Background: Severe acute respiratory syndrome coronavirus 2 (SARS-Cov-2), has been associated with higher fatality rate among elderly patients (pts). In pts >60 years (y), "early prediction for adverse outcome" is central to adopt therapeutic strategies to mitigate mortality. Multiorgan, but specifically, respiratory failure is associated with aging and known comorbid conditions [i.e obesity, cardiovascular disease]. Previous reports suggest that innate immunity effectors [i.e neutrophils and lymphocytes] "readapt" during acute infection. During acute COVID-19 infection, neutrophilia/lymphopenia are linked to "deregulated" cytokines, a phenomenon that seems to be associate with disease severity. In this study, we seek to investigate the predictive role of peripheral blood count modifications [absolute neutrophil (ANC), lymphocyte (ALC) and monocyte (AMC) counts], in risk evaluation for life-threatening COVID-19 among elderly pts. Methods: After IRB approval, 104 pts admitted at Baylor St Luke's Medical Center with confirmed COVID-19 PCR between April-June 2020 were selected for analysis. 66/104 (63.4%) pts were older than 60 y. In elderly pts, disease severity was estimated according to the World Health Organization and the National Health Commission of China guidelines (mild vs severe vs life-threatening). To detect at diagnosis differential ANC, ALC and AMC expression, cases were categorized into mild/severe (censor=0) vs life-threatening (censor=1). Student'st-test and chi-square analysis for continuous and categorical variables were performed. Univariate and multivariate logistic regression analysis including: (1) Hemopoietic factors [ANC, ALC, AMC] and (2) comorbid/inflammatory [CRP, body mass index (BMI), Charlson score] were obtained. SAS software was used for analysis. Results: 27/66 (40.9%) and 39/66 (50.09%) pts were mild/severe vs life-threatening COVID-19 disease. For mild/severe vs life-threatening disease, median age, BMI, Charlson comorbid score, were 68 vs 71 y, p=0.78; 27.7 vs 30, p=0.09; 5.40 vs 5.69, p=0.68, respectively. ANC, ALC, AMC, IL6 and CRP were 3.98 K/UL vs 6.79 K/uL, p=0.008; 0.90 K/UL vs 0.66 K/uL, p=0.01; 0.54 K/uL vs 0.43 K/uL, p=0.01; 43.2 vs 187.3, p=0.02; 10.1 vs 17.8, p=0.0001, for patients with mild/severe vs life-threatening disease (Fig 1). On univariate analysis, predictive variables for life-threatening disease were higher ANC (OR=1.24[1.02-1.49], p=0.02); lower ALC (OR=0.21[0.05-0.8], p=0.02), lower AMC (OR=0.11[0.01-0.86], p=0.03), and higher CRP (OR=1.19 [1.07-1.32], p=0.001). In multivariate analysis considering hemopoietic parameters, higher ANC (OR=1.53 [1.17-2.01], p=0.001) and lower AMC (OR=0.01 [0.001-0.28], p=0.006) retained predictive value. "Detectable evidence at diagnosis" for higher ANC, low AMC and ALC (3/3 hemopoietic factors) was discriminative for life-threatening COVID-19 disease (3 vs those with <3 factors, 100% vs 50%. Likelihood ratio= 7.73, p=0.01). A cutoff of 5.58 K/UL, 0.22 K/uL, 0.66 K/uL were associated severity prediction with sensitivity of 73.6%, 76.3%, 81.4% for ANC, AMC and ALC, respectively. Conclusions: In elderly COVID-19 pts exhibiting elevated CRP, a combination of high ANC, low monocyte and lymphocyte count is predictive for adverse COVID-19 outcome. Recent evidence suggests that monocytes "redistribute" from peripheral blood into the lungs to participate in tissue injury. Our data suggests that neutrophils/monocytes/lymphocytes induce differential "early signature" with ability for prediction of life-threatening disease. This signature may assist in identifying pts necessitating risk-adaptive interventions such as steroids, biologic agents and/or novel anti-inflammatory strategies.

Disclosures

Rivero:Incyte: Membership on an entity's Board of Directors or advisory committees; celgene: Membership on an entity's Board of Directors or advisory committees; agios: Membership on an entity's Board of Directors or advisory committees.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution