COVID-19 in Hospitalized Patients with Sickle Cell Disease

Introduction

Sickle cell disease (SCD) is the most common inherited hemoglobinopathy and is estimated to affect more than 100,000 Americans. Current Centers for Disease Control statistics indicate that Black Americans die from COVID-19 at a disproportionately high rate, 13.6% of 449, 373 deaths but only account for 12.54% of the United States Population (CDC COVID Data Tracker accessed 5/5/2021). A voluntary clinical reported registry of COVID-19 infections in patients with SCD reported both high hospitalization rates (69%) and case fatality rates (7%) (Panepinto, 2020), but only reported data from March 20-May 21, 2020. A retrospective cohort review from England demonstrated a 4-fold increased risk of hospitalization and 2.6-fold increased risk of death due to COVID-19 for those patients with sickle cell disease compared to the general population (Hippisley-Cox, 2021).

The unique constellation of SCD manifestations complicate both the diagnosis and management of COVID-19, particularly related to anticoagulation and transfusion practices. Understanding the impact of early exchange and anticoagulation would guide development of appropriate treatment guidelines and future understanding of pathophysiology. We report on the outcomes for all hospitalized inpatients with SARS-CoV-2 and SCD at institutions using Cerner electronic health record (EHR).

Methods

Exempted retrospective review approved by ChristianaCare IRB. We obtained deidentified data from the Cerner COVID Data Lab which includes information on all hospitalized inpatients with SARS-CoV-2 and sickle cell disease as documented by ICD 10 D57.XX from 12/10/2019-10/15/2020 at institutions that use the Cerner EHR. Those with sickle cell trait excluded. The data included 209 patients; 1 patient had separate 2 visits, only the first visit in the data. Evaluated anticoagulation use, not dose. Descriptive statistics are given: percentages for categorical variables and mean (standard deviation) for continuous variables. Comparisons between patients who died or went to hospice versus patients who recovered were done using unadjusted chi-squared tests or t-tests.

Results:

Admission:

124 (74.3%) were afebrile (temp <100)

52 (28.4%) were tachypneic (RR >22)

10 (6.2%) were hypoxic (<92%)

7 (3.3%) were intubated during hospitalization

Many patients had comorbidities including diabetes, hypertension, and congestive heart failure, but it was not clear if patients had multiple co-morbidities.

Treatment:

16 (7.7%) got transfused RBC between 1.77 and 589.18 hours into admission

5 (2.4%) got Remdesivir, none of these patients died/went to hospice.

No exchange transfusions, but maybe wasn't captured in coding data

149 (71.3%) received anticoagulant, dosing unable to be obtained so not known if this was prophylactic or treatment dosing.

Outcomes:

158 (80.2%) discharged home

18 (9.1%) discharged facility

8 (4.1%) died

2 (1.0%) discharged hospice

11 (5.6%) left against medical advice

12 (5.7%) missing disposition data

Those who died/went to hospice had longer hospital stays, were more likely to be hypoxic and initially tachypneic. None of these patients received remdesivir.

Study included small numbers but those who died more likely to have hypertension, diabetes w/ and w/out complications, CHF.

Discussion

This data only included hospitalized patients, doesn't account for patients who remained outpatient so case fatality rate likely lower. 10/209 patients died- 4.8% fatality rate, 12 patients missing final discharge disposition. 18 went to facility, may have died as outpatient. Only a small number of patients received remdesivir and there were overall low rates of anticoagulation, concerning given SCD is a hypercoagulable state.

Study limitations include only hospitalized patients in the dataset and only draws from Cerner EMR institutions, a 26% market share. Based on our own SCD population, only a small percentage of patients with SCD and COVID-19 hospitalized. Additionally, there is likely significant variability between institutions' treatment protocols, particularly in the early months of the data set. Finally, we could not adequately gauge severity of COVID-19 disease given notable variations in institutional resources.

TABLES (in process):

TABLE 1- Demographics- split by death/hospice vs. other dispositions

TABLE 2- Admission Characteristics- Death/hospice vs. others

TABLE 3- Treatments- Death/Hospice vs. others