS, Fine JP. Introduction to the analysis of survival data in Crit Care. 2021;25:25. the presence of competing risks. Circulation. 2016;133:601–9. 4. Pickens CO, Gao CA, Cuttica M, Smith SB, Pesce L, Grant R, Kang M, 18. Therneau T-M, Grambsch P-M. Modeling survival data: extending the Cox Morales-Nebreda L, Bavishi AA, Arnold J, Pawlowski A, Qi C, Budinger model. New-York: Springer; 2000. GS, Singer BD, Wunderink RG, Investigators for the NC. Bacterial 19. Willke RJ, Zheng Z, Subedi P, Althin R, Mullins CD. From concepts, theory, superinfection pneumonia in SARS-CoV-2 respiratory failure. medRxiv and evidence of heterogeneity of treatment effects to methodological 2021:2021.01.12.20248588. approaches: a primer. BMC Med Res Methodol. 2012;12:185. 5. Razazi K, Arrestier R, Haudebourg AF, Benelli B, Carteaux G, Decousser J, 20. van Burren S, Groothuis-Oudshoorn K. Multivariate imputation by Fourati S, Woerther PL, Schlemmer F, Charles-Nelson A, Botterel F, de Prost chained equations in R. 2011. J Stat Softw 2011;45:1–24. N, Mekontso Dessap A. Risks of ventilator-associated pneumonia and 21. Gladitz J, Rubin DB. Multiple imputation for nonresponse in surveys. Biom invasive pulmonary aspergillosis in patients with viral acute respiratory J. 1989;31:131–2. distress syndrome related or not to Coronavirus 19 disease. Crit Care. 22. Makris D, Desrousseaux B, Zakynthinos E, Durocher A, Nseir S. The impact 2020;24:699. of COPD on ICU mortality in patients with ventilator-associated pneumo‑ 6. Luyt C-E, Sahnoun T, Gautier M, Vidal P, Burrel S, Pineton de Chambrun M, nia. Respir Med. 2011;105:1022–9. Chommeloux J, Desnos C, Arzoine J, Nieszkowska A, Bréchot N, Schmidt 23. Bercault N, Boulain T. Mortality rate attributable to ventilator-associated M, Hekimian G, Boutolleau D, Robert J, Combes A, Chastre J. Ventilator- nosocomial pneumonia in an adult intensive care unit: a prospective associated pneumonia in patients with SARS-CoV-2-associated acute case-control study. Crit Care Med. 2001;29:2303–9. respiratory distress syndrome requiring ECMO: a retrospective cohort 24. Martin-Loeches I, Povoa P, Rodríguez A, Curcio D, Suarez D, Mira J-P, Cor‑ study. Ann Intensive Care. 2020;10:158. dero ML, Lepecq R, Girault C, Candeias C, Seguin P, Paulino C, Messika J, 7. Llitjos J-F, Bredin S, Lascarrou J-B, Soumagne T, Cojocaru M, Leclerc M, Castro AG, Valles J, Coelho L, Rabello L, Lisboa T, Collins D, Torres A, Salluh Lepetit A, Gouhier A, Charpentier J, Piton G, Faron M, Stoclin A, Pène F. J, Nseir S. Incidence and prognosis of ventilator-associated tracheobron‑ Increased susceptibility to intensive care unit-acquired pneumonia in chitis (TAVeM): a multicentre, prospective, observational study. Lancet severe COVID-19 patients: a multicentre retrospective cohort study. Ann Respir Med. 2015;3:859–68. Intensive Care. 2021;11:20. 25. Melsen WG, Rovers MM, Bonten MJM. Ventilator-associated pneumonia 8. Melsen WG, Rovers MM, Groenwold RHH, Bergmans DCJJ, Camus C, and mortality: a systematic review of observational studies. Crit Care Bauer TT, Hanisch EW, Klarin B, Koeman M, Krueger WA, Lacherade JC, Lor‑ Med. 2009;37:2709–18. ente L, Memish ZA, Morrow LE, Nardi G, van Nieuwenhoven CA, O’Keefe 26. Bekaert M, Timsit JF, Vansteelandt S, Depuydt P, Vésin A, Garrouste-Orgeas GE, Nakos G, Scannapieco FA, Seguin P, Staudinger T, Topeli A, Ferrer M, M, Decruyenaere J, Clec’h C, Azoulay E, Benoit D. Attributable mortality of Bonten MJM. Attributable mortality of ventilator-associated pneumonia: ventilator-associated pneumonia: a reappraisal using causal analysis. Am a meta-analysis of individual patient data from randomised prevention J Respir Crit Care Med. 2011;184:1133–9. studies. Lancet Infect Dis. 2013;13:665–71. 