Non-invasive ventilation in >= 70 year old COVID-19 patients in a level two unit: retrospective cohort analysis, comparison with COVIP substudy and defense of a multifaceted work of breathing control driven strategy

Introduction: The debate about optimal management of patients with COVID-19 ARDS remains, including medical treatment, ventilatory strategies, awake proning and others. COVIP is a multicentric observational study with over 3000 patients under NIV. A substudy by Polok and al. evaluated patients (PTS) >= 70 years old. At our intermediate care unit (IU) we used a strategy of high dose corticosteroid started when the work of breathing (WOB) increased, prolonged awake prone positioning (> 12 h) and high CPAP ventilatory strategy. We describe our cohort of >= 70 years old NIV PTS and compare it to COVIP substudy results. Method(s): Descriptive retrospective study. Data were collected from electronic medical records of 95 COVID-19 PTS aged 70 years old or above under NIV at the IU between September/20 and March/21. Categorical data are presented as frequency (percentage) and were compared using chi2-test. Continuous variables were compared using Mann-Whitney U test. Cohort results were compared with those from Polok et al. COVIP substudy (COVIPss). Result(s): 95 of PTS were submitted to NIV. Median age was 76 years and 49.5% were male, versus 75.7 and 71.4% in COVIPss. Median admission SOFA score was 4 and CFS was 3 with 14% considered frail (CFS > 5). In COVIPss median SOFA was 5 and 17% of PTS were frail. The preferred mode was CPAP with median maximum pressure of 13. Mean PaO2/fiO2 ratio after start of NIV was 125, 30% < 100. NIV failure occurred in 46.3% versus 74,7% in COVIPss. Our intra-unit mortality was 31.6%. 14 PTS (14.7%) were submitted to invasive mechanical ventilation and 57% of those died. In COVIPss mortality at 30d was 52.9% in NIV and 47.7 in IMV groups. Conclusion(s): We argue that NIV is a valid option for COVID ARDS management if supported by a multifaceted strategy such as ours, using prone and CPAP for WOB control. We agree with COVIPss authors as NIV trial should be short and intubation promptly if WOB not controlled. Comparison with COVIP substudy NIV failure and mortality results, support our belief.

Acquired infections in intensive care unit (ICU) COVID-19 patients

Introduction: Patients with COVID-19 admitted to the ICU are at high risk of developing infectious complications during their ICU stay. Data on acquired(AI) in Portuguese critical COVID-19 patients are scarce. The aim of this study was to investigate the characteristics and risk factors for AI in critical patients with COVID-19 pneumonia admitted to the ICU. Methods: Retrospective cohort of patients with COVID-19 pneumonia admitted to an ICU in a tertiary hospital, between September 2020 and June 2021. AI considered were ventilator-associated pneumonia (VAP) or tracheobronchitis (VAT), bacteremia, CVC associated infections, urinary tract infections and soft skin tissue infections. Baseline characteristics, 3-months previous antibiotic (ATB) exposure, ATB treatment at ICU-admission and clinical management of COVID-19 pneumonia were analyzed. Results: Of the 159 patients included, with a median (IQR) age of 66 (57-72) and 63.5% males, 14 (8.8%) had no known comorbidities. A total of 63 patients(39.6%) developed AI: 45(71.4%) VAP, 20(33.3%) VAT, 28 (45.2%) UTI, 6 (9.5%) CVC associated infections and 3(4.8%) soft skin tissue infections. In univariate analysis, both SOFA score at admission (p < 0.001), acute cardiovascular (p = 0.003) and neurologic (p = 0.006) disfunction at ICU admission were associated with the development of AI. AI were also correlated to need of tracheostomy(p < 0.001), development of delirium (p < 0.001) or shock (p < 0.001);and with longer ICU and in-hospital stay (p < 0.001) and ICU and hospital mortality (p = 0.011 and p = 0.011, respectively). None of the COVID-19 pharmacologic treatments considered (remdesivir, steroids and tocilizumab), neither different regimens of ATB therapy at ICU admission were significantly associated with AI. Conclusions: In this cohort, almost 40% of the patients developed AI, that was associated with 4 times higher hazard of needing mechanical ventilation and higher rate of adverse events such as delirium, shock during in-ICU stay and longer length of ICU and in-hospital stay.

Characteristics and outcomes of critical COVID patients in the intensive care unit

