Hematologic Characteristics of Surviving Versus Non-Surviving COVID-19 Patients in Critical Care Units: Inner-City Hospital Experience, Detroit, MI

Background

As the COVID-19 pandemic spread across the US, the city of Detroit rose as one of the earliest infection epicenters in the nation. Henry Ford Hospital serves a diverse multi ethnic population in the inner city and throughout metropolitan Detroit, and thus provided care to a significant percentage of COVID 19 patients in the state of Michigan.

Retrospective reviews have described different hematologic abnormalities in patients with COVID-19. Various markers, such as lymphopenia, neutrophilia, elevated ferritin, and DDimer levels have been identified as predictors of poor outcomes, intensive care unit (ICU) admission, and mortality.

The aim of this study was to evaluate the impact of hematologic characteristics on the risk of intubation and mortality in our unique patient population infected with COVID-19 who required ICU admission.

Methods: This was a retrospective medical record review of adult patients with laboratory-confirmed COVID-19 requiring admission to adult ICU at Henry Ford Hospital in Detroit, MI, from March 1 to April 15, 2020. The main outcomes assessed were death and need for intubation and mechanical ventilation.

Results: A total of 229 patients met the study inclusion criteria.

Of the 76 surviving patients analyzed, 40 (53%) were men and 36 (47%) were women, including 59 (78%) Black and 11 (14%) White patients. The mean age at presentation was 61 (SD, 15) years. Hypertension was the most common comorbidity (n=65 patients), followed by diabetes (n=39). Most surviving patients (n=60; 79%) were admitted to general practice units (GPU) and then transferred to ICU, while 49 (64%) patients required intubation.

Among the 153 patients who died, 95 (62%) were men and 58 (38%) were women, with 41 (27%) being White and 99 (65%) Black. The mean age at presentation was 69 years (SD, 14.3). Hypertension was the most common presenting co-morbidity (n=118), followed by diabetes (n=72). More non-surviving patients were admitted directly to the ICU (n=71; 46%) and/or required intubation (n=147; 96%).

Hematologic laboratory findings of the studied patients are represented in Table 1.

Multivariate analysis of variables predictive of death and intubation are summarized in Table 2.

Logistic modeling revealed an 18% increase in the odds of death and a 17% increase in odds of intubation for each unit increase in WBC. Also, for each unit increase in ANC, there was a 17% increase in odds of death and 21% increase in odds of intubation.

Conclusions:

Among patients with COVID-19 who required admission to ICU, Black patients were over-represented. Coronary artery disease (CAD), older age, and a lower oxygen saturation (SpO2) on initial triage in the emergency department were associated with increased mortality. A higher absolute neutrophil count (ANC) and white blood cell count (WBC) was associated with higher risk of intubation and death. Male sex and direct ICU admission were predictors of increased risk of intubation - and these patients were more likely to die. Contrary to other reports, lymphopenia did not increase odds of intubation or death in ICU, and neither the serum ferritin nor DDimer levels on admission were discriminators for death; however, higher peak levels during the hospitalization were linked to increased mortality.

Inner city Black populations with advanced age, multiple co-morbidities, and COVID-19 may be at increased risk for ICU admission, and thereby at an increased risk of death. However, not all hematologic characteristics are generalizable with regard to intubation and mortality in ICU.