Researchers have created a dataset of COVID-19 patients with acute respiratory distress syndrome to help characterize lung compliance and hypoxemia, which may help administer better treatments.
The severity of COVID-19, the disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), ranges from no symptoms in some patients to severe disease, including death in others. Some patients require mechanical ventilation for respiratory failure and acute respiratory distress syndrome (ARDS). How COVID-19 related ARDS is different from other ARDS is still a subject of debate. Previous studies have small sample sizes and often do not have data on lung mechanics.
To better understand how low blood oxygen level, or hypoxemia, and lung compliance varies in COVIDARDS and how they relate to disease outcomes, researchers developed a dataset of using more than 11,000 COVID-19 patients admitted to the hospital. A research paper published on the medRxiv* preprint server reports their results.
The authors performed a retrospective study of mechanically ventilated COVID-19 patients admitted to New York City hospitals between 1 March 2020 and 30 April 2020. They included oxygen saturation, fraction of inspired oxygen, static and dynamic lung compliance in their dataset and investigated different characteristics of COVIDARDS.
Of the more than 3,000 patients the team identified who were mechanically ventilated, 2020 were COVID-19 positive and 1,554 met the criteria the team set for inclusion in the study, which included reliable lung compliance data.
They found the average lung compliance was 24.44. mL/cm H2O. About 34.6% of the patients had deficient lung compliance, less than 20 mL/cm H2O, 63.2% with low-normal compliance (20-50 mL/cm H2O), and 2.2% had high compliance (more than 50 mL/cm H2O).
The patients' average age was 65 years, 32% were females, and 35% were white. There were more females in the very low compliance group and a larger proportion was non-white or multiracial. The most common co-morbidities were hypertension and diabetes. Almost all the patients were given hydroxychloroquine. About 83.5% of the patients in the very low compliance category and about 77% in the low-normal group received at least one vasopressor during the first two days.
COVIDARDS patients were intubated on average in about two days of admission and were intubated for an average of 14 days. The mean oxygen index 24 hours after intubation was 11.1 and a little worse for patients in the very low compliance category. They found that lung compliance decreased with time in all the patients. Of the 1554 patients, 67.5% died in the hospital, with the very low compliance group having the highest deaths.
Lung compliance not the same across patients
Of particular note in the dataset was the longer timer to intubate patients with very low lung compliance, the steeper decrease in compliance among those who died, and severe hypoxemia in high lung compliance.
It is possible that ARDS patients may have started out with normal lung compliance, which became worse with the disease or the treatments. Long exposure to high oxygen may contribute to low compliance in intubated patients. High oxygen has been shown to cause lethal lung injury in animals. Or, severe inflammation because of the disease could have led to lower compliance. Further studies that measure compliance over time before intubation and the progression of ventilation may help shed more light.
The ratio of oxygen partial pressure and fraction of inspired oxygen (P/F) improved after intubation in all groups, from a mean of 109 to 155. The very low compliance category showed the lowest improvement. However, the team did not see any correlation between the P/F ratio and lung compliance.
Although there have been questions about whether COVIDARDS should be treated differently than non-COVID ARDS, the authors write a more pertinent question would be if ARDS management should be different for patients with different lung compliance. An index that accounts for oxygen impairment and compliance over time may help doctors tailor treatments better.
The about 2% of patients in the high compliance group was lower than the 12% reported in a recent study of non-COVID ARDS patients, who did not receive ventilation until many days after hospital admission. So comparisons between studies should account for the timing of intubation.
Although the study's strength is its large sample size, one of the limitations is that it was a retrospective study and factors that influenced decisions to intubate could not be controlled. However, the dataset will help understand COVIDARDS further and provide treatment appropriately.
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.