Introduction: This study aimed to determine the mortality and morbidity of COVID-19 patients in an intensive care unit (ICU) until hospital discharge, and explore the factors that influence in-ICU and in-hospital mortality rates. Methods: Single center retrospective cohort regarding COVID-19 critical patients in a tertiary hospital ICU, from September/20 to June/21. Demographic data, clinical characteristics, admission SOFA score, frailty score (FS) and clinical management were analyzed. Results: We included 159 consecutive COVID-19 critical patients. The median (IQR) age was 66(57-72);101(63.5%) were male. A total of 126 (79.2%) patients received hospital discharge, ICU-mortality rate was 18.9%(30 deaths). The median (IQR) ICU length of stay was 12 days (6-20) and in-hospital stay was 21(13-35), and no significant differences were found in ICU and in-hospital length of stay between survivors and non-survivors. At admission to the ICU total SOFA score was 4(3-7). In univariate analysis, increased age, higher admission SOFA score, acute kidney injury and acute neurologic disfunction at admission were significantly associated with increased hazard of mortality. The need for mechanical ventilation were associated with higher risk of ICU and in-hospital mortality. Previous comorbidities (hypertension, diabetes, obesity, heart failure, COPD, renal, hepatic, oncologic or immunosuppression) or the FS were not significantly associated with in-hospital mortality. None of the COVID-19 pharmacologic treatments (remdesivir, steroids and tocilizumab) were significantly associated with in-hospital mortality. In a multivariable analysis with in-hospital death as the dependent variable, a 10 year increase in age was associated with a mortality OR of 2.9 (95 CI:1.5-5.5)( p = 0.002) and the development of shock during ICU stay was associated with a mortality OR of 8.8 (95 CI:1.5 to 53.3). Conclusions: In this cohort, only age and the development of shock during ICU stay were independently associated with higher risk of inhospital death.

Non-invasive ventilation in COVID-19 patients, the experience of a level 2 unit

Introduction: COVID-19 has a broad spectrum of severity and, although the majority of those infected are asymptomatic or have mild disease, many need hospitalization and organ support for respiratory failure. The approach to this dysfunction varied across the pandemic, influenced by retrospective data and centre experience. After initial unfavorable data, NIV resumed prominence during the 2nd wave, having been the modality of choice in our intermediate care unit (IU). We describe our NIV cohort and the results of our ventilatory strategy. Methods: Descriptive retrospective study. Data were collected from electronic medical records of 202 COVID-19 patients (PTS) under NIV at the IU between September/20 and March/21. Categorical data are presented as frequency (percentage) and were compared using χ2 -test. Continuous variables were compared using Mann-Whitney U test. Statistical significance was set at p < 0.05. Results: 202 of 469 PTS were submitted to NIV. Mean age was 66 years and 62.8% were male. Most common comorbidities were hypertension, dyslipidemia, obesity and diabetes. Mean admission SOFA score was 3.6. Most PTS underwent corticosteroid therapy, 86.7% in > 1 mg/ kg dosage equivalent. Remdesivir was used in 50%. In 88.6% NIV was the initial modality of ventilatory support, 11.4% after HFNC failure (23). The preferred mode was CPAP with mean maximum pressure of 13 (6-16), titrated to normalization of the work of breathing (WOB). Mean PaO2/FiO2 ratio at start of NIV was 122, < 100 in 43% of PTS. NIV failure occurred in 35.6%, intra-unit mortality was 25.6%. 35 PTS were submitted to invasive mechanical ventilation (IMV), 41% died. Advanced age, intolerance to awake prone and delirium were associated with higher mortality. Conclusions: NIV is a valid option for the management of respiratory failure secondary to COVID-19 ARDS, reducing the need for IMV. Elevated CPAP values, titrated to WOB control, complemented with prolonged periods of awake prone are essential for success.

Impact of remdesivir in COVID-19 patients under non-invasive ventilation in an intermediate care unit

Introduction: COVID-19 has generated enormous difficulties globally due to the high number of critically ill patients and uncertainty of the best therapeutic approach, even after 18 months of pandemic and multiple clinical trials. The antiviral remdesivir (RDV) has shown to reduce time to clinical recovery and, in a subgroup with low flow O2 at time of drug initiation, to reduce mortality by 70% (ACTT-1). Subsequent openlabel RCT, Solidarity and Discovery, didn't confirm these findings. In our unit, a strict protocol was used, including a 5-day cycle of 20 mg dexamethasone and start of HFNC/CPAP when increased work of breathing became noticeable, along with prolonged periods of awake prone position. The use of RDV was a point of significant variability, allowing us to compare outcomes. We describe our unit's experience and RDV impact on patients under non-invasive ventilation (NIV). Methods: Descriptive retrospective study. Data were collected from 202 COVID-19 patients under NIV at our intermediate care unit between September/ 2020 and March/2021, through medical records in the electronic clinical file. Categorical data are presented as frequency (percentage) and were compared using χ2 -test. Continuous variables were compared using Mann-Whitney U test. Statistical significance was set at p < 0.05. Results: Each group consisted of 101 patients, with the group not submitted to RDV being slightly older (mean age 70.5 vs 63 years), more frail (mean CFS 3.5 vs 2.8) and with higher mean SOFA at admission (4.0 vs 3.2). The RDV group had a lower mortality rate (20.8 vs 52.5%;p < 0.001), less NIV failure (20.8 vs 50.5%;p < 0.001), shorter duration of ventilation in survivors (7.0 vs 8.5 days;p = 0.036) and less need for intensive care admission (14.9 vs 23.8%), with favorable impact on mortality (26.6 vs 50%) in this subgroup. Conclusions: In our cohort of patients under NIV, RDV use was associated with lower mortality, less need for IMV and shorter duration of ventilation.