Elderly Male With Cardiovascular-Related Comorbidities Has a Higher Rate of Fatal Outcomes: A Retrospective Study in 602 Patients With Coronavirus Disease 2019

Abstract

Elderly with comorbidities have shown a higher rate of fatal outcomes when suffering coronavirus disease 2019 (COVID-19). However, a delineation of clinical significances of hematologic indices and underlying comorbidities in the progression and outcome of COVID-19 remains undefined. Six hundred two COVID-19 patients with established clinical outcomes (discharged or deceased) from Hankou Hospital of Wuhan, China between January 14, 2020 and February 29, 2020 were retrospectively analyzed. Of the 602 patients with COVID-19, 539 were discharged and 63 died in the hospital. The deceased group showed higher leukocyte and neutrophil counts but lower lymphocyte and platelet counts. Longer activated partial thromboplastin time (APTT) and prothrombin time (PT), as well as higher D-dimer and C-reactive protein levels, were found in non-survivors. Our observations suggest that these parameters could serve as potential predictors for the fatal outcome and in the discharged group. A higher neutrophil count and D-dimer level but lower lymphocyte were associated with a longer duration of hospitalization. A multivariable Cox regression analysis showed that higher neutrophil count, prolonged PT, and low lymphocyte count were risk factors for patients with COVID-19. Also, we found an association of lower lymphocyte count and higher C-reactive protein levels with the elderly group and those with cardiovascular-related comorbidities. The significantly different hematologic profiles between survivors and non-survivors support that distinct hematologic signatures in COVID-19 patients will dictate different outcomes as a prognostic marker for recovery or fatality. Lymphopenia and aggressive inflammatory response might be major causes for fatal outcomes in the elderly male and especially those with cardiovascular-related comorbidities.

Keywords: COVID-19; aggressive inflammatory response; cardiovascular-related comorbidities; elderly male; lymphopenia.

Figure 1

Different levels of hematologic indices (A) leukocyte count, (B) neutrophil count, (C) lymphocyte count, (D) platelet count, (E) CRP level, (F) D-dimer level, (G) APTT, and (H) INR between the discharged and deceased patients. Data are shown as a violin plot with median and 25 and 75% percentile lines. *P < 0.05, ****P < 0.0001. CRP, C-reactive protein; APTT, activated partial thromboplastin time; INR, international normalized ratio.

Figure 2

Kaplan–Meier survival curves for different prognostic factors. The curves according to (A) leukocytes, (B) neutrophils, (C) lymphocytes, (D) CRP levels, (E) platelets, (F) APTT, (G) PT, (H) TT, (I) FIB levels, (J) INR, and (K) D-dimer levels. The patient number of each group was indicated next to the curve. CRP, C-reactive protein; APTT, activated partial thromboplastin time; INR, international normalized ratio; PT, prothrombin time; TT, thrombin time; FIB, fibrinogen.

Figure 3

Correlations between hospitalization days and on admission levels of (A) leukocytes, (B) blood oxygen saturation, (C) neutrophils, and (D) D-dimers.

Figure 4

Correlation networks for hematologic indices. Networks showed different profiles of correlations in COVID-19 survivors (A) and non-survivors (B), on admission. (C) PCA biplot of hematologic indices on admission. Individuals are shown as dots and colored by outcomes (survivors and non-survivors). Indices showed as lines with arrows and colored by positive or negative contribution to PC1. The configuration of indices in biplot represented the relationship between variables and principal components. PCA, principal component analysis.

Figure 5

Association of underlying comorbidities and hematologic indices. (A) Kaplan–Meier survival curves for different underlying comorbidities. The number of patients in each group is indicated next to the curve. Hematologic indices in the four groups that showed to be better fit the survival curves including (B) neutrophils, (C) lymphocytes, (D) leukocytes, and (E) CRP levels. Data are shown as dots with median lines. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 6

Hematologic variations between the young and old. Kaplan–Meier survival curves for (A) age and (B) underlying comorbidities are shown. (C) The relationship of age and underlying comorbidities was analyzed by chi-square test. Kaplan–Meier survival curves for age in those patients (D) without any underlying comorbidities, (E) those without cardiovascular-related comorbidities, (F) those with only cardiovascular-related underlying comorbidities, and (G) those with cardiovascular-related underlying comorbidities. Hematologic indices that were found to have different levels between the young and old including (H) lymphocytes, (I) neutrophils, and (J) CRP and were also correlated with age (K–M). Data are shown as boxes and whiskers. Correlations are colored by positive (red) or negative (green). The numbers next to the survival curves indicate quantities of patients in such a group. ***P < 0.001 and ****P < 0.0001.

Figure 7

Hematologic variations between the male and female. (A) Kaplan–Meier survival curves for genders are shown. (B) Composition of different underlying comorbidities in the male and female. Different levels of hematologic indices (C) leukocytes, (D) neutrophils, (E) lymphocytes, (F) platelets, (G) CRP level, (H) APTT, (I) PT, (J) D-dimer level, (K) FIB, and (L) INR between the male and female. Data are shown as a violin plot with median and 25 and 75% percentile lines. **P < 0.01, ****P < 0.0001. The numbers next to the survival curves indicate quantities of patients in such a group.