27. Steen J, Vansteelandt S, De Bus L, Depuydt P, Gadeyne B, Benoit DD, 9. Safdar N, Dezfulian C, Collard HR, Saint S. Clinical and economic conse‑ Decruyenaere J. Attributable mortality of ventilator-associated pneu‑ quences of ventilator-associated pneumonia: a systematic review. Crit monia: replicating findings, revisiting methods. Ann Am Thorac Soc. Care Med. 2005;33:2184–93. 2021;18(5):830–7. 10. Nseir S, Di Pompeo C, Soubrier S, Cavestri B, Jozefowicz E, Saulnier F, 28. Barbier F, Lisboa T, Nseir S. Understanding why resistant bacteria are Durocher A. Impact of ventilator-associated pneumonia on outcome in associated with higher mortality in ICU patients. Intensive Care Med. patients with COPD. Chest. 2005;128:1650–6. 2016;42(12):2066–9. 11. Rouzé A, Boddaert P, Martin-Loeches I, Povoa P, Rodriguez A, Ramdane 29. Martin-Loeches I, Torres A, Povoa P, Zampieri FG, Salluh J, Nseir S, Ferrer M, N, Salluh J, Houard M, Nseir S. Impact of chronic obstructive pulmonary Rodriguez A. TAVeM study Group: the association of cardiovascular failure disease on incidence, microbiology and outcome of ventilator-associated with treatment for ventilator-associated lower respiratory tract infection. lower respiratory tract infections. Microorganisms. 2020;8:165. Intensive Care Med. 2019;45:1753–62. 12. Robba C, Rebora P, Banzato E, Wiegers EJA, Stocchetti N, Menon DK, 30. Nair GB, Niederman MS. Ventilator-associated pneumonia: present under‑ Citerio G. Collaborative European NeuroTrauma effectiveness research standing and ongoing debates. Intensive Care Med. 2014;41:34–48. in traumatic brain injury participants and investigators: incidence, risk 31. Papazian L, Forel J-M, Gacouin A, Penot-Ragon C, Perrin G, Loundou A, factors, and effects on outcome of ventilator-associated pneumonia Jaber S, Arnal J-M, Perez D, Seghboyan J-M, Constantin J-M, Courant in patients with traumatic brain injury: analysis of a large, multicenter, P, Lefrant J-Y, Guérin C, Prat G, Morange S, Roch A. ACURASYS Study prospective, observational longitudinal study. Chest. 2020;158:2292–303. Investigators: neuromuscular blockers in early acute respiratory distress 13. Papazian L, Klompas M, Luyt C-E. Ventilator-associated pneumonia in syndrome. N Engl J Med. 2010;363:1107–16. adults: a narrative review. Intensive Care Med. 2020;46:888–906. 32. Shintani AK, Girard TD, Eden SK, Arbogast PG, Moons KGM, Ely EW. 14. Martin-Loeches I, Povoa P, Rodríguez A, Curcio D, Suarez D, Mira J-PJ-PJ-P, Immortal time bias in critical care research: application of time- Cordero MLML, Lepecq R, Girault C, Candeias C, Seguin P, Paulino C, varying Cox regression for observational cohort studies. Crit Care Med. Messika J, Castro AGA​GAG​, Valles J, Coelho L, Rabello L, Lisboa T, Collins 2009;37:2939–45. D, Torres A, Salluh J, Nseir S, Fernández RO, Arroyo J, Gabriela M, Alvarez R, Reyes AT, Dellera C, Molina F, Franco DM, et al. Incidence and prognosis of ventilator-associated tracheobronchitis (TAVeM): a multicentre, prospec‑ Publisher’s Note tive, observational study. Lancet Respir Med. 2015;3:859–68. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 15. Guidelines for the management of adults with hospital-acquired. lished maps and institutional affiliations. ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med. 2005;171:388–416. 16. Paul M, Shani V, Muchtar E, Kariv G, Robenshtok E, Leibovici L. Systematic review and meta-analysis of the efficacy of appropriate empiric antibiotic therapy for sepsis. Antimicrob Agents Chemother. 2010;54:4851–63.