Changes in utilization and outcomes of mechanical ventilation of COVID-19 during the course of the pandemic in Germany in 2020: an observational study of 7,490 patients

2021 
BackgroundThe role of non-invasive ventilation (NIV) in severe COVID-19 requiring mechanical ventilation (MV) remains a matter of debate. MethodsIn this observational study of confirmed COVID-19 cases claims data of the Local Health Care Funds of MV patients were comparably analysed: spring period of 2020 (February-May) versus autumn period (October/November). FindingsIn a nationwide cohort 7,490 cases were included: median age 70 (IQR 60-79) years, 66% male, hypertension 67%, diabetes 42%, cardiac arrhythmia 43%, congestive heart failure 34%, renal failure 27%. Overall, 3,851 (51%) patients primarily received invasive MV without NIV, 1,614 (22%) patients received NIV without having been escalated to intubation, and 1,247 (17%) patients had NIV failure (NIV-F), defined by endotracheal intubation following NIV. Comparing cases of the first and second period, the proportion of patients who received invasive MV decreased from 74% to 39%. Accordingly, the proportion of patients with NIV without subsequent intubation increased from 10% to 28%, and those failing NIV increased from 9% to 21%. The overall median length of hospital stay decreased from 26 to 22 days, and the overall duration of MV decreased from 11.6 to 7.6 days. The NIV failure rate decreased from 49% (219/449) to 42% (927/2,185). Overall mortality remained unchanged (51% and 53% respectively). Mortality was 39% with NIV only, 52% with invasive MV and 66% with NIV-F with mortality rates steadily increasing from 58% in early NIV-F (day 1) to 75% in late NIV-F ([≥]5 days). InterpretationThe utilization of NIV rapidly increased during the autumn period compared to the spring period 2020, which was associated with a reduced duration of MV and hospital stay, but not with overall mortality. NIV-F rates are high and are associated with increased mortality rates, particularly in late NIV-F. In contrast, NIV success is associated with the lowest mortality rates. FundingInstitutional support and physical resources were provided by the University Witten/Herdecke and Kliniken der Stadt Koln and the Federal Association of the Local Health Care Funds. Research in contextO_ST_ABSEvidence before this studyC_ST_ABSCurrent management of ventilatory support in COVID-19 patients with respiratory failure is heterogeneous. Despite increasing use of non-invasive ventilation (NIV), defining intubation criteria still remains a matter of uncertainty and discussion, especially with regard to the balance between the NIV benefits and the risk of NIV failure. In addition, robust data concerning the influence of the duration and failure of NIV on intubation and mortality rates are still missing, although the time span between initiation of NIV and subsequent intubation in case of respiratory failure progression is suggested to influence patient outcome. Added value of this studyThis is the first large observational study describing differences of ventilatory strategies between the spring and autumn period of the SARS-CoV-2 pandemic in Germany and provides the in-hospital mortality rate of 7,490 patients who received mechanical ventilation. The increased utilization of NIV from 10% (first period) to 29% (second period) was associated with overall reduced durations of mechanical ventilation and length of hospital stay, but overall mortality remained comparably high and reached 51%, 53% respectively. Patients succeeding with NIV had lower mortality rates than those getting intubated without preceding NIV attempts, but those failing NIV had higher mortality rates, respectively, and this became even more predominant in late NIV failure. Implications of all available evidenceThe present observational study shows the increasing role of NIV in the concert of ICU medicine related to COVID-19, but also clearly addresses its risks in addition to its benefits, both impacting on mortality.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    32
    References
    6
    Citations
    NaN
    KQI
    []