It's Sunday...
This piece is courtesy of Dr. Laura Brenner.
|
|
|
The FLARE Four:
- As the global pandemic due to SARS-CoV-2 progresses, more and more data are becoming available about the clinical course and outcomes of patients with COVID-19.
- Tonight we review a recent case series from New York, New York.
- Clinical characteristics of patients described in this Open Letter to New York, are broadly similar to those described in earlier reports.
- Due to the short follow-up period, final outcome data are not yet available, but outcomes of ventilated patients thus far provide some reassurance.
|
|
|
|
Clinical characteristics of patients with COVID-19 in New York (Goyal et al. 2020)
|
|
|
|
This article presents a retrospective case series of 393 patients with confirmed COVID-19 admitted between March 5th and March 27th at a quaternary referral center (Cornell, so you can’t Take the ‘A’ Train) and an affiliated community hospital. Cases were confirmed via RT-PCR assays from nasopharyngeal swabs. Median age was 62.2 years. Presenting symptoms included cough (79.4%), fever (77.1%), dyspnea (56.5%), myalgias (23.8%), diarrhea (23.7%) and nausea and vomiting (19.1%). They adopted a self-described “early intubation” strategy with minimal use of high flow nasal cannula oxygen and non-invasive positive pressure ventilation.
|
|
|
A few interesting points stand out about this cohort:
Supplementary O2
Of patients who eventually received invasive mechanical ventilation during their hospitalization, just over 30% did not require supplemental oxygen during the first 3 hours of admission.
Gender
60% of the cohort was male, with 70% of those receiving mechanical ventilation being male. This is similar to Chinese cohorts (Guan et al. 2020), where 58.1% of all inpatients were male, and Italian studies, where 82% of patients admitted to the ICU were male (Grasselli et al. 2020).
Radiography
Like other cohorts, a majority (59.8%) of patients exhibited bilateral infiltrates on presentation while initial CXR was clear in 16.5% of all patients (and in 20% of patients who needed mechanical ventilation) and 15.5% had a unilateral infiltrate. 6.4% had a pleural effusion. In the largest Chinese cohort, 59.1% of patients had abnormal chest radiographs (Guan et al. 2020) at admission.
CBC
13% of all COVID patients had a WBC > 10K. 90% exhibited lymphopenia (<1500 per mm3), and 27% had thrombocytopenia.
Laboratory data correlated with need for mechanical ventilation on admission:
|
|
|
|
Complications during hospital stay:
|
|
|
|
|
N.b. Outcome data were incomplete for 23.7% of patients.
|
|
|
How does this compare to other COVID-19 ICU data?
|
|
|
|
Published case series vary in their inclusion criteria; some report all confirmed COVID-19 patients, some only ICU patients, and some results stratified by those who received mechanical ventilation and those who did not. However, if you take a New York Minute to pull some numbers together:
|
|
|
|
|
*median or mean, as reported
**reported metric varies between studies
***only reported in mechanically ventilated patients
MV = mechanical ventilation, RRT = renal replacement therapy
|
|
The above case series are not directly comparable across all data points, but several common features emerge. Overall, ICU patients are predominantly older, and presenting with moderate impairments in gas exchange. Also consistent across multiple reports are the high utilization of mechanical ventilation, low to moderate levels of PEEP and lower mortality than described in the earliest reports. While early series from China and Seattle (some that reported using relatively high levels of non-invasive support) reported high mortality, later series are reporting much lower initial mortality, but with prolonged periods of mechanical ventilation.
It is also important to note that in the NYC cohort, 4.3% of patients received remdesivir (April 16 FLARE) and 63.6% of all patients (97.7% of mechanically ventilated patients) received hydroxychloroquine (March 22 FLARE).
|
|
|
|
How do these common ARDS and sepsis complications compare to the pre-COVID-19 era?
|
|
|
Need for renal replacement therapy is reported in 13.3% of the patients who required mechanical ventilation in the NYC cohort. In previously reported cohorts of unselected ICU patients, the need for RRT ranged from 5-8% (Schneider and Bagshaw 2014). Analyses specifically querying the prevalence of RRT requiring AKI in ARDS have revealed rates of 21.8% (Panitchote et al. 2019) and 44.3% (Darmon et al. 2014).
The NYC cohort also suggested a substantial burden of arrhythmias (18% in mechanically ventilated patients), nearly all supraventricular in origin. These arrhythmias were primarily noted in mechanically ventilated patients, 95% of whom were treated with vasopressors. It is well-established that atrial fibrillation is common (25%) in sepsis (Walkey et al. 2013) and is associated with worse outcomes. The CENSER trial, which studied early norepinephrine administration in septic shock, noted that 20% of the usual care arm developed a new cardiac arrhythmia during the study period (Permpikul et al. 2019).
The NYC data are consistent with previous COVID-19 cohorts, where markers of inflammation were elevated in subjects who require mechanical ventilation. In this cohort, CRP, ferritin and LFTs are all higher in mechanically ventilated patients. This phenomena of increased inflammatory markers has been seen in other non-COVID cohorts. In a study that evaluated ferritin as a predictor of ARDS, a ferritin level of > 270ng/mL in women and > 680ng/mL in men predicted the development of ARDS in patients at risk for ARDS (Connelly et al. 1997).
|
|
|
|
The data from this NYC hospital and its associated community hospital provide important insight. Although the data are incomplete for many patients who remain critically ill, the rates of extubation and discharge to-date appear overall better than previously feared. It is challenging to directly compare cohorts between the published reports thus far given differences in proportion of incomplete data, patients populations, health system strain and resources, and mortality reporting metrics (i.e. overall mortality, ICU mortality, 28-day mortality). These early data from our NYC colleagues, if not quite a Fairytale of New York, provide some encouragement and reason for hope.
|
|
|
|
|
|
