Peripheral Blood Examination Findings in SARS-CoV-2 Infection

Alia Nazarullah, MD; Christine Liang, MD; Andrew Villarreal, MLS; Russell A. Higgins, MD; Daniel D. Mais, MD


Am J Clin Pathol. 2020;154(3):319-329. 

In This Article

Materials and Methods

The study was approved by the institutional review board. The list of all patients tested for SARS-CoV2 infection was retrieved. Clinical criteria recommended by the Centers for Disease Control and Prevention (CDC) were adopted by our institution for COVID-19 testing. Detection of SARS-CoV-2 was carried out using a modified CDC emergency use authorization–approved polymerase chain reaction (PCR) assay that had been successfully validated in our laboratory. This method involves nucleic acid extraction from nasopharyngeal swabs using Nuclisens easyMAG (bioMérieux) or Maxwell RSC 48 (Promega). Reverse transcription and then real-time PCR reactions were performed using 2 sets of CDC primers targeting the SARS-CoV-2 nucleocapsid genes N1 and N2 on the QuantStudio-3 Real-Time PCR system (ThermoFisher). Specimens were considered positive if either N1 or N2 was detected, and they were considered negative if neither was detected in the presence of appropriately detected controls.

As of April 22, 2020, a total of 1,584 cases were tested, among which SARS-CoV-2 virus was detected by PCR in 98 patients. Electronic medical records of the positive cases were reviewed for evidence of any hematologic tests performed. Twelve patients had peripheral blood smears available for review, and 6 of them had flow cytometric studies performed. Ten patients who met symptomatic criteria for COVID-19 testing, but who tested negative, had smears available for review. These 10 were reviewed as the control group. The peripheral blood smears were reviewed by 4 board-certified pathologists, which included 2 board-certified hematopathologists. The pathologists were blinded to the COVID-19 test result of patients during smear review. Statistical analysis was performed on the Microsoft Excel platform.

Complete blood cell counts (CBCs) were performed using Sysmex XN-9000. Leukocyte differential counts were performed using Sysmex DI-60 Cellavision software, and manual differential counts were performed when flagged for abnormalities. Peripheral blood smears were stained with Wright stain.

With regard to granulocytes, the presence of APHA, monolobate neutrophils, and toxic changes were recorded as a percentage of total granulocytes. These changes were reported as present if they were noted in more than 1% of granulocytes. Left shift of granulocytes is reported as present using the following cutoffs per total granulocyte count: bands, 10%; metamyelocytes and myelocytes, 5%; promyelocytes, 1%; and blasts, greater than 0%. Toxic changes in granulocytes were defined as presence of at least 2 of the following features: prominent cytoplasmic granulation, cytoplasmic vacuolization, and Döhle bodies, present in at least 5% of granulocytes.

The presence of atypical lymphocytes was noted, and atypical lymphocytes were further classified as Downey type I, Downey type II, Downey type III, plasmacytoid lymphocytes, large granular lymphocytes (LGL above normal ranges), and other atypical lymphocytes. The individual categories of atypical lymphocytes were reported as a percentage of total lymphocytes. Flow cytometric immunophenotyping for lymphocyte subsets was performed using BD FACSCanto II (Becton Dickinson). The antibodies analyzed included CD3, CD4, CD8, CD19, and CD16/56 with percentages and absolute counts reported.